Materials Science and Engineering C 56 (2015) 286–293

Contents lists available at ScienceDirect

Materials Science and Engineering C

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

Microwave assisted synthesis and characterization of magnesium

substituted calcium phosphate bioceramics
Nida Iqbal Khan a,b, Kashif Ijaz a, Muniza Zahid a, Abdul S. Khan a, Mohammed Rafiq Abdul Kadir b,
Rafaqat Hussain c, Anis-ur-Rehman d, Jawwad A. Darr e, Ihtesham-ur-Rehman f, Aqif A. Chaudhry a,⁎
a
Interdisciplinary Research Centre in Biomedical Materials, COMSATS Institute of Information Technology, M. A. Jinnah Campus, Defence Road, Off Raiwind Road, Lahore, Pakistan
b
Medical Implant Technology Group (MEDITEG), Faculty of Bioscience and Medical Engineering, Universiti Teknologi Malaysia, 81310 Skudai, Johor Darul Takzim, Malaysia
c
Department of Chemistry, Faculty of Science, Universiti Teknologi Malaysia, Skudai, Johore, Malaysia
d
Department of Physics, COMSATS Institute of Information Technology, Chakshahzad Campus, Islamabad, Pakistan
e
Clean Materials Technology Group, Department of Chemistry, University College London, Christopher Ingold Laboratories, 20 Gordon Street, London WC1H 0AJ, UK
f
The Kroto Research Institute, North Campus, University of Sheffield, Broad Lane, Sheffield S3 7HQ, UK

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

Article history: Hydroxyapatite is used extensively in hard tissue repair due to its biocompatibility and similarity to biological ap-
Received 2 May 2014 atite, the mineral component of bone. It differs subtly in composition from biological apatite which contains other
Received in revised form 23 March 2015 ions such as magnesium, zinc, carbonate and silicon (believed to play biological roles). Traditional methods of hy-
Accepted 7 May 2015
droxyapatite synthesis are time consuming and require strict reaction parameter control. This paper outlines syn-
Available online 9 May 2015
thesis of magnesium substituted hydroxyapatite using simple microwave irradiation of precipitated suspensions.
Keywords:
Microwave irradiation resulted in a drastic decrease in ageing times of amorphous apatitic phases. Time taken to
Magnesium substitution synthesize hydroxyapatite (which remained stable upon heat treatment at 900 °C for 1 h) reduced twelve folds
Microwave synthesis (to 2 h) as compared to traditionally required times. The effects of increasing magnesium concentration in the
Bioceramics precursors on particle size, surface area, phase-purity, agglomeration and thermal stability, were observed
Calcium phosphates using scanning electron microscopy, BET surface area analysis, X-ray diffraction and photo acoustic Fourier trans-
form infra-red spectroscopy. Porous agglomerates were obtained after a brief heat-treatment (1 h) at 900 °C.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction stimulate osteoblast proliferation [20,21]. During synthesis magnesium

Hydroxyapatite [HA] is a key bioceramic due to its excellent biocom-
patibility and its bioactivity which stems from its compositional and
structural similarity to the mineral part of bone, enamel and dentine
[1–4]. Although bone mineral is an apatitic phase it also contains addi-
tional ions which include magnesium, carbonate, strontium, zinc and
silicate ions [5–7]. The type and amount of ionic substitution in biolog-
ical apatite varies from wt% level (e.g. CO23 −) to the ppm–ppb level
(Mg2+ or Sr2+) [8–10]. These ionic substitutions affect the crystal lat-
tice structure of HA and influence the dissolution rate and hence the
resulting bioactivity of the bone mineral [11–13].
Magnesium substitution into HA is of interest due to its potential im-
pact on the mineralization process and crystallization of apatite [14–16].
It has been reported that in calcified tissues, the amount of magnesium
associated with the apatite phase is higher (5% at.) at the first stages of
the bone remodeling process and decreases with increasing calcification
and ageing of an individual [17–19]. Moreover, magnesium can directly
⁎ Corresponding author.
E-mail address: aqifanwar@ciitlahore.edu.pk (A.A. Chaudhry).
ions also bind to the surface of apatite crystals inhibiting their growth
resulting in smaller particles (higher bioactivity due to higher surface
area) [22,23]. Decrease in content of magnesium in bone and loss of
some crystallinity are associated with osteoporosis [24,25]. These
adverse effects related to magnesium deficiency can impair bone
growth and reduce bone quality (strength and density) and increase
bone fragility [5,20].
As the magnesium ion is smaller than the a calcium ion (0.86 Å vs.
1.14 Å), the incorporation of magnesium affects the crystal structure
of HA and substitution cannot occur over the full compositional range
(i.e. magnesium ions cannot replace all calcium ions in HA) [26,27].
Substituting magnesium ions into HA lattice reduces crystal cell param-
eters and hence cell volume as expected; above a concentration limit of
~0.15Mg2+/Ca2+ the apatite structure cannot accommodate any more
magnesium resulting in the formation of additional phases (typically
TCP or amorphous calcium magnesium phosphates), that act as segre-
gating phases for this ion, is observed [28,29]. TCP phases are known
to have higher solubility than that of HA [30].
Several methods can be used for preparing calcium phosphates in-
cluding magnesium substituted HA (Mg–HA), such as wet chemical
routes based on precipitation at low temperature [22,31–36] or solid-

http://dx.doi.org/10.1016/j.msec.2015.05.025
0928-4931/© 2015 Elsevier B.V. All rights reserved.
 N.I. Khan et al. / Materials Science and Engineering C 56 (2015) 286–293 287

state reactions at elevated temperatures [37,38] resulting in a range of
0.3–29 wt.% magnesium substitution. Traditional co-precipitation
routes often require aging times of 18 h or more in order for the Ca:P
phases to reach maturation (and hence thermal stability) [39,40].
Other methods such as sol–gel syntheses [41], batch hydrothermal syn-
theses [5], mechanochemical-hydrothermal synthesis methods [42,43]
and continuous hydrothermal flow synthesis [29] have been reported
to produce phase pure materials in the range of 0.5–28 wt.% magnesium
substitution. Some of these techniques require expensive starting
materials, strict reaction parameter control and monitoring or require
expensive equipment. Therefore, a technique that reduces the time of
hydroxyapatite synthesis without compromising on its desired proper-
ties and also ensures sufficient ion substitution into hydroxyapatite
would be of use.
The advantages of microwave irradiation during chemical reactions
include acceleration of the reaction, smaller particle sizes with narrow
particle distribution and finer nanostructures leading to improved me-
chanical properties when formed into dense test specimens [44–49].
Despite the benefits, ion substitutions using microwave assisted ap-
proaches have not been reported extensively in literature. Magnesium
[50] and silver [51] ion substitutions in HA have been reported using
microwave assisted methods. The authors previously reported on the
magnesium ion substitution into the HA lattice using Continuous
Hydrothermal Flow Synthesis (CHFS) Technology [29]. CHFS synthesis
of Mg–HA exploited the benefits provided by near-critical water to re-
sult in rapid maturation (increase in Ca:P ratio from an initial amor-
phous precipitate). CHFS is a high throughput technique, however it
requires an expensive setup which works at high temperature and pres-
sure [52–54]. The authors have also previously reported on the rapid
synthesis of thermally stable hydroxyapatite using microwave irradia-
tion and its deposition on metallic substrates [55,56]. In the current
work, microwave irradiation was used as a much simpler alternative
method to controllably produce a range of magnesium substituted cal-
cium phosphates with differing compositions and thermal stabilities
in a rapid manner.
2. Materials and methods
2.1. Materials
Diammonium hydrogen phosphate (NH4)2HPO4 (Applichem,
Darmstadt, Germany, 96–102%), calcium nitrate tetrahydrate Ca(NO3)2·
4H2O (Unichem, Guangzzhou, China 99%), magnesium nitrate hexahy-
drate Mg(NO3)2·6H2O (Pamrea quimica SA, Barcelona, Spain 98%) and
ammonium hydroxide NH4OH (Fisher Scientific, Loughborough, UK,
33%) were used. Deionized water was used in all reactions. The wet-
chemical precipitation method for synthesis of HA can be expressed by
the following equation:
xMgðNO3 Þ2 þ ð10−xÞCaðNO3 Þ2 þ 6 ðNH4 Þ2 HPO4 →Ca10−x Mgx ðPO4 Þ6 ðOHÞ2
þ 12NH4 NO3 þ 8HNO3 :
2.2. Synthesis methodology
For a typical reaction, calcium and magnesium nitrate solutions were
prepared by dissolving appropriate amounts in 100 mL deionized water
such that (Ca + Mg):P molar ratio was always ~1.67. A control sample
(for structural comparison based on XRD and FTIR) labeled Ca–P was
prepared by the same synthesis method but without any magnesium.
Sample Ids and amounts of calcium nitrate and magnesium nitrate used
are given in Table 1. Amounts were selected based on the following for-
mula for magnesium substituted HA [Mg–HA, Ca10 − xMgx (PO4)6OH2].
Samples were denoted as follows: Ca–P corresponds to a sample without
any Mg (no magnesium nitrate added), 0.5Mg–CaP corresponds to
0.5 wt.% Mg substituted in hydroxyapatite (based on theoretical 100%
magnesium substitution into HA lattice). Similarly 2Mg–CaP corresponds
to a sample with 2 wt.% Mg substituted into HA. Therefore x and 10 − x
represent Mg:Ca molar ratio. The sample labeled as Mg–P corresponds to
a powder made using magnesium nitrate and no calcium nitrate.
0.3 M diammonium hydrogen phosphate solution was prepared by
dissolving 3.96 g diammonium hydrogen phosphate in 100 mL deion-
ized water (pH adjusted to 10 by ammonium hydroxide addition).
Diammonium hydrogen phosphate solution was added drop wise into
the calcium nitrate (if any) and magnesium nitrate (if any) solutions.
During addition, the pH of suspension (as instantaneous precipitation
took place) was maintained at pH 10 by ammonium hydroxide addition
and once complete, the suspension was stirred for 30 min. The suspen-
sion was put into a modified household microwave oven (600 W,
2.45 GHz) modified with a refluxing system. Microwave irradiation
was performed under ambient air for 5 min (15 s on and 15 s off)
[49]. The suspensions were observed to bubble at the base of the cooling
condenser (due to boiling by heating due to microwave irradiation).
After completion of exposure to microwave irradiation the suspension
was then filtered and washed with deionized water. The precipitate
was dried at 80 °C in a drying oven for 24 h and sieved (below
74 μm). The sieved powder was then heat-treated at 900 °C for 1 h.
2.3. Characterization
Powder X-ray diffraction (PXRD) data was collected on a X'Pert Pro
PW3050/60 diffractometer in the 20–80° range, based on Cu K-α radia-
tion [1.54060 Å], a step size of 0.02 and a scan time of 0.15 s per step.

Table 1
Sample IDs, corresponding x values, theoretical substitution weight percentages and amounts of reagents used.

Sample ID x (Mg) 10 − x (Ca) Mg:Ca molar ratio Mg wt % Added to 100 mL deionized water (grams)
(x:10 − x)
Calcium nitrate Magnesium nitrate

Concentration (M) Amount (grams) Concentration (M) Amount (grams)

CaP – 10 – – 0.500 11.800 – –

0.1Mg–CaP 0.04 9.96 0.004 0.1 0.497 11.700 0.002 0.054
0.5Mg–CaP 0.20 9.80 0.020 0.5 0.490 11.570 0.010 0.256
1Mg–CaP 0.40 9.60 0.042 1.0 0.480 11.330 0.020 0.513
1.5Mg–CaP 0.60 9.40 0.064 1.5 0.470 11.100 0.030 0.769
2Mg–CaP 0.80 9.20 0.087 2.0 0.460 10.860 0.040 1.026
4Mg–CaP 1.60 8.40 0.191 4.0 0.420 9.920 0.080 2.051
5Mg–CaP 2.00 8.00 0.250 5.0 0.400 9.447 0.100 2.564
10Mg–CaP 4.00 6.00 0.670 10.0 0.300 7.085 0.200 5.128
16Mg–CaP 6.00 4.00 1.500 16.0 0.200 4.724 0.300 7.692
22Mg–CaP 8.00 2.00 4.000 22.0 0.100 2.362 0.400 10.256
Mg–P 10.00 0.00 – 29.0 0.000 0.000 0.500 12.821
288 N.I. Khan et al. / Materials Science and Engineering C 56 (2015) 286–293

Table 2 3. Results and discussion

Magnesium substitution levels determined by area scans
using an EDX detector attached to a scanning electron
microscope.
3.1. Scanning electron microscopy

Sample ID Mg wt % age The chemical composition of as-precipitated samples was deter-

0.1Mg–CaP 0 mined using EDX and the results are shown in Table 2. The results indi-
0.5Mg–CaP 0.03 cate that the measured weight percentages of substituted magnesium
1Mg–CaP 0.97
were close to the theoretical amounts and followed an increasing trend.
1.5Mg–CaP 1.59
2Mg–CaP 1.64 Scanning electron microscopy was carried out to determine the size
4Mg–CaP 2.33 and morphology, degree of agglomeration and effect of heat-treatment.
5Mg–CaP 3.64 Fig. 1(a)–(d) shows the SEM images of heat-treated sample 1Mg–CaP
10Mg–CaP 7.3 at various magnifications. Particles were present in sintered agglomerat-
16Mg–CaP 11.86
22Mg–CaP 16.26
ed forms (and will hence be referred to as grains) with agglomerates as
Mg–P 22.83 large as ca. 90 μm as observed in Fig. 1(a) [with open porosity b 10 μm].
Fig. 1(b) and (c) revealed an average grain size of 618 nm [(±182 nm),
25 particles sampled]. Fig. 1(d) reveals these particles to be sintered po-
lygonal grains with distinct necking regions. This suggests that a certain
degree of agglomeration exists after precipitation; heat-treatment trans-
Fourier transform infra-red spectra were collected on a Thermo Nicolet forms these soft-agglomerates (statically bound particles) into harder
P6500 spectrometer using a photoacoustic cell purged with helium agglomerates (sintered particles). The literature suggests that indeed,
gas. 128 scans with 4 cm− 1 resolution were collected for each sam- granular (porous agglomerates) are a preferred form of calcium phos-
ple. Powder morphology and particle size was investigated using a phate ceramics for clinical use [57–59].
scanning electron microscope (JEOL JSM6490A, Japan). UTHSCSA A further increase in magnesium content from 1 wt.% (sample 1 Mg–
image tool was used to measure grain sizes from SEM images (grains CaP) to 5 wt.% (5Mg–CaP) resulted in formation of smaller agglomerates
with distinct grain boundaries were selected for sizing). Samples for with minimal porosity [compared to Fig. 1(a)] as seen in Fig. 2(a). The
SEM analysis were gold plated with coating thickness 250 Å using coat- SEM image in Fig. 2(b) revealed grains with an average size of 261 nm
ing unit JFC 1500 Ion Sputtering Device [JEOL ltd., Tokyo, Japan]. EDX [(±89 nm), 25 grains sampled]. Fig. 2(c)–(d) shows the SEM images
analysis was carried out using an EDX detector attached to a Hitachi for sample 10Mg–CaP revealing similar agglomeration and an average
scanning electron microscope SU-1500. BET Surface Area was deter- grain size of 256 nm [(± 58 nm), 25 grains sampled]. Samples 5Mg–
mined using N2 adsorption in a Micrometrics TriStar Surface Area Ana- CaP and 10Mg–CaP therefore have similar morphology and sizes. Ag-
lyzer. Samples were degassed at 100 °C under N2 purging for 2 h prior to glomerates of sample 10Mg–CaP were observed to be nanoporous
analysis. (pores b 0.5 μm).

Fig. 1. Scanning electron microscope images of heat treated (at 900 °C for 1 h) sample 1Mg–CaP at (a) ×1k magnification [bar = 10 μm] (b) at ×5k magnification [bar = 5 μm], (c) at ×10k
magnification [bar = 1 μm] and (d) at ×20k magnification [bar = 1 μm].
 N.I. Khan et al. / Materials Science and Engineering C 56 (2015) 286–293 289

Fig. 2. Scanning electron microscope images of heat treated Mg–CaP samples 5 Mg–CaP at (a) ×5k magnification [bar = 5 μm] and (b) ×50k magnification [bar = 0.5 μm], 10Mg–CaP at
(c) ×5k magnification [bar = 5 μm] and (d) ×50k magnification [bar = 0.5 μm] and 22Mg–CaP at (e) ×5k magnification [bar = 5 μm] and (f) ×50k magnification [bar = 0.5 μm].

A further increase in nominal magnesium content from 10 to 22 wt.%
(for sample 22Mg–CaP) resulted in a size dispersion of seemingly
non-porous agglomerates (1 μm to 10 μm) as seen in Fig. 2(e)–(f). The
average grain size was measured to be 144 nm [(±32), 25 grains
sampled].
3.2. BET surface area
Fig. 3 shows the BET surface area of as-precipitated and heat-treated
samples as a function of nominal magnesium weight percentages (in
the solution). The highest surface area (146 m2/g) for as-precipitated
samples was observed for sample 1.5Mg–CaP. Indeed, magnesium
has been reported to decrease the particle size of calcium phosphates
[29]. In our case this holds true for samples 0.1Mg–CaP, 0.5Mg–CaP,
1Mg–CaP and 1.5Mg–CaP only. Further increase in nominal magne-
sium content resulted in a decreasing surface area for as-precipitated
samples. As explained later, an increase beyond 2 wt.% magnesium re-
sulted in the disappearance of the broad peak characteristic of poorly
crystalline apatitic phases in XRD patterns for as-precipitated samples
[seen in Fig. 4(a)–(f) but not observed in Fig. 4(g)–(l)]. This could be
attributed to different phases with more amorphous nature being pre-
cipitated (hence the decrease in surface area). It is therefore suggested
that the decrease in agglomerate size and their dispersion [as seen in
Fig. 2(e)–(f)] is possibly due to reduced sinterability as a result of
lower surface area.
3.3. Powder X-ray diffraction
Powder X-ray diffraction (PXRD) data was collected for all samples
for assessing phase purity and crystallinity. Fig. 4 shows the PXRD pat-
terns of as-precipitated samples. PXRD patterns for samples Ca–P,
0.1Mg–CaP, 0.5Mg–CaP, 1Mg–CaP, 1.5Mg–CaP and 2Mg–CaP in
290 N.I. Khan et al. / Materials Science and Engineering C 56 (2015) 286–293
Fig. 3. BET surface area of as-precipitated and heat treated (at 900 °C for 1 h) samples as a
function of theoretical magnesium weight percentages (in solution).
Fig. 4(a)–(h) reveal the presence of an apatitic phase (loosely matching
ICDD pattern 09-432 for HA). The broadness and low intensity of the
peaks suggest small particle size and poor crystallinity. It was observed
that with increasing magnesium content PXRD patterns became even
more poorly resolved in [Fig. 4(g)–(l)]. The center of the broad peak
shifted from 31.9° to 30.11° suggesting formation of non-apatitic calci-
um phosphate phases.
Fig. 5 shows the PXRD patterns for samples heat treated at 900 °C
for 1 h. PXRD patterns corresponding to heat treated samples Ca–P,
0.1Mg–CaP and 0.5Mg–CaP in Fig. 5(a)–(c) respectively, gave a
good match to ICDD pattern 09-432 corresponding to phase-pure
crystalline HA. Inability to provide long maturation times (N 18 h)
often leads to non-stoichiometric apatites which decompose at or
above 750 °C into water, HA and TCP phases [60]. In this case however,
phase-pure hydroxyapatite was observed for magnesium-less (sample
Ca–P) and low Mg content samples. This therefore elucidates that the
microwave irradiation of suspensions greatly accelerates the matura-
tion process. Peaks observed were better resolved as compared to
PXRD patterns in Fig. 4(a)–(c) (for the same samples in their as-
precipitated forms) due to improved crystallinity as a result of heat-
Fig. 4. Powder X-ray diffraction patterns of as-precipitated samples (a) Ca–P (b) 0.1Mg–CaP
(c) 0.5Mg–CaP (d) 1.0Mg–CaP (e) 1.5Mg–CaP (f) 2.0Mg–CaP (g) 4Mg–CaP (h) 5Mg–CaP,
(i) 10Mg–CaP (g) 16Mg–CaP (k) 22Mg–CaP and (l) Mg–P.
Fig. 5. Powder X-ray diffraction patterns of heat treated (at 900 °C for 1 h) samples (a) Ca–P
(b) 0.1Mg–CaP (c) 0.5Mg–CaP (d) 1.0Mg–CaP (e) 1.5Mg–CaP (f) 2.0Mg–CaP (g) 4Mg–CaP
(h) 5Mg–CaP (i) 10Mg–CaP (j) 16Mg–CaP (k) 22Mg–CaP and (l) Mg–P (Note: HA = ,
Mg-βTCP = , Stanfieldite = , Farringtonite = ).
treatment at 900 °C for 1 h [49]. A further increase in nominal magne-
sium content to 1 wt.% (for heat-treated sample 1Mg–CaP) resulted
in appearance of a very small amount of a new phase as seen in the
PXRD pattern shown in Fig. 5(d) which was identified to be Mg-βTCP
(i.e. magnesium stabilized β-TCP phase) [compared to ICDD patterns
70-068 and 13-0404] [22,61–63]. Further increase in nominal mag-
nesium contents from 1.5 to 2 wt.% for samples 1.5Mg–CaP and
2Mg–CaP resulted in increase in intensity of the peak around 31.4° corre-
sponding to main peak for Mg-βTCP as seen in Fig. 5(e) and (f). Magne-
sium contents of 4 wt.% and 5 wt.% (corresponding to heat treated
samples 4Mg–CaP and 5Mg–CaP respectively) resulted in formation of
phase pure Mg–TCP as seen in PXRD patterns shown in Fig. 5(g) and
(h). For the aforementioned theoretical compositional range of magne-
sium substitution (i.e. from 0.1 wt.% to 5 wt.%) it is suggested that there
are two routes of formation of the Mg-βTCP phase. PXRD patterns for
as-precipitated Mg–CaP samples 0.1Mg–CaP, 0.5Mg–CaP, 1Mg–CaP,
1.5Mg–CaP and 2Mg–CaP (0.1 wt.% to 2 wt.%) in Fig. 4(b)–(f) revealed
an amorphous apatitic phase. It is suggested that the Mg-βTCP phase in
heat-treated forms of these samples [detected in PXRD patterns
Fig. 5(b)–(f)] evolves from decreased thermal stability (hence partial
decomposition into Mg-βTCP) of the magnesium containing apatitic
lattice (first route). The destabilizing effect of Mg is possibly due to lattice
strain introduced as a result of smaller ionic radius of Mg2+ (0.65 Å)
in comparison with Ca2+ (0.99 Å) which favors the hydroxyapatite–
Mg-βTCP transition upon heating [29,50]. Non-apatitic phases were de-
tected for as-precipitated sample 4Mg–CaP and other samples with
higher magnesium content [Fig. 4(g)–(l)]. It is suggested that this amor-
phous non-apatitic phase/s may contain Mg-βTCP in the as-precipitated
form which later crystallizes upon heat-treatment at 900 °C for 1 h
hence showing up as phase-pure crystalline Mg-βTCP in Fig. 5(g) and
(h) for heat treated samples 4Mg–CaP and 5Mg–CaP respectively (second
route) [64,65]. Further increase in nominal magnesium contents to
10 wt.% for sample 10Mg–CaP resulted in appearance of Mg-βTCP and
an additional phase identified as Stanfieldite corresponding to ICDD pat-
tern 23642 [Fig. 5(i)]. Indeed for higher level of magnesium substitutions
in the TCP phase wherein Mg2+ / (Mg2+ + Ca2+) molar ratio is greater
than 14% this phase has been reported (based on amounts added to solu-
tions this ratio is 40% in our case) [66]. Further increase in magnesium
content to 16 wt.% (heat treated sample 16Mg–CaP) resulted in forma-
tion of phase pure Stanfieldite as seen in Fig. 5(j). Sample 22Mg–CaP
(nominally 22 wt.% magnesium) was identified as a biphasic mixture of
Stanfieldite and Farringtonite [ICDD # 33-0876, (Mg3(PO4)2] as seen in
Fig. 5(k). Indeed a reaction carried out in the absence of calcium resulted
 N.I. Khan et al. / Materials Science and Engineering C 56 (2015) 286–293
in formation of phase pure Farringtonite for sample Mg–P as seen in
Fig. 5(l) [67].
3.4. Fourier transform infrared spectroscopy
FTIR spectra for as-precipitated Mg–CaP samples are shown in
Fig. 6(a)–(l). For samples Ca–P, 0.1Mg–CaP and 0.5Mg–CaP typical
apatitic FTIR spectra were observed as seen in Fig. 6(a–c). The broad
bands centered at ca. 3450 cm−1 (in the range of 3650–3000 cm− 1)
and 1650 cm−1 (1687–1600 cm−1) corresponded to the presence of
absorbed water in HA. The band in the range of 1510–1390 cm− 1
corresponded to the asymmetric (υ3) stretching of the C\\O bond in
carbonate. This is unsurprising as water used for preparation of
solutions was not degassed prior to reaction. The band observed in
the range of 1110–1040 cm− 1 corresponded to the asymmetric (υ3)
stretching of the P\\O bond of phosphate group in HA. The peak at
960 cm−1 was attributed to the symmetric stretching (υ1) of the P\\O
bond of phosphate group in HA. Peak at 874 cm−1 was due to the bending
mode (υ2) of the O\\C\\O linkage in carbonate groups present in HA.
The position of the C\\O and O\\C\\O bands and peaks (υ2 and υ3)
indicates that B-type carbonate substitution into HA took place.
The peak doublet present in the range of 600–560 cm−1 was attributed
to the bending mode (v4) of the O\\P\\O linkage in phosphate group
of HA. No change was observed in Fig. 6(c)–(f) with increasing nom-
inal magnesium content from 0.5 to 2 wt.% (for samples 0.5Mg–CaP,
1Mg–CaP 1.5Mg–CaP and 2Mg–CaP). This suggested that for theoret-
ical magnesium substitution levels of up to 2 wt.% the phases were
apatitic (supported by PXRD patterns in Fig. 4(c)–(f)]. Further in-
crease in nominal magnesium content up to 4 wt.% (4Mg–CaP)
caused a broadening of the band initially observed in the range of
1110–1040 cm− 1 (corresponding to v3 asymmetric P\\O stretching)
and in disappearance of the peak at 961 cm− 1 corresponding to v1
stretching of the P\\O bond [see Fig. 4(g) for sample 4Mg–CaP].
The peaks of v4 bending mode of the O\\P\\O linkage were observed as a
doublet (at 609 and 565 cm−1) for sample 2Mg–CaP in Fig. 6(f) however
this doublet was not observed in Fig. 6(g) corresponding to sample
4Mg–CaP (a band was observed instead). The aforementioned broaden-
ing, disappearance of peaks (non-resolution) and changing of a doublet
peak into a band can possibly be due to similar vibrations of the P\\O
bond and/or O\\P\\O linkages, but from phosphate groups from a dif-
ferent (non-apatitic) calcium phosphate phase. Indeed a non-apatitic
phase was observed in Fig. 4(g) which crystallized into phase pure
Mg-βTCP after heat treatment at 900 °C as seen in Fig. 5(g). Further
Fig. 6. Fourier transform infrared spectra in the range of 4000–400 cm−1 for as precipitated
samples (a) Ca–P (b) 0.1Mg–CaP (c) 0.5Mg–CaP (d) 1.0Mg–CaP (e) 1.5Mg–CaP
(f) 2.0Mg–CaP (g) 4Mg–CaP (h) 5Mg–CaP (i) 10Mg–CaP (j) 16Mg–CaP (k) 22Mg–CaP
and (l) Mg–P.
291
increase in nominal magnesium content from 5 wt.% to 22 wt.% (for
samples 5Mg–CaP, 10Mg–CaP, 16Mg–CaP and 22Mg–CaP) resulted in
no discernible change in the FTIR spectra as seen in Fig. 6(h)–(k). The in-
creased water retention as observed in Fig. 6(h) for sample 5Mg–CaP may
possibly be due to very amorphous nature of the as-precipitated phase.
Indeed, magnesium has been shown to reduce crystallinity of calcium
phosphate phases [5]. The FTIR spectrum for sample MgP is shown in
Fig. 6(l). Most of its features are very similar to Fig. 6(k) for sample
22Mg–CaP [due to the presence of the Farringtonite phase as shown in
PXRD patter in Fig. 4(k)]. The doublet peak in the range of 650 cm−1–
550 cm−1 reappeared corresponding to bending vibration of the P\\O
bond in O\\P\\O linkages of phosphate group in the Farringtonite
phase (sample MgP).
Fig. 7 shows FTIR spectra of samples heat treated at 900 °C for 1 h. In
Fig. 7(a–b) for samples Ca–P and 0.1Mg–CaP the sharp peak at ca.
3567 cm−1 was assigned to the stretching mode (υs) of the hydroxyl
group of HA that indicated the crystalline nature of HA (heat-treatment
resulted in increase in crystallinity). The band in the range of 2360–
1990 cm−1 was attributed to the overtones and combination of v3 and
v1 phosphate modes [68]. The band observed in the range of
1510–1380 cm−1 was attributed to the asymmetric stretching (υ3) of
the C\\O bond of carbonate group in HA. The peaks at 1091 and
1034 cm−1 were ascribed to the asymmetric stretching (υ3) of P\\O
bond of phosphate group in HA. The peak at 960 cm−1 was assigned
to the symmetric stretching (υ3) of the P\\O bond in phosphate
group. Peak at 873 cm−1 was attributed to the bending mode (υ2) of
the O\\C\\O linkage in carbonate groups. The presence of carbonate
groups in these samples is unsurprising because water was not
degassed prior to use and all experimental steps were carried out in
air. These bands have a lower intensity probably due to CO2 loss upon
heating as compared to those observed in Fig. 6 for as-precipitated sam-
ples. The peak at 632 cm−1 corresponded to the librational mode (υL) of
the hydroxyl group. This peak was not observed in FTIR spectra for HA in
Fig. 6 (for as-precipitated samples). Indeed, this peak has been used as
an indicator of crystallinity [69]. Peaks at 606 and 582 cm− 1
corresponded to bending mode (υ4) of the O\\P\\O linkage in phos-
phate groups of HA. The peak at 470 cm−1 was ascribed to the O\\P\\O
bending mode (υ2). The sharpness of peaks at 632, 606 and 582 cm−1
indicated a well crystallized HA. No change was observed in
Fig. 7(b)–(d) with increasing magnesium content from 0.1 to 1 wt.%
(for samples 0.1Mg–CaP, 0.5Mg–CaP and 1Mg–CaP). This indicates
that magnesium substitution (theoretically calculated) up to 1 wt.%
Fig. 7. Fourier transform infrared spectra in the range of 4000–400 cm−1 for heat treated
(at 900 °C for 1 h) samples (a) Ca–P (b) 0.1Mg–CaP (c) 0.5Mg–CaP (d) 1.0Mg–CaP
(e) 1.5Mg–CaP (f) 2.0Mg–CaP (g) 4Mg–CaP (h) 5Mg–CaP (i) 10Mg–CaP (j) 16Mg–CaP
(k) 22Mg–CaP and (l) Mg–P.
292 N.I. Khan et al. / Materials Science and Engineering C 56 (2015) 286–293
(measured to be 0.97 wt.%, see Table 2) did not induce any significant
change in the apatite structure. Indeed only a small amount of
Mg-βTCP phase was observed in the PXRD pattern shown in
Fig. 5(d) for sample 1Mg–CaP (major phase being HA).
A further increase in nominal magnesium content to 1.5 wt.% (for
sample 1.5Mg–CaP) resulted in the appearance of new peaks at 989
and 945 cm−1 attributed to the phosphate groups in Mg-βTCP phase
[also confirmed in Fig. 5(e)]. The reduction in intensities of the O\\H
peak at 3569 cm−1 and υ3 C\\O (1510–1380 cm−1) band and widening
of the band associated with asymmetric υ3 P\\O stretching (and
appearance of new peaks in this range) from Fig. 7(b)–(f) were due to
successive reduction in amount of HA (and increase in amount of
Mg-βTCP) till phase-pure Mg-βTCP was obtained. It was observed
that O\\H peaks at 3569 and 634 cm− 1 were not present in
Fig. 7(g) for sample 4Mg–CaP (as it is phase-pure Mg-βTCP). In the
FTIR spectra shown in Fig. 7(g)–(h) for samples 4Mg–CaP and 5Mg–
CaP (identified earlier as phase-pure Mg-βTCP) the peaks correspond-
ing to O\\P\\O υ2 bending at 600 and 552 cm− 1 were observed as a
doublet. This doublet was not observed in Fig. 7(i)–(j) [for samples
10Mg–CaP and 16Mg–CaP] possibly due to the bending vibrations of
the O\\P\\O linkages in phosphate groups in Stanfieldite phase. The
reappearance of the doublet in Fig. 7(k)–(l) for samples 22Mg–CaP
and MgP can be attributed to the formation of Farringtonite phase.
These FTIR results support the PXRD data presented in Fig. 5 suggesting
that increased magnesium incorporation in HA leads to its phase de-
composition and formation of non-apatitic phases at higher magnesium
contents.
4. Conclusions
This work demonstrates that microwave irradiation of pre-
precipitated suspensions results in magnesium substituted phase-pure
HA (up to 0.5 wt.% Mg content) which is thermally stable up to
900 °C. In using this route, long synthesis durations required were
avoided (N18 h maturation time needed to obtain phase pure thermally
stable HA under standard co-precipitation route); only 2 h was required
to obtain a precipitate that gave phase-pure Mg–HA. It was observed
that a magnesium content of above 0.5 wt.% (i.e. at 1 wt.% and higher)
yielded non-apatitic magnesium containing calcium phosphates as
well. This study constitutes the first study of its kind which focuses on
such a large spectrum of substituting compositions and fully evaluates
microwave assisted synthesis (especially from a structural standpoint)
as a robust tool for tailorable synthesis of bioceramics with enhanced
potential bioactivity. Endeavors which focus on more ionic substitutions
(such as zinc and iron) for a range of biological applications will be
reported in due course of time.
Acknowledgments
Ministry of Science and Technology, Government of Pakistan is ac-
knowledged for its developmental funding (IRCBM PC-1). The Higher
Education Commission Pakistan is acknowledged for research (NRPU
AAC - 1834) and developmental (DIASF — CIIT Lahore - 0912252K047)
grants. School of Chemical and Materials Engineering (SCME), National
University of Science and Technology (NUST) is acknowledged for its
help regarding SEM imaging. CIIT Lahore is thanked for visiting professor-
ships (IR and JAD). Dr. Umair Manzoor is thanked for his help with EDX
analysis. Dr. Zenab Sarfaraz and Ayesha Idrees are thanked for their help
with BET surface area analysis.
References
[1] R.H. Doremus, J. Mater. Sci. 27 (1992) 285.
[2] L.L. Hench, J. Am. Ceram. Soc. 74 (1991) 1487.
[3] M.J. Olszta, X. Cheng, S.S. Jee, R. Kumar, Y.Y. Kim, M.J. Kaufman, E.P. Douglas, L.B.
Gower, Mater. Sci. Eng. R. Rep. 58 (2007) 77.
[4] E.B. Nery, K.L. Lynch, W.M. Hirthe, K.H. Mueller, J. Periodontol. 46 (1975) 328.
[5] A. Bigi, G. Falini, E. Foresti, M. Gazzano, A. Ripamonti, N. Roveri, J. Inorg. Biochem. 49
(1993) 69.
[6] A. Bigi, G. Cojazzi, S. Panzavolta, A. Ripamonti, N. Roveri, M. Romanello, K.N. Suarez,
L. Moro, J. Inorg. Biochem. 68 (1997) 45.
[7] N. Rangavittal, A.R. Landa-Canovas, J.M. Gonzalez-Calbet, M. Vallet-Regi, J. Biomed.
Mater. Res. 51 (2000) 660.
[8] E. Boanini, M. Gazzano, A. Bigi, Acta Biomater. 6 (2010) 1882.
[9] R.N. Correia, M.C.F. Magalhaes, P.A.A.P. Marques, A.M.R. Senos, J. Mater. Sci. Mater.
Med. 7 (1996) 501.
[10] R. Zapanta LeGeros, Prog. Cryst. Growth Charact. Mater. 4 (1981) 1.
[11] G. Daculsi, R.Z. LeGeros, D. Mitre, Calcif. Tissue Int. 45 (1989) 95.
[12] F.C.M. Driessens, Ber. Bunsenges. Phys. Chem. 90 (1986) 760.
[13] S. Kannan, A.F. Lemos, J.M.F. Ferreira, Chem. Mater. 18 (2006) 2181.
[14] L.A. Bretado, D.A. Cortes, W. Ortega, D. Renteria, Adv. Appl. Ceram. 108 (2009) 194.
[15] I.R. Gibson, W. Bonfield, J. Mater. Sci. Mater. Med. 13 (2002) 685.
[16] R.K. Rude, H.E. Gruber, H.J. Norton, L.Y. Wei, A. Frausto, J. Kilburn, Bone 37 (2005)
211.
[17] C.M. Serre, M. Papillard, P. Chavassieux, J.C. Voegel, G. Boivin, J. Biomed. Mater. Res.
42 (1998) 626.
[18] D.A. Straub, Nutr. Clin. Pract. 22 (2007) 286.
[19] Y.H. Yun, Z.Y. Dong, Z.Q. Tan, M.J. Schulz, Anal. Bioanal. Chem. 396 (2010) 3009.
[20] H. Aydin, O. Deyneli, D. Yavuz, H. Gozu, N. Mutlu, I. Kaygusuz, S. Akalin, Biol. Trace
Elem. Res. 133 (2010) 136.
[21] G. Qi, S. Zhang, K.A. Khor, S.W. Lye, X. Zeng, W. Weng, C. Liu, S.S. Venkatraman, L.L.
Ma, Appl. Surf. Sci. 255 (2008) 304.
[22] S. Kannan, I.A.F. Lemos, J.H.G. Rocha, J.M.F. Ferreira, J. Solid State Chem. 178 (2005)
3190.
[23] B. Tomazic, M. Tomson, G.H. Nancollas, Arch. Oral Biol. 20 (1975) 803.
[24] E. Odabasi, M. Turan, A. Aydin, C. Akay, M. Kutlu, Ann. Acad. Med. Singap. 37 (2008)
564.
[25] R.K. Rude, H.E. Gruber, H.J. Norton, L.Y. Wei, A. Frausto, J. Kilburn, Osteoporos. Int. 17
(2006) 1022.
[26] N. Kanzaki, K. Onuma, G. Treboux, S. Tsutsumi, A. Ito, J. Phys. Chem. B 104 (2000)
4189.
[27] A. Yasukawa, T. Yokoyama, K. Kandori, T. Ishikawa, Colloids Surf. A Physicochem.
Eng. Asp. 238 (2004) 133.
[28] R. Enderle, F. Gotz-Neunhoeffer, M. Gobbels, F.A. Muller, P. Greil, Biomaterials 26
(2005) 3379.
[29] A.A. Chaudhry, J. Goodall, M. Vickers, J.K. Cockcroft, I. Rehman, J.C. Knowles, J.A. Darr,
J. Mater. Chem. 18 (2008) 5900.
[30] G. Grandi, C. Heitz, L.A.D. Santos, M.L. Silva, M. Sant'Ana Filho, R.M. Pagnocelli, D.N.
Silva, Mater. Res. 14 (2011) 11.
[31] M. Tomozawa, S. Hiromoto, Y. Harada, Surf. Coat. Technol. 6 (7) (2015) 2787–2796.
[32] F. Ren, Y. Leng, R. Xin, X. Ge, Acta Biomater. 6 (7) (2015) 2365–2862.
[33] E. Landi, G. Logroscino, L. Proietti, A. Tampieri, M. Sandri, S. Sprio, J. Mater. Sci. Mater.
Med. 19 (2008) 239.
[34] S.V. Chiranjeevirao, J. Hemmerle, J.C. Voegel, R.M. Frank, Inorg. Chim. Acta 67 (1982)
183.
[35] I. Cacciotti, A. Bianco, M. Lombardi, L. Montanaro, J. Eur. Ceram. Soc. 29 (2009) 2969.
[36] E. Landi, A. Tampieri, M. Mattioli-Belmonte, G. Celotti, M. Sandri, A. Gigante, P. Fava,
G. Biagini, J. Eur. Ceram. Soc. 26 (2006) 2593.
[37] Y. Pan, J.L. Huang, C.Y. Shao, J. Mater. Sci. 38 (2003) 1049.
[38] L.J. Peng, C.Q. Ning, D.Y. Ding, X.Y. Liu, J. Biomater. Tissue Eng. 5 (162) (2015).
[39] S. Haque, I. Rehman, J.A. Darr, Langmuir 23 (2007) 6671.
[40] M.J. Phillips, J.A. Darr, Z.B. Luklinska, I. Rehman, J. Mater. Sci. Mater. Med. 14 (2003)
875.
[41] A.R. Toibah, I. Sopyan, M. Hamdi, S. Ramesh, 4th Kuala Lumpur International Confer-
ence on Biomedical Engineering 2008 IFMBE Proceedings 21 (2008) 314–317.
[42] H.G. Zhang, Q.S. Zhu, Z.H. Xie, Mater. Res. Bull. 40 (2005) 1326.
[43] W.L. Suchanek, K. Byrappa, P. Shuk, R.E. Riman, V.F. Janas, K.S. TenHuisen, J. Solid
State Chem. 177 (2004) 793.
[44] A. Siddharthan, T.S.S. Kumar, S.K. Seshadri, Biomed. Mater. (2009) 4.
[45] N. Rameshbabu, T.S.S. Kumar, K.P. Rao, Mater. Res. Innov. 14 (2010) 45.
[46] A. Lak, M. Mazloumi, M.S. Mohajerani, S. Zanganeh, M.R. Shayegh, A. Kajbafvala, H.
Arami, S.K. Sadrnezhaad, J. Am. Ceram. Soc. 91 (2008) 3580.
[47] S.J. Kalita, S. Verma, Mater. Sci. Eng. C 30 (2010) 295.
[48] M. Ipekoglu, S. Altintas, Biyomut: 2009 14Th National Biomedical Engineering Meeting
2009, p. 439.
[49] H. Arami, M. Mohajerani, M. Mazloumi, R. Khalifehzadeh, A. Lak, S.K. Sadrnezhaad,
J. Alloys Compd. 469 (2009) 391.
[50] I.V. Fadeev, L.I. Shvorneva, S.M. Barinov, V.P. Orlovskii, Inorg. Mater. 39 (2003) 947.
[51] A. Marti, Inj. Int. J. Care Inj. 31 (2000) S33–S36.
[52] A.A. Chaudhry, H. Yan, G. Viola, M.J. Reece, J.C. Knowles, K. Gong, I. Rehman, J.A. Darr,
J. Biomater. Appl. 27 (2012) 79.
[53] A.A. Chaudhry, J.C. Knowles, I. Rehman, J.A. Darr, J. Biomater. Appl. 28 (2013) 448.
[54] A.A. Chaudhry, H. Yan, K. Gong, F. Inam, G. Viola, M.J. Reece, J.B.M. Goodall, I. ur
Rehman, F.K. McNeil-Watson, J.C.W. Corbett, J.C. Knowles, J.A. Darr, Acta Biomater.
7 (2011) 791.
[55] R. Nazir, N. Iqbal, A. Samad Khan, A. Akram, A. Asif, A. Anwar Chaudhry, I. Rehman, R.
Hussain, Ceram. Int. 38 (2012) 457.
[56] N. Iqbal, A. Asif, A.A. Chaudhry, R. Nazir, R. Amin, M. Akram, G.Y. Fan, A. Akram, S.H.
Park, R. Hussain, Curr. Appl. Phys. 12 (3) (2011) 755–759.
[57] N.M.D.F. Costa, B.R. Melo, R.T. Brito, G.V.D.O. Fernandes, V.G. Bernardo, E.C. Fonseca,
M.B. Conz, G.R.A. Soares, J.M. Granjeiro, Mater. Res. 12 (2009) 245.
[58] A. Hafiz, K.A. Khalid, A. Yusof, M.A. Azril, A. Shukrimi, M.Y. Nazri, C.A. Aminudin, Z.
Zamzuri, F. Fazan, Eur. Cell. Mater.s 16 (2008) 55.
 N.I. Khan et al. / Materials Science and Engineering C 56 (2015) 286–293 293

[59] F.D. Burstein, S.R. Cohen, R. Hudgins, W. Boydston, Plast. Reconstr. Surg. 100 (1996)
869.
[60] K. Ishikawa, P. Ducheyne, S. Radin, J. Mater. Sci. Mater. Med. 4 (1993) 165.
[61] S. Kannan, J.M. Ventura, J.M.F. Ferreira, Ceram. Int. 33 (4) (2015) 637–641.
[62] S. Kannan, A.F. Lemos, J.H.G. Rocha, J.M.F. Ferreira, J. Am. Ceram. Soc. 89 (2006) 2757.
[63] S. Kannan, J.H.G. Rocha, J.M.F. Ferreira, J. Mater. Chem. 16 (2006) 286.
[64] S.L. Rowles, Bull. Soc. Chim. Fr. (1968) 1797.
[65] R.Z. LeGeros, A.M. Gatti, R. Kijkowska, D.Q. Mijares, J.P. LeGeros, Mg-Substituted
Tricalcium Phosphates: Formation and Properties, 2004.
[66] D. Clement, J.M. Tristan, M. Hamad, P. Roux, J.C. Heughebaert, J. Solid State Chem. 78
(1989) 271.
[67] U. Klammert, E. Vorndran, T. Reuther, F.A. Muller, K. Zorn, U. Gbureck, J. Mater. Sci.
Mater. Med. 21 (2010) 2947.
[68] S. Koutsopoulos, J. Biomed. Mater. Res. 62 (2002) 600.
[69] A.A. Chaudhry, S. Haque, S. Kellici, P. Boldrin, I. Rehman, A.K. Fazal, J.A. Darr, Chem.
Commun. (2006) 2286.