Transformation of Brushite (CaHPO4·2H2O) to Whitlockite (Ca9Mg(HPO4)(PO4)6) or Other CaPs in Physiologically Relevant Solutions

Article abstract of DOI:10.1111/jace.14069

Brushite (dicalcium phosphate dihydrate, DCPD, CaHPO₄·2H₂O) and whitlockite [WH, Ca₉Mg(HPO₄)(PO₄)₆] are usually found in the mammalian metabolism in the form of diverse pathological calcifi

Full text of DOI:10.1111/jace.14069

J. Am. Ceram. Soc., 99 [4] 1200–1206 (2016)
DOI: 10.1111/jace.14069
© 2016 The American Ceramic Society
ournal
J
Transformation of Brushite (CaHPO4Á2H2O) to Whitlockite
(Ca9Mg(HPO4)(PO4)6) or Other CaPs in Physiologically Relevant Solutions
A. Cuneyt Tas†,*
Department of Materials Science and Engineering, University of Illinois, Urbana, Illinois 61802
Brushite
(dicalcium
phosphate
dihydrate,
DCPD,
formed in the synthesis solutions (e.g., 1 h residence time in
solutions) exhibited the crystal structure and characteristic
X-ray diffraction pattern of ICDD PDF 9-0077, but these
crystals had a Ca/P molar ratio of only 0.8 in opposition to
the stoichiometric ratio of 1.0.5 Crystals soaked in the same
synthesis solution for 24 h reached a Ca/P molar ratio of
0.97.5 On the other hand, the in vivo and in vitro maturation
of apatitic calcium phosphate (Ap-CaP) was studied by
Cazalbou et al.6
CaHPO4Á2H2O) and whitlockite [WH, Ca9Mg(HPO4)(PO4)6]
are usually found in the mammalian metabolism in the form of
diverse pathological calcifications, dental calculi, urinary tract
stones, salivary gland deposits, cardiovascular or pulmonary
calcified deposits, and even as prostate or cartilage calcifica-
tions. The hydrothermal transformation of synthetic brushite
crystals into single-phase whitlockite, octacalcium phosphate,
or apatitic calcium phosphate was observed over the time per-
iod of 1 to 21 d and at 37°C, 70°C, and 115°C in nonstirred
physiologically relevant solutions developed for this work. The
Brushite is found in the human metabolism in the form of
pathological calcifications; such as, dental calculi (or tartar),7
urinary stones 8, and in chondrocalcinosis (together with
strong influence of the physiologically relevant ions such as
on hydrothermal transformations is
Mg2+ and HCO3
Ca2P2O7Á2H2O (CPPD) crystals ). Brushite has a high solu-
–
9
exposed. The formation of the nanoglobules and nanofibrils of
X-ray amorphous calcium phosphate or Mg-doped calcium
phosphate on the surfaces of brushite crystals are observed for
the first time in biomimetic solutions containing 10 mM Mg2+
and/or 27 mM HCO3 – . The experimental conditions leading to
the formation of such nanofibrils on brushite crystal surfaces
are also found to stop the further transformation of brushite
into any other calcium phosphate (CaP) phases even at high
solution temperatures. Samples were characterized by scanning
electron microscopy and powder X-ray diffraction.
bility (pKSP of 6.59 at 25°C) in comparison to stoichiometric
hydroxyapatite (pKSP of 116.8 at 25°C).10 The solubility of
brushite is also significantly higher than that of octacalcium
phosphate, OCP, Ca8(HPO4)2(PO4)4Á5H2O (with a pKSP of
96.6 at 25°C).10 Tang et al.11 experimentally measured the
dissolution rate of brushite in water at RT as
4.26 9 10À4 mol/m2/min. High solubility of brushite, in com-
parison to hydroxyapatite, led to the development of inject-
able orthopedic or dental paste formulations based on
brushite12–14 with b-TCP (b-tricalcium phosphate, b-
Ca3(PO4)2) and MCPM [monocalcium phosphate monohy-
drate, Ca(H2PO4)2ÁH2O] as the starting materials. Apelt
et al.15 reported in a comparative in vivo study that such
nonapatitic cements, with microstructures comprised of
coarse b-TCP granules embedded in a brushite matrix, were
rapidly biodegraded by macrophage activity and showed fas-
ter new bone formation compared to commercially available
hydroxyapatite cements.
I. Introduction
RUSHITE (DCPD, dicalcium phosphate dihydrate, CaH-
B
PO4Á2H2O), named after the mineralogist George Jarvis
Brush (1831–1912), is the predominant mildly acidic phase of
the Ca(OH)2–H3PO4–H2O system to precipitate between pH
2 and 6.5,1–3 when Ca2+ and HPO4 ions are simultane-
2À
ously present in an aqueous solution of the stated pH range.
When Ca-chloride (or Ca-nitrate or Ca-acetate) solutions are
mixed with those of basic Na-(or K- or ammonium-) phos-
phate at room temperature (RT, 22°C Æ 1°C), monoclinic
(ICDD PDF 9-0077) brushite crystals instantly form. The
crystal structure of brushite was ascertained by Curry and
Jones4 via neutron diffraction.
Brushite, as a mildly acidic CaP, is known to be not stable
in aqueous solutions of neutral pH and rapidly transforms to
apatitic calcium phosphate (Ap-CaP) in basic media contain-
ing dissolved NaOH, KOH, or NH4OH.16 Similarly, brushite
slowly transforms into carbonated Ap-CaP when soaked in
Tris [tris(hydroxymethyl)aminomethane, (HOCH2)3CNH2]-
and into
17
buffered SBF (synthetic body fluid) solutions
Brushite synthesis, if performed in the way outlined above,
does not require any strict pH control in stark contrast to
stoichiometric Ca-hydroxyapatite [HA, Ca10(PO4)6(OH)2 or
3Ca3(PO4)2ÁCa(OH)2] synthesis since the latter necessitates
the solution pH to be maintained above 12 at all times. For
instance, if the pH during stoichiometric Ca-hydroxyapatite
synthesis falls below 12, then Ca- and OH-deficient, apatite-
like (apatitic) calcium phosphates would precipitate. Precipi-
tated brushite crystals were recently shown to go through a
maturation process5 in their mother liquors. The maturation
of brushite revealed itself as follows. The very first crystals
OCP when immersed in Hepes [4-(2-hydroxyethyl)-1-piperazi-
neethanesulfonic acid, C8H18N2O4S]-buffered DMEM solu-
tion (Dulbecco’s Modified Eagle Medium) at the human
body temperature of 37°C and at the blood pH of 7.4 for
about 48 h.18 SBF and DMEM solutions contain Mg2+ ions
at the concentrations of 1.517 and 0.814 mM,18 respectively.
High solubility brushite, therefore, has the unique ability of
acting (in vitro) like a precursor to calcium phosphates of
lower solubility, such as HA or OCP, especially in solutions
mimicking the inorganic ion concentrations of blood plasma.
Brushite can be deposited on titanium19,20 or Mg-alloy21 sub-
strates at ambient temperatures in aqueous solutions.
Mechanically stable millimeter-sized brushite granules devel-
oped for use in orthopedic and oral surgery were also shown
to be suitable for high throughput production.22 Therefore,
the close monitoring of the hydrothermal transformations of
brushite, into other benign CaP phases, in physiologically rel-
J. McKittrick—contributing editor
evant abiotic solutions (containing Mg2+
,
Na+, ClÀ,
Manuscript No. 37068. Received June 18, 2015; approved November 10, 2015.
*Member, The American Ceramic Society
2À
2À
HCO3À, HPO4 and SO4 ions at concentrations equal to
†Author to whom correspondence should be addressed. e-mail: c_tas@hotmail.com
1200

April 2016
Brushite to Whitlockite
1201
those of the human blood) would be useful to future devel-
opment work on brushite cements, brushite coatings, and
brushite granules.
II. Experimental Procedure
The synthesis of brushite powders consisted of preparing two
solutions, as described elsewhere.18, 45 Solution A was pre-
pared as follows: 1.650 g of KH2PO4 (≥99.9%, Cat. No:
1.04873; Merck KGaA, Darmstadt, Germany) was dissolved
in 1400 mL of deionized water, in a 2L-capacity glass beaker,
followed by the addition of 6.026 g of Na2HPO4 (≥99.9%,
Cat. No: 1.06586; Merck), which resulted in a clear solution
of pH (Orion Star 215 pH-meter; Thermo Scientific,
Dreieich, Germany) 7.5 at RT. The pH meter was calibrated
by using solutions having standard pH values of 4.01 and 7
(Solutions PHA-4-CAL-PH and PHA-7-CAL-PH; Omega
Engineering Ltd., Manchester, UK).
Solution B (of pH 6.4) was prepared by dissolving 8.028 g
of CaCl2Á2H2O in (≥99.9%, Cat. No: 1.02382; Merck)
400 mL of deionized water. Solution B was then rapidly
added to solution A under stirring and the crystals formed
were aged in solution for 24 h at RT, while stirring at
500 rpm (final solution pH 5.3). Solids recovered by filtration
(and washing) from their mother liquors were dried at 35°C
(24 h) to obtain approximately 7 g of flat plate-shaped,18
optically transparent brushite crystals. Only these crystals
were used in the hydrothermal transformation experiments.
The hydrothermal reaction solutions were prepared, as
detailed in Table II of the next chapter, by dissolving appro-
priate amounts of NaCl (>99%, Cat. No: 106400, Merck),
MgCl2Á6H2O (>99%, Cat. No: 105833, Merck), KCl (>99%,
Cat. No: PX-1405, Merck), NaHCO3 (>99%, Cat. No: SX-
0320, Merck), Na2SO4 (>99%, Cat. No: SX-0761, Merck), or
NaH2PO4ÁH2O (>99%, Cat. No: 106349, Merck) in preboiled
(an essential step to eliminate dissolved bicarbonate ions in
water) deionized water. All of the hydrothermal transforma-
tion solutions were optically transparent and were free of
any turbidity prior to the reaction(s).
Brushite crystals (0.35 g) were placed in 100 mL aliquots
of the specific hydrothermal reaction solutions, and the solu-
tions were contained in 100 mL-capacity sterile glass media
bottles having plastic screw caps with O-rings. The brushite
suspensions were first ultrasonicated for 1 min by dipping
the sealed glass media bottles in an ordinary ultrasonic clean-
ing bath. The bottles were then placed in a microprocessor-
controlled oven for durations (and temperatures) specified in
Table II. The bottle contents were not stirred or agitated
during the course of hydrothermal transformation reactions.
Biological mineralization processes do never employ stir bars,
stir rods, stir plates, etc. The solids were finally separated
from the solutions by filtration (and washing with deionized
water), followed by drying at 35°C (24 h). All experiments
were repeated three times and the XRD and SEM analyses
were also repeated on such samples of reproduction.
The particle/crystal morphology analyses of the samples
were performed by scanning electron microscopy (SEM;
Neon 40-EsB, Zeiss, Jena, Germany). The samples for the
SEM analyses were sputter coated with a 5 nm-thick layer of
Au–Pd alloy prior to imaging at 5 kV. The samples for
energy-dispersive X-ray spectroscopy (EDXS) analyses were
not sputter coated. Samples for X-ray diffraction (XRD; D8
Advance, Bruker, Karlsruhe, Germany) runs were first
ground in an agate mortar with an agate pestle. All XRD
scans (k = 0.15406 nm) were performed at 40 kV and 40 mA
Rowles23 reported that brushite powders soaked in dis-
tilled water containing only 1 mM Mg2+, at 37°C for 24 d,
led to the formation of whitlockite [WH, Ca9Mg(HPO4)
(PO4)6 or 3Ca3(PO4)2ÁMgHPO4] at the expense of brushite.
The natural mineral whitlockite is named after Herbert Percy
Whitlock (1868-1948) and the trigonal (rhombic symmetry)
crystal structure (ICDD PDF 04-09-3397) of whitlockite is
similar to that of b-TCP (ICDD PDF 55-0898). The powder
X-ray diffraction spectra of WH and b-TCP look somewhat
similar to one another. Nevertheless, WH and b-TCP are not
isostructural due to the difference between their space
groups, as shown in Table I. This structural similarity caused
some researchers to inadvertently name Mg- and HPO4-free
b-TCP [i.e., b-Ca3(PO4)2] as whitlockite.24–30 Although the
crystallographic unit cell of WH was recently compared to
that of b-TCP by Jang et al.,31 a more comprehensive analy-
sis of the crystal structure of Mg-containing WH single crys-
tals was provided earlier by Gopal et al.32 The crystal
structure of Mg-free b-TCP, on the other hand, has been
described by Dickens et al.33
Whitlockite [Ca9Mg(HPO4)(PO4)6] was reported to be
present in human tooth enamel,34 in the knees of arthritic
35
patients
and in the calcifications of the articular carti-
lage,36 in the human supragingival calculus7 and dental pla-
que,37 as pathological calcification spots on aorta,38, 39 in the
pulmonary calcified deposits of tuberculosis patients,40 in
human prostate calculi,41 in urinary tract stones,42, 43 and
even in the stones of the parotid (salivary) glands.44
Although brushite and whitlockite both are phases of diverse
pathological calcifications in the human metabolism, the lit-
erature on the in vitro inorganic transformation of synthetic
brushite powders into whitlockite in Mg- and HCO3-contain-
ing saline solutions remained scarce, and this provided the
impetus for this study.
This study is designed to examine, for the first time, the
hydrothermal transformation of synthetic brushite (the start-
ing phase) to whitlockite in solutions, free of any organics,
such as Tris or Hepes, mimicking the inorganic ion concen-
trations of physiological fluids. Throughout this study, the
temperature of hydrothermal reaction solutions were varied
between 37°C and 115°C, whereas the same solutions con-
tained Mg2+ (from 1.5 to 10 mM) and Na+, K+, ClÀ,
2À
SO42À, HCO3À, or HPO4 ions in physiologically relevant
concentrations. The simple inorganic solutions and experi-
mental conditions reported here might be of interest to read-
ers who wish to further study
1. The transformation of brushite into whitlockite (WH),
or
2. The transformation of brushite into OCP, or
3. The transformation of brushite into apatitic CaP (Ap-
CaP), or
4. The transformation of brushite into whitlockite-apati-
tic CaP or whitlockite-OCP biphasic mixtures, or
5. The inhibition of the transformation of brushite to
monetite (DCPA, CaHPO4) or to any other calcium
phosphates in aqueous or alcoholic solutions heated
above 37°C.
Table I. Structural Similarity Between b-TCP and Whitlockite (WH)
Compound
Crystal structure
Lattice parameters
Space group
ICDD-PDF†
b-Ca3(PO4)2 (b-TCP)
Trigonal
Trigonal
a = 1.0418 nm
R-3c (167)
R3c (161)
00-55-0898
c = 3.7346 nm
a = 1.035 nm
c = 3.7085 nm
Ca9Mg(HPO4)(PO4)6 (WH)
04-09-3397
†ICDD-PDF = International Centre for Diffraction Data–Powder Diffraction File.

1202
Journal of the American Ceramic Society—Cuneyt Tas
Vol. 99, No. 4
Table II. Experimental Conditions and XRD Data of Products
Salt concentration in 100 mL aqueous solution (mM)
Experiment
Brushite (g)
MgCl2
NaCl
KCl
NaHCO3
Na2SO4
NaH2PO4
Temp. (°C)
Time (d)
Final pH†
Phase(s) by XRD
1
2
3
4
5
6
7
8
9
10
11
12
13
14
0.35
0.35
0.35
0.35
0.35
0.35
0.35
0.35
0.35
0.35
0.35
0.35
0.35
0.35
—
—
—
—
—
—
—
—
—
—
95
95
95
—
95
—
—
—
—
—
—
—
—
5
5
5
—
5
—
27
—
—
—
—
—
—
27
27
—
—
27
—
—
—
—
—
—
—
—
0.5
0.5
—
—
0.5
—
—
—
—
—
—
—
1
1
1
—
—
1
70
70
70
37
37
37
115
37
37
70
37
70
70
70
1
1
1
2
7
21
1
21
21
1
21
1
1
6.3
6.4
4.5
7.4
5.9
5.2
5.3
5.5
7.0
7.0
5.4
6.4
6.8
N/A
DCPA
OCP
WH
DCPD
DCPD + OCP
WH
WH
WH + OCP
Ap-CaP
WH + Ap-CaP
WH + Ap-CaP
DCPD
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
10
10
DCPD
DCPD
1.5 mM MgCl2Á6H2O in pure ethanol
1
†All pH values reported here showed the maximum variability of Æ0.1.
hydrogen bonds, but these hydrogen bonds are broken when
the bilayers of water molecules were interrupted by the crys-
tal surface termination, and the presence of such broken
hydrogen bonds explains the instability of brushite crystals
to heating (either dry or wet) at temperatures even below
50°C.
(2) Brushite to OCP
Experiment 2 provides, perhaps, the simplest and the most
economical method of OCP synthesis performed at 70°C.
The aqueous solution of Exp. 2, with an initial pH of
-
7.8 Æ 0.1, only contains 27 mM HCO3 (i.e., the bicarbonate
ion concentration of the human blood plasma) and the pro-
cedure of Exp. 2 does not involve any stirring and pH con-
trol during synthesis in contrast to more tedious yet popular
OCP synthesis techniques.48, 49 The XRD data of experi-
ments 1 and 2 are given in Fig. 1. Although the final solution
pH measured in Exps. 1 and 2, at the end of 24 h at 70°C,
were close to one another (Table II), the solution pH alone
Fig. 1. XRD traces of the samples of experiments 1 through 3 of
Table II.
does not strongly dictate which CaP phase will form. The
solution of Exp.
À
2
matched the HCO3 concentration
(27 mM) of blood plasma, while its Na+ concentration was
only 19% of that of blood.
with a step size of 0.02° and a 5 s preset time while using a
rotating specimen holder.
2À
The positive effect of CO3 ion in enhancing the forma-
tion of OCP was previously reported by Iijima et al.,50
although their synthesis experiments did not utilize DCPD as
the starting material. The carbonate content, if any, of the
powder samples of Exp. 2 were not measured.
III. Results and Discussion
The select experiments and their qualitative XRD phase anal-
ysis results are shown in Table II together with the solution
preparation details for each experiment. The presentation
and the discussion of the experimental findings are, therefore,
directly based on Table II.
(3) Brushite to Whitlockite
In order to form whitlockite (WH), out of brushite crystals,
the aging solution must have Mg2+ ions. The Mg2+ concen-
tration of 1.5 mM was selected, since it is the [Mg2+] of the
human blood plasma.51 Experiment 3 resulted in the forma-
tion of single-phase whitlockite crystals at a final solution
pH of 4.5 at the end of 24 h at 70°C. Figure 1 shows the
XRD trace of the Exp. 3 sample and all peaks present in that
trace do match with those listed in the ICDD PDF 04-09-
3397 of WH. Figure 2(a) shows the morphology of the start-
ing brushite flat plate-shaped crystals, whose XRD data were
previously reported elsewhere.45 Figures 2(b) and (c) depict
the particle morphology of hexahedral WH crystals (of about
200 nm in the largest dimension) formed in Exp. 3, from the
brushite crystals of Fig. 2(a), at two different magnifications.
The flat plates (looking like brushite crystals) seen in the
right half of Fig. 2(b) did not produce any brushite peaks in
the XRD data of Exp. 3 (Fig. 1). Stulajterova and Med-
(1) Brushite to Monetite
Brushite (CaHPO4Á2H2O) decomposes in dry air into mon-
etite (DCPA; dicalcium phosphate anhydrous, CaHPO4)
when heated to temperatures at or above 45°C17 and also
starts its decomposition to DCPA when it is heated in deion-
ized or distilled water at or above 40°C.46 Experiment 1
showed that 0.35 g of brushite crystals of this study (having
the characteristic crystal morphology and XRD pattern
reported previously5,17) completely transformed into DCPA
upon heating in 100 mL of deionized water at 70°C for 1 d.
Water is incorporated into the crystal structure of brushite
and the water layers of brushite appear as bilayers parallel to
the (020) faces of crystals.4, 47 Water molecules inside the
2À
crystal structure are linked to the HPO4 groups by bulk

April 2016
Brushite to Whitlockite
1203
(a)
(b)
(c)
(d)
Fig. 2. (a) flat plate morphology of starting brushite crystals, (b) and (c) hexahedral whitlockite crystals formed in experiment 3, (d) EDXS
spectrum of the whitlockite crystals of experiment 3.
vecky52 reported that brushite hydrolyzed for long times in
Ca2+ ion-containing aqueous solutions at elevated tempera-
tures (as we did; however, in our case the source of aqueous
Ca2+ was the dissolving brushite platelets themselves) may
cause the drastic loss of brushite reflections from the XRD
data of such samples while the particles still maintaining
their plate-like shapes, in strong confirmation of our observa-
tions in this study. The morphology of the synthetic, hexahe-
dral whitlockite crystals of this study [Fig. 3(c)] resemble
those found and reported in the human tooth enamel,34
human dental calculus,43, 53 calculi of human prostate,41 and
human articular cartilage.36 The mineralized sections of
mature bones do not contain whitlockite in opposition to
what is claimed by Jang et al.31, 54 The characteristic EDXS
data for the Exp. 3 sample is given in Fig. 2(d) and the semi-
quantitative EDXS analysis indicated the presence of
2.2 Æ 0.1 wt% Mg. The theoretical Mg wt% of WH is 2.31.
Would it be possible to synthesize WH, from brushite, at
37°C in less than 1 d in a nonstirred solution? The answer
was not affirmative. Increasing the aging time at 37°C, in the
1.5 mM Mg2+ solution of Exp. 4, to 2 d did not cause the
formation of any CaP phase other than DCPD. However,
one week of aging at 37°C in the 1.5 mM Mg2+ solution of
Exp. 5 resulted in the formation of the secondary phase of
OCP. In confirmation of the early yet inspirational contribu-
tion of Rowles,23 brushite crystals aged in a 1.5 mM Mg2+
solution at 37°C for 21 d (Exp. 6) resulted in the complete
disappearance of brushite in favor of the formation of single-
needs to consult two different solubility diagrams to compare
the solubilities of CaPs with one another, one diagram with
the axes of log [Ca] and pH, and the other with log [P] and
pH. Two such diagrams have previously been provided by
Chow and Takagi55 and they indicate higher rates for Ca
release in comparison to that of phosphate ions. The recently
reported Ca-deficiency in brushite crystals,5 and the associ-
ated phenomenon of the maturation of brushite during its
synthesis, actually helps the above discussion. When DCPA
(formed upon the initial conversion of DCPD into DCPA)
undergoes its dissolution in aqueous solutions, the solution
became progressively enriched in [Ca2+] and this by itself
explains the transition from DCPD/DCPA to OCP to WH
(in the presence of Mg2+), which represents a continuous
increase in the Ca/P molar ratio of the reprecipitating
phases.
(4) Brushite to Whitlockite + OCP Biphasics
Experiment 6, as mentioned previously, is a slow but sure
way of transforming brushite (DCPD) into single-phase whit-
lockite (WH) at the human body temperature of 37°C in
21 d in a simple aqueous solution only containing 1.5 mM
Mg2+. In accord with the above discussion, if brushite disso-
lution continuously increases the Ca2+ concentration of the
solution, one way to slowdown or to counter this [Ca2+
]
enrichment would be to add 1 mM phosphate ions to the
same solution and repeat the experiment with the new solu-
tion at 37°C for 21 d. The human blood plasma contains
1 mM phosphate ions. This physiologically relevant solution
initially containing 1.5 mM Mg2+ and 1 mM phosphate ions
was tested in experiment 8 and it resulted in the formation of
WH + OCP biphasics in lieu of single-phase WH. Since the
formulations of Exp. 6 and 8 were tested at the same temper-
ature for the same aging time, they thus provide a direct
comparison for the influence of 1 mM phosphate ions (in the
solution of Exp. 8) on the phase constitution of solids recov-
ered. The XRD traces of the samples of experiments 6 and 8
are given in Fig. 3(a). The characteristic morphology of a
phase WH. The message received from experiments
3
through 6 was: WH formation, taking place at the expense
of brushite crystals, proceeds in 1.5 mM Mg2+ solutions via a
dissolution–reprecipitation process, and the kinetics of which
is enhanced by an increase in the aging temperature from the
human body temperature of 37°C to 70°C. Increasing the
temperature of aging to 115°C (Exp. 7 of Table II) did again
result in the formation of single-phase WH.
Calcium phosphate dissolution has a unique character; cal-
cium and orthophosphate ions do not leach out, from any
CaP phase, at exactly the same rate, and this is why one

1204
Journal of the American Ceramic Society—Cuneyt Tas
Vol. 99, No. 4
(a)
Fig. 4. XRD traces of the samples of experiments 9 through 13 (A:
apatitic CaP, W: whitlockite).
(b)
(6) Brushite to Whitlockite + Apatitic CaP Biphasics
Experiment 10 helps to describe the synthesis of WH + Ap-
CaP biphasic samples (Fig. 4) by only changing the
hydrothermal transformation temperature from 37°C to
70°C, in combination with a drastic decrease in reaction
time. Note that the solution composition and the brushite
amount in both Exps. 9 and 10 were identical. Experiment
10 shows that it is possible to synthesize WH crystals at
70°C in a physiologically relevant, biomimetic solution hav-
ing a pH of 7. In comparison to Exp. 10, experiment 11 (per-
formed at the biomimetic temperature of 37°C for 21 d) uses
a solution free of bicarbonate, sulfate and phosphate ions to
facilitate a drop in pH to around 5, and the obtained sam-
ples were comprised of high crystallinity WH, only contami-
nated with small amounts of Ap-CaP (Fig. 4). The hump
(Fig. 4, Exp. 11) at around 32° 2h is indicative of the pres-
ence of Ap-CaP.
Fig. 3. (a) XRD traces of the samples of experiments 6 and 8; O
indicates OCP peaks in the trace of experiment 8; experiments 6
trace belongs to single-phase WH, (b) SEM photomicrograph of the
sample of experiment 8; the flake-like tiny crystals are of OCP,
where the background hexahedral crystals are of WH.
(7) Inhibition of Brushite Transformation
A water-based solution (100 mL) containing only 10 mM
Mg2+ was found to completely inhibit the transformation of
brushite crystals (0.35 g) into any other CaP phase, which
can be detected by powder XRD, such as DCPA, OCP, WH,
or Ap-CaP, even upon heating the said crystals in that solu-
tion at 70°C for 1 d (Exp. 12). The SIEM solution [of Sec-
tion III(5)] with a [Mg2+] raised to 10 mM was also found to
inhibit the transformation of brushite into any such CaP
phases within the detection ability of powder XRD (Exp. 13;
70°C, 1 d). The XRD traces of experiments 12 and 13 are
shown in Fig. 4, indicating that only the crystalline brushite
phase was present in both samples.
Experiments 3 and 12 are directly comparable with one
another. While experiment 3, which resulted in WH forma-
tion, employed a solution with a [Mg2+] of 1.5 mM, the same
concentration was increased to 10 mM in Exp. 12; as the
remaining experimental parameters were unchanged. Simi-
larly, the experiments 9 and 13 are comparable with one
another. In Exp. 9, the [Mg2+] of the SIEM solution was
1.5 mM and the resultant sample was comprised of phase-
pure Ap-CaP. Increasing the [Mg2+] in Exp. 13 to 10 mM
inhibited the Ap-CaP formation.
WH + OCP biphasic sample (e.g., Exp. 8) is shown in the
SEM photomicrograph of Fig. 3(b), where the occasional
OCP flake-like crystals seemed to contaminate the hexahe-
dral WH crystals. It should be emphasized that WH is not a
CaP phase, but it is rather a CaMgP (calcium–magnesium
phosphate) phase with a Ca + Mg/P molar ratio of 1.4286.
(5) Brushite to Apatitic CaP
The biomimetic synthesis of phase-pure Ap-CaP is not so
difficult when using the route of hydrothermal transforma-
tion of brushite at the human body temperature of 37°C.
The simple solution used in Exp. 9 has 124 mM Na+,
103 mM ClÀ, 1.5 mM Mg2+, 5 mM K+, 1 mM PO4, and
0.5 mM SO42À. This extremely stable (with respect to any
colloidal particle formation or decomposition) solution is
named as saline ionic essentials medium (SIEM).56 The
above-mentioned ions are all present in the human blood
plasma at the stated concentrations. The only slight mis-
match in the composition of the solution of Exp. 9 is seen
in its Na+ concentration, where the Na+ concentration of
blood plasma is equal to 142 mM. A quantity of 350 mg
of brushite was soaked in 100 mL of SIEM at 37°C for
21 d to synthesize single-phase, poor crystallinity Ap-CaP
(Fig. 4) at the final pH of 7. This is a nonstirred synthesis
process, which required no pH adjustments or control
whatsoever.
Although the strong influence of Mg2+ ions present in the
aqueous synthesis media exert to stop the phase transforma-
tion of brushite into any other CaPs, such as OCP or Ap-
CaP, was known,57 electron microscopical evidence on this
inhibition was not abundant in the literature. The SEM anal-
ysis performed on the samples of Exps. 12 and 13 was pow-
erful enough to disclose what was happening on the surfaces

April 2016
Brushite to Whitlockite
1205
(a)
(b)
(c)
(d)
Fig. 5. SEM photomicrographs of the samples of experiment 12, (a) and (b); experiment 13, (c) and (d).
of brushite crystals. The photomicrographs of Fig. 5, for the
samples of both Exps. 12 and 13, show the formation of
nanofibrils of X-ray amorphous calcium phosphate or Mg-
doped calcium phosphate (which do not show up in the pow-
der XRD runs) on the surfaces of large brushite crystals. The
surface nanofibrils actually start forming as a monolayer of
individual yet much smaller globules [Figs. 5(b) and (f)] on
the flat brushite surfaces, eventually coalescing with one
another to form the nanofibrils. With the progress in surface
coverage of brushite crystals by such nanoglobules/nanofib-
rils of X-ray-amorphous calcium phosphate or Mg-doped
calcium phosphate, the inherent solubility of brushite crystals
diminish and they would no longer be able to transform into
OCP, WH, or Ap-CaP. X-rays still see the underlying brush-
ite crystals, by readily penetrating through those nano-thick
layers (Fig. 5) of X-ray amorphous fibrils, as single-phase
brushite of high crystallinity (as shown in the XRD data of
Fig. 4). The role of Mg in biomineralization (in vitro) is dis-
cussed, in detail, elsewhere.56 Experiment 14, on the other
hand, shows the presence of such strong inhibition of brush-
ite transformation to persist at 70°C (1 d) in pure ethanol
solutions having even a much lower [Mg2+] of 1.5 mM.
In brief, the solutions and hydrothermal techniques
reported in this study can be adopted to transform brushite-
coated metallic implant materials into OCP, whitlockite, or
apatitic calcium phosphate. The same can also be used to
hydrothermally transform brushite granules, brushite cement
bodies, or 3D-printed brushite bioceramic scaffolds into whit-
lockite, OCP, or apatitic calcium phosphate.
1. DCPD crystals transformed to single-phase OCP, in
less than a day, when statically heated at 70°C in a
simpleÀ aqueous solution only containing 27 mM
HCO3
.
2. DCPD crystals transformed to single-phase WH, in
less than a day, when statically heated at 70°C in an
aqueous solution only containing 1.5 mM Mg2+
DCPD crystals also transformed to WH in less than
.
3 weeks when heated at 37°C in the same solution of
1.5 mM Mg2+
.
3. DCPD transformation at 70°C, to any other CaP, was
completely inhibited when the aqueous hydrothermal
solutions contained 10 mM Mg2+
. DCPD crystals
heated at 70°C in 1.5 mM Mg2+-containing ethanol
solutions did not transform to any other CaP as well.
4. DCPD crystals transformed to single-phase Ap-CaP,
in less than 3 weeks, when statically heated at 37°C in
a completely inorganic solution mimicking the Na+,
Mg2+, K+, ClÀ, HCO3À, SO42À, and phosphate ion
concentrations of the human blood plasma.
Disclaimer
Certain commercial instruments or materials are identified in this article solely
to foster understanding. Such identification does not imply recommendation
or endorsement by the author, nor does it imply that the instruments or mate-
rials identified are necessarily the best available for the purpose.
Acknowledgments
This study was conceived and performed in the Department of Biomedical
Engineering of Yeditepe University, Istanbul, Turkey, while the author was a
full professor there between Nov 2006 and Sep 2010. The aforesaid depart-
ment does not have a graduate program and is designed for undergraduate
education only.
IV. Conclusions
The hydrothermal transformation of brushite (DCPD) to
whitlockite (WH), OCP, apatitic calcium phosphate (Ap-
CaP), whitlockite + octacalcium phosphate (WH + OCP)
biphasics, and whitlockite + apatitic calcium phosphate
(WH + Ap-CaP) biphasic samples were separately studied in
unique solutions developed in this study. The solutions, free
of any organics, such as Tris or Hepes, were partially or
completely mimicking the inorganic ion concentrations of
physiological fluids.
References
1P. W. Brown, “Phase Relationships in the Ternary System CaO-P2O5-H2O
at 25°C,” J. Am. Ceram. Soc., 75, 17–22 (1992).
2R. I. Martin and P. W. Brown, “Phase Equilibria Among Acid Calcium
Phosphates,” J. Am. Ceram. Soc., 80, 1263–6 (1997).

1206
Journal of the American Ceramic Society—Cuneyt Tas
Vol. 99, No. 4
3E. Boanini, M. Gazzano, and A. Bigi, “Ionic Substitutions in Calcium
31H. L. Jang, et al., “Revisiting Whitlockite, the Second Most Abundant
Biomineral in Bone: Nanocrystal Synthesis in Physiologically Relevant Condi-
tions and Biocompatibility Evaluation,” ACS Nano, 8, 634–41 (2014).
32R. Gopal, C. Calvo, J. Ito, and W. K. Sabine, “Crystal Structure of Syn-
thetic Mg-Whitlockite, Ca18Mg2H2(PO4)14,” Can. J. Chem., 52, 1155–64
(1974).
Phosphates Synthesized at Low Temperature,” Acta Biomater., 6, 1882–94
(2010).
4N. A. Curry and D. W. Jones, “Crystal Structure of Brushite, Calcium
Hydrogen Orthophosphate Dihydrate: A Neutron Diffraction Investigation,”
J. Chem. Soc. A-Inorg. Phys. Theor., 23, 3725–9 (1971).
5M. A. Miller, M. R. Kendall, M. K. Jain, P. R. Larson, A. S. Madden,
and A. C. Tas, “Testing of Brushite (CaHPO4∙2H2O) in Synthetic Biomineral-
ization Solutions and In Situ Crystallization of Brushite Micro-Granules,” J.
Am. Ceram. Soc., 95, 2178–88 (2012).
33B. Dickens, L. W. Schroeder, and W. E. Brown, “Crystallographic Studies
of the Role of Mg as a Stabilizing Impurity in b-Ca3(PO4)2 I. The Crystal
Structure of Pure b-Ca3(PO4)2,” J. Solid State Chem., 10, 232–48 (1974).
34T. Kodaka, K. Debari, and M. Abe, “Hexahedrally Based Crystals in
Human Tooth Enamel,” Caries Res., 26, 69–76 (1992).
6S. Cazalbou, C. Combes, D. Eichert, C. Rey, and M. J. Glimcher, “Poorly
Crystalline Apatites: Evolution and Maturation in Vitro and in Vivo,” J. Bone
Miner. Metab., 22, 310–7 (2004).
35C. Austin, D. Hare, A. L. Rozelle, W. H. Robinson, R. Grimm, and P.
Doble, “Elemental Bio-Imaging of Calcium Phosphate Crystal Deposits in
Knee Samples From Arthritic Patients,” Metallomics, 1, 142–7 (2009).
36C. A. Scotchford, M. Vickers, and S. Y. Ali, “The Isolation and Charac-
terization of Magnesium Whitlockite Crystals From Human Articular Carti-
lage,” Osteoarthr. Cartilage, 3, 79–94 (1995).
7Y. Jin and H. K. Yip, “Supragingival Calculus: Formation and Control,”
Crit. Rev. Oral Biol. Med., 13, 426–41 (2002).
8D. B. Leusmann, “A Classification of Urinary Calculi With Respect to
Their Composition and Micromorphology,” Scand. J. Urol. Nephrol., 25, 141–
50 (1991).
37H. Newesely, “Calcifying Processes Within the Superficial Layers of the
Dental Plaque,” Caries Res., 2, 19–26 (1968).
9A. K. Rosenthal, E. Mattson, C. M. Gohr, and C. J. Hirschmugl, “Charac-
terization of Articular Calcium-Containing Crystals by Synchrotron FTIR,”
Osteoarthr. Cartilage, 16, 1395–402 (2008).
38J. D. Reid and M. E. Andersen, “Medial Calcification (Whitlockite) in the
Aorta,” Atherosclerosis, 101, 213–24 (1993).
10S. V. Dorozhkin and M. Epple, “Biological and Medical Significance of
Calcium Phosphates,” Angew. Chem. Int. Ed., 41, 3130–46 (2002).
11R. Tang, M. Hass, W. Wu, S. Gulde, and G. H. Nancollas, “Constant
Composition Dissolution of Mixed Phases II. Selective Dissolution of Calcium
Phosphates,” J. Coll. Interf. Sci., 260, 379–84 (2003).
39C. H. P’ng, R. Boadle, M. Horton, M. Bilous, and F. Bonar, “Magnesium
Whitlockite of the Aorta,” Pathology, 40, 539–40 (2008).
40R. Lagier and C. A. Baud, “Magnesium Whitlockite, a Calcium Phosphate
Crystal of Special Interest in Pathology,” Pathol. Res. Pract., 199, 329–35
(2003).
12B. Flautre, C. Maynou, J. Lemaitre, P. van Landuyt, and P. Hardouin,
“Bone-Colonization of b-TCP Granules Incorporated in Brushite Cements,” J.
Biomed. Mater. Res., 63B, 413–7 (2002).
41T. Kodaka, A. Hirayama, T. Sano, K. Debari, M. Mayahara, and M.
Nakamura, “Fine Structure and Mineral Components of Primary Calculi in
Some Human Prostates,” J. Electron Microsc., 57, 133–41 (2008).
42M. Daudon, J. C. Dore, P. Jungers, and B. Lacour, “Changes in Stone
Composition According to Age and Gender of Patients: A Multivariate Epi-
demiological Approach,” Urol. Res., 32, 241–7 (2004).
13J. M. Kuemmerle, et al., “Assessment of the Suitability of a New Brushite
Calcium Phosphate Cement for Cranioplasty – An Experimental Study in
Sheep,” J. Cranio Maxill. Surg., 33, 37–44 (2005).
14F. Theiss, et al., “Biocompatibility and Resorption of a Brushite Calcium
Phosphate Cement,” Biomaterials, 26, 4383–94 (2005).
43T. Kodaka, K. Debari, T. Sano, and M. Yamada, “Scanning Electron
Microscopy and Energy-Dispersive X-ray Microanalysis Studies of Several
Human Calculi Containing Calcium Phosphate Crystals,” Scanning Microsc.,
8, 241–57 (1994).
15D. Apelt, et al., “In Vivo Behavior of Three Different Injectable Hydraulic
Calcium Phosphate Cements,” Biomaterials, 25, 1439–51 (2004).
16K. Furutaka, H. Monma, T. Okura, and S. Takahashi, “Characteristic
Reaction Processes in the System Brushite-NaOH Solution,” J. Eur. Ceram.
Soc., 26, 543–7 (2006).
44S. Kraaij, K. H. Karagozoglu, T. Forouzanfar, E. C. I. Veerman, and H.
S. Brand, “Salivary Stones: Symptoms, Aetiology, Biochemical Composition
and Treatment,” Brit. Dent. J., 217, E23 (2014).
17A. C. Tas and S. B. Bhaduri, “Chemical Processing of CaHPO4∙2H2O: Its
Conversion to Hydroxyapatite,” J. Am. Ceram. Soc., 87, 2195–200 (2004).
18S. Mandel and A. C. Tas, ““Brushite (CaHPO4Á2H2O) to Octacalcium
Phosphate (Ca8(HPO4)2(PO4)4Á5H2O) Transformation in DMEM Solutions at
36.5°C,” Mater. Sci. Eng., C, 30, 245–54 (2010).
45A. C. Tas, “Calcium Phosphate Cement Composition and a Method for
the Preparation Thereof”; U.S. Patent No. 6,929,692, August 16, 2005.
46Z. Sheikh, Y. L. Zhang, L. Grover, G. E. Merle, F. Tamimi, and J. Bar-
ralet, “In Vitro Degradation and in Vivo Resorption of Dicalcium Phosphate
Based Grafts,” Acta Biomater., 26, 338–46 (2015).
19M. Kumar, J. Xie, K. Chittur, and C. Riley, “Transformation of Modified
Brushite to Hydroxyapatite in Aqueous Solution: Effects of Potassium Substi-
tution,” Biomaterials, 20, 1389–99 (1999).
47J. Arsic, D. Kaminski, P. Poodt, and E. Vlieg, “Liquid Ordering at the
Brushite-010-Water Interface,” Phys. Rev. B, 69, 245406 (2004).
48R. Z. LeGeros, “Preparation of Octacalcium Phosphate (OCP): A Direct
Fast Method,” Calcif. Tissue Int., 37, 194–7 (1985).
20A. C. Tas, “Electroless Deposition of Brushite (CaHPO4∙2H2O) Crystals
on Ti-6Al-4V at Room Temperature,” Int. J. Mater. Res., 97, 639–44 (2006).
21X. Xiao, H. Yu, Q. Zhu, G. Li, Y. Qu, and R. Gu, “In Vivo Corrosion
Resistance of Ca-P Coating on AZ60 Magnesium Alloy,” J. Bionic Eng., 10,
156–61 (2013).
49M. Iijima, Y. Moriwaki, T. Takagi, and J. Moradian-Oldak, “Effects of
Bovine Amelogenins on the Crystal Morphology of Octacalcium Phosphate in
a Model System of Tooth Enamel Formation,” J. Cryst. Growth, 222, 615–26
(2001).
22A. C. Tas, “Granules of Brushite and Octacalcium Phosphate From Mar-
ble,” J. Am. Ceram. Soc., 94, 3722–6 (2011).
50M. Iijima, H. Kamemizu, N. Wakamatsu, T. Goto, Y. Doi, and Y. Mori-
2-
waki, “Effects of CO3 Ion on the Formation of Octacalcium Phosphate at
23S. L. Rowles, “The Precipitation of Whitlockite From Aqueous Solu-
tions,” B. Soc. Chim. Fr., Suppl. I, 179, 7–802 (1968).
pH 7.4 and 37°C,” J. Cryst. Growth, 135, 229–34 (1994).
51H. A. Krebs, “Chemical Composition of Blood Plasma and Serum,” Annu.
Rev. Biochem., 19, 409–30 (1950).
24C. Frondel, “Whitlockite: A New Calcium Phosphate, Ca3(PO4)2,” Am.
Mineral., 26, 145–52 (1941).
52R. Stulajterova and L. Medvecky, “Effect of Calcium Ions on Transforma-
tion Brushite to Hydroxyapatite in Aqueous Solutions,” Colloid. Surface. A,
316, 104–9 (2008).
25L. H. Fuchs, “Occurrence of Whitlockite in Chondritic Meteorites,”
Science, 137, 425–6 (1962).
26C. P. A. T. Klein, K. de Groot, A. A. Driessen, and H. B. M. van der
Lubbe, “Interaction of Biodegradable b-Whitlockite Ceramics With Bone Tis-
sue: An in Vivo Study,” Biomaterials, 6, 189–92 (1985).
53T. Kodaka, K. Debari, and S. Higashi, “Magnesium-Containing Crystals
in Human Dental Calculus,” J. Electron Microsc., 37, 73–80 (1988).
54H. L. Jang, H. K. Lee, K. Jin, H. Y. Ahn, H. E. Lee, and K. T. Nam,
“Phase Transformation From Hydroxyapatite to the Secondary Bone Mineral,
Whitlockite,” J. Mater. Chem. B, 3, 1342–9 (2015).
27C. A. van Blitterswijk, J. J. Grote, H. K. Koerten, and W. Kuijpers, “The
Biological Performance of Calcium Phosphate Ceramics in an Infected Implan-
tation Site III: Biological Performance of b-Whitlockite in the Noninfected
and Infected Rat Middle Ear,” J. Biomed. Mater. Res., 20, 1197–217 (1986).
28L. M. Ryan, et al., “Cellular Responses to Whitlockite,” Calcif. Tissue
Int., 65, 373–7 (1999).
55L. C. Chow and S. Takagi, “Self-Setting Calcium Phosphate Cements and
Methods for Preparing and Using Them”; U.S. Patent No. 5,525,148, June 11,
1996.
56A. C. Tas, “The Use of Physiological Solutions or Media in Calcium
Phosphate Synthesis and Processing,” Acta Biomater., 10, 1771–92 (2014).
57H. E. L. Madsen, “Influence of Foreign Metal Ions on Crystal Growth
and Morphology of Brushite (CaHPO4Á2H2O) and its Transformation to Octa-
29S. Gomes, G. Renaudin, E. Jallot, and J. M. Nedelec, “Structural Char-
acterization and Biological Fluid Interaction of Sol-gel-Derived Mg-Substi-
tuted Biphasic Calcium Phosphate Ceramics,” Appl. Mater. Interf., 1, 505–13
(2009).
calcium Phosphate and Apatite,” J. Cryst. Growth, 310, 2602–12 (2008).
h
30E. E. Jay, et al., “Predicted Energies and Structures of b-Ca3(PO4)2,” J.
Solid State Chem., 183, 2261–7 (2010).