- Email: [email protected]
Inhibitory effect of polyelectrolytes on crystallization kinetics of hydroxyapatite
Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51 www.elsevier.com/locate/pcrysgrow
Inhibitory effect of polyelectrolytes on crystallization kinetics of hydroxyapatite ¨ ner*, O ¨ . Do M. O gan Yildiz Technical University, Chemical Engineering Department, Davutpasa, 34210 I_stanbul, Turkey
Abstract Modelling of the biologic materials, well-organized multifunctional structures and systems found in nature has attracted the interest of scientists working in many scientific disciplines. A new and rapidly growing field of biomimetics has stimulated an increased focus on biological materials as the researchers attempt to mimic the features, characteristics and growth of these naturally-occurring materials. This review discusses the principal features of biomineralization in relation to the controlled crystallization of inorganic materials and biomimetic routes to the formation of nanometer hydroxyapatite particles. This approach can be compared with biologic mineralization and has the potential for providing much greater control of particle size and distribution than would conventional methods. The constant-composition method has been used to study the influence of polyelectrolytes on the kinetics of crystal growth of hydroxyapatite (HAP), the thermodynamically most stable calcium phosphate phase, on HAP seed crystals at pH 7.4 and 37 C. The results indicate that polyelectrolyte concentration and the larger number of negatively charged functional groups markedly affect the growth rate. The fit of the Langmuir adsorption model to the experimental data supports a mechanism of inhibition through molecular adsorption of polymers on the surface of growing crystals. This system may allow insights into biomineralization processes. Ó 2005 Elsevier Ltd. All rights reserved. PACS: 81.10.-h Keywords: Crystal growth inhibition; Hydroxyapatite precipitation; Habit modification; Polyelectrolytes; Seed crystal
* Corresponding author. ¨ ner). E-mail address: [email protected] (M. O 0960-8974/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.pcrysgrow.2005.08.002
40
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
1. Introduction Biological systems are a rich source of ideas for scientists concerned with the synthesis of new materials and processes. Plants and animals have evolved a vast diversity of structures through strategies that often are very different from those used by the engineer. These naturally fabricated materials are invariably composites and are assembled from readily available materials, usually in aqueous media, at ambient conditions with high degrees of orientation, specific morphologies, and to specific polymorphs selected [1e4]. The field of biomimetics is based on the assumption that through evolution, nature has solved an optimization problem. Nature uses very few materials to create a vast variety of life forms at largely ambient temperature and pressure in order to synthesize a diverse range of mechanical and structural properties. The same material is used in many different ways to meet vastly different needs as exemplified by collagen [5,6]. Biological minerals employ a limited group of generic materials comprising organic, inorganic crystals and amorphous phases. Oyster shells, cuttlefish bone, ivory and magnetic crystals in bacteria are just a few of the vast variety of biological minerals engineered by living creatures [7,8]. Biominerals are normally produced very slowly and present a limited range of substances. The major mineralized tissues such as bone, teeth and shells are dominated by calcium phosphate or calcium carbonate minerals; nevertheless, more than 60 different types of biominerals are now known in organisms ranging from bacteria to mammals and more are being discovered at a surprising rate [9e11]. Hydroxyapatite is the major inorganic component of bone and teeth whilst the shells of invertebrates consist of calcite and aragonite. Bone, more than any other biomineral, reflects the greatest distinction between an inorganic and bioorganic mineral. Bone is a multifunctional two-phase substance, comprising flexible collagen and brittle hydroxyapatite, which is approximately 50% stronger in compression than in tension. Bone can be thought as a living mineral since it undergoes continual growth, dissolution, repair and remodeling. It performs many different functions such as skeletal homeostasis and mineral homeostasis. The skeletal homeostasis governs the growth and maintenance of the skeleton. Mineral homeostasis is dependent upon the high content of mineral ions in the bones to provide a constant concentration of essential ions in plasma. The mineral acts as an important calcium store as well as a supply of calcium in case of high demands [12,13]. Biomimetic approaches based on an understanding of the biomineralization process are aimed at synthesizing nanoparticles, polymeremineral composites and templated crystals. Studies over the last few years lead to the following distinct characteristics of biomineralization [13]: (a) Biomineralization occurs in well-constructed compartments or microenvironments; (b) These compartments have the ability to stimulate the nucleation and growth of crystals of the required inorganic material at certain functional sites while effectively inhibiting the formation of other crystals; (c) The crystal size and shape are well defined and show little variation; and (d) Formation of the macroscopic structure is accomplished by the packaging of many such units. Organisms have been manipulating minerals for almost as long as life has existed on earth. They use these minerals to fulfill functions as diverse as detoxification, temporary storage of vital ions, gravity perception, light reflection, navigation in the earth’s magnetic field and, perhaps most familiar, the construction of mechanically sound skeletal materials [14]. They have evolved sophisticated and varied means of controlling mineralization processes. This control is assumed to be mediated by specific interactions between certain crystal planes and biological macromolecules that are most conveniently referred to as acidic macromolecules [15]. There are several types of macromolecules identified depending on the organism. These are proteins or glycoproteins that are rich in acidic amino acids, either sulphate groups in polysaccharides and carboxylic
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
41
acid groups on aspartic acid proteins found in mollusks, or phosphate groups on proteins found in bone or tooth [16e18]. Direct observations of tissues show that acidic macromolecules are located at the interface between the crystal and the extracellular matrix, whereas others can actually be found within crystals themselves. Microstructural control is exerted on all levels, from the molecular and nanometer scale to the overall three-dimensional structure. On the nanometer scale, mineral crystals are formed with one or all axes oriented, specific shapes and sizes, highdensity packing, and specific polymorphs. Many of the acidic macromolecules isolated from biomineral have demonstrated the ability to regulate mineral growth in vitro. These are the main reasons why they are generally thought to be involved in manipulation and critical for the regulation of biomineral formation. The organic matrix is formed extracellularly and consists of two components: an underlying collagen sheet and individually attached macromolecules [19]. The acidic macromolecules usually comprise much less than 1% of the mass of the tissue and are often only a relatively minor component of all the macromolecules that make up the so called organic matrix. The major components are more hydrophobic framework macromolecules such as collagen, chitin and silk-fibroin-like proteins. The acidic molecules are frequently sandwiched between the surfaces of the crystal and these framework macromolecules [20]. Nucleation and growth of crystals occur within matrices. The size and shape of the crystals are thought to be controlled by the size and shape of the organic matrix, which acts as a template for the mineral microstucture. Nucleation may be initiated by the adsorption of cations onto functional sites of acidic macromolecules, which promotes the formation of critical nuclei. Crystallite orientation may be controlled by specific molecular interactions, which causes the arrangement of solution ions in specific locations relative to organic sites. Simple considerations of ionic charge, size and packing and their effect on hydration and lattice energies are sufficient to explain the thermodynamic stability of these minerals in biological environments [21]. Despite a great deal of study, the effect of biological polymers on the nucleation and growth process has not become clear. Arguments have been put forward that their effectiveness is due to particular molecular weights, hydrophobic and hydrophilic regions, a close match between spacings of acid groups and spacings of cations on the crystal surface [22]. The biological synthesis of inorganic solids often yields materials of uniform size, unusual habit, organized texture and defined structure and composition under moderate conditions of supersaturation and temperature. An understanding of biological solid-state interactions would therefore be of immense value in structural biology and medicine (for example, in the pathological mineralization of bones and teeth and formation of kidney stones) crystal growth, colloidal and solid-state science (as in the prevention of industrial scaling and controlled synthesis of electronic, magnetic and catalytic devices) and materials and engineering technology (organized composites and ceramic precursors, and the interrelationships between microstructure and mechanical properties). It seems that new chemical synthetic methods can be developed to form materials with highly controlled microstuctures if biomineralization processes can be understood and imitated [23,24]. One particularly promising approach is to use tailor-made additives that modify crystal habits through specific interactions at the crystal faces. Many of the soluble macromolecules associated with biominerals are considered to act in this manner. For example, acidic glycoproteins extracted from the adult sea urchin tests have been shown to bind specifically to the 110 faces of synthetic calcite crystals [14]. Diphosphonates and polyphosphates containing molecules are well known as calcification inhibitors. The direct precipitation of HAP on HAP seed crystals has been studied in the presence of EHDP (ethane-1-hydroxy-1,1,1-diphosphonic acid) and phytic acid [25]. The marked inhibitory influence of the additives on the HAP crystallization
42
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
was observed at physiological pH. DDPE (1,2-dihydroxy-1,2-bis(dihydroxyphosphonyl)ethane) has been successfully used as inhibitors of the crystal growth of HAP [26]. In vivo, HAP grows in solution that contains many essential trace elements besides calcium and phosphorus. Inhibitory effects of magnesium and zinc on HAP crystal growth have been investigated by many research groups [26e28]. It was confirmed that magnesium and zinc reduced the growth rate of HAP [26]. The adsorption behavior of single amino acids and their inhibitory effect on the crystal growth of HAP are of great importance for the calcification process [29]. It was found that phosphorine has the strongest affinity for HAP compared with aspartic and glutamic acids and serine [29e32]. The importance of protein eCOOH side chains in mineralization has been tested by several investigators using HAP as a model system [33e38]. It was found that the crystal growth rates of HAP decreased markedly in the presence of both aspartic and glutamic acid [35]. Lysine, which is an amino acid with a basic side group, was also used for hydroxyapatite crystallization [34]. It was found that the crystal growth of HAP was inhibited by lysine and this inhibition was attributed to adsorption and further blocking of the active growth sites on the crystal surface. Poly-L-glutamate and poly-L-aspartate have been shown to inhibit HAP growth when present in solution but also act as HAP nucleators when adsorbed on germanium surfaces [30]. Other workers have found that the presence of ferrocene complexes inhibited the crystal growth of HAP [36]. Studies have also been conducted on the role of serine, tyrosine and hydroxyproline amino acids with polar side groups. Depending on the side group of the amino acid different inhibiting activities were observed [33]. The precipitation and dissolution of calcium phosphate salts is of particular interest because of its importance in industrial water systems, in waste water treatment processes, in agriculture as fertilizers and in biological calcification processes [39]. Under physiological conditions the most stable calcium phosphate phase is hydroxyapatite [Ca5(PO4)3OH, HAP] [40]. The growth mechanism of HAP has received considerable attention in view of its importance in understanding the mechanism of hard tissue calcification such as bone and teeth and in many undesirable cases of pathological mineralization of articular cartilage, dental caries and kidney stones [41e43]. One problem in nucleation and crystal growth, of considerable interest in both industrial and biological situations, concerns the way in which certain polyelectrolytes inhibit or delay crystallization from supersaturated solutions of inorganic salts. Although preferential adsorption of a growth inhibitor at the crystal surface is known to be an essential step in its specific performance, the precise nature of the various factors influencing this step and the role they play in the process are not yet fully disclosed [44,45]. As many biological molecules contain oxyanion functional groups (carboxylate, phosphate and sulphate esters) we have been involved in crystallization experiments using polyelectrolytes. In this work we have prepared a range of acrylic and methacrylic polymers with different architectures to explore their relative effectiveness in inhibiting crystal growth of hydroxyapatite. These polymers consist of a hydrophilic block, which promotes water dissolution (polyethylene glycol), and a second hydrophilic block (acrylic block), which strongly interacts with minerals.
2. Experimental procedure A series of acidic acrylate block copolymers have been made, by radical polymerization, with defined molecular weight and structure. Radical polymerization of acrylic acid (AA) was carried out in the presence of a-thio polyethylene glycol monomethylether as a chain transfer agent to produce poly (ethylene glycoleblockeacrylic acid) copolymers. PEG block length in the copolymers
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
43
was controlled by using three different molecular weight chain transfer agents (Mn: 350, 750 and 2000 g/mole). The complete experimental procedures were reported previously [46]. Crystal growth experiments were made in magnetically stirred water-jacketed cells of 1 L capacity with water thermostated at 37 G 0.1 C. Reagent grade and triply distilled water were used and crystallization experiments were conducted in a nitrogen atmosphere. Carbon dioxide was excluded from the solutions by bubbling with presaturated nitrogen gas. The HAP seed material prepared as described in the literature had a molar TCa/Tp ratio of 1.66 G 0.1. Their specific surface area (SSA) was found to be 33.1 G 0.1 m2/g, as determined by B.E.T method (Micromeritics FlowSorb II-2300). The decrease in SSA during the precipitation reactions from 33.1 to 26.9 m2/g was taken into consideration for calculation of reaction rates. The seed crystals were characterized by scanning electron microscopy (JEOL-FEGSEM), X-ray powder diffraction (SHIMADZU XRD-6000) and infrared spectroscopy (KBr pellet method, PerkinElmer Spectrum One). In a typical constant-composition crystal growth experiment, stable supersaturated solutions of calcium phosphate with a molar ratio Cat/Pt Z 1.67 were prepared in water-jacketed reactor. The total molar concentration of calcium, Cat was 5.00 ! 104 mole/L with calcium/phosphate molar ratio 1.67. It should be noted that this solution was undersaturated with respect to other calcium phosphate phases such as OCP (orthocalcium phosphate) and DCPD (dicalcium phosphate dihydrate) [38]. The pH was adjusted to the required value by the addition of dilute potassium hydroxide solution. Following the pH adjustment and verification of the stability of the supersaturated solution for at least 2 h, known weights of HAP seed crystals were added to the solution. As growth commenced, the release of protons lowered the pH of the solutions and a pH drop triggered the addition of titrant solutions, which was controlled by means of pHstat (Radiometer pH-Stat Titrator PHM290, Auto-Burette ABU901). The crystal growth reaction was monitored by the addition of titrant solutions as a function of time from mechanically coupled automatic burettes. Throughout the course of the seeded growth experiments, the pH of the working solution and the added volume of titrants as a function of time were recorded and stored in the computer for further analysis. The titrant solutions in the burettes consisted of calcium chloride, potassium phosphate, potassium hydroxide and the inhibitor. Identical volumes of these solutions were introduced by burettes and the titrant molar concentration ratio was adjusted systematically so as to match the expected stoichiometry of the precipitated phase after correction for dilution of the supersaturated working solution. The molar concentration ratio of the titrant corresponded to the stoichiometry of the HAP phase, Ca:P:OH Z 5:3:1. As a result the HAP crystallization took place under conditions of constant supersaturation. The supersaturated working solution contained sufficient concentration of potassium chloride (inert electrode) so as to maintain the ionic strength of the solution during titrant addition. The course of the reaction was followed by removing homogeneous aliquots at various times. These were quickly filtered through Millipore filters of 0.22 mm pore size and the crystals removed by filtration were examined by scanning electron microscopy (SEM), X-ray powder diffraction, FT-IR and specific surface area. The rates of crystallization were determined from the rates of addition of mixed titrants and corrected for surface area changes.
3. Results and discussion The grown needlelike crystals were identified as hydroxyapatite by XRD (Fig. 1) and compared with that of the ASTM (09-0432) standards (Table 1). Lattice parameters of the crystals
44
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
Fig. 1. X-ray powder diffraction of the hydroxyapatite seed crystals.
precipitated are a Z b Z 0.9417 nm and c Z 0.6879 nm. These values are very close to the theoretical values (a Z b Z 0.9418 nm and c Z 0.6884 nm). FT-IR spectra of the seed crystals also showed that they consisted of HAP (Fig. 2). The bands at 3570 cm1 are characteristic of stretching and vibrational modes of the OH ions. 1 1 Characteristic frequencies derived from PO3 4 modes can be seen at 1094 cm , 1035 cm 1 1 1 and 962 cm , between 600 and 560 cm . The IR peak at 875 cm characteristic of 2 HPO2 4 was observed for the crystal. CO3 ions were detected in the precipitate from the peaks at 1453 cm1 and 1415 cm1. Crystal growth experiments were performed at constant-composition conditions. The ability to measure quantitatively the rates of mineralization at constant thermodynamic driving forces is one of the most important advantages of the constant-composition kinetic methods [47]. Typical plots of moles of HAP grown on HAP seed crystals as a function of time, after correction of the raw data for the observed changes in specific surface area are shown in Fig. 3. The slopes of the lines are used to calculate the growth rates as moles of HAP per square meter of surface. Table 1 List of the peaks observed in the X-ray powder diffraction pattern of HAP and the respective data from ASTM file HAP seed crystals
ASTM (09-0432)
2q
˚) d (A
I/II
hkl
2q
˚) d (A
I/II
hkl
25.846 31.891 32.843 34.017 39.752 46.650 49.440 50.387 52.015 53.171
3.444 2.804 2.725 2.633 2.265 1.945 1.842 1.809 1.757 1.721
50 100 60 27 23 29 35 14 11 21
002 211 300 202 310 222 213 321 402 004
25.879 31.773 32.902 34.048 39.818 46.711 49.468 50.493 52.100 53.143
3.440 2.814 2.720 2.631 2.262 1.943 1.841 1.806 1.754 1.722
40 100 60 25 20 30 40 20 16 20
002 211 300 202 310 222 213 321 402 004
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
45
Fig. 2. Infrared spectra of the seed crystals (a) and crystals grown on HAP seed (b).
The experiments have been carried out with methacrylate and acrylate block and homopolymers in order to see the differences in inhibitor performance. The effect of polyelectrolytes on the crystallization of HAP, from calcium phosphate solutions, supersaturated with respect to HAP, was studied by a series of experiments summarized in Table 2. Plots of HAP grown on HAP seed crystals, as a function of time for the crystal growth experiments in the presence of homopolymers and copolymers are shown in Fig. 3 and rate data are summarized in Table 2. It can be seen from Table 2 that polymers at a concentration of 3 ppm in the supersaturated solution resulted in reduction in the rate of crystallization. There is no difference between homo and block copolymers PMAA in terms of inhibition effectiveness (Fig. 4). Under conditions used in the present work, PAA and EO-b-AA copolymers at concentration of 3 ppm retarded crystal growth of HAP to a greater degree than PMAA and EO-b-MAA copolymers (Fig. 4).
Added Titrant Volume/Surface Area (ml/m2)
45 Blank
40
PMAA
35
EO-b-AA PAA
30 25 20 15 10 5 0
0
50
100
150
200
250
300
350
Time (minute) Fig. 3. Crystal growth of HAP on HAP seed crystal. Titrant added as a function of time in the presence of polyelectrolytes.
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
46
Table 2 Effect of EO/AA and EO/MAA ratios of polymers on crystal growth of HAP at 37 C and 3 ppm polymer concentration Polymer
EO/AA
EO/MAA
Mn
HAP growth rate ! 108 (mole/m2 min)
Blank PMAA EO-b-MAA EO-b-AA PAA
e e e 1.31 e
e e 8 e e
e 8000 17,500 4500 5000
15.44 10.40 10.39 3.64 2.48
PMAA, poly(methacrylic acid); PAA, poly(acrylic acid); EO-b-MAA, ethylene glycol-b-methacrylic acid; EO-b-AA, ethylene glycol-b-acrylic acid.
Results of precipitation experiments conducted in the presence of varying concentrations of PAA, PMAA and EO-b-AA are summarized in Table 3 and shown in Fig. 5. It can be seen from Fig. 5 that an increase in polymer concentration results in a decrease of rate values. The decrease of rate values depends on the nature of the inhibitor and its effective concentration at a given condition [48e51]. It can be seen in Fig. 5 that the adsorption strength for polymers leveled off when the polyelectrolyte concentrations reached a certain value. Similar findings were reported where the equilibrium surface concentration of human serum albumin plateaued at about 1.6 mg/m2 of HAP [52]. The crystallization was strongly inhibited even at low surface concentrations of PAA. Rate values decrease from 3.39 ! 108 (mole/m2min) to 2.27 ! 108 (mole/m2min) when the polymer concentration increases from 1 ppm to 5 ppm. In the case of PMAA, the rate of HAP crystallization was reduced to 12.43 ! 108 (mole/m2min) at 1 ppm concentration and reached an inhibition value of 6.99 ! 108 (mole/m2min) at a concentration of 5 ppm. In previous studies, it had been found that there was a direct relationship between the fractional site coverage of an adsorbed species and the fractional reduction of the rate of crystal growth. This behavior led to the conclusion that the adsorption takes place on the sites where crystal growth occurs [53,54]. Since the amounts of polymer in solution were small, the inhibition could probably be attributed to the blocking of the active growth sites of the seed crystals rather than to the binding to calcium ions in the solution [55]. This assumption was tested by fitting the kinetic results in 16 14
3
PAA
2
4
EO/AA=1.31
6
EO/MAA=8
8
PMAA
10
Blank
R (mole/m2min)
12
2 0
1
4
5
Fig. 4. Comparison of the HAP growth rate by pairs of similar polymers with different architecture.
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
47
Table 3 Effect of polymer concentrations on growth rate of HAP Polymer
HAP growth rate ! 108 (mole/m2 min)
Concentration (ppm)
EO-b-AA EO/AA Z 1.93; Mn Z 3700
PMAA Mn Z 8000
PAA Mn Z 5000
0.5 2 5 10
8.57 5.14 2.65 2.52
1 3 5 7.5
12.43 10.40 6.99 5.41
1 2 3 5
3.39 2.52 2.48 2.27
a langmuir-type kinetic isotherm [56]. When a polyelectrolyte poisons active growth sites on the crystal faces, the coverage by polymers, q, for adsorption model is described by qZKCi =ð1CKCi Þ
ð1Þ
where K is the adsorption or affinity constant which is the ratio of the rate constants for adsorption and desorption, kads/kdes, and can be considered as a measure for the adsorption affinity of the polymer for the crystal surface and Ci is the total equilibrium concentration of the polymer. Depending on the surface coverage, q, the rate of HAP crystal growth in the presence of the polymeric additive, Ri, is given by
R (mole/m2min)
Ri ZR0 ð1 qÞ
ð2Þ
13 12 11 10 9 8 7 6 5 4 3 2 1
EO-b-AA (Mn=3700, EO/AA=1.93) PMAA (Mn=8000) PAA (Mn=5000)
0
1
2
3
4
5
6
7
8
9
10
11
Polymer Concentration (mole/l) Fig. 5. Reduction of the crystallization rate of HAP at constant solution supersaturation as a function of the polymer concentration.
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
48
where R0 is the crystallization rate in the absence of additive. Combination of Eqs. (1) and (2) gives R0 kdes 1 Z1C R0 Ri kads Ci
ð3Þ
Eq. (3) shows that this model predicts a linear relationship between R0/(R0Ri) and 1/Ci. The linearity of the plots of Eq. (3) for HAP crystal growth in the presence of polymers (Fig. 6) suggests that the inhibitory effect of polymers is due to adsorption at active growth sites. kads/kdes can be evaluated from the slope of the resulting straight line. The values of the affinity constant as calculated for PAA, EO-b-AA and PMAA are 36.50 ! 106, 6.71 ! 106 and 1.99 ! 106 L/mole, respectively. The high value of the affinity constant for PAA may reflect stronger equilibrium adsorption of PAA than of PMAA on the HAP surface. Photographs were also taken by scanning electron microscope (SEM) for the subsequent visual analysis in order to assess the effects of polymers on crystal shape and size (Fig. 7). The crystals precipitated at 37 C have the form of tiny needles with average length of 293 nm and width of 65 nm (Fig. 7a). Fig. 7bed shows the effect of polymers on crystal habit. No apparent changes in morphology were observed in the presence of polyelectrolytes. The average length of the crystals grown in the presence of the polymers was significantly less than that of the control sample. The average length of the crystals was reduced to 274 nm, 200 nm and 191 nm for solutions containing 3 ppm PMAA, EO-b-AA and PAA, respectively. The average value of the width in the samples containing polymers was similar to that of the control sample. 4. Conclusions All polymers tested in this study are effective as growth inhibitors in this experimental condition. The higher affinity of PAA for HAP corresponds to the more significant effect of this polymer on the rate of HAP crystal growth. The PAA chains are more flexible compared with those containing methyl group, thus increasing the interaction ability of the appended anionic groups with the solid phase and producing an overall three-dimensional arrangement most suitable for strong binding. The polymer backbone can then act as a fence on the crystal surface, thus forming an obstacle for the propagating steps. 6 PAA EO-b-AA (EO/AA=1.93) PMAA
5
Ro/(Ro-Ri)
4 3 2 1 0 0
1
2
3
4
5
6
7
8
9
1/C 10-6 (mole/l)-1 Fig. 6. Crystal growth of HAP on HAP seed crystals. Langmuir-type adsorption isotherm for the effect of polymers.
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
49
Fig. 7. SEM of HAP crystals grown for 5 h from solution; pH 7.4 and at 37 C. (a) Growth of HAP on HAP seed crystals. (b) Crystals grown from a solution containing 3 ppm poly (ethylene glycol-b-acrylic acid) block copolymer. (c) Crystals grown from a solution containing 3 ppm PAA. (d) Crystals grown from a solution containing 3 ppm PMAA.
All the results of the present investigation indicate that polyelectrolytes inhibit crystal growth and that such an inhibition is directly related to the fractional coverage of adsorption sites. From a physicochemical point of view, efficient regulators of calcification would be those molecules that have a high enough adsorption affinity so that low concentrations can result in the adsorption on and blocking of a significant number of crystal growth sites on the crystallites of the tissue. Acknowledgements We appreciate the support of YTUAF (Project No: 22-07-01-01); TUBITAK (Project No: TBAG-AY/236(101T098); TBAG-AY/250(101T174)) and DPT (Project No: 24-DPT-07-0401) for the accomplishment of this work. The authors thanks Prof. Dr. P.G. Koutsoukos of Patras University for valuable discussions and help in various phases of this work.
References [1] A. Berman, L. Addadi, S. Weiner, Nature 331 (1988) 546. [2] A. Berman, J. Hanson, L. Leiserowitz, T.F. Koetzle, S. Weiner, L. Addadi, Science 259 (1993) 776.
50 [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51]
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51 S. Mann, Nature 332 (1988) 119. S. Mann, Nature 365 (1993) 499. S. Mann, J.P. Hannington, R.J. Williams, Nature 324 (1986) 565. S. Mann, B.R. Heywood, S. Rajam, J.D. Birchall, Nature 334 (1988) 692. P. Calvert, Materials Research Society Bulletin 17 (1992) 37e40. P. Calvert, Chemistry of Materials 8 (1996) 1715. A.H. Heuer, D.J. Fink, V.J. Laraıa, J.L. Arias, P.D. Calvert, K. Kendall, G.L. Messing, J. Blackwell, P.C. Rieke, D.H. Thompson, A.P. Wheeler, A. Veis, A.I. Caplan, Science 255 (1992) 1098. B.R. Heywood, S. Mann, Advanced Materials 6 (1) (1994) 9. S. Mann, Nature 349 (1991) 285. P. Calvert, Nature 334 (1988) 651. A.H. Heuer, D.J. Fink, V.J. Laraia, Science 255 (1992) 1098. A. Berman, J. Hanson, L. Leiserowitz, T. Koetzle, S. Weiner, L. Addadi, Journal of Physical Chemistry 97 (19) (1993) 5162. S. Mann, B.R. Heywood, J.M. Didymus, S. Rajam, V.J. Wade, J.B. Walker, Materials Research Society Symposium Proceedings 174 (1990) 25. L. Addadi, A. Berman, J. Moradianoldak, Croatica Chemica Acta 63 (1990) 539. A. Berman, L. Addadi, A. Kvick, Science 250 (1990) 664. L. Addadi, A. Berman, J.M. Oldak, Connective Tissue Research 21 (1e4) (1989) 457. S. Weiner, A. Berman, L. Addadi, Ultramicroscopy 23 (2) (1987) 231. S. Mann, Nature 351 (1991) 24. R.J. Davey, S.N. Black, L.A. Bromley, D. Cottier, B. Dobbs, J.E. Rout, Nature 353 (1991) 549. T. Aoba, E.C. Moreno, Journal of Colloid and Interface Science 106 (1) (1985) 110. G.H. Nancollas, W. Wu, Journal of Crystal Growth 211 (2000) 137. S. Mann, D.D. Archibald, J.M. Didymus, T. Douglas, B.R. Heywood, F.C. Meldrum, N.J. Reeves, Science 261 (1993) 1286. P.G. Koutsoukos, Z. Amjad, G.H. Nancollas, Journal of Colloid and Interface Science 83 (1981) 599. M. Dalpi, E. Karayianni, P.G. Koutsoukos, Journal of the Chemical Society Faraday Transactions 89 (6) (1993) 965. T.A. Fuierer, M. LoRe, S.A. Puckett, G.H. Nancollas, Langmuir 10 (1994) 4721. Z. Amjad, P. Koutsoukos, G.H. Nancollas, Journal of Colloid and Interface Science 101 (1984) 250. N. Spanos, P.G. Koutsoukos, Journal of Materials Science 36 (2001) 573. A. Tsortos, G. Nancollas, Journal of Colloid and Interface Science 250 (2002) 159. E. Moreno, M. Kresak, A. Gaffar, Journal of Colloid and Interface Science 168 (1) (1994) 173. M. Kresak, E. Moreno, R. Zahradnik, Journal of Colloid and Interface Science 59 (2) (1977) 283. S. Koutsopoulos, E. Dalas, Journal of Crystal Growth 216 (2000) 443. S. Koutsopoulos, E. Dalas, Journal of Colloid and Interface Science 231 (2000) 207. S. Koutsopoulos, E. Dalas, Journal of Crystal Growth 217 (2000) 410. S. Koutsopoulos, E. Pierri, E. Dalas, N. Tzavellas, N. Klouras, Journal of Crystal Growth 218 (2000) 353. P.G. Koutsoukos, G.H. Nancollas, Journal of Crystal Growth 53 (1981) 10. P.G. Koutsoukos, Z. Amjad, M.B. Tomson, G.H. Nancollas, Journal of the American Chemical Society 102 (5) (1980) 1553. Z. Amjad, Journal of Colloid and Interface Science 117 (1987) 98. L.B. Kong, J. Ma, F. Boey, Journal of Materials Science 37 (2002) 1131. J.V. Garcia-Ramos, P. Carmona, Journal of Crystal Growth 57 (1982) 336. J. Gomez-Morales, J. Torrent-Burgues, T. Boix, J. Fraile, R. Rodriguez-Clemente, Crystal Research Technology 36 (1) (2001) 15. E. Dalas, J.K. Kallitsis, P.G. Koutsoukos, Langmuir 7 (8) (1991) 1822. ¨ ner, O ¨ . Do ¨ ner, Journal of Crystal Growth 186 (1998) 427. M. O gan, G. O ¨ ner, J. Norwig, W.H. Meyer, G. Wegner, Chemistry of Materials 10 (2) (1998) 460. M. O ¨ ner, Polymers for Advanced Technologies, submitted for publication. E. Akyol, A. Bozkurt, M. O J. Zhang, A. Ebrahimpour, G. Nancollas, Journal of Colloid and Interface Science 152 (1) (1992) 132. Z. Amjad, Langmuir 3 (1987) 106. V.K. Sharma, M. Johnsson, J.D. Sallis, G.H. Nancollas, Langmuir 8 (1992) 676. Z. Amjad, Canadian Journal of Chemistry 67 (1989) 850. V. Garcia-Ramos, P. Carmona, A. Hidalgo, Journal of Colloid and Interface Science 83 (2) (1981) 479.
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
51
[52] A. Ebrahimpour, M. Johnsson, C.F. Richardson, G.H. Nancollas, Journal of Colloid and Interface Science 159 (1993) 158. [53] S.G. Rees, D.T.H. Wassell, R.P. Shellis, G. Embery, Biomaterials 25 (2004) 971. [54] N. Spanos, P.G. Klepetsanis, P.G. Koutsoukos, Journal of Colloid and Interface Science 236 (2001) 260. [55] Z. Amjad, Colloids and Surfaces 48 (1990) 95. [56] N. Kanzaki, K. Onuma, G. Treboux, S. Tsutsumi, A. Ito, Journal of Physical Chemistry B 104 (2000) 4189. ¨ ner is a professor of Chemical Engineering at Yildiz Technical University in Turkey. Mualla O She received a BS in chemical engineering from Istanbul Technical University in 1982 and an MS degree in chemical engineering from Bogazici University in 1985. She earned her Doctorate from Yildiz Technical University in 1989. She was engaged in post-doctoral research at Technical University of Clausthal and Technical University of Berlin in 1989. She spent one year in the Chemical Engineering Department at the University of Arizona as a visiting professor in 1990. In 1991 she was appointed Research Associate at the University of Arizona in the Arizona Materials Laboratories, where she worked on the development of polymers for controlling crystal growth. Prof. Oner spent about three months at Max-Planck Institute in Mainz to carry out research related to biomineralization in 1996. In 2005 she became Deputy Dean of the Faculty of Engineering at Yildiz Technical University. She was involved in a variety of research efforts including studies of coal liquefaction, degradation of polymers, thermophysical properties of polymers and biomineralization. Prof. Oner’s main current research effort is devoted to crystal growth and dissolution phenomena with applications in mineral scale formation such as, CaCO3, BaSO4, Ca3(PO4)2, CaSO4, and crystal growth and its inhibition related to biological systems such as the formation of hydroxyapatite from aqueous solution at conditions of sustained solution supersaturation. ¨ zlem Dogan is a Research Assistant of Chemical Engineering at Yildiz Technical University. O She received her MS and PhD in Chemical Engineering from Yildiz Technical University in 1997 and 2005. The subject of her PhD thesis concerned the controlling of nanosize hydroxyapatite crystallization by polyelectrolytes. Her research interests are separation processes, crystallization and polyelectrolytes.
Inhibitory effect of polyelectrolytes on crystallization kinetics of hydroxyapatite ¨ ner*, O ¨ . Do M. O gan Yildiz Technical University, Chemical Engineering Department, Davutpasa, 34210 I_stanbul, Turkey
Abstract Modelling of the biologic materials, well-organized multifunctional structures and systems found in nature has attracted the interest of scientists working in many scientific disciplines. A new and rapidly growing field of biomimetics has stimulated an increased focus on biological materials as the researchers attempt to mimic the features, characteristics and growth of these naturally-occurring materials. This review discusses the principal features of biomineralization in relation to the controlled crystallization of inorganic materials and biomimetic routes to the formation of nanometer hydroxyapatite particles. This approach can be compared with biologic mineralization and has the potential for providing much greater control of particle size and distribution than would conventional methods. The constant-composition method has been used to study the influence of polyelectrolytes on the kinetics of crystal growth of hydroxyapatite (HAP), the thermodynamically most stable calcium phosphate phase, on HAP seed crystals at pH 7.4 and 37 C. The results indicate that polyelectrolyte concentration and the larger number of negatively charged functional groups markedly affect the growth rate. The fit of the Langmuir adsorption model to the experimental data supports a mechanism of inhibition through molecular adsorption of polymers on the surface of growing crystals. This system may allow insights into biomineralization processes. Ó 2005 Elsevier Ltd. All rights reserved. PACS: 81.10.-h Keywords: Crystal growth inhibition; Hydroxyapatite precipitation; Habit modification; Polyelectrolytes; Seed crystal
* Corresponding author. ¨ ner). E-mail address: [email protected] (M. O 0960-8974/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.pcrysgrow.2005.08.002
40
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
1. Introduction Biological systems are a rich source of ideas for scientists concerned with the synthesis of new materials and processes. Plants and animals have evolved a vast diversity of structures through strategies that often are very different from those used by the engineer. These naturally fabricated materials are invariably composites and are assembled from readily available materials, usually in aqueous media, at ambient conditions with high degrees of orientation, specific morphologies, and to specific polymorphs selected [1e4]. The field of biomimetics is based on the assumption that through evolution, nature has solved an optimization problem. Nature uses very few materials to create a vast variety of life forms at largely ambient temperature and pressure in order to synthesize a diverse range of mechanical and structural properties. The same material is used in many different ways to meet vastly different needs as exemplified by collagen [5,6]. Biological minerals employ a limited group of generic materials comprising organic, inorganic crystals and amorphous phases. Oyster shells, cuttlefish bone, ivory and magnetic crystals in bacteria are just a few of the vast variety of biological minerals engineered by living creatures [7,8]. Biominerals are normally produced very slowly and present a limited range of substances. The major mineralized tissues such as bone, teeth and shells are dominated by calcium phosphate or calcium carbonate minerals; nevertheless, more than 60 different types of biominerals are now known in organisms ranging from bacteria to mammals and more are being discovered at a surprising rate [9e11]. Hydroxyapatite is the major inorganic component of bone and teeth whilst the shells of invertebrates consist of calcite and aragonite. Bone, more than any other biomineral, reflects the greatest distinction between an inorganic and bioorganic mineral. Bone is a multifunctional two-phase substance, comprising flexible collagen and brittle hydroxyapatite, which is approximately 50% stronger in compression than in tension. Bone can be thought as a living mineral since it undergoes continual growth, dissolution, repair and remodeling. It performs many different functions such as skeletal homeostasis and mineral homeostasis. The skeletal homeostasis governs the growth and maintenance of the skeleton. Mineral homeostasis is dependent upon the high content of mineral ions in the bones to provide a constant concentration of essential ions in plasma. The mineral acts as an important calcium store as well as a supply of calcium in case of high demands [12,13]. Biomimetic approaches based on an understanding of the biomineralization process are aimed at synthesizing nanoparticles, polymeremineral composites and templated crystals. Studies over the last few years lead to the following distinct characteristics of biomineralization [13]: (a) Biomineralization occurs in well-constructed compartments or microenvironments; (b) These compartments have the ability to stimulate the nucleation and growth of crystals of the required inorganic material at certain functional sites while effectively inhibiting the formation of other crystals; (c) The crystal size and shape are well defined and show little variation; and (d) Formation of the macroscopic structure is accomplished by the packaging of many such units. Organisms have been manipulating minerals for almost as long as life has existed on earth. They use these minerals to fulfill functions as diverse as detoxification, temporary storage of vital ions, gravity perception, light reflection, navigation in the earth’s magnetic field and, perhaps most familiar, the construction of mechanically sound skeletal materials [14]. They have evolved sophisticated and varied means of controlling mineralization processes. This control is assumed to be mediated by specific interactions between certain crystal planes and biological macromolecules that are most conveniently referred to as acidic macromolecules [15]. There are several types of macromolecules identified depending on the organism. These are proteins or glycoproteins that are rich in acidic amino acids, either sulphate groups in polysaccharides and carboxylic
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
41
acid groups on aspartic acid proteins found in mollusks, or phosphate groups on proteins found in bone or tooth [16e18]. Direct observations of tissues show that acidic macromolecules are located at the interface between the crystal and the extracellular matrix, whereas others can actually be found within crystals themselves. Microstructural control is exerted on all levels, from the molecular and nanometer scale to the overall three-dimensional structure. On the nanometer scale, mineral crystals are formed with one or all axes oriented, specific shapes and sizes, highdensity packing, and specific polymorphs. Many of the acidic macromolecules isolated from biomineral have demonstrated the ability to regulate mineral growth in vitro. These are the main reasons why they are generally thought to be involved in manipulation and critical for the regulation of biomineral formation. The organic matrix is formed extracellularly and consists of two components: an underlying collagen sheet and individually attached macromolecules [19]. The acidic macromolecules usually comprise much less than 1% of the mass of the tissue and are often only a relatively minor component of all the macromolecules that make up the so called organic matrix. The major components are more hydrophobic framework macromolecules such as collagen, chitin and silk-fibroin-like proteins. The acidic molecules are frequently sandwiched between the surfaces of the crystal and these framework macromolecules [20]. Nucleation and growth of crystals occur within matrices. The size and shape of the crystals are thought to be controlled by the size and shape of the organic matrix, which acts as a template for the mineral microstucture. Nucleation may be initiated by the adsorption of cations onto functional sites of acidic macromolecules, which promotes the formation of critical nuclei. Crystallite orientation may be controlled by specific molecular interactions, which causes the arrangement of solution ions in specific locations relative to organic sites. Simple considerations of ionic charge, size and packing and their effect on hydration and lattice energies are sufficient to explain the thermodynamic stability of these minerals in biological environments [21]. Despite a great deal of study, the effect of biological polymers on the nucleation and growth process has not become clear. Arguments have been put forward that their effectiveness is due to particular molecular weights, hydrophobic and hydrophilic regions, a close match between spacings of acid groups and spacings of cations on the crystal surface [22]. The biological synthesis of inorganic solids often yields materials of uniform size, unusual habit, organized texture and defined structure and composition under moderate conditions of supersaturation and temperature. An understanding of biological solid-state interactions would therefore be of immense value in structural biology and medicine (for example, in the pathological mineralization of bones and teeth and formation of kidney stones) crystal growth, colloidal and solid-state science (as in the prevention of industrial scaling and controlled synthesis of electronic, magnetic and catalytic devices) and materials and engineering technology (organized composites and ceramic precursors, and the interrelationships between microstructure and mechanical properties). It seems that new chemical synthetic methods can be developed to form materials with highly controlled microstuctures if biomineralization processes can be understood and imitated [23,24]. One particularly promising approach is to use tailor-made additives that modify crystal habits through specific interactions at the crystal faces. Many of the soluble macromolecules associated with biominerals are considered to act in this manner. For example, acidic glycoproteins extracted from the adult sea urchin tests have been shown to bind specifically to the 110 faces of synthetic calcite crystals [14]. Diphosphonates and polyphosphates containing molecules are well known as calcification inhibitors. The direct precipitation of HAP on HAP seed crystals has been studied in the presence of EHDP (ethane-1-hydroxy-1,1,1-diphosphonic acid) and phytic acid [25]. The marked inhibitory influence of the additives on the HAP crystallization
42
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
was observed at physiological pH. DDPE (1,2-dihydroxy-1,2-bis(dihydroxyphosphonyl)ethane) has been successfully used as inhibitors of the crystal growth of HAP [26]. In vivo, HAP grows in solution that contains many essential trace elements besides calcium and phosphorus. Inhibitory effects of magnesium and zinc on HAP crystal growth have been investigated by many research groups [26e28]. It was confirmed that magnesium and zinc reduced the growth rate of HAP [26]. The adsorption behavior of single amino acids and their inhibitory effect on the crystal growth of HAP are of great importance for the calcification process [29]. It was found that phosphorine has the strongest affinity for HAP compared with aspartic and glutamic acids and serine [29e32]. The importance of protein eCOOH side chains in mineralization has been tested by several investigators using HAP as a model system [33e38]. It was found that the crystal growth rates of HAP decreased markedly in the presence of both aspartic and glutamic acid [35]. Lysine, which is an amino acid with a basic side group, was also used for hydroxyapatite crystallization [34]. It was found that the crystal growth of HAP was inhibited by lysine and this inhibition was attributed to adsorption and further blocking of the active growth sites on the crystal surface. Poly-L-glutamate and poly-L-aspartate have been shown to inhibit HAP growth when present in solution but also act as HAP nucleators when adsorbed on germanium surfaces [30]. Other workers have found that the presence of ferrocene complexes inhibited the crystal growth of HAP [36]. Studies have also been conducted on the role of serine, tyrosine and hydroxyproline amino acids with polar side groups. Depending on the side group of the amino acid different inhibiting activities were observed [33]. The precipitation and dissolution of calcium phosphate salts is of particular interest because of its importance in industrial water systems, in waste water treatment processes, in agriculture as fertilizers and in biological calcification processes [39]. Under physiological conditions the most stable calcium phosphate phase is hydroxyapatite [Ca5(PO4)3OH, HAP] [40]. The growth mechanism of HAP has received considerable attention in view of its importance in understanding the mechanism of hard tissue calcification such as bone and teeth and in many undesirable cases of pathological mineralization of articular cartilage, dental caries and kidney stones [41e43]. One problem in nucleation and crystal growth, of considerable interest in both industrial and biological situations, concerns the way in which certain polyelectrolytes inhibit or delay crystallization from supersaturated solutions of inorganic salts. Although preferential adsorption of a growth inhibitor at the crystal surface is known to be an essential step in its specific performance, the precise nature of the various factors influencing this step and the role they play in the process are not yet fully disclosed [44,45]. As many biological molecules contain oxyanion functional groups (carboxylate, phosphate and sulphate esters) we have been involved in crystallization experiments using polyelectrolytes. In this work we have prepared a range of acrylic and methacrylic polymers with different architectures to explore their relative effectiveness in inhibiting crystal growth of hydroxyapatite. These polymers consist of a hydrophilic block, which promotes water dissolution (polyethylene glycol), and a second hydrophilic block (acrylic block), which strongly interacts with minerals.
2. Experimental procedure A series of acidic acrylate block copolymers have been made, by radical polymerization, with defined molecular weight and structure. Radical polymerization of acrylic acid (AA) was carried out in the presence of a-thio polyethylene glycol monomethylether as a chain transfer agent to produce poly (ethylene glycoleblockeacrylic acid) copolymers. PEG block length in the copolymers
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
43
was controlled by using three different molecular weight chain transfer agents (Mn: 350, 750 and 2000 g/mole). The complete experimental procedures were reported previously [46]. Crystal growth experiments were made in magnetically stirred water-jacketed cells of 1 L capacity with water thermostated at 37 G 0.1 C. Reagent grade and triply distilled water were used and crystallization experiments were conducted in a nitrogen atmosphere. Carbon dioxide was excluded from the solutions by bubbling with presaturated nitrogen gas. The HAP seed material prepared as described in the literature had a molar TCa/Tp ratio of 1.66 G 0.1. Their specific surface area (SSA) was found to be 33.1 G 0.1 m2/g, as determined by B.E.T method (Micromeritics FlowSorb II-2300). The decrease in SSA during the precipitation reactions from 33.1 to 26.9 m2/g was taken into consideration for calculation of reaction rates. The seed crystals were characterized by scanning electron microscopy (JEOL-FEGSEM), X-ray powder diffraction (SHIMADZU XRD-6000) and infrared spectroscopy (KBr pellet method, PerkinElmer Spectrum One). In a typical constant-composition crystal growth experiment, stable supersaturated solutions of calcium phosphate with a molar ratio Cat/Pt Z 1.67 were prepared in water-jacketed reactor. The total molar concentration of calcium, Cat was 5.00 ! 104 mole/L with calcium/phosphate molar ratio 1.67. It should be noted that this solution was undersaturated with respect to other calcium phosphate phases such as OCP (orthocalcium phosphate) and DCPD (dicalcium phosphate dihydrate) [38]. The pH was adjusted to the required value by the addition of dilute potassium hydroxide solution. Following the pH adjustment and verification of the stability of the supersaturated solution for at least 2 h, known weights of HAP seed crystals were added to the solution. As growth commenced, the release of protons lowered the pH of the solutions and a pH drop triggered the addition of titrant solutions, which was controlled by means of pHstat (Radiometer pH-Stat Titrator PHM290, Auto-Burette ABU901). The crystal growth reaction was monitored by the addition of titrant solutions as a function of time from mechanically coupled automatic burettes. Throughout the course of the seeded growth experiments, the pH of the working solution and the added volume of titrants as a function of time were recorded and stored in the computer for further analysis. The titrant solutions in the burettes consisted of calcium chloride, potassium phosphate, potassium hydroxide and the inhibitor. Identical volumes of these solutions were introduced by burettes and the titrant molar concentration ratio was adjusted systematically so as to match the expected stoichiometry of the precipitated phase after correction for dilution of the supersaturated working solution. The molar concentration ratio of the titrant corresponded to the stoichiometry of the HAP phase, Ca:P:OH Z 5:3:1. As a result the HAP crystallization took place under conditions of constant supersaturation. The supersaturated working solution contained sufficient concentration of potassium chloride (inert electrode) so as to maintain the ionic strength of the solution during titrant addition. The course of the reaction was followed by removing homogeneous aliquots at various times. These were quickly filtered through Millipore filters of 0.22 mm pore size and the crystals removed by filtration were examined by scanning electron microscopy (SEM), X-ray powder diffraction, FT-IR and specific surface area. The rates of crystallization were determined from the rates of addition of mixed titrants and corrected for surface area changes.
3. Results and discussion The grown needlelike crystals were identified as hydroxyapatite by XRD (Fig. 1) and compared with that of the ASTM (09-0432) standards (Table 1). Lattice parameters of the crystals
44
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
Fig. 1. X-ray powder diffraction of the hydroxyapatite seed crystals.
precipitated are a Z b Z 0.9417 nm and c Z 0.6879 nm. These values are very close to the theoretical values (a Z b Z 0.9418 nm and c Z 0.6884 nm). FT-IR spectra of the seed crystals also showed that they consisted of HAP (Fig. 2). The bands at 3570 cm1 are characteristic of stretching and vibrational modes of the OH ions. 1 1 Characteristic frequencies derived from PO3 4 modes can be seen at 1094 cm , 1035 cm 1 1 1 and 962 cm , between 600 and 560 cm . The IR peak at 875 cm characteristic of 2 HPO2 4 was observed for the crystal. CO3 ions were detected in the precipitate from the peaks at 1453 cm1 and 1415 cm1. Crystal growth experiments were performed at constant-composition conditions. The ability to measure quantitatively the rates of mineralization at constant thermodynamic driving forces is one of the most important advantages of the constant-composition kinetic methods [47]. Typical plots of moles of HAP grown on HAP seed crystals as a function of time, after correction of the raw data for the observed changes in specific surface area are shown in Fig. 3. The slopes of the lines are used to calculate the growth rates as moles of HAP per square meter of surface. Table 1 List of the peaks observed in the X-ray powder diffraction pattern of HAP and the respective data from ASTM file HAP seed crystals
ASTM (09-0432)
2q
˚) d (A
I/II
hkl
2q
˚) d (A
I/II
hkl
25.846 31.891 32.843 34.017 39.752 46.650 49.440 50.387 52.015 53.171
3.444 2.804 2.725 2.633 2.265 1.945 1.842 1.809 1.757 1.721
50 100 60 27 23 29 35 14 11 21
002 211 300 202 310 222 213 321 402 004
25.879 31.773 32.902 34.048 39.818 46.711 49.468 50.493 52.100 53.143
3.440 2.814 2.720 2.631 2.262 1.943 1.841 1.806 1.754 1.722
40 100 60 25 20 30 40 20 16 20
002 211 300 202 310 222 213 321 402 004
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
45
Fig. 2. Infrared spectra of the seed crystals (a) and crystals grown on HAP seed (b).
The experiments have been carried out with methacrylate and acrylate block and homopolymers in order to see the differences in inhibitor performance. The effect of polyelectrolytes on the crystallization of HAP, from calcium phosphate solutions, supersaturated with respect to HAP, was studied by a series of experiments summarized in Table 2. Plots of HAP grown on HAP seed crystals, as a function of time for the crystal growth experiments in the presence of homopolymers and copolymers are shown in Fig. 3 and rate data are summarized in Table 2. It can be seen from Table 2 that polymers at a concentration of 3 ppm in the supersaturated solution resulted in reduction in the rate of crystallization. There is no difference between homo and block copolymers PMAA in terms of inhibition effectiveness (Fig. 4). Under conditions used in the present work, PAA and EO-b-AA copolymers at concentration of 3 ppm retarded crystal growth of HAP to a greater degree than PMAA and EO-b-MAA copolymers (Fig. 4).
Added Titrant Volume/Surface Area (ml/m2)
45 Blank
40
PMAA
35
EO-b-AA PAA
30 25 20 15 10 5 0
0
50
100
150
200
250
300
350
Time (minute) Fig. 3. Crystal growth of HAP on HAP seed crystal. Titrant added as a function of time in the presence of polyelectrolytes.
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
46
Table 2 Effect of EO/AA and EO/MAA ratios of polymers on crystal growth of HAP at 37 C and 3 ppm polymer concentration Polymer
EO/AA
EO/MAA
Mn
HAP growth rate ! 108 (mole/m2 min)
Blank PMAA EO-b-MAA EO-b-AA PAA
e e e 1.31 e
e e 8 e e
e 8000 17,500 4500 5000
15.44 10.40 10.39 3.64 2.48
PMAA, poly(methacrylic acid); PAA, poly(acrylic acid); EO-b-MAA, ethylene glycol-b-methacrylic acid; EO-b-AA, ethylene glycol-b-acrylic acid.
Results of precipitation experiments conducted in the presence of varying concentrations of PAA, PMAA and EO-b-AA are summarized in Table 3 and shown in Fig. 5. It can be seen from Fig. 5 that an increase in polymer concentration results in a decrease of rate values. The decrease of rate values depends on the nature of the inhibitor and its effective concentration at a given condition [48e51]. It can be seen in Fig. 5 that the adsorption strength for polymers leveled off when the polyelectrolyte concentrations reached a certain value. Similar findings were reported where the equilibrium surface concentration of human serum albumin plateaued at about 1.6 mg/m2 of HAP [52]. The crystallization was strongly inhibited even at low surface concentrations of PAA. Rate values decrease from 3.39 ! 108 (mole/m2min) to 2.27 ! 108 (mole/m2min) when the polymer concentration increases from 1 ppm to 5 ppm. In the case of PMAA, the rate of HAP crystallization was reduced to 12.43 ! 108 (mole/m2min) at 1 ppm concentration and reached an inhibition value of 6.99 ! 108 (mole/m2min) at a concentration of 5 ppm. In previous studies, it had been found that there was a direct relationship between the fractional site coverage of an adsorbed species and the fractional reduction of the rate of crystal growth. This behavior led to the conclusion that the adsorption takes place on the sites where crystal growth occurs [53,54]. Since the amounts of polymer in solution were small, the inhibition could probably be attributed to the blocking of the active growth sites of the seed crystals rather than to the binding to calcium ions in the solution [55]. This assumption was tested by fitting the kinetic results in 16 14
3
PAA
2
4
EO/AA=1.31
6
EO/MAA=8
8
PMAA
10
Blank
R (mole/m2min)
12
2 0
1
4
5
Fig. 4. Comparison of the HAP growth rate by pairs of similar polymers with different architecture.
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
47
Table 3 Effect of polymer concentrations on growth rate of HAP Polymer
HAP growth rate ! 108 (mole/m2 min)
Concentration (ppm)
EO-b-AA EO/AA Z 1.93; Mn Z 3700
PMAA Mn Z 8000
PAA Mn Z 5000
0.5 2 5 10
8.57 5.14 2.65 2.52
1 3 5 7.5
12.43 10.40 6.99 5.41
1 2 3 5
3.39 2.52 2.48 2.27
a langmuir-type kinetic isotherm [56]. When a polyelectrolyte poisons active growth sites on the crystal faces, the coverage by polymers, q, for adsorption model is described by qZKCi =ð1CKCi Þ
ð1Þ
where K is the adsorption or affinity constant which is the ratio of the rate constants for adsorption and desorption, kads/kdes, and can be considered as a measure for the adsorption affinity of the polymer for the crystal surface and Ci is the total equilibrium concentration of the polymer. Depending on the surface coverage, q, the rate of HAP crystal growth in the presence of the polymeric additive, Ri, is given by
R (mole/m2min)
Ri ZR0 ð1 qÞ
ð2Þ
13 12 11 10 9 8 7 6 5 4 3 2 1
EO-b-AA (Mn=3700, EO/AA=1.93) PMAA (Mn=8000) PAA (Mn=5000)
0
1
2
3
4
5
6
7
8
9
10
11
Polymer Concentration (mole/l) Fig. 5. Reduction of the crystallization rate of HAP at constant solution supersaturation as a function of the polymer concentration.
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
48
where R0 is the crystallization rate in the absence of additive. Combination of Eqs. (1) and (2) gives R0 kdes 1 Z1C R0 Ri kads Ci
ð3Þ
Eq. (3) shows that this model predicts a linear relationship between R0/(R0Ri) and 1/Ci. The linearity of the plots of Eq. (3) for HAP crystal growth in the presence of polymers (Fig. 6) suggests that the inhibitory effect of polymers is due to adsorption at active growth sites. kads/kdes can be evaluated from the slope of the resulting straight line. The values of the affinity constant as calculated for PAA, EO-b-AA and PMAA are 36.50 ! 106, 6.71 ! 106 and 1.99 ! 106 L/mole, respectively. The high value of the affinity constant for PAA may reflect stronger equilibrium adsorption of PAA than of PMAA on the HAP surface. Photographs were also taken by scanning electron microscope (SEM) for the subsequent visual analysis in order to assess the effects of polymers on crystal shape and size (Fig. 7). The crystals precipitated at 37 C have the form of tiny needles with average length of 293 nm and width of 65 nm (Fig. 7a). Fig. 7bed shows the effect of polymers on crystal habit. No apparent changes in morphology were observed in the presence of polyelectrolytes. The average length of the crystals grown in the presence of the polymers was significantly less than that of the control sample. The average length of the crystals was reduced to 274 nm, 200 nm and 191 nm for solutions containing 3 ppm PMAA, EO-b-AA and PAA, respectively. The average value of the width in the samples containing polymers was similar to that of the control sample. 4. Conclusions All polymers tested in this study are effective as growth inhibitors in this experimental condition. The higher affinity of PAA for HAP corresponds to the more significant effect of this polymer on the rate of HAP crystal growth. The PAA chains are more flexible compared with those containing methyl group, thus increasing the interaction ability of the appended anionic groups with the solid phase and producing an overall three-dimensional arrangement most suitable for strong binding. The polymer backbone can then act as a fence on the crystal surface, thus forming an obstacle for the propagating steps. 6 PAA EO-b-AA (EO/AA=1.93) PMAA
5
Ro/(Ro-Ri)
4 3 2 1 0 0
1
2
3
4
5
6
7
8
9
1/C 10-6 (mole/l)-1 Fig. 6. Crystal growth of HAP on HAP seed crystals. Langmuir-type adsorption isotherm for the effect of polymers.
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
49
Fig. 7. SEM of HAP crystals grown for 5 h from solution; pH 7.4 and at 37 C. (a) Growth of HAP on HAP seed crystals. (b) Crystals grown from a solution containing 3 ppm poly (ethylene glycol-b-acrylic acid) block copolymer. (c) Crystals grown from a solution containing 3 ppm PAA. (d) Crystals grown from a solution containing 3 ppm PMAA.
All the results of the present investigation indicate that polyelectrolytes inhibit crystal growth and that such an inhibition is directly related to the fractional coverage of adsorption sites. From a physicochemical point of view, efficient regulators of calcification would be those molecules that have a high enough adsorption affinity so that low concentrations can result in the adsorption on and blocking of a significant number of crystal growth sites on the crystallites of the tissue. Acknowledgements We appreciate the support of YTUAF (Project No: 22-07-01-01); TUBITAK (Project No: TBAG-AY/236(101T098); TBAG-AY/250(101T174)) and DPT (Project No: 24-DPT-07-0401) for the accomplishment of this work. The authors thanks Prof. Dr. P.G. Koutsoukos of Patras University for valuable discussions and help in various phases of this work.
References [1] A. Berman, L. Addadi, S. Weiner, Nature 331 (1988) 546. [2] A. Berman, J. Hanson, L. Leiserowitz, T.F. Koetzle, S. Weiner, L. Addadi, Science 259 (1993) 776.
50 [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51]
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51 S. Mann, Nature 332 (1988) 119. S. Mann, Nature 365 (1993) 499. S. Mann, J.P. Hannington, R.J. Williams, Nature 324 (1986) 565. S. Mann, B.R. Heywood, S. Rajam, J.D. Birchall, Nature 334 (1988) 692. P. Calvert, Materials Research Society Bulletin 17 (1992) 37e40. P. Calvert, Chemistry of Materials 8 (1996) 1715. A.H. Heuer, D.J. Fink, V.J. Laraıa, J.L. Arias, P.D. Calvert, K. Kendall, G.L. Messing, J. Blackwell, P.C. Rieke, D.H. Thompson, A.P. Wheeler, A. Veis, A.I. Caplan, Science 255 (1992) 1098. B.R. Heywood, S. Mann, Advanced Materials 6 (1) (1994) 9. S. Mann, Nature 349 (1991) 285. P. Calvert, Nature 334 (1988) 651. A.H. Heuer, D.J. Fink, V.J. Laraia, Science 255 (1992) 1098. A. Berman, J. Hanson, L. Leiserowitz, T. Koetzle, S. Weiner, L. Addadi, Journal of Physical Chemistry 97 (19) (1993) 5162. S. Mann, B.R. Heywood, J.M. Didymus, S. Rajam, V.J. Wade, J.B. Walker, Materials Research Society Symposium Proceedings 174 (1990) 25. L. Addadi, A. Berman, J. Moradianoldak, Croatica Chemica Acta 63 (1990) 539. A. Berman, L. Addadi, A. Kvick, Science 250 (1990) 664. L. Addadi, A. Berman, J.M. Oldak, Connective Tissue Research 21 (1e4) (1989) 457. S. Weiner, A. Berman, L. Addadi, Ultramicroscopy 23 (2) (1987) 231. S. Mann, Nature 351 (1991) 24. R.J. Davey, S.N. Black, L.A. Bromley, D. Cottier, B. Dobbs, J.E. Rout, Nature 353 (1991) 549. T. Aoba, E.C. Moreno, Journal of Colloid and Interface Science 106 (1) (1985) 110. G.H. Nancollas, W. Wu, Journal of Crystal Growth 211 (2000) 137. S. Mann, D.D. Archibald, J.M. Didymus, T. Douglas, B.R. Heywood, F.C. Meldrum, N.J. Reeves, Science 261 (1993) 1286. P.G. Koutsoukos, Z. Amjad, G.H. Nancollas, Journal of Colloid and Interface Science 83 (1981) 599. M. Dalpi, E. Karayianni, P.G. Koutsoukos, Journal of the Chemical Society Faraday Transactions 89 (6) (1993) 965. T.A. Fuierer, M. LoRe, S.A. Puckett, G.H. Nancollas, Langmuir 10 (1994) 4721. Z. Amjad, P. Koutsoukos, G.H. Nancollas, Journal of Colloid and Interface Science 101 (1984) 250. N. Spanos, P.G. Koutsoukos, Journal of Materials Science 36 (2001) 573. A. Tsortos, G. Nancollas, Journal of Colloid and Interface Science 250 (2002) 159. E. Moreno, M. Kresak, A. Gaffar, Journal of Colloid and Interface Science 168 (1) (1994) 173. M. Kresak, E. Moreno, R. Zahradnik, Journal of Colloid and Interface Science 59 (2) (1977) 283. S. Koutsopoulos, E. Dalas, Journal of Crystal Growth 216 (2000) 443. S. Koutsopoulos, E. Dalas, Journal of Colloid and Interface Science 231 (2000) 207. S. Koutsopoulos, E. Dalas, Journal of Crystal Growth 217 (2000) 410. S. Koutsopoulos, E. Pierri, E. Dalas, N. Tzavellas, N. Klouras, Journal of Crystal Growth 218 (2000) 353. P.G. Koutsoukos, G.H. Nancollas, Journal of Crystal Growth 53 (1981) 10. P.G. Koutsoukos, Z. Amjad, M.B. Tomson, G.H. Nancollas, Journal of the American Chemical Society 102 (5) (1980) 1553. Z. Amjad, Journal of Colloid and Interface Science 117 (1987) 98. L.B. Kong, J. Ma, F. Boey, Journal of Materials Science 37 (2002) 1131. J.V. Garcia-Ramos, P. Carmona, Journal of Crystal Growth 57 (1982) 336. J. Gomez-Morales, J. Torrent-Burgues, T. Boix, J. Fraile, R. Rodriguez-Clemente, Crystal Research Technology 36 (1) (2001) 15. E. Dalas, J.K. Kallitsis, P.G. Koutsoukos, Langmuir 7 (8) (1991) 1822. ¨ ner, O ¨ . Do ¨ ner, Journal of Crystal Growth 186 (1998) 427. M. O gan, G. O ¨ ner, J. Norwig, W.H. Meyer, G. Wegner, Chemistry of Materials 10 (2) (1998) 460. M. O ¨ ner, Polymers for Advanced Technologies, submitted for publication. E. Akyol, A. Bozkurt, M. O J. Zhang, A. Ebrahimpour, G. Nancollas, Journal of Colloid and Interface Science 152 (1) (1992) 132. Z. Amjad, Langmuir 3 (1987) 106. V.K. Sharma, M. Johnsson, J.D. Sallis, G.H. Nancollas, Langmuir 8 (1992) 676. Z. Amjad, Canadian Journal of Chemistry 67 (1989) 850. V. Garcia-Ramos, P. Carmona, A. Hidalgo, Journal of Colloid and Interface Science 83 (2) (1981) 479.
M. O¨ner, O¨. Dogan / Progress in Crystal Growth and Characterization of Materials 50 (2005) 39e51
51
[52] A. Ebrahimpour, M. Johnsson, C.F. Richardson, G.H. Nancollas, Journal of Colloid and Interface Science 159 (1993) 158. [53] S.G. Rees, D.T.H. Wassell, R.P. Shellis, G. Embery, Biomaterials 25 (2004) 971. [54] N. Spanos, P.G. Klepetsanis, P.G. Koutsoukos, Journal of Colloid and Interface Science 236 (2001) 260. [55] Z. Amjad, Colloids and Surfaces 48 (1990) 95. [56] N. Kanzaki, K. Onuma, G. Treboux, S. Tsutsumi, A. Ito, Journal of Physical Chemistry B 104 (2000) 4189. ¨ ner is a professor of Chemical Engineering at Yildiz Technical University in Turkey. Mualla O She received a BS in chemical engineering from Istanbul Technical University in 1982 and an MS degree in chemical engineering from Bogazici University in 1985. She earned her Doctorate from Yildiz Technical University in 1989. She was engaged in post-doctoral research at Technical University of Clausthal and Technical University of Berlin in 1989. She spent one year in the Chemical Engineering Department at the University of Arizona as a visiting professor in 1990. In 1991 she was appointed Research Associate at the University of Arizona in the Arizona Materials Laboratories, where she worked on the development of polymers for controlling crystal growth. Prof. Oner spent about three months at Max-Planck Institute in Mainz to carry out research related to biomineralization in 1996. In 2005 she became Deputy Dean of the Faculty of Engineering at Yildiz Technical University. She was involved in a variety of research efforts including studies of coal liquefaction, degradation of polymers, thermophysical properties of polymers and biomineralization. Prof. Oner’s main current research effort is devoted to crystal growth and dissolution phenomena with applications in mineral scale formation such as, CaCO3, BaSO4, Ca3(PO4)2, CaSO4, and crystal growth and its inhibition related to biological systems such as the formation of hydroxyapatite from aqueous solution at conditions of sustained solution supersaturation. ¨ zlem Dogan is a Research Assistant of Chemical Engineering at Yildiz Technical University. O She received her MS and PhD in Chemical Engineering from Yildiz Technical University in 1997 and 2005. The subject of her PhD thesis concerned the controlling of nanosize hydroxyapatite crystallization by polyelectrolytes. Her research interests are separation processes, crystallization and polyelectrolytes.