Growth of strontium hydrogen phosphate/gelatin composites: a biomimetic approach

Recent research has focused on the crystal growth of strontium phosphates via different methods due to the bioactivity and biocompatibility of these materials with bone tissue. Here, we use a biomimetic method to synthesize strontium hydrogen phosphate/gelatin composites via single diffusion in gelatin. We compare the composite crystals with analytical-grade strontium hydrogen phosphate using infrared spectroscopy (IR), scanning electron microscopy (SEM), thermal gravimetric analysis, and X-ray diffraction. The IR spectra reveal that gelatin is incorporated into strontium hydrogen phosphate crystals due to CH2 and amide I vibrations. The SEM data exhibit arrow-like crystals and microspheres of strontium hydrogen phosphate with bundled nano-flakes and flower-like surface arrangements of crystals. Furthermore, we find that the thermal stability of the strontium hydrogen phosphate/gelatin composite crystals is higher than that of the analytical-grade strontium hydrogen phosphate. The ion concentration and the preparation method had only a small influence on the broadening of the X-ray Bragg reflections. Our study confirms that gelatin is a useful hydrogel for studying the crystallization of strontium hydrogen phosphate.


Introduction
2][3][4] SrHPO 4 (SrHP) as one of its main compounds has been considered to be an ion exchanger biomaterial for holding both hydrogen phosphate (HPO 4

2À
9][20][21][22][23][24][25] Taher et al. studied the hydrothermal synthesis of SrHP from a mixture of Li 2 CO 3 , SrCO 3 and H 3 PO 4 in water after conducting a thermal procedure at 180 1C for 6 h. 15More recently, Zhang et al. developed SrHP nano-belts without any surfactant as the template at 160 1C. 16It has been shown that the SrHP coating can be controlled on the cathode via the electrochemical reduction of alkaline earth metal-containing electrolytes. 17First introduced by Mooney et al., 18,19,21 growing solutions of SrHP has been the most popular approach in recent decades.This process consists of the slow evaporation of a solution containing Sr(NO 3 ) 2 and NH 4 H 2 PO 4 , which leads to the precipitation of SrHP. 24Furthermore, Roming and Feldman prepared nanoscale SrHP by rapidly mixing Sr(NO 3 ) 2 , diethylene glycol and H 3 PO 4 in water.These authors found that the rapid cooling of the process temperature from 170 to 15 1C resulted in the formation of a-SrHP and a continuous slow-heating process resulted in the appearance of b-SrHP. 25The only study about SrHP crystallization in gelatin was conducted by Banks et al. who focused on double diffusion in a gel.These authors, who found that polycrystalline spherulites of SrHP were formed in gelatin, did not discuss the morphology of the crystals or the influence of the gel on the crystallization. 26he question, therefore, arises of how the polar domains of gelatin influence the SrHP nucleation and morphology.In 2013, we aimed to study the effect of glutamic acid on the biomineralization of brushite in gelatin hydrogel.Interestingly, we found that glutamic acid capable of changing the morphology of brushite from spherulitic growth to mushroom-like crystals.Such structures were shown to be very similar to brushite kidney stones which play an important role in the study of crystal deposition diseases.8][29][30] In this regard, we reported the novel morphologies of barium hydrogen phosphate in gelatin for different metal phosphate crystals.As a result, the gel pH, the gel type, the concentration of ions as well as the method of crystallization strongly contributed to morphological changes of different inorganic-organic composite crystals.Our results strongly revealed that the morphology of crystals is altered and controlled using the gel growth method.
The mineralization of inorganic biocompounds in living organisms typically occurs in collagenous scaffolds. 31Here, we use gelatin as a model system for the gel growth of SrHP because it is an irreversibly hydrolyzed form of collagen, the primary organic compound found in bone.Gelatin contains peptides and proteins that are excellent biocompatible and biodegradable materials for bone tissue engineering applications.Furthermore, the gel diffusion method has been shown to be a straightforward, inexpensive and controllable model for the reproduction of biomineralization conditions that occur in nature. 32,33This method has been previously employed by several groups to gain new insights into the formation mechanisms of fluorapatite, 34 brushite, 27 octacalcium phosphate, 30 calcium carbonate and calcite. 35Hydrogels provide an effective medium with high cross-linking density for controlling the morphology of a crystal by adjusting the ratio of ion diffusion. 36On the other hand, SrHP is known as a transient in vivo phase to produce apatite.Strontium ions bear a chemical resemblance to calcium ions due to their charge-to-size ratio, and strontium ions play a special role in bone remodeling via apatite lattice distortion, osteoblast proliferation, osteoclast differentiation, bone strength promotion and alkaline phosphatase activation. 10,11These benefits arise because strontium has a larger ionic radius than calcium, which strongly suppresses the biological performance of strontiumcontaining cements. 12,13ere, we developed new types of SrHP crystals as key compounds in bone tissue engineering based on our previously stated approaches to calcium hydrogen phosphate and barium hydrogen phosphate.This supports the further understanding of how different parameters in biomineralization via the gel diffusion method affect the morphology of biomimetic composite materials.Therefore, this study may be useful for explaining the mechanisms of biomineralization of phosphate and strontium ions in vivo.

Synthesis of composites
We dissolved 1 g of pig skin gelatin (300 bloom, Sigma-Aldrich, USA) in 10 mL deionized water, and 7.5 cc of 0.5 M SrCl 2 solution (Sigma-Aldrich) was added to the dissolved gelatin.The solution was then stirred for 2 h at 50 1C.The pH of the mixture was changed using tris(hydroxymethyl)aminoethane/ HCl (Sigma-Aldrich) to adjust it to a value of 7.4.The mixture was then poured into the tube and left at room temperature for 24 h until gelation occurred.Next, we added 7.5 cc of 0.5 M Na 2 HPO 4 solution (Sigma-Aldrich) buffered at pH 7.4 to the top of the gelatin tube.The diffusion was allowed to take place at 25 1C for 1 month.At the same time, gelatin was mixed with a Na 2 HPO 4 solution to allow the gelation to come into contact with a buffered SrCl 2 solution.In the third single tube, a mixture of gelatin and 0.8 M Na 2 HPO 4 was placed in contact with buffered 1 M SrCl 2 .SrHP/gelatin composite crystals were produced inside the gel near the interface between the gel and the top liquid in all of the tubes.The samples were washed several times with hot water to remove the excess gel, and the experiments were repeated 10 times for each set of samples.

Characterization
We carried out the chemical characterization of the gelatin and SrHP crystals using a Perkin Elmer Spectrum One IR spectrometer integrated with an IBM personal computer.
We analyzed the surface morphology of the composite crystals by means of a Hitachi scanning electron microscope S-3000 N. The samples were covered with a layer of gold under vacuum conditions prior to the measurement.
We performed thermogravimetric analysis on a Mettler-Toledo TG/SDTA 851e analyser.In each case, a 5 mg sample was examined under an N 2 atmosphere at a heating rate of 5 1C min À1 from 30 to 600 1C.
We collected powder X-ray diffraction patterns at ambient temperature using a STOE StadiP diffractometer in the reflection Bragg-Brentano geometry (monochromatic Cu Ka 1 radiation, wavelength l = 0.15406 nm).The scattered intensity was detected using a linear position sensitive detector.

Chemical characterization
The infrared (IR) spectra of pure gelatin, SrHP (Sigma-Aldrich) and SrHP/gelatin composite crystals are shown in Fig. 1.In gelatin, evidence for amide I and amide A consists of two strong peaks at 1644 and 3423 cm À1 for a random coil and the a-helix conformation of gelatin, respectively. 37,38These bands are obvious based on CQO stretching coupled with COO À vibrations and NH stretching coupled with hydrogen bonding.The signature of amide II in gelatin is seen in a band at 1534 cm À1 for an out-ofphase combination of CN stretching and in-plane NH deformation modes of the peptide groups. 39On the other hand, the NH bending of amide III is evidenced by the peak at 1233 cm À1 .Moreover, all of the bands at 1384, 1400, 1437 and 2933 cm À1 appeared for CH 2 wagging and the asymmetrical CH 2 stretching of proline.We assign the bands at 1163, 1079 and 1030 cm À1 to the carbonyl and the skeletal vibrations of gelatin. 40,41nalytical-grade SrHP provided by Sigma-Aldrich (Fig. 1b) was used as a reference for comparison with different SrHP/ gelatin composite crystals obtained in this study (Fig. 1, panels c-e).Four intense bands at 3428, 2813, 2390 and 1719 cm À1 are related to H 2 O bending, hydrogen bonding, O-H stretching of the hydrogen phosphate anions and vibration of the hydrogen ions in the H + -PO 4 3À contacts, respectively. 28,30,42,43The bands located in the range 1400-1200 cm À1 are attributed to the in-plane P-O-H bending modes.The PO asymmetric and symmetric stretching modes in the PO 4 3À fragment of SrHP are associated with the three strong bands at 1131, 1077 and 991 cm À1 .The symmetric stretching of P-O(H) is represented by a strong band at 877 cm À1 .The bands in the range 600-400 cm À1 arise from the P-O anti-symmetric and symmetric bending modes in the PO 4 3À fragment. 25e IR spectrum for the crystals collected from tube 1 (Fig. 1c) revealed the appearance of two new peaks at 2919 and 2850 cm À1 after the crystallization of SrHP in gelatin and the intensity of the band at 3430 cm À1 increased.Furthermore, the peak at 1719 cm À1 (in Fig. 1b) was shifted to 1650 cm À1 (in Fig. 1c) and the band at 511 cm À1 disappeared compared with analytical grade SrHP.For the crystals produced in tube 2 (Fig. 1d), we observed several changes including the appearance of a new peak at 2850 cm À1 , shifting of a band from 1719 to 1666 cm À1 and disappearance of a peak at 511 cm À1 .On the other hand, the collected crystals from tube 3 (Fig. 1e) showed an increase in the intensity of the band at 3439 cm À1 ; the appearance of several new peaks at 2919, 2851, 1654, 1011, 972, 547, 542 and 529 cm À1 ; and the disappearance of a peak at 511 cm À1 as well as shifting of bands at 1719 and 877 cm À1 to 1736 and 844 cm À1 , respectively.
As previously stated, polypeptide chains generally produce several characteristic bands as amide A (about 3500 cm À1 ), amide B (about 3100 cm À1 ), amide I (1600-1700 cm À1 ), amide II (1550 cm À1 ) and amide III (1300-1350 cm À1 ).In other words, CH bonds can be vibrated at 2850-3100 cm À1 depending on the vibration mode.According to all the FTIR results, all the changes in the FTIR spectra of crystals associated with the above-mentioned regions are due to the fact that gelatin is entrapped within the crystal sheets.In this regard, we previously measured the gelatin content in barium hydrogen phosphate and calcium hydrogen phosphate crystals to be about 2-3 wt%.We also stated that the changes occurred in the FTIR spectra of composite crystals produced in biopolymeric gels depend on the ion concentration, gel pH and ion diffusion method.This observation is confirmed in the FTIR spectra of SrHP/gelatin composites where the type of ion firstly diffused and the ion concentration was different.
Note that we have previously demonstrated similar peak changes in amide I after the incorporation of gelatin into calcium hydrogen phosphate and barium hydrogen phosphate minerals.Such intermolecular interactions were due to hydrogen bonds and the coupling of positive ions with the gelatin carboxylate anions. 27,28The ionic strength of proteinous gels can be further enhanced via salts, which may lead to more rigid conformations of peptides and changes in the ordering of triple helices, as was found by Feng et al. 44 Furthermore, the intensity of the amide A bands and the CH 2 stretching of gelatin were not the same for the SrHP samples in different tubes.This result clearly demonstrates that the interaction between gelatin and ions is dependent on the type of ion diffusing in the gel and the ion concentration.

Microscopic observation of composite crystals
Fig. 2 illustrates the morphology of SrHP/gelatin composites obtained from the gel diffusion.We observed two different morphologies for the SrHP/gelatin composites: arrow-like and spherulitic crystals.In the first tube, arrow-like crystals approximately 60 mm in width and 300 mm in length were obtained.These crystals were composed of aligned bundles of single crystals, which functioned as the building blocks of a three-dimensional supercomposite crystal.We also observed highly porous microspheres in all of the tubes.The surface of the spheres in tube 1 (Fig. 2c and d) included bundled nano-flakes originating from a common center with a thickness of approximately 140 nm.It was obvious from tubes 2 and 3 that the crystalline structure of SrHP was similar to that of the spheres in tube 1.However, their surface morphology was different depending on the variation of the molarity of the ions and the type of ion diffused in the gelatin.It should be noted that the molarities of HPO 4 2À and Sr 2+ were similar for tubes 1 and 2 but different for the diffusion procedure in the gel.This situation resulted in changes in the crystallization and was responsible for the absence of arrow-like crystals in tube 2.Moreover, the molarity of ions in tube 3 was higher than that in tube 2, which possibly changed the surface arrangement, roughness and porosity of the spheres.The spheres in tube 2 were more uniform compared with those in tube 3, and twodimensional bundled sheets were arranged on the surface of spheres via self-assembly.As a result, spheres with flower-like surface arrangements were generated (Fig. 2, panels e-h).
We previously demonstrated that the crystal growth habits of barium hydrogen phosphate and calcium hydrogen phosphate crystals in gels depended on several factors including the electrostatic charge density, the topology of the binding surface, van der Waals forces, the surface distribution of biopolymers and the type and sequence of the functional groups. 27,28,45,46Researchers have found that gelatin macromolecular chains have randomly coiled and untangled single conformations in water at about 40-50 1C (i.e., their melting point region).These chains have a reduced energy level at lower temperatures.In this regard, three individual helices are organized into a clockwise triple-helix rich in glycineproline-hydroxyproline amino acids.Hydrogen bonds among the peptide groups play a pivotal role in the stabilization of this structure.It has been shown that the nucleation and growth of crystals in gelatin are influenced by the impregnation of calcium, barium and phosphate ions.In this regard, pre-ordering of the gelatin network is related to the type of the first ion impregnated in the gel prior to the diffusion of a second ion.Since the HPO 4 2À ion was mixed with gelatin in tubes 2 and 3, it interacted with the outer surface of the protein helices, which resulted in the bending of the triple-helices and the formation of hydrogen and electrostatic bondings. 47The Sr 2+ cations, which are diffused later, were able to react with the aggregated HPO 4 2À ions in gelatin to induce the crystallization of SrHP.This mechanism differs from the as-prepared mixture of gelatin and Sr 2+ , which possibly resulted in changes in the crystallization mechanism and morphology of the SrHP/gelatin composites.We conclude that the experimental conditions included the ion concentrations and the type of ion impregnated in the gel control, the reactant diffusion and the morphology of the SrHP crystals.

Thermal studies
Fig. 3 shows the thermal degradation of SrHP/gelatin composites and analytical-grade SrHP crystals.We observed three zones or regions of decomposition -initial, main and char -for all of the samples.In the first zone, the weight loss is the result of the removal of surface-adsorbed water molecules. 48The SrHP/gelatin composites lost about 3% of their weight in the first region, which was negligible for the analytical-grade SrHP crystals. 25ignificant weight loss for the analytical-grade SrHP crystals occurred at earlier temperatures of the second region between 310 and 350 1C due to the generation of Sr-O-P 2 O 5 , as stated by Sales et al.However, a higher thermal stability of the SrHP/ gelatin composites was observed by shifting the second zone to temperatures between 410 and 470 1C. 23This finding may be due to strong interactions between the inorganic and organic compounds.We observed that the degradation of composites started at 470 1C, which was most definitely higher than the  degradation level of 450 1C observed for analytical-grade SrHP pure crystals.We have previously shown the pyrolytic split between the polar domain of amine and carboxylic acids in gelatin on the one hand and the strontium and phosphate ions of SrHP on the other hand in composites of barium hydrogen phosphate/gelatin and calcium hydrogen phosphate/gelatin. 28urthermore, we observed that the char content was dissimilar for the SrHP/gelatin composites produced in different tubes at temperatures higher than 470 1C.This result further confirms the interaction between gelatin and SrHP, which is in agreement with the findings of the IR spectra.
We used pattern decomposition to investigate the influence of the amount of gelatin and the preparation method on the crystals' structure and the line broadening of the SrHP Bragg reflections.The monotonic increase in the intensity of the (120) reflection as a function of the amount of gelatin (Fig. 4) may well suggest a gelatin-induced preferential (textured) growth of the SrHP crystals with respect to the (120) planes.
In terms of the hypothesis of negligible micro-strain, we applied the Scherrer eqn (1) to quantitatively estimate the size of the coherent scattering domains (i.e., the crystallite size): where K is the shape constant (here K = 1), l is the X-ray wavelength (here l = 0.15406 nm, Cu K a radiation), b is the fullwidth at half maximum (FWHM, in radians) and y hkl is the Bragg angle of the diffraction line with Miller indices (hkl).
In Table 1, we relate G 002 to the mean crystallite size normal to the (002) plane; G 120 refers to the mean crystal size along a direction perpendicular to the (120) planes.
The difference in the crystallite size values in different directions suggests that the crystallites possess a needle or plate-like shape, which is in agreement with the scanning electron microscopy (SEM) results (see, for example, Fig. 2d, f  and h).The values of the lattice parameters did not change significantly as a function of the preparation method or the ion concentration (Table 2).

Summary and conclusion
In this work, we have investigated the influence of gelatin and the concentration of ions and their diffusion on the preparation of SrHP/gelatin biocomposite crystals using various analytical techniques.IR spectra revealed strong interactions among the gelatin functional groups and the strontium and phosphate ions, which promoted the formation of composites.The IR results also confirmed that the carboxylate, amine and hydroxyl groups of gelatin provided the crystal backbone precursors for the nucleation and growth of SrHP.As a result, various morphologies were produced such as micro-crystals, spheres and nano-flakes.The macromolecular conformation of double helices in gelatin in the presence of strontium cations or phosphate anions induced pre-structural disorder for SrHP deposition via the production of highly porous spheres.Our findings clearly show that gelatin can produce novel SrHP crystal morphologies depending on the first ion diffused and the ion concentrations.Such morphologies have not yet been reported elsewhere.Irrespective of the ion concentrations, the single-diffusion method leads to the formation of single-phase a-SrHPO 4 .Our SrHP-gelatin composites contribute to a better understanding of the mechanisms of mineralization of strontium and phosphate ions in bone tissue.However, additional studies  should focus on different factors such as the gel type and pH, the type of diffusion method, the duration of the diffusion process and the gel concentration.

Fig. 2
Fig. 2 SEM images of SrHP/gelatin composites: (a) arrow-like composites obtained in tube 1, (b) spherulitic composites obtained in tube 1,(c and d) the surface of spherulitic crystals obtained in tube 1 with higher magnification, (e) spherulitic composites obtained in tube 2 at low magnification, (f) spherulitic composites obtained in tube 2 at high magnification, (g) spherulitic composites obtained in tube 3 at low magnification, and (h) spherulitic composites obtained in tube 3 at high magnification.

Fig. 4 X
Fig. 4 X-ray diffraction patterns collected on specimen tubes 1-3 and on the reference sample (Sigma-Aldrich).

Table 1
Crystallite size for samples in tube 1-3 and a Sigma-Aldrich reference sample

Table 2
Lattice parameters for samples in tube 1-3 and a Sigma-Aldrich reference sample