Biogenic Hydroxyapatite Obtained from Bone Wastes Using CO2-Assisted Pyrolysis and Its Interaction with Glyphosate: A Computational and Experimental Study.

In this work, biogenic hydroxyapatite (BHap) obtained from cattle bone waste is proposed as an adsorbent of this dangerous pollutant. Density functional theory (DFT) and calorimetric studies were developed to study the interaction between BHap and glyphosate (GLY). A strong interaction was found in the experiments through the measurement of immersion enthalpy, confirmed by the exothermic chemisorption obtained with DFT calculations. These results suggest that hydroxyapatite is a promising adsorbent material for GLY adsorption in aqueous solutions. In addition, it was determined that the GLY-hydroxyapatite interaction is greater than the water-hydroxyapatite interaction, which favors the GLY adsorption into this material.

Introduction

Agricultural industry is one of the most important industries since it has made a major contribution to worldwide development. Meeting the global demand caused by the rapid population growth is one of the sustainable challenges faced by this industry as well as the efficient management of waste and residue.[1] Innovation and technological development of agriculture is essential to deal with high food demand, for example, designing and utilizing agrochemicals that increase the nutrients in soils and improve crop growth by the eradication of crop-damaging organisms such as insects, fungi, pathogen bacteria, and undesirable weeds. However, the uncontrolled and excessive usage of these agrochemicals has led to serious environmental pollution, aquatic ecosystem degradation, and poisoning of some organisms.[2]

For instance, GLY is a broad spectrum herbicide and one of the most-used active weed killer substances.[3] Recently, awareness of its potential hazardous side effects such as endocrine disruption,[4] oxidative stress, DNA damage,[5] dermatological and respiratory issues,[6] and probably being carcinogenic to humans[7] have been a central discussion around the world, to such an extent that some countries have banned its use.[8] Thus, it is important to develop technologies and strategies for the efficient removal of GLY from ecosystems to protect human health and restore the affected habitats by this compound.

Wastes and residues are produced in our everyday activities, and it is vital to properly manage them to mitigate their environmental impact. In the meat industry, one of the most abundant byproducts is bones, which have historically been used in animal feed and fertilizer production.[9] However, as it is a biological residue, some problems arise due to its inadequate biological stability, potential for rapid oxidation, high level of enzyme activity, and potentially pathogenic nature.[10] In fact, the food crisis that arose in Europe due the outbreak of bovine spongiform encephalopathy is a clear example of the pathogenic nature of this type of biological residues.[11] Therefore, other applications such as adsorbents’ synthesis have been used as an alternative for bone waste management.[12−15]

Bone is a complex multifunctional connective tissue that plays several highly specialized functions in animal bodies. It is considered a natural organic–inorganic nanocomposite, the organic phase (25–37% wt) of which consists in collagen, glycoproteins, proteoglycans, sialoproteins, and bone “gla” proteins.[16] In contrast, the inorganic or biomineral phase (65–70% wt) is principally composed of calcium apatite nanocrystals very similar to the geologically occurring hydroxyapatite (Hap) but with some important chemical and structural differences.[17] These characteristics make bone waste a striking precursor for low-cost material obtention.

Biogenic hydroxyapatite (BHap) is a nanomaterial as it is found as nanocrystals embedded in collagen fibrils. It has received much attention because of its potential in biotechnological and biomedical applications. Thus, Hap surface interaction with some biomolecules and biochemicals, such as amino acids, has been studied by some researchers, finding high affinity between these kind of compounds and the Hap surface. Because of the similarity between GLY and amino acid molecular structures, it was of our interest to study the interaction between GLY and the BHap surface, obtained from cattle bone wastes via CO2-assisted pyrolysis. The interaction was investigated by computational and experimental studies using density functional theory (DFT) calculations and immersion calorimetry, respectively.

Theoretical and Experimental Details

Density Functional Theory Calculations

To study the interaction between GLY and BHap, DFT, as implemented in the GAUSSIAN 09 package,[18] was used. The ground state of each system was obtained by calculating the total energy; to do that, the B3LYP[19] hybrid functional and 6-31g(d) basis set were employed. To model the interaction between GLY and BHap, the GLY molecule was set in different sites around the BHap structure, without any symmetry constraints. The total energy was calculated; that of the lowest energy was used to obtain the physicochemical properties. All the inputs were generated with GaussView 6.0. The absence of imaginary frequencies in the lowest total energy configuration was the criterion for ensuring stability. The structural parameters and the theoretical IR spectrum are also reported.

BHap Preparation

BHap was obtained from cattle bone waste, collected from a local butcher shop (Bogotá, Colombia), using the following methodology. First, bone wastes were subjected to hydrothermal treatment under controlled pressure and temperature to remove the marrow, soft tissues, fats, and fluids. They were then dried at 100 °C for 48 h. Size reduction was done by crushing bone wastes with a hand-drilling hammer and subsequent milling using a blade mill of local construction. The obtained powders were washed several times with warm distilled water, dried at 100 °C for 48 h, and sieved. Finally, the raw bone powder (RBP) with a particle size between 1000 and 2000 μm was subjected to pyrolysis under a partially oxidative atmosphere (CO2) using a tubular furnace Barnstead Thermolyne 21100 at a maximum temperature of 850 °C for 2 h, a heating rate of 5 °C·min–1, and a CO2 flow of 150 mL·min–1.

Physicochemical Characterization

Thermogravimetric analysis was performed using a Hitachi STA 7200 thermal analysis system under a constant CO2 flow of 150 mL min–1 and a heating rate of 5 °C·min–1 from 20 to 900 °C. Elemental analysis (C, H, N) was performed using a Thermo Scientific Flash 2000 elemental analyzer. Fourier transformed infrared (FT-IR) spectroscopy measurements were taken using a Shimadzu IRTracer-100 spectrometer equipped with a PKI Technologies attenuated total reflectance (ATR) accessory. Spectra were acquired between 400 and 4000 cm–1 with a resolution of 1.93 cm–1 and 100 scans to increase the spectra resolution. Raman spectroscopy measurements were taken with a Horiba Xplora Raman microscope using a 523 nm laser, 100× magnification, and 100% of laser power. X-ray powder diffraction (XRPD) was performed using a PANalytical X’Pert Pro MPD diffractometer with an X-ray source of Cu (λ = 1.5406 Å). Diffractograms were acquired in the 2θ range of 5–90° with a step size of 0.02°. Scanning electron microscopy (SEM) images were obtained using a TESCAN LYRA3 FIB-SEM microscope, equipped with a microanalysis system of energy-dispersive X-ray spectroscopy (EDS).

Determination of the point of zero charge (pHPZC) was done following the salt addition method with slight modifications.[20] First, a stock solution of NaCl 0.01 M was prepared. The initial pH0 was adjusted to values between 2 and 11 by the addition of some droplets of HCl o NaOH 0.10 M solutions. N2 gas was bubbled into each solution to remove the dissolved CO2 and avoid interferences in the pH measurements. Then, 25 mg of BHap was added to 15 mL of each solution with different pH0. After 72 h of constant shaking, solutions were filtered and final pHf was measured. pHPZC was determined by plotting the difference between pHf and pH0 (ΔpH) against pH0. The pH in which ΔpH equals zero corresponds to pHPZC. These measurements were performed using a Hanna Instruments HI 2211 pH-meter; the electrode was calibrated using standard buffer solutions of pH 4, 7, and 10.

The nitrogen physisorption isotherm at −196 °C was measured using a gas sorption analyzer Quantachrome Autosorb IQ2. Before analysis, 100 mg of BHap was outgassed under vacuum (10–5 mbar) using two heating steps: the first one up to 100 °C with a heating rate of 2 °C·min–1 (evacuation phase) and the second one from 80 to 200 °C with a heating rate of 5 °C·min–1, holding this temperature for 9 h (heating phase). Then, sorption analysis was carried out acquiring a 51-point isotherm, 30 of the adsorption branch, and 21 of the desorption branch. Specific surface area was determined applying the Brunauer–Emmett–Teller (BET) equation in the 0.05–0.3 range of relative pressure.

Immersion calorimetry was performed using a Calvet-type microcalorimeter of local construction.[21] A sample of 200 mg of BHap was introduced in a glass ampule with a filament at the end. Then, the stainless steel cell was filled with 10 mL of the immersion liquid, and the ampule was placed inside. The lid was set on the top of the calorimeter, and the electrical potential was measured every 2 s until the baseline was constant and uniform. Afterward, an ampule filament was broken with the bottom of the cell to immerse the solid into the liquid (solution) and measure the electric potential change that results of the solid–liquid interaction (thermal effect). After immersion, calibration was done by supplying an electrical power to the system [known potential (V), current intensity (I), and time (t)] and measuring the electric potential change to determine the calorimeter constant Kcal (eq ). Immersion enthalpies were calculated using the area under the peak (immersion energy) in each potentiogram as follows (eqs –3)

Immersion liquids were water, HCl at different concentrations, and solutions of GLY analytic reagent and commercial formulation. Each measurement was carried out in triplicate.

Results and Discussion

Computational Simulation Results

Structural Properties

Figure shows the optimized molecular structure of GLY and Hap; the initial structure for GLY was taken from Knuuttila et al.’s work[22] and for Hap from Ozturk et al.’s work.[23] In all cases, no imaginary frequencies were found, which confirm the system stability. Neutral charge (q = 0) and multiplicity (singlet) yielded the molecules to the ground state. Further studies were developed with these molecular structures.

Figure 1

(A) Glyphosate and (B) hydroxyapatite optimized structures.

The molecular structure of glyphosate (GLY) has the typical zig–zag form; Table S1 contains their ground-state intramolecular distances and angles, which is in good agreement with experimental and theoretical values.[22,24] The P–C bond has the value 1.822 Å, while the reported experimental and theoretical values are 1.823 and 1.827 Å, respectively. The C–H bonds range from 1.096 to 1.109 Å, while the N–H bond has the value 1.017 Å. Small deviations from experimental values were found in the phosphonic group. The obtained P11–O9 and P11–O10 bond lengths have very similar values (1.627 Å), which indicates a resonance between the two O atoms;[22] the experimental value is 1.500 Å. A larger discrepancy was found in the P11=O8 bond, with 1.627 Å being our result and 1.576 Å being the experiment value; some divergences between our results (0.972 Å) and the experimental value (0.711 Å) were also noted in the O–H bonds. This result may be due to the fact that the calculations were done in the isolated form of GLY,[22] and the interaction with neighboring molecules was neglected. Additionally, our calculated angle bonds also coincide with the reported values; as an example, the theoretical O–P–O angles range from 102.485 to 118° and the experimental values are in the range 106.0–118°.

Six phosphate groups, whose P–O bonds are in the range 1.53–1.63 Å, compose Hap; other DFT calculations obtained with this parameter are in the range 1.54–1.66 Å.[23] Our calculated Ca–O bonds (2.32–2.52 Å) and O–Ca bonds (2.28–2.50 Å) as well as O–H (0.97 Å) are also in good accordance with the reported values (presented in Table S2). The angles of the O–P–O bonds are reported in the range 102.05–113.26°, while our calculated values are in the range 102–120.2°; similar agreement was found in the O–Ca–O angle bonds (75.06–143.55° in the literature and 54.23–143.01° in our calculations).

Hydroxyapatite Interaction with GLY

The interaction between these two systems was studied placing the optimized structure of GLY on the optimized structure of Hap; this approach has been widely used to model the interaction of GLY with many other systems with successful results.[25−27] The complex was analyzed using the B3LYP functional with the 6-31g(d) basis set. The as-obtained system was optimized to the lowest energy structure (Figure ). The adsorption energy (Eads) of GLY on Hap, obtained with eq , has the value −2.672 eV, which is associated with exothermic chemisorption.where Eads is the energy of adsorption, EHap–GLY is the total energy of the Hap–GLY system, EHap is the Hap total energy, and EGLY is the GLY total energy.

Figure 2

Lowest energy configuration of the interaction of Hap–GLY.

The Hap–GLY interaction is developed between the Ca37 (Hap) atom and O60 of the carboxyl group (GLY), whose distance is 2.45 Å; this is the average value for the Ca–O bond in Hap. In the isolated form of Hap and GLY, Ca37 has the charge 1.12 e and O60 0.45 e; in the interaction, the charges of Ca and O atoms change to 1.07 and −0.66 e, respectively. This fact generated the attraction of GLY to Hap. Also, one hydrogen atom (H18) transferred from the carboxyl group to the phosphate group (H62). Besides this phenomenon, the absence of structural changes in Hap as well as in GLY after the interaction was noted.

The interaction between Hap and GLY was also analyzed by the infrared (IR) spectrum. Normal modes of individual molecules as well as the coupled system after optimization are shown in Figure . Our results of Hap and GLY are very close to data reported in the literature.[23,26] The main normal modes found in the range 500–1500 cm–1 of the Hap–GLY system are consistent with that found for the isolated Hap and GLY. The H atom that transferred from GLY to Hap formed an O–H stretching bond, characterized by the signal at 3202 cm–1. The interaction between Ca37 (Hap) and the atom O60 of the carboxyl group (GLY) is related to the 1704 and 1417 cm–1 vibrational modes. This result confirmed the existence of coupled vibrations in the Hap–GLY system, as predicted in the adsorption energy calculations. Finally, intense peaks at 2528 and 3003 cm–1 correspond to the O–H stretching bonds in the phosphonic group.

Figure 3

Infrared intensities calculated of Hap and GLY isolated and the Hap–GLY system.

Experimental Results

Physicochemical Characterization

Figure shows the ThermoGravimetric (TG) and Differential ThermoGravimetry (DTG) plots for RBP under a CO2 atmosphere. It is possible to note three distinctive weight loss zones. Zone I (20–200 °C) is attributed to loss of adsorbed volatile molecules and moisture, with a maximum mass loss of 6.2% and a maximum mass loss rate of −0.04%/°C reached at 92.5 °C. Thermal degradation of the organic structure composed of substances such as molecular chains of collagen, proteins, and fat tissue is presented between 200 and 576 °C (Zone II). In this region, the highest mass loss percentage (22.4%) is found, with also the highest mass loss rate at 331 °C. Finally, Zone III goes from 576 to 880 °C. The weight loss in this zone is owing to the partial dehydroxylation process of Hap and the decomposition of carbonates.[28] The mass loss in this temperature range is about 6.6% and the maximum rate of mass loss is located at 781 °C.

Figure 4

TGA-DTG of RBP under a CO2 atmosphere.

These results show the thermal behavior of RBP under a CO2 atmosphere and are useful for the proper selection of the pyrolysis conditions to obtain BHap. As mentioned previously, in the third zone, the mass loss is due to the decomposition of carbonates and partial dehydroxylation of Hap. It is worthwhile to mention that to obtain BHap with a structure and chemical composition similar to that of geological Hap, it is necessary to guarantee the complete elimination of the organic content from the RBP as well as the decomposition of carbonates. According to the analysis, 850 °C is an optimal pyrolysis temperature to obtain BHap under a partially oxidative atmosphere (CO2).

Pyrolysis yield was around 60–62%, which is in accordance with the final mass percentage calculated with the TGA. Samples of RPB are yellow-pale, and after thermal treatment, the typical white color of Hap is obtained, as observed in Figure . This color change is associated with the degradation of the organic matrix, leaving BHap as the product.

Figure 5

Photographs of (A) RBP and (B) BHap (photographs taken by the authors).

Elemental analysis of RBP shows that it has carbon, nitrogen, and hydrogen contents of 16.61, 6.90, and 2.75% wt, respectively, while in BHap, these percentages are reduced to 0.51, 0.39, and 0.01% wt, respectively. The decrease of these percentages is owing to the thermal decomposition of proteins presented in the RBP structure as well as the degradation of carbonates during thermal treatment under a CO2 atmosphere, showing that the resulting BHap corresponds to the inorganic matrix of RBP.

The FT-IR spectra are shown in Figure , where several differences are identified. The RBP spectrum shows a broad band between 3600 and 3000 cm–1, characteristic of the hydroxyl (−OH) functional group; bands between 1700 and 1400 cm–1 correspond to C=O, C=C, and C=N vibrations, all of them associated to collagen and other proteins. The inorganic phase of RBP is evidenced in the strong band between 1150 and 900 cm–1 that corresponds to phosphate (PO43–) vibrational modes, the band at 565 cm–1 corresponds to calcium and phosphate bonds, and the bands at 1300–1200 and 720 cm–1 are due to the presence of carbonate (CO32–).[29] The bands associated with the inorganic phase are also found in the BHap spectrum, with insignificant differences, confirming that Hap was generated successfully with the CO2-assisted pyrolysis. Some authors attribute the shoulder located at 1100 cm–1 to the degree of crystallinity of the Hap;[30] because of the absence of this shoulder in the BHap spectrum, there is an increase in the crystallinity of BHap after thermal treatment. The clear nonappearance of the bands associated with C=O, C=C, and C=N bonds (between 1700 and 1400 cm–1) in BHap confirms the thermal degradation of the organic matrix. The dehydroxylation is confirmed by the lack of the −OH band, while the reduction in the intensities of the CO32– bands indicates the thermal degradation of carbonates.

Figure 6

FT-IR spectra of RBP and BHap.

The Raman spectra of RBP and BHap are shown in Figure . The intense characteristic band at 960 cm–1 corresponds to the stretching ν1 of the phosphate group, and as can be seen, there is an increase in the intensity of this band in the BHap spectrum. This result demonstrates that the crystallinity indeed increases as proposed previously with FT-IR. On the other hand, the band located at 590 cm–1 is attributed to the ν4 stretching of the phosphate tetrahedra. The weak band located at 1070 cm–1 is assigned to the CO32– presence, which agrees with the band of the FT-IR spectra and with the literature.[31]

Figure 7

Raman spectra of RBP and BHap.

The BHap diffractogram is shown in Figure together with the reported Hap diffraction patter (Crystallographic Open Database COD#2300273). From the XRDPD patterns, it is possible to prove the presence of Hap in the BHap sample, presenting a characteristic hexagonal Bravais lattice with space group P63/m. The most intense diffraction peaks in the diffractogram are at two values of 25.8, 28.9, 31.7, 32.1, 32.9, 34.0, 39.8, 46.7, 49.4, and 50.5°, which are described by the following Miller indices (002), (021), (211), (300), (202), (310), (222), (312), and (213), respectively. It is worth noting that there is no peak shifting, indicating that the unit cell had neither shrunk nor expanded compared to Hap. The difference in the peaks’ relative intensities indicates a variation in the BHap’s atomic structure, probably by vacancies of some groups such as Ca2+, PO43–, and OH.

Figure 8

X-ray diffraction pattern of BHap and Hap (COD#2300273).

SEM micrographs of BHap are shown in Figure . The BHap surface is characterized by roughness and heterogeneity (Figure A); some cavities and fractures on its surface are also observed (Figure B,D). In some regions, disorganized laminar-shaped structures are found (Figure B), which come from the original bone matrix (RBP). A magnification of 40,000× (Figure C) permits one to observe in detail the micro–nano structure of BHap. As seen there, it is composed of nanosized particles of Hap that are attached to each other due to the coalescence process that takes place at the pyrolysis temperature used (850 °C).[32] It is possible to observe the grain boundaries between each nanoparticle and some voids or cavities formed between them. In Figure D,E a different BHap particle is observed, in general, with the same characteristics as the other one, indicating homogeneity in the sample. A special feature is highlighted in Figure E, revealing the presence of collagen fiber-like structures that remain after thermal treatment, clearly without organic composition, thus composed of BHap.

Figure 9

SEM micrographs of BHap with a backscattering electron detector and 10 kV. (A) Magnification of 500× (particle I), (B) magnification of 5000× (particle I), (C) magnification of 40,000× (particle I), (D) magnification of 1000× (particle II), and (E) magnification of 5000× (particle II).

The EDS analysis confirms the chemical composition of the obtained material. The elemental mapping (Figure ) shows that even though BHap has a heterogeneous and rough surface, its chemical composition is homogeneous, as it is observed that all the atoms are distributed uniformly over the surface. The EDS spectrum shows that O, Ca, P, C, and traces of Mg and Na are present in BHap, in a mass percentage of 41.69, 37.21, 16.45, 3.23, 0.85, and 0.57% wt, respectively. Mg and Na are natural ions commonly found in a bond matrix, and so they are present in BHap. The Ca/P ratio is a very important parameter commonly determined when Hap is synthesized because it is directly related to the stoichiometry of Hap. According to its molecular formula, the Ca/P atomic ratio must be 1.67. In this case, Ca and P have atomic percentages of 21.15 and 12.07%, respectively, which gives a Ca/P ratio of 1.75. This value shows that BHap is not stoichiometrically identical to Hap, probably to a smaller content of phosphate groups or a higher content of calcium, which explains some of the differences observed in XRPD patterns. Similar results have been reported for Hap obtained from other natural resources (i.e., fish scales, eggshells, sea shells, etc.).[33]

Figure 10

EDS mapping, EDS spectra, and element composition percentage.

Nitrogen physisorption isotherm at −196 °C of BHap is shown in Figure . According to the IUPAC technical report,[34] it is classified as a Type III isotherm, characteristic of systems with no porosity or macroporosity, with a weak interaction between N2 and the BHap surface. The BET specific surface area was determined to be 5 m2·g–1, which is consistent with the low surface areas reported for this type of materials.[35,36] Finally, it was determined that BHap presents a basic nature because of its high pHPZC of 9.7. The basicity of this material is due to the presence of groups such as phosphates (PO43–) and carbonates (CO32–). pHPZC is a very important property in surface processes because it describes the behavior of the surface charge according to the solution pH. When pH > pHPZC, the surface is negatively charged, and conversely, when pH < pHPZC, the surface has a positive charge.

Figure 11

Nitrogen adsorption–desorption isotherm at −196 °C.

With the results presented so far, it may be concluded that the pyrolysis of bone waste assisted with CO2 is an appropriate technique to obtain this type of material. An advantage of using CO2 as the pyrolysis atmosphere gas is that the organic matrix is totally degraded due to the gasification reactions promoted by CO2 at 850 °C. In contrast, pyrolysis under an inert atmosphere (N2) and under the same experimental conditions gives as the product a dark material, known as bone char, whose carbon content is in the range of 10–12%.[37] The gasification reaction that takes place in CO2 pyrolysis is known as the Boudouard reaction, described in the following reaction

The thermal degradation of carbonates is expected to happen at temperatures higher than 700 °C, and it is described by the following reaction

However, according to Le Chatelier’s principle, this reaction is not favored as pyrolysis is under a CO2 atmosphere. This is an advantage of this methodology in comparison to thermal treatment under oxidative atmosphere (calcination) because it avoids the formation of calcium oxide and, in general, other oxides such as magnesium oxide. In addition, as determined via FT-IR, in this case, partial degradation of carbonates happens under the pyrolysis conditions used.[38]

Immersion Calorimetry: Interaction of GLY and BHap

The interaction between GLY and BHap was investigated by means of immersion enthalpies, as a way of looking for potential adsorbents of GLY from water. Immersion calorimetry consists of measuring the heat transferred when a solid is immersed in a liquid, such as water, organic solvents, or any aqueous solution. The thermal effect is measured and correlated to the immersion enthalpy, which is a thermodynamic parameter that depends on the system under study. Principally, it depends on the textural properties and surface chemistry of the solid as well as on the nature of the immersion liquid.[39]

In general, immersion enthalpy is proportional to the surface area of the solid, which in our case is a very low surface area (5 m2·g–1). This was a limiting factor in our experiments, so we had to increase 4 times the mass of BHap in each experiment. However, in the case of immersion in water, no thermal effect was observed, indicating a weak interaction between water molecules and the BHap surface. It is well known that bone Hap must have a hydrophilic nature owing to the cell proliferation that must be guaranteed as it is a biological tissue.[40] In calorimetric studies of different samples of Hap immersed in water and n-decane, the hydrophilic character of the Hap was demonstrated.[41] This discrepancy is explained by the difference in the surface chemistry, structure, and morphology between the BHap obtained from bone waste and the synthetic Hap obtained from the hydrothermal reaction of calcium acetate and potassium hydrogen phosphate. Specifically, the surface area of the Hap obtained by Skwarek et al.[41] is almost 6 times higher than BHap’s surface area, expounding the lower or absence of thermal effect of the immersion of BHap in water.

The same behavior was observed when BHap was immersed in other solvents such as benzene, ethanol, and NaOH 0.1 M solution, which is also related to the low surface area, hydrophilicity, and basicity. By contrast, a thermal effect was observed when BHap was immersed in acidic solutions of HCl, as shown in Figure .

Figure 12

Potentiograms obtained from immersion of BHap in HCl solutions of different concentrations.

The thermal effect measured with the immersion of BHap in HCl solutions is not only due to the neutralization of surface basic groups but also the dissolution of BHap. Calcium apatites are not stable in acidic medium, and this process is influenced by factors such as the pH, concentration, saturation, chemical composition, solubility, particle size, surface defects, and the crystal phase, among others. Eight models have been reported in the literature, namely, diffusion and kinetically controlled, mono- and polynuclear, stoichiometric/non-stoichiometric dissolution, chemical model, self-inhibition, etch pit formation, ion exchange model, and hydrogen catalytic model. All of them have advantages and drawbacks, and their validity depends on the different assumptions and specific conditions in which they are based.[42] Recently, a generalized dissolution mechanism was proposed, in which the previous models were summarized and combined to explain completely the acid dissolution process.[42] Concisely, after Hap is placed into an acidic aqueous solution, sorption of water molecules and ions (H+ and Cl–) takes place; then, a chemical transformation occurs on the Hap surface, the detachment of lattice ions from a kink site, and finally diffusional steps away from the crystal toward the bulk solution.

The thermodynamics of acid dissolution have been studied by some authors, specially related to the kinetics and rate of dissolution of Hap.[43−45] In addition, an investigation about the enthalpies of solubilization of Hap was made by Ardhaoui et al.[46] They found that there is a proportional relation between the acid concentration and the measured enthalpy. It is worth noting that conditions used there differ from the conditions used in the present work. We observed no total dissolution of BHap due to the high amount of solid that was required to detect the thermal effect in a cell of 10 mL, and for that reason, results are not totally comparable. However, the same trend was obtained with an enthalpy of immersion (−ΔHimm) of 6.6 ± 0.9, 29.6 ± 1.7, and 41.4 ± 1.5 J·g–1 for the HCl solution of 0.1, 0.5, and 1.0 M, respectively. The tendency is caused by the less quantity of basic groups of BHap required to neutralize the H+ ions, while a higher concentration of HCl reacts with a higher quantity of basic groups, solubilizing more BHap.

On the other hand, immersions in 20 mM solutions of GLY were carried out, and the corresponding potentiogram is shown in Figure . Interestingly, a thermal effect was found unlike the immersion experiment in water, despite them being aqueous solutions. This means that there is an interaction between GLY and the BHap surface, strong enough to be measured and clearly stronger than that of water molecules and BHap. From the potentiogram (Figure ), the calculated −ΔHimm for each solution has average values of 8.3 ± 0.8 and 4.1 ± 1.1 J·g–1 for the solutions of commercial GLY and the analytical reagent, respectively. The −ΔHimm of commercial GLY solution is almost twice the enthalpic value of the analytical reagent system; this difference is associated with the composition of the commercial formulation, which contains other substances besides GLY, named as “inert” components. These substances are in greater proportion surfactants that promote the uptake of the herbicide by plants cells, such as polyoxyethylene tallowamine and metals such as As, Co, Cr, Ni, and Pb.[47] This co-formulant can interact with BHap during the immersion, resulting in a more exothermic process compared to analytical GLY solutions.

Figure 13

Potentiogram obtained from the immersion of commercial GLY and the analytical GLY reagent (20 mM).

With the obtained results so far, it may be concluded that BHap has potential applications in some areas like adsorption for water depuration, specifically in the adsorption of GLY. Another important feature is that the interaction with BHap and GLY is stronger than the interaction of this material with water, which is favorable in adsorption processes in the aqueous phase, because it is necessary to exchange the surface adsorbed water molecules with the adsorbate molecules (GLY) for adsorption to take place.

Other Haps have been obtained using different methodologies. For example, Hap obtained by the reaction of eggshell powder (CaCO3) with HNO3 at 800 °C, in conjunction with aqueous ammonia and H3PO4, is also effective for adsorbing GLY.[48] This Hap presented a BET specific surface area of almost 100 times higher (566 m2·g–1) than the one obtained in the present work. With this Hap, the authors obtained a promising GLY adsorption capacity of 890 mg·g–1.

Owing to the biological applications of Hap, its adsorption capacity has been studied using compounds such as amino acids, which motivated the present work due to their molecular structure similarity with GLY. In fact, GLY’s IUPAC name is N-(phosphonomethyl) glycine, being a derivative of the amino acid glycine. As an example, adsorption of glycine and glutamic acid was studied by means of molecular dynamics.[49] In that work, they found that at an atomistic level, amino acids are adsorbed over specific crystals facets of Hap, with their positive-charged amino groups on calcium vacant sites, while negatively charged carboxylate groups occupy phosphate vacancies. In addition, adsorption of l-lysine and dl-leucine was studied onto a poorly crystallin Hap obtained using the precipitation method.[50] The main conclusions were that the charged carboxylate and amino moieties are the groups responsible in the interaction with the Hap surface, being governed by electrostatic interactions. Finally, binding free energies and binding mechanisms of 20 natural amino acids were studied using dynamic simulations,[51] concluding that charged amino acids bind stronger than neutral amino acids. Also, it was confirmed that the affinity of this kind of molecules with the Hap surface depends directly on its ability to replace adsorbed water molecules.

Conclusions

BHap obtained from cattle bone waste was successfully obtained via CO2-assisted pyrolysis, and it was fully characterized as a calcium apatite very similar to Hap. We conclude that this is a promising material for the adsorption of GLY in aqueous solutions. It was demonstrated experimentally that there is an interaction between GLY molecules and BHap and that, indeed, it is stronger than the interaction with water molecules, which is vital for adsorption to take place. Complementarily, a computational study was carried out with the purpose of studying the interaction between BHap and GLY, finding that there is strong adsorption (chemisorption) between these compounds, which proves the high affinity of GLY and the BHap surface.

Here, we showed that BHap is a promising adsorbent for remediation of water polluted with GLY. The positive environmental impact of this work is evident in the fact that we propose the depuration of water sources by adsorption, utilizing and valorizing animal bone wastes as adsorbents. In this way, two different problems are faced at the same time: polluted water treatment and solid waste management.

For further works, it is necessary to study the adsorption of GLY into Hap by the experimental determination of the adsorption isotherms in the aqueous phase. In addition, as reviewed in literature, surface area is a critical parameter during adsorption, so modifications of BHap must be performed to increase the surface area, without losing the surface functional groups of BHap.