1 Introduction

Many pharmaceutical residues are hazardous in nature and intrinsically dangerous as well as persistent in the environment because they are poor substrates for indigenous microorganisms [1, 2]. They have been found worldwide in soil and water. Among them, antibiotics and hormones have been used in large quantities for several decades for human and veterinary medicine and as growth promoters in animal husbandry. The eleminination of these toxic species from wastewater is therefore of major importance [3,4,5].

Recently, several physico-chemical treatments have been proposed for efficient pharmaceutical removal, such as adsorption [6, 7], wet oxidation [8], ozonation [9], and heterogeneous photocatalysis [10, 11]. The simultaneous adsorption and heterogeneous photocatalysis have attracted special attention for the removal of micro-contaminants in the environment using low-cost raw materials. It is based on the activation of a semiconductor by light of those porous TiO2 is commonly used, while ZnO is considered an alternative material despite being non-porous [12]. TiO2 nanoparticles are an efficient photoactive material to reduce environmental pollution through to its high photo-catalytic activity. The catalytic efficiency of pure TiO2 is limited by a low rate of electron transfer to oxygen and a strong recombination of electron–hole pairs [13]. Association with various metal oxides shifts the activity of TiO2 towards the visible range and improves its photocatalytic activity [14]. Rare-earth-doped-TiO2 photocatalyzed 17α-methyltestosterone degradation in aqueous solution, with the best photocatalytic mineralization performance for Sm-doped TiO2 [15]. Compared with pure ZnO, rare-earth doped ZnO nanocomposite improved the photocatalytic degradation of tetracycline antibiotic under visible light and maintained high stability for catalyst regeneration [16]. Various xenobiotic organic compounds (XOCs) are also eliminated by photocatalytic using ZnO nanoparticles [17]. ZnO particles were tested for the effective piezo-photocatalytic degradation of the emerging contaminants testosterone and β-estradiol in water [18]. However, ZnO nanoparticles have a strong tendency to aggregate, which limits the photocatalytic degradation. ZnO modification like the association with a porous mineral happens to overcome this issue [19]. Although hydroxyapatite has good retention efficiency for the majority of organic pollutants, it is less effective as a photocatalyst although some empirical and statistical work describes the opposite [20,21,22]. In this context, the association of ZnO with porous hydroxyapatite (HAP) prepared from Moroccan natural phosphate (ZnHAP) at low cost can prove very effective in degrading certain pharmaceutical residues and trapping by-products resulting from photocatalytic degradation, understanding that these can be even more toxic than the parent drug.

A photocatalytic degradation of emerging antibiotic pollutants in water by TiO2/HAP [23] and ZnO-HAP [24, 25] nanocomposite materials was achieved where 40TiO2/HAP and 40ZnO/HAP are efficient catalysts for the enhanced adsorption and photocatalytic degradation of ciprofloxacin and ofloxacin. The relationship between structural, surface properties and chemical composition of nanocomposites with their photodegradation efficiency has been discussed. These nanocomposites were easily regenerated by heat treatment at 500°C without significant loss of catalytic efficiency. Therefore, the 40ZnO-HAP catalyst (called 40ZnHA) is efficient and environmentally friendly to remove various antibiotics from water. The extension to other pharmaceutical residues should make it possible to analyse the effectiveness of the 40ZnHAP nanocomposite to clarify y the different reactions involved between the pollutant and its catalyst cataltytic surface as well as the underlying mechanisms which occur within it. Hormones are also an important contaminants in water ressources,whose effects on health and the environment are now fully considered and estimated [26, 27]. In fact, their release into the environment without pre-treatment disrubts the endocrine systems, causing toxic effects for the aquatic environment [28,29,30]. This study develops and evaluates the 40ZnHAP nanocomposite prepared from natural phosphate in the presence of Zn2+ ions and evaluated its photodegradation efficiency against a steroid hormone 17α ethinylestradiol (named EE2). Kinetics of adsorption and photocatalytic degradation of EE2 as homonal model using three photocatalysts as ZnO, HAP, and 40ZnHAP composites.

2 Materials and Methods

2.1 Synthesis

The pure hydroxyapatite powder (HAP) was prepared by a dissolution/precipitation method from a natural phosphate coming from the Bengurir region (Morocco) as previously described elsewhere [31]. ZnO was prepared by precipitation of zinc nitrate dihydrate Zn(NO3)2.2H2O with a solution of ammonia in water (NH4OH, 25%) which is perfectly compatible with the synthesis conditions of HAP. The 40ZnHAP powder was obtained from the natural phosphate by a dissolution/reprecipitation method in the presence of Zn2+ ions as described elsewhere [24].

2.2 Techniques

The crystalline phases were identified using a powder X-ray diffractometer (XRD) (Philips PW131 diffractometer) with CuKα source (λ = 1.5406Ǻ) operating at 40kV and 30mA. The N2 adsorption–desorption isotherms for dried powders were obtained by multi-point N2-gas sorption experiments at 77 K using a Micromeritics ASAP 2010 instrument. The specific surface areas were calculated according to the Brunauer–Emmett–Teller (BET) method using adsorption data in the relative pressure range from 0.05 to 0.25 whereas the pore size and volume were estimated using the Barret-Joyner-Halenda (BJH) approximation. The sample powder was chemically analyzed by inductively coupled plasma (ICP-AES) emission spectroscopy (ICPS-7500, Shimadzu, Japan). The particle morphologies were determined by SEM/EDX on a JEOL/EO IP100, and TEM on a Tecnai G2 apparatus operating at 100 kV.

2.3 Adsorption and Photocatalytic Degradation Experiments

The adsorption kinetics were followed in batch experiments carried out at 23°C by adding 200 mg of each adsorbent (HAP, ZnO and 40ZnHAP) to 100 mL of water containing 10 mg L−1 of 17α-ethinylestradiol hormone (EE2); its characteristics are given in Table 1. The pH of the mixture is raised without adjustment (towards 6.5) and the suspension is kept under magnetic stirring. At selected time interval, samples of 5 ml were taken, filtered using a 0.45 μm membrane filter and the EE2 concentration in supernatant was analyzed using a UV–visible spectrophotometer (VWR UV-3100PC) at 365 nm. The amount of adsorbed hormone qt (mg g−1) was measured by the difference between the initial (C0) and instantaneous (Ct) concentration of the hormone molecule in the solution according to the equation \({{\text{q}}}_{{\text{t}}}= \frac{\left({{\text{C}}}_{0}-{{\text{C}}}_{{\text{t}}}\right)}{{\text{m}}}.{\text{V}}\), where m and V are the mass of adsorbent and the volume of solution respectively. The obtained kinetics data were fitted with the linear form of the Lagergren pseudo first-order equation \({\text{log}}\left({{\text{q}}}_{{\text{e}}}-{{\text{q}}}_{{\text{t}}}\right)={{\text{logq}}}_{{\text{e}},1}-\frac{{{\text{k}}}_{1}}{\mathrm{2,303}}\mathrm{ t}\), where qe,1 (mg g–1) is the amount at equilibrium of adsorbed hormone per gram of sorbent and k1 is the pseudo-first order rate constant. The linear form of the Lagergren pseudo-second-order equation \(\frac{{\text{t}}}{{{\text{q}}}_{{\text{t}}}}=\frac{1}{{{\text{k}}}_{2}{{{\text{q}}}_{{\text{e}},2}}^{2}}+\frac{1}{{{\text{q}}}_{{\text{e}},2}}{\text{t}}\), with qe,2 (mg g–1), the amount at equilibrium of adsorbed molecule per gram of sorbent and k2 the pseudo-second order rate constant.

Table 1 Physicochemical characteristics of 17α-ethinylestradiol hormone
Full size table

For photocatalysis experiments, the photoreactor used is an open-necked Pyrex cylinder with a capacity of 150 mL, giving the possibility of introducing solutes and taking samples. It is equipped with a HPK high pressure mercury vapor dipping lamp (Philips 125 W) and a cooling system through the circulation of water in a double-walled envelope equipped with a stirring system. The photocatalytic activity of three powders ZnO, HAP and 40ZnHAP heated at 500°C was evaluated under UV A-B-C irradiation of 125W (200–600 nm). Each 200 mg sample was initially dispersed in 100 ml EE2 solution (10 mg L−1) and left to react in the dark for 60 min to establish an adsorption/desorption equilibrium on he catalyst surface before UV-irradiation. Then, the solution was exposed to UV light under magnetic stirring. At selected time, the suspensions were centrifuged at 4000 rpm for 20 min, and the supernatants were kept in the dark before analysis.

3 Results and Discussion

3.1 Characterization Overview

The resulting powders of ZnO, HAP and their 40ZnHAP nanocomposite were characterized by XRD technique (Fig. 1). After calcined at 500°C, broad diffraction peaks of the hydroxyapatite structure were well identified for the pure HAP (JCPDS 09–0432) [33] while ZnO sample showed the main diffraction peaks of the zincite structure (JCPDS 36–1451) [34]. These two structures are found in the 40ZnHAP nanocomposite powder calcined at 500°C. No secondary phase such as zinc phosphates was detected.

Fig. 1
figure 1

XRD patterns of photocatalysts calcined at 500°C

Full size image

The textural properties of the prepared samples were investigated by SEM and TEM analsyses. Before being analyzed, the powders are deposited on an adhesive carbon membrane. SEM images of ZnO calcined at 500°C show only hexagonal aggregates but sometimes fused particles of different morphologies while the HAP and 40ZnHAP particles agglomerate and difficult by SEM analysis to determine the particle size of the materials studied. However, this analysis provides information on the homogeneity of the samples and, more particularly, on the dispersion of the ZnO in porous HAP network. Compared to ZnO and HAP, the EDX analysis confirmed the presence of Ca, P, O and Zn chemical elements in 40ZnHAP powder without taking into consideration the carbon peak from the carbon adhesive membrane.

TEM studies indicate that these powders consist of aggregated nanoparticles (Fig. 2). The pure HAP powder consists of agglomerates with particle sizes between 20–50 nm, while 40ZnHAP powder consists of visibly separated rod-sHAPed particles (10–20) with platelet morphologies of about 5 nm. Notably, 40ZnHAP exhibits a rather open structure with easily distinguishable individual particles and the aggregation density seems to increase especially visible at high magnification (scale bar = 20 nm). It should be noted that the association of ZnO with HAP reduces the size of the particles.

Fig. 2
figure 2

a, a’ SEM/EDX and (b) TEM images of calcined powders

Full size image

ICP-AES analyzes were performed to determine the amounts of Ca, P and Zn in the 40ZnHAP nanocomposite (Table 2). In 40ZnHAP, the experimental and theoretical values of the Zn/Ca molar ratio are the same and close to 0.48, which suggests that there is no loss of Ca2+ and Zn2+ ions during the preparation of the composite.

Table 2 Chemical analyses, Specific surface area (SBET) and Average pore size (Dp)
Full size table

The specific surface area of the powders was calculated from N2 sorption measurements for dried and calcined powders by applying the BET method (Table 2). After drying at 100°C, the 40ZnHAP nanocomposite has a specific surface area SBET of 250 m2g−1, greater than that of pure HAP (165 m2g−1) and much greater than that of ZnO (20 m2g−1) (Table 2). This may be related to the inhibition of grain crystal growth by the addition of Zn2+ ions to the HAP precipitation solution which promotes the creation of pores. Average pore size Dp was 11 for 40ZnHAP, slightly larger (14 nm) for HAP and significantly smaller (4 nm) for ZnO. After heat treatment at 500°C, all SBET values have significantly decreased due to crystalline grain growth, but the nanocomposite powder keeps good surface properties than HAP and ZnO.

The optical properties of the prepared powders were determined by diffuse reflection spectroscopy using the Wood-Tauc’s equation (αhν)2 vs (hν) [35] where α is absorption coefficient, Eg is the optical band gap of the material. Eg is identified by extrapolating the linear portion of the plots of (αhυ)2 against (hυ) to the energy axis. The calculated Eg values of ZnO and HAP powders were 3.2 eV and 5.3 eV respectively, while that of 40ZnHAP is around 3.7 eV such as published in our previous work [35]. Therefore, the composite presents an intermediate energy gap between ZnO and HAP.

3.2 Hormone Adsorption

To understand the movement of EE2 into and through the sites on the surface of ZnO, HAP and 40ZnHAP adsorbents, the adsorption kinetics were studied at fixed 10 ppm of EE2 in the dark. Compared to ZnO and HAP powders, 40ZnHAP showed a signficant adsorption capacity and the time to reach equilibrium was around 60 min for HAP and 40HAP and 90 min for ZnO (Fig. 3). To further analyze these results, two kinetic models, Lagergren pseudo-first-order and pseudo-second-order equations, were evaluated to fit the experimental data. The best fitting model was selected based on the values of the linear regression correlation coefficient. The correlation coefficients, R2, the rate ki, and sorption capacity, qe,i, are gathered in Table 3. The amount of adsorbed EE2 hormone does not exceed about 1.1 mg g−1 for the nanocomposite and it is lower for ZnO and HAP, which requires another process based on heterogeneous photocatalysis.

Fig. 3
figure 3

Adsorption kinetics of 17α-ethinylestradiol onto different photocatalysts. Error bars represent the standard deviation of triplicate experiments

Full size image
Table 3 Parameters obtained by simulating the sorption data of the hormone EE2 on HAP, ZnO and the 40ZnHAP nanocomposites according to the two kinetic models proposed
Full size table

Compared to the pseudo-second order fit, the first order model fits better the experimental data, especially for HAP, which evidences a physisorption mechanism. This indicates that the adsorption of EE2 on the adsorbents is controlled by external and internal diffusion processes followed by a surface reactivity between the surface functions of the adsorbent and the active groups of the hormone. Bulk porous diffusion, surface diffusion or both can influence adsorption kinetics depending on the porosity and composition of the adsorbent. Moreover, this diffusion is also strongly influenced by nanostructuring the adsorbents, with kinetics that depend on the nature of the Van de Waals interactions and the hydrogen bonds between solute/ and adsorbent. This reveals the kinetic differences between the three materials ZnO, HAP and 40ZnHAP composite that are characterized by a variation of porosity and oxygenated species on their surface, of which 40ZnHAP composite combines the best characteristics of the two materials. For this, the kinetic alteration observed shows that the mechanism of adsorption of EE2 on 40ZnHAP differs slightly from that on HAP and ZnO.

Overall, the adsorption process can be rationalized as follows: the hormone is first adsorbed on the surface of the apatite grains where interactions occur between EE2 and the apatite surface being the most porous and containing more the active sites. This step depends not only on the porosity of the material but also on the availability of oxygenated species at the adsorbent surface and the formulation of EE2 which also contains only two OH functions. At pH 6.5 close to that of the majority of wastewater which does not exceed pH = 6 due to the dilution phenomenon, the EE2 hormone retains its OH character in its structure. Therefore, adsorption is governed by the interactions between the basic sites on the 40ZnHAP surface or even HAP with the EE2 molecules through the oxygenated groups on the surface of the adsorbent by hydrogen bonds. This hypothesis is supported by previous reports on the physisorption of phenol and its derivatives on hydroxyapatites [34, 36].

3.3 Photocatalytic Degradation of the Hormone

The direct photolysis of the hormone in water under UV light was carried out before any photocatalytic experiment in order to evaluate its contribution to the degradation of EE2 under only UV irradiation without a catalyst (Fig. 4). Results show no change in residual EE2 concentration with illuminating time indicating that any significant EE2 degradation without catalyst was observed. In the dark, all the catalysts show a slight decrease in the initial EE2 concentration, which was in good agreement with their previously obtained adsorption capacities. The small difference is explained by the variation in porosity between the three catalysts.

Fig. 4
figure 4

Kinetics of photodegradation of 17α-ethinylestradiol hormone by the 40ZnHAP nanocomposite powder heated at 500°C compared to pure HAP and ZnO during a first 60 min period in the dark following by UV-irradiation

Full size image

Using catalyst and under UV light, the degradation kinetics of the EE2 hormone supported on 40ZnHAP calcined at 500°C was studied and compared with those of HAP and ZnO under the same operating conditions. A complete EE2 degradation was observed using 40ZnHAP and ZnO photocatalysts at 90 min and 120 min as lighting time, respectively. The best catalytic efficiency is obtained on the 40ZnHAP catalyst, but very closed to that of ZnO. The advantage of the 40ZnHAP composite is related to the HAP porosity added to the ZnO semiconductor, which undoubtedly makes it possible to trap the photoproducts if they are form after catalytic degradation. However, the low photocatalytic efficiency of HAP is related to its high gap energy (5.3 eV) compared to those of 40ZnHAP (3.7 eV) and ZnO (3.2 eV) [35]. Eventually, 40ZnHAP catalyst showed the highest abilities to degrade the EE2 molecules. This suggests that the hydroxyapatite associated with ZnO improves the surface of the formed nanocomposite and activates the fixation of the EE2 and subsequently its photocatalytic efficiency. This approach is very interesting because the use of natural calcium and phosphate precursors from natural phosphate with a small ZnO amount reduces the manufacturing cost of 40ZnHAP as a photocatalytic agent for the remediation of hormones and other pharmaceutical species. Its highly interconnected porous structure and the homogeneous dispersion of ZnO in the 40ZnHAP nanocomposite catalyst are favorable for instantaneous adsorption followed by effective photodegradation. Based on this observation, the nanocomposite catalyst can be an alternative to ZnO in several photocatalytic reactions.

The kinetics data were fitted with the linear form of the Langmuir–Hinshelwood equation: LnCt/C0 = -kappt, where C0 and Ct are the initial and instantaneous EE2 concentrations (mg L−1), respectively and kapp is the apparent pseudo-first-order rate constant (min−1) determined from slope of the linear plot Ln Ct/C0 against time t (Fig. 5). Based on the correlation coefficients (R2), this model is most likely to characterize the degradation kinetics of the target molecules supported on the catalysts studied, indicating that the reaction follows the first order very well. For this first order kinetic, t1/2 being a half-life time is calculated by the following relationship: t1/2 = Ln2/kapp providing a clear view of the persistence of this molecule in the environment. The rate constant kapp and t1/2 are gathered in Table 4.

Fig. 5
figure 5

Langmuir–Hinshelwood kinetic modeling curves of EE2 degradation by the synthesized nanocomposites compared to ZnO and HAP photocatalysts

Full size image
Table 4 Kinetic constants and half-life of the EE2 hormone in the presence of different catalysts
Full size table

Results show that the rate constant and the time t1/2 depend on the nature of the catalyst and it should be noted that the EE2 hormonal degradation kinetic over 40ZnHAP nanocomposite was of the same order of magnitude and are similar to that of ZnO. Indeed, a synergy between the transfer of the hormone towards the catalyst surface and its porosity accelerates the kinetics of degradation under UV light which further favors the collisions of EE2 molecules with 40ZnHAP particles in decreasing the required activation energy.

According to our previous experimental results and those published elsewhere [24], it was found that photogenerated holes and superoxide radicals are the main active species responsible for the photocatalytic degradation of organic molecules on 40ZnHAP and ZnO photocatalysts. According to the results published elsewhere, the photocatalytic degradation of some antibiotics supported on the wZnHAP nanocomposites was very efficient and no HPLC peaks were detected related to the by-products at the end of the degradation reaction. [24]. As already proposed, this is explained by the adhesion of small amounts of fragmentation by the porous surface of the apatite, whereas the ZnO catalyst is a non-porous material. Therefore, advanced water treatment processes under UV light can generally achieve the complete removal of various toxic organic materials with the adsorption of by-products by the porous apatite surface. As detailed in our previous study [35], recycling of the 40ZnHAP catalyst was carried out to evaluate its stability and ensure the conservation of its photocatalytic efficiency against degradation of the pharmaceutic pollutants after several cycles. After each photodegradation cycle the 40ZnHAP powder was washed with distilled water, and then dried in an oven overnight at 100 °C and heat treated at 500 °C for 3 h for the next degradation test. After four cycles, a 6% decrease in photocatalytic activity was recorded, and is linked to the loss of mass of the 40ZnHAP powder during washing, calcination, or even to the existence of by-products in the porous catalyst. The photocatalytic efficiency of 40ZnHAP was compared to other catalysts cited in the literature (Table 5). This nanocomposite is considered one of the most efficient photocatalysts, even for high hormonal concentrations, unlike other materials.

Table 5 Comparison of the photocatalytic efficiency of various catalysts towards EE2 degradation
Full size table

Under UV irradiation, the ZnHAP particles are at the origin of the energetic activation between the two bands CB ↔ VB, giving rise to very reactive species, which lead to the degradation of the hormone in the solution (Fig. 6). Moreover, •OH hydroxyl radicals cannot be produced directly by the photogenerated holes, but through the superoxides •O2− radicals which have a strong reducing power to react with H2O or OH− to obtain •OH [36, 45]. Based on the published studies and the discussions above, it is found that photogenerated holes and superoxide radicals are the main active species responsible for the photocatalytic degradation of the hormone on the 40ZnHAP photocatalyst. As has been demonstrated elsewhere on the degradation of antibiotics supported on 40ZnHAP, no toxic species were found by bacteriological analysis carried out on an aqueous solution after photodegradation and subsequently the discharged water is healthy and that it does not contain toxic by-products, which is consistent with published HPLC analysis [24]. The absence of by-products in the final solution is explained either by a complete mineralization or by the trapping of these products by the porous apatite surface after such photodegradation of the pharmaceutical substances. In summary, the steps of the photocatalysis mechanism are as follows: (i) formation of charge carriers by absorption of photons; (ii) recombination of charge carriers; (iii) trapping an electron from the conduction band; (iv) trapping of a valence band hole at a porous 40ZnHAP surface; (v) initiation of an oxidative process.

Fig. 6
figure 6

Schematic diagram of the photodegradation mechanism of EE2 by porous 40ZnHAP nanocomposite showing the process of excitation and transfer of charge carriers in the nanocomposite

Full size image

4 Conclusions

The removal of 17α-ethinylestradiol (EE2) from water as a hormonal model using an efficient 40ZnHAP nanocomposite was studied. The adsorption, photolysis and photocatalytic degradation of the hormone EE2 by the 40ZnHAP, ZnO and HAP catalysts were assessed using a multiple analysis approach. The main findings of the study are summarized as:

  • The high ZnO content in 40ZnHAP nanocomposite induces efficient photocatalytic activity. This is related to the small size and better dispersion of ZnO in the HAP network, inducing a good porosity in the nanocomposite.

  • Thanks to the gap energy of the 40ZnHAP nanocomposite close to that of ZnO, the electronic process occurs on the catalyst surface under UV photonic-radiation. The polluting particles are adsorbed on the catalyst surface and are then completely oxidized.

  • The adsorption and photocatalytic degradation of EE2 by ZnO, HAP and 40ZnHAP nanocomposite follows the first-order kinetic model. The quantity of EE2 adsorbed is low on all the catalyst surfaces whereas a complete degradation was achieved using 40ZnHAP and ZnO at 90 min and 110 min, respectively.

  • The association of ZnO with porous hydroxyapatite offers an innovative approach to improve photocatalytic performance for the degradation of organic contaminants, but also to capture the by-products formed after the end of the photocatalytic reaction.

Further studies on the photocatalytic oxidation of this and other hormones will be conducted under UV and natural sunlight conditions.