Adsorption of a Mixture of Daily Use Pharmaceuticals on Pristine and Aged Polypropylene Microplastics

Abstract

The main goal of this study is the examination of polypropylene (PP) microplastics (MPs) as possible carriers of daily use pharmaceutical compounds. The selected compounds can be separated into three groups: (i) antibiotics (Trimethoprim, Metronidazole, Indomethacin, Isoniazid), (ii) anti-inflammatories (Ketoprofen, Diclofenac), and (iii) anti-hypertensive (Valsartan). Two types of PP MPs (virgin and UV-aged) were used in the experimental procedure, and the effect of time and the effect of the initial concentrations of the drugs were examined. The impact of various environmental factors such as pH, salinity, and natural organic matter were also explored. The last two factors were studied using real aqueous matrices such as wastewater and seawater. According to the obtained results, the highest uptake was observed in indomethacin (9.3 mg/g) and diclofenac (7.3 mg/g), owing to their physiochemical properties. Aged particles showed enhanced adsorption ability in accordance with the existing literature, as their adsorption capacity was between 0.5–1.5 times greater than that of the virgin ones. Regarding the desorption of compounds from the virgin and aged PP MPs at three different pH values, diclofenac and indomethacin exhibited the highest desorption capacity, while alkaline conditions favored the desorption ability of PP MPs for most of the target compounds.

1. Introduction

Aquatic contamination arouses the concerns of the global scientific community, as various pollutants are detected in natural waters [1,2]. In such environments, a “contaminant” can be defined as any physical, chemical, biological, or radiological substance [2], such as plastics, pharmaceuticals, heavy metals, etc. [3]. Regarding plastic pollution, polymers are released in surface waters and they can break down into smaller particles owing to physical, chemical, or biological degradation [1]. Small particles with dimensions of $\le$5 mm can be defined as microplastics (MPs) [3]. MPs are referred to as emerging contaminants [4] owing to their omnipresence in water, air, and soil [5,6,7]. Various polymer types of MPs can be found in freshwater environments, with polypropylene (PP), polyethylene (PE), poly(ethylene terephthalate) (PET), polyvinyl chloride (PVC), polyamide (PA), and polystyrene (PS) representing approximately 80% of the annual polymers released in aquatic environments [8]. Among them, PP is the most frequently detected polymer (detection frequency >50%); owing to its low density, it is able to float at the surface of water matrices where different environmental interactions/processes can occur [5].

In the light of the above, the environmental behavior of PP is of great interest, and especially its interactions with co-existing contaminants in natural waters. Many studies have been published focusing on the adsorption behavior of PP MPs toward different organic pollutants, like pharmaceuticals, phenols, dyes, or antibacterial agents [9,10,11,12]. Most of these studies investigated the adsorption process under laboratory conditions, exploring various affecting factors such as: the aging of MPs by UV radiation [6], the presence of dissolved organic matter [13,14,15,16]—using fulvic or humic acid—and the salinity [13,17]. Moreover, crucial parameters of adsorption like pH, kinetics, isotherms, and MP size were also explored [13,18]. Despite these great research efforts, the existed adsorption studies restrict their study to one, two, or three target compounds, which are usually pharmaceuticals from the same category, such as antibiotics or anti-inflammatory drugs. However, the examination of a complex mixture of pharmaceuticals with different physicochemical characteristics is necessary since more complicated interactions may take place as organic micropollutants act competitively on each other in real matrices. Hence, increasing the number of examined drugs in adsorption studies is fully recommended.

To fulfil this gap in the existing literature, this study aimed to examine the adsorption process of seven daily use pharmaceutical compounds onto the surface of virgin and UV-aged PP MPs. Specifically, the target compounds belong to three different classes (antibiotics, anti-inflammatories, and antihypertensives), and they were selected based on their high detection frequency in natural waters. Particularly, the complex mixture of target compounds consisted of Trimethoprim (TMP), Metronidazole (MTZ), Indomethacin (IMC), Isoniazid (ISO), Diclofenac (DCF), Ketoprofen (KTF), and Valsartan (VLS). Among the selected drugs, ISO and IMC have never been examined before for their adsorption on PP MPs, while the rest of the target compounds have never been studied in a so complex a mixture up until now. Furthermore, crucial factors of adsorption were examined such as pH, contact time, and the initial concentrations of the drugs, while the PP MPs were aged through UV light irradiation for the evaluation of the aging effect. Moreover, the effect of the water matrix was also examined by conducting experiments in distilled water, in wastewater effluent, and in real seawater. Finally, desorption studies were also conducted in aqueous solvents with different pH values so as to investigate whether PP MPs can act not only as sinks of pharmaceutical compounds, but as sources too.

2. Materials and Methods

2.1. Chemicals and Targeted Compounds

All pharmaceutical compounds (

Table 1. The molecular formula, the chemical structure, and the physicochemical properties of the pharmaceuticals used.

Compound	Molecular Formula	Chemical Structure	M.W. (g/mol)	log Kow	pKa
ISO	C6H7N3O		137	−0.70	1.82
IMC	C19H16ClNO4		358	4.27	4.50
MTZ	C6H9N3O3		171	−0.02	2.38
TMP	C14H18N4O3		290	0.91	7.12
DCF	C14H11Cl2NO2		296	4.51	4.15
KPF	C16H14O3		254	3.12	4.45
VLS	C24H29N5O3		436	4.00	4.73

) were provided by Sigma-Aldrich (Steinheim, Germany). Stock drug solutions were prepared in methanol and stored at −20 °C. Methanol was used as an HPLC-grade eluent for the analytical methods (supplied by Merck (Darmstadt, Germany)). Acetic acid (purity > 98%) was purchased by Sigma-Aldrich and was used for the preparation of the phosphoric buffer used in the motivation phase of HPLC analytical method. Ultrapure water was used as a solvent for the adsorption experiments and HPLC analysis (obtained by a Mili-Q water purification system (Millipore, Temecula, CA, USA)). HCL (37% w/w), and NaOH (≥98%) were purchased by Sigma-Aldrich and were used for pH adjustment.

2.2. Aging of Polypropylene MPs

Polypropylene (PP) was purchased from Hellenic Petroleum (Athens, Greece) under the trade name ECOLEN^® HZ40 P with a melt flow rate of 12 g/10 min (ASTM D 1238). An amount of 50 g of PP was milled through a cutting rotor (Variable Speed Rotor Mill PULVERISETTE 14 premium line, FRITSCH GmbH Milling and Sizing Industriestrasse, Germany), and the resulting MPs were sieved through 200 and 75 μm sieves in order to achieve a size classification of 0.200–0.075 mm. After this step, 10 g of the milled MPs was settled into a glass petri dish. Virgin PP MPs were exposed under simulated sunlight (SSL) irradiation through an Atlas Suntest CPS+ (Germany) equipped with an air-cooled Xenon Lamp (1.5 kW), providing SSL with an irradiation intensity at 500 W/m². The MPs were placed at a distance of approximately 30 cm from the lamp and were stirred once per hour [19]. The SSL irradiation includes a calculated percentage of UV-B. Consequently, the microplastics (MPs) were exposed to UV-B irradiation for 30 h, which is equivalent to approximately 3 months of environmental exposure to solar radiation based on the following equation (Equation (1)) [20]:

$$t={\displaystyle \frac{Annual UV-B irradiance}{Intensity of lamps}}$$

(1)

where t is time measured in hours, the annual UV-B irradiance is 5% of the total European irradiance (1200 KWh/m²) [21], and the intensity of the lamps is equal to 500 W/m².

2.3. Batch Adsorption Experiments

The adsorption process was performed using glass jars in which V = 30 mL of each pharmaceutical solution was prepared in distilled water at an initial concentration (C₀) of 10 mg/L. Then, m = 30 mg of PP MPs was added. The experiments were carried out using a water bath stirrer in order to keep the temperature (T = 25 °C) and agitation (200 rpm) stable. Each adsorption experiment was carried out in triplicate. Samples of the treated solutions were filtered through nylon membrane filters (0.22 μm) prior to instrumental analysis.

2.3.1. Effect of Contact Time

Kinetic experiments were carried out to study the equilibrium time of the adsorption process. Following the experimental procedure described in Section 2.3, the treated solutions were analyzed at many time intervals ranging from 5 to 360 min in order to study the effect of time on the adsorption process. Each sample was filtered, and the remaining concentrations (C_e) of the target pollutants were determined. The obtained results were extrapolated using three fundamental kinetic equations: (i) the pseudo-1st order equation (Equation (2)) [22,23], (ii) the pseudo-2nd order equation (Equation (3)) [22,23], and the intraparticle diffusion model [10,24]. Briefly, the mathematical expression of the aforementioned models can be described as follows:

$$C_t=C_0-\left(C_0-C_e\right)\left(1-e^{-k_{1}t}\right)$$

(2)

$$C_t=C_0-\left(C_0-C_e\right)\left(1-{\displaystyle \frac{1}{1+k_{2}t}}\right)$$

(3)

$$q_t={k_i}^{0.5}+C$$

(4)

where C_t (mg/L), k₁ (min⁻¹), and k₂ (g/(mg×min)) are the concentrations of pollutants in the examined solutions at a certain time interval and the kinetic constants obtained from the pseudo-1st order and pseudo-2nd order equations, respectively. Concerning the interparticle diffusion model, k_i is the rate constant, C is the intercept, and q_t (mg/g) is the adsorption amount of the drugs at time t.

The fitting of the applied models was evaluated through the R² or the χ² values. The Chi-square (χ²) statistic test was based on the mathematical equation (Equation (5)):

$${\chi }^2={\displaystyle \sum \left(\frac{{\left(Q_e-Q_{em}\right)}^2}{Q_{em}}\right)}$$

(5)

2.3.2. Isotherms

For the isothermal experiments, the aqueous solutions were prepared at the same pH value as previously, while the C₀ of the pharmaceuticals ranged from 2–20 mg/L. Then, the samples were placed into a water bath for 360 min under constant agitation conditions (N = 200 rpm) at room temperature (T = 25 °C). After six hours, the supernatant was collected and filtered carefully through membrane filters in order to determine the concentration of the drugs and to measure the accuracy of the fit of an isotherm model to the experimental equilibrium data. The isotherm models used were Langmuir (Equation (6)) [11,25] and Freundlich (Equation (7)) [12,26]

$$Q_e={\displaystyle \frac{Q_{max}{K}_L{C}_e}{1+K_L{C}_e}}$$

(6)

$$Q_e=K_F{C_e}^{1/n}$$

(7)

where Q_max (mg/g) is the maximum adsorption capacity, K_L (L/mg) is the Langmuir constant of adsorption equilibrium, K_F (mg^1−(1/n) L^1/n g⁻¹) is the Freundlich constant of adsorption capacity, and n determines the intensity of adsorption (dimensionless), respectively. To estimate the Q_e of the pharmaceuticals, another equation (Equation (8)) was used:

$$Q_e={\displaystyle \frac{\left(C_0-C_e\right)·V}{m}}$$

(8)

where C₀ (mg/L) represents the initial concentration of the pharmaceutical compound; V (L) symbolizes the volume of aqueous solution; and m (g) is the mass of MPs.

2.3.3. Effect of pH

Aiming to explore the effect of pH, experiments were undertaken with initial pH adjustment employing micro-additions of HCl 0.1 M or NaOH 0.1 M. Slightly acidic and alkaline conditions were examined, such as 4, 6, and 8, since very low/high pH values are not possible in real aqueous samples. Then, the samples were kept in a bath at T = 25 °C with continuous agitation at N = 200 rpm for t = 6 h. At the end of the process, the samples were analyzed in order to detect the C_e of the selected drugs. The adsorbed quantity of the pharmaceuticals was calculated in removal percentage, using the following equation (Equation (9)) [27]:

$$R=\left({\displaystyle \frac{C_0-C_e}{C_0}}\right)·100\%$$

(9)

2.4. Aqueous Matrices

For the evaluation of the effect of the matrix on the adsorption process, wastewater (WW), and seawater (SW) were employed for the experiments. Wastewater was collected from the effluent stream of the Wastewater Treatment Plant in northern Greece, while seawater was collected from the port of Michaniona near Thessaloniki (Greece). A volume of 1 L of each aqueous matrix was filtered through GF (glass fiber) filters with a pore size of 1.6 μm in order to remove MPs or materials in the real samples that might adsorb a quantity of the selected drugs. The physiochemical properties of the employed matrices are presented in

Table 2. The physicochemical properties of aqueous matrices used in adsorption experiments.

Parameters	Distilled Water	Seawater	Wastewater Effluent
pH	5.9	8.3	6.9
Conductivity (μs/cm3)	35	83,200	1388
TOC (mg/L)	<2	4.718	10.72

.

2.5. Desorption Experimental Procedure

For the implementation of the desorption experimental procedure, aqueous solutions were prepared with various pH values (4, 6, and 8) to be used as solvents. Although organic solvents are widely used solvents in desorption processes, they were not selected since the adsorption and desorption processes were being studied under environmental conditions. Regarding the latter, the criteria pertaining to the selection of pH values are explained in Section 2.3.3. Initially, an adsorption step was carried out in which 30 mg of MPs was added to 30 mL of the aqueous solution of the selected pharmaceutical blend (C₀ = 10 mg/L). The pH value was adjusted to 6, while the samples were kept in a bath at T = 25 °C with continuous agitation at N = 200 rpm for t = 6 h. Then, the MPs particles were settled to dry for 24 h at room temperature inside a laboratory cabinet. Then, the dried MPs were added to glass jars with 30 mL of an aqueous desorption solvent under the same experimental conditions (T = 25 °C, N = 200 rpm, t = 6 h). After this step, the supernatant was removed and filtered through membrane filters (pore size 0.22 μm) and then analyzed by the UPLC-DAD system. The percentage of desorption for each compound was calculated based on the difference between the loaded amount of the selected drugs onto MP particles and the desorbed amount of the pharmaceuticals.

2.6. Analytical Methods

For the determination of the concentration of the target compounds, a Dionex Ultimate 3000 UPLC device (Thermo Fisher Scientific, Bremen, Germany) was used. The equipment was employed with a binary gradient system of pumps, a temperature-controlled autosampler, and the column compartment. Analysis was performed with a Thermo Hypersil GOLD aQ (50 mm × 2.1 mm, 1.9 μm) column while the temperature was maintained at 40 °C during the analysis. For the elution of the mobile phase consisting of ultra-pure water mixed with phosphoric buffer (0.1 mM) (A) and methanol (B), a gradient method was used at a flow rate of 0.2 mL/min, while the injection volume was set at 20 μL. The gradient elution started at 10% of (B) and was kept stable until 5.5 min. Then, (B) was increased to 45% until 13.5 min, and then up to 55% until 19 min. From 19 min to 20 min, (B) reached 70% and for the next minute only (B) flowed inside the column in order to clean up the system, since the retention time of the last pollutant was 18 min. In the last 4 min, the system returned to the initial conditions when total run time was 25 min. The detection of four out of the seven pollutants was realized at 221 nm, while MTZ, VLS, and TMP were detected at 320 nm, 256 nm, and 271 nm, respectively.

2.7. Characterizations

For the examination of the effect of aging on PP MPs, two characterization techniques were used. FTIR spectra of virgin and UV-aged PP MPs were acquired using a Perkin-Elmer FTIR spectrometer, model Spectrum One (Dresden, Germany) in absorbance mode and in the spectral region of 450–4000 cm⁻¹ using a resolution of 4 cm⁻¹ and 64 co-added scans. The samples under investigation were prepared as KBr tablets. Specifically, a portion of the microplastic (MP) samples was thoroughly mixed with KBr powder, and the tablets were then formed using a press instrument. These tablets were subsequently analyzed using FTIR instrumentation to obtain the spectra of the MPs. Then, the virgin and UV-aged PP MPs were explored through a scanning electron microscope (SEM), the JEOL JMS 7610 F (Freising, Germany), employed with an energy dispersive X–ray (EDX), the Oxford ISIS 300 micro-analytical system, providing information about the morphological changes of the MPs before and after UV exposure.

Moreover, the carbonyl index for the PP MPs was measured using the following Equation (10):

$$CI={\displaystyle \frac{Carbony Area under band 1850–1650 {\mathrm{cm}}^{-1}}{Area under band 1500–1420 {\mathrm{cm}}^{-1}}}$$

(10)

The CI was calculated in order to determine the degree of aging of the PP MPs. The methodology used is named the Specified Area Under Band (SAUB) calculation of CI, as described in previous studies [13].

3. Results and Discussion

3.1. Characterization and Aging Effect on MPs

3.1.1. FTIR

UV-light irradiation can change the chemical properties of polymers [28]. Herein, PP MPs were exposed for 30 h under UV-B irradiation in order to study the effect of the aforementioned exposure on polymeric chains. FTIR instrumentation was used to assess the the chemical changes in the PP MPs after the end of aging process, while the virgin particles were also measured as references. Generally, PP can be degraded by light irradiation through two main mechanisms. The first mechanism is called Norrish I, and it is related to the creation of carbonyl groups, while the second one (Norrish II) is associated with the formation of vinyl- and hydroxyl- hydroxyl peroxide species [15].

According to the obtained FTIR spectra from the PP MPs (Figure 1), it can be noticed that the differences and the intensity of peaks were increased in the region of 3300–3500 cm⁻¹, related to the stretching of the vibration of -OH bonds [15]. Moreover, alternations were observed in the region of 1600–1800 cm⁻¹, caused by the formation of new groups of esters, and/or ketones [15]. Furthermore, the increase in the absorbance intensities of the peaks that appeared in region 1200–1400 cm⁻¹ is related to the oxidation of PP functional groups creating more oxygen functional groups, for instance (C–O to C=O) [16]. Finally, calculations of the carbonyl index (CI) of the virgin and UV-aged PP MPs revealed an increase in the CI values for the aged particles (CI_UV-AGED = 0.3 > CI_VIRGIN = 0.2), since the aging process formed more oxygen groups on the surface of the PP MPs [17].

3.1.2. SEM

Concerning SEM analysis, micrographs were taken from the surface of the virgin and aged PP particles in order to observe the changes caused by UV-light exposure. The observed morphological alternations are presented in Figure 2. Specifically, the surface of pristine particles was initially smooth and wrinkleless. Conversely, several holes and defects can be noticed on the surface of the aged PP MPs. Furthermore, the surface of the aged particles was rougher compared to the virgin MPs. These findings are related to the decomposition of polymers that occurred during the aging process and the photo-oxidation mechanism, as described in the Section 3.1.1. Furthermore, an increase in the specific surface area (SSA) and porosity has been recorded in the aged particles after SEM analysis [15]. The observed changes on the polymer’s surface can affect its ability to adsorb organic compounds [16,29]. According to the existing literature, the increase of the SSA [30,31], holes, and porosity of MPs can cause an increase of their adsorption capacity [32,33]. The latter will be described more thoroughly in the following sections, in which the adsorption ability of pristine and aged particles will be further compared.

3.2. Sorption Kinetics

Kinetic studies were performed to evaluate the time required for adsorption of the selected pharmaceuticals on virgin and UV-aged PP MPs. The initial concentration of the drugs, for all kinetic experiments, was set at 10 mg/L, while pseudo-first order, and pseudo-second order models were employed for the evaluation of the obtained data results as illustrated in Figure 3. The calculated kinetic parameters are presented in

Table 3. The kinetics parameters for the adsorption of pharmaceuticals onto PP MP surfaces (time = 360 min, T = 25 °C).

		Pseudo-First Order			Pseudo-Second Order
MPs	Pharmaceuticals	k1 min−1	R2	X2	k2 g mg−1 min−1	R2	X2
Virgin PP	TMP	0.028	0.979	0.060	0.053	0.966	0.099
	MTZ	0.034	0.991	0.022	0.070	0.965	0.090
	IMC	0.026	0.968	2.708	0.043	0.988	1.581
	ISO	0.037	0.971	0.088	0.063	0.972	0.074
UV-aged PP	TMP	0.037	0.994	0.021	0.064	0.973	0.083
	MTZ	0.026	0.995	0.014	0.043	0.969	0.096
	IMC	0.022	0.990	0.098	0.047	0.993	0.071
	ISO	0.016	0.968	0.119	0.026	0.995	0.014
Virgin PP	DCF	0.025	0.958	0.701	0.048	0.986	0.233
	KTF	0.031	0.956	0.142	0.063	0.987	0.046
	VLS	0.018	0.959	0.515	0.035	0.980	0.248
UV-aged PP	DCF	0.026	0.994	1.290	0.052	0.983	3.736
	KTF	0.035	0.985	0.551	0.066	0.991	0.441
	VLS	0.030	0.992	0.544	0.057	0.977	1.727

. The aforementioned models were used to investigate the rate limiting step of the adsorption of the target compounds on PP MPs. The first model explains the adsorption as a physical process, while the second one explains it as a chemical process [34]. The evaluation of data fitting in the applied kinetic models was conducted using the correlation coefficient parameter (R²) and the Chi-square parameter (χ²).

Concerning the first group of pharmaceuticals, the adsorption process can be separated into three subsequent stages: (i) the initial intense/rapid adsorption stage (first 45 min), (ii) the slow/gradual adsorption stage (45–180 min), and (iii) the equilibrium/plateau stage (180–360 min). The rapid reaction in the first stage can be explained by the mass transfer forces and the available adsorption sites existing on the surface of PP MPs. When the drug molecules (liquid phase) come in contact with the solid surface of the adsorbents, instant strong forces are developed among the functional groups, and mass transfer begins (the mass transfer force is caused by the difference in the concentration of drug molecules in the microplastic phase and the water phase). Then, the adsorption rate falls as the adsorption sites are being covered, until the equilibrium phase is reached (slow diffusion of drug molecules in the microplastic particles). In the latter two stages, the drug concentration difference existing in the solid (PP) and the adsorbate (solution) decreases, and the number of adsorption sites on the surface of MPs are cut down, leading to the gradual weakening of the interaction between PP and the drugs. Similar findings were observed by Ciu et al. [22]. The values of C_e measured for TMP, MTZ, IMC, and ISO at the end of process using virgin PP MPs were: 8.52 mg/L, 8.9 mg/L, 3.5 mg/L, and 8.7 mg/L, respectively, while for the UV-aged PP MPs the C_e values were slightly reduced: 8.3 mg/L, 8.5 mg/L, 3 mg/L, and 8.6 mg/L, indicating higher adsorption of the drugs on the aged MPs. Regarding the experimental data fitting to the applied kinetic models, the correlation coefficient (R²) values were sufficiently high and the χ-square values low (χ²), (0.968 ≤ R²ps1 ≤ 0.995; 0.965 ≤ R²ps2 ≤ 0.995 and 0.014 ≤ χ²ps1 ≤ 2.708; 0.014 ≤ χ²ps2 ≤ 1.581), indicating that the selection of these models to describe the adsorption kinetics for the selected antibiotics was absolutely correct [27]. More specifically, for TMP and MTZ the pseudo-first kinetic model presents better fitting, revealing that their adsorption occurs as a more physical process, while for the other two drugs the pseudo-second kinetic model appears to be optimum, suggesting that chemical forces dominate the process (chemisorption) [35]. Finally, although an increase of the adsorbed amount of antibiotics is observed with aging, the adsorption rate constants do not proportionally increase, suggesting antagonistic behavior among the target compounds competing for the same (residual) available binding sites.

For the other group of pharmaceuticals (DCF, KTF, VLS), the adsorption process proceeds through a sharp decrease in their concentration within the first 45 min followed by a slower adsorption rate until 180 min, when the equilibrium point is reached for all compounds. The measured C_e values at the end of the process—using Virgin PP MPs—for DCF, KTF, VLS were: 6.0 mg/L, 8.4 mg/L, and 7.0 mg/L, respectively, while the aged particles showed better adsorption performance, as the C_e values detected were: 4.3 mg/L, 7.3 mg/L, and 6.1 mg/L. Comparison between the applied kinetic models revealed that pseudo-second order exhibits better data fitting for virgin PP MPs (0.956 ≤ R²ps1 ≤ 0.959; 0.980 ≤ R²ps2 ≤ 0.987 and 0.142 ≤ χ²ps1 ≤ 0.701; 0.046 ≤ χ²ps2 ≤ 0.248), whereas for UV-aged MPs the pseudo-first kinetic model outperforms the pseudo-second for DCF and VLS (0.992 ≤ R²ps1 ≤ 0.994; 0.977 ≤ R²ps2 ≤ 0.983 and 0.544 ≤χ²ps1 ≤ 1.290; 1.727 ≤ χ²ps2 ≤ 3.736).

Moreover, comparing the amount of pharmaceuticals adsorbed on virgin and aged PP MPs, UV-aged MPs showed better adsorption capacity since the aging process increased their specific area. It is well-known that aging increases the oxygen content of the surface of MPs; the latter was confirmed by the study of Liu et al. [36], based on which the increase of oxygen content is more evident in summer, indicating that the functional groups on the surface of aged PPs were oxidized to form more oxygen-containing functional groups, especially for summer-aged PPs. So, in the present study, the highest oxygen content on the surface of aged particles induced the ability of hydrogen bonding in MPs, leading to the increased adsorption of selected drugs as well [37].

In order to have an insight into the adsorption mechanism and to deeply understand the rate-limiting step of the process, the intraparticle diffusion model was also applied. The fitting results are presented in Figure S1. Three obvious linear regions were identified in the drugs’ uptake plots, as previously reported in other studies [14]. The adsorption process undergoes three distinct phases: The first stage indicates the external mass transfer between liquid and solid phases, the second one refers to the intraparticle diffusion, and the third refers to the final adsorption/desorption equilibrium stage. According to the calculated k values, presented in Table S1, the k₁ values in the first stage are higher, indicating that external mass transfer is the faster stage and consequently plays a pivotal role in the process mechanism. Consequently, the intraparticle diffusion is not only the rate-controlling step in the adsorption process of the target compounds onto virgin or aged PP MPs [38].

3.3. Adsorption Isotherms

In order to study the adsorption mechanism of the target compounds on the virgin and UV-aged PP MPs, Linear, Langmuir, and Freundlich models were employed to fit the experimental data obtained. In Figure 4 the curves of the equilibration isotherms at 25 °C are presented, while

Table 4. Th The isotherms parameters for the adsorption of pharmaceuticals onto PP MPs’ surfaces (time = 360 min, T = 25 °C).

		Freundlich Model				Langmuir Model
MPs	Pharmaceuticals	KF (mg1−1/n L1/n g−1)	n	R2	X2	Qmax (mg/g)	KL (L/mg)	R2	X2
Virgin PP	TMP	0.65	2.98	0.969	0.677	1.65	0.643	0.962	0.825
	MTZ	0.40	2.45	0.970	0.343	1.25	0.437	0.957	0.490
	ISO	0.53	2.83	0.955	0.679	1.45	0.544	0.955	0.822
	IMC	3.98	4.04	0.991	2.902	7.83	1.005	0.983	5.328
UV-aged PP	TMP	0.96	4.05	0.997	0.038	1.92	1.019	0.988	0.187
	MTZ	0.60	3.12	0.994	0.054	1.51	0.565	0.990	0.095
	ISO	0.76	3.20	0.998	0.020	1.82	0.461	0.995	0.065
	IMC	4.28	3.50	0.994	1.107	9.30	0.776	0.989	2.077
Virgin PP	DCF	1.67	2.37	0.991	0.451	5.5	0.412	0.987	0.644
	VLS	0.93	1.95	0.996	0.155	4.1	0.293	0.989	0.830
	KTF	0.73	2.80	0.997	0.027	2.1	0.478	0.979	0.107
UV-aged PP	DCF	2.63	2.62	0.974	2.281	7.3	0.528	0.981	1.477
	VLS	1.61	2.48	0.988	0.399	5.5	0.337	0.997	0.095
	KTF	1.03	2.29	0.993	0.069	3.6	0.382	0.989	0.113

presents the calculated equilibrium parameters obtained from the applied aforementioned models.

Comparing the three applied models, it was noticed that the Freundlich model exhibited better fitting to the experimental data based on the calculated R² and Chi–square values. The exported R² values were higher using the Freundlich model (0.969 ≤ R²_F ≤ 0.998) compared to those obtained from the Langmuir model (0.955 ≤ R²_L ≤ 0.995), or the Linear one (0.744 ≤ R²_LN ≤ 0.928), while the Chi–square values were lower than those obtained from the other models (0.020 ≤ χ²_F ≤ 2.902; 0.065 ≤ χ²_L ≤ 5.328; 2.758 ≤ χ²_LN ≤ 69.512). The Freundlich fitting model is a widely used non-linear sorption model, which indicates multi-layer coverage of an adsorbent’s surface, while the Langmuir model is a non-linear fitting model which describes that the adsorption process as occurring in monolayers with sorption sites that are constant.

The calculated Κ_F values indicate the adsorption capacity of the material differentiated depending on the physicochemical properties of each pharmaceutical. Since the drug uptake on the surface of the MPs follows approximately the pharmaceuticals’ LogKow values (DCF > IMC > VLS > KTF > TMP > MTZ > ISO (Table 1)) it can be concluded that the hydrophobicity of micropollutants may play a crucial role in their adsorption onto PP MPs [18]. Similar findings have been previously published in other studies [16,39,40]. However, electrostatic forces, Van der Waals forces, or π-π interactions may also affect the adsorption of some drugs on PP MPs, since the adsorption uptake of pharmaceuticals did not exactly follow their hydrophobicity trends.

Concerning the effect of aging on the adsorption of pharmaceuticals, it was noticed that the K_F values were increased with the UV-aged MPs in all cases, meaning that the aging process enhanced the adsorption capacity of the MPs. This can be attributed to the cracks and pits formed after UV-light exposure, which offer more adsorption sites on the PP MPs’ surfaces. Moreover, the calculated CI value of the MPs was also increased after the photo-aging process, leading to the formation of more oxygen containing groups on the MPs’ surfaces. This might favor the formation of hydrogen bonding with the studied compounds [41].

3.4. Effect of pH

Typically, the pH observed in aquatic environments ranges from 5 to 9. To investigate the role of pH, experiments were conducted at three different pH levels, ranging from 4 to 8. Results are shown in Figure 5. As it can be observed, changes of pH from acidic conditions (pH = 4) to slightly acidic conditions (pH = 6) and alkaline conditions (pH = 8) can significantly impact the entire process. The latter can be explained by electrostatic interactions, which may play a crucial role in the adsorption processes between MPs and contaminants and are dependent on their surface charges. As observed, an alkaline pH environment negatively impacts all studied compounds. Microplastics (MPs) are always negatively charged in alkaline media [11,18], and most of the studied compounds have a pH of less than 6. This means that at alkaline pH values, the electrostatic repulsion between the negatively charged MPs and the target compounds reduces their adsorption on the MPs’ surfaces.

For the three hydrophobic compounds, DCF, VLS, and KTF, the adsorption on PP MPs decreases by more than 50% as pH increases from 4 to 8. Under acidic pH conditions, the functional groups of these molecules can be protonated, allowing them to attract the negatively charged MPs more easily. For the other four antibiotics, a slight increase in adsorption is observed as pH increases from acidic to neutral, with an optimum pH of 6. This is due to the protonation of amino groups (in the drugs), which offers electrostatic attraction at near-neutral pH values.

Moreover, increasing the pH from neutral to alkaline levels results in excess [OH⁻] competing with the adsorption of the drug molecules onto the functional groups of the MPs. In general, it can be concluded that electrostatic forces play a significant role in the adsorption process; however, other forces like pi-pi stacking, hydrogen bonding, Van der Waals forces, etc. may also occur depending on the nature of each drug.

3.5. Effect of the Environmental Matrix

To investigate the effect of the water matrix, the adsorption of the target pharmaceuticals on both virgin and aged PP MPs was examined in three different aqueous matrices: distilled water (DW), wastewater (WW), and seawater (SW). The physicochemical properties of these matrices are detailed in Table 2, and the results are illustrated in Figure 6.

The nature of the water matrix significantly influences the adsorption of the target compounds on both virgin and aged PP MPs. Higher removals rates were observed in distilled water, while lower adsorption efficiencies were noted in wastewater, and finally in seawater. Comparison between the virgin and aged MPs indicated that aging still had a positive impact on the adsorption in all matrices, while the order of the adsorption efficiency of the drugs remained consistent and was not affected by the matrix. Consequently, the reduced adsorption efficiency observed in wastewater and seawater can only be attributable to the physicochemical characteristics of the matrices, like pH, TOC, and salinity.

Regarding the pH of the selected matrices, only seawater has an alkaline pH (pH = 8.3), while wastewater’s pH is almost neutral (pH = 6.9) and distilled water’s slightly acidic (pH = 5.8). Considering that alkaline pH values are less favorable for adsorption, as observed in Section 3.4, it can be assumed that the pH of the matrix may be one factor that leads to the lower adsorption efficiency in seawater compared to the other matrices employed. Furthermore, the organic content observed in wastewater and in seawater have a negative effect on the uptake of the pharmaceuticals since it may compete with the target compounds for the adsorption sites on the MPs’ surfaces. Additionally, the size of the molecules of the natural organic constituents (usually humic or fulvic acids) is higher than the size of the target micropollutant molecules, and as a result the former can cover the surface of the MPs and block the interaction between adsorbents and pollutants [18,42] Lastly, the salinity in these matrices negatively affects adsorption. High concentrations of Na+ ions can interact with the negatively charged microplastic surfaces, decreasing drug adsorption, or replace the hydrogen ions in acidic groups, thus inhibiting hydrogen bond formation. Furthermore, increased salinity can cause the agglomeration of MPs with pharmaceuticals during adsorption, reducing the effective surface area and partially blocking the adsorption sites as previously confirmed in other studies [39,43,44]. However, other factors may also come into play due to the increased NaCl concentration. This increase can reduce the thickness of the electric double layer, neutralizing the surface charge of the MPs and weakening the electrostatic interactions between the drugs and the MPs. Additionally, the higher viscosity and density of the solution can hinder the mass transfer from the aqueous phase to the solid phase, further inhibiting the adsorption capacity of the MPs [45,46].

3.6. Desorption Process

Τhe desorption process of the target compounds was studied in order to evaluate their possible release from the surface of PP MPs under environmental conditions. The evaluation of desorption was carried out using three aqueous solvents with different pH values (pH = 4, 6, and 8). The desorption experiments were conducted for six hours, and the exported results are presented in Figure 7.

The results indicate that as pH rises, the desorption of the target pollutants also increases. Notably, the highest desorption of all the pharmaceuticals was observed at pH 8, while at acidic pH values, desorption efficiency was significantly reduced for most of the target compounds. Given that desorption is the reverse process of adsorption, it is expected that the electrostatic repulsion forces present in alkaline environments will promote the desorption of organic pollutants. However, ISO presents an opposite trend, being more easily desorbed in acidic conditions. Moreover, ISO adsorption was the only process that was not negatively affected in alkaline pH, indicating that interactions other than electrostatic forces may govern the process. Nevertheless, an overall glance at the desorption results reveal that the desorption percentages were not significantly high, ranging up to <14% for the virgin PP MPs. Even though the aged particles showed higher desorption ability, no significant increase in the desorption percentages of the target compounds were recorded, ranging up to <18%.

4. Conclusions

This study demonstrated that PP MPs can serve as potential carriers for various pharmaceutical compounds, as they can adsorb drugs with diverse physicochemical characteristics simultaneously. Hydrophobic interactions and electrostatic forces predominantly drive the adsorption mechanism, though other interactions such as π-π interactions and Van der Waals forces may also be involved. The kinetic study showed that the pseudo-second order kinetic model exhibited better fitting for most of the compounds, whereas for MTZ and TMP, the pseudo-first order model was more appropriate, indicating that their adsorption occurred as a physical process. The intraparticle diffusion model revealed that external mass transfer was the faster stage, playing a crucial role in the process mechanism. Furthermore, isotherm curves fit the Freundlich model best, suggesting multi-layer coverage on the MPs’ surfaces. Additionally, factors such as pH, salinity, and aging significantly influence the whole process. Adsorption is favored in acidic pH and low salinity conditions, while the aging of MPs appears to enhance drugs’ adsorption due to the formation of cracks and pits that increase active sites and the creation of more oxygen-containing functional groups that can form hydrogen bonds with the pollutants’ polar groups. Desorption of the pharmaceuticals from PP MPs is also possible in aqueous media, particularly under alkaline conditions, although to a small extent.