Solar photodegradation of carbamazepine from aqueous solutions using a compound parabolic concentrator equipped with a sun tracking system

2.1 Materials

Carbamazepine (CBZ ≥ 99%) was acquired from Arasto Pharmaceutical Chemicals Inc. (Tehran, Iran). The chemical structure and specification of CBZ are shown in Figure 1. All other used chemicals in this study, such as hydrochloric acid and sodium hydroxide, was purchased from Merck, Darmstadt, Germany.

Figure 1

Chemical structure and specifications of Carbamazepine.

2.2 Installation and Experimental procedure

A schematic diagram of the experimental pilot CPC reactor equipped with sun tracker system is shown in Figure 2. The pilot consisted of a quartz tube (number 1 in Figure 2) with a length of 80 cm, an outside diameter of 3.2 cm and inside diameter of 2.8 cm; a compound parabolic collector (CPC) (number 2 in Figure 2); and a dual axis solar tracker system. The CPC aperture area was 0.38 m². The tracking system pursues the angular height position of the sun in the sky in addition to pursuing the sun’s east-west motion by rotating the CPC reactor. As shown in Figure 2, the solar tracker system has a photo sensor (number 3 in Figure 2), a vertical rotator (number 4 in Figure 2), a horizontal rotator (number 5 in Figure 2), anda controller (number 6 in Figure 2). The rotators are principally conducting the action of sun tracking. The horizontal rotator tracks the east-west movement of the sun, and the vertical rotator tracks the sun’s up-down motion. These rotators and the photo sensor are interfaced with a controller, which is controlling the rotators according to the sensor’s input. The photo sensor uses a light-dependent resistor (LDR) to acquire the sun’s position. The LDR sensor senses the light intensity and sends a signal to the controller. The controller compares the signals received from the photo sensor and decides the rotation direction of the rotators according to the stored program.

Figure 2

Schematic diagram of pilot scale CPC with tracking system reactor; 1- quartz tube (with inlet and outlet valves), 2 - CPC, 3 - photo sensor, 4 - Vertical rotator, 5 - Horizontal rotator, 6 - Controller , 7 - Thermometer, 8 - Copper rod

The reactor temperature was recorded at the beginning and end of the reaction using a mercury thermometer (number 7 in Figure 2) placed inside the quartz tube. To determine the role of temperature in the removal of carbamazepine, a black copper rod (number 8 in Figure 2) was placed inside the quartz tube. The reactor without the black copper tube was called CPC1, while the reactor with the tube was called CPC₂. The CPC (1 and 2) experiments were replicated with the same CPC and tracking system but one different days and with different quartz tubes (with or without the copper rod). The conditions on the different days were similar (sunny and calm). As a negative control for the effect of CPC on CBZ removal, a simple quartz tube without CPC or copper rod was used (this situation is termed the “blank”). Natural sunlight experiments were carried out during three consecutive weeks during the period of May–July 2017 at the University of Shahid Beheshti University of Medical Sciences in Iran, local latitude 35° 47’ 50’’ N, longitude 51° 23’ 50’’ E.

2.3 Solar radiation dose

In this study, the solar light intensity and the reactor’s temperature changes were measured and recorded. Solar irradiance was measured with a global radiometer (Model HAGNER, Sweden).To estimate the solar radiation dose, the cumulative solar radiation dose (CSRD) into the reactor over the period of experiment per unit time and volume was calculated. The CSRD was calculated by integrating the average of measured solar irradiance (Q) during the different exposure times using following equation:

where CSRD is cumulative solar radiation dose (kWh/L), C is reflection coefficient (typically 0.8), A is the CPC aperture area (m²), V is the volume of treated wastewater (L) and Q is the measured solar irradiance (W/m²).

2.4 CBZ-contaminated water preparation

A standard stock solution of CBZ (150mg/L) was prepared from analytical grade CBZ powder and methanol/ deionized water (20% volume/volume). This solution was stored in the refrigerator at 7°C for further use and refreshed at 5-day intervals. The stock solution of CBZ was spiked into the reaction solution in a 500 mL volumetric flask to yield a selected concentration from 0 (control) to 5 mg/L with intervals of 0.5 mg/L. Generally, the volumetric percentage of methanol present in reaction solutions was less than 0.1% (v/v). The solution pH was adjusted with 0.01N H₂SO₄ and 0.01N NaOH. CBZ removal experiments were carried out according to the following approaches: (1) Blank reactor (quartz tube without CPC, tracking system and copper rod, (2) CPC₁ reactor (quartz tube in the focal axis of CPC) and (3) CPC₂ reactor (CPC₁ plus copper rod to quartz tube). All experiments were performed in duplicate.

2.5 Analytical methods

CBZ concentration measurements were carried out by HPLC (Knauer) equipped with an Eclipse plus C18 column (3.5 ×4.6×100 mm) and a UV detector at λ= 280 nm. Its retention time (RT) was 7.2 ± 0.1 min. A mixture of methanol/water solution (60/40 %v/v) was used as the mobile phase at a flow rate of 0.8 mL/min, and the injection volume was 20 μL. The lowest detection and quantification limits were estimated to be approximately 50μg/L and 100 μg/L respectively. CBZ products in photochemical reactions were also monitored by liquid chromatography mass spectrometry (LC-MS, Quartro Micro API micro-mass Waters 2695) with a Thermo Quest Finnigan LCQ Duo mass spectrometer system.

Ethical approval: The conducted research is not related to either human or animal use.

3.1 Solar radiation

The cumulative solar radiation doses (CSRD) over the period of experiments are shown in Figure 3. The CSRD values were calculated using equation (1). As shown in this Figure, the solar radiation doses fell within the range of 0.5 to 4 kWh/L.

Figure 3

Estimated cumulative solar radiation dose (CSRD) in the period of experiments into the reactor per unit of time and volume.

3.2 Effect of initial CBZ concentrations

FigureThe efficiency of removal of CBZ during exposure to three hours of sunlight (10:30-13:30) at pH 7±0.01 is shown in Figure 4. As can be seen in this Figure, there was no significant change in the phototransformation rate of CBZ with the initial concentrations of 0.5 mg/L to 2 mg/L. When the initial concentration was increased from 2 to 4 mg/L, the phototransformation rate of CBZ has increased remarkably in all three reactors (blank, CPC1, and CPC2), after which the trend was invariant. These results suggest that the CBZ phototransformation reactions follow pseudo-first-order kinetics. Also, these observations may be explained by the exposure of specific binding sites of CBZ at higher concentrations, assuming that binding sites were limiting. It is also seen in Figure 4 that conversion rates in CPC1 and CPC2 reactors are about twice as high as that for the blank reactor. Therefore, it can be concluded that concentrating solar radiation and use of the sun tracking system could dramatically increase CBZ removal efficiency. However, there is no significant difference in CBZ removal between CPC1 and CPC2, despite the higher temperature induced by the presence of the copper rod in CPC2. It thus seems that temperature has little effect on CBZ removal.

Figure 4

Effect of initial CBZ concentrations on removal efficiency in three reactors (solar irradiation time 3 hours (10:30-13:30 pH = 7±0.01).

3.3 Effect of pH and Temperature

Solution pH significantly affects the photolysis of organic contaminants, including the nature of photoproducts as well as reaction kinetics [26, 27]. The effects of pH and temperature variation during CBZ solar photolysis by CPC1, CPC2, and blank reactors are shown in Figure 5. As shown in Figure 5a, in all three reactors, the CBZ removal rate is low at low pH. However, there is a significant difference in the removal efficiency of CPC1 and CPC2 reactors relative to the blank reactor under all three pH conditions. This

Figure 5

Comparison of the Efficiency of Blank, CPC1 and CPC₂ reactors in removal of Carbamazepine (Mean ± SD) at contact Time of 0.5 to 5 hours - start time: 9:30 a.m., end time 2:30 p.m. Initial CBZ concentration = 3 mg/L (a) pH= 4± 0.1; (b) pH=7± 0.1; (c) pH= 9± 0.1 and (d) Temperature variations in these reactors.

further shows that the concentration of solar radiation by CPC reflectors dramatically increases the effectiveness of CBZ removal. However, the efficiency of the blank reactor in CBZ removal is higher than that previously reported for similar conditions (see refs. [28] and [29]). This difference is likely due to the different initial CBZ concentrations and differences in solar radiation angles. Upon increasing the pH to neutral and basic, the efficiency of CBZ removal was enhanced in all three reactors (see Figure 5b and c). Thus, solution pH played an important role in CBZ photodegradation, as removal efficiency increased slightly with pH from about 49% at pH 7 to almost 61% at pH 9.

Although temperature differences between CPC1 and CPC2 reactors were significant (Figure 4 d), there is no meaningful difference in removal efficiency between the two reactors. From the results above, it seems that the temperature has a minor effect on CBZ removal efficiency. This is consistent with other published findings [30, 31].

As can be seen in Figure 5, using a CPC reactor equipped with sun tracking system promote CBZ transformation by 2-3 folds higher than without CPC and sun tracking system.

3.4 CBZ transformation products and pathway

Solar photodegradation reactions can occur directly or indirectly. In direct photodegradation, solar radiation received by the compound itself leads to phototransformations; in indirect photodegradation, highly reactive radicals, such as atomic oxygen and hydroxyl radicals (HO^•), are generated using catalysts [32, 4].

The direct solar photodegradation of CBZ starts after absorption of photons from solar radiation by the compound. The interaction between sunlight and CBZ leads to the generation of hydroxyl radicals (HO^•) [33]. The proposed pathway for HO^• generation is presented in Equations 2 and 3 [3, 34].

CBZ radicals may be generated by direct photolysis, and then the CBZ radical becomes able to oxidize water to^• generate HO^• (Equation 2). Hydroxyl radicals then participate in CBZ hydroxylation.

In this study, the main intermediates of CBZ were monitored by LC/MS followed by detection via UPLC (Synapt G2-S HDMS in positive ion mode). To investigate the by-products formed by solar photo-degradation of CBZ in the CPC2 reactor, photo-transformation of CBZ (3 mg/L) was accomplished at a solar irradiation time of 3 h and initial pH value of 9. In Figure 6 a scan LC/MS analysis of CBZ intermediates is shown. CBZ photo-transformation intermediate molecular weights were assigned on the basis of the m/z quantity of the protonated molecule [M + H] ⁺. Generally, eight intermediates of CBZ were recognized. The proposed structures and pathways of the main photodegradation intermediates were shown in Figure 7. For simplification in this Figure and further, the detected intermediates (DI) are numbered from 1 to 8 in Figure 7.

Figure 6

Scan LC/MS analysis of CBZ intermediates in the CPC₂ (initial CBZ concentration 3 mg/L, the solar irradiation time of 3 h and initial pH value of 9).

CBZ photo-transformation under solar irradiation is likely to progress with photoionization (facilitated by dissolved oxygen to generate the radical cation of CBZ) and hydration of the double bond on the central heterocyclic ring of CBZ [34]. The potential structures of reaction products were tentatively proposed based on the accurate masses and similar studies [34, 35, 36, 37]. Hence, an extensive description of MS data analysis is not provided here. Hydration of the double bond on the central heterocyclic ring is a minor route leading to 10, 11-Dihydro-10-Hydroxy Carbamazepine (DI₁). This hydration can occur in the

Figure 7

The proposed structures and pathways of the main photo-transformation detected intermediates (DI) (based on LC/MS scan of CBZ intermediates and Chiron et al. [34]).

absence of light at near-neutral pH [34]. The photo ring expansion reaction was suggested to be the main pathway for the other detected intermediates (DI2 - DI8) [38]. A photo ring expansion reaction likely participates in the formation of carbamazepine-9-carboxaldehyde (DI2) resulting from a hydroxylation. Complementary hydroxylation of DI2 with loss of the side chain (CH₂NO) yielding hydroxyacridine-9-carboxaldehyde (DI6), Acridone (DI7) and Acridine-9-carbaldehyde (DI8). The concurrent loss of the carboxyaldehyde group and the side chain (CONH2) leads to DI5 (acridine). Also, Acridine (DI5) oxidation can yield Acridone (DI7).

DI2 (carbamazepine-9-carboxaldehyde) dimerization and reduction of the CHO side chain lead to DI3 and more reduction of the CHON side chain of DI3, yielding DI4[34]. According to this analysis of CBZ transformation by-products, it seems that complete mineralization of drug did not occur. However, the use of a CPC reactor equipped with a tracking system had a significant effect on the photo-transformation of CBZ.