Effectiveness of Filtrasorb Activated Carbon in Removing Selected Pharmaceuticals from Water ^†

Abstract

This paper deals with the removal of selected micropollutants from water in the laboratory, namely the removal of pharmaceuticals using the sorption materials Filtrasorb F100 and Filtrasorb F400. A group of well-known and available pharmaceuticals was selected for the experiment, which were the over-the-counter analgesics Ibuprofen, Diclofenac, Naproxen and Paracetamol. The model water was created by mixing drinking water from the water supply system of the city of Brno and standards of these pharmaceuticals prepared in the accredited laboratory of ALS Czech Republic. Water filtration was carried out through two filter columns, each filled with a different type of Filtrasorb sorbent. The filtered water was collected at selected time intervals (1, 2, 4 and 6 min) for analysis. The measurements showed that Filtrasorb F100 and Filtrasorb F400 activated carbons have comparable efficiency for the removal for Ibuprofen, Diclofenac, Naproxen and Paracetamol (around 83%). Both activated carbons have proven to be reliable sorbents for the removal of selected micropollutants from water.

1. Introduction

Pollutants are harmful substances that have an undesirable effect on some components of the environment. Because some pollutants occur in small amounts, they are called micropollutants. These include, in particular, pesticides, pharmaceuticals, PPCPs, detergents, disinfectants and drugs [1]. During drinking water treatment, the pollutant may be transformed into various compounds that may be toxic, persistent and less biodegradable than the original pollutants. Micropollutants can thus become harmful to human and animal health, as their residues can eventually enter and accumulate in the food chain [2].

For the research described in this article, pharmaceuticals were selected from the above-mentioned micropollutants. Pharmaceuticals can enter drinking water sources in several ways. Surface water may contain pharmaceuticals mainly originating from sewage treatment plants that do not treat pharmaceuticals or from veterinary products that are excreted by animals. Another possibility for groundwater and surface water pollution is agriculture, where the source of pharmaceuticals and production-enhancing substances is mainly manure [3]. Leachates from poorly secured landfills also contribute to micropollutants in water.

The increasing consumption of pharmaceuticals worldwide, coupled with a continuously aging population and improved access to health care, has made pharmaceuticals a critical and indispensable element of modern society. Pharmaceuticals are used in both human and veterinary medicine and can accumulate in the environment depending on the degradation processes [4].

It is now known that pharmaceuticals can be removed from water by advanced oxidation processes, membrane processes or adsorption [5]. For our experiment of removing pharmaceuticals from water, we chose the adsorption process. We have previously addressed the issue of pharmaceutical removal from water and in [6] we described the results of a static (beaker) test for the removal of selected pharmaceuticals from water using four adsorbents. From these adsorbents, the two most effective were selected (Filtrasorb F100 and Filtrasorb F400) and a dynamic test for the removal of selected pharmaceuticals from water was performed on them, which is described in this paper. The aim of the dynamic test was to assess the effectiveness of two Filtrasorb activated carbons in the removal of four selected pharmaceuticals from water. The dynamic test was carried out in the laboratory of the Institute of Municipal Water Management of the Faculty of Civil Engineering, Brno University of Technology.

2. Materials and Methods

2.1. Characteristics of Selected Activated Carbons

Filtrasorb F100 and Filtrasorb F400 are the two types of activated carbon that were selected for our experiment. Filtrasorb F100 (Figure 1) and Filtrasorb F400 (Figure 2) are granular activated carbons for the removal of dissolved organic compounds from water and wastewater as well as industrial and food processing streams. The activated carbons are made from select grades of bituminous coal through a process known as reagglomeration to produce a high-activity, durable, granular product capable of withstanding the abrasion associated with repeated backwashing, hydraulic transport and reactivation for reuse. The raw coal is mined and subsequently manufactured into GAC in the United States to ensure the highest quality and consistency in the finished product [7,8]. Technical and physical parameters of Filtrasorb F100 and F400 (Calgon Carbon, Pittsburgh, PE, USA) are provided in

Table 1. Technical and physical parameters of Filtrasorb F100 and F400 activated carbons [6,7,8].

Parameter	Unit	F100	F400
Specific adsorption surface	m2/g	850	1050
Methylene blue	mg/g	230	300
Atrazine 1 µg/L	mg/g	40	40
Trichloroethylene 50 µg/L	mg/g	25	20
Effective size	mm	0.8–1.0	0.55–0.75
Bulk mass	kg/m3	500	450
Iodine number	mg/g	850	1000

.

2.2. Characteristics of Selected Pharmaceuticals

For our research, a group of pharmaceuticals that are known and available to everyone is selected. This is a group of OTC analgesics, which are divided according to their clinical effect into analgesic–antiphlogistic (Ibuprofen, Naproxen, Diclofenac, Dexketoprofen) and analgesic–antipyretic (Acetylsalicylic acid, Paracetamol, Propyphenazone) and according to the number of substances into single and combination analgesics [9]. For the laboratory experiment reported in this paper, four simple analgesics, namely Ibuprofen, Diclofenac, Naproxen and Paracetamol, were used.

Ibuprofen is a well-known pharmaceutical that has analgesic effects (relieves pain) and antipyretic effects (lowers temperature). Because it also has anti-inflammatory effects, it belongs to the group of non-steroidal anti-inflammatory drugs (NSAIDs). It is used for mild and moderate pain of various origins, e.g., joint pain, muscle pain, tooth pain, etc. The drug is available in the Czech Republic without a prescription, but only in certain doses. A prescription is required for higher doses of Ibuprofen [10]. In 2013, Ibuprofen was the best-selling drug in the Czech Republic, with more than 8 million packages sold [11].

Diclofenac is a proven, commonly prescribed, non-steroidal anti-inflammatory drug that has analgesic, anti-inflammatory and antipyretic properties, and has been shown to be effective in treating a variety of acute and chronic pain and inflammatory conditions. Diclofenac is well resorbed and due to its analgesic effects, it has broad use in a large number of patients [12].

Naproxen is a non-steroidal anti-inflammatory agent advocated for use in rheumatoid arthritis, degenerative joint disease and ankylosing spondylitis. It has a good antiphlogistic (anti-inflammatory) effect and rarely causes side effects. Very rarely, gastrointestinal bleeding may occur. Because it is eliminated slowly, it is particularly suitable for chronic therapy [12,13].

Paracetamol (acetaminophen) is one of the world’s most widely used non-prescription medicines from cradle to grave. It is readily available and inexpensive. As an analgesic, Paracetamol is better tolerated than non-steroidal anti-inflammatory drugs, although it may be somewhat less efficacious. During the 1980s, a decline in the use of Aspirin due to its association with Reye’s syndrome allowed Paracetamol to become the antipyretic and analgesic of choice in children and it is now the standard antipyretic and analgesic in all age groups [14,15].

2.3. Laboratory Experiment

The filtration kit (Figure 3) consisted of a 30 L barrel with model water, a pump, two filter columns, a pipe with caps and containers for the filtered water. The inner diameter of the columns was 4.4 cm and the bottom of the columns was filled with a drainage layer to prevent leakage of sorption material. The drainage layer consisted of a 1 to 2 cm layer of grit, a 4 mm layer of glass beads above and a 2 mm layer of glass beads below. Above the drainage layer was a layer of filter fill, which was chosen to meet the minimum fill height requirement of Filtrasorb material, which is stated to be 0.75 m. The new filter material was soaked for 24 h and then washed. Washing (Figure 4) was carried out by reversing the water flow from the bottom to the top until clean water flowed out (about 8 min). Filtration was then performed on both materials.

The model water was created by mixing drinking water from the water supply system of the city of Brno and standards of these drugs prepared in the accredited laboratory of ALS Czech Republic. The drinking water quality indicators used for the preparation of model water for the dynamic test are summarized in

Table 2. Quality of drinking water used for model water.

Parameter	Unit	Value	Limit
Color	mg Pt/L	4	20
Turbidity	FNU	0.6	5
Iron	mg/L	0.03	0.2
pH	-	7.39	6.5–9.5
Total hardness	mmol/L	2.66	2–3.5
Ammonium ions	mg/L	<0.03	0.5
Nitrates	mg/L	31.6	50
Nitrites	mg/L	<0.005	0.5
Chlorides	mg/L	20.4	100
Coliform bacteria	CFU/100 mL	0	0
Escherichia coli	CFU/100 mL	0	0

, where the limit for drinking water in the Czech Republic is given by Decree No. 252/2004 Coll., as amended [16]. The concentrations of the selected pharmaceuticals in the model water were similar to the real concentrations in raw water according to the literature [17]. The concentration value of Ibuprofen in the model water was 0.391 μg/L, Diclofenac was 0.346 μg/L, Naproxen was 0.526 μg/L and Paracetamol was 0.465 μg/L.

During filtration through the two activated carbons, water samples were taken at 1, 2, 4 and 6 min (Figure 5) and the pH, temperature, turbidity and concentrations of each pharmaceutical were measured. Only the concentrations of the pharmaceuticals were determined by the accredited laboratory ALS Czech Republic; the other parameters were determined in the laboratory of the Institute of Municipal Water Management of Faculty of Civil Engineering at Brno University of Technology.

Water pH measurements were performed using an Adwa AD14 pH meter with a thermometer (Adwa Instruments, Szeged, Hungary), which is a high-quality microprocessor-controlled portable pH meter with built-in temperature measurement with automatic temperature compensation. The pH and temperature can be monitored simultaneously on the tester’s two-line display. The Adwa meter is water- and moisture-proof.

Turbidity was determined using the HACH 2100Q IS turbidimeter (HACH, Prague, Czech Republic). Turbidity values are measured by this instrument in FNU, which is the official internationally accepted unit for measuring turbidity.

3. Results and Discussion

The resulting pharmaceutical concentrations in the filtered water during the measurements are shown in

Table 3. Pharmaceutical concentrations in the model water (time 0 min) and after filtration through Filtrasorb F100 and Filtrasorb F400.

Time [min]	Ibuprofen [µ/L]	Diclofenac [µ/L]	Naproxen [µ/L]	Paracetamol [µ/L]
0	0.391	0.346	0.526	0.465
1	<0.05	<0.05	<0.1	<0.1
2	<0.05	<0.05	<0.1	<0.1
4	<0.05	<0.05	<0.1	<0.1
6	<0.05	<0.05	<0.1	<0.1

. The limit of quantification for Ibuprofen and Diclofenac was 0.05 μg/L, while for Naproxen and Paracetamol it was 0.1 μg/L. The limit of quantification (LOQ) corresponds to the concentration at which the precision of the determination is such that it allows quantitative evaluation [18]. The measured values showed that both Filtrasorb F100 and Filtrasorb F400 activated carbon removed pharmaceutical concentrations from water within one minute.

Equation (1) was used to determine the efficiency [19] of Filtrasorb F100 and Filtrasorb F400 sorbents in removing pharmaceuticals from water:

$$\mathsf{\eta }={\displaystyle \frac{{\mathrm{C}}_{\mathrm{R}\mathrm{W}}-{\mathrm{C}}_{\mathrm{F}}}{{\mathrm{C}}_{\mathrm{R}\mathrm{W}}}}100,$$

(1)

where η is the drug removal efficiency from water in %, C_RW is the drug concentration in raw (model) water in µg/L and C_F is the drug concentration in filtered water in µg/L.

Since the values of pharmaceutical concentrations in water were measured below the limit of quantification, it was the limit of quantification values that were substituted into the above Equation (1). Although the Filtrasorb F400 sorbent material had a larger specific adsorption surface than the Filtrasorb F100 sorbent material, the efficiency of removing drugs from water using these sorbents was the same. Thus, the removal efficiencies of Ibuprofen, Diclofenac, Naproxen and Paracetamol from water were greater than 87.2%, 85.5%, 81.0% and 78.5%, respectively, as shown in

Table 4. Efficiency of Filtrasorb F100 and Filtrasorb F400 for pharmaceutical removal from water.

Time [min]	Ibuprofen [%]	Diclofenac [%]	Naproxen [%]	Paracetamol [%]
1	>87.2	>85.5	>81.0	>78.5
2	>87.2	>85.5	>81.0	>78.5
4	>87.2	>85.5	>81.0	>78.5
6	>87.2	>85.5	>81.0	>78.5

. According to the lower limit of quantification for Ibuprofen and Diclofenac, the removal efficiency of these two drugs from water was higher than that of Naproxen and Paracetamol. Overall, the efficiency of Filtrasorb F100 and Filtrasorb F400 activated carbons in removing pharmaceuticals from water averaged around 83%.

Other parameters evaluated in the laboratory regarding the removal of pharmaceuticals from water were pH, temperature and turbidity. The pH value is defined as the negative logarithm of hydrogen ions and affects the efficiency of chemical, physicochemical and biological processes used in water treatment. In terms of taste perception, the ideal pH is in the neutral range, i.e., between 6.5 and 7.5 [20]. Decree No. 252/2004 Coll. that lays down the sanitary requirements for drinking and hot water and the frequency and scope of monitoring of drinking water states a limit value for pH in the range of 6.5 to 9.5 [16]. The measured pH values in the water filtered through Filtrasorb F100 ranged from 6.73 to 6.77. The pH values measured in water filtered through Filtrasorb F400 ranged from 6.73 to 6.84. All measured water pH values met the requirements of the drinking water quality decree. During the removal of pharmaceuticals from water, a higher pH was measured in water samples filtered through Filtrasorb F400 activated carbon than in water samples filtered through Filtrasorb F100 (Figure 6). Already, during the static (beaker) test carried out earlier, it was noted that the water pH values in contact with Filtrasorb F400 were higher than those in contact with Filtrasorb F100. The increase in water pH values during the dynamic test was apparently due to the nature of the sorption material.

Temperature is also one of the important indicators of drinking water. The optimum temperature for drinking water is between 8 and 12 °C [21]. The water temperature during the laboratory test ranged from 17.6 °C to 19.1 °C (Figure 7). The higher measured water temperature during filtration through Filtrasorb F400 was due the fact that the experiment with this sorbent was the second in the series. The temperature of the model water therefore increased due to the temperature in the laboratory. During filtration, the water was heated by a running pump.

Turbidity is one of the main sensory characteristics of water and is caused by inorganic or organic colloidal and finely suspended particles [22]. Turbidity is calculated photochemically by measuring the scattered or absorbed light at a given intensity [23]. Decree No. 252/2004 Coll. states a limit value for the turbidity of drinking water of 5 FNU. A turbidity value of 1.03 FNU was measured in the model water, and filtration through Filtrasorb F100 reduced the turbidity to 0.72 FNU at the sixth minute of measurement. The turbidity value of water filtered through Filtrasorb F400 was 0.89 FNU at the same time. All measured turbidity values during the dynamic test met the Czech drinking water quality regulation.

At the end of the experiment, the evaluation of the adsorption process during the flow of model water through the immobile adsorbent layer was carried out. The empty bed contact time, the total volume of sorption material, the filtration rate and the flow rate of the model water through the column were the design parameters on which the ideal adsorption progress depended. Equations (2)–(4) were used for the calculation [24]:

EBCT = h/v_f = V_R/Q_t,

(2)

V_R = A_R × h,

(3)

v_f = Q_t/A_R.

(4)

In the above equations, EBCT denotes the contact time with the empty bed, h the adsorbent height, v_f the filtration rate, A_R the bed area, Q_t the water flow rate and V_R the total volume of sorption material. The measured and calculated adsorption parameters are shown in

Table 5. Adsorption parameters for the removal of pharmaceuticals from water.

Parameter	Unit	F100	F400
h	m	0.8	0.9
Qt	m3/h	0.03	0.03
AR	m2	0.00152	0.00152
EBCT	min	2.43	2.74
VR	m3	0.0012	0.0014
vf	m/hr	19.73	19.73

.

The minimum recommended spreading height of Filtrasorb material was given as 0.75 m, which was observed in the laboratory experiment. The recommended filtration rate for Filtrasorb sorption materials was given as between 5 and 20 m/h [25]; thus the maximum recommended rate was not exceeded during the laboratory experiment with a filtration rate of 19.73 m/h. The different EBCT values for the sorption materials were due to the different spreading heights.

4. Conclusions

The removal of selected pharmaceuticals from water dynamically via Filtrasorb F100 and Filtrasorb F400 sorbents was very successful and beyond expectations. Pharmaceutical concentrations were measured only in the model water. In the other time samples, the concentration values were below the limit of determination. Thus, the specific measurements showed that the removal efficiencies of Filtrasorb F100 and Filtrasorb F400 activated carbons were comparable, around 83%, for Ibuprofen, Diclofenac, Naproxen and Paracetamol. Both activated carbons were proven to be reliable sorbents for the removal of selected micropollutants from water.

The measured pH values during the experiment ranged from 6.73 to 6.84, which complied with the drinking water regulation in the Czech Republic. The water temperature increased slightly for both sorbent materials during the dynamic test, but more for the F400 material than for the F100 sorbent, which was due to the fact that the experiment with the column filled with Filtrasorb F400 sorbent was only conducted second in the sequence and the temperature of the model water was increased by the temperature of the laboratory room. The water temperature did not exceed 19.1 °C during the experiment with both sorbents. The assumption of [20] that the pH of water is less than 7 at water temperatures below 25 °C was consistent with the pH of the water throughout the dynamic test.

The turbidity of the model water was already below the drinking water limit. Both activated carbons were able to keep turbidity below this limit during the experiment, and at low levels.

Evaluation of the adsorption process showed that the contact time with the empty bed was sufficient to remove the selected pharmaceuticals from the water, following the manufacturer’s instructions for Filtrasorb F100 and Filtrasorb F400 activated carbons for both the spreading height and flow rate.