Peroxide impact on the fate of veterinary drugs in fertilizers

Vol.:(0123456789) 1 3 Chemical Papers (2020) 74:311–322 https://doi.org/10.1007/s11696-019-00883-x ORIGINAL PAPER Peroxide impact onthefate ofveterinary drugs infertilizers JolantaTurek‑Szytow1,2 · D.Marciocha3,4· J.Kalka1· J.Surmacz‑Górska1 Received: 13 February 2019 / Accepted: 17 July 2019 / Published online: 26 July 2019 © The Author(s) 2019 Abstract The presence of veterinary medicines in organic manure causes soil contamination which contributes to increasing resistance of indigenous microflora to drugs and results in greater susceptibility of people to allergies. The main aim of the study was to assess the efficiency of inorganic peroxide mixtures (PM) with calcium peroxide content (CaO2) in the stabilization process of manure contaminated with antiparasitic agents: albendazole (ALB) and levamisole (LEV). As a solid, CaO2 is relatively stable against decomposition. In contact with water, however, it hydrolyzes with release of oxygen. The hydrolyzation of CaO2 proceeds very slowly in soil, which guarantees the constant release of hydrogen peroxide that subsequently becomes the source of free radicals (chemical oxidation) and oxygen (aerobic conditions for the microbes). It may contribute to con‑ tinuous elimination of drugs from manure. The study has demonstrated that there were significant differences in ALB and LEV conversion stimulated by the PM addition. PM supplementation increased the drug availability (on average 15% and 25% increase in the initial concentration for ALB and for LEV, respectively), thereby increasing the initial rate of reaction. Elimination of ALB and LEV from the manure sorption complex is followed by Ca2+ saturation. The initial degradation rate was affected by PM for both drugs, but the mechanisms of decomposition have been modified only for ALB. The loss of ALB in the peroxide supplemented samples was 92%, and in the samples, without the PM, it did not exceed 61%. Loss of LEV was over 90% irrespective of PM supplementation. * Jolanta Turek‑Szytow [email protected] 1 Environmental Biotechnology Department, Faculty ofEnergy andEnvironmental Engineering, Silesian University ofTechnology, ul. Akademicka 2, 44‑100Gliwice, Poland 2 Centre forBiotechnology, Silesian University ofTechnology, Krzywoustego 8, 44‑100Gliwice, Poland 3 Energy Research Center, Technical University ofOstrava, 17 listopadu 15/2172, 70833Ostrava, CzechRepublic 4 PPU “CHEMCO” Sp. z o.o. ingredients forcosmetics andhousehold chemicals, ul. Kościuszki 19, 83‑033Sobowidz, Poland 312 Chemical Papers (2020) 74:311–322 1 3 Graphic abstract Levamisole (LEV) Albendazole (ALB) NS N N N S N H O O poultry manure (Ca2+) Peroxide mixture with CaO2 SOIL Keywords Rate· Degradation· Albendazole· Levamisole· Calcium peroxide (CaO2)· Poultry manure Introduction The use of veterinary medicines in animal husbandry increases the risk of their presence in organic fertilizers (manure and liquid manure), which results in their occur‑ rence in soil environment (Kumar etal. 2005; Sarmah etal. 2006; Song etal. 2010). Antiviral, antifungal, and antiparasitic agents can penetrate the soil at varied rates in an unchanged or partially degraded form (Boxall etal. 2003, 2004; Capelton etal. 2006). Consequently, surface runoff from the soil and resulting water contaminated with drugs causes gradual development of drug resistance and endocrine disorders in soil micro‑organisms. The build‑up of drug residues in the environment makes it necessary to use drugs at higher pharmacological doses, which in effect constantly increases environmental pollution (Kools etal. 2008a; Varel etal. 2012). The use of manure for fertilizing soils is a common agricultural technique and has been enhancing quantity and quality of crops (Miller and Berry 2005). Polluting potential of pharmaceutical residues in soils is determined by their bioavailability, which is mostly defined by sorp‑ tion, desorption, and migration with water (Hamscher etal. 2005; Popova etal. 2013). Their distribution can be influenced by a range of factors and processes, includ‑ ing physico‑chemical properties of the environment and drugs (Jørgensen and Halling‑Sørensen 2000;Chefetz etal. 2008). Drugs infiltrating into the soil with organic fertilizers influence the quality of both the soil and the crop (Davis etal. 2006; Oliver and Gregory 2015). The negative environmental impact is mainly due to organisms’ ability to immobilize and accumulate medicines in their tissues (Boxall etal. 2003; Kumar etal. 2005; Marciocha etal. 2013). Albendazole (ALB) is an antiparasitic drug with a broad activity range. It is a benzimidazole derivative and is used against roundworms and some flatworms (Dayan 2003; Prasad etal. 2010). The mechanism of action is 313Chemical Papers (2020) 74:311–322 1 3 based on inhibition of specific enzymes (e.g., tubulin poly‑ merase andfumarate reductase). ALB is eliminated from the organism mainly with urine as albendazole sulfoxide (Capece etal. 2009; Bartikova etal. 2011). Levamisole (LEV) is used as a selective agent for the control of nematodes in the treatment of parasitic infections in humans and animals (Kamal etal. 2005). It is mainly applicable as a deworming agent designed to fight gastro‑ intestinal and pulmonary nematodes in poultry, pigs, and ruminants (Grønvold etal. 2004). It also shows anti‑cancer and immunomodulation activity (Sadeghi etal. 2007). From the body, LEV is mainly excreted in an unchanged form in the urine. Metabolites constitute about 20% of LEV (Barker 2008; Kools etal. 2008b). ALB and LEV were selected for investigations in this study due to their high consumption as veterinary medica‑ tion, unknown metabolic paths, and migration in soil, as well as evidence in scientific literature of high concentration in the environment (Capletonet al. 2006; Spychaj‑Fabisiak etal. 2007; Kools etal. 2008a). The environmental risk of these drugs has been assessed in this study, because pre‑ dicted environmental concentration (PEC) for both selected drugs was in excess of 100μg/kg. At such PEC, according to the European Medicines Agency (EMA) guidelines for ALB and LEV, the process of assessing the environmental risk should be performed (EMA 2007; Oliver and Gregory 2015). The average concentration of ALB found in soil and sediment samples was determined in the range 0.2–25mg/ kg (Thiele‑Bruhn 2003). Studies showed that the environ‑ mental concentrations of pharmaceuticals are in the range of 0.034µg/kg to 500mg/kg (Thiele‑Bruhn 2003; Babić and Mutavdžić Pavlović 2013; Wohde etal. 2016; Mutavdžić Pavlović etal. 2018). Calcium peroxide (CaO2), a compo‑ nent of inorganic peroxide mixtures (PM), is a source of oxygen (aerobic condition for autochthonic microflora) and free radicals (chemical interaction) in the environment. Free radicals, especially hydroxyl radicals, have been reported to have strong chemical activity (2.8V) which leads to a non‑ selective degradation of both persistent organic pollutants (POPs) and organic substances (Biń and Zieliński 2000; Liu etal. 2014; Miksch etal. 2015). Low solubility of PM results in gradual secretion of molecular oxygen and free radicals to achieve slow yet complete dissolution (Walawska and Gluzińska 2006). Sub‑ strate generation is at its maximum in the first 7days after introduction PM to the soil environment and then slowly declines, but usually takes about 30days (Walawska etal. 2007; Romero etal. 2011). The disintegration rate depends on the pH and increases together with the increase of soil moisture content. Once the PM is introduced into the soil, the released oxygen improves aerobic environmental condi‑ tions and stimulates the activity of indigenous microflora, which gives rise to the rate of the biological pollutants degradation. Due to low PM water solubility, the oxidation process occurs slowly and results in the slow and continuous formation of hydrogen peroxide (reactions1, 2): or In turn, H2O2 molecules are the source of free radicals (chemical oxidation) and oxygen (due to aerobic conditions for micro‑organisms and biological oxidation) (reactions3, 4): Hydroxyl radicals can react with organic contaminants reactions5 and 6 (Deng and Zhao 2015; Wang etal. 2016): During the degradation of PM and H2O2, reactive oxygen species (ROS) are produced, which is the source of oxygen for micro‑organisms taking an active part in the biodegrada‑ tion of contaminants and at the same time being the source of free radicals capable of oxidizing organic contaminants. However, it has been shown that the presence of ROS in low concentrations is essential for the proper course of many physiological processes. The apparent paradox is defining the role of ROS as a transformation regulator of the com‑ pounds. On the other hand, the toxicity of the products is mainly dependent on their concentration. As a solid substance, the PM‑containing CaO2 is far less problematic and safer in handling than hydrogen peroxide solutions (Solvay Chemicals and Inc 2013; Malachowska‑ Jutsz and Neisler 2015; Wang etal. 2016). The oxidizing potential of ROS generated from CaO2 is summarized in Table1. ALB and LEV as pharmaceutical products are among the “emerging compounds”. Contaminants of emerging concern (CECs) are increasingly being detected at low levels in the environment, and there is a concern that these compounds may have a negative impact on living organisms. Horvat (1) 2CaO2+2H2O → 2Ca(OH)2+O2, (2) CaO2+2H2O → Ca(OH)2+H2O2. (3) H2O2 → H2O+O⋅, (4) 2H 2O2 → 2H 2O+O2. (5) H2O2+OH− → HOO⋅ +H2O, (6) HOO⋅ +organic compounds → oxidation products +HO⋅. Table 1 The oxidizing potential of ROS (Pera‑Titus etal. 2004; Deng and Zhao 2015) Substance Oxidizing potential (V) OH·2.80 1.23 (pH < 7) 0.4 (pH > 7) O22.42 H2O21.77 314 Chemical Papers (2020) 74:311–322 1 3 etal. present the analysis, occurrence, and fate of anthel‑ mintics and their transformation product including ALB and LEV in the environment. Their wide range of applica‑ tions in the protection of livestock health and prevention of diseases and equally widespread practice of managing organic fertilizer justifies carrying out research on the fate of drugs in the soil environment. The key aim of this research was to determine the influence of PM on the fate of veteri‑ nary drugs from the group of antiparasitic agents (ALB and LEV). Second, the study was aimed at developing an alter‑ native bioremediation method to increase the efficiency of manure stabilization by the addition of PM‑containing CaO2. The degradation of pharmaceuticals have been respon‑ sible for the generation of many transformation products (TP) (Boxall etal. 2003; Sadeghi etal. 2007; Capece etal. 2009; Prasad etal. 2010; Horvat etal. 2012; Oyeduntan and Uwalaka 2015; Technical Evaluation Report 2015; Mutavdžić Pavlović etal. 2018). The presented research justifies the desirability of com‑ bining chemical and biological processes that would lead to the disappearance of drugs in poultry manure before their deposition into the soil. Materials andmethods The selected properties of studied drugs ALB and LEV are summarized in Table2. Both tested drugs were purchased from Sigma‑Aldrich, LEV CAS Number 16595‑80‑5 and ALB CAS Number 54965‑21‑8. The experiment used a sol‑ vent Sigma‑Aldrich (methanol, dimethyl sulfoxide—DMSO, and acetonitrile) and chemical reagents purchased from POCH (K2HPO4, NaHCO3, and Na2CO3). In the experiments, natural chicken manure (veterinary ID 30 17 84 01) produced by PPHU CDN Poland was used. The manure properties are summarized in Table3. The manure was activated using water that was dosed to obtain 60% of full manure water capacity, and then incubated for 7days to keep the constant level of respiration activity. Manure activ‑ ity was assessed by respiration activity SIR method deter‑ mined by OxiTop® Control System developed by WTW. Biodegradability tests were carried out in accordance with the method DIN EN 29 408/ISO 9408/OECD 301 F. The initial concentration of both drugs in manure was 45mg/kg. This dose was selected to detect for a large mar‑ gin above the analytical detection limit and allow for long Table 2 Physico‑chemical properties of veterinary drugs a http://www.drugb ank.ca/drugs /DB008 48 b Horvat etal. (2012) c Boxall etal. (2006) d http://www.drugb ank.ca/drugs /DB005 18 e http://www.chemi calla nd21.com/ f https ://pubch em.ncbi.nlm.nih.gov/compo und/Levam isole #secti on=Synon yms Active substance Levamisole hydrochloride Albendazole LEV ALB Systematic name (by IUPAC) (6S)‑6‑phenyl‑2H,3H,5H,6H‑imidazo[2,1‑b] [1,3]thiazoleaMethyl N‑(6‑propylsulfanyl‑1H‑benzimidazol‑ 2‑yl) carbamatec Brand names Levamisol, Ergamisol, Ketrax, Tetramisol, Wormicid, DecarisfValbazen, Albendazole, Zentel, Bilutac, Eskazolec Molecular formula C11H12N2SaC12H15N3O2Sd Structural formula Molar mass 204.291g/mola265.3314g/mold Half life 4.4–5.6ha8–12hd Melting point 264–265°Ca208–210°Cd Solubility in water 1.44g/la 1.116g/l (25°C)b2.28e−02g/ld 0.010g/l (25°C)b Octanol–water partition coefficient Log KOW 1.84c 2.87b3.14e 3.07b Organic carbon normalized sorption coefficient Log KOC 1.88b2.94b Toxicity LD50 = 40mg/kg (pig, subcutaneously) LD50 = 180mg/kg (rat, orally)aLD50 = 2.400mg/kg (rat, orally)c 315Chemical Papers (2020) 74:311–322 1 3 reaction time to compare varied test conditions. The tested manure (10kg dry mass, d.m.) was thoroughly mixed with ALB and LEV (45mg of each the drug/kgd.m.). ALB is insol‑ uble in water; therefore, DMSO was used as a co‑solvent for homogeneity samples with this drug. LEV is water‑soluble and distilled water was used to obtain solutions at tested concentrations. The PM dose was calculated to raise the pH of the manure by one unit assuming manure’s buffering capacity of 0.172mol H+/kgd.m. (Table3). The total dose of PM was 0.5g per 1kg of manure dry mass. This dosage of PM stimu‑ lated chemical changes by the addition of active oxygen pro‑ duced from CaO2. PM is strongly alkaline and using larger amounts might significantly change the pH and abrupt pH changes could affect the homeostasis of the environment or distress indigenous microflora by acting toxic (Turek‑Szytow etal. 2015; Małachowska‑Jutsz and Neisler 2015). The research included the following test series: MA— manure with ALB, MACa—manure with ALB and PM, ML—manure with LEV, and MLCa—manure with LEV and PM. During the experiment, the moisture level in all samples was maintained at the level of 60% fwc (full water capacity). To determine the removal rate of the drugs, the concentra‑ tions of ALB and LEV were tested on every second day for the total period of 35days. All test series underwent three replicate tests to assure statistical significance of the results. The influence ofPM ontheloss ofALB andLEV inthemanure Determination of ALB ad LEV was carried out with use of the HPLC method applying the RP‑18 chromatography column, length 250mm, which was preceded by the RP‑18 Hypersil Gold pre‑column, length 25mm, manufactured by Thermo Fisher. The system was provided with the UVD 340U detector by Gynkotek, the ASI‑100 autosampler by Dionex, and the P 580 LPG pump by Gynkotek. The results obtained were processed using the Chromilion software package. To analyse tested samples for the LEV content, 1g of the homogenized material was collected and underwent a 3‑h extraction process with 10ml of methanol (MeOH). In the ALB analyses, 1g of the homogenized sample under‑ went several steps of the extraction process using firstly distilled water (200ml for 1h) and chloroform (20ml for 1h), and then a 3‑h extraction process using 1‑ml DMSO solution. The samples for both LEV and ALB were filtered using a glass 0.2‑µm fiber filter. Acetonitrile and phos‑ phate buffer (15‑mmol K2HPO4, pH 3.8) were the eluents; the flow through the column was maintained at 1ml/min. The method recovery reached 99.6% and 89.7% for ALB and LEV, respectively. To determine the concentration of veterinary pharmaceuticals in the manure, the five‑point analytic curve was plotted in the range between 0.20 and 10mg/l. The linear affinity was 99.987 and 99.973 for LEV and ALB, respectively. It was assumed that both compounds the limit of quantification (LOQ) correspond to the first point of the calibration curve with the lowest concentra‑ tion (LOQ = 0.20mg/l). Limit‑of‑detection (LOD) level was 0.02mg/l. ALB was determined using a mixture of acetonitrile and phosphate buffer at a 50:50 ratio in the mobile phase. The spectra were recorded at a wavelength of 292nm. The analy‑ sis was conducted for 8min, whereas the time of ALB reten‑ tion was about 6.5min. The LEV concentration was determined using the eluent mixture at a ratio of 85:15 (buffer: acetonitrile) changeable flow gradient. The multi‑gradient flow of an isocratic (vari‑ able) and ramp (constant) was used. Flow changes were as follows: (1) isocratic, i.e., decreasing from 1 to 0.4ml/min in the time interval 4–7min, (2) isocratic, i.e., increasing from 0.4 to 1ml/min for the period 10–13min, and (3) ramp, i.e., 1ml/min at intervals of 1–4min and 13–19.5min and a constant 0.4ml/min in the period 7–10min. The analyti‑ cal detection wavelength was 217nm, and the total analysis duration was 19.5min. The retention time for LEV was esti‑ mated at 13.8min. The kinetic model was used to analyse drug loss rates: where C(t) is the concentration in the function of time, C0 is the initial concentration, and k1 is the kinetic constant. The value of the constant k was also calculated by the following: In this way, it was possible to check whether the given reaction is the reaction of the first order. The effect of PM on the availability of drugs was inves‑ tigated. The relationship between the highest identified C (t)=C 0 (e ∧ (k 1 t)) , k=1∕tln [C0∕(C0−C)] = 2303∕t∗log[C0∕C0−C]. Table 3 Physico‑chemical properties of the manure Parameter Value pH 11.6 Buffering 171.6mmol H+/kgd.m. ADW (absolutely dry weight) 62% NKjeldahl 48.39g/kg Organic substance 87.2% Organic carbon 13.2% Hh (hydrolytic acidity) 52.9mmol H+/kg S (alkalis sum) 705.1mmol H+/kg T (sorption capacity of the fertilizer) 758.0mmol H+/kg 316 Chemical Papers (2020) 74:311–322 1 3 concentration of drugs and Co was determined. These assessments were based on determining the correlation coef‑ ficient “R” Pearson’s between initial concentration rates of reaction for the ALB and LEV with and without PM addi‑ tion. The value of “a” coefficient and statistical significance analyses were calculated using the REGLIN function in Microsoft Excel Software package. The value of “a” coefficient and statistical significance analyses were calculated using the REGLIN function in Microsoft Excel Software package. Duncan’s test (program XLSTAT) was used for the statistical analysis of all results. Both Duncan’s test and REGLIN function methods were performed with 5% error margin. The influence ofPM onthephysico‑chemical properties ofmanure contaminated withveterinary drugs ALB andLEV Physico‑chemical parameters were measured at two inter‑ vals, namely at the start day (i.e., day 0) and the experi‑ ment end (day 35). Based on previous studies (Walawska and Gluzińska 2006; Małachowska‑Jutsz and Neisler 2015), 35days of test duration was chosen as adequate to assure stabilization of chemical reactions related to sorption and desorption, and biochemical activity associated with per‑ oxide decomposition. Each sample was homogenized prior to the analysis by grinding in a Testchem’s mill for 3min. The following parameters were measured and analysed at selected time intervals: • organic substance and absolute dry mass—using the gravimetric method, • pH—using potentiometric method, • Kjeldahl N—using Kjeltec 8200 system, • organic carbon—using Turin’s method, • nitrite and nitrate nitrogen concentration—were deter‑ mined in soil solution (PN‑ISO 14255:2001) using ion chromatography (IC) by the Dionex system ICS‑900, with conductivity detector DS5, applying Rfic TM IonPac®AG22, length 2 × 50mm Guard column. The results were analysed using the Chromilion software package. The eluent was used to the analysis, and it con‑ tained 4.5mM Na2CO3 and 1.4mM NaHCO3; applied flow was 0.25ml/min. The retention time for nitrite ions was 6.7min, while that for nitrate ions was equal to 9min. Results anddiscussion The application of PM as a chemical oxidant in manure can contribute to changes in physico‑chemical properties of both the selected substances (ALB, LEV) and the manure itself. The gradual progression of the reaction makes the PM an ideal stimulant of aerobic processes in manure. The manure stabilization lasts usually up to 35days and occurs in the period between its removal from the breeding room to the release onto the soil. Therefore, the tests conducted as part of this research lasted for 35days and corresponded to the most intense PM decomposition process. Apart from influence of PM on the dissolution velocity of drugs, the effectiveness and their removal are linked with soil solution–manure split factor and solubility constants for all reaction intermediates. The initial concentration of LEV and ALB was twice as high of magnitude above the determined average concen‑ tration in the environment. It was the simulation of events when the drugs are applied to all animals on the farm and due to the precision of the analytical methods. At the same time, the results of the study at such a high initial concentra‑ tion of pharmaceuticals (45mg/kg manure) can be used for intervention cases, such as those when the fertilizer origi‑ nate from excrements of treated animals should be stabi‑ lized. The analysis showed the desirability of the use of PM in such cases. We found that the positive effect obtained for ALB, which is slightly soluble in water and has a high coefficient of Kow (Table2). The use of PM significantly increases the desorption of this compound and its removal. In contrast, LEV is well soluble in water, which has a low Kow (Table2), and the use of PM, in this case, increases desorption. Probably, the kinetics described in high concentrations may occur in the environment due to the consumption of ALB and LEV. Benzimidazoles (among them ALB) are introduced into the environment when they are excreted by animals treated with this compound. It is expected that 100% of the prescribed dose is excreted within 7days. For cattle farms of ten animals per acre, it is conventionally recom‑ mended to treat the animals three times a year using approxi‑ mately 3.5g of benzimidazoles per animal per treatment. The amount of benzimidazole excreted onto 1 acre is about 110g/year. Because the benzimidazoles bind to humic mate‑ rial, they are not expected to run off into aquatic environ‑ ments (Technical Evaluation Report 2015). The influence ofPM (with CaO2) onthephysico‑chemical properties ofmanure contaminated withveterinary drugs The test result analysis presented in Fig.1 has demonstrated that for samples at day 35, with and without PM addition, the determined physical–chemical parameters of manure are not significantly different (taking variation coefficients into account). The basic parameter of soil fertility is the pH; it deter‑ mines the organic and mineral transformations, conver‑ sion of humic substances, and availability of nutrients and 317Chemical Papers (2020) 74:311–322 1 3 toxic components. Based on stoichiometric calculations, a calcium peroxide dose of 0.5‑g PM per 1kg of manure dry mass was determined; this dose was responsible for the increase of pH by one pH unit. It was determined that such dosage of the PM should not significantly influence important biological processes in the manure. Hence, it would not be toxic to the micro‑organisms that live in the manure, but it would accelerate the aerobic processes of biological transformations and degradation of the stud‑ ied drugs (LEV and ALB) due to better oxygenation of the environment. At day 35, pH values were lower than calculated based on theoretical assumptions. In the ALB samples, an increase of pH was about 0.3 units, and in the case of LEV samples, it was 0.17 units. Most probably, the buffering capacity of the manure caused such a result. The obtained results indicate a beneficial effect of the PM on the loss of the studied veterinary drugs (ALB and LEV) during manure stabilization, without a negative influence on the physico‑chemical parameters character‑ ising organic matter transformation (Fig.1). Neutralization of distinctive, unpleasant odor released during stabilization of organic fertilizers can be an addi‑ tional benefit of the PM addition process (Walawska etal. 2007). Fig. 1 Change of chosen physico‑chemical parameters of manure contaminated with LEV and ALB over time under influence of PM; a pH, b organic substances, c organic carbon (%), d total nitrogen (g/kg), e alkalinity (mmol H+/kg), and f hydrolytic acidity (mmol H+/kg) 318 Chemical Papers (2020) 74:311–322 1 3 The influence ofPM (with CaO2) ontheLEV andALB detection During the fertilizer stabilization process, the concentra‑ tions of drugs were measured. The initial concentration (day 0, time 0) was measured to be much lower than the introduced dose of 45mg/kgd.m. Drug concentration did not exceed 7.66‑mg LEV/kgd.m. and 2.65‑mg ALB/kgd.m. in the PM samples. Samples without PM contained 5.76‑mg LEV/kgd.m. and 2.25‑mg ALB/kg d.m. (Fig.2). At the start of the experiment, higher drug concentration was deter‑ mined in the series consisting of PM with ALB and LEV than in the series without the PM addition. This could have been related to the separation of the compounds from the manure sorption complex due to saturation with cal‑ cium (Ca2+) ions, which resulted in the compounds being more available and detected in higher concentrations. The PM addition resulted in an increase of the solution ionic strength. Consequently, it affected the liquid‑manure parti‑ tion coefficient and increased the solubility of drugs. The percentage loss of drugs was calculated as a difference in the concentration at day 0 and day 35, and accounted for sorption and PM activity. The net rate of drug loss was 95%, 94%, 87%, and 83% for samples MA, MACa, ML, and MLCa, respectively. Due to high removal rates achieved, it has been assumed that a dominant part of the drugs loss is associated with sorption. However, the sorp‑ tion process promotes accumulation (Kümmerer 2008), which in effect only delays the adverse impact of drugs as biologically active substances still persist in the environ‑ ment (Halling‑Sørensen etal. 1998). The sorption mechanisms and kinetic properties of ALB were investigated through sorption equilibrium and sorption rate experiments in the range of concentrations between 1‑ and 20‑mg/kg soil/sediments (Mutavdžic Pav‑ lovic etal. 2018). This study demonstrates that the ALB sorption affinity is strongly governed by physico‑chemi‑ cal characteristics of the sediment matrix (soil, sediment) as well as physico‑chemical characteristic of ALB. The benzimidazoles are generally insoluble in water, sticks to humic material in terrestrial and aquatic environments, and are readily photodegradable (Horvat etal. 2012). Levamisole is highly soluble in water and can runoff into the aquatic environment. It may decompose non‑ enzymatically. Depending on the temperature and pH, its decomposition results in information of one of three degra‑ dation products (Technical Evaluation Report 2015). One of the products shows the immunomodulatory activity. The presence of pharmaceutical compounds beyond the sorption complex, e.g., in the soil water, increases the like‑ lihood of their loss as a result of biological and chemical reactions. Based on the obtained results, differences in physico‑chemical properties of both drugs were observed. The ALB reached a maximum concentration in the studied solution much later than the LEV. The highest concentra‑ tion of LEV in manure was observed at the start of the experiment in MLCa samples (with PM) and ML samples (without PM); whereas, in ALB tests, the concentration was the highest on the 4th day in MACa samples (5.83‑mg ALB/kgd.m.) and on the 6th day in the MA series (3.42‑mg ALB/kgd.m.). The PM addition increased the ALB release to the environment. In the series with added PM, the ALB concentration increased twofold between day 0 and day 4, and then, it was gradually decreased (Fig.2). The manure sorption complex saturation with calcium ions was sig‑ nificantly slower for ALB than for LEV, which suggests lower ALB mobility in relation to the LEV. Until day 4, the increase of ALB concentration was attributed to two consecutive processes: desorption and degradation. From day 6 onwards, a decrease in ALB concentration was observed, which implies that the degradation process occurred at a faster rate than the desorption process. Desorption is of great importance for removal pro‑ cesses. The presence of substances in the solution increases their availability and thus increases the possi‑ bility of degradation, both chemical and biological. Fig. 2 The disappearance of pharmaceuticals (C/C0) over time and fitting the kinetic equation to the disappearance of pharmaceuticals for a ALB (MA series without PM, MACa series with PM) and b LEV (ML series without PM, MLCa series with PM) 319Chemical Papers (2020) 74:311–322 1 3 Loss ofdrugs duringmanure stabilization During the experiment, a reduction in the ALB content over 35days was found in all series (Fig.2). The initial ALB con‑ centration was 2.25mg/kgd.m. of manure for the MA series without the PM and 2.65mg/kgd.m. of manure for the MACa series with the PM. The ALB loss in the series with peroxide addition (MACa) was 84% (0.43‑mg/kgd.m. on day 35). In the MA series, the ALB removal did not exceed 41% (1.36‑mg/ kgd.m. on day 35), which was three times less compared to the MACa series. The effect of PM on drug loss was calculated taking into account the highest determined concentrations of drugs (time 0 for ML and MLCa, day 4 for MACa, and day 6 for MA). Estimated 92% loss of ALB in PM series in compari‑ son to 61% loss in non PM series is an evidence of a measur‑ able effect of PM. In the LEV case, the beneficial effect of PM used was less prominent. This study has shown a 1.4% difference in LEV removal (losses 98.4% for series ML and 97.0% for series MLCa). It could be assumed that in the case of this drug, the use of PM cannot be justified. The LEV loss exceeding 97% was obtained in both ML and MLCa series. However, given that the drug dose was the same (45mg/kgd.m.) in samples with and without the PM, the observed amount of LEV at time 0 was significantly below the dose (7.66mg/kgd.m. in samples with PM and 5.76mg/kgd.m. in samples without PM). Availability of LEV was 25% higher when the PM was added. These results indicate a supportive role of the PM (with CaO2) in the capacity to remove LEV from manure. Time series of drug concentrations over time enabled to establish tangents to the initial losses of drugs from the equa‑ tion y = ax + b (Fig.3). Based on the determined slope (incli‑ nation angle a = − tgα) of the tangents, initial reaction rates were compared. Based on the calculated rates, the influence of PM on the loss of LEV and ALB was determined. For both ALB and LEV, the PM (with CaO2) addition resulted in changes of the initial degradation rate. Negative α tangents were 0.0985 for ALB and 0.137 for LEV, whereas for the samples without PM, these values were lower for ALB (tg α = 0.034) and higher for LEV (tg α = 0.152). The above values confirm the augmenting influence of the PM on the loss of ALB and LEV in manure. The obtained results suggest a stronger PM influence on the degradation of a hydrophobic substance (ALB), as the rate of drug loss (value of tg α) was near twofold higher in comparison with the sample without the PM addition. In the case of a substance with mainly hydrophilic properties (LEV), the change of tg α value was not statistically sig‑ nificant (coefficient of variation was 7% between ML and MLCa). The kF constant as kLEV for LEV and k ALB for ALB was calculated according to Eq.(2). The resulting values of k ALB were variable and did not show any trend with its concentration. In contrast, kLEV in the range of 4–21days was 143.5day‑1 and 86.7day‑1 for ML for MLCa, respectively. To demonstrate the impact of PM on ALB and LEV con‑ versions in poultry manure, the influence of determined ini‑ tial concentration (C0) as a change in “a” coefficient values in the formula y = ax + b was assessed (at confidence level 95% and critical value tα = 3.1824) (Table4). Statistically significant differences in conversions of ALB and LEV influ‑ enced by the addition and absence of inorganic PM were observed. Further investigations focused on assessing the impact of PM on the reaction rates. Linear correlation of initial conver‑ sion rates separately for the samples with and without PM was also tested by determining Pearson’s “R” coefficient. It was assumed that in the absence of drugs, the reaction rate would be 0, and hence, the intersection was at point 0 (Fig.4). Fig. 3 Determination if the initial reaction rate for a ALB (MA series without PM and MACa series with PM) and b LEV (ML series with‑ out PM and MLCa series with PM)