Screening of indicator pharmaceuticals and personal care products in landfill leachates: a case study in Shanghai, China

Highlights

● A systematic framework was developed to identify i-PPCPs for landfill leachate.

● The wide-scope target analysis offered a basis for comprehensive i-PPCP screening.

● Source-specificity and representativeness analysis helped to refine i-PPCPs.

● Erythromycin, gemfibrozil and albendazole were identified as i-PPCPs for leachate.

Abstract

Identifying potential sources of pharmaceuticals and personal care products (PPCPs) in the environment is critical for the effective control of PPCP contamination. Landfill leachate is an important source of PPCPs in water; however, it has barely been involved in source apportionment due to the lack of indicator-PPCPs (i-PPCPs) in landfill leachates. This study provides the first systematic framework for identifying i-PPCPs for landfill leachates based on the wide-scope target monitoring of PPCPs. The number of target PPCPs increased from < 20 in previous studies to 68 in the present study. Fifty-nine PPCPs were detected, with median concentrations in leachate samples ranging from below the method quantification limit (MQL) to 41 μg/L, and 19 of them were rarely reported previously. A total of 29 target compounds were determined to be PPCPs of high concern by principal component analysis according to multiple criteria, including occurrence, exposure potential, and ecological effect. Coupled with source-specificity and representativeness analysis, erythromycin, gemfibrozil, and albendazole showed a significant difference in their occurrence in leachate compared to other potential sources (untreated and treated municipal wastewater and livestock wastewater) and correlated with total PPCP concentrations; these were recommended as i-PPCPs for leachates. Indicator screening procedure can be used to develop a sophisticated source apportionment method to identify sources of PPCPs from adjacent landfills.

Key words： Landfill leachates; PPCPs; Indicator; Screening criteria; Source-specificity

1 Introduction

The prevalence of pharmaceuticals and personal care products (PPCPs) in the environment has generated increasing concern due to the potential threats they pose to the ecosystem and human health (Daughton and Ternes, 1999; Meyer et al., 2019; Meng et al., 2021). PPCPs consist of different classes of compounds, including antibiotics, adrenergic agents, anthelmintics, anticoagulants, antidepressants, hypoglycemic agents, and lipid regulators, etc. They are continuously discharged from various emission sources (Kasprzyk-Hordern et al., 2009; Sui et al., 2017; Yu et al., 2020b; Arvaniti et al., 2022). Because of this, PPCP source identification is indispensable for effective control of PPCP discharge, to reduce the risks to aquatic environments (Sun et al., 2016; Mei et al., 2019; Zhu et al., 2022).

Numerous approaches have been developed to track PPCPs, and indicator-based methods have been applied widely for source identification (Nakada et al., 2008; Sun et al., 2016; Currens et al., 2019). To date, there have been a few studies on screening indicators (Table S1), the majority of which have been based on several criteria: concentration, detection frequency, and detection ratio (defined as measured concentration divided by the limit of quantification) (Dickenson et al., 2011; Jekel et al., 2015; Cantwell et al., 2018; Tran et al., 2019). In most cases, compounds chosen as indicators due to their higher concentrations and higher detection frequency were not investigated to verify their source-specificity (Fenech et al., 2013; Yang et al., 2017). Sun et al. (2016) screened acetaminophen as an indicator for domestic wastewater due to its high concentration (0.10–3.5 μg/L), but an even higher concentration (0.60–90 μg/L, Wu et al. (2021)) was reported in livestock wastewater in China. Lack of source specificity may lead to bias in source apportionment when different emission sources are involved in the studied region. Likewise, the fate and transport of indicators should be considered meticulously in indicator screening as these factors influence their occurrence in the aquatic environment (Jekel et al., 2015). If contaminant concentrations are easily reduced by hydrolysis, photodegradation, biodegradation and sorption (Scheurer et al., 2011; Foolad et al., 2015; Yuan et al., 2019), their persistence and transportability will be impacted, making them poor indicators. For instance, tetracyclines, which have high consumption rates in animal husbandry and aquaculture (Wan et al., 2021), were found to be prone to adsorption and deposition in soil and sediments (Mejías et al., 2021; Scaria et al., 2021), making them poor indicators for livestock wastewater.

Recently, investigations on PPCP occurrence and characteristics in municipal solid waste (MSW) landfills have indicated that landfill leachates are an under-recognized source of PPCPs (Yi et al., 2017; Chung et al., 2018; Wang et al., 2021; Yu et al., 2021). Landfill leachate discharge reaching surrounding aquatic environments unintentionally could result in high environmental risks (Qi et al., 2018; Stefania et al., 2019; Yu et al., 2020a). Notably, in most cases, landfill leachate is not the only source of PPCPs in the adjacent region. Rural areas, where MSW landfills are usually located, can also be contaminated by domestic wastewater, livestock wastewater and other potential PPCP sources. Source apportionment, therefore, is an effective approach to identify where PPCPs in the surface water and groundwater originated from. Unfortunately, while there are many studies on the indicators for other emission sources that can be used for source apportionment, indicators for landfill leachate are limited. Our previous study (Wu et al., 2021) developed an analytical method using wide-scope target PPCPs for landfill leachate and proposed albendazole as a promising indicator as it had the highest concentration in landfill leachates among the target PPCPs. Despite progress in this area, a more comprehensive framework of indicator screening is needed to identify the most suitable indicators of PPCPs in landfill leachates.

This study aimed to develop a systematic framework for the identification of indicator-PPCPs (i-PPCPs) in raw landfill leachate samples. A total of 68 PPCPs were simultaneously analyzed in the leachate samples collected from an MSW landfill in Shanghai, China. Principal component analysis (PCA) was conducted to identify PPCPs of high concern according to the occurrence, exposure potential, and ecological effect, to ensure the practicality of using the proposed indicators in the aquatic environment. Finally, the source-specificity and representativeness of i-PPCPs were verified by comparison to other emission sources. Applying the screening framework with statistical analysis, the results will be helpful to implement source apportionment in the vicinity of landfills.

2 Materials and methods

The key concepts of the indicator screening method are illustrated in Fig.1. First, the occurrence of PPCPs in landfill leachates was investigated, following which the detected PPCPs were ranked by PCA according to their abundance in landfill leachates, persistence and risk to the aquatic environment. From this, PPCPs of high concern were screened, and the source-specificity of PPCPs of high concern was verified by comparing their detection frequencies and concentrations with other emission sources. Finally, i-PPCPs were identified through correlation analysis and significance analysis.

Fig.1 Method for the screening of i-PPCPs in landfill leachates.

Full size|PPT slide

2.1 Sampling and analysis of environmental samples

Understanding the occurrence of PPCPs in landfill leachates is the basis of screening indicators. In the previous studies of PPCPs in landfill leachates, only 10–20 target compounds were included (Clarke et al., 2015; Yao et al., 2020). It is insufficient to screen the appropriate i-PPCPs from the limited number of analytes. Thus, we analyzed 68 PPCPs in the leachate samples collected from a typical MSW landfill to expand the knowledge of their occurrence in landfill leachates.

2.1.1 Sampling

Landfill leachates were taken from the MSW landfill (31°06′ N, 121°87′ E) in Shanghai, China, which is equipped with an advanced leakage prevention system and a leachate collection system. It receives approximately 8000 t of MSW per day, accounting for 70% of the MSW in Shanghai, and generates 1000–1500 m³ of leachates per day. In total, five sampling campaigns were conducted between 2019–2020 (November and December of 2019, and in September, October and November of 2020) to obtain comprehensive profiles of PPCPs in raw landfill leachates. To conduct source-specificity analysis for PPCPs of high concern, wastewater influent and effluent samples from six municipal WWTPs were also collected in November 2020, along with livestock wastewater samples from seven concentrated animal feeding operations during 2019–2020 in the studied region (Tables S2 and S3). At all sampling sites, grab samples were filled in 500 mL amber glass bottles, immediately sent to the laboratory in ice-packed containers, and pretreated within 48 h.

2.1.2 Pretreatment and analysis

Sixty-eight PPCPs belonging to 19 therapeutic classes (Tab.1) were chosen as target compounds in this study. All PPCP standards of high purity grade (> 98%) and 12 isotopically labeled internal standards (ILISs) were purchased from Cayman (Ann Arbor, MI, USA), CDN Isotopes (Quebec, Canada), Cambridge Isotope Laboratories (Andover, MA, USA), Dr. Ehrenstorfer (Augsburg, Germany), Sigma-Aldrich (St. Louis, MO, USA), TCI (Shanghai) (Shanghai, China) and Toronto Research Chemicals (Toronto, Canada). The information on ILISs is provided in the supporting information.

Tab.1 Target PPCPs detected in landfill leachate (n = 5), untreated municipal wastewater (n = 6), treated municipal wastewater (n = 6) and livestock wastewater (n = 7) in the studied region

Category	No	Compound	Abbreviation	CAS	Therapeutic class
Antibiotic PPCPs	1	Azithromycin	ATM	83905-01-5	Macrolides
	2	Clarithromycin	CTM	81103-11-9	Macrolides
	3	Erythromycin	ETM	23893-13-2	Macrolides
	4	Leucomycin	LEM	1392-21-8	Macrolides
	5	Roxithromycin	RTM	80214-83-1	Macrolides
	6	Tylosin	TYL	1401-69-0	Macrolides
	7	Flumequine	FQ	42835-25-6	Fluoroquinolones
	8	Oxolinic acid	OA	14698-29-4	Fluoroquinolones
	9	Clinafloxacin	CLX	105956-97-6	Fluoroquinolones
	10	Ciprofloxacin	CPX	85721-33-1	Fluoroquinolones
	11	Danofloxacin	DAX	112398-08-0	Fluoroquinolones
	12	Difloxacin	DIX	98106-17-3	Fluoroquinolones
	13	Enrofloxacin	EFX	93106-60-6	Fluoroquinolones
	14	Lomefloxacin	LFX	98079-51-7	Fluoroquinolones
	15	Marbofloxacin	MAX	115550-35-1	Fluoroquinolones
	16	Norfloxacin	NFX	70458-96-7	Fluoroquinolones
	17	Ofloxacin	OFX	82419-36-1	Fluoroquinolones
	18	Pefloxacin	PFX	70458-92-3	Fluoroquinolones
	19	Sarafloxacin	SFX	98105-99-8	Fluoroquinolones
	20	Sparfloxacin	SPX	110871-86-8	Fluoroquinolones
	21	Sulfaclorazina	SC	80-32-0	Sulfonamides
	22	Sulfadiazine	SD	68-35-9	Sulfonamides
	23	Sulfadimethoxine	SDM	122-11-2	Sulfonamides
	24	Sulfameter	SF	651-06-9	Sulfonamides
	25	Sulfaguanidine	SG	57-67-0	Sulfonamides
	26	Sulfisomidine	SIM	515-64-0	Sulfonamides
	27	Sulfisoxazole	SIX	127-69-5	Sulfonamides
	28	Sulfamerazine	SMR	127-79-7	Sulfonamides
	29	Sulfamethazine	SMT	57-68-1	Sulfonamides
	30	Sulfamethoxazole	SMX	723-46-6	Sulfonamides
	31	Sulfamethiazole	SMZ	144-82-1	Sulfonamides
	32	Sulfaphenazole	SPZ	526-08-9	Sulfonamides
	33	Sulfaquinoxaline	SQX	59-40-5	Sulfonamides
	34	Sulfathiazole	STZ	72-14-0	Sulfonamides
	35	Chlortetracycline	CTC	57-62-5	Tetracyclines
	36	Demeclocycline	DMC	127-33-3	Tetracyclines
	37	Doxycycline	DTC	564-25-0	Tetracyclines
	38	Oxytetracycline	OTC	79-57-2	Tetracyclines
	39	Tetracycline	TC	60-54-8	Tetracyclines
	40	Chloramphenicol	CP	56-75-7	Other antibiotics
	41	Florfenicol	FF	73231-34-2	Other antibiotics
	42	Lincomycin	LIN	154-21-2	Other antibiotics
	43	Tiamulin	TIA	55297-95-5	Other antibiotics
	44	Trimethoprim	TP	738-70-5	Other antibiotics
Non-antibiotic PPCPs	45	Salbutamol	SAL	18559-94-9	Adrenergic agent
	46	Cimetidine	CIM	51481-61-9	Antagonist
	47	Albendazole	ABZ	54965-21-8	Anthelmintics
	48	Fenbendazole	FBZ	43210-67-9	Anthelmintics
	49	Theophylline	THP	58-55-9	Anti-asthmatic agent
	50	Triclosan	TCS	3380-34-5	Antibacterial agent
	51	Warfarin	WAR	81-81-2	Anticoagulant
	52	Fluoxetine	FLU	54910-89-3	Antidepressants
	53	Sulpiride	SP	15676-16-1	Antidepressants
	54	Diltiazem	DIL	42399-41-7	Antihypertensive agent
	55	Crotamitone	CRO	483-63-6	Antipruritic
	56	Carbamazepine	CBZ	298-46-4	Anti-seizure
	57	Gliclazide	GLI	21187-98-4	Hypoglycemic agents
	58	Glyburide	GLY	10238-21-8	Hypoglycemic agents
	59	Tolbutamide	TOL	64-77-7	Hypoglycemic agents
	60	DEET	DEET	134-62-3	Insect repellent
	61	Bezafibrate	BF	41859-67-0	Lipid regulators
	62	Gemfibrozil	GF	25812-30-0	Lipid regulators
	63	Acetaminophen	ACE	103-90-2	NSAID a)
	64	Diclofenac	DCF	15307-86-5	NSAID
	65	Naproxen	NAP	22204-53-1	NSAID
	66	Caffeine	CF	58-08-2	Stimulant
	67	Atenolol	ATE	29122-68-7	β-Blockers
	68	Metoprolol	MTP	51384-51-1	β-Blockers

Notes: a) NSAIDs: Non-steroidal anti-inflammatory drugs.

Target PPCPs were divided into two groups (Group I and Group II), extracted with different solid-phase extraction (SPE) procedures, and analyzed by high-performance liquid chromatography with tandem mass spectrometry (HPLC-MS/MS, LCMS-8050, Shimadzu, Japan) in multiple reaction monitoring (MRM) mode (Wu et al., 2021). Briefly, samples were diluted with Milli-Q water if necessary (landfill leachate: 50 times; livestock wastewater: 25 times) and were filtered using 0.45 μm glass fiber filters (Whatman, UK). Duplicate samples (100 mL for diluted landfill leachate, diluted livestock wastewater, and WWTP influent; 400 mL for WWTP effluent) were spiked with Na₂EDTA (1.0 g/L sample) and ILIS solution (100 μL × 1000 μg/L), and then adjusted to pH = 6.5 ± 0.2 (for Group I) and pH = 2.5 ± 0.2 (for Group II) using 1 mol/L H₂SO₄ and NaOH solutions. Samples were loaded on Oasis HLB SPE cartridges (500 mg, 6 mL, Waters) at a flow rate of 1 mL/min after the cartridges were preconditioned with 6 mL MeOH followed by 3 washes with 6 mL Milli-Q water. The cartridges were then rinsed with 5 mL 5% MeOH (v/v) solution and dried by vacuum for over 1 h. Finally, the SPE cartridges were eluted by washing twice with 5 mL MeOH at a flow rate of 1 mL/min. Extracts were evaporated to approximately 0.2 mL under a gentle nitrogen stream at 40 °C, reconstituted by adding MeOH to a final volume of 1.0 mL, and filtered through a PTFE membrane filter (Millex-FG, Millipore, USA). The working conditions of HPLC-MS/MS analysis are provided in the supporting information.

2.1.3 Quality assurance and quality control (QA/QC)

Calibration curves were established using at least five solutions of different concentrations for each chemical, and the correlation coefficients (R²) were over 0.99. The method detection limit (MDL) was calculated by the standard deviation of the replicates of low concentration standards, and t-statistic with (n–1) degree of freedom (n represents the number of replicates) at a 99% confidence level (USEPA, 2016); method quantification limit (MQL) was estimated as three times the MDL value (Renganathan et al., 2021). The MDLs and MQLs were 0.08–1.4 ng/L and 0.23–4.0 ng/L, respectively; recoveries of most PPCPs (in landfill leachates, influent and effluent of WWTPs, and livestock wastewater) ranged between 60% and 140%, and the relative standard deviations of most duplicate samples were less than 25% (Table S4). In each sampling campaign, two laboratory blanks and one field blank were extracted and analyzed identically to leachate samples, and none of the analytes were quantifiable in the blanks. For the statistical analysis, concentrations below the MQL were considered to be zero (Köck-Schulmeyer et al., 2021).

2.2 Ranking of PPCPs in landfill leachates

2.2.1 Criteria for PCA analysis

The criterion used to rank PPCPs consisted of occurrence, exposure potential, and ecological effect (Tab.2). Occurrence included concentration (median value) and detection frequency of PPCPs in landfill leachates. Exposure potential was composed of persistence and transportability, represented by the degradation half-lives and organic carbon adsorption coefficients (K_oc), respectively; substances with higher persistence and lower K_oc were assigned a higher score for exposure potential. The ecological effect was composed of bioaccumulation and eco-toxicity. The index of bioaccumulation was estimated by the octanol/water partitioning coefficient (K_ow) (Li et al., 2019), where a higher K_ow suggested a higher possibility of bio-concentration. Eco-toxicity was estimated from data on the toxicity to aquatic species, including fish, daphnia, and green algae.

Tab.2 Criteria and corresponding utility functions used to rank PPCPs in landfill leachates

Criteria	Index	Data required	Utility functions
Occurrence (O)	Detection frequency (DF)	Overall detection frequency (F)	\text{U} \left(\text{DF}\right)=\mathrm{F}
	Detection concentration (DC)	Median concentration (C)	\text{U} \left(\text{DC}\right) \text{=}{\displaystyle \frac{{\text{log}}_{\text{10}}(C)-{\text{log}}_{\text{10}}\text{(}{C}_{\text{min}}\text{)}}{{\text{log}}_{\text{10}}\text{(}{C}_{\text{max}}\text{)}-{\text{log}}_{\text{10}}\text{(}{C}_{\text{min}}\text{)}}}
Exposure potential (P)	Persistence (P)	Half-life (H)	\text{U} \left(\text{P}\right) \text{=}{\displaystyle \frac{{\text{log}}_{\text{10}}(H)-{\text{log}}_{\text{10}}(H_{\text{min,2}}\text{)}}{{\text{log}}_{\text{10}}\text{(}{H}_{\text{max,2}}\text{)}-{\text{log}}_{\text{10}}\text{(}{H}_{\text{min,2}}\text{)}}}
	Transportability (T)	Koc	\text{U} \left(\text{T}\right) \text{=}{\displaystyle \frac{{{\text{log}}_{\text{10}}\text{(}{K}_{\text{oc}}\text{)}}_{\text{max}}-{\text{log}}_{\text{10}}\text{(}{K}_{\text{oc}}\text{)}}{{{\text{log}}_{\text{10}}(K_{\text{oc}}\text{)}}_{\text{max}}-{{\text{log}}_{\text{10}}\text{(}{K}_{\text{oc}}\text{)}}_{\text{min}}}}
Ecological effect (E)	Bioaccumulation (B)	Kow	\text{U} \left(\text{B}\right) \text{=}{\displaystyle \frac{{\text{log}}_{\text{10}}{K}_{\text{ow}}{-{\text{log}}_{\text{10}}\text{(}{K}_{\text{ow}}\text{)}}_{\text{min}}}{{{\text{log}}_{\text{10}}\text{(}{K}_{\text{ow}}\text{)}}_{\text{max}}-{{\text{log}}_{\text{10}}\text{(}{K}_{\text{ow}}\text{)}}_{\text{min}}}}
	Eco-toxicity (E)	RQ	\text{U} \left(\text{E}\right) \text{=}{\displaystyle \frac{{\text{log}}_{\text{10}}(RQ)-{\text{log}}_{\text{10}}\text{(}{RQ}_{\text{min}}\text{)}}{{\text{log}}_{\text{10}}\text{(}{RQ}_{\text{max}}\text{)}-{\text{log}}_{\text{10}}\text{(}{RQ}_{\text{min}}\text{)}}}

2.2.2 Data collection and processing

Besides the data obtained by analyzing the collected leachate samples in this study, the occurrence data of 68 PPCPs in leachates detected in the investigated landfill in our previous study (Yu et al., 2021) were also included. Half-life and K_oc data of PPCPs were accessed from the QSAR model in OPERA (version 1.5). The K_ow data of PPCPs were mainly extracted from literature in addition to the ECOTOX database (USEPA). No observed effect concentration (NOEC), half-maximal effective concentration (EC₅₀), or median lethal concentration (LC₅₀), were used to indicate the toxicity of PPCPs, obtained mainly from the ECOTOX database, and partly predicted from the ECOSAR model (version 2.2). All the data employed in this study are listed in Tables S5 and S6.

The data were normalized to dimensionless terms in the range of 0–1 by utility functions (Tab.2), to provide a reasonable distribution of each index and facilitate the ranking of PPCPs. The values of concentration (median), persistence and eco-toxicity were transformed by {\text{log}}_{\text{10}}\text{(}\text{x} \text{)} to avoid excessive cluster decentralization and poor parameter discrimination. For the index of eco-toxicity, the risk quotient (RQ), calculated according to the median concentration value, was applied to construct a utility function, as described in the supporting information. The normalized data are provided in Table S7.

The distribution of data for each index (Fig. S1) presented a reasonable numerical distribution and similar numerical range, which improved the comparability of different indices. For the half-life index, the second maximum and minimum were utilized to replace the maximum and minimum during the calculation to avoid excessive cluster decentralization; the maximum and minimum outliers were assigned as 1 and 0 respectively. For the PPCPs of which the half-life and eco-toxicity were not available, utility function values were set to 1 (Kumar and Xagoraraki, 2010).

The significance analysis of PPCPs tested for source-specificity and representativeness was performed by the Kruskal–Wallis test (p < 0.05) in SPSS 26.0 (IBM).

2.2.3 Uncertainty

The uncertainty of individual indices was set at 0, 0.25, and 0.5 for indices based on experimental data, predicted data, and unavailable data (Li et al., 2019; Zhong et al., 2022), respectively. Considering the half-life and K_oc data of PPCPs could only be acquired from the model, uncertainty scores of available persistence and transportability were set to 0 (Li et al., 2019). The arithmetic mean of uncertainty scores for six indexes was calculated as the overall uncertainty score for each PPCP.

3 Results and discussion

3.1 Overall occurrence of PPCPs in landfill leachates

The concentrations and detection frequencies of 68 target PPCPs in landfill leachates are presented in Fig.2. Fifty-nine out of the 68 target PPCPs were detected at least once, and 35 had detection frequencies > 60% demonstrating pervasiveness in leachates. For instance, DCF, GF, and DEET were found at high concentrations (2.3–58 μg/L) in landfill leachates, similar to those reported in China and USA (0.081–46 μg/L) (Lu et al., 2016; Masoner et al., 2016). It is worth noting that 19 PPCPs (ATM, CLX, DAX, SPX, SDM, SMZ, SPZ, SQX, DTC, LIN, CIM, ABZ, FBZ, WAR, SP, CRO, GLI, GLY, and TOL), rarely monitored in previous studies, were detected in 82% of leachate samples with some found in high abundance. Among 19 PPCPs, macrolide ATM, antidepressants SP, and anthelmintic ABZ were detected at 3.9–9.7, 2.23–11.2, and 0.53–30 μg/L, respectively, in all leachate samples. Some of the 19 PPCPs (GLY, SP, and ATM) may present high environmental risks due to their low values of predicted no effect environmental concentration (PNEC) (GLY: 0.017 ng/L, SP: 16 ng/L, and ATM: 20 ng/L). The high abundances and environmental risks found for PPCPs that were rarely monitored previously verified the significance of the wide-scope analysis, and supported it providing the basis for more comprehensive i-PPCP screening.

Fig.2 Median concentration and detection frequency of 68 PPCPs in the collected landfill leachates (n = 5).

Full size|PPT slide

3.2 PPCPs of high concern in leachates

3.2.1 Ranking of PPCPs

Conducting the PCA analysis, the first two components explained 84.1% of the cumulative variance, indicating that PC1 and PC2 played a major role in the total variance (Fig.3(a)). Occurrence and ecological effect were positively correlated with PC1 (coefficients were 0.52 and 0.72 respectively), suggesting that the PC1 score reflects the level of occurrence and ecological effect. The positive correlation between exposure potential and PC2 indicated the strong likelihood of the PC2 score representing exposure potential, referring to the retained capacity of PPCPs in the dissolved phase. Together, the PC1 and PC2 scores were combined to calculate the overall scores (Sharma, 2008; Thomaz and Giraldi, 2010) of 68 PPCPs and to obtain the ranking list of PPCPs in landfill leachates, as detailed in the supporting information (including Fig. S2 and Table S8).

Fig.3 PCA analysis for three criteria of occurrence, exposure potential and ecological effect (triangle symbols refer to individual PPCPs) (a). Ranking list of PPCPs in landfill leachate (the cut-offs for inclusion in groups I, II, III, and IV were 0.55, 0.45, and 0.30, respectively) (b). Contribution of individual criteria (occurrence, exposure potential and ecological effect) to the total score of each PPCP of high concern (c).

Full size|PPT slide

The 68 PPCPs were classified into four groups based on overall score distribution with cut-offs of 0.55, 0.45, and 0.3 (Li et al., 2019; Zhong et al., 2022) (Fig.3(b)): Group I (29 compounds, top importance), Group II (15 compounds, moderate importance), Group III (14 compounds, low importance), and Group IV (10 compounds, no importance). Twenty-nine PPCPs were classified in Group I as PPCPs of high concern, and contributions of individual criteria to the total score of each PPCP of high concern are exhibited in Fig.3(c).

The contribution of the three criteria for most PPCPs of high concern was balanced, suggesting these PPCPs were abundant, persistent and of high risk. Several PPCPs such as DCF, ATM, and ETM were identified as high-concern contaminants due to their abundance and high ecological effect (Sui et al., 2012; Li et al., 2019; Li et al., 2020). The PPCPs of high concern in leachates were quite different from the results in other aquatic environments. The discrepancy could be attributed to PPCP characteristics in leachates, and more importantly, the integration of results by using multiple criteria. In most studies, exposure potential is hardly involved in the screening criteria, leading to contaminants of high concern impractical for source identification. Kumar and Xagoraraki (2010) ranked demeclocycline as a contaminant of high concern in American waters, even though demeclocycline tends to be adsorbed or deposited in soil and sediments (Scheurer et al., 2011; Foolad et al., 2015) due to the lack of exposure potential as the screening criterion. Here, the multiple criteria-based ranking approach circumvents these issues and is recommended for the exploration of distinctive and practical PPCPs of high concern.

3.2.2 Uncertainty

The uncertainty of the ranking list was mainly attributed to the data gaps in bioaccumulation (K_ow), eco-toxicity (RQ) and persistence (half-life) (Fig. S3a). The predicted K_ow of 12 PPCPs including EFX, DAX, and SFX had a large contribution (52.2%) to the total uncertainty; 39.1% of the uncertainty was explained by predicted RQ values of seven PPCPs (DAX, DIX, MAX, PFX, SPX, SF, and CRO) and inaccessible RQ values for SC and TOL due to the unavailable toxicity data. In future study, assessments of bioaccumulation and eco-toxicity should be improved to reduce uncertainties associated with data gaps. As 74% of PPCPs had total uncertainty scores of 0 and 97% of PPCPs had uncertainty scores of < 0.10 (Fig. S3b), the overall uncertainty of the ranking list was acceptable.

3.3 Source-specificity of PPCPs of high concern for landfill leachates

Source-specificity is one of the prerequisites for indicator screening (Fenech et al., 2013; Yang et al., 2017); both detection frequency and concentration of indicators in leachates should be significantly higher than those found in other pollution sources adjacent to the landfill. Among 29 PPCPs of high concern, 20 PPCPs had detection frequencies > 80% in leachate samples (Fig. S4); their concentrations were compared with those in untreated municipal wastewater, treated municipal wastewater and livestock wastewater in the studied region, as shown in Fig.4 and Table S9.

Fig.4 Median concentrations of PPCPs in landfill leachates, untreated municipal wastewater, treated municipal wastewater, livestock wastewater.

Full size|PPT slide

Twelve PPCPs (ATM, ETM, CIM, ABZ, SP, CRO, CBZ, GLI, DEET, BF, GF, MTP) were found at the highest concentration in leachates and significantly differed from other pollution sources (p < 0.05, Table S10). For non-antibiotic PPCPs, nonprescription pharmaceutical CIM, CRO, and insect-repellent DEET were detected up to 5.7–41 μg/L in landfill leachates, distinctly greater than those in other three sources (0.01–5.0 μg/L) by two orders of magnitudes. This was in accordance with the frequent household use of these non-antibiotics and the largest proportion of household waste ending up in MSWs in landfills, leading to massive repositories in landfill leachates (Clarke et al., 2015; Yao et al., 2020). The distinctly high concentration of non-antibiotic PPCPs compared to other pollution sources suggested their source-specificity in landfill leachates.

By contrast, most antibiotics of high concern exhibited lower concentration levels in landfill leachates than those in livestock wastewater. Sulfonamides (SD, SMT and SMZ) were detected at 3.3–29 μg/L in livestock wastewater, much higher than the concentrations measured in leachates (1.2–3.6 μg/L); this is likely due to the extensive antibiotic use (accounting for 52%–82% of the total) in the animal breeding industry (Zhang et al., 2015; Charuaud et al., 2019; Song et al., 2022). Thereby, most antibiotics were prone to be source-specific for livestock wastewater, consistent with the findings of Duan et al. (2021) and Sim et al. (2013); however, macrolide ATM and ETM were two exceptions, as their concentrations in landfill leachates were significantly higher than those in livestock wastewater according to the Kruskal–Wallis test (p = 0.018 and 0.002 respectively). Because of this, ATM and ETM could serve as i-PPCP candidates for landfill leachates.

3.4 Representativeness of indicator PPCPs in landfill leachates

Representativeness analyses, including correlation analysis and significance analysis, were used to refine i-PPCPs for landfill leachates. The total PPCPs in leachates were significantly correlated with three compounds (Fig.5(a)): ETM with ∑antibiotic PPCPs (Pearson’s r = 0.953, p < 0.05), and GF and ABZ with ∑non-antibiotic PPCPs (Pearson’s r = 0.939–0.974, p < 0.05). Linear fittings of ETM, GF and ABZ with the two therapeutic classes (R² = 0.909 and 0.683, respectively; p < 0.05) were generated, suggesting that ETM, GF, and ABZ could represent the characteristics of total PPCPs in leachates (Fig.5(b) and 5(c)); therefore, these compounds were recommended as i-PPCPs for landfill leachates in the studied region.

Fig.5 Representative analysis of i-PPCP candidates and sum concentrations of antibiotics and non-antibiotics. The result of correlation analysis (a); linear fitting between concentrations of ETM and total concentrations of antibiotics (b); linear fitting between sum concentrations of GF and ABZ and total concentrations of non-antibiotics (c).

Full size|PPT slide

The representativeness of ETM and GF as i-PPCPs for landfill leachates can be verified by their predominance in leachates found elsewhere (Clarke et al., 2015; Yi et al., 2017; Chung et al., 2018; Wang et al., 2021). In addition, the concentration of GF exhibited no significant seasonal variation (Yu et al., 2021), indicating that it could be reliable as i-PPCPs regardless of the sampling seasons. As ABZ had hardly been investigated in landfill leachates, its occurrence in landfill leachates in other regions should be investigated in the future to evaluate the applicability of ABZ as an i-PPCP for landfill leachates in different locations. As i-PPCPs of landfill leachate were screened based on the data in Shanghai during 2019–2020, there is the possibility that the data may be inappropriate for extrapolation to other periods. The regionally representative i-PPCPs at different periods should be studied according to the developed framework for indicator screening.

The i-PPCPs screened in this study (ETM, GF, and ABZ) should be added to the target list of routinely monitored compounds to investigate the influence of landfill leachate on PPCP contamination in regions adjacent to landfills. The narrowed number of target PPCPs can save time, labor, and cost for monitoring and analysis. Moreover, the identified i-PPCPs for landfill leachate can be used in source apportionment (by introduction into the characteristic matrix model, as done by Kan et al. (2022)) to expand the source types that can be traced and to quantify the contribution of diverse emission sources, especially in the vicinity of landfills.

4 Conclusions

In this study, a systematic indicator screening method was developed to identify i-PPCPs for landfill leachates. In the wide-scope target monitoring of PPCPs, 59 compounds were detected with median concentrations of < MQL–41 μg/L in the landfill leachate samples, out of which 19 had rarely been reported previously. A total of 29 target compounds were identified as PPCPs of high concern by a multiple criteria-based ranking approach, due to their abundance in leachates, persistence and risk to the aquatic environment. Coupled with source-specificity analysis, i-PPCPs for leachates were further refined to 12 candidates, whose concentration in leachates significantly differed from those in untreated municipal wastewater, treated municipal wastewater, and livestock wastewater. Finally, ETM, GF and ABZ were selected as the chosen i-PPCPs for the studied region, due to their representativeness in landfill leachates. This framework of indicator screening provides a comprehensive approach for identifying practical and distinctive i-PPCPs for leachates and will help conduct the further source apportionment of landfill leachates in the water environment.

Acknowledgements

This research was partly supported by the National Natural Science Foundation of China (Nos. 21777042 and 22076045), the Open Research Fund of State Environmental Protection Key Laboratory of Ecological Effect and Risk Assessment of Chemicals (China) (No. 2022KFYB03), the Science and Technology Commission of Shanghai Municipality’s Yangfan Special Project (China) (No. 23YF1408400), the project supported by Shanghai Talent Development Funding (China).

Electronic Supplementary Material

Supplementary material is available in the online version of this article at https://doi.org/10.1007/s11783-023-1716-y and is accessible for authorized users.

References

1 Arvaniti O S, Dasenaki M E, Asimakopoulos A G, Maragou N C, Samaras V G, Antoniou K, Gatidou G, Mamais D, Noutsopoulos C, Frontistis Z, Thomaidis N S, Stasinakis A S. (2022). Effectiveness of tertiary treatment processes in removing different classes of emerging contaminants from domestic wastewater. Frontiers of Environmental Science & Engineering, 16(11): 148

2 Cantwell M G, Katz D R, Sullivan J C, Shapley D, Lipscomb J, Epstein J, Juhl A R, Knudson C, O’mullan G D. (2018). Spatial patterns of pharmaceuticals and wastewater tracers in the Hudson River Estuary. Water Research, 137: 335–343

3 Charuaud L, Jarde E, Jaffrezic A, Thomas M F, Le Bot B. (2019). Veterinary pharmaceutical residues from natural water to tap water: sales, occurrence and fate. Journal of Hazardous Materials, 361: 169–186

4 Chung S S, Zheng J S, Burket S R, Brooks B W. (2018). Select antibiotics in leachate from closed and active landfills exceed thresholds for antibiotic resistance development. Environment International, 115: 89–96

5 Clarke B O, Anumol T, Barlaz M, Snyder S A. (2015). Investigating landfill leachate as a source of trace organic pollutants. Chemosphere, 127: 269–275

6 Currens B J, Hall A M, Brion G M, Fryar A E. (2019). Use of acetaminophen and sucralose as co-analytes to differentiate sources of human excreta in surface waters. Water Research, 157: 1–7

7 DaughtonC G, Ternes T A (1999). Pharmaceuticals and personal care products in the environment: agents of subtle change? Environmental Health Perspectives, 107(suppl 6): 907–938

8 Dickenson E R V, Snyder S A, Sedlak D L, Drewes J E. (2011). Indicator compounds for assessment of wastewater effluent contributions to flow and water quality. Water Research, 45(3): 1199–1212

9 Duan L, Zhang Y Z, Wang B, Zhou Y T, Wang F, Sui Q, Xu D J, Yu G. (2021). Seasonal occurrence and source analysis of pharmaceutically active compounds (PhACs) in aquatic environment in a small and medium-sized city, China. Science of the Total Environment, 769: 144272

10 Fenech C, Nolan K, Rock L, Morrissey A. (2013). An SPE LC–MS/MS method for the analysis of human and veterinary chemical markers within surface waters: an environmental forensics application. Environmental Pollution, 181: 250–256

11 Foolad M, Ong S L, Hu J. (2015). Transport of sewage molecular markers through saturated soil column and effect of easily biodegradable primary substrate on their removal. Chemosphere, 138: 553–559

12 Jekel M, Dott W, Bergmann A, Dunnbier U, Gnirss R, Haist-Gulde B, Hamscher G, Letzel M, Licha T, Lyko S. . (2015). Selection of organic process and source indicator substances for the anthropogenically influenced water cycle. Chemosphere, 125: 155–167

13 Kan X P, Feng S Y, Mei X B, Sui Q, Zhao W T, Lyu S G, Sun S Y, Zhang Z W, Yu G. (2022). Quantitatively identifying the emission sources of pharmaceutically active compounds (PhACs) in the surface water: method development, verification and application in Huangpu River, China. Science of the Total Environment, 815: 152783

14 Kasprzyk-Hordern B, Dinsdale R M, Guwy A J. (2009). The removal of pharmaceuticals, personal care products, endocrine disruptors and illicit drugs during wastewater treatment and its impact on the quality of receiving waters. Water Research, 43(2): 363–380

15 Köck-Schulmeyer M, Ginebreda A, Petrovic M, Giulivo M, Aznar-Alemany Ò, Eljarrat E, Valle-Sistac J, Molins-Delgado D, Diaz-Cruz M S, Monllor-Alcaraz L S. . (2021). Priority and emerging organic microcontaminants in three Mediterranean river basins: occurrence, spatial distribution, and identification of river basin specific pollutants. Science of the Total Environment, 754: 142344

16 Kumar A, Xagoraraki I. (2010). Pharmaceuticals, personal care products and endocrine-disrupting chemicals in U.S. surface and finished drinking waters: a proposed ranking system. Science of the Total Environment, 408(23): 5972–5989

17 Li F, Chen L, Bao Y, Zheng Y, Huang B, Mu Q, Feng C, Wen D. (2020). Identification of the priority antibiotics based on their detection frequency, concentration, and ecological risk in urbanized coastal water. Science of the Total Environment, 747: 141275

18 Li Y, Zhang L, Liu X, Ding J. (2019). Ranking and prioritizing pharmaceuticals in the aquatic environment of China. Science of the Total Environment, 658: 333–342

19 Lu M C, Chen Y Y, Chiou M R, Chen M Y, Fan H J. (2016). Occurrence and treatment efficiency of pharmaceuticals in landfill leachates. Waste Management (New York, N.Y.), 55: 257–264

20 Masoner J R, Kolpin D W, Furlong E T, Cozzarelli I M, Gray J L. (2016). Landfill leachate as a mirror of today’s disposable society: Pharmaceuticals and other contaminants of emerging concern in final leachate from landfills in the conterminous United States. Environmental Toxicology and Chemistry, 35(4): 906–918

21 MeiX B, Sui Q, ZhangZ W, YaoG J, HeM D, ChenZ C, Lyu S G (2019). Identification of indicator pharmaceutical and personal care products (PPCPs) in different emission sources. Zhongguo Huanjing Kexue, 39(3): 1173–1180 (in Chinese)

22 Mejías C, Martín J, Santos J L, Aparicio I, Alonso E. (2021). Occurrence of pharmaceuticals and their metabolites in sewage sludge and soil: a review on their distribution and environmental risk assessment. Trends in Environmental Analytical Chemistry, 30: e00125

23 Meng Y, Liu W, Fiedler H, Zhang J, Wei X, Liu X, Peng M, Zhang T. (2021). Fate and risk assessment of emerging contaminants in reclaimed water production processes. Frontiers of Environmental Science & Engineering, 15(5): 104

24 Meyer M F, Powers S M, Hampton S E. (2019). An evidence synthesis of pharmaceuticals and personal care products (PPCPs) in the environment: imbalances among compounds, sewage treatment techniques, and ecosystem types. Environmental Science & Technology, 53(22): 12961–12973

25 Nakada N, Kiri K, Shinohara H, Harada A, Kuroda K, Takizawa S, Takada H. (2008). Evaluation of pharmaceuticals and personal care products as water-soluble molecular markers of sewage. Environmental Science & Technology, 42(17): 6347–6353

26 Qi C D, Huang J, Wang B, Deng S B, Wang Y J, Yu G. (2018). Contaminants of emerging concern in landfill leachate in China: a review. Emerging Contaminants, 4(1): 1–10

27 Renganathan J, S I U H, Ramakrishnan K, Ravichandran M K, Philip L. (2021). Spatio-temporal distribution of pharmaceutically active compounds in the River Cauvery and its tributaries, South India. Science of the Total Environment, 800: 149340

28 Scaria J, Anupama K V, Nidheesh P V. (2021). Tetracyclines in the environment: an overview on the occurrence, fate, toxicity, detection, removal methods, and sludge management. Science of the Total Environment, 771: 145291

29 Scheurer M, Storck F R, Graf C, Brauch H J, Ruck W, Lev O, Lange F T. (2011). Correlation of six anthropogenic markers in wastewater, surface water, bank filtrate, and soil aquifer treatment. Journal of Environmental Monitoring, 13(4): 966–973

30 Sharma S. (2008). Principal component analysis (PCA) to rank countries on their readiness for e-tail. Journal of Retail & Leisure Property, 7(2): 87–94

31 SimW J, Kim H Y, ChoiS D, KwonJ H, OhJ E (2013). Evaluation of pharmaceuticals and personal care products with emphasis on anthelmintics in human sanitary waste, sewage, hospital wastewater, livestock wastewater and receiving water. Journal of Hazardous Materials, 248–249: 219–227

32 Song L, Jiang G Y, Wang C, Ma J B, Chen H. (2022). Effects of antibiotics consumption on the behavior of airborne antibiotic resistance genes in chicken farms. Journal of Hazardous Materials, 437: 129288

33 Stefania G A, Rotiroti M, Buerge I J, Zanotti C, Nava V, Leoni B, Fumagalli L, Bonomi T. (2019). Identification of groundwater pollution sources in a landfill site using artificial sweeteners, multivariate analysis and transport modeling. Waste Management (New York, N.Y.), 95: 116–128

34 Sui Q, Wang B, Zhao W T, Huang J, Yu G, Deng S B, Qiu Z F, Lu S G. (2012). Identification of priority pharmaceuticals in the water environment of China. Chemosphere, 89(3): 280–286

35 Sui Q, Zhao W, Cao X, Lu S, Qiu Z, Gu X, Yu G. (2017). Pharmaceuticals and personal care products in the leachates from a typical landfill reservoir of municipal solid waste in Shanghai, China: occurrence and removal by a full-scale membrane bioreactor. Journal of Hazardous Materials, 323: 99–108

36 Sun Q, Li Y, Li M Y, Ashfaq M, Lv M, Wang H J, Hu A Y, Yu C P. (2016). PPCPs in Jiulong River Estuary (China): spatiotemporal distributions, fate, and their use as chemical markers of wastewater. Chemosphere, 150: 596–604

37 Thomaz C E, Giraldi G A. (2010). A new ranking method for principal components analysis and its application to face image analysis. Image and Vision Computing, 28(6): 902–913

38 Tran N H, Reinhard M, Khan E, Chen H, Nguyen V T, Li Y, Goh S G, Nguyen Q B, Saeidi N, Gin K Y. (2019). Emerging contaminants in wastewater, stormwater runoff, and surface water: application as chemical markers for diffuse sources. Science of the Total Environment, 676: 252–267

39 USEPA (2016). Definition and procedure for the determination of the method detection limit, Revision 2. Washington DC: United States Environmental Protection Agency

40 Wan Y P, Liu Z H, Liu Y. (2021). Veterinary antibiotics in swine and cattle wastewaters of China and the United States: features and differences. Water Environment Research, 93(9): 1516–1529

41 Wang P L, Wu D, You X X, Su Y L, Xie B. (2021). Antibiotic and metal resistance genes are closely linked with nitrogen-processing functions in municipal solid waste landfills. Journal of Hazardous Materials, 403: 123689

42 Wu D Q, Sui Q, Yu X, Zhao W T, Li Q, Fatta-Kassinos D, Lyu S G. (2021). Identification of indicator PPCPs in landfill leachates and livestock wastewaters using multi-residue analysis of 70 PPCPs: analytical method development and application in Yangtze River Delta, China. Science of the Total Environment, 753: 141653

43 YangY Y, Liu W R, LiuY S, ZhaoJ L, ZhangQ Q, ZhangM, Zhang J N, JiangY X, ZhangL J, YingG G (2017). Suitability of pharmaceuticals and personal care products (PPCPs) and artificial sweeteners (ASs) as wastewater indicators in the Pearl River Delta, South China. Science of the Total Environment, 590–591: 611–619

44 Yao L H, Li Y, Li Z Q, Shen D S, Feng H J, Zhou H H, Wang M Z. (2020). Prevalence of fluoroquinolone, macrolide and sulfonamide-related resistance genes in landfills from East China, mainly driven by MGEs. Ecotoxicology and Environmental Safety, 190: 110131

45 Yi X, Tran N H, Yin T, He Y, Gin K Y. (2017). Removal of selected PPCPs, EDCs, and antibiotic resistance genes in landfill leachate by a full-scale constructed wetlands system. Water Research, 121: 46–60

46 Yu X, Sui Q, Lyu S G, Zhao W T, Cao X Q, Wang J S, Yu G. (2020a). Do high levels of PPCPs in landfill leachates influence the water environment in the vicinity of landfills? A case study of the largest landfill in China.. Environment International, 135: 105404

47 Yu X, Sui Q, Lyu S G, Zhao W T, Liu J G, Cai Z X, Yu G, Barcelo D. (2020b). Municipal solid waste landfills: an underestimated source of pharmaceutical and personal care products in the water environment. Environmental Science & Technology, 54(16): 9757–9768

48 Yu X, Sui Q, Lyu S G, Zhao W T, Wu D Q, Yu G, Barcelo D. (2021). Rainfall influences occurrence of pharmaceutical and personal care products in landfill leachates: evidence from seasonal variations and extreme rainfall episodes. Environmental Science & Technology, 55(8): 4822–4830

49 Yuan X, Li S, Hu J, Yu M, Li Y, Wang Z. (2019). Experiments and numerical simulation on the degradation processes of carbamazepine and triclosan in surface water: a case study for the Shahe Stream, South China. Science of the Total Environment, 655: 1125–1138

50 Zhang Q Q, Ying G G, Pan C G, Liu Y S, Zhao J L. (2015). Comprehensive evaluation of antibiotics emission and fate in the river basins of China: source analysis, multimedia modeling, and linkage to bacterial resistance. Environmental Science & Technology, 49(11): 6772–6782

51 Zhong M M, Wang T L, Zhao W X, Huang J, Wang B, Blaney L, Bu Q W, Yu G. (2022). Emerging organic contaminants in Chinese surface water: identification of priority pollutants. Engineering (Beijing), 11: 111–125

52 Zhu M Y, Wang J W, Yang X, Zhang Y, Zhang L Y, Ren H Q, Wu B, Ye L. (2022). A review of the application of machine learning in water quality evaluation. Eco-Environment & Health, 1(2): 107–116