Occurrence and health risk of phthalates in groundwater and surface water matrix: a case study of Kampung Daraulin, Citarum, Indonesia Open Access

Emerging pollutants are compounds or chemical substances that are newly recognized as pollutants that can have potential impacts and are not yet fully addressed by existing environmental regulations (Deblonde et al. 2011; Khan et al. 2020). In recent decades, emerging pollutants have gained a lot of attention among researchers and policymakers. They come from various sources, including pharmaceutical and personal care products (Deng et al. 2021), hospital waste (Khan et al. 2020), industrial chemicals, and household products (Kumari & Pulimi 2023). These pollutants often cannot be removed by conventional wastewater treatment methods, so they can pollute and accumulate in the environment (Ghosh & Sahu 2022). Phthalates, or phthalic acid esters (PAEs), are chemicals of emerging concern widely found in daily household products (Zhang et al. 2020). PAEs are commonly used in the plastic industry to enhance flexibility, transparency, durability, and longevity of plastic products (Net et al. 2015; Bi et al. 2021). PAEs can be found in polyvinyl chloride products, perfumes, hair sprays, adhesives, cable coatings, toys, medical tubes, blood storage bags, paint fixative food packaging materials, and others so that plastic materials have the desired flexibility (Wang & Qian 2021; ECHA 2023).

Although PAEs are considered non-persistent compounds, their widespread use results in continuous introduction into the environment, making them ubiquitous in both point and non-point sources (Dueñas-Moreno et al. 2022). Their presence is widespread in the environment and can have significant health impacts (Gao & Wen 2016). Abtahi et al. (2019) found that phthalates, such as di(2-ethylhexyl) phthalate (DEHP) and dibutyl phthalate (DBP), were often detected in significant concentrations in various water samples. PAEs are physically, rather than chemically, bound to the polymer matrix (Tran et al. 2022). Therefore, PAEs can be easily released into environmental media and distributed into the aquifer system directly or indirectly, during the manufacturing, use, and disposal processes (Baloyi et al. 2021).

PAEs are classified as endocrine disruptor chemicals (EDCs), which are detrimental to the body's metabolic, neurological, immune, and reproductive systems (Abtahi et al. 2019). Studies suggest that PAEs can cause several negative health impacts, including hormonal changes in children, increased body mass index, increased eczema risks in children, a decrease in semen quality and concentration, and reduced female fertility (Eales et al. 2022). In addition, PAEs exposure is positively associated with an increased risk of cardiovascular disease (Zhang et al. 2021; Mariana et al. 2023). Although DBP and benzyl butyl phthalate (BBP) are not listed in the carcinogen category by International Agency for Research on Cancer (IARC), they are classified by the European Chemicals Agency (ECHA) as Substances of Very High Concern (SVHC), especially since they are considered harmful to fertility and fetal development ‘Reproductive Toxicant 1B’ (Repr. 1B) (EFSA 2019). Despite these risks, many low- to middle-income countries (LMICs), such as Indonesia, currently lack specific regulations that establish limits for phthalate use or contamination.

Studies on PAEs' occurrence in the aquatic environment in LMICs have highlighted the need to lower their concentrations in water matrices, although the human health risks of most PAE compounds in drinking water are still below the acceptable limit (Elaiyaraja et al. 2022; Vasseghian et al. 2023). Even so, PAEs might pose risks to aquatic organisms (Kingsley & Witthayawirasak 2020; Dueñas-Moreno et al. 2022; Elaiyaraja et al. 2022). Scholars have also detected PAEs in the groundwater matrix (Kotowska et al. 2020; Dueñas-Moreno et al. 2022). Despite the growing body of literature, more light should be shed to understand PAE occurrence and environmental and health risks associated with their presence in specific watersheds in LMICs, which generally have looser pollution control measures and regulations. Moreover, risks related to the presence of PAEs in groundwater should be better understood as they may pose a direct risk to human health (Zhang et al. 2015a, b).

This study focuses on two types of PAEs, DBP and BBP, which are the most commonly used phthalates in plastic products and personal care items, making them highly prevalent in the environment (Wang & Qian 2021; Eales et al. 2022). Additionally, DBP and BBP are known to exhibit significant environmental persistence and pose potential risks to human health due to their endocrine-disrupting properties (Abtahi et al. 2019). To the best of our knowledge, this is the first study to investigate the presence and potential health risks of PAEs, particularly DBP and BBP, in the aquatic environment of a critical river watershed in Indonesia. This study highlights the Citarum River, which is central to approximately 25 million populations who depend on the river for agriculture, domestic, and industrial activities, and power generation. Citarum is the longest river in West Java, encompassing 297 km along nine regencies and three municipalities. In 2022, Citarum River water quality was given the status of ‘moderately polluted’ (Marselina et al. 2025). Poor sanitation and poor disposal are among the leading causes of Citarum River pollution (Sholeh et al. 2018). Moreover, Citarum is home to 2,822 industries with potential access to discharge their wastewater into the river, but, providing vital support to 18.64 million people living along its stream (Sembiring et al. 2020). Textile manufacturing represents 68% of all factories in the Upper Citarum area. The industries undertake polyester weaving and wet processing, such as dyeing, printing, and finishing of polyester (Greenpeace International 2012), which may contribute to phthalate discharge into the aquatic environment (Sahnsarayi et al. 2025). The Daraulin Oxbow is a significant remnant of the old Citarum River course, located in Daraulin Village, Nanjung, West Java. Formed during river normalization efforts, this oxbow stretches approximately 2,890 m and is densely populated. The Daraulin Oxbow's water quality was indicated to suit activities like recreation, aquaculture, and irrigation, but not for direct consumption (Astuti et al. 2022).

There are several studies documenting the presence of emerging pollutants in the Citarum River (Jeong et al. 2021; Astuti et al. 2022; Utami et al. 2022; Wilkinson et al. 2022). Currently, no specific studies document the presence of PAEs in the Citarum River. However, scholars have identified a number of pollutants in the river, including plastic waste and industrial chemicals (Pamungkas et al. 2021; Chazanah et al. 2024), which are potential sources of PAEs. Moreover, although a comparison with surface water was made, this study also focuses on groundwater, which serves as a key source for domestic purposes in the area, emphasizing the direct relevance of our findings to public health. This study focuses on Kampung Daraulin, located in the southern part of the Citarum River. Kampung Daraulin is an area that has experienced flooding and surface water and groundwater pollution. Daraulin was selected as the study site due to its strategic location along the Citarum River and its likely exposure to substantial domestic, industrial, and agricultural waste discharges. This makes it a critical hotspot for studying organic pollutants such as phthalates. This study aims to measure the concentration of PAEs and evaluate the health risks associated with groundwater use in Kampung Daraulin. In addition, we also assess the community's perceptions of groundwater pollution and its risk.

Given this watershed's ecological and public health significance, the findings will provide essential baseline data on PAEs contamination, inform strategies for mitigating potential risks, and contribute to a better understanding of how these emerging contaminants behave in tropical aquatic systems. This will raise awareness of emerging pollutants in Indonesia and could inform targeted intervention as part of the ongoing government-led river cleanup program, Citarum Harum.

The village is surrounded by sodetan (a river diversion channel designed to redirect river flow or mitigate flooding) of the Citarum River, commonly referred to by the local community as the ‘oxbow’, as well as the meandering section of the river, known as the new Citarum River. The diversion channel is no longer operational, as it has been permanently closed by the government (Ridhosari & Roosmini 2011). The community in Kampung Daraulin experiences inadequate sanitation, with septic tanks positioned <10 m from dug wells, the primary source of clean water at depths of 10–15 m. Furthermore, wastewater from daily activities is commonly discharged into the nearby oxbow and main river streams, exacerbating the environmental and health risks (Ridhosari & Roosmini 2011). Despite these conditions, the residents still rely on water from the river channel for household activities such as washing clothes and dishes. This may pose health risks to the local community. The stagnant water flow in this area creates favorable conditions for pollutant accumulation. Daraulin hosts a plastic waste separation facility, which may contribute to local environmental contamination risks. Consequently, Kampung Daraulin is a critical sampling site due to its high contamination potential, making it an essential location for evaluating the impacts of PAEs on groundwater quality and assessing long-term health risks.

In this study, water samples were taken from Daraulin Village, Citarum River watershed. The sampling points for groundwater and surface water were determined using the stratified sampling method to ensure that each type of water source was proportionally represented in the study. A grid sampling strategy was employed to systematically collect water samples. This approach involved dividing the study area into a predefined grid of uniform sections, ensuring the representativeness of both surface water and groundwater sources. Sampling points were strategically focused on grid cells that are occupied by residents to capture risks to humans. Time and cost-effectiveness were also considered part of the research requirements.

Water samples were collected directly into conventional 500-mL glass bottles, immediately placed on ice, and stored at 4 °C in a laboratory refrigerator prior to analysis. All samples should be extracted within 7 days after collection and fully analyzed within 40 days after extraction.

Solvents used in this work include hexane, methanol, acetone, and dichloromethane, which were high-performance liquid chromatography (HPLC) grade, from Waters, USA. Phthalate standards of DBP and BBP were purchased from RESTEK (EPA Method 8061A Phthalate Esters Mixture), and solid phase extraction (SPE) cartridge silica was purchased from Hawach Scientific (SPE 1 g 6 cc).

The analysis of DBP and BBP contained in all the liquid extract samples was performed by using a gas chromatograph-mass spectrometer (GC–MS), Agilent model 7820A GC–5977B MSD (Agilent Technologies, USA) operating in electron impact (EI) and selective ion monitoring (SIM) modes with a DB-5 MS (30 m × 0.25 mm × 0.25 mm) fused-silica capillary column for chromatographic separation (Zhang et al. 2015a, b).

To prevent PAEs background contamination, contact between the sample and plastic materials was strictly avoided during the procedure (Lee et al. 2019). The pre-treated laboratory equipment was thoroughly cleaned with acetone three times before use. One procedural blank was processed alongside the water samples as a quality control measure. Procedural blanks were prepared by running ultrapure water through the same analytical process, including sample preparation, extraction, and analysis, to monitor potential contamination introduced by reagents, equipment, or the laboratory environment. Procedural blanks showed no detectable levels of the two targeted PAEs, indicating no contamination occurred during sample handling. The limit of detection (LOD) and limit of quantification (LOQ) for individual PAEs congeners were estimated based on a signal-to-noise ratio of 3 and 10 times, respectively.

Descriptive statistics were utilized to present the concentration data of PAEs in water samples, summarizing the distribution, central tendency, and variability of the data. These statistics were also employed to calculate and display key parameters required for health risk analysis, providing a comprehensive overview of exposure levels and their potential impacts.

The survey was deployed through direct face-to-face structured interviews. The sample selection was conducted using the probability sample method, in which the sample is taken in a simple random manner so that each sample can represent an equal chance of being statistically selected in a population (Oppenheim 2000). The inclusion criteria for this survey were (1) residents of Daraulin Village; (2) using groundwater; and (3) agreed to participate in the survey. A total of 100 respondents were interviewed. The sample size was determined using the Cochran formula to represent the diversity within the population, considering a confidence level of 95% and a desired error rate of 10%. Descriptive analysis was employed to describe the responses of the community regarding groundwater pollution. This study has been approved by Research Ethics Committee, Faculty of Psychology, University of Muhammadiyah Malang Research Ethic Committee with Ethical Approval Number E.6.m/155/KE-FPsi-UMM/VII/2024.

This study has several limitations that should be acknowledged. First, water sampling was conducted during a single period or season, which may not capture seasonal variations in PAE concentrations. Second, the number of water samples collected was relatively small, potentially limiting the robustness and generalizability of the findings. Thirdly, the grab sampling method used to collect water samples only captures water quality at the time of sampling and does not account for temporal or seasonal variations. Fourthly, detecting and quantifying small concentrations of PAEs in water depends heavily on the quality of the device setup and calibration. Consequently, inadequate sensitivity may result in these small concentrations going undetected. Lastly, there is uncertainty regarding the precise concentration of pollutants in the environment, how these concentrations may affect the population during the exposure period, and the ambiguity surrounding the characteristics of the exposed population.

Table 1 shows summary statistics of the two types of PAEs, DBP and BBP, concentration in groundwater and surface water of Daraulin Village, Citarum River. The smallest quantified total concentration of PAEs in groundwater was 1.10 μg/L and the largest was 4.50 μg/L. Each type of PAEs concentration ranged from 0.16 to 0.89 μg/L for DBP, and 0.73 to 3.94 μg/L for BBP. Both types of PAEs were quantified in 100% of the total groundwater samples. The average concentration of each type of PAE was 0.43 ± 0.26 μg/L for DBP and 2.18 ± 1.04 μg/L for BBP in groundwater. The elevated BBP levels in groundwater compared to DBP suggest its greater environmental persistence and sorption potential, particularly in subsurface environments (EPA 2024a, b). BBP's higher hydrophobicity compared to DBP enhances its affinity for organic carbon in soils and sediments (EPA 2024a, b).

The distribution of PAEs in groundwater showed that the highest total concentration of PAEs was recorded at sampling site G1 at 4.5 μg/L with the highest quantified concentrations of DBP and BBP at 0.56 and 3.94 μg/L, respectively. G1 is located adjacent to one of the household-scale industries in the form of a junk collector, which may indicate that the high concentration of PAEs at that location comes from the migration of PAEs compounds from wastewater to the groundwater environment. BBP compounds are more persistent than DBP, so the risk of pollution to groundwater is higher (Dueñas-Moreno et al. 2022). This result is consistent with earlier studies indicating that proximity to plastic recycling and processing facilities is associated with elevated levels of BBP contamination (Net et al. 2015).

The data suggest that surface water shows consistently higher concentrations than groundwater, likely due to surface runoff and localized industrial discharge. Right around locations O4 and O3, there is a paint manufacturer factory in the field of automotive resurfacing, so it can be indicated that the high concentration of PAEs in these locations comes from poorly managed waste disposal and paint industry wastewater runoff containing paint coating residues (Becky Miriyam et al. 2022). This spatial distribution reflects land-use vulnerability and inadequate wastewater management, a theme seen in other studies. In a study of the Wangyang River basin, a typical wastewater irrigation area in North China, PAEs were found to be ubiquitous in both water and soil samples. The highest concentrations of PAEs were detected near villages where dense human activity and random garbage disposal occurred (Zhang et al. 2015a, b). Research conducted in an e-waste recycling area in Tianjin, China, found that PAEs exposure was greater in areas with less centralized management and poorer waste handling practices (Li et al. 2019). This suggests that poorly managed waste and anthropogenic activities are significant sources of PAE contamination in both ground and surface waters.

The concentrations of PAEs obtained in this study can be compared with DBP and BBP data previously published. Kotowska et al. (2020) reported DBP concentrations ranging from not detected to 12.7 μg/L in groundwater near municipal solid waste landfills in Poland. Saha et al. (2022) found DBP concentrations in agro-climatic zones in India ranging from 93.35 to 255.33 μg/L, while BBP concentrations were between 0.76 and 40.85 μg/L. Similarly, in groundwater from Hanoi and Ho Chi Minh City, Vietnam, Duong et al. observed DBP concentrations ranging from 0.39 to 10 μg/L, and BBP concentrations between 0.1 and 0.11 μg/L (Duong et al. 2015). DBP concentrations in this study were found to be 2-fold higher than those reported in Delta State, Nigeria, and 12- to 100-fold lower than those reported in monitoring wells in Landfill, Poland, and Agro-Climatic Zone, India (Edjere et al. 2020; Kotowska et al. 2020; Saha et al. 2022). Meanwhile, BBP concentrations were found to be 7- to 15-fold higher than studies reported in Hanoi and Ho Chi Minh City, Vietnam, and Delta State, Nigeria. DBP concentrations in surface water were 5- to 30-fold higher than studies reported in Nigeria, and Pearl and Jiangsu Rivers in China, but 70-fold lower than studies reported in the Agro-Climatic Zone, India (Edjere et al. 2016; Cheng et al. 2019; Fan et al. 2021). Meanwhile, BBP concentration in surface water was 1-fold higher than the study reported in Pearl River, China (Cheng et al. 2019). However, the BBP concentration was 5-fold lower than the study reported in the Agro-Climatic Zone, India (Saha et al. 2022). A study in Iran measured PAEs levels in various water sources at concentrations ranging from 0.10 to 0.52 μg/L, on average (Abtahi et al. 2019). Consequent analysis showed that the health risks were generally low at these concentrations. As this study recorded the average PAEs level in groundwater used in daily life of 2.62 g/L, caution should be taken as potential health risks might occur.

The relatively high concentrations of BBP in this study compared to those found in Vietnam and Nigeria reflect differences in several possible factors, such as the proximity of sampling points with industrial or recycling activities and the regulations and their enforcement. Meanwhile, DBP concentrations in this study were found to be lower than those in India. This trend might be explained by the differences in plastic usage and regional soil permeability affecting leaching patterns. Regions with more intensive plastic usage may have higher DBP concentrations in soil and water compared to regions with less plastic use (Wang et al. 2021). Therefore, contextual factors should be considered when analyzing transport and persistence of PAEs in the environment.

The non-carcinogenic exposure values (HQ) and total non-carcinogenic exposure (HI) values of the water samples for exposure to PAEs via ingestion and dermal absorptions are presented in Table 2. In general, if the non-carcinogenic risk value (HQ) > 1 indicates that the assessed pollutant has adverse effects on public health further monitoring and assessment is recommended. Based on the calculation, non-carcinogenic exposure to PAEs is <1. As the current exposure to PAEs in groundwater and surface water samples via the oral and dermal routes is acceptable, the current PAEs presence does not pose a public health risk to adults and children. The results of this study are in line with the studies conducted by Li et al. (2016) and Selvaraj et al. (2015) in which PAEs showed no human health risk through surface water exposure.

Although the non-carcinogenic risk obtained in this study is considered safe for adults and children, the potential for long-term and cumulative exposure to PAEs should not be overlooked. The implications of chronic exposure to PAEs are significant and multifaceted, affecting human health through endocrine disruption, reproductive toxicity, developmental issues, and increased risks of certain cancers. Chronic exposure to PAEs has been linked to serious health problems such as congenital disabilities (Kingsley & Witthayawirasak 2020) and may interfere with immune functions, neurodevelopment and cardiovascular health, and may contribute to cancer development (Kumar et al. 2020; Chen et al. 2024).

Moreover, it is important to note that this study exclusively evaluated the non-carcinogenic risks associated with PAEs through oral and dermal exposure. While the results indicate no significant non-carcinogenic health risks, additional research is recommended to assess potential carcinogenic effects, particularly given the evidence from other studies suggesting that some PAEs may act as endocrine disruptors with carcinogenic potential at low concentrations (Chen et al. 2016; Abdolahnejad et al. 2019).

There are reports describing the toxicity of PAEs at levels relatively lower than the RfD using in vitro models. At low concentrations, DEHP, BBP, and DBP compounds can induce proliferation in human breast cancer via Phosphoinositide 3-Kinase (PI3K) or Protein Kinase B (AKT) signaling pathways and exhibit estrogenic effects (Chen et al. 2016). This necessitates enhanced capacity and continued environmental and health monitoring to measure chronic exposure to PAEs at a low dose. Moreover, additional studies should shed light on the potential impact of PAEs on vulnerable populations, such as pregnant women, the elderly, and people with compromised immune systems. Phthalate exposure during pregnancy is associated with endocrine disruption in mothers and offspring, increased risk of pregnancy complications, adverse birth outcomes, and impaired fetal reproductive and developmental health (Gao et al. 2017; Qian et al. 2020; Al-Saleh et al. 2024).

All the risk estimation methods involve varying degrees of uncertainty. There are two main sources of uncertainty in the assessment of human exposure to pollutants: how much the concentration of pollutants present in the environment can affect the population during the exposure period and the ambiguity of the exposed population (Dong et al. 2015; Vandenberg et al. 2023). In this study, both can significantly affect the results. The actual exposed population could not be studied, so the health risk was only based on a hypothetical exposure scheme. Thus, the uncertainty in the calculation results obtained could be significant. In addition to the factors already mentioned, uncertainty in the dose response from non-carcinogenic data may also be a contributing factor. Various assumptions used in this study, such as a life span of 70 years, a body weight of 60 kg for adults and 20 kg for children, may increase the uncertainty in the assessment process. Therefore, the results provided in this study are only an indication of the potential health risks associated with potential exposure to ground and surface water when used for drinking and bathing purposes. These estimated risks may be an underestimate of the actual health risks as they only describe the risks associated with the population exposed to the DBPs and BBPs detected in the study. No other pollutants in ground and surface water were considered in this study, but there may be other pollutants that are more harmful to humans. Therefore, the health risks may be higher than those depicted in this study.

While the current risk appears low, the evidence of BBP's higher persistence and the proximity to pollutant sources highlight a critical need for continued monitoring in the study area. The exposure pathways evaluated here may not fully capture inhalation risks (see Wang et al. 2018 for example) or bioaccumulation through the food chain, which should be addressed in future studies.

In this study, we also investigated how the community perceived risks associated with groundwater pollution. While this study includes a comparative analysis of phthalate concentrations in both groundwater and surface water sources, the risk perception component is focused on groundwater. This decision is rooted in the socio-environmental context of the study site where groundwater remains used for domestic purposes. A total of 100 respondents from Daraulin Village participated; all respondents met the inclusion criteria and completed all items on the questionnaire given in the risk perception survey on water pollution. Most respondents were of productive age, mostly female, and at the junior and senior high school education levels in lower-middle-class economic conditions. The risk perception scores among respondents are shown in Table 2, with scores ranging from 1.00 to 5.00.

The results suggest that the respondents have a high awareness of groundwater pollution, indicating they are already well-informed about the issue. While respondents are moderately optimistic about future improvements, the relatively low trust in institutional actions and information points to a need for stronger engagement from policymakers and researchers. The next step is how to translate this awareness into action in two areas: enhancing self-preservation to protect themselves from the health impact of emerging pollutants and reducing their contributions as potential sources of pollution while advocating for stronger government action to address emerging pollutants, such as PAEs.

In summary, the results suggest that BBP is the more persistent and dominant pollutant in both groundwater and surface water in Daraulin Village, likely due to specific industrial activities. Despite currently acceptable health risk levels, the persistent nature of these compounds and their endocrine-disrupting potential at low concentrations call for precautionary measures. Public awareness is relatively high, but limited institutional trust may hinder coordinated mitigation. These findings underline the importance of integrating scientific monitoring with governance reforms and public engagement to effectively manage emerging pollutants in peri-urban environments.

This study found the presence of both types of PAEs at an average concentration of 2.62 and 10.11 μg/L in the groundwater and surface water of a critical watershed in Indonesia, respectively. This study also underscores the importance of continuous monitoring of PAEs in regions with high ecological and human dependence on aquatic resources. The total HQ and HI values of groundwater samples for exposure to PAEs through ingestion and dermal routes in Kampung Daraulin were found under the value of 1, through ingestion and dermal routes. Research in Indonesia on the content of various types of PAEs in water, including DBPs and BBPs, is still very limited. This study fills a critical knowledge gap by providing the first insight into the occurrence and risks of PAEs in the aquatic environment of a key river watershed in Indonesia. This finding is significant given the complex challenge of pollution control measures and high anthropogenic pressure in Citarum, which might make such environments more vulnerable to contamination over time. Although the health risks are acceptable, the potential for long-term accumulation in the oxbow area of the river highlights the need for strengthening pollution control measures and environmental regulations in LMICs. The high awareness of the residents of groundwater pollution and the importance of preventing it serves as an opportunity to mobilize community action, promote sustainable practices, and advocate for stronger regulatory measures to protect groundwater resources.

Future research could concentrate on longitudinal studies with multiple sampling periods across different seasons to capture temporal variations in contaminant levels. Expanding the sample size and including a wider range of PAE compounds, along with other potentially interacting contaminants, will provide a more comprehensive understanding of groundwater quality. Moreover, oxbows serve as natural sediment traps and offer an excellent baseline for evaluating the extent of phthalate contamination in sediment. Given the lipophilic nature of PAEs, future research should also consider sediment sampling to understand the fate of phthalates in the aquatic environment. Being relatively stable water bodies, oxbows should also be considered for long-term monitoring of pollutant trends. Policymakers should consider establishing guidelines for PAE contamination in groundwater, promoting proper waste disposal practices, and implementing community awareness programs to mitigate pollution risks.

This research is funded by Beasiswa Unggulan 2023, Program of the Ministry of Education, Culture, Research, and Technology Republic of Indonesia.

All relevant data are available from an online repository or repositories. The dataset and supplementary materials used in this study are currently stored in a private Zotero library.

The authors declare there is no conflict.