Benzene homologues contaminants in a former herbicide factory site: distribution, attenuation, risk, and remediation implication

  • Shuo Yang
  • Xiulan YanEmail author
  • Lirong Zhong
  • Xuejiao Tong
Original Paper


Benzene homologues often used as organic raw materials or as detergents in chemical industry are prone to accidental release into the environment which can cause serious long-term soil pollutions. In a large former herbicide factory site, we investigated 43 locations for benzene homologues contaminations in soil, soil gas, and groundwater and studied the hydrogeological conditions. An inverse distance weighted interpolation method was employed to determine the pollutants three-dimensional spatial distribution in the soils. Results showed that benzene homologues residues were mainly originated from the herbicide production workshop and that the pollution had horizontally expanded at the deeper soil layer. Contaminants had already migrated 15 m downward from ground surface. Contaminant phase distribution study showed that NAPL was the primary phase (> 99%) for the pollutants accumulated in the unsaturated zone, while it had not migrated to groundwater. The primary mechanism for contaminant transport and attenuation included dissolution of “occluded” NAPL into pore water and pollutant volatilization into soil pore space. Risk assessment revealed that the pollutants brought unacceptable high carcinogenic and non-carcinogenic risks to public health. In order to convert this former chemical processing factory site into a residential area, a remediation to the polluted production workshop sites is urgently required.


Benzene homologues Soil NAPLs Distribution Attenuation Health risk 


Leakages from chemical industry factories are an important source of anthropogenic pollutants (Li et al. 2014). The multi-phase flow in the unsaturated zone after the release of liquid organic pollutants may induce complex contaminations sources (Choi et al. 2002; Molins et al. 2010; Oostrom et al. 2007). The typical agrochemical industry relies on the use of benzene homologues, such as toluene, benzene, ethylbenzene and xylenes. Toluene and benzene are normally used as solvents for organic raw materials, while ethylbenzene and xylenes are used as raw materials in organic synthesis (Hellweg et al. 2004). These chemicals are often found at the production sites. Due to leakages from chemical processing, benzene homologues are often detected as liquid source at the sites of former chemical factories (Huang et al. 2013; Jeong et al.; Li et al. 2012; Masih et al. 2016). Because benzene homologues are volatile and have relatively high water solubility and low Kow values (Dou et al. 2008; Hellweg et al. 2004), they tend to distribute into the air, soil and groundwater after release (Chen and Brune 2012; Yuan et al. 2010). For example, Wang et al. (2014) monitored benzene homologues in air, soil, dust and groundwater samples at three pesticide factories and found although the concentrations of benzene homologues in multimedia matrixes below the levels of standards established in China, imposing high occupational health risks on factory workers. Therefore, benzene homologues contaminants often pose a significant impact on the environment.

In polluted field, high levels of benzene homologues can exist in soils as non-aqueous phase liquids (NAPLs) (Hsieh et al. 2006; Perez-Rial et al. 2009) and may persist for decades (Rivett et al. 2011). Benzene homologues NAPL as non-wetting fluids may penetrate the unsaturated zone or be restricted to shallower horizons depending on the permeability of layering and whether the NAPL driving head exceeded the entry pressure of porous media (Parker 2003). A study suggested excess head is needed for NAPL to displace pore water (Gooddy et al. 2002). Meanwhile, NAPLs can volatilize to form a high concentration vapor plume, partition into pore water, and potentially infiltrate to groundwater (Mendoza and Frind 1990; Pankow and Cherry 1996). Since benzene homologues were detected in air, soil, dust and groundwater, they will cause adverse impacts on human health through inhalation, dermal and ingestion routes. Researchers adopted various models such as USEPA health risk model (Wang et al. 2014), a risk-based corrective active (RBCA) model (Liu et al. 2016) to assess health risk and considered benzene homologues having potential to cause the high cancer risk at the contaminants sites.

While data on contamination source distribution and vertical profile exist at some sites (Hsieh et al. 2006; Lawrence et al. 2006; Wang et al. 2014), systematic evaluation of transport, attenuation, risk, and implication on remediation of pollution in the unsaturated zone is lacking. In this study, a field of a former herbicide factory with more than 30-year production history and possibility of organic chemical leakage was chosen to study the transport and attenuation of benzene homologues in unsaturated zone. We measured the concentrations of benzene homologues in air, soil and groundwater samples at this site and employed an inverse distance weighted (IDW) interpolation model to obtain the continuity distribution of soil contaminants. Furthermore, the morphology of benzene homologues was established to understand the scenario of infiltration and attenuation process of these contaminants on the air–water–NAPL phases through unsaturated zone as well as the potential of migration to underlying water table. The carcinogenic and non-carcinogenic risks to human health were quantified to evaluate their risks to public health.

Materials and methods

Description of the study area

The study was conducted in a small industrial city of Xuanhua, once hosted more than 30 large factories, in northwestern Hebei province, China (Fig. S1). Those factories were in the business of metallurgy, machine manufacturing, paper-making and chemical engineering. This city is located on the silt and loess deposits of an alluvial plain of the Yang River, a tributary of the upper Yongding River. There are different lithological features between vertical layers at this field and the groundwater flows from northwest to southeast (Fig. S2) at a depth of 60–70 m.

A former large herbicide factory site covering 120,000 m2 was chosen as the study area. This factory produced atrazine, tribenuron, acetochlor, clomazone, and related products from 1992 to 2009. It was one of the largest herbicide factories in northern China. During production, benzene homologues were used as raw materials and solvents. The factory was divided into office and production areas (Fig. 1). The production areas included six production units, five workshops and one sewage treatment plant, which were divided by roads (Fig. 1, right-hand panel). Underground sewage pipelines were constructed to connect the workshops with the treatment plant, but the burying depth of these pipelines is unknown. Prior to our investigation, all the factory buildings on this site had been demolished.
Fig. 1

Sampling locations at the former herbicide factory site

Sampling procedure

Soil samples were collected from 43 locations on the factory site. Seven uniformly distributed locations were within the area occupied by the office buildings (outside the orange frame in the left-hand panel of Fig. 1). These areas were not considered as the source regions for the pollutants. The main sampling locations were located both inside and outside each of the production units (Fig. 1, right-hand panel) with the aim of locating the points of emission or leakage of pollutants. Using dense grid sampling method, locations 8–16, 17–20, 21 and 22, 23–26, 27–35 and 36–43 were inside the sewage treatment plant, clomazone workshop, tribenuron workshop, acetochlor workshop 1, ethylamine workshop and acetochlor workshop 2, respectively.

The stratified soil samples (0–22 m) were collected from four layers. The first layer with the depth of 0–1 m was miscellaneous filling soil; the second layer with the depth of 1–3 m was sand; the third layer with the depth of 3–15 m was silt; the fourth layer with the depth of 15–22 m was gravel. A 5-m-long soil core was taken at each site using a hydraulic drill rig, and soil samples were collected at depths of 0–0.5, 0.5–1, 1–3, 3–5, 5–10, 10–15, 15–17, 17–22 m. Each undisturbed sample was packed into a 40-mL brown glass jar containing 5 mL of methanol. The jars were completely filled by the soil samples to minimize the evaporation of the benzene homologues. Soil samples were also collected from deeper layers near the office buildings (location 1), the tribenuron workshop (location 22) and the acetochlor workshop (location 40) to observe the vertical migration of pollutants to deeper layers. The sampling profiles extended to a depth of 22 m at locations 1, 22 and 40.

Samples of soil gas were also collected at locations 22 and 40. Screens (3 m long and 25 mm in diameter) were deployed at each depth in the sampling well, with PVC tubes extended up to ground level. Soil gas was pumped via the PVC tubes at 0.2 ml/min for 2 min, and the gas samples were drawn up through an absorption tube filled with activated carbon.

Water samples were taken from a well to the south of the production areas to assess the possible leakage of benzene homologues to groundwater. Groundwater samples were collected in 40-mL vials with a headspace after flushing the sampling line for 3 min. The flow rate was reduced to avoid introducing bubbles during sampling. Atmospheric samples were also collected at a sampling rate of 20 ml/min for 20 min at locations 1, 22 and 40 to determine contaminant evaporation from the soil. All the samples were transported in a cooled container and stored at − 20 °C until analysis.

Analytical approaches

All samples were analyzed within 5 days after collection. Toluene (TLE), ethybenzene (EBZ), m/p-xylene (MPX) and o-xylene (OX) were analyzed for each soil sample. Extraction of the benzene homologues from the soil and water samples was based on US Environmental Protection Agency (US EPA) Methods 5035A and 5035C, respectively. The concentrations of contaminants were measured by Agilent 7890B-5977A GC/MS using US EPA Method 8260C.

Quality assurance and quality control

A soil sample, a soil gas (air) sample, and a groundwater sample taken from outside the contamination site were used as blank samples. The concentrations of the target compounds in the blanks were less than the detection limits of the GC-MS. The detection limits were 0.05 μg/g, 0.5 μg/L and 0.5 mg/m3 for soil, groundwater and gas (air) samples, respectively.

The internal standards used for soil, groundwater, and gas samples were phenol-D6, nitro-benzene-D5, and p-terphenyl-D14, with average recovery of 56%, 68%, and 81%, respectively. One check standard was injected for every 10 samples and the relative deviation of the target compound in parallel samples was < 10%.

Inverse distance weight interpolation methods

Inverse distance weighting (IDW) method is used extensively in spatial interpolation. It is straightforward to interpret and easy to compute. In this research, we use the IDW method to generate a surface with less background in spatial statistics and geostatistics and 43 available sampled locations. Its general assumption is that the predicted value of an unsampled point is the weighted average of measured values surrounding the prediction location, and the weights are inversely related to the distances between the prediction location and the sampled locations (Lu and Wong 2008). The interpolation algorithms can provide many value of variables X(s0), X(s1), X(s2),…, at unmeasured locations, which has been used extensively in soil pollution mapping. Inverse distance weight interpolation method is a linear interpolator: the predictor at any location s0 is assumed to be a linear combination of the available data, i.e.,
$$X\left( {s_{0} } \right) = \mathop \sum \limits_{i = 1}^{n} \lambda_{i} X\left( {s_{i} } \right)$$
The weights are given by
$$\lambda_{i} = {\raise0.7ex\hbox{${\left[ {d\left( {s_{i} ,s_{0} } \right)} \right]^{ - p} }$} \!\mathord{\left/ {\vphantom {{\left[ {d\left( {s_{i} ,s_{0} } \right)} \right]^{ - p} } {\mathop \sum \nolimits_{1}^{n} \left[ {d\left( {s_{i} ,s_{0} } \right)} \right]^{ - p} }}}\right.\kern-0pt} \!\lower0.7ex\hbox{${\mathop \sum \nolimits_{1}^{n} \left[ {d\left( {s_{i} ,s_{0} } \right)} \right]^{ - p} }$}}$$
where \(d\left( {s_{i} ,s_{0} } \right)\) is the distance between \(s_{i}\) and \(s_{0}\). The power p is selected to be 2.

Calculations on contaminant distribution

Based on the concentration of soil and soil gas (\(C_{\text{gas}}\)) at different depths, the concentration in the aqueous phase (\(C_{\text{aq}}\)) was calculated using Eq. 3:
$$C_{\text{aq}} = \frac{{C_{\text{gas}} }}{H}$$
where H is the constant of Henry’s law (Table S1). To calculate the amount of contaminant in the solid and in the NAPL, use Eq. 4.
$$m_{{s + {\text{napl}}}} = m_{\text{tot}} - C_{\text{gas}} \times V_{\text{gas}} - C_{\text{aq}} \times V_{\text{aq}}$$
where \(m_{\text{tot}}\) is the total amount of benzene homologues in the soil, \(V_{\text{gas}}\) is the volume of the gas phase, and \(V_{\text{aq}}\) is the volume of the aqueous phase.

Health risk assessment model

The health risks of the surface soil layers (0–1 m) were assessed in this field. The indicators, CRs and HQs, were used to quantify the carcinogenic and non-carcinogenic risks, respectively, and the methods are described in details in the supplementary material. In this study, the 95% confidence levels of the mean concentration of the pollutants in soils and the highest concentrations of the pollutants in the soil gas were used as the input data. The health risks from inhalation, oral uptake and dermal adsorption were considered, and both carcinogenic and non-carcinogenic risks were estimated. The health risk assessment followed the Technical Guidelines for Risk Assessment of Contaminated Sites (HJ 25.3-2014) published by Ministry of Environmental Protection of China (2014).

Results and discussion

Soil properties in the study area

The main soil properties in the study area are given in Table 1. The nature of the soils in the herbicide factory site is weakly alkaline and sandy with very low amounts of natural organic matter. The soil porosity remains constant; however, soil water content increases with depth.
Table 1

Main properties of soils in the study area

Depth (m)


f om a (%)

Bulk density (g/cm3)

Water content (%)

Porosity (%)

Textural class






















afom fraction of organic matters in soil

Concentration levels and compositions of benzene homologues

Four species of benzene homologues pollutants were detected in the soil samples. A summary of description of mean and range values for individual pollutant is presented in Table 2. The average concentrations of TLE, EBZ, MPX and OX in the soil samples were 16, 1035, 1643 and 484 μg/g dry weight (d.w.), respectively. A site investigation carried out at one other former factory sites in Hebei province, China (Wang et al. 2014), showed maximum concentrations of 7.2, 7.9 and 20.0 μg/g d.w. for TLE, EBZ and MPX + OX, respectively. An urban site investigation reported 0.011–0.041, 0.013, 0.013–0.039 and 0.010–0.014 μg/g d.w. concentrations for TLE, EBZ, MPX and OX, respectively (Zouir et al. 2009). These data indicate that our research site had one to three orders of magnitude higher contaminants concentrations than the other abandoned factory sites and at least one thousand times higher pollutants level than the urban region (Table 2).
Table 2

Statistical data for benzene homologues in soil profiles at the former factory site (unit: μg/g dry weight) and comparisons with other studies






Studied depth (m)

< 0.5




















> 10





Mean of whole profile





H/L ratiob





Other regions

Pesticide chemical factory in Hebei, Chinac



BDL–20.0 (MPX + OX)

Chemical plant in Jilin, China (soil profiles)d



Tangier city, Moroccoe





aBDL below detection limits

bH/L ratio, the ratio of the highest concentration to the lowest concentration at the study site. The detection limit of this study was 0.05 μg/g, and the lowest concentrations (below detection limits) were assumed to be half the detection limit (0.025 μg/g) and were used to calculate the H/L ratios

cWang et al. (2014)

dZhou et al. (2013)

eZouir et al. (2009)

For individual compounds, EBZ and xylene had similar concentration ranges at different depth with the former herbicide factories in Hebei province, whereas the concentrations of TLE at the current site were slightly lower than those observed elsewhere (e.g., at the chemical plant in Jilin, China; Table 2). This pattern may be related to the production characteristics. Tribenuron and acetochlor at the current factory were produced at rates of 300 and 2000 tons/year, respectively. TLE is one of the raw materials used to produce atrazine, whereas a xylene mixture (MPX + OX) is an important solvent in the synthesis of tribenuron and acetochlor. More TLE was consumed than xylenes because xylenes were expected to left as solvent. Hence, the concentration level of TLE was relatively lower than that of xylenes in soil. Meanwhile, the residual levels and spatial variability of pollutants might also be affected by soil composition and compound properties. This knowledge is critical for health risk assessment and for development of a remediation strategy.

The horizontal and vertical distribution of the benzene homologues in soil will be further discussed in the following, in an attempt to describe the emission and leaching of these pollutants from the different workshop areas and to characterize the extent of soil contamination at the study site.

Horizontal distribution of benzene homologues

The spatial distribution of benzene homologues is presented in Fig. 2. Although samples were obtained from both the former office and production areas, benzene homologues were only detected in samples from the production areas. None of the target compounds was detected in soil samples taken from near the clomazone and ethylamine workshops. The pesticide clomazone was produced for a short period of time at this factory. Figure 2 shows that the highest concentrations in surface soils were detected near the acetochlor and tribenuron workshops. The peak concentrations of TLE in the surface soils were found in the northern and southern tribenuron workshops, while the peak concentrations of EBZ, MPX, and OX were mainly found in the acetochlor and southern tribenuron workshops. It can be seen that the benzene homologues residues in the soil were mainly from the herbicide production. The large values measured for the highest to lowest ratios (H/L ratios) for each compound (3236–265,600; Table 2) also implied significant leakage of benzene homologues in this factory (Wang et al. 2010), with subsequent migration. Vertical migration might be an important pathway of transport for benzene homologues. In short, the horizontal spreading of benzene homologues is limited to the acetochlor and tribenuron workshops areas based on the interpolation model analysis (Fig. 1). This result is consistent with the results of the previous study, suggesting the pollution hotspots on contaminated sites were close to the production process area (Liu et al. 2013, 2016; Sinha et al. 2007).
Fig. 2

Three-dimensional spatial distribution of concentrations (mg/kg) for 4 benzene homologue pollutants in soils

Vertical distribution of the benzene homologues

To study the vertical migration, soil contamination concentration profiles were measured near the polluted workshops. Sampling locations 1, 22 and 40 were chosen as the representative locations for the office area, the tribenuron workshops, and acetochlor workshops, respectively (Fig. 1). The concentrations of the target compounds at deeper layers in the office area (location 1) were below the limits of detection. However, at the tribenuron and acetochlor workshops, the TLE and EBZ concentrations ranged from 2.5 and 334 (surface) to 71.7 and 5470 mg/kg (15 m depth), respectively (Fig. 3), while the xylenes ranged from 332 to 6640 mg/kg, with two peaks at depth of 3 m and 15 m (Fig. 3). More than one concentration peaks of the pollutants in deep layers had been reported in laboratory simulations for gasoline (Ostendorf et al. 1993) and in field observations for petroleum hydrocarbons (Zhang et al. 2014). According to those studies, slow leaching on contaminants from the shallow layers of soil reached the deep low permeable layers over long periods and formed concentration peaks associated with these layers. Moreover, based on the lithological profile in this site (left panel in Fig. 3), it can be inferred that lithologic characters and boundaries of different soil layers might have played important roles on the formation of this vertical patterns of pollutants. The depth of 3 m in this site is the boundary between sand and silt layers, and the depth of 15 m is the boundary between silt and gravel layers. At the depth of > 15 m, the concentrations of pollutants dropped sharply almost to the detection limits. The concentrations of pollutants in groundwater (~ 70 m depth) were below the detection limits, probably because the low permeability silt layer intercepted the migration of benzene homologues; therefore, there was no or very limited downward migration to the groundwater. Many researchers observed complicated vertical distribution of immiscible liquid pollutants at different sites. They attributed the distribution complexity to many influential factors including moisture (Albergaria et al. 2010; Ostendorf et al. 1993), lithologic characters (Zhang et al. 2014), adsorption by soil particles (Zhang et al. 2018) and so on.
Fig. 3

Vertical distribution of benzene homologues in soil samples

Leaching is supported indirectly by the vertical distribution of the concentration ratios (Fig. S3). If the benzene homologues have same sources, due to the similar physicochemical properties, the ratios of individual compounds may be in small variations between locations or between soil layers. The EBZ/MPX ratios are similar at the same depth at locations 22 and 40, indicating that the residual EBZ and MPX may be influenced by same application time and application type (Pan et al. 2017). The ratios slightly increased from the surface to deeper layers. This fractionation pattern for individual pollutants suggests continuous migration along transport pathways. The pollutants have transported vertically about 15 m downward from their original source locations.

In summary, the residual benzene homologue in the research site had high concentrations in the soil and the distribution was significantly influenced by the production process. Contaminants had migrated 15 m downward from ground surface but had not reached groundwater. Their maximum concentrations were found at depths of 3 and 15 m. These pollutants were mostly accumulated on the interface between two different lithological layers, especially between the silt and gravel layers.

Benzene homologues transport and attenuation in unsaturated zone

The knowledge of the transport and fate of benzene homologue in the unsaturated zone is critical for accurate risk assessments and design of effective remedial actions. In the unsaturated zone, the target compounds are present in four phases: gas, dissolved in aqueous liquid, non-aqueous phase liquid (NAPL), and solid adsorption (Albergaria et al. 2010). The soils in this study area were mainly sandy and silty soils with negligible amounts of natural organic matter (< 0.5%). Therefore, the contaminants were mainly in gas phase, dissolved phase and NAPL phase.

Assuming all the benzene homologues mass is dissolved in pore water, the calculated concentrations are much greater than the aqueous solubility of these compounds (Table S1). Thus, NAPL phase must be present in the soil system. Based on the calculations from Eqs. 3 and 4, the distribution of the target compounds in each phase in the soil matrix was determined (Fig. 4). More than 99% of the pollutants existed in NAPL phase, while the proportion of other phases was less than 1%. Regarding NAPL mass transfer in unsaturated zone, local equilibrium between NAPL and soil gas phase is assumed to volatilize to pore space and generates high concentration vapor, close to saturated vapor pressures (Rivett et al. 2011). However, the concentrations of the target compounds in the soil gas were detected much lower than its saturated vapor pressures concentration (Table 3), suggesting pore water separated NAPL from the air in the pore space. It means that a water coating formed on the NAPL, which was in the “occluded” state (Petri et al. 2015). The “occluded” NAPL might be formed by co-release infiltrating water causing NAPL to be trapped, where the NAPL chemicals must transfer through the aqueous phase before volatilizing into gas phase.
Fig. 4

The distribution of the benzene homologues in different phase at different depth of location 22

Table 3

Concentrations of benzene homologues in soil gas at locations 22 and 40 (mg/m3)


Depth (m)





Saturated vapor pressures

















































NAPL dissolution may be rate-limited where the NAPL-water interfacial area is small. Although water saturation increased from 38.47% at the depth of 1.5 m to 66.84% at the depth of 15 m, percentage of NAPL volume was less than 3% (Table S2). Contaminant partitioning between dissolved and gas phase, including water to gas volatilization and the reverse vapor dissolution into aqueous process, is also a primary transport and attenuation process in unsaturated zone system. At equilibrium, mass transfer between the phase boundaries may be described by the temperature-dependent Henry’s law constant, H [Eq. (1)]. In addition, the porous structure of soil will provide pathways for the gas-phase pollutants advection and diffusion where concentration gradients exist. The significant linear relationship between the concentration ratio of TLE and EBZ in gas phase (Fig. 5) provided evidence of transport in the gas and aqueous phases rather than in NAPL phase. Hence benzene homologues vapor transport may lead to either reduction of groundwater impacts through diffuse into the atmosphere, or alternatively arrival at the water table through downward vapor migration. Greater lateral flow with increased travel distance was observed through inverse distance weighting model (Fig. 2), resulting from advection, dispersion caused by geological heterogeneity and vapor transport. It may bring health risk beyond the production workshop areas. Furthermore, biodegradation is considered as another natural attenuation process of benzene homologues in pore water. The availability of electron acceptors and a residual NAPL source determine the effectiveness of remediation by natural attenuation (Chiu et al. 2013; Jacome and Van Geel 2013; Kolhatkar and Schnobrich 2017; Wei et al. 2018; Zhao et al. 2015).
Fig. 5

Scatter plot of TLE and EBZ concentrations in air of the pore space

Health risks assessment

The study site is to be converted into a residential area, and some constructions have already been started (Fig. S4). Thus, an assessment of health risks from the benzene homologues compounds and soil remediation are urgently required.

Figure 6 shows the estimated carcinogenic risks (CRs) and hazard quotients (HQs) in the surface soil (0–1 m), which describe the carcinogenic and non-carcinogenic risks of the benzene homologues pollutants. Because there is no evidence of carcinogenicity for xylenes, only the carcinogenic risks of EBZ were assessed (Fig. 6a), while both EBZ and xylenes were assessed for non-carcinogenic risks (Fig. 6b). The high CRs appeared at the former tribenuron and acetochlor workshops (0.02 and 0.14, respectively, Fig. 6a), related to the high concentrations of EBZ in soil and in soil air at these two areas (Fig. 2 and Table 3). These values are three to four orders of magnitude higher than the acceptable value (10−6) regulated by the Ministry of Environment Protection of China (MEP-China 2014). Of these risks, the risk of inhalation was the primary contributor to the total carcinogenic risk (Fig. 6a). In addition, the CRs of the soils at the other workshops (5 × 10−8 to 7 × 10−7) were lower than the acceptable level, suggesting the low risks at those workshop areas.
Fig. 6

Health risk of benzene homologue pollutants: a carcinogenic risks; and b non-carcinogenic risks

HQs and CRs have similar patterns in the study field (Fig. 6b). The HQs of all the pollutants at tribenuron and acetochlor workshops (143–954 and 984–9765, respectively) are much higher than the threshold value of 1 for non-carcinogenic risk, and inhalation was the primary exposure pathway of pollutants for human body (Fig. 6b).

In the meantime, the health risk assessment for workers during sampling process uses a different calculation because exposure estimates are different from the residents. Based on the inhalation risks calculation, the average exposure intensity for the sampling workers should be doubled comparing to the residents, while the exposure duration for these workers only lasts 1–2 years comparing to 24 years for adult residents. Therefore, the CRs and HQs for the sampling workers are about 1/10 of the ones for residents.


Unexpectedly high concentrations of benzene homologues including toluene, ethylbenzene, m/p-xylene and o-xylene, were found at the study site of a former herbicide factory, posing risks to public health. The plan of developing this field site into a residential area brings up tremendous concerns regarding the high health risks from the contaminations. This site can be one outstanding example of many extremely polluted industrial sites in China.

From the IDW interpolation model, it was shown that benzene homologues residues in the soil were mainly from the herbicide production workshop and that pollution horizontal expanding occurred in the deeper soil layer. In the soil lithologic profile, benzene homologues had already migrated downward 15 m from ground surface to the silt layer, but had not reached groundwater. The contaminants were distributed in air, water (dissolved), and NAPL phases in the unsaturated zone. NAPL phase held more than 99% of the total contaminant mass and it could be described with the “occluded NAPL” phase. NAPL phase benzene homologues undergo dissolution into pore water and volatilization from pore water within the unsaturated zone. Both processes can increase the risks to human health. Based on the health risk assessment, EBZ brings unacceptable carcinogenic risks, and EBZ and xylenes bring non-carcinogenic risks at the tribenuron and acetochlor workshop areas. Hence, controlling each phase including dissolution, NAPL and gas is all important to reduce health risk for both residents and workers. It is important to survey, assess and remediate these sites before this field is developed into a residential area as planned.

Although a variety of treatment approaches including physical, chemical, biological and thermal have been employed to remove contaminants in dissolved and gas phases, removal of NAPL phase is still a challenge. However, as we suggested in this research, NAPL would continuously transform into other phases in a long period. Therefore, sustainable remediation is needed not only because it can assist in accelerating the efficient clean-up but also minimize the potential health risk in the long run (Hou and Al-Tabbaa 2014; O’connor et al. 2018; Song et al. 2019).



This work was financially supported by National Nature Science Foundation of China (Grant Nos. 41571309, 41702255). The authors thank Dr. Ping Gong for providing comments with good insights and improving the writing. We also thank Mr. Teng Quan for drawing the 3-D spatial distribution plots of benzene homologues.

Supplementary material

10653_2019_342_MOESM1_ESM.docx (1014 kb)
Supplementary material 1 (DOCX 1014 kb)


  1. Albergaria, J. T., Alvim-Ferraz, M. D. M., & Delerue-Matos, M. C. F. (2010). Estimation of pollutant partition in sandy soils with different water contents. Environmental Monitoring and Assessment, 171, 171–180.CrossRefGoogle Scholar
  2. Chen, S., & Brune, W. H. (2012). Global sensitivity analysis of ozone production and O-3-NOx-VOC limitation based on field data. Atmospheric Environment, 55, 288–296.CrossRefGoogle Scholar
  3. Chiu, H. Y., Hong, A., Lin, S. L., Surampalli, R. Y., & Kao, C. M. (2013). Application of natural attenuation for the control of petroleum hydrocarbon plume: Mechanisms and effectiveness evaluation. Journal of Hydrology, 505, 126–137.CrossRefGoogle Scholar
  4. Choi, J. W., Tillman, F. D., & Smith, J. A. (2002). Relative importance of gas-phase diffusive and advective trichloroethene (TCE) fluxes in the unsaturated zone under natural conditions. Environmental Science and Technology, 36, 3157–3164.CrossRefGoogle Scholar
  5. Dou, J., Liu, X., Hu, Z., & Deng, D. (2008). Anaerobic BTEX biodegradation linked to nitrate and sulfate reduction. Journal of Hazardous Materials, 151, 720–729.CrossRefGoogle Scholar
  6. Gooddy, D. C., Bloomfield, J. P., Harrold, G., & Leharne, S. A. (2002). Towards a better understanding of tetrachloroethene entry pressure in the matrix of Permo-Triassic sandstones. Journal of Contaminant Hydrology, 59, 247–265.CrossRefGoogle Scholar
  7. Hellweg, S., Fischer, U., Scheringer, M., & Hungerbühler, K. (2004). Environmental assessment of chemicals: Methods and application to a case study of organic solvents. Green Chemistry, 6, 418–427.CrossRefGoogle Scholar
  8. Hou, D., & Al-Tabbaa, A. (2014). Sustainability: A new imperative in contaminated land remediation. Environmental Science & Policy, 39, 25–34.CrossRefGoogle Scholar
  9. Hsieh, L.-T., Yang, H.-H., & Chen, H.-W. (2006). Ambient BTEX and MTBE in the neighborhoods of different industrial parks in Southern Taiwan. Journal of Hazardous Materials, 128, 106–115.CrossRefGoogle Scholar
  10. Huang, L., Mo, J., Sundell, J., Fan, Z., & Zhang, Y. (2013). Health risk assessment of inhalation exposure to formaldehyde and benzene in newly remodeled buildings, Beijing. PLoS One, 8(11), e79553.CrossRefGoogle Scholar
  11. Jacome, L. A. P., & Van Geel, P. J. (2013). An initial study on soil wettability effects during entrapped LNAPL removal by surfactant flooding in coarse-grained sand media. Journal of Soils and Sediments, 13, 1001–1011.CrossRefGoogle Scholar
  12. Kolhatkar, R., & Schnobrich, M. (2017). Land application of sulfate salts for enhanced natural attenuation of benzene in groundwater: A case study. Ground Water Monitoring and Remediation, 37, 43–57.CrossRefGoogle Scholar
  13. Lawrence, A., Stuart, M., Cheney, C., Jones, N., & Moss, R. (2006). Investigating the scale of structural controls on chlorinated hydrocarbon distributions in the fractured-porous unsaturated zone of a sandstone aquifer in the UK. Hydrogeology Journal, 14, 1470–1482.CrossRefGoogle Scholar
  14. Li, H., Zhang, Q., Wang, X. L., Ma, X. Y., Lin, K. F., Liu, Y. D., et al. (2012). Biodegradation of benzene homologues in contaminated sediment of the East China Sea. Bioresource Technology, 124, 129–136.CrossRefGoogle Scholar
  15. Li, P., Wu, J., Qian, H., Lyu, X., & Liu, H. (2014). Origin and assessment of groundwater pollution and associated health risk: A case study in an industrial park, northwest China. Environmental Geochemistry and Health, 36, 693–712.CrossRefGoogle Scholar
  16. Liu, G., Bi, R., Wang, S., Li, F., & Guo, G. (2013). The use of spatial autocorrelation analysis to identify PAHs pollution hotspots at an industrially contaminated site. Environmental Monitoring and Assessment, 185, 9549–9558.CrossRefGoogle Scholar
  17. Liu, G., Niu, J., Zhang, C., & Guo, G. (2016). Characterization and assessment of contaminated soil and groundwater at an organic chemical plant site in Chongqing, Southwest China. Environmental Geochemistry and Health, 38, 607–618.CrossRefGoogle Scholar
  18. Lu, G. Y., & Wong, D. W. (2008). An adaptive inverse-distance weighting spatial interpolation technique. Computers & Geosciences, 34, 1044–1055.CrossRefGoogle Scholar
  19. Masih, A., Lall, A. S., Taneja, A., & Singhvi, R. (2016). Inhalation exposure and related health risks of BTEX in ambient air at different microenvironments of a terai zone in north India. Atmospheric Environment, 147, 55–66.CrossRefGoogle Scholar
  20. Mendoza, C. A., & Frind, E. O. (1990). Advective-dispersive transport of dense organic vapors in the unsaturated zone: 2. Sensitivity analysis. Water Resources Research, 26, 388–398.CrossRefGoogle Scholar
  21. Ministry of Environmental Protection (2014) Technical guidelines for risk assessment of contaminated sites (HJ 25.3-2014). Beijing: China Environmental Science Press. (in Chinese).
  22. Molins, S., Mayer, K. U., Amos, R. T., & Bekins, B. A. (2010). Vadose zone attenuation of organic compounds at a crude oil spill site—interactions between biogeochemical reactions and multicomponent gas transport. Journal of Contaminant Hydrology, 112, 15–29.CrossRefGoogle Scholar
  23. O’connor, D., Hou, D., Ok, Y. S., Song, Y., Sarmah, A. K., Li, X., et al. (2018). Sustainable in situ remediation of recalcitrant organic pollutants in groundwater with controlled release materials: A review. Journal of Controlled Release, 283, 200–213.CrossRefGoogle Scholar
  24. Oostrom, M., Rockhold, M. L., Thorne, P. D., Truex, M. J., Last, G. V., & Rohay, V. J. (2007). Carbon tetrachloride flow and transport in the subsurface of the 216-Z-9 trench at the hanford site. Vadose Zone Journal, 6, 971–984.CrossRefGoogle Scholar
  25. Ostendorf, D. W., Richards, R. J., & Beck, F. P. (1993). LNAPL retention in sandy soil. Ground Water, 31, 285–292.CrossRefGoogle Scholar
  26. Pan, H., Liu, H., He, X., Xi, B., Han, Y., & Xu, Q. (2017). Levels and distributions of organochlorine pesticides in the soil-groundwater system of vegetable planting area in Tianjin City, Northern China. Environmental Geochemistry and Health, 39, 417–429.CrossRefGoogle Scholar
  27. Pankow, J. F., & Cherry, J. A. (1996). Dense chlorinated solvents and other DNAPLs in groundwater: History, behavior, and remediation. FAO of the UN: Waterloo Press.Google Scholar
  28. Parker, J. C. (2003). Physical processes affecting natural depletion of volatile chemicals in soil and groundwater. Vadose Zone Journal, 2, 222–230.CrossRefGoogle Scholar
  29. Perez-Rial, D., López-Mahía, P., Muniategui-Lorenzo, S., & Prada-Rodríguez, D. (2009). Temporal distribution, behaviour and reactivities of BTEX compounds in a suburban Atlantic area during a year. Journal of Environmental Monitoring, 11, 1216–1225.CrossRefGoogle Scholar
  30. Petri, B. G., Fučík, R., Illangasekare, T. H., Smits, K. M., Christ, J. A., Sakaki, T., et al. (2015). Effect of NAPL source morphology on mass transfer in the vadose zone. Groundwater, 53, 685–698.CrossRefGoogle Scholar
  31. Rivett, M. O., Wealthall, G. P., Dearden, R. A., & Mcalary, T. A. (2011). Review of unsaturated-zone transport and attenuation of volatile organic compound (VOC) plumes leached from shallow source zones. Journal of Contaminant Hydrology, 123, 130–156.CrossRefGoogle Scholar
  32. Sinha, P., Lambert, M. B., & Schew, W. A. (2007). Evaluation of a risk-based environmental hot spot delineation algorithm. Journal of Hazardous Materials, 149, 338–345.CrossRefGoogle Scholar
  33. Song, Y., Kirkwood, N., Maksimović, Č., Zheng, X., O’connor, D., Jin, Y., et al. (2019). Nature based solutions for contaminated land remediation and brownfield redevelopment in cities: A review. Science of the Total Environment, 663, 568–579.CrossRefGoogle Scholar
  34. Wang, T., Bo, P., Bing, T., Zhaoyun, Z., Liyu, D., & Yonglong, L. (2014). Benzene homologues in environmental matrixes from a pesticide chemical region in China: Occurrence, health risk and management. Ecotoxicology and Environmental Safety, 104, 357–364.CrossRefGoogle Scholar
  35. Wang, X.-P., Gong, P., Yao, T.-D., & Jones, K. C. (2010). Passive air sampling of organochlorine pesticides, polychlorinated biphenyls, and polybrominated diphenyl ethers across the Tibetan Plateau. Environmental Science and Technology, 44, 2988–2993.CrossRefGoogle Scholar
  36. Wei, Y., Thomson, N. R., Aravena, R., Marchesi, M., Barker, J. F., Madsen, E. L., et al. (2018). Infiltration of sulfate to enhance sulfate-reducing biodegradation of petroleum hydrocarbons. Ground Water Monitoring and Remediation, 38, 73–87.CrossRefGoogle Scholar
  37. Yuan, B., Shao, M., Lu, S., & Wang, B. (2010). Source profiles of volatile organic compounds associated with solvent use in Beijing, China. Atmospheric Environment, 44, 1919–1926.CrossRefGoogle Scholar
  38. Zhang, Q., Wang, G. C., Sugiura, N., Utsumi, M., Zhang, Z. Y., & Yang, Y. N. (2014). Distribution of petroleum hydrocarbons in soils and the underlying unsaturated subsurface at an abandoned petrochemical site, North China. Hydrological Processes, 28, 2185–2191.CrossRefGoogle Scholar
  39. Zhang, H., Zuo, R., Wang, J., Jin, C., Guo, X., & Teng, Y. (2018). The underground migration and distribution of petroleum contamination at a gas station. China Environmental Science, 38, 1532–1539.Google Scholar
  40. Zhao, Y., Qu, D., Hou, Z., & Zhou, R. (2015). Enhanced natural attenuation of BTEX in the nitrate-reducing environment by different electron acceptors. Environmental Technology, 36, 615–621.CrossRefGoogle Scholar
  41. Zhou, Y.-Y., Yu, J.-F., Yan, Z.-G., Zhang, C.-Y., Xie, Y.-B., Ma, L.-Q., et al. (2013). Application of portable gas chromatography-photo ionization detector combined with headspace sampling for field analysis of benzene, toluene, ethylbenzene, and xylene in soils. Environmental Monitoring and Assessment, 185, 3037–3048.CrossRefGoogle Scholar
  42. Zouir, A., Esteve-Turrillas, F. A., Chafik, T., Morales-Rubio, A., & De La Guardia, M. (2009). Evaluation of the soil contamination of Tangier (Morocco) by the determination of BTEX, PCBs, and PAHs. Soil and Sediment Contamination, 18, 535–545.CrossRefGoogle Scholar

Copyright information

© Springer Nature B.V. 2019

Authors and Affiliations

  • Shuo Yang
    • 1
    • 2
  • Xiulan Yan
    • 1
    Email author
  • Lirong Zhong
    • 3
  • Xuejiao Tong
    • 4
  1. 1.Institute of Geographic Sciences and Natural Resources ResearchChinese Academy of SciencesBeijingChina
  2. 2.Institute of Process EngineeringChinese Academy of SciencesBeijingChina
  3. 3.Energy and Environment DirectoratePacific Northwest National LaboratoryRichlandUSA
  4. 4.Yuhuan Environmental Technology Co., Ltd.ShijiazhuangChina

Personalised recommendations