Unveiling the Potential: Selecting Optimal Materials for Physical Pools in a Pavement-Runoff-Integrated Treatment System

Abstract

Pavement runoff contains complex pollutants that can lead to environmental pollution and health risks. A pavement-runoff-integrated treatment system has been recognized as an effective way to deal with pavement runoff pollution. However, there is little support for selecting appropriate materials for physical pools due to a lack of understanding of the selective filtration and physical adsorption characteristics. In this study, gravel and activated carbon were chosen as the substrate materials for physical filtration and adsorption pools, and their corresponding purification characteristics were investigated using an indoor scaled down model. The results showed that the removal rate of all pollutants was related to the size of the gravel used. This was mainly due to the increased gravel particle size and voids, which resulted in a higher water velocity, shorter hydraulic retention time, and inadequate filtration. Compared with coconut shell granular activated carbon (GAC) and coal column activated carbon (EAC), analytically pure granular activated carbon (ARAC) showed a better removal rate for petroleum and heavy metals. This is mainly because ARAC has a larger specific surface area, higher pore volume, and wider pore size distribution, resulting in a remarkable adsorption capacity for pollutants. Overall, the combination of 0.3 mm gravel and ARAC was found to be the most suitable for use as filtration and adsorption materials for physical pools. These findings offer a gravel- and ARAC-based pavement-runoff-integrated treatment system, which has excellent potential to enhance the removal of pollutants from pavement runoff.

1. Introduction

As the global traffic infrastructure and capacity continue to expand, their key role in economic development is becoming more and more prominent. However, these advancements also bring about a number of problems. For instance, pollutants from pavement runoff can flow into drinking water source protection areas and other water-sensitive areas, causing harm to the environment and potentially affecting human health. Wang et al. [1] demonstrated a strong positive correlation between the water quality near highways and both pavement construction and traffic flow. The pavement has been identified as the primary source of runoff pollutants in the Feitsui Reservoir. Kriech et al. [2] found that coal tar sealers, commonly used in military air bases, parking lots, and pavement, contain high levels of polycyclic aromatic compounds that are not easily soluble in water. These pollutants can be washed off by stormwater runoff and enter local water sources, which degrades the quality of the receiving waters and causes genotoxic effects to living fish and other aquatic organisms. Hwang et al. [3] suggested that the daily operation of vehicles on roads releases a wide variety of pollutants, which are often picked up by stormwater runoff and delivered into local water sources. Nikolaeva et al. [4] studied the correlations between traffic-related contaminants in roadside soils and their ecotoxicity, finding that higher plant and animal toxicity was correlated with total petroleum hydrocarbons, polycyclic aromatic hydrocarbons, and some heavy metals, while microorganism toxicity was correlated with total petroleum hydrocarbons. Müller et al. [5] revealed that vehicular transportation and road maintenance were the major sources of urban runoff pollution, including vehicle operation, exhaust gases and particles, vehicle wear, tires, brakes, road abrasion, road salts, combustion, oils, and other factors. These pollutants can have negative effects on human and animal health if they reach high concentrations. Additionally, pavement runoff [6] has also been recognized as an important source of micropollutants in groundwater. These micropollutants not only have ecotoxic effects on aquatic organisms but also negatively impact water biodiversity.

The composition and concentration levels of pollutants in pavement runoff vary greatly and depend on factors such as road materials, traffic type, traffic intensity, climate, pollutant treatment methods, and distance to the receiving water. Martin et al. [7] showed that asphalt mixtures used in road construction have the potential to release particulate matter, polycyclic aromatic hydrocarbons, and other organic pollutants into the environment as contaminants. Li et al. [8] found that volatile organic compounds are released from asphalt pavement construction, with the emission concentrations varying from 4.24 to 104.16 mg/m³, which can have a negative impact on both the environment and human health. Kou et al. [9] maintained that traffic characteristics have a significant effect on the types of pollutants present, with higher concentrations of suspended solids (SS) and heavy metals often found at intersections and higher concentrations of organic pollutants often found in commercial areas. Kibblewhite [10] indicated that road sediments are the main source of pavement runoff pollution, containing metals, toxic organics, and plastics. The levels of metals, such as Cu and Zn, near the highways are often higher than the normal value, which can affect the production of agricultural crops. Baah et al. [11] suggested that road dust is a complex combination of heavy metals, such as Cu and Zn, with the highest metal concentrations observed at a distance of approximately 30 to 50 m from the road. Hou et al. [12] found that tire abrasion, lubricants, and vehicular corrosion are thought to be major sources of Cu and Zn pollution in road dust. Baensch-Baltruschat et al. [13] believed that organic compounds, such as tire and road wear particles, also show highly variable concentrations in road runoff.

In recent years, numerous treatment methods have been developed to address pavement runoff pollution. These methods include sedimentation, filtration, adsorption, biological degradation, and others. Tan et al. [14] designed an ecological bridge surface runoff treatment three-tank system, which consisted of a storage regulation tank, an oil separation sedimentation tank, and an emergency storage tank. Wu et al. [15] investigated the physical filtration ability of five types of media, namely zeolite, ceramic granules, slag, diatomaceous earth, and vesuvianite, on pollutants, finding that the performance is highly dependent on the type of media material used in the system. The physical filtration ability of different filtration media is related to the layer thickness and particle size; thicker layers and smaller particle sizes result in better filtration ability. Therefore, choosing the right filtration materials is crucial for improving the overall filtration performance. Xu et al. [16] reported on the physical filtration ability of four types of infiltration materials, diatomite, volcanic, slag, and zeolite, and showed that the pore adsorption and the mechanical interception of the gap between materials were two modes of infiltration materials to intercept runoff pollutants. They found that the average rate of SS removal is maintained at a high level of around 70%, while the average rate of COD removal is poor, at about 40~50%. Shukla et al. [17] used the chemical oxygen demand (COD) as an index to describe the organic compounds and showed that the activated carbon in a chemical lab can absorb SS and COD at a relatively high rate, ranging from 43% to 100%. Lin et al. [18] indicated that physical retention in the surface layer of the bricks plays a crucial role in the SS filtration process. However, it is difficult for one method to achieve the ideal treatment effect. Wang et al. [19] proposed a combined low-impact development (LID) treatment system to capture and treat onsite stormwater runoff. Their results showed that Zn metal appeared to be highly particulate-bound and easy to remove, while Cu metal was mostly dissolved and more difficult to remove. The LID treatment system performed well in carrying peak runoff flows and removing pavement runoff pollutants compared to a single medium. Ziajahromi et al. [20] showed that synthetic rubber particles are one of the pollutants in pavement runoff and that the size and concentration of synthetic rubber particles were reduced in the outlet compared to the inlet of stormwater treatment wetlands.

A pavement-runoff-integrated treatment system is a promising eco-technology due to its low construction cost, environmental friendliness, and good comprehensive purification effect. It is composed of three parts, namely the physical pool, chemical pool, and biochemical pool, which can be used to remove pollutants from varying pavement runoff. Wang et al. [21] analyzed the role of aquatic plants in a pavement-runoff-integrated treatment system for the purification of pavement runoff. Their results showed that aquatic plants can effectively remove SS, COD, petroleum, Zn, and Cu. The treatment effect increased with the increasing planting density. Qian et al. [22] studied the removal effect of three types of constructed wetland structures in the pavement-runoff-integrated treatment system. They found that the tandem wetland structures were more effective than the general vertical flow wetland structures, with removal rates of heavy metals exceeding 90%, the removal rate of petroleum exceeding 60%, and the removal rate of COD exceeding 70%. However, the function and mechanism of physical pools in removing pavement runoff pollution are still unclear. In this study, a well-designed runoff purification test was carried out under different combinations of gravel and activated carbon using an indoor scaled down experimental model, and the pollutant removal efficiency and working mechanisms of the physical pools were also investigated. The aim of this study was to provide a guide for choosing the right material for the physical pools in the pavement-runoff-integrated system.

2. Materials and Methods

2.1. Design of an Indoor Scaled Down Model of a Pavement-Runoff-Integrated Treatment System

The prototype of the pavement-runoff-integrated treatment system is located in Zengcheng Expressway, Guangzhou, China, as shown in Figure 1. It is composed of a physical filtration pool, a physical adsorption pool, a chemical treatment pool, and a biochemical pool. In this system, the pavement runoff first moves to the physical filtration pool, which acts as a sediment trap to remove floaters and SS. Then, the runoff flows to the physical adsorption pool, where it undergoes adsorption to remove molecular or ionic substances such as petroleum and heavy metals. From there, it flows into the chemical treatment pool, which is only used in the event of toxic or hazardous substances leaks. Next, it flows to the biochemical pool to further purify the runoff to meet the water-quality discharge standards. Finally, it is discharged into the natural water system. To better understand the effectiveness of this system, an indoor scaled down model was built based on the prototype of the pavement-runoff-integrated treatment system (Figure 2). The scaling fraction of this model is 1:10. Additionally, an overflow port was added to the retaining wall of the physical filtration pool to handle sudden rainstorms or clogging risks.

2.2. Material Selection of the Physical Filtration and Adsorption Pools

The materials used in physical pools are regularly replaced, providing a platform for physically removing pollutants through filtration and adsorption. The selection of these materials plays a crucial role in determining the cost and effectiveness of a pavement-runoff-integrated treatment system. In order to meet the requirements of such a pavement-runoff-integrated treatment system, the gravel is available from Yueyang Zhongzhou Quartz Sand Factory, Yueyang, China, in several sizes as the filtration material in the physical filtration pool, which is often used for decorative purposes due to the low cost and naturally smooth shape of the stones [23,24,25]. Additionally, the activated carbon is available in several types from Pingdingshan Lvzhiyuan Activated Carbon Co., Ltd., Pingdingshan, China, including coal-based columnar activated carbon (EAC), coconut shell granular activated carbon (GAC), and analytically pure granular activated carbon (ARAC). These activated carbons are utilized as the adsorption material in the physical adsorption pool due to its low cost and high adsorption ability [26,27,28,29,30].

Twelve combinations of gravel and activated carbon were tested to select the best combination for maximizing the purification capability of physical pools. These combinations were named 0.3EAC (0.3 mm gravel and EAC), 0.3GAC (0.3 mm gravel and GAC), 0.3ARAC (0.3 mm gravel and ARAC), 0.6EAC (0.6 mm gravel and EAC), 0.6GAC (0.6 mm gravel and GAC), 0.6ARAC (0.6 mm gravel and ARAC), 1.18EAC (1.18 mm gravel and EAC), 1.18GAC (1.18 mm gravel and GAC), 1.18ARAC (1.18 mm gravel and ARAC), 2.36EAC (2.36 mm gravel and EAC), 2.36GAC (2.36 mm gravel and GAC), and 2.36ARAC (2.36 mm gravel and ARAC), as shown in Figure 3.

2.3. Preparation of Artificial Pavement Runoff Pollutant

In order to obtain the artificial pavement runoff pollutant required for the indoor scaled down model, sediment was collected from the curbstone of the entrance to Wanjiali South Road, Changsha Bypass Expressway, China, at a range of 0–50 cm. This is because most of the road sediment is deposited at both sides of the road within 50 cm of the curbstone [31]. The collected area was 5 m². The process of preparing the pavement runoff pollutant is described in Figure 4. First, the sediment was collected by hand-sweeping after one week of sunny weather. The large floaters were removed, such as leaves. Next, 2.5 kg of collected road sediment was dissolved into 160 L of tap water and stirred to create the artificial pavement runoff pollutants. Finally, the as-prepared pavement runoff pollutant was divided into a control group and an experimental group.

2.4. Purifying Procedure

When the experimental group (as shown in Figure 4) was poured into the pavement runoff pool, the pavement runoff pollutants flowed to the physical pools (as shown in Figure 3). The large particles of SS in the pavement runoff pollutants were deposited in the gravel of the physical filtration pools. Then, most of the pollutants, including petroleum and heavy metals, ran into the physical adsorption pools and were absorbed by activated carbon. Finally, the pavement runoff was purified and flowed to the chemical treatment pool.

2.5. Evaluation Indicator and Methodology

The evaluation of the purifying effect is based on the monitoring of the pollutants’ concentrations both in the inlet water and outlet water, including SS, COD, petroleum, and heavy metals (Zn and Cu). The removal performance of each pollutant was evaluated by introducing the removal rate η, as follows.

$$\eta =\frac{C_0-C_1}{C_0}\times 100\%,$$

(1)

where C₀ is the influent concentration of the pollutant in the inlet, and C₁ is the effluent concentration of the pollutant in outlet 1.

The standard gravimetric method of GB 11901-89 [32] was used to evaluate the SS with an electronic balance from Lichen Technology Co., Ltd., Shanghai, China, while the concentration of heavy metals (Cu and Zn) was determined using the standard atomic absorption spectrometry of GB 7475-87 with a GDSYS-201M multiparameter water quality analyzer from Changzhou Yitong Analytical Instruments Manufacturing Co., Ltd., Changzhou, China. The COD was measured using the standard dichromate method of HJ 828-2017 [33] with the GDSYS-201M multiparameter water quality analyzer, and the concentration of petroleum was assessed using the standard infrared spectrophotometry of HJ 637-2018 [34] with the MAI-50G infrared oil meter analyzer from Changzhou Yitong Analytical Instruments Manufacturing Co., Ltd., Changzhou, China.

Table 1. Referenced standard methods and testing equipment for the pavement runoff pollutants.

Number	Evaluation Indicator	Referenced Standards	Testing Equipment
1	SS	GB 11901-89 [32]	Oven, Electronic Balance
2	COD	HJ 828-2017 [33]	GDYS-201M Water Quality Analyzer
3	Petroleum	HJ 637-2018 [34]	MAI-50G Infrared Oil Meter Analyzer
4	Cu	GB 7475-87 [35]	GDYS-201M Multiparameter Water Quality Analyzer
5	Zn	GB 7475-87 [35]	GDYS-201M Multiparameter Water Quality Analyzer

displays the referenced standard methods and testing equipment for each pollutant.

3. Results and Discussion

3.1. Results

Figure 5a illustrates the SS removal rate for various combinations of gravel and activated carbon. These combinations had varying effects on the purification of pollutants in the as-prepared pavement runoff. The removal rates for SS, petroleum, COD, Cu, and Zn ranged from 71.49 to 91.14%, 34.85% to 60.93%, 44.01 to 49.80%, 6.67 to 33.33%, and 12.24 to 30.61%, respectively. It was observed that the treatment effect of the physical pools on pollutants decreased with the increase in gravel size, which indicated the gravel size affects the purification process. Furthermore, the treatment effect of the ARAC combinations was found to be superior to that of the EAC and GAC combinations, which could be attributed to the varying adsorption capacities of the activated carbons. The highest removal rates for SS, petroleum, COD, Cu, and Zn were achieved with the combination of 0.3 mm gravel and activated carbons.

To better illustrate the effect of gravel size on water quality, the average purification effect of different types of activated carbon was extracted from Figure 5. This was conducted to more clearly present the relationship between gravel size and water quality. The mean removal rate (M) for each pollutant under the same gravel size can be calculated using the following equation.

$$M=\frac{C_{ARAC}+C_{GAC}+C_{EAC}}{3}$$

(2)

The relationship between the mean removal rate of pollutants and gravel size is shown in Figure 6. It is evident that as the gravel size increases, the removal rate of all pollutants decreases. The gravel size played a more significant role in the removal of pollutants. The removal rate of SS was over 72%, indicating that the gravel had an excellent filtration capacity for SS. However, the filtration capacity for the other pollutants was not as high. The removal rate for COD ranged from 46% to 65%, while the removal rate for petroleum varied from 44% to 63%. The removal rates for Zn and Cu fluctuated between 15% and 40%.

To determine the optimal combination of gravel size and activated carbon type, the pollutant removal effects for the combinations of 0.3 mm gravel and activated carbons are presented in Figure 7. This is important because, as noted, the combinations of 0.3 mm gravel and activated carbons had the highest removal rates for SS, petroleum, COD, Cu, and Zn. The results showed that these combinations were effective in removing pollutants from pavement runoff, with a removal rate of more than 82% for SS, 49% to 75% for COD, 60% to 67% for petroleum, and 30% to 47% for heavy metals. The removal rate for SS, petroleum, COD, Cu, and Zn in combination were found to be superior to those of permeable bricks (20–32%) [18]. The average concentrations at 0.3EAC were reduced from 92.6 mg/L to 8.2 mg/L for SS, 12.08 mg/L to 4.72 mg/L for petroleum, 45.1 mg/L to 22.64 mg/L for COD, 1.5 mg/L to 1 mg/L for Cu, and 0.49 mg/L to 0.34 mg/L for Zn. The average concentrations at 0.3GAC were reduced from 60.6 mg/L to 10.8 mg/L for SS, 7.26 mg/L to 2.68 mg/L for petroleum, 39.17 mg/L to 10.37 mg/L for COD, 0.67 mg/L to 0.37 mg/L for Cu, and 0.16 mg/L to 0.09 mg/L for Zn. The average concentrations at 0.3ARAC were reduced from 79.8 mg/L to 12.7 mg/L for SS, 6.43 mg/L to 2.05 mg/L for petroleum, 40.98 mg/L to 10.12 mg/L for COD, 1.4 mg/L to 0.77 mg/L for Cu, and 0.19 mg/L to 0.1 mg/L for Zn. The combination of 0.3EAC had the highest removal rate for SS, while the combination of 0.3ARAC had the highest removal rates for petroleum, COD, Cu, and Zn. Overall, the purification capabilities of these combinations can be ranked as follows: 0.3ARAC > 0.3GAC > 0.3EAC.

3.2. Discussion

To clarify the mechanism of the impact of gravel size on the rate of pollutant removal, we conducted a study on the porosity and water flow rate of the gravels.

The porosity of gravel, denoted as P, can be determined using the following formula.

$$P=(1-\frac{D_B}{D_A})\times 100\%,$$

(3)

where D_B represents the bulk density of the gravels, and D_A represents the apparent density of the gravels.

The D_B of the gravels is calculated by dividing the weight of the gravels by the total volume they occupy, while the D_A of the gravels is calculated by dividing the weight of the gravels by the volume of the impermeable portion they occupy. The test method for determining D_B and D_A follows the standard test methods for aggregates (JTG E42-2005) [36]. The results of our study are presented in

Table 2. The porosity of the gravels.

Material	Bulk Density (g/cm3)	Apparent Density (g/cm3)	Porosity (%)
0.3 mm gravel	1.5850	2.6549	40.2974
0.6 mm gravel	1.5770	2.6549	40.5999
1.18 mm gravel	1.5450	2.6316	41.2914
2.36 mm gravel	1.5187	2.6549	42.7949

.

Figure 8 shows the relationship between porosity and gravel size. It is evident that as the gravel size increases, the porosity also increases gradually. This can be attributed to the fact that larger gravels have more spaces between them, allowing for incompact deposit and ultimately resulting in a higher porosity.

In order to gain a better understanding of the relationship between the maximum water flow velocity and gravel size, a water meter at was used to control the water flow velocity of the artificial pavement runoff in the model. The physical filtration pool was filled with gravel of different particle sizes, while the physical adsorption pool remained empty. The maximum water flow rate (V) of the gravel was measured when the water reached the overflow port at the physical filtration pool.

$$V=\frac{Q}{A}$$

(4)

where Q is the volume of runoff passing through the connecting pipe of the physical filtration pool during a specific time period, and A is the cross-sectional area of the connecting pipe. The applied hydraulic loads ranged from 0.003 to 0.008 m³, while the hydraulic residence time was maintained between 1 and 2 min.

Figure 9 displays the relationship between the maximum water flow velocity and gravel size, where the maximum water flow velocity gradually increases with the increasing gravel size. Combined with Figure 8, it is evident that introducing an increase in the particle size of gravel is beneficial for increasing the porosity and flow velocities. Therefore, the mechanism by which increasing the gravel size is detrimental to pollutant removal originates from the shortening of the hydraulic residence time and weakening of the filtration capacity.

Figure 10a displays the N2 adsorption–desorption isotherms of EAC, GAC, and ARAC at 77 K, measured using the BET method with a fully automated Micromeritics ASAP 2460 Automatic Specific Surface and Porosity Analyzer from Micromeritics Instruments Corporation, Atlanta, Georgia, USA. The isotherms of all three activated carbon samples exhibit a typical Ⅳ type curve. The curve initially rises steeply at lower relative pressures, indicating the presence of micropores for monomolecular layer adsorption. As the relative pressure increases, the curve gradually levels off, indicating multilayer adsorption. As the relative pressure further increases, a H4-type hysteresis loop appears, indicating the occurrence of the capillary condensation phenomenon and the presence of mesopores. When the relative pressure approaches 1.0, the growth rate of the curve steepens, indicating coalescence occurs in all pores, and adsorption occurs only on the outer surface as macroporous adsorption. The specific surface area of each activated carbon was calculated by using the BET equation [37], and the specific surface areas of EAC, GAC, and ARAC were 370 m²/g, 1002 m²/g, and 1480 m²/g, respectively. This indicates that ARAC has the largest specific surface area and therefore the best adsorption performance.

The pore volume distributions of the three activated carbons were calculated based on the N2 adsorption at a relative pressure of P/P0 = 0.95, as shown in Figure 10b. The inset shows an enlargement of Figure 10b. It can be seen that the ARAC has the highest overall pore volume peak intensity, followed by GAC, and that of EAC has the lowest, indicating that ARAC has the most pores, followed by GAC and EAC. All three types of activated carbon have a wide range of pore sizes, from 2 nm to 40 nm. GAC and ARAC have a peak at around 1.5 nm, indicating a predominance of micropores, while EAC peaked at both 1.5 nm and 15 nm, indicating a significant presence of both micropores and mesopores. Significantly, GAC and ARAC also contain macropores. This confirms that ARAC and GAC are more effective at adsorbing molecular or ionic substances, such as petroleum and heavy metals, while the EAC has a stronger filtration capacity for SS due to its lower number of micropores and mesopores.

Therefore, the adsorption capacity of activated carbon is closely related to its specific surface area, total pore volume, and pore size distribution. A larger specific surface area provides more active adsorption sites and a stronger adsorption capacity. A higher pore volume and wider pore size distribution also contribute to a better adsorption capacity. Therefore, the excellent pollutant removal ability of ARAC is due to its large specific surface area, high pore volume, and wide pore size distribution.

4. Conclusions

This study primarily aims to evaluate the selection of filtration and adsorption materials for physical pools in the treatment of pavement runoff pollution. In order to evaluate the selective filtration and adsorption characteristics and explore their corresponding mechanisms, a scaled down model of the pavement-runoff-integrated treatment system was proposed. The results show that:

(1)

The filtration characteristics of gravel are size-dependent. This is mainly due to the fact that larger gravels have more spaces between them, allowing for incompact deposits and higher porosity, resulting in a higher water velocity, shorter hydraulic retention time, and inadequate filtration.

(2)

The three adsorption materials exhibit different adsorption capabilities. The combination of ARAC had the highest removal rates for petroleum, COD, Cu, and Zn. This is mainly because ARAC has a larger specific surface area, higher pore volume, and wider pore size distribution, resulting in the best adsorption performance for pollutants.

(3)

An optimal combination of gravel and activated carbon is proposed based on the removal rate of pollutants in this system. The combination of 0.3ARAC presents superior capability to remove pollutants and is more suitable for physical filtration and adsorption pools.

It is important to choose suitable filtration and adsorption materials in physical pools for pavement-runoff-integrated treatment systems, as suggested by the results. However, the replacement period of filtration and adsorption materials could be a key factor for the stable operation of the pavement-runoff-integrated treatment system. Therefore, the long-term performance of these filtration and adsorption materials should be further investigated.