Enhanced Benzofluoranthrene Removal in Surface Flow Constructed Wetlands with the Addition of Carbon.

Polycyclic aromatic hydrocarbons (PAHs), as hazardous pollutants, could be removed by constructed wetlands (CWs). While the traditional substrate of CWs has a weak adsorption capacity for PAHs, in this study, the carbonous fillers-activated carbon (AC) and biochar-were added into the substrate of surface flow CWs to improve the removal performance of benzofluoranthrene (BbFA), a typical PAH. The results showed that the BbFA removal efficiencies in CWs with the addition of AC and biochar were 11.8 and 1.2% higher than those in the Control group, respectively. Simultaneously, the removal efficiencies of NO3 --N were 42.8 and 68.4% in these two CWs, while the BbFA content in the substrate and plants with the addition of carbon was lower than that in the Control group. The addition of carbonous filler reduced the absorption of PAHs by plants in CWs and enhanced microbial degradation. The microbial community results showed that the relative abundance of Proteobacteria, especially γ-proteobacteria, was higher with the addition of fillers, which related to PAH degradation.

Introduction

Polycyclic aromatic hydrocarbons (PAHs) are typical persistent organic pollutants with biotoxicity and teratogenic carcinogenicity. PAHs accumulate in the biological chain and thus are harmful to the ecological environment and the health of organisms.[1] It has been detected in watershed environments around the world, e.g., Yangtze River (China),[2] Elbe River (Germany),[3] Brisbane River (Australia),[4] Ovia River (Southern Nigeria),[5] and other rivers. PAHs (>3 rings) have a stable chemical structure with the character of lipotropy and hydrophobicity and thus are difficult to be biodegraded.[6] The current methods for PAH removal mainly include chemical methods, such as coagulation,[7] chemical oxidation,[8] ultrasonic,[9] and adsorption.[10] However, these methods are not suitable for water pollution treatment in large river basins. Constructed wetland (CW) is an ecological project for pollution treatment in an aquatic environment with low energy consumption, high removal efficiency, and simple operation management.[11] It has been proved to have an effective removal capacity for nitrogen, phosphorus, and organic pollutants from water by the synergistic effect of plants, substrates, and microorganisms.[12]

Previous research revealed that substrate adsorption and microbial degradation played important roles in organic contaminant treatment in CWs, which contributed to 28.67–61.00% of pollutant removal.[13] As the main framework of CWs, the substrate provided a medium for plant roots and microorganisms’ culture and provided the location for contaminant degradation by microbes. Meanwhile, due to the sedimentation and hydrophobicity of PAHs, PAH pollutants mainly accumulated in the substrate.[14] Thus, improving the adsorption capacity and microbial degradation of pollutants in the substrate were important for the removal of PAHs in CWs. An effective method to enhance the performance of the substrate is the addition of a filler. The main substrate fillers in CWs are gravel, sand, and soil,[15] but these fillers showed lower adsorption capacity for PAHs. Some other absorbents, such as montmorillonite and clay, have good adsorption properties for PAHs but cause blockage of CWs when added as substrate fillers.[16]

The carbon adsorbent has stable physical and chemical properties, well-developed specific surface area, abundant surface functional groups, and shows good performance for the PAH adsorption from wastewater.[17] As a filler, the carbon adsorbent was predicted to improve the adsorption capacity of the substrate as well as promote the PAH transformation and degradation by microbes in the habitat provided by the well-developed pore structures.[12] Thus, the removal efficiency of PAHs can be further improved in CWs. Moreover, carbon adsorbent played an important role in plant growth and microbial culture.

Based on the potential role of carbon adsorbent for PAH removal in CWs, the biochar and activated carbon (AC) were added to the substrate to establish CW microcosms. Benzofluoranthrene (BbFA), a typical PAH with 5 rings, was selected as a PAH pollutant in this work. The objectives of this research were to (1) observe the treatment performance of CWs with biochar and AC; (2) evaluate the influence of these two materials on PAH removal in wastewater, substrate, and plants in CWs; and (3) explore the effect of the two materials on the microbial community. This research provides useful information for PAH removal in an aquatic environment.

Results and Discussion

Performance of CWs in Wastewater

The BbFA concentration in the wastewater of each group is shown in Figure a. The BbFA removal efficiencies in the experimental treatments followed the order A-CW > B-CW > Control. The average removal efficiency of BbFA in A-CW was up to 99.0 ± 0.23%, which was 10.6 and 1.2% higher than that in the Control and B-CW groups, respectively (p > 0.05). The results reflected that both AC and biochar promoted the removal of BbFA in CWs, and could be attributed to the immobilization and adsorption of organic contaminants due to their hydrophobic properties.[18] Brennan et al.[19] found that the AC has better performance for PAH reduction in porewater than biochar, while the biochar had larger particle sizes.

Figure 1

Concentrations of BbFA (a), NO3––N (b), NH4+–N (c), and total phosphorus (TP) (d) and in the wastewater of each group.

The concentrations of the main pollutants were also measured to evaluate the operation conditions of CWs. As shown in Figure , there was a great variation in concentrations of each inorganic nitrogen among the three groups. Both biochar and AC showed better performance for nitrate (NO3––N) removal, especially B-CW, when the NO3––N concentration in the effluent was less than 16.9 mg/L; the removal efficiency was 30.4 ± 2.77%, which was 68.4% higher than that in the Control group. The addition of AC in CWs also increased the NO3––N removal efficiencies, which was 42.8% higher than that of the Control group. For ammonium (NH4+–N) concentrations, as shown in Figure c, the B-CW group showed better performance of NH4+–N reduction with an average removal efficiency of 86.3 ± 3.29%, while no significant difference of NH4+–N reduction in A-CW was observed when compared to the Control group (p > 0.05). As shown in Figure d, the CWs also has a good reduction performance of total phosphorus (TP), especially B-CW. Approximately 91.6 ± 3.82% of the phosphorus was removed in B-CW. Generally, the B-CWs showed a better reduction of nitrogen and phosphorus than A-CWs. The ion-exchange reactions between the functional groups and contaminants make biochar an effective adsorbent for inorganic nitrogen and phosphorus.[20]

Performance of CWs in the Substrate

Previous studies showed that the nitrogen element contents in AC and biochar were decreased due to the denitrification process in CWs.[21] Both AC and biochar could reduce the water-soluble fraction of pollutants and store the organic matter in the pores of the carbon. As shown in Figure , the Control group had the highest BbFA content in the substrate (1.10 ± 0.04 μg/g), followed by A-CW (0.84 ± 0.02 μg/g) and B-CW (0.64 ± 0.03 μg/g). The reduction in the BbFA content of the substrate treated with AC or biochar can be attributed to the immobilization of the bioavailable fraction of organic contaminants. Meanwhile, both AC and biochar could adsorb and retain contaminants due to their hydrophobic character and micropores.[22] Hence, the addition of AC and biochar in the substrate could reduce the adverse impact of PAHs in CWs; for example, it could reduce the toxicity of PAHs in plant growth and microbial reproduction.

Figure 2

BbFA content in the dry substrate of each group.

Performance of CWs in Plants

Previous studies proposed that wetland plants could accelerate PAH degradation in sediments due to the rhizosphere exudates and radial oxygen loss.[23] As an important removal pathway, the PAH content stored in plants was determined in each microcosm. The uptake of BbFA in the stem and root parts varied in each group. As shown in Figure , the BbFA content in the plant stem in the Control group was highest (3.15 ± 0.78 μg/g), followed by A-CW (1.76 ± 0.44 μg/g) and B-CW (1.58 ± 0.39 μg/g). The BbFA content in plant roots of each group showed a similar order: Control (9.23 ± 1.02 μg/g) > A-CW (7.41 ± 1.4 μg/g) > B-CW (6.16 ± 0.93 μg/g). The results reflected that the uptake of BbFA in the root was higher than that in the stem because the roots were in direct contact with the pollutants in the substrate settled from wastewater. The BbFA uptake by plants in CWs involved several steps. First, the PAHs were absorbed into the substrate and fillers such as AC or biochar, then absorbed by roots, and finally transferred to stems.[24] Further, the higher content of BbFA retained in the roots was also due to the lower weight of the roots. Among the three groups, the total amount of BbFA was highest in the Control group, indicating that the substrate modification by AC and biochar in CWs may reduce BbFA accumulation in plants. There are two possible reasons for lower BbFA contents accumulated in plants with biochar or AC: first, the BbFA settled in the substrate and adsorbed into biochar or AC. Hence, the pollutants retained in the substrate were relatively lower and could be absorbed by wetland plants.[25] Second, the release of carbon source from the biochar or AC enhanced the root exudates of plants, mainly due to the modified surface structures or provided the organic matter for BbFA.[26] The root exudates could change the rhizosphere microbial activity and enhance PAH desorption in the substrate.[27] Meanwhile, the adsorption of carbon fractions on the root surface might prevent plants from absorbing pollutants.[25] The lower BbFA accumulated in plants with biochar and AC could reduce the toxic effects of PAH pollutants on plants.

Figure 3

BbFA content in stem (upper) and roots (bottom) of wetland plants.

Microbial Community

The biological scanning electron microscopy (SEM) images of AC and biochar before and after the experiment showed a rough surface morphology and high internal surface area (Figure ). A large number of microbes were obviously attached in the pores and outside of AC and biochar after the experiment period, presented in all randomly selected images, including rod-shaped, micrococcus, and hyphal microorganisms (Figure c,d). Hence, both the AC and biochar possessed porosity for pollutants’ absorption and provided habitat for microbial growth.[28]

Figure 4

SEM images and biological SEM images before and after the experimental period of AC (a, c) and biochar (b, d).

To realize the effect of AC and biochar for microbes in the substrate, the microbial community from each group were investigated. The overall analysis of high-throughput sequencing results is shown in Table . As demonstrated, lower species were found in A-CW and B-CW groups, indicating that the addition of AC and biochar could reduce the microbial species, which might be due to the re-establishment of dominant microbial groups under the changed substrate. Also, biochar and AC might increase the organic carbon content in the substrate by the release of carbon and polymerization, which was negative to microbial richness.[29] A little lower community abundance was also found in CWs with the addition of AC or biochar according to ACE and Chao 1 values, which showed the following order: Control > B-CW > A-CW; a part of the community was also affected by the addition of carbon.[30] As illustrated by the Shannon values, A-CW had the highest community diversity and correlated with higher metabolic versatility. B-CW had a lower Simpson value and also reflected higher community diversity. As reported in a previous research study, higher community diversity reflected that the microbes could utilize different organic sources.[31]

Table 1

Microbial Diversity Values and Richness of Each Group at a Phylogenetic Distance of 3%

groups	observed species	Shannon	Simpson	Chao 1	ACE	good’s coverage
control	1224	5.753	0.911	1641.541	1644.59	0.988
A-CW	766	3.831	0.729	1051.109	1093.103	0.992
B-CW	1074	6.032	0.928	1385.019	1362.864	0.991

The 16S rRNA gene sequencing analysis reflected the microbial community in each group. As shown in Figure a, Proteobacteria and Actinobacteria were predominant at the phylum level and contributed to 50.7 ± 11.9 and 29.3 ± 6.2% of all detected OTUs in the three groups, respectively. Among them, the phylum of Proteobacteria was most abundant in A-CW (63.6%), followed by B-CW (48.4%). Both of them were higher than those in the Control group (40.2%). The Proteobacteria was highly related to nitrogen or carbon cycling, which was predominant in most wastewater treatment processes, including CWs.[32] Furthermore, Proteobacteria contains many denitrifiers, which contributed to NO3––N removal with the addition of AC and biochar.[33]

Figure 5

Bacterial community composition at the phylum level (a) and subgroups of Proteobacteria at the class level (b), as revealed by high-throughput sequencing analyses.

As an important microbial type, the subgroups of Proteobacteria were analyzed under the class levels. Significant differences in Proteobacteria community composition were observed between three groups, as shown in Figure b. Among the Proteobacteria phylum, γ-proteobacteria and Actinobacteria were identified as the most abundant classes with the relative abundance of 44.2 ± 13.8 and 23.1 ± 3.7% of the three groups, respectively. Obviously, the A-CW and B-CW groups contained more γ-proteobacteria than the Control group, accounting for 59.9 and 38.5% of the microbial community, respectively, which contributed to the enhanced NO3––N removal performance with AC and biochar. The γ-proteobacteria also enhanced the biodegradation of organic matter and is presented as potential PAH degraders.[34] The results for microbial community changed by AC and biochar in CWs were highly toward enhanced nitrogen and carbon removal performance.

Conclusions

The addition of AC and biochar in the substrate can effectively enhance the removal performance of PAHs in surface-flow CW microcosms. At the same time, NO3––N removal performance was enhanced with these two carbonous materials. The BbFA contents absorbed by plants and retained in the substrate were decreased due to the addition of carbonous fillers. The carbonous materials could provide a breeding habitat for microbes and improve the microbial community toward enhanced denitrification and the PAH removal performance.

Materials and Methods

Construction and Operation of CW Systems

To stimulate nature CWs, three surface flow CW microcosms were established under a transparent canopy. The CW microcosms were of PVC material with a length of 60 cm, a width of 40 cm, and a depth of 50 cm. The substrate was washed gravel (particle size 1–2 mm) with 30 cm depth. In experimental groups, equal volumes of biochar and AC with 20 cm depth were added to the substrate. The typical wetland plants acorus calamus were planted in microcosms, with a density of 10 individuals/per unit. The microcosms were divided into three groups as follows: (1) Control group, with no biochar or AC added; (2) A-CW, with AC added as the substrate; and (3) B-CW, with biochar added as the substrate.

The batch-operated procedure was implemented in the experiment, with the hydraulic retention time (HRT) of 3 days. After planting, the synthetic wastewater was added to each microcosm. The pollutant concentrations were according to the Wastewater Discharge Standard (GB 18918-2002), detailed as follows: NO3––N, 24.4 ± 0.48 mg/L; NH4+–N, 14.3 ± 0.38 mg/L; TP, 3.21 ± 0.11 mg/L; and chemical oxygen demand (COD), 79.5 ± 2.86 mg/L. During the experimental period, the depth of overlying wastewater was maintained at 15 cm. The concentration of BbFA was 0.08 ppm in each microcosm. The BbFA was dissolved in acetonitrile before adding it to the synthetic wastewater due to its hydrophobicity.

Sample Collection and Analysis

Wastewater Samples

The overlying wastewater was collected using a 100 mL polyethylene bottle. After the filtration through a 0.45 μm filter membrane, the water quality indexes such as TP, TN, NH4+–N, NO3––N, NO2––N, COD, etc., were determined according to the standard method.[35] The BbFA in the wastewater was extracted using a solid-phase extraction apparatus (Extrapid, Labtech, China). Before extraction, 2-fluorobiphenyl and p-terphenyl-d14 were added into the wastewater as substitutes to determine the recovery rate of pretreatment. Then, 500 mL of the wastewater was passed through a solid-phase extraction membrane (SPMEM, C18 Disks), and then eluted into 20 mL of dichloromethane. The solution was concentrated to 1 mL by the Vacuum concentrator (Vortex 600, China) and the internal standard substance was added. The final concentration was measured by gas chromatography/mass spectrometry (gas chromatography/mass spectrometry (GC/MS), Shimadzu, Japan) as outlined in our previous work.[12,36]

Substrate Samples

The substrate samples were collected for the analysis of the BbFA content and microbial community. To obtain typical samples, five individual samples of equal amounts were collected and homogenized. After drying in a freeze dryer (Unicryo MC 2 L freeze dryer, Germany) for 72 h at −60 °C, the substrate samples were stored at −20 °C for further analysis. The BbFA in 10 g of the substrate was extracted into 20 mL dichloromethane/methanol in a 1:1 volume ratio by the accelerated solvent extractor (Lebtech, China). Then, the BbFA concentration in the extract was determined by GC/MS according to the U.S. Environmental Protection Agency (US EPA) 8270 C.

The biological scanning electron microscopy (SEM) images of the AC and biochar were obtained by Hitachi SU-8010 scanning electron microscope. Before the test, the samples were fixed overnight in a 2.5% glutaraldehyde solution at 4 °C. Then, the solution was discarded, and the samples were rinsed three times in 0.1 M phosphate buffer (pH 7.0) for 15 min. Then, the samples were fixed in 1% osmic acid solution for 1–2 h and they were repeatedly rinsed three times. Several gradients of concentrated ethanol solutions (including 30, 50, 70, 80, 90, and 95%) were used to dehydrate the samples, each concentration for 15 min. Then, they were treated with 100% ethyl alcohol solution twice, each for 20 min. After that, the samples were treated with a mixture of ethanol and isoamyl acetate (volume ratio of 1:1) for 30 min, and then with pure isoamyl acetate for 1 h. Then, the samples were dried at a critical point, and the coating was observed. The samples were examined for at least five particles, and the representative images were selected.

The total DNA of the substrate sample in each microcosm was extracted using the MOBIO PowerSand DNA Isolation Kit (MoBio Laboratories, Inc., Carlsbad, CA) from 0.5 g of the dry substrate following standard methods. Due to the large particle size, the substrate was first mixed with deionized water and ultrasonically treated for 0.5 h. Then, the mixed solution was passed through a 0.22 μm filter membrane. The DNA retained on the membrane was extracted. Before use, the DNA samples were stored at −80 °C. The microbial community was analyzed by the high-throughput sequencing method. The V4-V5 region of 16S rRNA genes was performed on the Miseq (Illumina) platform in the Novogene Co., LTD (Beijing, China). The measurement and data processing methods were similar as reported by a previous.[37]

Plant Samples

The stem and root of plant samples were harvested after the experimental period. After drying at −60 °C for 72 h, the plants were cut into pieces using liquid nitrogen. The BbFA in 10 g of plant samples were extracted into 20 mL dichloromethane/methanol in a volume ratio of 1:1 by accelerated solvent extraction similar to the substrate samples. After purifying through a silica gel column, the concentration of BbFA in organic solution was determined by GC/MS.

Chemicals

The biochar and AC were prepared by wetland plants according to the previous methods.[38,39] The BbFA (>98% purity) was purchased from the Aladdin Reagent (Shanghai, China). The standard substitute for pretreatment extraction and internal standards were obtained from the ANPEL Laboratory Technologies (Shanghai, China) Inc. The solid-phase extraction membrane (SPMEM, C18 Disks) was purchased from Horizon Technology Inc. All of the organic solvents used in the experiment were of chromatographic grade and were purchased from Shanghai Sinopharm Pharmaceutical Co. LTD.

Statistic Analysis

The experimental results were analyzed by SPSS 11.0 (SPSS Inc., Chicago), and analysis of variance (ANOVA) was used for statistical analyses. The results were considered to be statistically significant when p < 0.05.