Performance of bio-nest by modified basalt fiber carriers on the enhanced remediation of low-carbon and -nitrogen urban black-smelling water

Urban industrialization continually increases wastewater discharge, particularly in densely populated and developed cities where high nitrogen content is common in both domestic and industrial effluents. High nitrogen levels in aquatic environments lead to eutrophication and the deterioration of water quality (Zhang et al. 2022; He et al. 2023). Urban rivers in China typically exhibit low carbon-to-nitrogen (C/N) ratios and elevated total nitrogen (TN) concentrations, typically near 10 mg/L. These concentrations predominantly consist of nitrate nitrogen (⁠⁠) (Tu et al. 2019). As urban lifestyles evolve, nitrogen concentrations are suggested to keep increasing and further decreasing the C/N ratio (Tian & Yu 2020). However, optimal denitrification requires a C/N ratio between 6 and 11 and can operate minimally at 3.5–4.5 (Liu et al. 2019). Low C/N-ration conditions pose substantial challenges to denitrification processes (Zheng et al. 2023). Organic carbon deficiency necessitates the addition of external organic carbon sources, such as methanol and sodium acetate, to facilitate heterotrophic denitrification, transforming nitrate into dinitrogen. However, these conventional nitrogen removal methods are costly and inefficient (Ni et al. 2021; Zhai et al. 2022).

Attached growth processes, such as biofilm reactors, are gaining attention due to their sustainability, cost-effectiveness, and eco-friendly nature, offering an easily scalable and self-sufficient solution (Ni et al. 2021; Thuy et al. 2022). Nutrient concentration, influent water quality, and carrier nature are vital factors influencing biomass and the nutrient removal of biofilm reactors (Leyva-Diaz et al. 2020; Ouyang et al. 2023). Biofilm systems, leveraging the benefits of increased microorganism–wastewater contact, surpass suspended growth variants in terms of volumetric loads due to the extensive specific surface area of carriers and enhanced biomass retention (Roche et al. 2017; Gao et al. 2021). Among various carrier materials, modified basalt fiber (MBF) is emerging as a promising eco-friendly inorganic fiber option for wastewater treatments due to its substantial specific area, minimal hydraulic resistance, and excellent permeability (Liu et al. 2018). Notably, continuous MBF has been shown to outperform glass and polymeric fibers in terms of tensile strength and chemical stability, with studies indicating rapid microbial attachment and proliferation, thus affirming its potential as an effective biocarrier material (Ni et al. 2018).

The ability of MBF bioreactors to effectively remove nitrogen, particularly across diverse nitrogen concentrations, offers valuable guidance for bioreactor refinement. For example, the MBF carrier introduced in the reactor captured more versatile microorganisms and achieved a 85% chemical oxygen demand (COD) removal rate at low C/N. A review showed that with the introduction of MBF bio-carriers, the baffled anaerobic–aerobic reactors exhibited TN removal efficiencies comparable to those of the anoxic/anoxic biological aerated filter (BAF) process, reaching 81% (Zhou et al. 2022; Cai et al. 2023). MBF-based bioreactors enhance the microbial community dynamics and functionality, leading to improved nitrification stability and denitrification efficacy (Ni et al. 2018). For example, the reduction of Saccharibacteria and a concomitant increase in Betaproteobacteria from the external to internal regions of the bio-nest were observed, highlighting the inherent aerobic/anaerobic interplay (Gao et al. 2021). The multifaceted construct of MBF bioreactors fosters an enriched bacterial biodiversity (Zhang et al. 2019a). MBF has a water contact angle of 61.64 and is spontaneously hydrophilic. Therefore, bacteria adhered more strongly to the surface of MBF in the initial stage, and microscopic observation showed that MBF was tightly adhered to the surface of the MBF biocarrier by a large number of bacteria (Zhang et al. 2019b). When juxtaposed with conventional biofilms, MBF-based bio-nests manifest pronounced heterogeneity, especially in terms of oxygen gradients and bacterial community spatial distribution. Crucially, the optimized mass transport within bio-nests facilitates superior biomass immobilization, as evidenced in studies on calcium-MBF bio-carriers (Gao et al. 2021).

Despite the promising advances, research into the roles of MBF as bio-carriers, particularly for urban river remediation, remains nascent. It is postulated that MBF might amplify microbial adhesion and biological degradation, thereby elevating remediation efficiency. Consequently, this study aims to examine the impact of carriers on water treatment and microbial enrichment. We meticulously evaluated these carriers’ long-term water processing efficacy of across varying nitrogen concentrations. In addition, employing high-throughput 16S ribosomal ribonucleic acid (rRNA) gene sequencing, we deciphered the bacterial community residing on these carriers. Our endeavor aims to provide deeper insights into the operational nuances and nitrogen elimination capabilities of bioreactors that employ ultrafine fiber carriers.

The sediment was sourced from the riverbed sediment of an urban river in Xiangzhou, Zhuhai, Guangdong, China. Synthetic urban water was used as feed water. The concentrations of components in synthetic urban water were determined based on the previous studies on water quality monitoring results of urban polluted rivers in Guangdong provincial surveys (Gao et al. 2023). Glucose (C₆H₁₂O₆) is used as a carbon source. TN concentration was maintained at 20 mg/L. The C/N ratio was maintained at 5. Particularly, the concentrations of ammonium and nitrate in urban rivers located in Guangdong varied (Xuan et al. 2020). Therefore, two concentration ratios of to were set at 1:1 (MBF-1) and 1:3 (MBF-2) for two MBF reactors, respectively. Table S2 shows the concentrations of components in the two types of synthetic urban water.

Furthermore, an electromagnetic oxygenation pump was employed for aeration and oxygen supply. Both reactors initially had a 72-h hydraulic retention time (HRT) and a 2:1 aeration ratio. To determine and control the HRT, the working volume of the reactor is measured and divided by the system flow rate. By carefully controlling the outlet flow rate, the HRT can be precisely adjusted. The working volume of the reactor is 70 L, and if the HRT is set to 72 h, the flow rate of the control system is 0.98 L/h. During the reactor operation, HRTs and aeration duration ratios were adjusted based on the performance of pollutant removal. Aeration control is performed in units of 24 h a day; the aeration ratio indicates the ratio of aeration to unaerated time, and the aeration time indicates the total aeration time of the day. For example, if the aeration ratio is 2:1, the aeration time would be two-thirds of 24 h, equating to 16 h/day. Detailed operation parameters are shown in Table 1. After changing HRT or aeration duration, each stage had a 10-day stabilization period. The 10-day experimental period was chosen based on empirical evidence, literature support, preliminary findings, and practical considerations (Duan & Kravaris 2017; Lu et al. 2020). It allows for a robust assessment of throughput stability while acknowledging the complexities of microbial community dynamics. The reactors consistently operated with a dissolved oxygen (DO) of 2–3 mg/L, a pH of 6.9–7.5, and a room temperature of 20–25 °C. Effluent water samples were collected and measured daily.

The concentration of COD was measured using the standard digestion spectrophotometric methods. concentration was measured employing Nessler's reagent spectrophotometry. Ultraviolet spectrophotometry combined with an alkaline potassium persulfate digestion was used to determine the TN levels. The total phosphorus (TP) concentration was assessed through the digestion-ascorbic acid technique. Detailed experimental methods could be referred to the ‘Water Analysis Handbook’ by Hach (Hach 2016).

To systematically investigate and compare alterations in the surface morphology and roughness of MBF, MBF specimens were affixed directly onto a conductive adhesive substrate. Subsequently, a 45-s gold sputtering treatment at 10 mA was administered using the Quorum SC7620 sputtering coater. The surface characteristics of these fibers were subsequently captured through imaging utilizing a TESCAN MIRA LMS scanning electron microscope (SEM), manufactured by Oxford Instruments in the United Kingdom. Following the acquisition of surface morphology images for MBF, an in-depth analysis of the elemental composition was conducted employing energy spectrum line scanning techniques. This analysis was executed utilizing the TESCAN MIRA LMS SEM, which was equipped with an X-ray energy dispersive spectrometer (SEM-EDS).

16S rRNA high-throughput sequencing was applied to determine the microbial diversity and community structure. Microbial samples adhered to the MBF fillers were collected at the end of each stage (Table 1). DNA was extracted using the ALFA-SEQ Advanced Soil DNA Kit (mCHIP BioTech CO., LTD, China). The integrity, purity, and concentration of the extracted DNA were verified using 1% agarose gel electrophoresis and the Thermo NanoDrop One spectrophotometer. The V3–V4 variable regions of the bacterial 16S rRNA gene were polymerase chain reaction (PCR)-amplified with primers 338F (5′-ACTCCTACGGGGAGGCAGCA-3′) and 806F (5′-GGACTACHVGGGGTWTCTAAT-3′). The PCR comprised 25 μL of Premix Taq (2×), 1 μL each of both the forward and reverse primers (10 μM), 50 ng of DNA, and nuclease-free water added to make up a total volume of 50 μL (Chen et al. 2018). The PCR protocol entailed an initial denaturation at 94 °C for 5 min, followed by 30 cycles of denaturation at 94 °C for 30 s, annealing at 52 °C for 30 s, and extension at 72 °C for 30 s, culminating in a final extension at 72 °C for 10 min. The reaction was then stored at 4 °C. Post-amplification, PCR products were pooled and purified utilizing the E.Z.N.A.^® Gel Extraction Kit, and a sequencing library was constructed as per the NEBNext^® Ultra™ DNA Library Prep Kit for Illumina^® protocol (Amir et al. 2017). Sequencing was executed on the Illumina Nova 6000 platform. Operational taxonomic units (OTUs) with over 97% similarity were clustered using the UPARSE software (Edgar 2013). Redundancy analysis (RDA) was performed using the Microeco package by R 4.2.2 and RStudio to characterize the microbial community composition of the two groups as affected by water quality.

Figure 2(b) illustrates the COD removal efficiency performance of MBF-2. A decrease in both HRT and aeration duration resulted in a slight reduction in the COD removal rate. However, this decline was not pronounced, and the system consistently maintained a high and relatively stable COD removal rate. Specifically, at an HRT of 72 h with aeration at a 2:1 ratio, the MBF system demonstrated the highest average COD removal rate at 91.58%. Shortening the HRT led to a reduction in the average COD removal rate to 88.99%. Furthermore, when the aeration duration was shortened to 8 h, the average COD removal rate reached 87.18%. The packing within the system facilitated biomass enrichment, resulting in higher biomass retention within the reactor. Consequently, under all three conditions, the average COD removal rate exceeded 85%, indicating the efficacy of MBF-2 in organic matter removal.

As illustrated by the data in Figure 2(c), a conspicuous elevation in effluent concentrations was observed in the MBF-1 reactor when both HRT and aeration duration were reduced. Specifically, under an HRT of 72 h, the recorded average effluent concentrations were 0.8 and 1.92 mg/L for aeration-to-discontinued aeration ratios of 2:1 and 1:2, respectively. These patterns suggest that prolonged aerobic conditions stimulate the growth and metabolism of nitrifying bacteria, thereby enhancing nitrification efficiencies (Mariane de Morais et al. 2020; Nemeth et al. 2023). Under optimal conditions, removal rates surged beyond 82%. However, a reduction in the HRT to 36 h triggered an increase in organic loading, promoting the proliferation of heterotrophic bacteria. This, coupled with diminished contact time with pollutants due to the shortened HRT, disadvantaged autotrophic nitrifying bacteria. Consequently, there was a notable increase in effluent concentrations, leading to dwindling removal rates, which reached 54.21 and 48.19% for HRTs of 36 and 24 h, respectively.

Figure 2(d) underscores the detrimental impact of the reduced HRT and aeration duration on nitrification efficacy in MBF-2. Under an HRT of 72 h and an aeration ratio of 2:1, MBF-2 demonstrated commendable removal rates, peaking at 97.93%, with effluent concentrations oscillating between 0.14 and 0.48 mg/L. However, a decrease in the HRT to 36 h significantly diminished the interaction time between the influent substrate and microbes. This, coupled with an augmented influent flow, potentially compromised biofilm integrity and depleted nitrifying bacteria populations, resulting in a decline in the removal rate to 83.13%. Further reducing the aeration duration to 8 h led to a progressive drop in removal rates, reaching 68.70%. The diminishing trend in the removal rate became evident with the shortened HRT and the aeration time.

Comparatively, while both reactors exhibited impressive removal rates under optimal conditions, MBF-2 displayed better removal rates. The performance of MBF-1 showed a more pronounced decline with reduced HRT and aeration duration compared to MBF-2. This suggests that MBF-2 may have a relatively more stable performance under varied operational conditions. The differing ratios between MBF-1 and MBF-2 could be contributing to these observed variations in performance.

In MBF-2, reducing the HRT from 72 to 36 h enhances the mean TN removal rate from 34.96 to 54.23% (Figure 3(b)). The diminished HRT intensifies the organic burden within the reactor, fostering the proliferation of heterotrophic denitrifying bacteria and thereby optimizing nitrogen removal (Niu et al. 2023). Maintaining the HRT at 36 h and adjusting the aeration-to-discontinued aeration time ratio from 2:1 to 1:2 induce a nearly 4% increase in the mean TN removal rate, accompanied by minimal deviations in the effluent concentration. Comparatively, the TN removal rate of MBF-2 was relatively lower than that of MBF-1, in contrast to removal, where MBF-2 exhibited superior removal performance. This result might be because the denitrification activity of MBF-2 was lower than that of MBF-1, while MBF-2 exhibited higher nitrification activity.

The removal rate of TP in MBF-1 exhibited an initial increase followed by a decrease with the extension of the HRT (Figure 3(c)). The best TP removal rate was observed when the HRT was 72 h and the aeration ratio was 2:1 (day 1–day 10), with the TP removal rate reaching almost 80% on day 10. Shifting the aeration rate from 2:1 to 1:2 led to a decrease in the TP removal rate from 73.52 to 59.44%. This significant decrease suggests that the TP removal activity benefited from extended aerobic stages. On the one hand, lower DO concentrations can compromise the activity of phosphorus-aggregating bacteria (Kim et al. 2003; Gong et al. 2023). On the other hand, longer aerobic periods might bolster the ability of polyphosphorus bacteria to uptake and store phosphates. With a consistent aeration ratio, shortening the HRT from 72 h to 36 and 24 h resulted in a continuous decrease in the TP removal rate from 68.52 to 42.25%. This could be attributed to the short HRT hindering the interaction between microorganisms and substrates, as mentioned earlier. The short HRT provided a limited duration of time for polyphosphorus bacteria to absorb phosphate (Chen et al. 2022).

Figure 3(d) shows the TP removal variations in MBF-2. The manipulation of HRT and aeration duration did not significantly impact the average TP removal rate. However, the stability of phosphorus removal was profoundly influenced by variations in aeration duration. Under the conditions of a 72-h HRT and a 2:1 aeration ratio, the average TP removal rate in MBF-2 was recorded as 63.02%. Maintaining the HRT and reducing the aeration ratio to 1:2 resulted in a decline in the TP removal rate to 59.99%, accompanied by greater fluctuations in the removal curve. One plausible explanation is that the 2:1 aeration ratio was adequate for polyphosphorus bacteria to conclude the anaerobic phosphorus release process. Further reductions in aeration time impeded the aerobic phosphorus absorption process, thereby diminishing the stability of phosphorus removal.

The hydrophilicity or hydrophobicity of a material significantly influences the propensity of microorganisms to adhere to it. A critical parameter in determining the hydrophilicity or hydrophobicity of biofilm carriers is the water contact angle (Al-Amshawee et al. 2021). It is noted that microorganisms tend to selectively colonize surfaces of carriers exhibiting hydrophilicity, thereby enhancing the likelihood of biofilm formation. According to measurements obtained with a contact angle meter, the contact angle of MBF was determined to be 61.5°. Materials with a contact angle exceeding 90° are classified as hydrophobic, while those with angles below 90° are considered hydrophilic (Ni et al. 2022). The magnitude of the angle is inversely proportional to the degree of hydrophilicity. Therefore, the MBF utilized in this study possessed pronounced hydrophilicity, implying a propensity for dispersion in water. This characteristic suggests that MBF was conducive to microbial adhesion and subsequent biofilm development.

The microscopic morphology and elemental composition of MBF are meticulously illustrated in Figure S1 and Table S3. The results of SEM-EDS line spectra revealed that MBF primarily consists of inorganic materials, encompassing elements such as Si, Ti, Al, Na, Ca, K, Mg, O, and Fe, with Si and O being the predominant constituents. The diameter of MBF filaments was approximately 12 μm. Notably, there was a conspicuous increase in surface roughness, which was characterized by the presence of pits and irregular protrusions. Adhering to the fiber filament surfaces are fine granular substances. These morphological attributes significantly amplify the likelihood of interactions between microorganisms and the fiber surface, thereby facilitating microbial adhesion (Kalia et al. 2013). Furthermore, the amplified specific surface area of the material enhances the potential for microbial attachment.

Upon concluding the initial phase for both reactors, fiber samples were extracted for a detailed examination of the biophase using an optical microscope. The graphical representation emphasizes a consistent distribution of yellow-brown activated sludge over the basalt fiber filaments. The manifested biofilm exhibits a dense architecture, with the sludge flocs remaining undisturbed – displaying neither disintegration nor expansion. With the gradual operation of the reactors, an array of microorganisms, including rotifers, bellworms, and schizothoracines, began to flourish, suggesting a diverse and healthy ecosystem within the reactors.

Figure 5(b) shows the dominant microbial composition at the genus level for each sample. The predominant genera observed across the groups encompassed Pseudomonas, Aeromonas, SM1A02, SC-I-18, and so on. Key denitrifying bacteria genera identified from the samples included SM1A02, Dechloromonas, Flavobacterium, Hyphomicrobium, Hydrogenophaga, Rhodobacter, Pseudomonas, and Aeromonas. In addition, nitrifying bacteria, Nitrospira, was also enriched in the reactors. Rhodobacter, notably, functions as an aerobic denitrifying bacterium, facilitating the simultaneous nitrification–denitrification process. Hydrogenophaga, an autotrophic bacterium, actively participates in denitrification. SM1A02, frequently encountered in water treatment, exhibits anaerobic ammonia oxidation functionality, enabling the conversion of to N₂, thus facilitating denitrification (Hoshino et al. 2005). Aeromonas, known for its resistance to both acid and alkali and remarkable temperature adaptability, demonstrates the capability to degrade polycyclic aromatic hydrocarbons. Recent research reported the potential denitrification characteristics of Aeromonas, contributing to nitrate nitrogen reduction for TN removal (Tan et al. 2021). Samples TF.1.2 and A3, which showcased optimal nitrogen removal, also exhibited the highest proportions of denitrifying bacteria genera at 52.98 and 47.01%, respectively. This implies that the MBF system performs optimally in nitrogen removal under these specific conditions.

RDA1 accounted for 72.8% of the dataset, while RDA2 explained 21.9%; together, they comprehensively covered 94.7% of the data. This comprehensive explanation contributed to a clear demarcation between the two groups. The microbial composition of MBF-1 exhibited a significant correlation with NH₄⁺–N and TN. While Pseudomonas, Rhizobacter, and Allorhizobium in the MBF-2 microbial community were positively correlated with and TN, Flavobacterium and Hydrogenophaga were positively correlated with COD.

This study underscores the efficacy of MBF carriers in enhancing the remediation of low carbon and nitrogen urban black-smelling water through biological contact oxidation reactors. Experimental findings demonstrated that an HRT of 36 h and an aeration ratio of 1:2 are optimal for removing COD, ⁠, TN, and TP. MBF-2 reactors maintained both high COD (>85%) and removal (97.93%) rates under optimal conditions. TN removal, driven by microbial denitrification, was the most efficient in the MBF-1 reactor. The hydrophilicity and rough surface morphology of MBF significantly facilitated microbial adhesion and biofilm formation. High-throughput 16S rRNA sequencing revealed diverse microbial communities, with the highest diversity under longer HRT conditions. MBF reactors predominantly hosted Proteobacteria, with a high abundance of denitrifying genera such as Pseudomonas, Aeromonas, and SM1A02, underscoring their robust denitrification capacity. Overall, MBF bioreactors present a promising solution for urban river water treatment.

This work was supported by the Open Research Fund of the Key Laboratory of Water Security Guarantee in Guangdong-Hong Kong-Marco Greater Bay Area of Ministry of Water Resources (Grant No. WSGBA-KJ202303), Foshan Shunde District Core Technology Breakthrough Project (2230218004273), 2022 Zhuhai Social Development Science and Technology Program Project (2220004000355), and Guangdong Basic and Applied Basic Research Foundation (2023B1515040028).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.