Sulfur-Based Composite Fillers Enable Adaptive Autotrophic Denitrification for Nitrogen Removal in Photovoltaic Wastewater: From Laboratory to Pilot Scale

Qingguo Zhou, Zhensheng Xu, Feng Shan, Yanchai Zhao, Dongxu Chen, Jian Su, Hao Wu, Lin He, Xialian Shi, Jiaxiang Yang, Mu LiuWater 2026 · DOI

TL;DR

This paper introduces novel sulfur-based composite fillers (S⁰-CFs) incorporating industrial waste (boron mud) for autotrophic denitrification, a process that removes nitrogen from wastewater without needing external carbon sources. These fillers demonstrated high nitrogen removal efficiency in both lab and pilot-scale tests, even under challenging conditions, and significantly reduced operational costs compared to traditional methods.

What Problem Does This Paper Address?

Photovoltaic (PV) wastewater is notoriously low in organic carbon, making conventional biological nitrogen removal processes inefficient and costly due to the need for supplemental carbon sources. Sulfur-based autotrophic denitrification (SAD) offers a promising alternative by utilizing elemental sulfur as an electron donor, but existing SAD systems often suffer from poor stability, low efficiency, and issues like pH drops and sulfate accumulation. This study aims to overcome these limitations by developing robust and cost-effective sulfur-based composite fillers specifically for carbon-deficient industrial wastewater like that from PV plants.

Methods

The researchers developed four types of sulfur-based composite fillers (S⁰-CFs) by blending elemental sulfur with mineral additives, notably including boron mud (an alkaline industrial byproduct), magnesite, and siderite. These were fabricated using a melt mixing-jet granulation process.

Key experimental approaches included:

  • Lab-scale testing: Reactors packed with the four S⁰-CF types were operated under varying hydraulic retention times (HRTs) to evaluate nitrogen removal performance (TN, NO₃⁻-N, NO₂⁻-N, NH₄⁺-N), pH, COD, and TP.
  • Pilot-scale testing: The most promising filler (S⁰-CF 4#) was tested in a 270-day pilot-scale reactor with actual PV wastewater to assess long-term performance, adaptability to temperature fluctuations, and cost-effectiveness.
  • Microbial community analysis: 16S rRNA gene sequencing and FAPROTAX analysis were used to identify the dominant microbial communities and their metabolic functions, confirming the prevalence of autotrophic denitrifiers.
  • Material characterization: Techniques like SEM, XRD, BET surface area analysis, and compressive strength testing were employed to understand the physical and chemical properties of the S⁰-CFs.

Datasets

The study utilized two main types of "datasets":

  • Simulated Wastewater: Used in the lab-scale experiments, this contained specific concentrations of nitrate, alkalinity, iron, ammonium, calcium, magnesium, and phosphate.
  • Actual Photovoltaic (PV) Wastewater: Collected from a PV industrial park, this wastewater was characterized by fluctuating concentrations of COD, NO₃⁻-N, NH₄⁺-N, TP, and F⁻, with a relatively stable pH. This real-world wastewater provided a more challenging and representative testbed for the pilot-scale study.

Key Insights and Ideas

  • Valorizing Industrial Waste: The successful use of boron mud, an alkaline industrial byproduct, as a key component in the S⁰-CFs is a significant contribution. This "waste-to-waste" approach offers a sustainable solution for both wastewater treatment and industrial waste management.
  • Enhanced Mechanical Stability: The S⁰-CF 4#, primarily composed of sulfur and boron mud, exhibited the highest compressive strength. This is crucial for preventing filler disintegration and ensuring long-term stability in packed-bed reactors, a common limitation in previous SAD fillers.
  • Adaptive Denitrification: The S⁰-CFs demonstrated adaptive nitrogen removal, maintaining high efficiencies even with short HRTs (1 hour) in lab tests. The pilot study further showed adaptability to varying temperatures and HRTs, with strategies like extending HRT to 1 hour proving effective during cold periods.
  • Microbial Community Shift: 16S rRNA sequencing revealed a significant enrichment of autotrophic denitrifiers like Thiobacillus and Sulfurimonas, confirming that SAD was the dominant pathway. This highlights the tailored microbial community development facilitated by the S⁰-CFs.
  • Cost-Effectiveness: The S⁰-CF 4# proved to be the most cost-effective filler, and the overall SAD process using this filler was found to be 34.5% cheaper than conventional heterotrophic denitrification due to the elimination of external carbon sources.

Key Results

  • Lab-Scale Performance: At a 1-hour HRT, all S⁰-CFs achieved high total nitrogen (TN) removal (89.1–93.8%) with effluent NO₃⁻-N below 1.5 mg/L (>93% nitrate removal efficiency). S⁰-CF 2# (containing siderite) showed the highest TN removal efficiency.
  • Pilot-Scale Performance: Over 270 days, the S⁰-CF 4# reactor achieved an average nitrate removal efficiency of 88.2% during warm periods (25.8–27.2 °C) at 1-hour HRT. Efficiencies dropped to 13.5–38.1% under cold conditions (7.3–16.0 °C) at 0.5-hour HRT but stabilized at 64.1% when HRT was adjusted to 1 hour.
  • Fluoride Removal: The pilot-scale reactor demonstrated fluoride removal of 0.51–1.49 mg/L.
  • Cost Reduction: The operational cost for electron donor consumption was approximately 34.5% lower than conventional heterotrophic denitrification.
  • Stability: Effluent pH remained stable, and no significant increases in COD or TP were observed, indicating minimal secondary contamination.

Limitations and Future Work

  • Temperature Sensitivity: While the pilot study showed adaptability, the performance still varied significantly with temperature, particularly under cold conditions. Adjusting HRT was an effective strategy, but further research into temperature-resilient microbial communities or filler designs could be beneficial.
  • Phosphorus Removal: While TP removal was observed, effluent concentrations generally exceeded the stringent limit of 0.5 mg/L. The low biomass yield of sulfur-autotrophic denitrifiers limits phosphorus uptake via cell synthesis.
  • Trace Element Monitoring: Although the study suggests minimal risk of boron leaching, the authors acknowledge that continuous monitoring of trace elements in the effluent will be conducted in future engineering applications to fully validate long-term environmental safety.
  • Further Optimization: While S⁰-CF 4# was selected for its mechanical properties and cost, S⁰-CF 2# showed slightly better nitrogen removal in lab tests. Further investigation into optimizing the composition of S⁰-CFs could potentially yield even higher removal efficiencies.

Summary updated

Abstract

Sulfur-based autotrophic denitrification (SAD) is limited by low efficiency and poor stability in carbon-deficient photovoltaic (PV) wastewater treatment. This study developed four sulfur-based composite fillers (S0-CFs) comprising 75% elemental sulfur and mineral additives (boron mud, magnesite, and/or siderite) fabricated via melt mixing–jet granulation. Lab-scale operation showed that at a hydraulic retention time (HRT) of 1 h, all S0-CFs achieved high TN removal (89.1–93.8%) with effluent NO3−-N below 1.5 mg/L (>93% nitrate removal efficiency) and stable pH. Although effluent COD increased with a short HRT (1 h) due to biofilm detachment, no leaching of organic or inorganic pollutants from the fillers was observed, and TP was consistently removed. 16S rRNA sequencing confirmed enrichment of autotrophic denitrifiers Thiobacillus and Sulfurimonas, verifying SAD as the dominant pathway. In a 270-day pilot-scale operation, nitrate removal varied with temperature (7.3–27.2 °C) and HRT, reaching 88.2% on average (range: 86.6–90.0%) at 1 h HRT during warm periods (25.8–27.2 °C), dropping to 13.5–38.1% under cold conditions (7.3–16.0 °C) at 0.5 h HRT, and then stabilizing at 64.1% by adjusting HRT to 1 h. Fluoride was removed at 0.51–1.49 mg/L. Additionally, operational cost was 34.5% lower than heterotrophic denitrification. These results demonstrated that S0-CF enabled efficient, stable, and cost-effective nitrogen removal, making SAD more suitable for low-carbon industrial wastewater treatment.
BibTeX
@inproceedings{zhou2026sulfur,
  author = {Qingguo Zhou and Zhensheng Xu and Feng Shan and Yanchai Zhao and Dongxu Chen and Jian Su and Hao Wu and Lin He and Xialian Shi and Jiaxiang Yang and Mu Liu},
  title = {{Sulfur-Based Composite Fillers Enable Adaptive Autotrophic Denitrification for Nitrogen Removal in Photovoltaic Wastewater: From Laboratory to Pilot Scale}},
  booktitle = {Water},
  year = {2026},
  doi = {10.3390/w18030345},
}
Figure