Research Interest

Machine Learning-Assisted Optimization of Mixed Carbon Source Compositions for High-Performance Denitrification 

Appropriate mixed carbon sources have a great potential to enhance denitrification efficiency and reduce operational costs in municipal wastewater treatment plants (WWTPs). However, traditional methods struggle to efficiently select the optimal mixture due to the variety of compositions. Herein, we developed a machine learning-assisted high-throughput method enabling WWTPs to rapidly identify and optimize mixed carbon sources. Taking a local WWTP as an example, a mixed carbon source denitrification dataset was established via a high-throughput method and employed to train a machine learning model. The composition of carbon sources and the types of inoculated sludge served as input variables. The XGBoost algorithm was employed to predict the total nitrogen removal rate and microbial growth, thereby aiding in the assessment of denitrification potential. The predicted carbon sources exhibited an enhanced denitrification potential over the single carbon sources in both kinetic experiments and long-term reactor operations. Model feature analysis shows that the cumulative effect and interaction among individual carbon sources in a mixture significantly enhance the overall denitrification potential. Metagenomic analysis reveals that the mixed carbon sources increased the diversity and complexity of denitrifying bacterial ecological networks in WWTPs. This work offers an efficient method for WWTPs to optimize mixed carbon source compositions and provides new insights into the mechanism behand enhanced denitrification under the supply of multiple carbon sources.

Yuan Pan, Tian-Wei Hua, Rui-Zhe Sun, Ying-Ying Fu, Zhi-Chao Xiao, Jin Wang, and Han-Qing Yu Environmental Science & Technology 2024 58 (28), 12498-12508 

Carbon source shaped microbial ecology, metabolism and performance in denitrification systems

The limited information on microbial interactions and metabolic patterns in denitrification systems, especially those fed with different carbon sources, has hindered the establishment of ecological linkages between microscale connections and macroscopic reactor performance. In this work, denitrification performance, metabolic patterns, and ecological structure were investigated in parallel well-controlled bioreactors with four representative carbon sources, i.e., methanol, glycerol, acetate, and glucose. After long-term acclimation, significant differences were observed among the four bioreactors in terms of denitrification rates, organic utilization, and heterotrophic bacterial yields. Different carbon sources induced the succession of denitrifying microbiota toward different ecological structures and exhibited distinct metabolic patterns. Methanol-fed reactors showed distinctive microbial carbon utilization pathways and a more intricate microbial interaction network, leading to significant variations in organic utilization and metabolite production compared to other carbon sources. Three keystone taxa belonging to the Verrucomicrobiota phylum, SJA-15 order and the Kineosphaera genus appeared as network hubs in the methanol, glycerol, and acetate-fed systems, playing essential roles in their ecological functions. Several highly connected species were also identified within the glucose-fed system. The close relationship between microbial metabolites, ecological structures, and system performances suggests that this complex network relationship may greatly contribute to the efficient operation of bioreactors. 

Tapping the Potential of Wastewater Treatment with Direct Ammonia Oxidation (Dirammox) 

Microbial-driven nitrogen transformation process are essential for maintaining nutrient balance in global ecosystems and have been extensively employed in the wastewater treatment to remove nitrogen compounds. A recent discovery has unveiled a promising microbial pathway for dinitrogen gas production, referred to as Dirammox (direct ammonia oxidation, NH3→NH2OH→N2), which aligns with the prediction for complete aerobic ammonia oxidation. the discovery of the Dirammox process presents a promising avenue for simplifying wastewater treatment procedures and enhancing treatment effectiveness. In the future, the following research is needed to further strengthen the wastewater treatment efficiency with Dirammox process. 

Microbial mixotrophic denitrification using iron(II) as an assisted electron donor

Mixotrophic denitrification processes have a great potential in nitrogen removal in biological wastewater treatment processes. However, so far, few studies have focused on the mixotrophic denitrification system using Fe(II) as an exclusively assisted electron donors and the underlying mechanisms in such a process remain unclear. Furthermore, the mechanisms by which microorganisms cover carbon, nitrogen, phosphorus and iron in an iron-assisted mixotrophic system remain unrevealed. In this work, we explore the feasibility of using Fe(II) as an assisted electron donor for enhancing simultaneous nitrogen and phosphorus removal via long-term reactor operation and batch tests. The results show that Fe(II) could provide electrons for efficient nitrate reduction and that biological reactions played a predominant role in these systems. In these systems Thermomonas, a strain of nitrate-reduction Fe(II)-oxidation bacterium, was enriched and accounted for a maximum abundance of 60.2%. These findings indicate a great potential of the Fe(II)-assisted mixotrophic denitrification system for practical use as an efficient simultaneous nitrogen and phosphorus removal process.

Re-Evaluation and Modification of Dehydrogenase Activity Tests in Assessing Microbial Activity for Wastewater Treatment Plant Operation

Reliable and cost-effective methods for monitoring microbial activity are critical for process control in wastewater treatment plants. The dehydrogenase activity (DHA) test has been recognized as an efficient measure of biological activity due to its simplicity and broad applicability. Nevertheless, the existing DHA test methods suffer from imperfections and are difficult to implement as routine monitoring techniques. In this work, an accurate and cost-effective modified DHA approach was developed and the procedure for the DHA test was critically evaluated with respect to the standard construction, sample pretreatment, incubation and extraction conditions. The feasibility of the modified DHA test was demonstrated by comparison with the oxygen uptake rate and adenosine triphosphate in a sequencing batch reactor. The sensitivities of the two typical tetrazolium salts to toxicant inhibition by heavy metals and antibiotics were compared, revealing that 2,3,5-triphenyltetrazolium chloride (TTC) exhibited a higher sensitivity. Furthermore, the sensitivity mechanism of the two DHA tests was elucidated through electrochemical experiments, theoretical analysis and molecular simulations. Both tetrazolium salts were found to be effective artificial electron acceptors due to their low redox potentials. Molecular docking simulations revealed that TTC could outperform other tetrazolium salts in accepting electrons and hydrogens from dehydrogenase. Overall, the modified DHA approach presents an accurate and cost-effective way to measure microbial activity, making it a practical tool for wastewater treatment plants. 

A video about the BES system.

Won the second of "The Digital Knowledge Competition" for WaterJam 2019 in Virginia Beach.

Minimizing effects of chloride and calcium towards enhanced nutrient recovery from sidestream centrate in a decoupled electrodialysis driven by solar energy 

To advance nutrient recovery from waste streams by electrodialysis (ED), a decoupled ED system with separated anode/cathode units was developed in this study. The use of an 26 additional cation exchange membrane in the anode unit was designed to prevent chloride oxidation and collect phosphate at a higher concentration. The results show that the decoupled ED removed 92 ± 2% of ammonia and 81 ± 3% of phosphate from a synthetic solution, or 75 ± 4% (ammonia) and 62 ± 2% (phosphate) from a real centrate. Both current generation and nutrient removal could be improved with increasing the applied voltage from 3 to 5 V, and a higher voltage of 6 V did not pose positive effects on removal efficiency but resulted in higher energy consumption. A shorter electrodes distance benefited the decoupled ED operation with a lower internal resistance. The solar energy was successfully applied to power the decoupled ED that exhibited comparable performance (removing 74 ± 4% of ammonia and 60 ± 2% of phosphate) to that by a power supply, although the use of solar energy would depend on the illumination condition. The quality of the recovered struvite was enhanced by either a pre-treatment step to precipitate calcium ions or using a small quantity of the catholyte for struvite formation. These results have demonstrated a promising approach to recover nutrients from sidestream centrate as struvite and ammonium sulfate, encouraging further exploration of ED application towards enhancing flexibility and utilizing clean energy. 

Energy advantage of anode electrode rotation over anolyte recirculation for operating a tubular microbial fuel cell

Mixing plays a key role in both electricity generation and organic removal in microbial fuel cells (MFCs) via affecting substrate distribution and internal resistance. Herein, two mixing methods, anode electrode rotation and anolyte recirculation, were investigated in terms of energy consumption and production. Anode electrode rotation could increase the maximum power density and COD removal by 81.5 and 45.7%, respectively, when the rotating speed increased from 0 to 45 rpm. Likewise, anolyte recirculation also improved the power density and COD removal by 43.1 and 30.1%, respectively, at an increasing rate from 0 to 300 mL min−1. The enhancement of electricity generation became less significant at a high mixing level, likely because that substrate supply was relatively sufficient and other factors posed more effects on electricity generation. The MFC with anode electrode rotation achieved a higher energy balance (e.g., 0.254 kWh kg COD−1 at 35 rpm) than the one without any mixing (0.124 kWh kg COD−1), while anolyte recirculation led to a lower or even negative energy balance compared to that with no mixing. The results of this study have demonstrated energy advantages of anode electrode rotation and encouraged further exploration of energy-efficient mixing methods for MFC operation.  

Removal of azo dye in an up-flow membrane-less bioelectrochemical system integrated with bio-contact oxidation reactor

An up-flow membrane-less bioelectrochemical system integrated with bio-contact oxidation (BES-BCO) was developed for degradation and/or mineralization of the Acid Orange 7 (AO7) in wastewater.  The study found that the BES-BCO system can degrade AO7 and COD effectively. Further, the toxic decolorized byproducts of azo dye, which are not generally degrade in the traditional BES system, can be further mineralized by the BES-BCO system. In addition, the COD removal efficiency increased with the increase of aeration at BCO. However, higher DO concentration at anode can hinder the electrons transfer between anode and exoelectrogen, resulting in a significant decrease of decolorization removal efficiency. This study provided a novel compact technology for advanced treatment of azo dyes wastewater for engineering applications, and concluded that the integrated BES-BCO system is effective for advanced treatment of azo dye-containing wastewater.

Enhanced performance and microbial community analysis of bioelectrochemical system integrated with bio-contact oxidation reactor for treatment of wastewater containing azo dye

 In this study, the effects of hydraulic retention time (HRT), applied voltage, and dissolved oxygen (DO) concentration at the bioanode on the performance of BES-BCO and traditional BES were investigated. Using the response surface methodology, the optimum values of HRT, applied voltage, and DO concentration at the bioanode of BES-BCO were investigated to obtain the maximum decoloration and COD removal efficiency and minimum specific energy consumption (SEC). The microbial community structure in BES-BCO was studied for analyzing the change following the introduction of oxygen. The optimised solution was an applied voltage of 0.59 V, HRT of 12 h, and DO concentration of 0.96 mg/L at the bioanode. Under such conditions, the DE, COD removal efficiency, and SEC values were 94.62 ± 0.63%, 89.12 ± 0. 32%, and 687.57 ± 3.86 J/g, respectively. In addition, after changing from BES to BES-BCO, the bacterial community structure of the bioanode underwent significant changes. Several aerobic aniline-degrading bacteria and anode-respiration bacteria (ARB) were found to dominate the community of the anode biofilm. The results showed that the removal of azo dye degradation by-products was closely correlated with the o-bioanode and the BCO bacterial community structure.