Thriving under toxicity: How partial denitrification adapts to fluoride exposure.

Abstract

Introduction

In recent years, the photovoltaic industry has expanded rapidly to achieve carbon neutrality (e.g., global installed photovoltaic capacity reaching 345.53 GW by 2023, marking a 74 % increase over the capacity recorded in 2022 [1]), yet it has concurrently caused increasingly severe environmental pollution issues. During the manufacturing of crystalline silicon solar panels, nitric acid (HNO₃), and hydrofluoric acid (HF) are commonly used for surface texturing and etching processes, which are subsequently followed by high-purity water rinsing [2]. Additionally, in dry etching processes, the incomplete utilization of ammonia (NH₃) can result in the generation of ammonia-nitrogen-containing wastewater [3]. As a result, this production process generates wastewater containing significant concentrations of fluoride and nitrogen compounds. Excessive fluoride in water has been demonstrated to not only induce dental and skeletal fluorosis in humans, but also lead to genetic mutations and apoptosis in aquatic organisms [4,5]. And the nitrogen in water could affect water quality by depleting dissolved oxygen (DO), thereby posing significant threats to both ecosystems and human health [6].To address these challenges, current treatment strategies typically adopt a sequential approach for treating nitrogen- and fluoride-containing wastewater follows a sequential strategy: “fluoride removal first, followed by nitrogen removal [7]”. Fluoride is typically removed via physicochemical methods, such as chemical precipitation, coagulation-sedimentation, adsorption, and electrocoagulation [1,2]. For nitrogen removal, biological technologies are preferred, particularly the energy-efficient anaerobic ammonium oxidation (Anammox) process, in which autotrophic bacteria oxidize ammonium (NH^₄+-N) using nitrite (NO₂ -N) as an electron acceptor [[8], [9], [10]]. However, since industrial wastewater typically contains both ammonia and nitrate, the Anammox process must be coupled with partial denitrification (PD), which selectively reduces nitrate to nitrite. PD process is achieved by regulating the electron donor supply, COD/

-N ratio, environmental conditions, and the structure of the denitrifying microbial community, in order to control the significant activity discrepancy between upstream nitrate reductase and downstream nitrite reductase [11,12]. Compared to conventional nitrification-denitrification, the PD/Anammox (PD/A) system offers distinct advantages: no DO requirement, 79 % less carbon consumption, low sludge production, 100 % theoretical nitrogen removal efficiency, and a near-zero carbon footprint [6].

In practice, fluoride- and nitrogen-rich wastewaters are often co-treated to lower treatment costs. Depending on the specific production processes and chemical reagents used, the nitrate concentrations in these wastewaters can vary significantly—ranging from as low as 20 mg N/L to over 500 mg N/L [1,3]. The fluoride concentration in such mixed wastewater typically ranges from several hundred to several thousand mg F^–/L and must be reduced to approximately 10 mg F^–/L before entering biological treatment units [1,2]. However, variability in production activities can lead to substantial fluctuations in both the quality and quantity of fluoride-laden wastewater. Elevated fluoride concentrations in the effluent from defluorination units not only increase the complexity of physicochemical treatment processes but also pose a considerable risk to downstream biological systems, particularly by inducing acute microbial stress or inhibition. Therefore, our short-term exposure experiment was designed to simulate fluoride shock events that are commonly encountered in practical operations. Previous studies have shown that fluoride levels above 0.4 g F^–/L can significantly inhibit Anammox activity [7], and anaerobic microorganisms are highly sensitive to fluoride exposure [13]. While the inhibitory effects of fluoride on Anammox are well-documented [7,14], its impact on the partial denitrification (PD) process remains poorly understood. This gap in knowledge is particularly critical given the PD process’s role in maintaining nitrite availability for PD/A systems.

Accordingly, the present research aimed to evaluate both the performance and resilience of the PD process when subjected to different levels of fluoride exposure. Three lab-scale sequencing batch reactors (SBRs), inoculated with PD granules, were exposed to fluoride concentrations of 0.3, 0.6, and 1.2 g F^–/L for durations ranging from 1 to 7 days. Nitrite accumulation, physicochemical responses, and microbial community dynamics of the PD granules were systematically evaluated. The findings of this study aim to elucidate the complex interactions among PD performance, granular physicochemical characteristics, microbial community succession, and fluoride exposure conditions, while revealing the adaptive mechanisms of PD granules under fluoride exposure. The insights gained are expected to provide critical guidance for the optimization of PD-based biological nitrogen removal technologies applied to nitrogen- and fluoride-rich wastewater treatment.