Long term performance and dynamics of microbial biofilm communities performing sulfur-oxidizing autotrophic denitrification in a moving-bed biofilm reactor

Abstract

Sulfide-oxidizing autotrophic denitrification (SOAD) implemented in a moving-bed biofilm reactor (MBBR) is a promising alternative to conventional heterotrophic denitrification in mainstream biological nitrogen removal. The sulfide-oxidation intermediate – elemental sulfur – is crucial for the kinetic and microbial properties of the sulfur-oxidizing bacterial communities, but its role is yet to be studied in depth. Hence, to investigate the performance and microbial communities of the aforementioned new biosystem, we operated for a long term a laboratory-scale (700 d) SOAD MBBR to treat synthetic saline domestic sewage, with an increase of the surface loading rate from 8 to 50 mg N/(m2·h) achieved by shortening the hydraulic retention time from 12 h to 2 h. The specific reaction rates of the reactor were eventually increased up to 0.37 kg N/(m3·d) and 0.73 kg S/(m3·d) for nitrate reduction and sulfide oxidation with no significant sulfur elemental accumulation. Two sulfur-oxidizing bacterial (SOB) clades, Sox-independent SOB (SOBI) and Sox-dependent SOB (SOBII), were responsible for indirect two-step sulfur oxidation (S2−→S0→SO4 2−) and direct one-step sulfur oxidation (S2−→SO4 2−), respectively. The SOBII biomass-specific electron transfer capacity could be around 2.5 times greater than that of SOBI (38 mmol e−/(gSOBII·d) versus 15 mmol e−/(gSOBI·d)), possibly resulting in the selection of SOBII over SOBI under stress conditions (such as a shorter HRT). Further studies on the methods and mechanism of selecting of SOBII over SOBI in biofilm reactors are recommended. Overall, the findings shed light on the design and operation of MBBR-based SOAD processes for mainstream biological denitrification.