Resíduos de gesso da construção civil como meio suporte de wetlands construídos de fluxo vertical descendente empregado no tratamento de efluente sintético

Abstract

The use of construction waste in wetlands (CW) has been investigated as an alternative to conventional materials. However, no studies are reported in the literature related to applying gypsum waste in CW in effluent treatment. Gypsum is composed of calcium, oxygen, sulfur, carbon, silicon, and aluminum, with beneficial properties for soil and plants. Moreover, the Al, Si, Ca, Fe, and Mg composition can assist in phosphorus removal by adsorption, precipitation, and ion exchange processes. In this study, the performance of constructed wetlands (DVCW) of vertical descending flow was evaluated in the treatment of synthetic effluent simulating sanitary sewage of low concentration in terms of COD. The DVCW were vegetated with Eichhornia crassipes (density of 27.23 plants m-²) on gypsum plasterboard fragments (CW-P) and modified gypsum plasterboard fragments (CW-MP) as support medium. The microcosm scale systems had a surface area of 0.1836 m² and a volume of 7 L (C-P) and 8 L (CW-MP). The systems were fed in a sequential batch with 48-48-72 h cycles from October/2021 to August/2022, totaling 308 days. The operation of the DVCW was divided into Phase I, with a cycle time of 24 h, and Phase II, with a cycle time of 48 h. The gypsum board fragments were chemically and texturally characterized by particle size analysis, bulk density, scanning electron microscopy (SEM), X-ray energy dispersive spectroscopy (SED), X-ray fluorescence (XRF), X-ray diffractometry (XRD) and Fourier transform infrared (FTIR). The systems were operated under anoxic conditions during all operating phases, with OD concentration < 1.6 mg L-1 and POR from -100 to +100 mV. Higher average COD removal efficiencies of 60% in CW-P and 70% in CW-MP were obtained in Phase II. The best performance was observed in Phase I (24 h) regarding the removal of KTN (52% and 41%), TAN (41% and 19%), and TN (46% and 37%) in CW-P and CW-MP, respectively. TP removal efficiencies resulted in 64% and 54% in CW-P and 62% and 54% in CW-MP for Phases I (24 h) and II (48 h), respectively. Reduction in organic loading rates in terms of COD, TAN, TN, and TP was observed with increasing operation time. Firmicutes and Proteobacteria were the most abundant species found in the evaluation of the systems’ microbial community structure. The principal genera identified in the systems were Bacillus and Pseudomonas in CW-P and Lactobacillus, Staphylococcus, and Sulfurimonas in CW-MP. No nitrifying genera were observed, only denitrifying bacteria. In the mass balance, it was observed that E. crassipes was responsible for the removal of 0.83% and 10.8% in CW-P and 0.96% and 8.62% in CW-MP of the TN and TP of the total removed by the system. Regarding the substrate, the removal was 0.40% and 1.06% in CW-P and 0.16% and 0.97% in CW-MP of the TN and TP of the total removed by the system. The removal of TN was attributed to microorganisms, assimilation in the biofilm, and adsorption. Adsorption, precipitation, and ion exchange probably acted in the removal of TP.