Enhanced anaerobic digestion of dairy wastewater in a granular activated carbon amended sequential batch reactor

Abstract

This study investigated the potential of granular activated carbon (GAC) supplementation to enhance anaerobic degradation of dairy wastewater. Two sequential batch reactors (SBRs; 0.8 L working volume), one control and another amended with GAC, were operated at 37°C and 1.5-1.6 m/h upflow velocity for a total of 120 days (four cycles of 30 days each). The methane production at the end of each cycle run increased by about 68%, 503%, 110%, and 125% in the GAC-amended SBR, compared with the Control SBR. Lipid degradation was faster in the presence of GAC. Conversely, the organic compounds, especially lipids, accumulated in the absence of the conductive material. In addition, a reduction in lag phase duration by 46%-100% was observed at all four cycles in the GAC-amended SBR. The peak methane yield rate was at least 2 folds higher with GAC addition in all cycles. RNA-based bacterial analysis revealed enrichment of Synergistes (0.8% to 29.2%) and Geobacter (0.4% to 11.3%) in the GAC-amended SBR. Methanolinea (85.8%) was the dominant archaea in the biofilm grown on GAC, followed by Methanosaeta (11.3%), at RNA level. Overall, this study revealed that GAC supplementation in anaerobic digesters treating dairy wastewater can promote stable and efficient methane production, accelerate lipid degradation and might promote the activity of electroactive microorganisms.

Keywords: conductive materials; dairy wastewater; granular activated carbon; methane production; sequential batch reactor.

FIGURE 1

Schematic diagram (a) and image (b) of the UASB reactors (1 L total and 0.8 L working volume) treating dairy wastewater operated in sequential batch cycle mode for a total of 120 days (four cycles of 30 days each) at 37°C and an upflow velocity at 1.5–1.6 m/h. The granular activated carbon (GAC)‐amended reactor was provided with a polypropylene container with 1 mm diameter holes to contain the GAC material (2 g/L). The control reactor was also provided with a similar polypropylene container without GAC

FIGURE 2

Methane production at all four cycles in the control and granular activated carbon (GAC)‐amended sequential batch reator (SBR). The profile of the control SBR is represented by the black line with solid circle, whereas GAC‐amended SBR is represented by the red line with open circle. Chemical oxygen demand (COD) fraction converted to methane (CH₄) was measured and converted to its COD equivalent following the ideal gas law (PV = nRT) and the conversion factor 1.45 g COD/g CH₄

FIGURE 3

Chemical oxygen demand (COD) mass balance for each 30‐day cycle run for the control (a–d) and granular activated carbon (GAC)‐amended (e–h) reactor. COD fraction converted to methane (CH₄), total lipids, and total volatile fatty acid (VFA) were measured, quantified and converted to their COD equivalent for each sampling point. COD accumulated was calculated from subtracting all measured values (CH₄, total lipids, and total VFA) from the initial COD input

FIGURE 4

Fluorescence emission excitation matrix (FEEM) characterization analyses of extracellular polymeric substances extracted from the initial inoculum (a), sludge from the control sequential batch rector (SBR) after 120 days or end of Cycle 4 (b) and sludge from the granular activated carbon–amended SBR after 120 days or end of cycle 4 (c). Prior to FEEM analyses, the total organic carbon (TOC) content was normalized to 10 mg/L

FIGURE 5

Electron microscopic images. Scanning electron microscopy (SEM) images at different magnification of granular activated carbon (GAC) taken from the GAC‐amended sequential batch reactor (SBR) at the end of 120 days of operation (a–c), as well as transmission electron microscopy (TEM) images of particles present in dairy wastewater (d), suspension of the control SBR mixed liquor (e) and suspension from the GAC‐amended SBR mixed liquor (f). Both suspensions were taken at the end of the reactor run (day 120)

FIGURE 6

Relative abundance of bacteria at genus level in samples collected from dairy wastewater, inoculum, control and granular activated carbon (GAC)‐amended sequential batch reactor (SBR) (at the end of each cycle run), and biofilm grown in the GAC at the end of the experiment. Genus level with relative abundance lower than 2% were included in unclassified groups. Sample designation labels are as follows: Dairy WW—Dairy wastewater; inoculum—Inoculum used as seed sludge in both SBRs at the start of the operation; CON‐C1, CON‐C2, CON‐C3, CON‐C4 are suspensions from the control SBR after Cycles 1, 2, 3, and 4, respectively; GAC(sus)‐C1, GAC(sus)‐C2, GAC(sus)‐C3, GAC(sus)‐C4 are suspensions from the GAC‐amended SBR after Cycles 1, 2, 3, and 4, respectively; and GAC(bio)‐C4—Biofilm in GAC at the end of the final Cycle 4

FIGURE 7

Relative abundance of archaea at genus level in samples collected from dairy wastewater, inoculum, control and granular activated carbon (GAC)‐amended sequential batch reactor (SBR) (at the end of each cycle run), and biofilm grown in the GAC at the end of the operation. Genus level with relative abundance lower than 2% were included in unclassified groups. Description for the sample designations is the same as that of Figure 6

FIGURE 8

Nonmetric multidimensional scaling (NMDS) ordination with (a) DNA‐based bacterial communities, (b) RNA‐based bacterial communities, (c) DNA‐based archaeal communities and (d) RNA‐based archaeal communities. The Bray–Curtis index was performed to generate NMDS to visualize microbiome similarities. The red and blue cluster denote the control and granular activated carbon (sus) microbiome, respectively. A stress value <0.05 is considered an excellent fit; 0.05–0.1 indicates a good fit; >0.2 indicates a poor fit. Description for the sample designations is the same as that of Figure 6