Effect of anode material and dispersal limitation on the performance and biofilm community in microbial electrolysis cells

Abstract

In a microbial electrolysis cell (MEC), the oxidization of organic compounds is facilitated by an electrogenic biofilm on the anode surface. The biofilm community composition determines the function of the system. Both deterministic and stochastic factors affect the community, but the relative importance of different factors is poorly understood. Anode material is a deterministic factor as materials with different properties may select for different microorganisms. Ecological drift is a stochastic factor, which is amplified by dispersal limitation between communities. Here, we compared the effects of three anode materials (graphene, carbon cloth, and nickel) with the effect of dispersal limitation on the function and biofilm community assembly. Twelve MECs were operated for 56 days in four hydraulically connected loops and shotgun metagenomic sequencing was used to analyse the microbial community composition on the anode surfaces at the end of the experiment. The anode material was the most important factor affecting the performance of the MECs, explaining 54-80 % of the variance observed in peak current density, total electric charge generation, and start-up lag time, while dispersal limitation explained 10-16 % of the variance. Carbon cloth anodes had the highest current generation and shortest lag time. However, dispersal limitation was the most important factor affecting microbial community structure, explaining 61-98 % of the variance in community diversity, evenness, and the relative abundance of the most abundant taxa, while anode material explained 0-20 % of the variance. The biofilms contained nine Desulfobacterota metagenome-assembled genomes (MAGs), which made up 64-89 % of the communities and were likely responsible for electricity generation in the MECs. Different MAGs dominated in different MECs. Particularly two different genotypes related to Geobacter benzoatilyticus competed for dominance on the anodes and reached relative abundances up to 83 %. The winning genotype was the same in all MECs that were hydraulically connected irrespective of anode material used.

Keywords: Bioanode; Bioelectrochemical system; Graphene; Metagenomics; Microbial community assembly.

Fig. 1

a) Schematic illustration of one hydraulic loop. b) Photograph of the experimental setup.

Fig. 2

a) Current density during the experimental run. G1-G4 are the MECs with graphene anodes, N1–N4 are the MECs with nickel anodes, and C1–C4 are the MECs with carbon cloth anodes. The numbering for each MEC refer to the system in which it is located. b) A bar graph depicting the number of days before each reactor started producing current (at least 1 A/m²). c) The cumulative charge generated in each MEC.

Fig. 3

Cyclic voltammetry measurements of the anodes from the start (a, c, e) and end (b, d, f) of the experiment. Graphene (a, b), nickel (c, d), and carbon cloth (e, f) are shown in different panels while hydraulic loops are indicated by colour of the lines. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 4

Relative abundance (%) of the most abundant MAGs in the anode biofilms. The MAGs are grouped based on phylum: Actinobacteriota (A), Bacteroidota (B), Desulfobacterota (C), Firmicutes (D), Proteobacteria (E), and Spirochaetota (F). Taxonomy is based on GTDB Release 07-RS207. Statistically significant effect on the relative abundances of the MAGs by either hydraulic loop (HL) or anode material (AM) is shown in the right panel (p < 0.05, ANOVA).

Fig. 5

a-c) Diversity calculated as Hill numbers with diversity order 0 (a), 1 (b), and 2 (c); d) evenness; and e-f) PCoA ordination based on dissimilarity matrices calculated with diversity order 0 (e), 1 (f), and 2 (g). The colors in a-d corresponds to the colors in e-g. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 6

Dominance analysis showing the contribution of the explanatory variables hydraulic loop and anode material to the variance of several response variables.