War. Res. Vol. 23, No. 12, pp. 1599-1602, 1989 0043-1354/89 $3.00+0.00 Printed in Great Britain. All rights reserved Copyright © 1989 Pergamon Press plc RESEARCH NOTE THE ROLE OF FORMATE IN THE ANAEROBIC BAFFLED REACTOR A. GROBICK1 and D. C. STUCKEY Department of Chemical Engineering, Imperial College, London SW7 2AZ, England (First received October 1988; accepted in revised form May 1989) Abstract--The anaerobic baffled reactor (ABR) contains a mixed anaerobic culture segregated into compartments. During pseudo-steady state runs, formate was detected in the first two or three compartments, thereafter dropping off sharply. Under conditions of shock loading, formate was detected in the reactor effluent, up to peak concentrations of 2500 mg/l. There are indications that formate may play an important role as an intermediate in the anaerobic digestion process, and that its production may contribute significantly to reactor stability. Key words--anaerobic digestion, formate, reactor design, reactor stability INTRODUCTION Formic acid has been known for many years to be produced by mixed cultures during anaerobic fermen- tation (Hungate et al., 1967; Marty, 1984). It is also well known that there are many species of methanogens which utilize formate directly to pro- duce methane (Schauer et al., 1982; Daniels et al., 1984). A new model of interspecies electron transfer has now been proposed (Thiele and Zeikus, 1988) which shows formate transfer to be far more signifi- cant than hydrogen transfer, for selected species of micro-organisms. In this paper we present experimen- tal evidence of formate production and utilization, by a mixed culture, in an anaerobic baffled reactor (ABR). This is a novel type of reactor first described by Bachmann et al. (1983). EXPERIMENTAL DATA The experiments reported in this paper were run on two similar anaerobic baffled reactors of l0 I. reactor volume (see Fig. 1), referred to as reactor 3 and reactor 4. Note that the numerous gas offtakes are for research purposes only: the basic design has a single undivided gas space. The reactors were seeded with digester sludge from a sewage treatment works. After the start-up period the micro-organ- isms were observed to have flocculated. The feed composi- tion (for l0 l.) was 26.67 g sucrose, 8 g peptone, 2.67 g Lab Lemco, 0.8 g K2HPO 4 and 25 g NaHCO 3. The total COD concentration was 4 g/1. In addition to the standard analyses (COD, total sus- pended solids) the bulk liquid was centrifuged, deep-frozen and analysed by high pressure liquid chromatography (HPLC). Liquid samples of 20 ml were withdrawn from each compartment at steady state, via sampling tubes of adjustable length set in the reactor cover. Sampling took place from the last compartment back towards the first, maintaining anaerobic conditions throughout. During shock loading, samples were taken from the reactor effluent only. The gas production rate was measured electronically, but the gas composition was not monitored, due to the difficulties of measuring and analysing the gas output from 16 compartments simultaneously. Formate was observed in the anaerobic baffled reactors, both at pseudo-steady state, and during shock loading. At steady state (Figs 2 and 3), there was no formate detectable in the effluent, but in the analyses of bulk liquid samples from the individual compartments it appeared in the first two or three compartments. The concentration of formate was highest in the first compartment (1833 mg/l, reactor 3), dropping off sharply thereafter. Figures 4 and 5 show analyses of the effluent from the two reactors at intervals of 0.5h, during shock load (101/h flowrate) and on returning to the normal loading at 20 h HRT (0.5 l/h). The shock loading took place from 0 to 3 h. In both reactors the formate concentration rises rapidly, peaks at 4 h (2466 mg/1, reactor 3) and disappears after 11 h. The concentration of acetate increases rapidly, remains high up to 7 h, then drops off. However, propionate builds up rather slowly, remains high up to I 1 h, then drops rapidly. Figure 6 shows the total COD and the COD contributed by the VFAs for reactor 3. The total COD contributed by the individual VFAs rises rapidly when the shock load begins, remains high until 13 h, and then drops off to a very low level. DISCUSSION Using the standard method of VFA analysis, an FID gas chromatograph, formic acid is difficult to detect as the peak is very close to the solvent peak. Many workers prepare samples for analysis by acid- ifying with formic acid (e.g. Cohen et al., 1980). Wet chemical methods to determine formic acid in a mixture of volatile fatty acids are lengthy (Jorgensen, 1981). This may explain why formate has not been widely noted in the literature as a significant interme- diate in the anaerobic digestion process. The ABR used in this experimental work is a unique design in that the gas space is partitioned. This may give rise to an atmosphere in a given compartment which is unusual for anaerobic di- 1599