Rapier (10) has determined that a mixture of 80% municipal solid waste (MSW) and 20% sewage sludge (SS) provides the optimal combination of energy and nutrients for a mixed culture of acid-forming microorganisms; therefore, this ratio was used in this study.

A series of semi-solid fermentations were operated using custom fermentors. The fermentors were horizontal, stainless-steel cylinders of 17.5-cm length and 10-cm diameter. A center shaft had finger-like projections that extended nearly to the cylinder wall. As the shaft rotated, it "kneaded" the fermentor contents through finger-like projections located on the cylinder wall. The 1.5-L fermentor was filled with about 0.5 L of fermenting MSW and SS.

determined as the weight loss from 2-g biomass samples placed in fine-mesh nylon bags located in the rumen of a fistulated steer.

Treatment conditions: lime loading = 0.1 g Ca(OHyg dry biomass, temperature = 100°C, water loading = 9 g HjO/g dry biomass, time = 1 h (bagasse) or 2 h (African millet straw, sorghum straw, tobacco stalks).

Two or four fermentors were operated in series with solids flowing countercurrently to the liquid. Solid/liquid separation was achieved by centrifuging the fermentor contents and decanting the liquid in an anaerobic hood. This countercurrent operation allows high VFA concentrations to be generated in the fermentor receiving fresh, highly reactive solids. Because inhibition is low, it also allows high conversion in the fermentor receiving fresh liquid.

Table ІП shows the results from four countercurrent fermentations. Compared to rumen fermentations which typically require only a couple of days, the fermentor residence times are significantly longer due to the inhibition from the high VFA concentration (20 — 30 g/L versus 8-10 g/L). Fortunately, because the fermentors can be very inexpensive, the long residence time does not impose a severe economic penalty. The process time scales are similar to those for composting; thus, the process may be viewed as an anaerobic composting operation. The fermentor volume is proportional to the liquid residence time whereas the conversion is proportional to the solids residence time.

Fermentor A used two countercurrent stages and Fermentor В used four countercurrent stages. Similar VFA concentrations, conversions, yields, and selectivities were obtained; however, Fermentor A required 55% longer liquid residence time. This comparison shows the beneficial effect of increasing the number countercurrent stages.

Fermentor C also employed four countercurrent stages. Compared to Fermentor B, the ratio of solid:liquid residence times increased allowing the conversion to increase while holding the product concentration constant.

Fermentor D used essentially the same liquid and solid residence times as Fermentor B, but increased the nutrient content allowing both the VFA concentration and conversion to increase substantially. The conversion of Fermentor D represents 84% of the maximum possible. (Considering ash and lignin content, the maximum digestibility of the MSW/SS mixture is 77.5%.) Further research is required to determine the optimal nutrient package.