Anaerobic digestion (AD) is normally considered to be a stable process in which the metabolic routes are moderated and directed through syntrophic reactions between different groups of microorganisms. These involve interspecies hydrogen transfer, and some of the enzyme systems that control steps within the flow of carbon to methane and other gaseous products are dependent on enzyme systems in which trace elements play an important role. There are situations where this metabolic stability is upset, however, as a result of conditions within the digester which may be strongly influenced by feedstock characteristics. One of the most common occurrences in this respect is due to a low carbon: nitrogen ratio which when the digester is fed at a normal carbon loading leads to an excess nitrogen load in the form of ammonia which has been recognised for at least 45 years as being toxic to the process (McCarty, 1964), although the levels quoted at that time are now recognised as much lower than can be tolerated in an anaerobic system.
There are numerous examples where the AD process has suffered as a result of high ammonia concentrations, for example in cattle manure, pig slurry, chicken litter, abattoir wastes and food wastes. In the case of the latter the effects of high ammonia were clearly seen in technical-scale trials using kitchen waste collected in the UK (Banks et al., 2008), and were almost certainly the cause of digester failure in a 30,000 tonne per year food waste digester operated by Valorsul SA in Lisbon, Portugal (Neiva Correia et al., 2008). The same conditions are observed in small-scale plant as the phenomenon is caused not by the size of digester but by the microbiological and biochemical processes occurring within the system. The problems are fundamentally related to the C:N ratio of the input material and food waste is in this category because of its high protein content which leads to a C:N ratio usually around 10:1. Problems are not normally encountered with food wastes from certain manufacturing processes e.g. vegetable wastes are unlikely to generate high ammonia concentrations although they may cause problems due to deficiencies in trace elements. The current research, however, is directed towards source segregated food waste collected from domestic kitchens. Compositional analysis has shown this to have a C:N ratio in the critical range, and it is known to have caused problems in digesters operating on this feedstock in a number of locations and at all scales. The problems resulting from high ammonia are known to be due to the inhibition of acetoclastic methanogens, a fact recently confirmed for mesophilic systems as well as thermophilic (Schnurer, 2008).
The second problem associated with digestion of certain substrate types is deficiencies in trace elements, the absence of which can result in metabolic blockages in the carbon flow and the irreversible accumulation of acid products causing direct toxicity and pH change. A good example of a full-scale system believed to have failed for this reason is the BTA Toronto digester which was fed on source segregated food waste: the process was eventually stabilised by addition of a cocktail of trace elements the detail of which have not been released (Climenhaga, personal communication). In the case of trace elements the effect is often delayed as digesters are often started off with seed sludges derived from municipal wastewater treatment plants or animal slurries neither of which usually suffer from metal deficiencies. This also explains why co-digestion of foodwaste with these materials may give a better and more stable result.