The biodegradability and bioavailability of hydrolysis-limited substrates under anaerobic (and aerobic) conditions can be represented by two key parameters – degradability (fd), or the percentage that can be effectively be destroyed during digestion, and first order hydrolysis coefficient (khyd), or the speed at which material breaks down. Biochemical methane potential (BMP) testing uses a batch test (in triplicate), and by fitting against a first order model, can fit both parameters in the same test. BMP testing is now being widely used for anaerobic process feasibility and design purposes, and standardisation efforts are ongoing. In this paper, we address a number of key issues relating to the test method and its analysis. This includes proposal of a new fitting and parameter estimation method, evaluation of the impact of inoculum to substrate ratio on fitted parameters, and comparison to performance in continuous systems. The new parameter estimation technique provides an estimate of parameter uncertainty and correlation, and is clearly more suitable than model transformation and linear regression. An inoculum volume ratio of at least 50% (2:1 on VS basis) was required on a cellulose substrate to use methane production as primary indicator, as found by comparing methane production and solubilisation of cellulose. Finally, on a typical material, waste activated sludge, the batch test was slightly conservative in terms of degradability and rate, indicating a bias in the BMP test. The test is a cost-effective and capable method to evaluate potential substrates, but it should be noted that it is generally conservative, especially if sub-optimal inoculum is used.
On the parameter estimation of first order IMA model corrupted with AR (1) errors
