Online monitoring of concentration and dynamics of volatile fatty acids in anaerobic digestion processes with mid-infrared spectroscopy.

An ATR-MIR-FTIR spectrometer was integrated into a laboratory scale anaerobic digestion setup. Automatically, a sludge sample from the digester was transferred to a measurement cell; an IR spectrum was recorded and evaluated by chemometric models to estimate the concentration of the individual volatile fatty acids (VFA). The calibration set included semi-artificial samples spiked with known concentrations of the VFA as well as original samples from a continuous fermentation. High-performance liquid chromatography (HPLC) was used as a reference analysis of the samples. The models were optimized for a low root mean square error of prediction (RMSEP). R(2) for acetic acid, propionic acid, isobutyric acid, butyric acid, valeric acid, and isovaleric acid were 0.94, 0.88, 0.83, 0.75, 0.59, and 0.90, respectively. The accuracy of the models was validated in a second experiment. Considering the complex and heterogeneous sludge composition and the chemical similarity of VFA, absolute concentration and dynamic (increasing and decreasing concentration of VFA) was predicted well for acetic, propionic, isobutyric, and isovaleric acid (in their respective concentration range); Butyric acid could not be detected. The installed setup was able to gather and measure native samples from the digester (every 2 h) automatically over a period of 6 months without problems of clogging or biofouling. The instant and continuous analysis of the concentration of the VFA made it possible to evaluate the current bioprocess status and adjust the organic loading rate accordingly.