Near infrared and acoustic chemometrics monitoring of volatile fatty acids and dry matter during co-digestion of manure and maize silage

In this study, two process analytical technologies, near infrared spectroscopy and acoustic chemometrics, were investigated as means of monitoring a maize silage spiked biogas process. A reactor recirculation loop which enables sampling concomitant with on-line near infrared characterisation was applied. Near infrared models resulted in multivariate models for total and volatile solids with ratio of standard error of performance to standard deviation (RPD) values of 5 and 5.1. indicating good on-line monitoring prospects. The volatile fatty acid models had slopes between 0.83 and 0.92 (good accuracy) and RPD between 2.8 and 3.6 (acceptable precision). A second experiment employed at-line monitoring with both near infrared spectroscopy and acoustic chemometrics. A larger calibration span was obtained for total solids by spiking. Both process analytical modalities were validated with respect to the total solids prediction. The near infrared model had an RPD equal to 5.7, while the acoustic chemometrics model resulted in a RPD of 2.6. (C) 2008 Elsevier Ltd. All rights reserved.