A novel, wireless, automated system for measuring fermentation gas production kinetics of feeds and its application to feed characterization

This study describes a novel automated method of measuring fermentation gas production kinetics of feeds using sensors that intermittently measure the pressure arising from fermentation of feeds within culture bottles, and relays them to a server using a wireless radio frequency (RF) signal. Measurements can be downloaded from the server at any location that has internet access and saved in ASCII or Excel format. Consequently, once the measurement frequency is set, and the incubation commences, there is no need for attention to the system until the end of the fermentation. In order to validate the system, fermentation parameters of three I mm ground feeds (i.e., corn grain, dehydrated citrus pulp (DCP), Pensacola bahiagrass hay (Paspalum notatum) were determined using the RF sensors and compared to those determined with a digital manometer. A second experiment used both pressure measuring devices to determine fermentation parameters of I rum ground bermuda grass hay (Cynodon dactylon) treated with an esterase enzyme at 0, 1 and 2 g/100 g DM. Feed samples were incubated in buffered rumen fluid in quadruplicate (Experiment 1) or triplicate (Experiment 2) in 250 ml gas tight culture bottles at 39 degrees C. Each culture bottle was connected to a pressure sensor with a three-way Luer lock that was secured into a hole in the culture bottle cap. Pressure sensors made hourly pressure measurements to 96 h. A digital manometer was used to take pressure readings from the culture bottles after 0, 2, 4, 6, 8, 12, 24, 48, 60, 72 and 96 h of incubation, and an exponential model was fitted to fermentation gas production data from the sensors and the digital manometer. Fermentation parameters were compared using a factorial (Experiment 1) or completely randomized design (Experiment 2). In both experiments, method of gas pressure measurement did not affect fermentation parameters, and there was no methodxfeed interaction. In Experiment 1, corn grain and DCP had bigger (P<0.001) gas pool sizes, faster fermentation rates and a longer lag phases than hay. Compared to DCP, corn had a longer (P<0.05) lag phase, similar fermentation rate and a bigger (P<0.01) gas pool size. In Experiment 2, increasing esterase enzyme levels did not affect fermentation rate, but increased (P<0.05) the lag phase and tended (P=0.063) to increase gas pool size. There was a good relationship between RF sensor and manometer-based estimates of gas pool size (r(2) = 92), fermentation rate (r(2) = 83), and lag phase (r(2) = 60). Total measured gas volume correlated to in vitro digestibility (r(2) = 0.60) and to stoichiometrically predicted total gas volumes (r(2) = 0.74). This study demonstrates the potential of this RF sensor technique to differentiate fermentation kinetics among feeds. (c) 2005 Elsevier B.V. All rights reserved.