Over the last 25 years, biodegradable polymers have been developed to replace petrochemical polymers. Because European directives like 94/62/CE seek to reduce the impact of packaging on the environment and because European consumers are increasingly demanding eco-friendly products, more and more resources are invested by industry in the field of biodegradable packaging. But just how biodegradable are these products? Have all of them really been tested or is 'biodegradable' just a marketing argument? There is an urgent need for a reproducible and cost effective test for the biodegradability of plastic material. Therefore, our aim was to develop a device that is able to function with a high degree of autonomy, that limits labour costs and has a high degree of scientific accuracy through auto-calibration and continuous validation. The device can be used both by research centres and testing and certification laboratories. Biodegradability of polymers can be examined by the method described in ISO standard 14855-1 “Determination of the ultimate aerobic biodegradability of plastic materials under controlled composting conditions Method by analysis of evolved carbon dioxide-”. In this procedure pieces of a defined size of the test material (2x2 cm) are mixed with mature municipal compost and incubated at 58°C for up to 6 months. The biodegradability of the material is assessed by comparing the amount of carbon dioxide produced by the mixture to the theoretical maximum evolution of carbon dioxide from the test material, corrected for the evolution from the compost itself and the organic and inorganic matter remaining. Three 3L glass vessels with air tight lids and one inlet and outlet were used per experimental group. The first experimental group, the positive control group, contained 46 g of microfibrous cellulose (Sigma-Aldrich) in 400 g of compost. The second experimental group contained 46 g of poly-lactic acid (PLA), cut in pieces of 2x2 cm, in 400 g of compost. The last experimental group, the blank group, only contained 400 g of compost. At the start of the experiment all mixtures were humidified to the extend that manual squeezing of the compost produced a small amount of fluid. Vessels were aerated from the bottom by a diffuser and were made independent by virtue of the use of a separate pressure regulator for each individual vessel. Aeration flow was set individually for each vessel with a high precision manual valve. During the experiment humidity of the compost was kept at around 50%, and the oxygen concentration was maintained above a minimum of 6% by regulating the flow of aeration. The compost was regularly mixed to ensure maximal homogeneity and minimize the formation of preferential routes for the aeration. For all vessels the aeration flow rate was set at 0.5 L/min initially. The vessels in the positive control group with cellulose started to produce significantly more carbon dioxide than the control group after 4 days. In order for the carbon dioxide concentration to be within range of the sensor, the flow rate was increased in the positive control group to 0.6 L/min after 5 days, but was decreased again to 0.5L/min after 7 days. The vessels in the experiment group with PLA only started to produce significantly more carbon dioxide than the control group after 15, 28 and 32 days respectively. Initially carbon dioxide production in the experimental group was slightly inhibited as compared to the control group which resulted in a difference in cumulative production of carbon dioxide only being evident after 26, 36 and 42 days respectively. Carbon dioxide production in the blank group remained very stable throughout the experiment with a small difference between the vessels. In the blank group 1.75 +/- 0.35 g of CO2 was produced in the first 10 days. In the positive control group 70.3 +/- 1.72 g of CO2 was produced after 108 days (2.5% variability). For the 3 tests, we have a reduction of the quantity of carbon dioxide produced in the beginning of the experiment. It seems that we are in a “wake up” phase due to the fact that the compost was frozen. Positive control: At the beginning of the experiment, we observed an intense activity evidenced by the strong CO2 increase. This phase is facilitated by the fact that the cellulose is in the form of powder. After this phase,a strong, but stable, activity remains before reaching a plateau phase due to the depletion of cellulose. PLA: The PLA was in the form of film of 2x2 cm and not in the form of powder or in crushed form. The advantage of using the PLA in this way is that we were able to observe its degradation visually. The launching phase is slower than in the positive control group, this can be explained by the difficulty of the micro-organisms of degrading the PLA thus presented. According to literature, the first phase of mineralization comes from the degradation of short chains which are immediately available by the micro-organisms. We noted this by the PLA pieces becoming first opaque and the subsequent formation of blisters on the material. The remaining fragments of highly crystalline PLA are much more resistant to degradation. Therefore, after a period of weak mineralization, which is relatively long, we note an increase in CO2 evolution and progressive fragmentation of the material. Blank: The CO2 release comes only from the internal activity of the already matured starting compost without addition of degradable material. A weak and stable release of carbon dioxide is thus observed over the experimental period. NB: Each agitation and re-humidification of the bioreactors excited the activity of the micro-organisms temporarily, which resulted in a spike in CO2 production. The newly developed device was shown to function autonomously over an extended experimental period with limited intervention required (periodic agitation and humidification). Sensor integrities were maintained throughout the period and validation values remained OK. Values measured showed a high degree of accuracy and small variability was observed among the vessels of the positive controls and among the vessels of the blanks. Variability among the vessels of the clearly high biodegradable PLA was shown to be primarily due to the onset of mineralization.