Compartment modeling of MTBE in the generic environment and estimations of the aquatic MTBE input in Germany using the EQC model

The use of the gasoline additive methyl tert-butyl ether (MTBE) has caused serious concern about groundwater and surface water contamination. The behavior of MTBE in the two most relevant compartments, surface water and air in a generic environment and in a simulated German environment is investigated using the equilibrium criterion (EQC) model. Due to lack of literature data, the half-life time of MTBE in river water is estimated to about 80-120 d (105 d) at 18 degrees C and roughly 1.5 a (year)(533 d) at 4 degrees C from a batch experiment. The EQC model considers four compartments, air, surface water, soil and sediment in an environment of typically 100,000 km2 with about 10% of the area covered with water. The user can progress through the tiered sequence of Level I to III with increasing complexity which reveals more information about the the fate of the considered chemical. The equilibrium mass distribution of MTBE calculated with the Level I model shows that 87% partitions into air and 13% into surface water at 10 degrees C. The results of the Level II calculations indicate that 50% of MTBE in the air is transported from the system and 38% in the air is degraded at 10 degrees C. The resulting total persistence time of 3 d for MTBE in the generic environment of the Level II model can be compared to the calculated value for chlorobenzene. The MTBE input into water is significantly more sensitive to the 'mode of entry' than input into air. The MTBE concentration in surface water is almost exclusively the result of direct emission into water, whereas the atmosphere can additionally be loaded by volatilization from water. The total aquatic MTBE emission in Germany and the average MTBE concentration in German surface waters were roughly estimated to 20-80 t a(-1) (tons per year)(50 t a(-1)) and 50 ng L(-1), respectively. Surface water concentrations calculated with the underlying assumptions of the model can neither be explained by exposure through waste water and industrial effluents nor with an estimated loss of industrially used MTBE in Germany. For the year-round scenario at 10 degrees C, MTBE concentrations of 19 ng L(-1) (surface water) and 167 ng m(-3) (air) result. However, it remains unclear whether the assumptions of the model, the lack of analyses from industrial effluents or both are responsible for the difference. Additional aquatic emission sources could result from gasoline transport on and storage near rivers. The comparison of winter and summer scenarios shows that in summer, atmospheric (25%) and aqueous (50%) concentrations are lower than in winter due to higher degradation rates.