An evaluation of several breakage rates from the literature based on single drop experiments was carried out. This data was collected in a single drop breakage cell under turbulent conditions, comparable to those in a stirred tank. For a constant initial diameter and flow velocity at least 750 single drops have been investigated to measure the breakage time and probability, using high-speed imaging. These results were used for the determination of breakage rates by the product of the inverse of breakage time and the breakage probability. The same subdivision was carried out for the literature models. These differentiations in the analysis showed that published models for the breakage probability are more or less similar and in good agreements with the experimental results. Proposed approaches for the breakage time are contrary. The experiments support the assumption of some researchers that the breakage time rise with increasing drop diameter. The magnitude of the predicted values of the breakage time for all kind of models is one or more magnitudes higher than experimental results in this study and from literature. Furthermore the influence of the physical properties, like viscosity or interfacial tension, is only poorly reflected in the available models. These analysis results lead to an improved breakage time model, which takes into account different breakage mechanisms and the influence of viscosity and interfacial tension. Combined with a breakage probability from literature, this new model leads to an excellent prediction of the breakage rate for the investigated single drops. (C) 2011 Elsevier Ltd. All rights reserved.