Application of microreactor systems could be the next break-through in the intensification of chemical and biochemical processes. The common flow regime for organic solvent-aqueous phase two-phase systems is a segmented flow. Internal circulations in segments cause high mass transfer and conversion. We analyzed slug flow in seven systems of organic solvents and aqueous phase. To analyze how slug lengths in tested systems depend on linear velocity and physical and chemical properties of used organic solvents, regression models were proposed. It was shown that models based on linearization of approximation by potentials give low correlation for slug length prediction; however, application of an essential nonlinear model of multiple layer perceptron (MLP) neural network gives high correlation with R2=0.9. General sensitivity analysis was applied for the MLP neural network model, which showed that 80% of variance in slug length for the both phases is accounted for the viscosity and density of the organic phases; 10% is accounted by surface tension of the organic phase, while molecular masses and flow rates each account for 5%. For defined geometry of microreactor, mass transfer has been determined by carrying out the neutralization experiment with NaOH where acetic acid diffuses from organic phase (hexane) into aqueous phase. Estimated mass transfer coefficients were in the range kLa=4,652-1,9807 h.1.