Combinatorial or high-throughput experimentation is improving numerous scientific domains intensely, such as biological chemistry, material science, catalysis. In the past research, most of researchers focus on the novel configuration and microfabrication rather than the transfer mechanism in the microcosm. In this study, we proposed a mass transfer model among droplets in a narrow space. In the derivation, the molecular mass flow rate (J(M)) formula was used to calculate droplet mass transfer rate, which was controlled by the concentration gradient of the surface vapor pressure (P-S) near the droplets. Especially, in the simplification of the expression of the simulating results, all droplets were divided into groups following their radii subtly. In order to assess the accuracy and predictive capability of the proposed model, we carried out some designed experiments to investigate distribution and size changes of the droplets at different temperatures and times. After capturing and analyzing the images of the droplets gathering on a hydrophobic glass slide in an airtight narrow space, we verified the simulating results with the experimental data. The less than 5% deviation indicated the proposed mass transfer model could describe the mass transfer phenomenon among droplets in a narrow space reliably. Furthermore, we found that the total droplet number (N-D) decreases with the running time, which should be taken into account discreetly when designing the parallel micro-reactors. (C) 2015 Elsevier Ltd. All rights reserved.