How to deal with the long-range electrostatic interactions theoretically and computationally has been well studied due to their importance in biological processes and time consuming summations in computer simulations. The main focus of our research has been on the design and application of a new type of hybrid model that combines both the explicit and implicit solvent models using a reaction field (RF) approach, for accurate and efficient electrostatic calculations. This hybrid model, named as Image Charge Solvation Model (ICSM), replaces an infinite Coulomb summation by two finite sums over direct interactions plus image charges for RF. To characterize the ICSM, the electrostatic torques and forces using different model parameters are compared through various histogram distributions. The contributions of RF are 20% and 2% of the total electrostatic torques and forces, respectively, suggesting that the main effect of RF is to maintain the orientation of water dipoles in the solution. Considering systematic artifacts of the discontinuous dielectric constant at the edge of the cavity, we modified the image charge formula in an optimal way to better account for the continuously changing dielectric profile near the boundary, which provides a computational procedure to determine the most accurate RF possible for a specified water model. The Periodic Boundary Conditions (PBC) in ICSM reduces the size of the productive region and introduces unphysical correlations between ions in ionic solution. With combination of finite boundary conditions, mean field theory for short-range forces and multiple constraint forces applied to water molecules in a buffer layer, bulk water properties are maintained without problems from imaged ions in a much bigger usable region than before. To summarize, the results presented in this work provide a complete characterization, optimization and improvement of the ICSM for electrostatic calculations.