Neuromorphic electronics have been developed and innovated in recent years due to their critical contributions to the future intelligent computing systems. The human brain has outstanding information processing functions with neural networks which can operate at a low energy consumption. Potentiation and depression are brain activities that refer to the process by which synaptic connections between neurons become stronger or weaker with frequent and periodic stimulating pulses. This concept is correlated with the learning and memory functions and can be simulated with various electronics nowadays. Among them, organic electro-double-layer transistors provide an opportunity since they are biologically compatible, well tunable and low power consumption for switching. 2-D monolayer Molybdenum disulfide ($MoS_2$) have recently been applied on the organic FET due to its high effective mass, large direct band gap around 1.8 eV, ultra-low power dissipation and a high on/off ratio. It is an ideal material designed for the semiconductor part of the device. Although there are several research of $MoS_2$ on the transistors currently, the exploration of $MoS_2$ application on the transistors is still in an incipient stage. To accomplish potentiation and depression activities, hysteresis is the crucial characteristic we are seeking in the devices. 2-D polymer materials have the proper microstructure for membrane applications which is capable to cage the ions inside the electrolyte. The porous structure can also fulfill membrane ion selectivity. 2-D polymers have not been applied wildly on the transistors yet, but we think it is intriguing and provides an unprecedented opportunity to modify the neuromorphic memory transistors. We study the potential of combining $MoS_2$ and crown ether, a 2-D polymer thin film, to construct a bilayer memory transistor. Through a comprehensive investigation of the synthesize process, characterization and device integration of 2D polymers, our work tends to unravel some principal mechanisms including dielectric constant of solvents, concentration-dependent dynamics and ion-selectivity of membranes. This research aims to facilitate the rational design of a bilayer structure memory transistor for neuromorphic computing.