New dry and wet Zn-polyaniline bipolar batteries and prediction of voltage and capacity by ANN

Chemically synthesized polyaniline doped with perchlorate ion was used as the electroactive material of the cathode in the construction of bipolar rechargeable batteries based on carbon doped polyethylene (CDPE) as a conductive substrate of the bipolar electrodes. A significant improvement in the originally poor adherence between the polymer foil and electroactive material layer of the anode was achieved by chemical pretreatment (etching) and single-sided metallization of the polymer foil with copper. A thin layer of optalloy was electroplated onto the surface of the copper-coated polymer foil to increase the battery overvoltage. A mixture of 1 wt% electrochemically synthesized optalloy, 92 wt% electrochemically synthesized zinc powder, 2 wt% MgO, 4 wt% ZnO and 1 wt% sodium carboxymethyl cellulose (CMC) was used as the anode mixture. Then, the electroactive mixture of the anode was coated onto the metallized surface of the CDPE. Graphite powder was used to coat the other side of the CDPE at 90 °C at 1 t cm−2 pressure This side was coated with a cathode mixture containing 80 wt% polyaniline powder, 18 wt% graphite powder and 2 wt% acetylene black. The battery electrolyte contained 1 M Zn(ClO4)2 and 0.5 M NH4ClO4 and 1.0 × 10−4 M Triton X-100 at pH 3.2. Both 3.2 V dry and wet bipolar batteries were constructed using a bipolar electrode and tested successfully during 200 charge–discharge cycles. The battery possessed a high capacitance of 130 mAh g−1 and close to 100% columbic efficiency. The loss of capacity during charge–discharge cycles for the wet bipolar battery was less than that for the dry bipolar battery. Self-discharge of the dry and wet batteries during a storage time of 30 days was about 0.64% and 0.45% per day, respectively. An artificial neural network (ANN) was used to model the voltage and battery available capacity (BAC) only for the dry bipolar battery at different currents and different times of discharge.