Introduction Water electrolysis is expected a key energy conversion system to introduce large-scale renewable electricity under power grid management. Here, less expensive large system is needed. Alkaline water electrolysis (AWE) systems are well-developed large system to combine hydropower generation, but fluctuated operation with start and stop operation is not considered for conventional systems. Therefore, improvement of durability to combine photovoltaic and/or wind turbine generation is significant issue. In this study, we have been investigated reverse current behavior of bipolar alkaline water electrolyzers and proposes accelerated durability test (ADT) protocol. Theoretical and Experimental Figure 1 shows configuration of bipolar alkaline water electrolyzer, principle of reverse current. An equivalent circuit to analyze reverse current is also shown in the figure. During electrolysis, surface Ni of anode oxidizes to Ni(III) and that of cathode reduces to Ni(0). They electronically connect on a bipolar plate and ionically connect through manifold to distribute electrolyte. In the equivalent circuit, electrochemical potential difference between anode and cathode on a bipolar plate correspond to electromotive force of Uk(t). The R 1 and R 2 correspond ionic resistances of branch and communicating tubes, respectively. Relationship among these parameters can describe as eq. 1 in Fig. 1 based on the Kirchhoff's laws, so reverse current would be estimated. In this study, we measured electrode potential and reverse current after 1 h water electrolysis at 0.6 Acm-2 for 2 to 4 cells stack bipolar electrolyzer using commercially available industrial anode and cathode (De Nora Permelec Ltd). The reverse current was measured ionic current through communicating tube with D. C. clamp meter.As bipolar electrolyzer evaluation, we measured electrode potential and reverse current after 1 h water electrolysis at 0.6 Acm-2 for 2 to 4 cells stack bipolar electrolyzer using the commercially available anode and cathode. Effective electrode size was 27 cm2. The reverse current was measured ionic current through communicating tube with D. C. clamp meter (KEW2510, Kyoritsu).As start & stop operation simulated ADT protocol, we propose the combination of constant current electrolysis, potential sweep, and chronoamperometry as the inset of illustration in Fig. 4, because constant current measurement is easier to get reproductivity in high current that need accurate iR correction for constant potential measurement and current control measurement never simulate stop operation. Results and discussion Figure 2 shows the setup of lab-scale electrolyzer and cell performances in the stacks. All cells showed almost the same performance. Figure 3 shows reverse currents and electrode potentials as a function of time for 2 to 4 cells stack electrolyzer. The longer cell stack had lager reverse current at initial period of stop operation. As the potential difference between anode and cathode decreased, the reverse current decreased. The anode potential significantly decreased certain time, and that for longer cell stack was shorter. For the 4 cells stack, the time for the BP2 was shorter than that for the BP1 and BP3. Simultaneously, the I 2 was larger than the I 1 and I 3. These behaviors could be explained with eq. 1 in the Fig. 1 with the same electromotive force of the U(t=0)s for initial period and time function of U(t)s with same surface condition for all cells at the end of electrolysis. Remarkable point is that the cathode potential of BP2 was higher than the anode, which is impossible with discharge of BP2 itself. This opposite potential of the BP2 would be forced by electromotive force of the BP1 and BP3.Figure 4 shows the anode potential at 10 mAcm-2 (E @10mAcm-2) of oxygen evolution reaction (OER) current during the ADT protocol with various minimum potential: E min. The all anodes significantly degraded for around 3-days test protocol, although DN851 was developed as an active and durable anode for conventional OER electrode. The degradation increased with decreases of the E min. Above 0 V vs. RHE of the E min, The E @10mAcm-2 increased after ca. 800 cycles of introduction period and reached around 1.69 V vs. RHE after 1900 cycles, although it increased from initial period and the increase continued around 2000 cycles for -0.3 V vs. RHE of the E min. From these experimental results, the E min of between 0 to 0.3 V vs. RHE might be suitable for the ADT protocol because of rapid degradation with robustness to the E min. Acknowledgements This study was based on results obtained from the Development of Fundamental Technology for Advancement of Water Electrolysis Hydrogen Production in Advancement of Hydrogen Technologies and Utilization Project commissioned by the New Energy and Industrial Technology Development Organization (NEDO). Figure 1