The maximum allowable temperature difference inside a nickel-hydrogen battery to avoid water relocation was calculated by using a graphical method together with a vapor pressure vs. temperature correlation equation for water vapor over potassium hydroxide solution. An equation was developed for this maximum allowable temperature difference for vessel-wall temperatures from 0 to 30~ and potassium hydroxide concentrations from 20 to 32%. A heat-generation equation for the nickel-hydrogen battery was used to investigate the effect of the location of heat generation on the maximum temperature in the cell and the temperature distribution in the cell.

2781cal response. This is certainly the case for the above system since the Li~NiOx/WQ eleetroehromic cell can be cycled several thousand times with about 50% change in optical transmission. 23 Some concern could still be addressed to the charge-discharge rates since they directly influence the device's response time. However, for most applications, response times of the order of one or few seconds are appropriate and thus even materials with relatively slow lithium diffusion rates are still acceptable, providing they offer the required optical properties.Therefore, the first generation rocking chair batteries, proposed in the early eighties as novel lithium, low-rate reehargeable systems capable of offering long cycle life, 3 have indeed found continued interest. So, the old concept does hold. The question is whether the new concepts for rocking chair batteries, proposed as alternatives to highenergy, high-power lithium and nickel-cadmium batteries, 13't4 will eventually find their way in the competitive portable power source market.

The reversibility of the octanoate adsorption was assessed by stepping the potential from -1 to +1 V vs. SHE (positive sweep), and then back from +1 to -1 V (negative sweep), and measuring friction and inverse capacitance for a 0.1 w/o 'solution of octanoic acid (Fig. 8). Friction and capacitance displayed hysteresis between the sweeps, though both returned to a value near the original at the end of the negative sweep, indicating desorption of the octanoate film. ConclusionStatic friction, current, capacitance, and radiotracer measurements have all shown the potential-dependent nature of lubricant-film formation in aqueous solutions of octanoic acid neutralized with sodium hydroxide. In a solution of 1 w/o octanoic acid, the static coefficient of friction between mild steel and iron was reduced by more than 50% on changing the electrode potential from -400 to 300 mV vs. SHE. This may be due to electrostatic interactions between negatively charged octanoate species and the electrode; at potentials above the pzc of iron (ca. -400 mV vs. SHE), the electrode is positively charged, enabling the adsorption of a lubricating oetanoate film. At potentials below -400 mV vs. SHE, film formation was inhibited, and friction consequently increased.