| description abstract | Lubricant ignition is a highly undesirable event in any mechanical system, and surprisingly minimal work has been conducted to investigate the auto-ignition properties of gas turbine lubricants. To this end, using a recently established spray injection scheme in a shock tube, two gas turbine lubricants (Mobil DTE 732 and Lubricant A from Cooper et al. 2021, “Auto-Ignition of Gas Turbine Lubricating Oils in a Shock Tube Using Spray Injection,” ASME J. Eng. Gas Turbines Power, 143(5), p. 051008) were subjected to high-temperature, post-reflected-shock conditions, and OH* chemiluminescence was monitored at the sidewall location of the shock tube to measure ignition delay time (τign). A combination of an extended shock-tube driver and driver-gas tailoring were utilized to observe ignition between 1183 K and 1385 K at near-atmospheric pressures. A clear, two-stage-ignition process was observed for all tests with Mobil DTE 732, and both first and second stage τign are compared. Second stage ignition was found to be more indicative of lubricant ignition and was used to compare τign values with lubricant A. Both lubricants exhibit three ignition regimes: a high-temperature, Arrhenius-like regime (>1275 K); an intermediate, negative-temperature-coefficient-like regime (1230 –1275 K); and a low-temperature ignition regime (<1230 K). Similar τign behavior in the high-temperature regime was seen for both lubricants, and a regression analysis using τign data from both lubricants in this region produced the Arrhenius expression τign(μs) = 4.4 × 10−14 exp(96.7(kcal/mol)/RT). While lubricant A was found to be less reactive in the intermediate-temperature regime, Mobil DTE 732 was less reactive in the low-temperature regime. As the low-temperature regime is more relevant to gas turbine conditions, Mobil DTE 732 is considered more desirable for system implementation. Chemical kinetic modeling was also performed using n-hexadecane models (a lubricant surrogate suggested in the literature). The current models are unable to reproduce the three regimes observed and predict activation energies much lower than those observed in the high-temperature regime, suggesting n-hexadecane is a poor surrogate for lubricant ignition. Additionally, experiments were conducted with Jet-A for temperatures between 1145 and 1419 K around 1 atm. Good agreement is seen with both literature data and model predictions, anchoring the experiment with previously established τign measurement methods and calculations. A linear regression analysis of the Jet-A data produced the Arrhenius expression: τign(μs) = 6.39 × 10−5exp(41.4(kcal/mol)/RT). | |
