Spark Assisted Compression Ignition Engine with Stratified Charge Combustion and Ozone Addition

Performance and emissions characteristics for stratified charge spark assisted compression ignition (SACI) with 30 ppm of added ozone (O3) were explored in a single-cylinder, optically accessible, spray-guided, research engine. For the present study, intake pressure and temperature were fixed at 1.0 bar and 42°C respectively, with a range of engine loads (1.5 – 5.5 bar indicated mean effective pressure) and speeds (800 – 1600 revolutions per minute) explored. Fuel stratification – achieved by a late-cycle injection of ~ 10-25% of the total fuel – was used to maintain stable operation at lower engine loads. For each condition spark timing, second injection SOI, and fuel split ratio between the main and second injection were optimized to maximize engine performance while maintaining nitrogen oxide emissions (NOx) below 5 g/kg-fuel. Ozone addition was found to decrease specific fuel consumption by up to 9%, with across the board improvement in combustion stability relative to similar conditions without O3. The effect of O3 addition was most substantial for the lowest loads. Moreover, because a higher fraction of the fuel burned was due to end-gas auto-ignition, specific NOx emissions likewise decreased by up to 30%. From complementary measurements of in-cylinder O3 decomposition acquired via an ultraviolet light absorption diagnostic, it was observed that rapid decomposition of O3 into molecular and atomic oxygen coincided with the onset of end-gas auto-ignition. The burst of resultant atomic oxygen was thought to accelerate low-temperature heat release (LTHR) reactions in the end gas. Optimal end-gas auto-ignition started between 20 and 30 crank angles before top dead center with temperatures at LTHR onset estimated to be between 575 and 700 K. An included analysis indicates that the spark deflagration was needed to add between 10 and 40 J of additional thermal energy to the end gas to achieve optimal auto-ignition. INTRODUCTION Current automotive gasoline engine research trends are focused on combustion modes that use some amount of bulk-gas auto-ignition. Until recently, most investigations were into advanced compression ignition (ACI) concepts that exclusively feature auto-ignition such as gasoline compression ignition (GCI) [1-3] or homogeneous charge compression ignition (HCCI) [4, 5]. Relative to conventional spark ignited (SI) combustion, these low-temperature combustion (LTC) strategies achieve substantially improved fuel economy and much lower engine-out emissions of nitrogen oxides (NOx) and particulate matter (PM) through a combination of reduced throttling losses, lower heat transfer, higher compression ratios, and increased charge specific heat ratios. The central challenge for ACI implementation has been to maintain stable and knock-free combustion across the load-speed map. At lower loads where combustion stability is problematic, charge heating via retained residuals along with mixture preparation from late-cycle direct injections and internal flows are often used to create reactive, heterogeneous mixtures at top dead center (TDC) [1, 6-8]. However, the improved stability comes at the cost of increased heat transfer losses [9] and more complex valve train requirements. For elevated loads, engine knock can be mitigated with substantial dilution by cooled exhaust gas recirculation (EGR) to slow heat release rates [10, 11], but requires extensive intake boost to meet power density requirements. As a result, expansion efficiency is reduced, and mechanical losses are increased due to the higher peak cylinder pressure requirements. An alternative to ACI engine combustion is the so-called mixed-mode combustion strategy, where some combination of spark-assisted compression ignition (SACI) and pure ACI are used at part-load operation with exclusive SI combustion used for high power-density conditions [12, 13]. For the SACI strategy, end-gas auto-ignition is induced by compression heating from an expanding spark initiated flame kernel [14-23]. Compression ratios JSAE 20199089 SAE 2019-01-2253 must generally be held to below 14 to limit high-load knock, which is much lower than the 16+ compression ratios commonly used for conventional ACI. As a result, some combination of mixture stratification and charge heating at intake valve closure (IVC) is once again required to ensure that gasoline mixtures near TDC are sufficiently reactive. Another potential method to improve charge reactivity is to seed the intake charge with small amounts (~ 10 − 70 ppm) of ozone (O3) – a powerful oxidizing chemical agent. Ozone addition enhances the reactivity of gasoline-like fuels, which thereby enables stable auto-ignition with less initial charge heating [2, 24-29]. Because charge reactivity is proportional to the initial O3 concentration and targeted O3 concentrations can be generated by onboard coronal discharges of intake air [30], the engine transient response can be very high. The authors previously demonstrated that 50 ppm of intake added O3 for homogeneous charge SACI operation led to excellent combustion performance for moderate loads and NOx emissions below 1 g/kg-fuel [29]. However, poor combustion stability prevented operation at loads below ~4 bar indicated mean effective pressure (IMEP). Ozone addition works by effectively acting as an atomic oxygen (O) carrier that is rapidly released near TDC when temperatures reach around 600 K. The newly formed O then initiates heat release through fuel hydrogen (H) abstraction to form fuel (R) and hydroxyl (OH) radicals [28, 31]. The OH likewise abstracts additional H to make additional R. For temperatures below ~800 K, chain branched low-temperature heat release (LTHR) pathways dominate. For these reactions, R combines with molecular oxygen (O2) to form peroxy radicals (RO2). The RO2 continues fuel H abstraction through the reaction: RO2 + RH → ROOH + R, with the formed fuel radical peroxide (ROOH) decomposing into oxy radical (RO) and OH [32]. Depending on initial intake O3 concentrations, these early LTHR reactions can advance combustion phasing by more than 20 crank angles (CA). For the present study, the effect of intake seeded O3 (30±4 ppm) was investigated as a way to replace charge pre-heating for stable lean SACI operation with an early and late direct injection (DI). Experiments were performed in a spray-guided single-cylinder research engine with optical access. A naturally aspirated intake pressure and 42°C intake temperature were maintained for all conditions, with internal residual fractions of between 10 and 18% achieved through a combination of positive valve overlap (PVO) and moderate exhaust backpressures. Low to moderate engine loads of between 1.5 and 5.5 bar indicated mean effective pressure (IMEP) and speeds of between 800 – 1600 revolutions per minute (rpm) were examined. Each load/speed operating condition was optimized to maximize engine performance while maintaining nitrogen oxide emissions (NOx) below 5 g/kg-fuel. For selected optimized operating points, the effect of four parameters were studied in detail: a) intake O3 concentration, b) spark timing (ST), c) timing of the second injection with respect to ST, and d) the fuel split ratio between the main and second injections. Performance and engine-out emissions measurements were complemented by CA resolved O3 measurements performed via ultraviolet (UV) light absorption. Finally, estimates of end gas auto-ignition onset and corresponding temperature are used to evaluate the amount of thermal energy required to initiate auto-ignition with O3 addition.