Figure 1. Research areas within advanced combustion systems and fuels 63

Figure 2. Example diffused back-illumination extinction imaging images using the original setup (top) and optimized setup (bottom). The light and dark regions in the top image are caused by beam steering... 65

Figure 3. Vapor mass fraction contours through a cut-plane passing through a single orifice of Spray G showing the propensity of ethanol blends to flash boil significantly more than the pure components.... 66

Figure 4. Improved model of the spark channel elongation and energy deposition at non-quiescent flow conditions (Scarcelli, report I.10) 66

Figure 5. Enthalpy fraction of H₂, CO, and CH₄ at the catalyst outlet relative to the initial fuel with 700°C initial temperature for iso-octane (Szybist, report I.12) 67

Figure 6. New constant pressure flow rig with air exhaust, injector cooling, and instrumentation installed (Agrawal, report I.24) 69

Figure 7. Premixed flame propagation and end-gas ignition in a constant volume configuration. A base conditional moment closure front propagation formulation solver has been coupled with an large... 69

Figure 8. Laser-based measurements allow the detection of fuel wall films for various combinations of fuels and operating conditions that are prone to pool fires (Sjoberg, report II.4) 71

Figure 9. Full factorial experimental design fuel matrix. Tested fuels are indicated by red dots; aromatic hydrocarbon vapor pressures are at 443 K. (Ratcliff, report II.7) 71

Figure 10. Schematic diagram of the heavy-duty, single-cylinder optical diesel engine and optical setup with infrared emission and visible natural combustion luminosity cameras (Musculus, report II.16) 73

Figure 11. Two-color soot pyrometry image processing. (Left) double image of burner showing tip in focus, no filters. (Middle) double image of flames at two wavelengths. (Right) Pixel-by-pixel intensity... 74

Figure 12. Example calculation of the constant volume ignition delay times (IDT) used to construct the phi-sensitivity metric for iso-octane (McNenly, report II.24) 75

Figure 13. CONVERGE computation fluid dynamics model of the diesel Ricardo Hydra engine (Lawler, report III.2) 76

Figure 14. Catalyst locations within small cross-sections of a commercial selective catalytic reduction filter (catalyst shown in red false color; note that inlet channels are larger and outlet channels are... 78

Figure 15. Three-dimensional reconstruction of a small section of the C2 material from X-ray computed tomography data and an associated flow field (color key indicates local gas velocities in cm/s) (Stewart, report IV.7) 79

Figure 16. Gen3 GDCI powertrain (Confer, report V.5) 81

Figure 17. Critical hardware components (Battiston, report V.7) 81

Figure 18. Electrically assisted variable speed electric waste heat recovery unit (Patil, report V.9) 82

Figure 19. Plasma transfer wire arc coated cylinder bores in a linerless engine block (Gangopadhyay, report VI.7) 84

Figure 20. Splitter plate dimensions and three-vehicle platoon in the Army 7 ft × 10 ft wind tunnel located at National Aeronautics and Space Administration Ames (Salari, report VII.1) 85

Figure 21. ISX450 engine with UltraShift Plus automated manual transmission under test in the powertrain test cell at Oak Ridge National Laboratory's Vehicle Systems Integration Laboratory (Deter, report VII.5) 86

Figure I.1.1. In a diesel engine, increasing the work extracted by the piston is significantly more effective at improving thermal efficiency than reducing the amount of energy lost through the combustion chamber walls 89

Figure I.1.2. Thermal efficiency (the fraction of fuel energy extracted as work) is higher with the stepped-lip piston for some injection timings. The efficiency differences correlate with differences in the degree of... 89

Figure I.1.3. As the main injection timing is delayed, the combustion takes place later in the expansion stroke. Piston bowl geometry does not affect the first half of the combustion event (until CA50), but the... 90

Figure I.1.4. Comparison between fuel vapor concentrations measured experimentally by planar, laser-induced fluorescence of a fuel tracer (left), and fuel vapor concentration predicted by the CFD simulation... 91

Figure I.1.5. Simulated vertical-plane projection of flow fields (shown with colored vectors) and the fuel concentrations (false-colored field data) for the conventional (left) and stepped-lip (right) pistons.... 92

Figure I.2.1. Optical engine schematic showing LED beam path (propagating from top to bottom) and two-camera setup for simultaneous DBI and NL imaging 95

Figure I.2.2. Mean PB for using various single-and post-injection schedules. 18% intake O2, 15.8 kg/m3 and 910 K TDC motored 96

Figure I.2.3. KL evolution within the DBI FoV from ensemble-averaged images with post-injections of various DSE added to a main injection of 2,350 μs DSE, as indicated in the legend. 18% intake... 96

Figure I.2.4. Correlation between the late-cycle FoV-averaged KL and engine-out PB data in Figure I.2.2 that have three replicates. Dotted lines show the 95% confidence interval 97

Figure I.2.5. Composite graphic showing injector solenoid energizing schedule (top left), in-cylinder geometry (bottom left), soot KL measured by ensemble-averaged DBI (top row of images)... 98

Figure I.2.6. Top row: measured 2D NL image (left), predicted 2D NL image (right), predicted three-dimensional (3D) soot distribution (right). Bottom row: transfer function map overlaid on... 98

Figure I.3.1. Example DBIEI images using the original setup (top) and optimized setup (bottom). The light and dark regions in the top image are caused by beam steering and introduce... 103

Figure I.3.2. A collage of images demonstrating three high-speed, quantitative diagnostics. The liquid length and soot images were obtained using a Sandia-developed LED driver that... 103

Figure I.3.3. Instantaneous ensemble-averaged (top) and single-injection (bottom) liquid phase images, 2.5 ms after start of injection of Spray D (magenta) and Spray C (green) at standard... 104

Figure I.3.4. Ensemble averaged reacting and non-reacting vapor penetration rate for Sprays D and C with error bars indicated. The nominal ambient conditions are stated in the figure 104

Figure I.3.5. Scatter plots of the quasi-steady lift-off length (left) and ignition delay time (right) as a function of ambient temperature (top) and ambient density (bottom). The grey markers... 105

Figure I.4.1. CA50 control using DDI-PFS to vary charge reactivity for a wide range of intake pressures for (a) CR = 16:1 and (b) CR = 14:1. The curves have been offset to align the CA50s... 110

Figure I.4.2. NOx and PM measurements for the CA50 control sweeps in Figure I.4.1a. PM data are those provided by AVL smoke meter measurements 111

Figure I.4.3. (a) Spark timings required for SA to compensate for a reduction in Tin for Pin = 1.0-1.3 bar, or to compensate for a reduction in intake O₂ for Pin = 1.6 bar. (b) Maximum... 112

Figure I.4.4. CA50 control authority with spark timing for Φ = 0.38, 0.42, and 0.45 at Pin = 1.0 bar 113

Figure I.4.5. Uncertainties in the cylinder pressure measurement through a typical engine cycle, (a) in kPa and (b) as a fraction of the total pressure 114

Figure I.5.1. Fuel energy utilization breakdown for two low-load (1.4 and 3.0 bar indicated mean effective pressure [IMEP]) ACI operating points with a fixed 265 J fuel injection during the... 117

Figure I.5.2. Map of CA50 for a range of intake temperatures and O₃ concentrations with a fixed -230 °CA start of injection (SOI) and 1,000 rpm engine speed. Charge φ was maintained... 118

Figure I.5.3. Quantitative in-cylinder O₃ measurements compared to chemistry modeling results of O₃ and O 119

Figure I.5.4. Measured apparent heat release rate and O₃ profiles compared to simulation O₃ results and important radicals (top). Model-predicted fuel ROC by O, OH, and HO₂ (bottom).... 119

Figure I.5.5. Pressure rise calorimetry for 20 kV LTP discharges at a 2.8 bar initial pressure and varying H₂O/CO₂ concentrations 120

Figure I.5.6. Qualitative images of excited state O for two initial pressures along with a schlieren image of temperature gradients for a 20 kV LTP discharge (top) compared to VizGlow predictions... 121

Figure I.6.1. Gas phase axial velocity vs. time is plotted for simulations performed using both LES and RANS turbulence models compared against experimental PIV data obtained from Sandia... 125

Figure I.6.2. Vapor mass fraction contours through a cut-plane passing through a single orifice of Spray G show the propensity of ethanol blends to flash boil significantly more than the pure components... 125

Figure I.6.3. Ignition delay vs. ambient temperature predicted using TFM and HR-MZ models and compared against experimental data from Sandia National Laboratories 126

Figure I.6.4. Flame structure in terms of CH₂O, OH, and temperature predicted using TFM approach for both 750 K and 900 K cases. The stoichiometric mixture fraction line is also shown 126

Figure I.6.5. Ignition delay and flame lift-off lengths for a new four-component diesel surrogate mechanism developed and validated against experimental data from Army Research Laboratory 127

Figure I.7.1. Simulations of flow through two diesel nozzles and the emerging liquid jets, courtesy of M. Battistoni at the University of Perugia. The results at left show the flow simulated using... 131

Figure I.7.2. A cutaway though a GDI injector measured using a combination of X-ray and neutron tomography. Five of the eight spray holes are visible at the center of the image 132

Figure I.7.3. A cross section of the fuel density as the liquid jet emerges from a single-hole diesel injector 132

Figure I.8.1. Experimentally measured (symbols) and modeled (lines) ignition delay times as a function of inverse temperature for FACE-F and various surrogate blends 137

Figure I.8.2. Representative pressure-time histories for syngas/O₂/diluent mixture illustrating the influence of hot-spot ignited flames. Inset time-shifts pressure traces corresponding to... 138

Figure I.9.1. Influence of turbocharger duty cycle (DC) upon engine efficiency 142

Figure I.9.2. Influence of injection pressure upon smoke (FSN) and combustion noise 142

Figure I.10.1. Improved model of the spark channel elongation and energy deposition at non-quiescent flow conditions 146

Figure I.10.2. Impact of in-cylinder turbulence on the performance of laser ignition and conventional spark 146

Figure I.10.3. Improved dilution tolerance with non-equilibrium plasma systems versus conventional spark (top) and occurrence of arc events during the non-equilibrium discharge (bottom) 147

Figure I.10.4. Effect of ambient conditions on the characteristics of the streamers (top) and transition from LTP to arc regime (bottom) 148

Figure I.10.5. Qualitative comparison between modeling and experiments in terms of chemical (left) and thermal (right) LTP properties 149

Figure I.11.1. Advanced combustion strategies shown as a continuum based on the level of fuel stratification at start of combustion 152

Figure I.11.2. Concept of fuel reactivity stratification with single-fuel (GCI) to dual-fuel 153

Figure I.11.3. CFD modeling of RCCI across a span of delta in fuel reactivies 153

Figure I.11.4. Control authority of RCCI allows for rapid and stable mode switches 154

Figure I.12.1. Schematic of the in-cylinder reforming process in which one cylinder has an isolated intake and exhaust, feeds the reforming catalyst, and is incorporated into the intake for the other three cylinders 157

Figure I.12.2. Enthalpy fraction of H₂', CO, and CH₄ at the catalyst outlet relative to the initial fuel with 700°C initial temperature for iso-octane 158

Figure I.12.3. Transient response of the exit concentration of catalytic (Cat) reforming products and monolith temperature profile at an O₂ catalyst flow rate of 12 g/min and Fcatalyst = 7 159

Figure I.12.4. PMEP as a function of intake manifold pressure for conventional EGR and catalytic reforming 159

Figure I.12.5. Brake thermal efficiency as a function of intake manifold pressure for conventional EGR and the catalytic reforming strategy 160

Figure I.12.6. In-cylinder temperature of (a) conventional SI combustion and (b) the dilute combustion strategy supported by reforming. The high dilution reduces in-cylinder temperature by more than 550 K 160

Figure I.13.1. (a) Mass-attenuation coefficients versus atomic number, and (b) schematic of neutron imaging apparatus 162

Figure I.13.2. System used to study intra-nozzle dynamics of fuel injection including (a) high-pressure fuel delivery system and (b) the aluminum spray chamber with optical viewport. The spray... 163

Figure I.13.3. Detection and scaling of neutron attenuation by fluid to obtain cumulative injector spray mass 164

Figure I.13.4. (a) Sagittal slice of neutron CT data, with X-ray CT domain marked, and (b) internal fluid velocities from a computational fluid dynamics simulation (Courtesy of Argonne National Laboratory) 165

Figure I.13.5. Results from successive GPF regeneration study to date for (a) E0 and (b) E30, including the recently imaged 60% regeneration state, with (c) comparison of activation energies for... 166

Figure I.13.6. Neutron attenuation at various axial locations for a single channel both with the original field-loaded DPF (ash+soot) and after regeneration (ash only). Inset images show attenuation... 166

Figure I.14.1. Experimentally measured (symbols) and simulated (curves) IDTs of n-decane in shock tubes 170

Figure I.14.2. Experimentally measured (symbols) and simulated (curves) IDTs of decalin in a shock tube and RCM 170

Figure I.14.3. Measured (symbols) and simulated (lines) IDTs of a 1-methylnaphthalene (AMN)/n-dodecane stoichiometric blend for a pressure of 15 bar 171

Figure I.14.4. Simulated (lines) and experimentally measured (symbols) of IDTs for iso-octane. The dash lines are from the previous LLNL iso-octane detailed chemical kinetic model and the solid... 172

Figure I.14.5. Simulated (curves) and experimentally measured (symbols) IDTs for gasoline primary reference fuel PRF90. The symbols are experimental measurements from the ANL RCM.... 172

Figure I.14.6. Simulated (lines) and experimentally measured (symbols) IDTs for TPRF for stoichiometric fuel/air mixtures at 20 bar. The dash lines are from the LLNL detailed chemical kinetic... 173

Figure I.15.1. Improvement in emissions prediction as a function of simulation complexity. (a) Results for baseline simulations, (b) baseline simulations with more chemical detail, (c) baseline simulations... 176

Figure I.15.2. (a) Baseline and (b) revised estimates of in-cylinder pressure uncertainty in LTGC engine experiments 177

Figure I.16.1. Parallel scaling efficiency of the adaptive preconditioner method applied to a fully-coupled, multi-species transport solver used to model the unsteady flame dynamics in a millimeter-scale flow reactor 182

Figure I.16.2. Fraction of species in a detailed gasoline surrogate chemistry model that have a mass fractions below a specified extinction level over the course of a constant volume ignition delay calculation 183

Figure I.16.3. The dormant order classification of the system states representing the species mass fraction composition and temperature evolution of iso-octane in a constant volume ignition delay calculation 183

Figure I.17.1. Computational grid using GridPro and overset valve and piston surfaces. In close-up, the valves and the injector module are easily married to the cylinder domain 188

Figure I.17.2. Two-Port engine. (a) Intake valve operating showing the magnitude of velocity of the fluid entering at 80° after top dead center and (b) exhaust valve operating showing magnitude of velocity... 189

Figure I.17.3. Multiphase flow simulation with VOF method, gasoline injected into quiescent air at 3 bar. (a) Gasoline jet primary break-up into ligaments and (b) primary break-up and w-component of velocity... 189

Figure I.17.4. The Spray G ECN case with the grid in cut-away showing the nozzles 190

Figure I.17.5. The ECN Spray A case: injection of diesel in quiescent nitrogen at 2.2 MPa, KH-RT spray model 190

Figure I.17.6. KIVA-hpFE's ever increasing advantage in computational speed over KIVA-4mpi; at 0.38 s of simulation time, KIVA-hpFE is 1.44× faster (for the problem being solved by both systems with exactly the same settings) 191

Figure I.18.1. Comparison of experimentally measured emissions index for NOx and HC with simulation predictions using different levels of modeling detail. (Figure Credit) 196

Figure I.19.1. Instantaneous cross-section of the scalar dissipation field showing modeled instantaneous fluctuations of the equivalence ratio versus time at two locations in the flow 201

Figure I.19.2. Transient ignition sequence predicted by LES at conditions identical to the experiment 201

Figure I.20.1. (a) Axial velocity and (b) temperature contours on horizontal slices at z = -3.75 mm and z = -0.9375 mm (distances below the head) at 346° after top dead center. (This is Figure 4 of Schmitt et al) 205

Figure I.20.2. Left: Schematic of the experimental setup for the Hencken burner spectral radiation measurements at the University of Michigan. Right: Measured infrared spectra for a one atmosphere... 205

Figure I.20.3. Schematic of the metal engine configuration at ORNL 206

Figure I.20.4. Schematic of a simplified spectral box model for high-pressure combustion systems 206

Figure I.21.1. Temporal history of multi-component droplet vaporization for ethanol-isooctane binary system under atmospheric pressure at 200°C (E0 = pure ethanol; E85 = 85% ethanol, 15% iso-octane by volume) 211

Figure I.21.2. The density contours (a) iso-octane, (b) ethanol, (c) E30, (d) E85 across a horizontal plane placed 10 mm downstream of the injector tip when subjected to Spray G2 condition 211

Figure I.21.3. Tower chamber and heating system 212

Figure I.21.4. Spray evolution for flash and non-flash boiling at 400 K and varying injection (Pinj) and environment (Penv) pressures: (a) Pinj = 2 bar, Penv = 0.5 bar; (b) Pinj = 5 bar, Penv = 0.5 bar; (c)... 213

Figure I.21.5. The droplet and film chamber components and system 213

Figure I.21.6. Droplet diameter evolution during vaporization for various ethanol-iso-octane blends (E0: 0% ethanol, 100% iso-octane by volume; E100: 100% ethanol, 0% iso-octane by volume) 214

Figure I.22.1. The impinged spray radius (left) and the impinged spray height (right) versus ASOI with various ambient densities (14.8 kg/m³, 22.8 kg/m³, and 30 kg/m³) (top) and various injection pressures... 217

Figure I.22.2. Effects of ambient density at injection pressure of 1,500 bar (left) and injection pressure at ambient density of 22.8 kg/m³ (right) on film thickness 218

Figure I.22.3. Effects of ambient density at injection pressure of 1,500 bar on heat flux at three locations 218

Figure I.22.4. Comparison of axial (top) and radial (bottom) impinged spray quantities against experiments, O'Rourke and Amsden's model (OA), and new implementation (YW) based on Yarin and Weiss' theory 219

Figure I.22.5. DNS results of the surface impingement of a train of 109-micron ethanol drops at non-splashing (top row) and splashing (bottom row) regimes at grid resolution of (a, c) 2.5 microns and... 220

Figure I.22.6. Splashed mass ratio versus time in surface impingement of a train of diesel droplets of 5.96 micron diameter impacting at 77 m/s from two DNS studies at 40 CPD and 80 CPD resolutions.... 220

Figure I.23.1. Validation of LES model through comparisons with (a) non-reacting fuel distributions, (b) bulk combustion characteristics (cylinder pressure and heat release rate [HRR]), and (c) locations of... 225

Figure I.23.2. Overview of detailed soot model and Lagrangian soot parcels 226

Figure I.23.3. (Top) soot parcels with velocity vectors showing implementation of soot model. (Bottom) output of soot parcel model late in the cycle for HCCI and conventional diesel combustion 227

Figure I.23.4. Comparison of measured and predicted particle size distributions. The predictions use the soot model developed in the current program in the one-way coupled approach 228

Figure I.23.5. (a) Comparison of a baseline mode, our current model, and measured particle size distribution at several times. (b) Resulting particle size distributions with several levels of stratification... 228

Figure I.24.1. New CPFR with air exhaust, injector cooling, and instrumentation installed 232

Figure I.24.2. First reacting sprays captured in new CPFR, one (left) with longer exposure time and another (right) with less exposure time 232

Figure I.24.3. Image processing steps for RSD images at 0.625 ms aSOI for n-heptane injected at 1,000 bar for 4 ms into ambient air at 30 bar and 825 K. A total of 50 injections were captured in this case 233

Figure I.24.4. RSD images during combustion at 1.3 ms aSOI for n-heptane injected at 1,000 bar for 4 ms into ambient air at 30 bar and 825 K. A total of 50 injections were captured in this case 234

Figure I.24.5. (a) Fuel mass fraction and (b) temperature 235

Figure I.24.6. (a) Molecular viscosity and (b) thermal conductivity 235

Figure I.24.7. (a) Specific heat and (b) fuel mass fraction along jet center 235

Figure I.24.8. Two-dimensional distributions of pressure, density, and velocity at 0.18 ms aSOI 236

Figure I.25.1. Premixed flame propagation and end-gas ignition in a constant volume configuration. A CMC-FPF solver has been coupled with an LES solver. Total enthalpy and density fields at the... 239

Figure I.25.2. DNS datasets. (a) and (b) 2D reduced chemistry DNS of end-gas ignition. Temperature fields at (a) t = 2.5 ms and (b) 2.84 ms for a representative case with the temperature inhomogeneity... 239

Figure I.25.3. Assessment of CMC reaction rate closure models using 2D reduced chemistry DNS of end-gas ignition. Sensible-enthalpy-based CMC performs very well for the entire duration of the ignition... 240

Figure I.25.4. Experimental data of knock limited combustion phasing with 15% EGR (red) and without EGR (black) at two different air flow rates (i.e., loads). Data highlights that as airflow and thus load... 242

Figure I.25.5. Evolution of flame fronts in LES of an SI engine. The model parameters in a turbulent subfilter flame speed model [8] are (top) b₁ = 10 and b₂ = 2; and (bottom) b₁ = 10 and b₃ = 2 242

Figure I.26.1. (a) High pressure fuel delivery system and (b) spray chamber installed at HFIR. (c) Computer-aided drawing of injector holder and the necessary sweep gases to minimize fuel build-up on... 245

Figure I.26.2. Left: Bubble shapes in time results from Verma et al. (top) and current BU OpenFOAM simulations (bottom) for TC2. Non-dimensional times displayed are (a) t = 0, (b) t = 1.5 and (c)... 247

Figure I.26.3. Left: LDV signal of fuel injector tip excitation from pintle opening and closing. Right: Max amplitude of LDV at the tip and 10, 15, 20, 25, and 30 mm away from the free end 248

Figure I.26.4. Contrast-enhanced radiograph showing fluid in the sac and the area of interest in the sac-nozzle interface (left). Radiograph with traverse used in calibrating neutron attenuation intensity... 249

Figure I.26.5. Attenuation within the area of interest, with a guideline defining the threshold of liquid detection, and conversion to the deviation from this threshold representing detected mass (left).... 249

Figure I.27.1. USAXS measurements of spray centerline SMD within the first 20 mm of Spray D over a wide range of ambient backpressures and injection pressures 253

Figure I.27.2. SAMR measurements of spatially resolved SMD in Spray D at ramb = 2.4 kg/m³, Pinj = 50 MPa 253

Figure I.27.3. Validation of SMD predictions by new KH-Faeth primary atomization model against USAXS measurements. Also shown are predictions by two benchmark models in current use within... 254

Figure II.1.1. Illustration of repeating Fire20-Skip80 sequence used to mimic the thermal state that an engine may experience during vehicle acceleration. Figure by Magnus Sjoberg 257

Figure II.1.2. Knock-limited combustion phasing as a function of intake pressure for both steady-state and transient operation with 30°C intake temperature. Transient operation is shown in dashed lines.... 258

Figure II.1.3. Indicated thermal efficiency as a function of intake pressure for all nine fuels for both steady-state and transient operation. Transient operation is shown in dashed lines. Figure by David Vuilleumier, SNL 259

Figure II.1.4. Best-fit linear regression between octane index and knock-limited combustion phasing for an intake pressure of 146 kPa under transient operation. Figure by David Vuilleumier, SNL 260

Figure II.1.5. Distribution of R2 values at the 146 kPa intake pressure, 30°C intake temperature, transient operating mode condition. Figure by David Vuilleumier, SNL 260

Figure II.1.6. K-values as a function of intake pressure and operation type accompanied by the 10th and 90th percentiles of the K distribution, as well as the mode of the distribution. Figure by David Vuilleumier, SNL 261

Figure II.2.1. The effect of ethanol blending on the base fuel RON requirement for PRFs to get a blended RON 98 264

Figure II.2.2. Temperature drop across the carburetor with increasing level of ethanol, causing super-saturation 265

Figure II.2.3. Minimum temperature for full vaporization with increasing level of ethanol and calculated uncertainty 265

Figure II.2.4. Change in ΔRON for PRF-ethanol blends between standard RON operating conditions (Case 1), compensated mixture temperature (Case 2), and pressure at spark timing (PST) (Case 3) 266

Figure II.2.5. Change in ΔRON for PRF-ethanol blends between standard RON operating conditions (Case 1) and each, compensated mixture temperature (Case 2), pressure at spark timing for that... 267

Figure II.3.1. Crank angle at 50% fuel burned (CA50) phasing versus brake mean effective pressure at 2,000 rpm for the 97 RON fuels using CR11.4 pistons 270

Figure II.3.2. Volumetric heating value for the 97 RON expanded matrix fuels 270

Figure II.3.3. Modeled volumetric fuel economy for a typical mid-size sedan using CR11.4 and the 97 RON fuels 271

Figure II.4.1. For boosted, stratified charge, direct injection SI operation at 2,000 rpm, engine-out soot increases monotonically with the fuels' PMI, as the soot formation is primarily occurring in the... 275

Figure II.4.2. A change of NOx-PM trade-off indicates a change of the dominating soot production pathway with fuel type and operating conditions. Figure by Magnus Sjoberg 275

Figure II.4.3. Laser-based measurements allow the detection of fuel wall films for various combinations of fuels and operating conditions that are prone to pool fires. Figure by Carl-Philipp Ding, Xu He, and Magnus Sjoberg, SNL 276

Figure II.4.4. Conceptual comparison of fuel injection and spark timing strategies for operation with partial fuel stratification and full stratification. Figure by Magnus Sjöberg and Zongjie Hu, SNL 277

Figure II.4.5. IR fuel-vapor imaging reveals extent of fuel stratification relative to the flame spread for operation with a late injection of 3.6 mg gasoline at the time of spark. Figure by Magnus Sjoberg, SNL 277

Figure II.4.6. Injection of 1.6 mg of fuel at the time of spark stabilizes combustion, enabling studying autoignition of lean end-gas with φ = 0.50, 1,400 rpm, Pin = 100 kPa. Figure by Magnus Sjoberg, SNL 278

Figure II.5.1. Effect of adding high LFS ethanol component on combustion stability with EGR dilution 282

Figure II.5.2. Variation of EGR tolerance with various fuel mixtures plotted against their LFS at 5.6 bar IMEPn 282

Figure II.5.3. Variation of EGR tolerance with various fuel mixtures plotted against their LFS at 3.2 bar IMEPn 282

Figure II.5.4. Comparison of timing of key combustion events for various fuel blends at constant 25% EGR 283

Figure II.5.5. Comparison of timing of key combustion events with E30 fuel mixture at 5.6 bar and 3.2 bar load at 23% EGR 283

Figure II.6.1. Knock-limited combustion phasing as a function of octane index for a single engine operating condition 286

Figure II.6.2. Cylinder pressure and heat release rate for engine operating Conditions 1 and 6 for select fuels 287

Figure II.6.3. Notional PT trajectories for changing intake manifold pressure. Solid lines 9.2:1 compression ratio, dashed lines represent continuation of trajectory for compression ratio of 14:1 288

Figure II.6.4. Ignition delay contour in the PT space for the aromatic fuel without EGR highlighting the different ignition zones 289

Figure II.6.5. PT path of the unburned fuel-air mixture during compression for different intake manifold pressure (Pin). Also shown are lines of constant ignition delay (8 ms) for 0%, 10%, and 20% EGR 289

Figure II.7.1. Full factorial experimental design fuel matrix. Tested fuels are indicated by red dots; aromatic hydrocarbon vapor pressures are at 443 K 292

Figure II.7.2. PM results from cumene at 20 vol%; error bars are 95% confidence intervals 294

Figure II.7.3. PM results from 4-t-butyltoluene at 10 vol%; error bars are 95% confidence intervals 294

Figure II.7.4. PM results from 4-t-butyltoluene at 20 vol%; error bars are 95% confidence intervals 295

Figure II.7.5. ADC results for p-cymene fuel blends with and without ethanol, showing ethanol suppression of p-cymene evaporation while ethanol remains in the liquid phase 296

Figure II.7.6. Modeled vapor phase concentrations of ethanol, paraffins, and aromatics from the surface of a droplet evaporating near the end of compression 296

Figure II.8.1. Total hydrocarbon conversion as a function of temperature for an E10 surrogate fuel over a hydrothermally aged commercial TWC at lambdas ranging from 0.995 to 0.999 301

Figure II.8.2. Comparison of total hydrocarbon conversion light-off curves over a hydrothermally aged commercial TWC for selected fuel components representative of alcohols (ethanol), linear ketones... 302

Figure II.8.3. Comparison of T50 and T90 total hydrocarbon light-off temperatures over the hydrothermally aged commercial TWC for all of the fuel components investigated. Error bars represent 95%... 303

Figure II.8.4. Comparison of T50 and T90 CO light-off temperatures over the hydrothermally aged commercial TWC for all of the fuel components investigated. Error bars represent 95% confidence... 303

Figure II.9.1. Six fuels, including the BOB and five bio-fuel blends, were tested for PM mass emissions on a GDI engine during a 90-s cold-start transient. The average mass measurements can be... 307

Figure II.9.2. Six fuels, including the BOB and five bio-fuel blends, were tested for PM mass emissions on a GDI engine. All five fuels were at or above the California Air Resources Board's 1 mg/mi gravimetric... 308

Figure II.10.1. Three-dimensional chromatogram of ionic species extracted from soot. Each color band represents a compound and the amount of the compound is represented by the color with red being... 311

Figure II.11.1. Three-dimensional surface maps of ID for two gasoline surrogate blends, a 25% ethanol blend into TRF88 on the left and neat TRF88 on the right 315

Figure II.11.2. Constant mass injection temperature sweeps at 1.0 MPa are shown with data plotted every 25 K. A 10% ethanol blend increases ID compared to iso-octane, but it maintains negative... 316

Figure II.11.3. Constant mass injection temperature sweeps at 1.0 MPa are shown with data plotted every 25 K. Ethanol blending effects into PRF95 are much more linear than with neat iso-octane (illustrated in Figure II.11.2) 316

Figure II.12.1. LSPI count for the four tested fuel combinations. (solid) LSPI event count, (crosshatch) LSPI cluster count (Figure Credit: Derek Splitter) 321

Figure II.12.2. The effect of different fuels on LSPI event start location (Figure Credit: Derek Splitter) 322

Figure II.12.3. Impact of various fuels on LSPI dwell (Figure Credit: Derek Splitter) 322

Figure II.12.4. Example of LSPI dwell times for EB25 exhibiting short and long dwell effects knocking behavior (Figure Credit: Derek Splitter) 323

Figure II.12.5. LSPI intensity observed with various fuels (Figure Credit: Derek Splitter) 323

Figure II.13.1. Fueling rate and CMT as a function of EGR for each fuel 327

Figure II.13.2. Tradeoff between intake air temperature and intake boost for constant combustion phasing (CA50) 328

Figure II.13.3. Engine out smoke (Filter Smoke Number) as a function of EGR for the three 98 RON fuels 328

Figure II.13.4. Soot luminosity for three 98 RON fuels as a function of engine crank angle 329

Figure II.13.5. Heat release profiles for each 98 RON fuel 329

Figure II.14.1. Combustion phasing (CA50) as a function of intake temperature (Tin) for various fuels for fully premixed LTGC operation at Pin = 1.0 bar, Φ = 0.4, 1,200 rpm 332

Figure II.14.2. CA50 for Tin = 154°C as a function of OI = RON - KS. Based on the Tin = 154°C data in Figure II.14.1 332

Figure II.14.3. CA50 as a function of Tin for early-DI fueling compared to premixed prevaporized fueling for the Co-Optima E30 and high-aromatic fuels. The larger HOV of E30 is evident in the greater difference... 333

Figure II.14.4. TBDC required for a CA10 of 368.7°CA (solid lines) or 371.5°CA (dashed lines) as a function of Pin for the Co-Optima E30 and high-aromatic fuels and regular E10. The E30 and high-aromatic fuels... 334

Figure II.15.1. Diagram of the multi-cylinder 1.9-L ACI engine 338

Figure II.15.2. Collaboration between metal engine and optical engine experiments 339

Figure II.15.3. RCCI injection strategy used in the present study: single PFI during early intake stroke and single DI with timing swept from bottom dead center to near-top dead center of compression stroke 339

Figure II.15.4. Mode transitions into RCCI with various combinations of PRF blends as a function of SOI timing 340

Figure II.16.1. Schematic diagram of the heavy-duty, single-cylinder optical diesel engine and optical setup with IR emission and visible natural combustion luminosity cameras 343

Figure II.16.2. Combustion phasing for both the single-cylinder optical engine (with standard deviation over 36 fired cycles), and the multi-cylinder metal engine (with standard deviation over four cylinders during steady-state) 344

Figure II.16.3. Simultaneous IR and visible NL images during HTHR period for 300 CAD CR SSE near the early injection control authority limit. Each pair of images corresponds to different combustion cycles... 345

Figure II.16.4. Simultaneous IR and visible NL images, during HTHR period for 310 CAD CR SSE, slightly retarded from the early injection control authority limit. Each pair of images corresponds to different... 345

Figure II.16.5. Simultaneous IR and visible NL images during HTHR period for 320 CAD CR SSE slightly advanced from the late injection control authority limit. Each pair of images corresponds to different combustion... 346

Figure II.16.6. Simultaneous IR and visible NL images during HTHR period for 330 CAD CR SSE near the late injection control authority limit. Each pair of images corresponds to different combustion cycles, not... 346

Figure II.17.1. Quantitative images of soot optical thickness (KL) from diffused back illumination movies of free-spray combustion (left column) vs. DFI combustion (right column). False color is used to indicate higher... 349

Figure II.17.2. Comparison of duct inlet and outlet shape effects on DFI performance. The duct shape that corresponds to each Greek letter is shown for reference. All data are for D2L8G3.79 ducts at a nominal... 350

Figure II.17.3. (a) Shortening duct length from 16 mm to 8 mm does not have a detrimental effect on soot mass for a duct with 2-mm diameter and δ configuration. (b) Decreasing the standoff distance from... 351

Figure II.17.4. Lines show predicted HC emissions indices from overlean fuel during the mixing-controlled burn, for 21% oxygen and φent values of 0.35 and 0.5 for the 180-MPa and 80-MPa injection pressures... 352

Figure II.18.1. Two-color soot pyrometer schematic (left) and optical hardware (right) 356

Figure II.18.2. Two-color soot pyrometry image processing. (Left) double image of burner showing tip in focus, no filters. (Middle) double image of flames at two wavelengths. (Right) Pixel-by-pixel intensity ratio... 356

Figure II.18.3. New CFR with air exhaust, injector cooling, and instrumentation installed 357

Figure II.18.4. First reacting sprays captured in new CFR, one (left) with longer exposure time and another (right) with less exposure time 357

Figure II.18.5. Image processing steps for RSD images at 0.625 ms after start of injection (aSOI) for n-heptane injected at 1,000 bar for 4 ms into ambient air at 30 bar and 825 K. A total of 50... 358

Figure II.18.6. RSD images during combustion at 1.3 ms aSOI for n-heptane injected at 1,000 bar for 4 ms into ambient air at 30 bar and 825 K. A total of 50 injections were captured in this case 359

Figure II.19.1. Representative models for DCN and T10. Observed values are compared to predicted using the full data set (Full Set, red circles), and leave-one-out predictions (LOO, blue asterisks)... 362

Figure II.19.2. Blending model development. The charts above show the pairwise interactions between two sets of functional groups on the value of DCN. In both charts, the population of Functional... 362

Figure II.19.3. Comparison of the solid-liquid equilibria for four diesel fuel surrogates, V0a, V0b, V1, and V2, and GTL-Diesel. Data points for each material are shown in various colors. Compositions... 363

Figure II.19.4. Comparison of the solid-liquid equilibria for diesel fuel surrogate V0b with eight cold flow improvers. The larger graph documents cold flow improver performance across a broad rangr... 364

Figure II.20.1. The predictions of octane blending of Co-Optima Tier 3 HPFs in a four-component base gasoline using the Co-Optima chemical kinetic model (curves) compared experimentally measured... 367

Figure II.20.2. RON is correlated to the IDT (Tau) computed by the kinetic model for stoichiometric fuel/air mixtures at 775 K and 25 atm (lower left plot). Octane sensitivity (SEN) is correlated to the... 368

Figure II.20.3. Laminar flame speeds (LFS) computed by the LLNL chemical kinetic model for gasoline for a high-octane and moderate-octane sensitivity over a wide pressure and temperature range relevant to ASI engines 368

Figure II.21.1. Experimentally measured (symbols) and simulated (lines) ignition delay times of methyl acetate in a shock tube at NUIG (LLNL) 372

Figure II.21.2. Simulated (curves) and experimentally measured (symbols) IDTs of n-heptane in a shock tube over a range of pressures and fuel/air equivalence ratios (Phi). The predictions by the... 373

Figure II.21.3. Simulated and measured IDTs for a toluene/n-heptane 50:50 molar blend at a fuel/air equivalence ratio (Phi) of 0.5. The IDTs were measured in a shock tube and an RCM at NUIG. (LLNL) 373

Figure II.21.4. Simulated (curves) and experimentally measured (symbols) laminar flame speeds for n-heptane and iso-octane. The measurements are from Ji et al. (LLNL) 373

Figure II.22.1. Experimentally measured pressure-time histories for Co-Optima core fuels indicating reactivity trends for the fuels in different combustion regimes, (a) high pressure, lower temperature... 377

Figure II.22.2. Computationally derived isopleths of ignition delay times (t) and normalized heat release rates (e) for PRF70 - PRF100 covering a wide range of thermodynamic conditions relevant to... 378

Figure II.23.1. PCA model validation via leave-one out cross validation for prediction of RON and S 381

Figure II.23.2. Gaussian process regression model validation via leave-one out cross validation for prediction of RON and S 382

Figure II.23.3. RON and S classification results for linear model and Gaussian process 382

Figure II.23.4. Comparison of isooctane in National Renewable Energy Laboratory (NREL) flow reactor to LLNL 1D plug flow model 383

Figure II.23.5. Comparison of reactor results and simulations for (a) anisole conversion, (b) phenol formation, (c) benzaldehyde formation, and d) benzene formation. A skeletal reaction mechanism... 383

Figure II.24.1. Mathematical definition of the phi-sensitivity metric "big-phi," which is developed to estimate the ability of a fuel blend to improve the controllability of a PSCCI engine 387

Figure II.24.2. Example calculation of the constant volume ignition delay times used to construct the phi-sensitivity metric for iso-octane 387

Figure II.24.3. Trade-off between high octane sensitivity and phi-sensitivity for three four-component gasoline blends with a model-based RON estimate greater than 95. The blends containing 1-pentanol... 388

Figure II.25.1. Predicted second cycle cylinder pressure showing good agreement between the baseline simulations and those conducted with species reduced to ~40 during gas exchange; but less... 393

Figure II.25.2. Reduction in computational flow (a) and kinetics costs (b) using dynamic species reduction during an exploratory homogeneous charge compression ignition simulation. (Proof of concept results with KIVA.) 393

Figure II.26.1. Micro-combustion at elevated pressure: unsteady FREI and stable flames at 4.46 bar (I. Schoegl/LSU) 396

Figure II.26.2. Preliminary results for four-wavelength TFP: 5.6 mil (142 μm) SiC filament yields calibration-free temperature measurements within 20 K of reference measurements, while 3 mil... 396

Figure II.26.3. Flame speeds of TPRF95.6-air mixture (T. Lu/UConn) 397

Figure II.27.1. YSI measured for several oxygenated hydrocarbons formed by adding oxygen atoms to toluene 399

Figure II.27.2. YSIs measured for the five Co-Optima test gasolines (TGs) 400

Figure II.27.3. Comparisons between YSI results for surrogate fuel components obtained using three different methodologies: physical measurement (Yale), empirical estimate (NREL), and kinetic simulation (Penn State) 401

Figure II.27.4. Comparisons between YSI results for surrogate fuels obtained using three different methodologies: physical measurement (Yale), empirical estimate (NREL), and kinetic simulation (Penn State) 402

Figure II.27.5. Predicted YSIs for a range of chemical compounds at 1 atm and 2 atm 402

Figure III.2.1. PRF mapping that allows the determination of an effective PRF number of any liquid or gaseous fuel within this range of compression ratio (CR) and intake temperature 408

Figure III.2.2. Determination of the effective PRF number for each of the reformate fuel mixtures 409

Figure III.2.3. GT-POWER model of the diesel Ricardo Hydra engine used for dual-fuel RCCI and single-fuel RCCI 410

Figure III.2.4. CONVERGE CFD model of the diesel Ricardo Hydra engine 410

Figure III.2.5. CFD results of pressure and heat release rates of single-fuel RCCI using diesel as the parent fuel 410

Figure III.3.1. Passive SCR concept. Ammonia generation mode is created by running stoichiometric or nominally 5-6% rich, generating ammonia across the TWC, which is stored in the SCR. In high... 413

Figure III.3.2. High dilution concepts results 414

Figure III.3.3. Camera arrangement for optical in-cylinder measurement. Two optical probes were positioned toward the spark plug for viewing of ignition point and flame propagation. Cylinder #6 (rear cylinder)... 414

Figure III.3.4. Two optical probes are positioned toward spark plug for viewing of ignition point and flame propagation. The cylinder #6 (rear cylinder) is used for instrumentation access. Combustion... 414

Figure III.3.5. Clearancing of piston crown is required for optical probe protrusion into the combustion chamber and for obtaining field of view for ignition and combustion. Concerns were that modification... 415

Figure III.3.6. Verification of combustion. Comparison of combustion parameters prior to and after combustion chamber modifications indicates post modification performance equivalent to normal cycle-to... 415

Figure III.3.7. Image processing progression: → Raw → Background Subtracted → Median Filtered → Binarized → Threshold 415

Figure III.3.8. CoV images for five crank angles for two operating conditions. Left: 1,000 rpm, 6.8 bar IMEP, l = 1.0, minimum (3%) EGR, CoVIMEP = 0.32%. Right: 1,000 rpm, 6.8 bar IMEP, l = 1.6, 10% EGR, CoVIMEP = 1.60% 416

Figure III.3.9. Process used to assess each system's projected results over various drive cycles. Assessment of each system will be performed by a combination of simulation and actual test results. Engine... 417

Figure III.3.10. Current (rescoped) milestone 417

Figure III.5.1. Sensitivity of BTE to multiple fuel injection events for different ethanol and gasoline (E0) blends 425

Figure III.5.2. Summary of the benefits of ethanol blends on direct injection engine performance 425

Figure III.5.3. Indicated thermal efficiency for air and EGR dilution versus syngas supplementation 426

Figure III.5.4. Particle size distributions and total number emission for gasoline, and 20 % blends of ethanol, iso-butanol and dimethyl furan 426

Figure IV.1.1. Average priority scores for CLEERS organizational activities from the 2017 CLEERS Industry Priorities Survey. Responses were scored as 10 for high priority, 5 for medium priority, and... 430

Figure IV.1.2. Average priority scores across technologies (vertical axis) and market sectors (horizontal axis) from the 2017 CLEERS Industry Priorities Survey. Responses were scored as 10 for high priority... 431

Figure IV.1.3. Average priority scores across research areas (vertical axis) and market sectors (horizontal axis) from the 2017 CLEERS Industry Priorities Survey. Responses were scored as 10 for high priority... 432

Figure IV.1.4. NO storage capacity as a function of inlet NO concentration (X-axis) and temperature (data series) for a commercially relevant passive NOx adsorber material 432

Figure IV.1.5. NO temperature programmed desorption profiles obtained with a commercially relevant passive NOx adsorber material after exposure to 1,000 ppm NO in the presence of (a) 0% H₂O, 0% O... 433

Figure IV.2.1. Complete redox cycling of low-temperature NH3-SCR that involves two Cu(I) centers in the oxidation half-cycle. Key intermediates are highlighted 436

Figure IV.2.2. Estimation of Cu₂+, Cu(OH)+, and CuOx in fresh and hydrothermally aged (HTA) samples. SCR catalyst with Si/Al = 12 and Cu loading of 2.1 wt% was used. HTA-T represents sample... 437

Figure IV.2.3. Low-temperature aftertreatment test protocol structure 438

Figure IV.2.4. NOx adsorption at 100°C for 10 min followed with temperature programmed desorption (10°C/min up to 500°C). The feed gas mixture contains 200 ppm of NOx (185 ppm of NO and 15 ppm of NO₂),... 438

Figure IV.2.5. Scanning transmission microscopy images of the freshly calcined (left), reduced (by 1% H₂/He at 400°C for 30 min, middle), and reoxidized (by 10% O₂/He at 500°C for 1 h, right) Pd/zeolite samples,... 439

Figure IV.2.6. Catalyst locations within small cross-sections of a commercial SCR filter (catalyst shown in red false color - note that inlet channels are larger and outlet channels are smaller in this asymmetric honeycomb filter design) 440

Figure IV.3.1. (a) DRIFTS spectra of CO-binding over Pt/Al₂O₃ (sulfated) and CCC+Pt/Al₂O₃ physical mixture (sulfated) and CCC (degreened) catalysts. (b) Time-on-stream CO oxidation conversion during 5 ppm... 444

Figure IV.3.2. (a) Overview of storage and release efficiencies for Ag and Pd-based zeolites under draft ACEC Tech Team trapping protocol. (b) THC storage and release profile (red) for Pd/ZSM-5 as determined by... 445

Figure IV.3.3. Light-off comparisons of DOC (dashed) and dual-bed trap-DOC system (solid) after (a) 800°C, 4 h aging and (b) 800°C, 50 h ageing with repeated sulfations/desulfations. (c) Schematic of dual-bed trap-DOC system 446

Figure IV.4.1. Pseudo-transient drive cycle used for engine-based studies of the passive SCR system 450

Figure IV.4.2. NOx and NH3 concentration data over the passive SCR lean-rich cycle for (a) a TWC with O₂ and NOx storage components and (b) a TWC without O₂ and NOx storage components 450

Figure IV.4.3. Concentration of NOx entering the TWC ("NOx in") and NH3 downstream of the TWC ("NH₃ out") as a function of the equivalence ratio during rich operation of a lean-rich cycle for average exhaust... 451

Figure IV.5.1. Spatiotemporal comparison of model results with measurements at 3/8L, 1/2L and 3/4L, where L is the overall catalyst length. Comparison is over the four-step protocol which includes step transients... 454

Figure IV.5.2. Spatiotemporal comparison of model results with measurements at 1/16L, 1/8L, 3/16L and 1/4L, where L is the overall catalyst length. Comparison is over the four-step protocol which includes step... 455

Figure IV.5.3. (Left) Spatiotemporally resolved SCR-onset transients throughout a field-aged commercial honeycomb-monolith-supported Cu/CHA catalyst, where L is the overall catalyst length. Field-aged commercial... 456

Figure IV.5.4. Ammonia conversion distributions for a commercial Cu/CHA SCR catalyst in degreened (DeG), hydrothermally aged (HTAged), and field-aged (FAged) states. Conversion inflections occur in the front catalyst... 457

Figure IV.6.1. Gaseous emissions conversions of fresh and aged TWC-GPFs with temperatures at the space velocity of 40,000 h-1 from the lab-flow system 460

Figure IV.6.2. Scanning transmission electron microscopy-electron dispersive X-ray spectroscopy measurements of (a) and (b) fresh TWC-GPF, and (c) and (d) field-aged TWC-GPF 461

Figure IV.6.3. PXRD patterns of TWC-GPF samples: "lab-aged" was aged with conventional engine oil, while "ZDDP" was aged with zinc dialkyldithiophosphate (ZDDP)-strengthened engine oil 461

Figure IV.6.4. Ce₃+ percentages in a combined Ce₃+ and Ce4+ for TWC-GPF samples using Ce L₃-edge X-ray absorption near-edge structure measurements: "ZDDP" was aged with ZDDP-strengthened engine oil, while... 462

Figure IV.6.5. H2-temperature programmed reduction results of lab-and field-aged TWC-GPFs 463

Figure IV.7.1. Properties of filter substrates studied (measured permeabilities in m² shown in parentheses) 466

Figure IV.7.2. Summary of initial filtration data for ultrafine particles across seven different filter substrates 467

Figure IV.7.3. Distributions of controlling pore throat diameters obtained by capillary flow porometry 467

Figure IV.7.4. Distributions of pore body sizes in pore network models generated for the seven filter substrates examined 468

Figure IV.7.5. Three-dimensional reconstruction of a small section of the C₂ material from X-ray CT data and an associated flow field (color key indicates local gas velocities in cm/s) 468

Figure IV.7.6. Distributions of streamline tortuosities in flow fields through the various filter media studied 469

Figure IV.8.1. (a) Powder X-ray diffraction plots of double salts synthesized by varying molar compositions of lithium chloride and magnesium chloride and (b) ammonia release profiles of corresponding... 472

Figure IV.8.2. TGA profile for ammonia release from MgCl2.6NH3 and corresponding material containing 10% CoCl₂ and NiCl₂, respectively 473

Figure IV.9.1. NOx adsorption and release on various zeolite supported Pd materials 476

Figure IV.9.2. Methane oxidation on supported Pd catalysts 477

Figure IV.9.3. PM number and concentration in the exhaust from diesel particulate filter 478

Figure IV.10.1. Impact of different SCO phases on the NO₂ out of the SCR reaction 482

Figure IV.10.2. Effect of SCO phase on standard (NO only) and fast (equimolar NO, NO₂) SCR performance 483

Figure IV.10.3. Effect of catalyst washcoat symmetry on SCR NOx reduction performance 484

Figure IV.10.4. Effect of passive soot oxidation on fast SCR reaction behavior 484

Figure IV.11.1. (a) Top view and (b) cross-section view SEM images of low-temperature processed TiO₂ nano-array. (c) Photograph of field-size Pt/TiO2 nano-array samples (5.66-in diameter × 3 in) for engine dynamometer testing 487

Figure IV.11.2. (a) DOC activity in LTC-D simulated exhaust, (b) X-ray diffraction patterns, and (c) transmission electron microscope and energy-dispersive X-ray spectroscope scanning transmission electron microscope elemental... 488

Figure IV.11.3. Aberration-corrected high-angle annular dark-field scanning transmission electron microscopy images of Pt/rutile TiO₂ nano-arrays: (a) fresh and (b) after hydrothermal aging at 700°C for 100 h... 489

Figure IV.11.4. DOC activity of ultra-low Pt loading on rutile TiO₂ nano-array in the CDC simulated exhaust. The addition of H₂ in the exhaust as an upstream promoter can significantly reduce the light-off temperature 490

Figure IV.11.5. Sulfur poisoning effects on CO and THC oxidation activity of Pt-Pd/TiO₂ nano-array (900 cpsi substrate, 70 g PGM/ft³, weight ratio of Pt/Pd = 3/5) 490

Figure V.1.1. Project schedule and phasing (Volvo) 494

Figure V.1.2. SuperTruck evolves into SuperTruck 2 (Volvo) 495

Figure V.1.3. System simulation plan for SuperTruck 2 concept selection (Volvo) 496

Figure V.2.1. Powertrain layout, highlighting the location and arrangement for the M/G system coupled to the rear facing power take off 502

Figure V.2.2. Heat release rate in response to increased injection rate. The lowest rate is the baseline. Increased peak rate corresponds to increased injection rate 503

Figure V.2.3. Powertrain mule milestones 504

Figure V.3.1. Results of LCHP kit evaluation. (a) BSFC comparison of baseline and LCHP engines at rated. (b) Friction power comparison of each component 506

Figure V.3.2. Comparison of DEGR and HP EGR 507

Figure V.3.3. Advanced fuel injection strategy simulation with CONVERGE 2.2 508

Figure V.3.4. ORC configurations for 55% BTE: (a) HT loop ORC, (b) LT loop ORC, (c) BTE gain projection 508

Figure V.3.5. Effects of oil and coolant temperatures on BSFC and exhaust energy distribution 509

Figure V.3.6. Efficiency breakdown and loss comparison between two gasoline fuels 509

Figure V.4.1. ST2 program phases 511

Figure V.4.2. ST2 proposed freight efficiency test cycle 512

Figure V.4.3. Detroit simulation and design process 513

Figure V.4.4. Detroit roadmap to reach the 55% BTE target 514

Figure V.4.5. DTNA roadmap to reach the 115% freight efficiency target 514

Figure V.5.1. Gen 3 GDCI powertrain 518

Figure V.5.2. BSFC as a function of BMEP for Gen 1, Gen 2, and Gen 3 GDCI engines 519

Figure V.5.3. Gen 3 gasoline direct injection injector spray and computational fluid dynamics simulation showing wetless injection process 519

Figure V.5.4. HC, CO emissions and combustion efficiency for GDCI early and late injection timings 520

Figure V.5.5. Smoke emissions as a function of injection dwell for Gen 3 gasoline direct-ignition injectors 520

Figure V.6.1. The Envera VCR engine 523

Figure V.6.2. Fuel consumption for an F-150 pickup truck on the EPA city (FTP75) and highway test cycle 524

Figure V.7.1. Sensitivity of two different spray designs to injection timing 528

Figure V.7.2. Sensitivity of flame turbulence and fuel/air ratio to dwell variation in a two-pulse injection strategy 528

Figure V.7.3. CFD mapping of EQR and burn rate comparing 2 ms and 0.5 ms dwell times 529

Figure V.7.4. Cycle fuel economy predictions based on weighted test points 529

Figure V.7.5. Critical hardware components 530

Figure V.8.1. Production-intent RF sensor including control unit and antennas 533

Figure V.8.2. RF catalyst cavity simulation results and comparison with experimental measurements 534

Figure V.8.3. Cavity resonance response of SCR samples in the presence and absence of stored ammonia 534

Figure V.8.4. Cavity resonance response of TWC samples in the presence and absence of stored oxygen 535

Figure V.8.5. Calibrated RF sensor output and correlation to estimated SCR ammonia storage levels (a) and RF antennas installed on full-size SCR for vehicle testing (b) 536

Figure V.8.6. Analysis of RF signal response to individual exhaust gas species 537

Figure V.9.1. Isuzu engine unmounted from the vehicle, mounted on the dynamometer 540

Figure V.9.2. Engine dynamometer test setup schematic 541

Figure V.9.3. EAVS unit 542

Figure V.9.4. eWHR unit 542

Figure V.9.5. System control coordinated between Eaton and Isuzu ECUs 543

Figure V.9.6. Control hardware 543

Figure V.9.7. Dynamometer setup with EAVS and EGR valve installation 544

Figure V.9.8. Engine torque, fuel, and NOx versus boost pressure at an engine speed of 2,000 rpm 544

Figure V.10.1. Effect of turning PCN flow on/off 547

Figure V.10.2. Fuel injection rate shape comparison 548

Figure V.10.3. Apparent heat release rate (AHRR) comparison 548

Figure V.10.4. Closed cycle efficiency comparison 548

Figure V.11.1. VW EA888 four-cylinder GT-POWER engine model 552

Figure V.11.2. VW Jetta GT-SUITE vehicle model 552

Figure V.11.3. Full valvetrain in-cylinder head packaging 553

Figure V.11.4. Probe test solenoid force vs. stroke comparison 554

Figure V.11.5. e-DEAC system mechanization 554

Figure VI.1.1. Micrographs of gray cast iron liner surfaces subjected to a scuffing load test at flow rates of (a) 0.1 mL/min, (b) 0.2 mL/min, and (c) 0.5 mL/min 558

Figure VI.1.2. Micrographs of surface of field-operated engine liner: (a) optical, (b) SEM, and (c) EDAX of tribofilm 559

Figure VI.1.3. TEM micrograph of tribofilm: (a) overview and (b) substrate-film interface 559

Figure VI.1.4. Tribofilm micrographs: (a) scanning transmission electron microscopy and (b) EDAX elemental mapping 560

Figure VI.1.5. SEM micrograph of (a) polishing and fatigue wear and (b) sub-surface structure 561

Figure VI.1.6. SEM micrograph of (a) local tribofilm removal, (b) plastic strain pattern on the surface, (c) sub-surface plastic deformation below scuffed surface, and (d) sub-surface structure with severe scuffing damage 561

Figure VI.2.1. Effect of MoDTC and OFM on friction for a steel-bronze and steel-Teflon contacts and impact of AWs on OFM for the steel-bronze contact 565

Figure VI.2.2. Friction and wear results of a steel-bronze contact lubricated by base oils containing a ZDDP, an IL, or a ZDDP+IL combination 566

Figure VI.2.3. While CB alone significantly increased the wear of both contact surfaces with little change in the wear ratio between the 52100 steel ball and M2 tool steel flat, CB+ZDDP together surprisingly... 567

Figure VI.2.4. The wear scars on the M2 (top) and A2 (bottom) flats lubricated by the PAO+CB+ZDDP (right) appeared much smoother compared with those lubricated by PAO+CB (left) 568

Figure VI.3.1. Average wear volume for each sample where the error bars represent the standard deviation. Three runs were averaged (UCM) 574

Figure VI.4.1. Kinematic viscosity for binary ester-SHC composite fluids 578

Figure VI.4.2. Friction and wear measurement under unidirectional and reciprocating sliding 579

Figure VI.4.3. Wear measurement under four-ball testing for binary ester-PAO4 composite fluids 579

Figure VI.4.4. Transmission electron micrographs from: (a) and (b) typical colloidal additives, and (c) typical fully formulated oil 580

Figure VI.4.5. (a) Transmission electron and (b) optical micrographs of ZnFe₂O₄ colloidal additives. (c) Profilometry of wear track of flat specimen tested with ZnFe₂O₄ colloidal additives 580

Figure VI.4.6. (Left) Coefficient of friction (COF) and (right) wear of a VN-Cu coated steel pin against steel flat during tests in a used oil (taken after 10,000 miles). Uncoated steel has slightly higher friction, and... 581

Figure VI.5.1. Relationship between PM organic and elemental carbon (OC, EC) for the three lubricants evaluated. Points are averages of three filter samples and error bars span the max and min values for each lubricant 585

Figure VI.5.2. Chromatogram of the thermal desorption step for PM collected under cold-start conditions from a GDI engine with three different viscosity lubricants. The mass spectroscopy signal has been... 585

Figure VI.5.3. Formation of CO₂ in WGS reaction between 200°C and 550°C for TWCs aged with different lubricant additives 586

Figure VI.5.4. Fuel economy improvement by test cycle for four vehicles with 5W-30 test lube compared to ASTM base lube 586

Figure VI.6.1. Pinion drive spin loss data for AW0704-A and 75W-140. Depicted is power loss (hp) at various vehicle speeds (mph) at multiple temperatures. Graph (a) presents data at 185°F, (b) at 165°F, and (c) at 145°F.... 590

Figure VI.6.2. Graphical display of XPS analytical data for post L-42 test gear tooth sections that completed testing in AW0704-R and AU6615-E fluids. Each bar represents three measurements and the error bars represent... 590

Figure VI.6.3. High speed load carrying capacity and shock load testing post-test hardware ratings per ASTM D7452. Inspection results for AW0704-A, AW0704-R, and AU6615-E. Passing requirements are indicated by the dashed lines 592

Figure VI.7.1. High porosity PTWA coating shows friction benefit under mixed lubrication regime 596

Figure VI.7.2. DLC-coated rings show additional friction benefit with high porosity PTWA coatings under boundary and mixed lubrication regimes 596

Figure VI.7.3. Nano-composite VN-Cu coated rings show additional friction benefit with high porosity PTWA coatings under boundary and mixed lubrication regimes 596

Figure VI.7.4. PTWA coated cylinder bores in a linerless engine block 597

Figure VI.7.5. Polyalkylene glycol (PAG) engine oil showed significant friction benefits over GF-5 5W-20 oil 597

Figure VI.7.6. Linerless engine block with PTWA coating shows friction benefits compared to cast iron liner engine block 598

Figure VI.8.1. Optical photos of (A) a 1.0 wt% freshly prepared dispersion and three 1.0 wt% dispersions of PC13-NP-9.7k in PAO after storage for 60 d at -15°C, room temperature (r.t.), and 100°C. (B) 1.0 wt% dispersion... 601

Figure VI.8.2. Friction curves for PAO SpectraSynTM 4 (i), PAO additized with 1.0 wt% PC18-NP-13.9k (ii), PC16-NP-8.3k (iii), PC13-NP-9.7k (iv), PC12-NP-9.5k (v), ZDDP (vi), and PC13-20.8k free polymer (vii). The tribological... 602

Figure VI.8.3. (A) Friction coefficient traces of the neat PAO base oil and PAO + 0.50 wt% oleyamine-Ag NP-MUA. (Inset) Wear rates corresponding to the tribotests. (B) Friction coefficient traces showed that organic-modified... 602

Figure VI.8.4. Finite element model of a rough surface sliding with a nanoparticle-additized lubricant. (a) Representative model geometry and (b) close up view of the deformation of interacting asperities and a nanoparticle 603

Figure VI.9.1. Comparison of friction modifiers using the Mini-Traction Machine at 50% slide-to-roll ratio 607

Figure VI.9.2. Schematic diagram of the direct soft mask fabrication process with only one ultraviolet irradiation 609

Figure VI.9.3. Conformal contact press jig for putting on soft mask on highly concave bearing surface 610

Figure VI.10.1. MD simulation results: interaction energy (a) and surface coverage (b) of alkyl-cyclen (dark solid-dot curves) and alkylamine (light open-dot curves) FM molecules at different temperatures 614

Figure VI.10.2. MSD of the alkyl-cyclen (solid-dot curves) and alkylamine (open-dot curves) FM molecules recorded during the MD simulation at 120ºC 615

Figure VI.10.3. (a) Variation of friction coefficients versus S/R ratio in the presence of different diblock VMs at 125ºC under 1 GPa contact pressure. (b) Variation of friction coefficients versus S/R ratio in the presence of different... 616

Figure VI.10.4. (a) Variation of viscosity ratio [η(mixture)/η(PAO4)] versus VM concentration at 100ºC. (b) Dynamic light scattering measurements of hydrodynamic diameters for different dilute VM-PAO4 oil solutions 617

Figure VI.10.5. Film thickness measurements at 125ºC at 0% (a) and 10% (b) S/R ratio 618

Figure VI.11.1. Statistical analysis of FE improvement on engine oils 623

Figure VII.1.1. Splitter plate dimensions and three-vehicle platoon in the Army 7 ft × 10 ft wind tunnel located at National Aeronautics and Space Administration Ames 626

Figure VII.1.2. Model instrumentation for body-axis force measurements and engine cooling air supply 627

Figure VII.1.3. Particle image velocimetry setup in the wind tunnel test section and a sample trailer wake flow field 628

Figure VII.1.4. Thermal images (top view) of a two-vehicle platoon with an 81-ft full-scale spacing at 0° and 9° yaw 628

Figure VII.1.5. Wind-averaged drag and pressure coefficient vs. vehicle spacing for a two-vehicle platoon with and without trailer boattails 629

Figure VII.1.6. (a) Percent benefit in CDwa for the entire two-vehicle platoon. (b) CPwa as a function of spacing for a two-vehicle platoon with and without trailer boattails 630

Figure VII.1.7. (a) CDwa for a 40-ft separation distance between the first and the second vehicle as a function of vehicle spacing for a three-vehicle platoon. (b) CDwa for the entire three-vehicle platoon as a function of vehicle spacing 630

Figure VII.1.8. (a) CPwa for a 40-ft separation distance between the first and the second vehicle as a function of vehicle spacing for a three-vehicle platoon. (b) CPwa for the third vehicle with multiple separation distances... 631

Figure VII.1.9. CDwa for a two-vehicle platoon (30-ft and 50-ft spacing) as a function of lateral misalignment 631

Figure VII.2.1. Response effects obtained on design of experiment study 635

Figure VII.3.1. Schematic images of organically modified GnPs 641

Figure VII.3.2. Experimental setup for the synthesis of SnFs 641

Figure VII.3.3. SEM images of synthesized silica fibers using the experimental setup in Figure VII.3.2. The diameter of the fibers is smaller than 100 nm 642

Figure VII.3.4. (a) Modulus values of the synthesized composites at a broad temperature range. (b) Tand values of the synthesized composites at a broad temperature range. The respective values of a commercial elastomer... 642

Figure VII.3.5. Transmission electron microscopy images of the (a) unfilled SBR, (b) SBR filled with GnPs, and (c) SBR filled with functionalized GnPs 643

Figure VII.3.6. SEM images of the (a) unfilled SBR, (b) SBR filled with GnPs, and (c) SBR filled with functionalized GnPs and SnFs 643

Figure VII.4.1. Filler vs. tan δ at 60°C, normalized to N330 646

Figure VII.4.2. Flex fatigue of different fillers 646

Figure VII.4.3. Ozone resistance by bent loop and crack growth 646

Figure VII.4.4. Discoloration test 647

Figure VII.4.5. 6-PPD concentration profile after cure and aging 647

Figure VII.5.1. ISX450 engine with UltraShift Plus AMT under test in the powertrain test cell art ORNL's VSI Laboratory 651

Figure VII.5.2. Distribution of Class 8 Trucks by On-Road Vehicle Weight 651

Figure VII.5.3. Engine operating area for the three high speed tests considered in this study: 65 mph cruise, 55 mph cruise, and HTDC 652

Figure VII.5.4. Fuel consumption for different cycles and vehicle weight 653

Figure VII.5.5. ORNL is beginning to work on modelling the new single supervisory controller using the offline validated model 654