Figure 1. Research areas within Advanced Combustion Engines and Fuels Program 59

Figure 2. Schematic of energy balance considered in the derivation of the new cavitation erosion metric (Som, I.6) 62

Figure 3. Impact of electrode geometry and electron seeding on low-temperature plasma discharge (Scarcelli, I.10) 63

Figure 4. Log plot of ignition delay versus ambient temperature for n-dodecane at Spray A conditions. Larger detailed mechanisms are able to capture the ignition delay at low... 64

Figure 5. Back-illumination images of sprays with different ethanol-iso-octane blend ratios (top row labels: E0 is pure iso-octane, E30 is 30-70 ethanol-iso-octane blend, and... 65

Figure 6. Measured particle size distributions for ACI combustion (Kokjohn, I.20) 65

Figure 7. Fuel properties impacting boosted SI engine efficiency (Farrell, II.1) 66

Figure 8. Knock-limited combustion phasing as a function of octane index for each fuel at the ACI condition (Szybist, II.3) 67

Figure 9. A comparison between two different modeling approaches to spray formation (Arienti, II.8) 69

Figure 10. Mean apparent heat release rate plotted for intake temperatures from 40-180°C for 2,000r/min operation; arrow denotes spark timing, shaded region denotes one... 70

Figure 11. Natural luminosity image from experiment with single-duct holder confirms that conventional diesel combustion spray produces significantly more incandescence from... 71

Figure 12. Molecular-level solution structure and Reid vapor pressure (RVP). The average number of molecules in clusters was determined by using nuclear magnetic resonance... 72

Figure 13. Prediction of RON and octane sensitivity (OS) for the simulated blending of nine high-performance fuels into the Co-Optima core fuels (Pitz, II.26) 73

Figure 14. Demonstration of simultaneous two-color pyrometry, rainbow schlieren defectometry, and OH* chemiluminescence measurements using a simple Bunsen burner... 74

Figure 15. Cut-plane equivalence ratio distribution and temperature distribution from the CFD simulations of single-fuel reactivity-controlled compression ignition combustion with... 75

Figure 16. Custom high-efficiency research engine with 1:5:1 stroke-to-bore ratio and high compression ratio (Szybist, III.5) 76

Figure 17. CO oxidation light-off performance showing excellent low-temperature activity and stability of the hydrothermally treated Pt/CeO₂ catalyst under exhaust conditions... 77

Figure 18. SCR-onset conversion infections for (a) a commercial Cu-CHA (chazabite) and (b) a model Cu-Beta SCR catalyst (Partridge, IV.5) 78

Figure 19. Pathway toward improved NOx reduction performance by a surface-active NOx species (Rappe, IV.8) 79

Figure 20. Baseline Model Year 2009 vehicle (left) and VEV3 text mule (right) at rest stop during fuel economy test (Amar, V.1) 79

Figure 21. Complete dynamic skip fire cylinder head assembly on test stand (Fernandez, V.10) 81

Figure 22. Scanning electron micrograph images of the abraded surface of (a) unfilled styrene-butadiene-styrene and (b) styrene-butadiene-styrene-graphene oxide samples after... 82

Figure I.1.1. Combustion image velocimetry results show the evolution of the azimuthally averaged radial component of flow (red/blue false color strips; note the piston profile on the... 86

Figure I.1.2. CFD results shown on a vertical cutting plane containing one spray axis. False color represents the fuel-air equivalence ratio; the black contour represents the stoichiometric... 87

Figure I.1.3. Components of the radial acceleration equation visualized on a vertical cutting plane containing a spray axis. Bottom: the radial acceleration; blue colors represent acceleration... 88

Figure I.2.1. Optical engine schematic showing the orientation of the PAH-PLIF and soot-PLII laser sheets in the combustion chamber and the orientation of the cameras and beamsplitter... 91

Figure I.2.2. Representative instantaneous images of simultaneous 633-nm PAH-PLIF (false-colored green) and soot-PLII (red) at an elevation of 15mm below the fredeck for single... 92

Figure I.2.3. Cylinder pressure (top) and injection schedules (bottom) for CC or LD double-injection conditions at 10% intake O₂. Single-injection pressure data using only the first injection... 93

Figure I.2.4. Representative instantaneous images of simultaneous 633-nm PAH-PLIF and soot-PLII at three laser-sheet elevations for the CC condition (top three sets of images, each... 94

Figure I.2.5. Ensemble-averaged images of simultaneous PAH-PLIF for excitation at 355nm (false-colored blue), 532nm (green), and 633nm (red) and soot-PLII (grayscale) at three... 95

Figure I.2.6. Predictions of multi-dimensional computational fluid dynamics simulations of the residual jet of the first injection of a double-injection condition with either negative ignition-dwell... 96

Figure I.3.1. (left) Normalized liquid-length and (right) lift-off length measurements for surrogate fuels at different temperatures. Spray A conditions, but with 80MPa fuel injection... 101

Figure I.3.2. (left) Extinction measurement at various times after the start of injection. (right) Tomographic reconstruction of extinction measurement for fuel mass fraction, compared... 101

Figure I.3.3. Extinction imaging of Spray G (Spray G injector and ambient conditions) with three different injection durations. Blue dashed lines show the line-of-sight projection of the... 102

Figure I.3.4. Measured plume direction (defined at right) for different injection durations or with multiple injections. Same conditions as Figure I.3.3 103

Figure I.4.1. A comparison of temperature-map images of the thermal stratification in a vertical cut-plane for CONVERGE-LES simulations of the all-metal engine, which has a shallow,... 107

Figure I.4.2. Comparison of CHEMKIN single-zone internal combustion engine results with experimental data for well pre-mixed LTGC engine operation (i.e., HCCI). Figure I.4.2a shows... 109

Figure I.4.3. A comparison of the f-sensitivity of the proposed fuel blend given at the top of the plot (Blend 4) with that of RD5-87A (regular E10 gasoline) for intake pressures from... 110

Figure I.4.4. A series of cylinder-pressure traces that show how the AMFI system can adjust the combustion phasing from retarded to advanced by increasing the amount of EHN additive... 110

Figure I.4.5. Demonstration of the ability of the AMFI system to control CA50 through a load sweep. (a) Required changes in CA50 with load change and the change in the amount of... 111

Figure I.5.1. Schlieren image of post-discharge transient plasma streamers for a 19.2kV discharge in 2.0 bar ambient air along with an image of O atom distributions from complementary... 116

Figure I.5.2. Image of the groundless partial DBD electrode and complementary filtered imaging of excited-state O from resultant negative coronas that surround the insulator during... 116

Figure I.5.3. Sequence of images showing ignition kernel development for pin-to-pin and groundless partial DBD TPI as compared to inductive spark ignition for different voltages and... 117

Figure I.5.4. Comparison of ISFC as a function of load for lean conditions with and without O₃ addition for engine speeds between 800rpm and 1,400rpm 118

Figure I.5.5. Plots of IMEP, CA50, and coefficient of variation of IMEP as a function of spark timing for a sweep of intake temperatures (42-80°C) and a fixed 1,000rpm engine speed 119

Figure I.5.6. Comparison of measured in-cylinder O₃ concentration and early-stage apparent heat release rate profiles for a sweep of intake temperatures 120

Figure I.6.1. Schematic of energy balance considered in the derivation of the new cavitation erosion metric, which considers the transfer and storage of energy in the solid material... 124

Figure I.6.2. Comparison of erosion patterns between experimental images from Skoda et al. (top) and homogeneous relaxation model predictions of local maximum pressure (middle)... 125

Figure I.6.3. (a) Ignition delay and (b) flame liftoff as a function of ambient temperature conditions with LES models using ANNs. (c) Memory consumption over all nodes for a conventional... 126

Figure I.6.4. Temporal evolution of CH₂O mass fraction contours predicted by the TFM-ANN LES model (right) is compared against the CH₂O PLIF data (left) from a single shot injection 126

Figure I.7.1. Near-nozzle spray surface area in ECN Spray C (a) and Spray D (b) injectors measured using ultra-small-angle X-ray scattering 130

Figure I.7.2. Near-nozzle spray surface area in ECN Spray D as a function of transverse position, shown for several distances from the injector nozzle 131

Figure I.7.3. A view of the outside surface and hole counter bores of an ECN Spray G injector 131

Figure I.7.4. Measurements of the ECN Spray C injector, including nozzle geometry (left), high-speed X-ray imaging of cavitation inside the nozzle (center), and a cross-section of... 132

Figure I.8.1. Calculated normalized HRRs for PRF90/O₂/diluent mixture at Tc = 735K, presented as a function of aHR, illustrating exothermic trends across a range of Pc for (a)... 138

Figure I.9.1. Results of pilot injection quantity sweep at constant 90dB combustion noise 144

Figure I.9.2. Overview of single-, double-, and triple-injection strategy injection timings and combustion phasings 144

Figure I.9.3. Comparison of single-, double-, and triple-injection results with constant 90dB combustion noise 145

Figure I.9.4. Effects of EGR on combustion and emissions while maintaining constant 90dB combustion noise 145

Figure I.10.1. Impact of electrode geometry and electron seeding on LTP discharge 149

Figure I.10.2. Comparison between simulations and experiments in terms of post-discharge plasma regime 149

Figure I.10.3. VizGlow results highlighting the distribution of atomic oxygen (O) and temperature 150

Figure I.10.4. Impact of number of pulses on LTP ignition as calculated using CONVERGE 151

Figure I.10.5. CONVERGE results showing the impact of number of pulses on the flame kernel growth 151

Figure I.11.1. 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... 154

Figure I.11.2. Brake thermal efficiency as a function of catalyst inlet O₂ concentration and Φcatalyst at 2,000rpm and 4 bar BMEP 155

Figure I.11.3. Axial temperature profile of reforming catalyst at 2,000rpm, 4 bar BMEP over a range of catalyst inlet O₂ concentrations and catalyst Φ conditions 155

Figure I.11.4. Schematic of two possibilities of reaction zones in reforming catalyst. In scheme (a), the oxidation and steam reforming occur sequentially. In scheme (b), the partial... 156

Figure I.11.5. Axial temperature profile for increasing engine load, from 4 bar to 10 bar BMEP, at an engine speed of 2,000rpm. Catalyst inlet O₂ was held constant at 1.8%, and... 157

Figure I.11.6. Exhaust temperature, intake manifold pressure, BMEP, and intake manifold pressure as a function of load for the reforming strategy and the baseline engine operation 157

Figure I.12.1. (a) Mass-attenuation coefficients versus atomic number, and (b) schematic of neutron-imaging apparatus 159

Figure I.12.2. System used to study intra-nozzle dynamics of fuel injection include (a) high-pressure fuel delivery system and (b) the aluminum spray chamber with optical viewport... 160

Figure I.12.3. Normalized neutron radiographs of fuel injection from a single-hole GDI-style injector (ECN Spray G) highlight ability to see opening and closing of the check ball as well as... 161

Figure I.12.4. (a) Approach for fitting normalized image data to analytical path length model. (b) Needle displacement results show oscillation during injection and large movement after... 162

Figure I.12.5. Slices of neutron and X-ray CT reconstructions of large-bore Bosch injector performed in collaboration with Argonne National Laboratory 163

Figure I.12.6. Soot cake thickness down GPF channel length extracted from neutron CT scans shows continuous decrease with regeneration percent with E0 particulate matter (left)... 163

Figure I.13.1. Comparison of old and new trimethyl benzene (red) and ortho-xylene (blue) LLNL kinetic model simulations with ignition delay times (IDTs) measured in the ANL RCM... 166

Figure I.13.2. Evolution of major PAHs during the pyrolysis of toluene primary reference fuel 97.5 (n-heptane/iso-octane/toluene = 14.5/8.0/77.5, by mole, respectively, 97.5 research... 167

Figure I.13.3. Comparison of LLNL gasoline surrogate model results with IDTs measured in the Stanford high-pressure shock tube for a high-octane, moderate sensitivity (RON = 101,... 168

Figure I.14.1. Log plot of ignition delay versus ambient temperature for n-dodecane at Spray A conditions. Larger detailed mechanisms are able to capture the ignition delay at low... 171

Figure I.14.2. Total simulation time for Spray A simulation with detailed reaction model, colored by time spent in chemistry and non-chemistry routines 172

Figure I.15.1. Four-valve DISI engine: (a) turbulent structures shown by magnitude of vorticity (1/s) during intake; (b) pressure rise as a function of crank angle as compared to... 178

Figure I.15.2. The ECN Spray A case: (a) injection of diesel in quiescent nitrogen at 2.2MPa, KH-RT spray model; (b) the penetration depth of the spray compared to ECN experimental... 178

Figure I.15.3. 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... 179

Figure I.15.4. FEARCE's super-linear algorithm scaling versus the ideal scaling curve 180

Figure I.15.5. FEARCE's beam-warming (BW) system versus use of Trilinos Multigrid preconditioned GMRES, a weak scaling study 180

Figure I.16.1. Model transitioned from cylinder sector to full-cylinder geometry with positionable intake swirl flap for improved simulation of charge motion and mixing 185

Figure I.16.2. Simulation results from full-cylinder model show improved local mixing, resulting in increased oxidation of CO and soot late in the cycle 185

Figure I.16.3. For the full-cylinder model, improved charge motion and detailed piston geometry with valve cut-outs results in earlier and faster combustion within the squish region,... 185

Figure I.16.4. Comparison of predictions for key combustion metrics and engine-out emissions 186

Figure I.16.5. Flow chart for iterative strategy used with CHT model 187

Figure I.16.6. Initial results with CHT model predict higher combustion chamber wall temperatures than assumed values used in simulations with uniform, constant wall temperatures,... 187

Figure I.17.1. Time-resolved (at intervals of two crank angle degrees of rotation) measured IR spectral radiative intensities in the transparent combustion chamber engine at 40kPa... 191

Figure I.17.2. Comparison of measured and simulated spectral intensities for the transparent combustion chamber engine operating at 40kPA manifold absolute pressure, 1,300r/min... 191

Figure I.17.3. AHRR of the knocking operating condition for each temperature and three diluents at 10%: (a) N₂, (b) H₂O, and (c) CO₂. All AHRR plots represent a single engine... 192

Figure I.17.4. Computed root-mean-square temperature profiles for a fully developed turbulent channel flow between parallel plates held at different fixed temperatures, with and without... 193

Figure I.18.1. Back-illumination images of sprays with different ethanol-iso-octane blend ratios (top row labels: E0 is pure iso-octane, E30 is 30-70 ethanol-iso-octane blend, and E100... 197

Figure I.18.2. Time-resolved liquid penetration lengths for ethanol-iso-octane fuel blend sprays (E100 is pure ethanol, E30 is 30% ethanol plus 70% iso-octane, and E0 is pure iso-octane)... 197

Figure I.18.3. Droplet distribution curves for ethanol-iso-octane fuel blend sprays (E100 is pure ethanol, E30 is 30% ethanol plus 70% iso-octane, and E0 is pure iso-octane) at chamber... 198

Figure I.18.4. Graphical visualization of the simulated spray result (left) and ethanol mass fraction profile on the plume centerline (right) at 1ms after start of injection time for ethanol... 198

Figure I.18.5. Iso-octane injection at Pambient = 50kPa. Experimental spray image (left), simulated spray (center), and fuel vapor mass fraction profile (right) at 1ms after start of... 199

Figure I.19.1. (a) n-Heptane film thickness at Pinj = 150MPa, (b) n-heptane film thickness at density = 22.8kg/m³, and (c) Sauter mean diameter of diesel spray at Pinj = 150MPa and... 202

Figure I.19.2. Splashing criteria of diesel, water, n-dodecane, n-heptane, and diesel 202

Figure I.19.3. Gas velocity fields and spray shape for the hydro-dynamically smooth plate (left figures) and rough plate (right figures) 203

Figure I.19.4. Experiments vs. CFD comparison of liquid penetration (left), rebound radii (center), and rebound heights (right) 204

Figure I.19.5. Splashed mass ratio results for simulations at u = 15, 16, 17, and 24. Comparison of the splashed mass ratios at u = 15, 16, and 17 shows the transition to splashing... 204

Figure I.19.6. Time-averaged secondary droplet characterization of the 80CPD diesel droplet train impingement simulation with a pre-existing liquid film at the nondimensional velocity... 206

Figure I.20.1. Comparisons of measured and predicted particle size distributions for mixing controlled compression ignition combustion 209

Figure I.20.2. Measured particle size distributions for advanced compression ignition combustion 210

Figure I.20.3. (left) Simulated and (right) measured particle size distributions for neat diesel fuel and 52% syngas 210

Figure I.20.4. Comparisons of measured and predicted PSDs and representative simulated soot aggregates for several different sized particles 211

Figure I.20.5. Ensemble-averaged 633-nm PAH PLIF and soot PLII images for the long dwell operating condition 212

Figure I.21.1. Detection method illustrating liquid, two-phase, and vapor regions of the spray based on RSD measurements from 95 injections: (a) instantaneous RSD image showing dark... 216

Figure I.21.2. Gas-phase adiabatic mixing of fuel and air as a function of fuel molar mixture fraction using real-fluid properties and real-fluid mixing models for Engine Combustion Network... 217

Figure I.21.3. Relationship between the mixture's normalized refractive index and equivalence ratio and temperature for the Engine Combustion Network Spray A conditions 217

Figure I.21.4. Density contours for n-dodecane/nitrogen mixtures at various fuel mole fractions 218

Figure I.21.5. Comparison of simulation and experimental (left) vapor and liquid penetration results and (right) ignition delay behavior for n-heptane injected at 1,000 bar, 363K into air... 219

Figure I.21.6. Comparison of experimental RSD images (top in each pair) and synthetic RSD images (bottom in each pair) processed from CFD results for n-heptane injected at 1,000 bar,... 219

Figure I.22.1. (left) Multi-cycle LES using FPF reproduces CCV of in-cylinder pressure in an engine experiment under a non-knocking condition (colored thick lines: 15 LES cycles, black... 222

Figure I.22.2. Comparison of the flame front evolution in fast and slow cycles in LES. Also shown is the subfilter kinetic energy at top dead center. In the fast cycle, where the flame... 222

Figure I.22.3. Verification of the coupling of CMC with CONVERGE CFD. The evolution of species mass fractions is linear both in time and in scalar (conditioning variable) space. The... 224

Figure I.22.4. Test of CMC-FPF-CONVERGE in a simple configuration. A premixed flame kernel is deposited initially at the center of the computational domain and grows in homogenous... 224

Figure I.23.1. Comparisons of terminal bubble shape for a range of parameter space from SPH (BU), VOF (BU), front tracking (Hua), and experiments (Bhaga) 228

Figure I.23.2. The differences between VOF and SPH simulations for low Reynolds numbers 228

Figure I.23.3. Time-averaged velocity field from numerical simulations of nozzle flow for varying cylindrical outlet diameters 229

Figure I.23.4. Normalized neutron radiographs comparing the max-cavitating (flash boiling) and non-cavitating (non-flash) conditions in the single-hole, large-bore injector 230

Figure I.23.5. Normalized neutron intensity in sac region for non-flash and flash boiling conditions. Shaded bands represent one standard deviation 230

Figure I.24.1. Predicted path-integrated SMD along the centerline of ECN Spray D (dnozz = 180mm, dodecane fuel) using the newly developed KH-Faeth model. Comparison is shown... 234

Figure I.24.2. Near-nozzle surface area measurements for ECN Spray D at several distances from the nozzle exit (x). Pinj = 1,500 bar, Pamb = 1 bar (left) and 20 bar (right) 235

Figure I.24.3. SAMR measurements of path-integrated SMD in ECN Spray D at three different viewing angles. Measurements are at an axial position of 10mm from the nozzle exit... 235

Figure II.1.1. Fuel properties impacting boosted SI engine efficiency 239

Figure II.2.1. Projected volumetric fuel economy improvements offered by Co-Optima Tier 3 blendstocks 248

Figure II.3.1. Pressure-temperature trajectories for the five engine operating conditions investigated for three of the Co-Optima core fuels 251

Figure II.3.2. Knock-limited combustion phasing as a function of octane index for each fuel investigated at (a) the boosted SI condition, (b) the RON-like condition, and (c) the MON-like... 253

Figure II.3.3. Knock-limited combustion phasing as a function of octane index for each fuel at the ACI condition 253

Figure II.3.4. Constant-volume ignition delay in milliseconds, calculated from kinetic modeling, as a function of pressure and temperature for each of the five fuels investigated at... 254

Figure II.3.5. Ignition delay differences between the alkylate fuel and either aromatic, E30, or the Tier III fuel at stoichiometric conditions (F = 1.0). Red areas indicate that the alkylate... 255

Figure II.4.1. SI merit function for fuel properties developed in Co-Optima 258

Figure II.4.2. Structure, RON, and S for the fuels investigated in this study 259

Figure II.4.3. RON of 20vol% blends of the fuel components investigated, as well as lines of linear blending for the BOB and each of the pure components. Amylene and prenol exhibit... 259

Figure II.4.4. S of 20vol% blends of the fuel components investigated, as well as lines of linear blending for the BOB and each of the pure components. Amylene and prenol exhibit... 260

Figure II.4.5. Pressure-temperature trajectories of each of the operating conditions investigated 261

Figure II.4.6. Knock-limited combustion phasing as a function of octane index for each fuel investigated at (a) the boosted SI condition, (b) the RON-like condition, and (c) the MON-like... 261

Figure II.5.1. For naturally aspirated stratified-charge, direct-injection SI operation at 1,000rpm, smoke emissions for E30 are much higher than the average trend line due to the... 264

Figure II.5.2. Effect of engine coolant temperature on piston-top wall wetting and associated formation of sooting pool fires and exhaust smoke emissions. Intake [O₂] = 18% 265

Figure II.5.3. Schematic of diffused back-illumination setup for in-cylinder soot quantification 265

Figure II.5.4. Detection of in-cylinder soot for stratified operation with a RON98 fuel containing 19.6% DIB by volume 266

Figure II.5.5. Effect of intake oxygen mole fraction on combustion phasing rendering "trace autoignition" for lean SI operation with Φm = 0.50 in the end-gas 267

Figure II.5.6. Example of the application of the octane-index framework for lean mixed-mode combustion utilizing end-gas autoignition. Pin = 100kPa 267

Figure II.6.1. (left) Cross-section of continuous-flow heated spray chamber, with capabilities depicted; (right) spray chamber installed in laboratory on optical table and behind operator... 271

Figure II.6.2. Schematic illustrating the geometry of Spray G and the process of plume interaction and spray collapse 272

Figure II.6.3. (Top) LVF simulations at axial distance cut plane of z = 15mm. Middle is with injector at 0° rotation; left at 22.5° rotation. Right is tomographic reconstruction at 0° rotation... 272

Figure II.6.4. Time sequence of schlieren images from the same injection. Spray G fuel injector with iso-octane fuel and 0.8ms injection duration. Injector is oriented at 0° rotation in... 273

Figure II.7.1. 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, 1,200rpm. For the scale... 277

Figure II.7.2. 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 high-cycloalkane, high-aromatic, and E30 fuels, and... 278

Figure II.7.3. Comparisons of CA50 as a function of TBDC for computational results using the S1 and S2 surrogates with experimental data at Pin = 1.0 bar for the E30, high-cycloalkane,... 279

Figure II.7.4. Comparison of the TBDC values required for simulations with the S1 (dotted lines) and S2 (dashed lines) surrogates to match the CA50 values of the experimental data... 280

Figure II.8.1. Cavitation interaction with time-accurate, resolved, liquid surface dynamics. Shown from left to right: a schematic of the experiment; three snapshots at increasing... 285

Figure II.8.2. Spray G geometry seen through a mid-plane cut. The mesh used in the CFD simulation is a blend between the high-resolution tomography of the holes (1.7μm) and... 285

Figure II.8.3. A comparison between two different modeling approaches to spray formation. Left: snapshot of the liquid surface, colored by temperature, from the time-resolved CLSVOF... 286

Figure II.8.4. Plot of vapor quality (non-dimensional units) from a data slice through one of the orifices in the simulation with BOB4. The black line marks the intersection of the liquid... 287

Figure II.9.1. Multi-mode concept showing ACI at part-load with engine running in SI mode for low loads and near idle and at higher engine loads and speeds 290

Figure II.9.2. Schematic of single-cylinder engine (left) and picture of engine installed (right) 291

Figure II.9.3. A single-zone temperature model is used to present experimental results of partially stratified ACI with gasoline-range primary reference fuels (PRFs), highlighting areas... 291

Figure II.9.4. Examples of range of conditions considered across ACI range and locations examined. The left figure shows a smaller ACI range and the different load locations (from... 292

Figure II.9.5. Results showing diminishing returns as ACI range is increased with different ACI efficiencies 292

Figure II.10.1. Engine oil pressure reduction as a function of LSPI segment for various engine loads and injector orientations; the reduced engine load conditions required increased... 296

Figure II.10.2. Recorded LSPI events per segment for matched load, varied injector orientation operation 297

Figure II.10.3. Oil pressure reduction for different engine loads and fuel distillation (left) and corresponding LSPI event counts (right) 297

Figure II.10.4. Lubricant additive metal content (left) and corresponding LSPI number count, color coded to lubricant (right) 298

Figure II.10.5. Oil pressure reduction for different engine loads and fuels (left) and corresponding LSPI event counts (right) 299

Figure II.11.1. Mean apparent heat release rate plotted for intake temperatures of 40-180°C for 2,000r/min operation; arrow denotes spark timing, shaded region denotes one standard... 303

Figure II.11.2. Quantified trends in PSHR and CA50 phasing for 2,000r/min operation (black, circle markers) and 1,000r/min operation (grey, diamond markers) as a function of intake... 304

Figure II.11.3. Quantified trends in KLSA CA50 and bulk gas temperature at CA2 of the deflagration as functions of PSHR for 2,000r/min operation (black, circle markers) and 1,000r/min... 304

Figure II.11.4. Pressure-temperature trajectory up to 2% of heat release (CA2) of the deflagration (blue star marker); spark discharge timing denoted (red circle markers) for an intake... 305

Figure II.12.1. Knock intensity as a function of intake air temperature at 1,500rpm, 9 bar net IMEP, and constant CA50 (crank angle at 50% mass fraction burned) of 20 crank angle degrees... 309

Figure II.12.2. Intake air temperature requirement as a function of intake manifold pressure at 12.6:1 and 15.3:1 compression ratio. Combustion phasing maintained at CA50 = 12 CAD... 310

Figure II.12.3. Intake air temperature change for knock-limited SI operation at 9 bar net IMEP and ACI operation at 3 bar net IMEP 311

Figure II.12.4. Motored compression trajectories during ACI operation for A30 and O30 test fuels 311

Figure II.13.1. The projected density distribution in sprays emerging from an eight-hole GDI injector. At left is a non-evaporating spray of gasoline-type calibration fluid; at right is a spray... 314

Figure II.13.2. Synchrotron X-ray radiography measurements of iso-octane injection. The measurements were taken 2mm from the fuel injection nozzle (left). Under conditions designed... 315

Figure II.13.3. Plot showing slices through the density distributions 1mm downstream of the fuel injector for three fuels: iso-octane, iso-octane with 20% butanol, and iso-octane with... 315

Figure II.14.1. Ethanol and isobutanol effects on the required base fuel RON to obtain constant RON 98 of the fInal fuel blends and the effects on the cylinder-pressure-transducer-based... 319

Figure II.14.2. Effective octane ratings of non-PRF fuels between their peak knocking lambdas and stoichiometry based on the CFR knockmeter knock intensities (A and B) and the cylinder... 320

Figure II.14.3. Three PRF 90 test conditions with the standard knockmeter knock intensity, using variations to intake temperature (Tin) and pressure (Pin) 321

Figure II.15.1. Isopleths of RCM-measured and chemical kinetically modeled ignition delay times under static (i.e., constant volume) conditions compared against isopleths of ACI engine... 326

Figure II.15.2. Comparison of relative fuel reactivities where static conditions (in RCM and model) are ranked against engine-intake requirements at Φ = 0.38. Baseline fuel is ALK. Intake... 327

Figure II.15.3. Comparison of RCM-measured and model-computed HRRs (normalized by mixture lower heating value), plotted as functions of accumulated heat release. Experiments... 328

Figure II.15.4. Normalized phi-sensitivity vs. temperature for RD5-87 measured under two pressure protocols. Model results utilize a multi-component surrogate for the E10 full... 329

Figure II.16.1. Schematic of the DFI concept on one fuel spray within an engine 332

Figure II.16.2. Single-duct holder (left) and two-duct holder (right) attached to cylinder head. All ducts have an inside diameter, length, and standoff distance from the injector orifice... 332

Figure II.16.3. Natural luminosity image from experiment with single-duct holder. Camera is viewing the cylinder head through a window in the piston (compare to left side of Figure II.16.2)... 333

Figure II.16.4. CDC results show trade-off between soot and NOx emissions as dilution increases, whereas DFI breaks the soot/NOx trade-off by simultaneously attenuating soot and... 334

Figure II.16.5. DFI dramatically lowers soot at constant dilution and combustion phasing. Plot shows the change in indicated-specific emissions of soot, NOx, hydrocarbons, and carbon... 334

Figure II.17.1. Example autoignition products of isooctane: reactor versus model results 337

Figure II.17.2. Autoignition products of ethanol: model versus experimental data 338

Figure II.17.3. Autoignition products from dimethylfuran: model versus experimental data 339

Figure II.17.4. Retro-Diels-Alder pathways for the three isomers of methyl cyclohexenes 340

Figure II.17.5. Experimental data for methyl cyclohexene isomers and formation of the soot precursor cyclopentadiene 340

Figure II.18.1. Arrhenius plot of ignition delay (log scale) versus inverse temperature (1,000/K) for iso-octane at 10 bar initial pressure in the AFIDA. The AFIDA simulations with full... 344

Figure II.18.2. Correlation of RON predicted from AFIDA ignition delay to RON measured by CFR engine method. AFIDA-predicted RON vs. RON measured on the CFR engine, showing... 345

Figure II.19.1. Raw mass spectrometer ion current data for several species: (a) E0-10% cumene blends and (b) E30-10% cumene blends in FACE B gasoline 351

Figure II.19.2. (a) Predicted time-dependent droplet diameter, (b) droplet temperature, and (c) liquid aromatic additive mass during evaporation of droplets of FACE B blends with... 353

Figure II.19.3. PN concentration as a function of particle diameter at Condition A for fuels blended with aromatics at 20vol%. Error bars are 95% confidence intervals 354

Figure II.19.4. Comparison of particle size distributions between Conditions A and B for the 20vol% 4-t-butyl toluene blends. Error bars are 95% confidence intervals 354

Figure II.19.5. Results of the linear model using optimal combined explanatory variables determined through regularized regression. Large dots indicate the mean experimental PM... 356

Figure II.20.1. Ignition delay time calculation using FACE A multi-component surrogate, PRF surrogate, and TPRF surrogate 362

Figure II.21.1. Tradeoff curves for bi-objective optimization with uncertainty in fuel component cost (synthetic cost data used to only show capabilities) 365

Figure II.21.2. Tradeoff curves for data-informed GP bi-objective optimization 366

Figure II.22.1. Molecular-level solution structure and Reid vapor pressure. The average number of molecules in clusters was determined by using NMR diffusion measurements in (a)... 369

Figure II.22.2. Hydrogen-bond clustering of alcohols in n-heptane determined by molecular dynamics simulations. Left: Number of hydrogen bonds per alcohol with increasing alcohol... 369

Figure II.22.3. Comparison of the solid-liquid equilibria for mixtures of diesel surrogate fuel V0b with complex diesel fuels, CFA or GTL1. Volume percentages of diesel surrogate fuel... 370

Figure II.23.1. T50 and T90 of surrogate BOB (baseline); 10%, 20%, and 30% ethanol blended into the BOB; unblended (100%) ethanol; 10%, 20%, and 30% isobutanol blended into... 376

Figure II.23.2. Comparison of CO light-off temperatures (T50 and T90) over the hydrothermally aged commercial TWC for all the fuel blends investigated. Error bars represent 95%... 376

Figure II.23.3. Comparison of NOx light-off temperatures (T50 and T90) over the hydrothermally aged commercial TWC for all the fuel blends investigated. Error bars represent 95%... 377

Figure II.24.1. (a) Explains the labeling format for this report in reference to the different conditions studied. The shape indicates which fuel, the shading indicates the % EGR, the... 381

Figure II.24.2. (a) Plot comparison of PM mass rate for each condition tested. PM mass rates were calculated from both gravimetric (black lines) and EC/OC (gray lines) filter analyses... 381

Figure II.24.3. Stacked bar graphs in (a) compare C1 HC emissions in order of increasing air-fuel stratification. The open bars represent the hydrocarbons measured by a FID, and... 382

Figure II.24.4. Size distribution plots for particulate flux at each condition tested. (a) RON 87 fuel and (b) RON 70 fuel from EEPS real-time exhaust sampling downstream of 2-stage... 383

Figure II.24.5. The particle flx (a) sum of all particle sizes measured by EEPS at each engine condition. The stacked bar graph (b) shows the fraction distribution of the total particle... 383

Figure II.25.1. Experimental (symbols) and simulated (lines) ignition delay times of cyclopentanone in air at Φ= 0.5, 1.0, and 2.0, and P = 15 bar. Solid and open symbols are shock... 387

Figure II.25.2. Experimental (symbols) and simulated (lines) ignition delay times of cyclopentanone in air at Φ= 0.5, 1.0, and 2.0, and P = 30 bar. Solid and open symbols are shock... 387

Figure II.25.3. Experimental and simulated pressure histories from the autoignition of an E10 research-grade 87 anti-knock index gasoline at various pressures at the end of compression... 388

Figure II.26.1. Prediction of RON and OS for the simulated blending of nine HPFs into the Co-Optima core fuels 392

Figure II.26.2. Predicted (curves) and measured (symbols) laminar flame speeds of cyclopentanone at different fuel-air equivalence ratios and unburned gas temperatures. Experiments:... 393

Figure II.26.3. Predicted (curves) and measured (symbols) laminar flame speeds of methyl acetate and ethyl acetate at different fuel-air equivalence ratios and unburned gas temperatures... 393

Figure II.27.1. Composition of the four-component Co-Optima BOB and the five virtual BOBs that maximize the volume fraction of each of the PIONA classes. The octane numbers... 397

Figure II.27.2. Variation of the merit score across the four-component Co-Optima BOB and the five virtual BOBs as a function of blending level of four blendstocks: (a) ethanol, (b)... 398

Figure II.28.1. (left) CFR engine geometry (red: knockmeter port cavity; green: intake valve with 180° shroud; blue: spark plug and cavity). (right) Temporal evolution of in-cylinder... 403

Figure II.28.2. Contour plots of OH and CH₂O mass fractions and local pressure difference on a horizontal cut plane passing through the spark plug electrode, at knock onset. Intake... 403

Figure II.28.3. (left) Knock intensity as a function of spark timing. Grey dot: knock intensity from each cycle CFD result with Mach CFL 50; red dot: maximum knock intensity at each... 405

Figure II.28.4. (left) Maximum knock intensity for different HoV values. KLSA predicted by the slope change point is highlighted by red dot. (right) Relative difference in BSFC for two... 405

Figure II.29.1. Updated two-color pyrometer design illustrating equal path lengths and equal number of components (left) and sample images of a sooting n-heptane flame with the... 409

Figure II.29.2. Computer-aided drafting illustration of how the three diagnostics are set up in relation to the CPFR 409

Figure II.29.3. Demonstration of simultaneous two-color pyrometry, rainbow schlieren defectometry, and OH* chemiluminescence measurements using a simple Bunsen burner 410

Figure II.29.4. New high-pressure air compressor that can supply up to 150 bar continuous air flow and enables testing at the full capacity of the current CPFR at 60 bar 410

Figure II.29.5. Demonstration of neural network prediction of liquid length data from the Engine Combustion Network hosted by Sandia National Laboratories. Symbols are data from... 411

Figure II.30.1. Cycle diagram showing gas exchange and open valve periods where species reduction is applied 414

Figure II.30.2. (a) CONVERGE mesh for CFR combustion chamber; (b) Direct injection (DI)/ACI chamber mesh based on Ford 1.6 L, 4-valve pent roof design 414

Figure II.31.1. Microcombustion results for Co-Optima fuel blends with matched RON and partially matched octane sensitivity. Fuel samples courtesy of Szybist/Oak Ridge National... 418

Figure II.31.2. Uncertainty quantification for temperature measurements via pyrometry (Schoegl/LSU) 419

Figure II.31.3. Operating conditions tested for nano-liter fuel delivery at ambient and elevated pressures (S. Menon/LSU) 420

Figure II.32.1. Sooting tendencies of amines (C₄H11N isomers) compared with structurally analogous alcohols and ethers (C₄H10O isomers). The sooting tendencies of the oxygenates... 424

Figure II.32.2. Soot yields for mixtures of a conventional diesel fuel and the optimal NREL acid-upgrading blendstock. The results are normalized to the pure diesel fuel. Each mixture... 424

Figure II.32.3. YSIs predicted for dioxolanes and related hydrocarbons with the NREL structure-property model. The panel on the left is with the original database, while the panel on... 425

Figure II.32.4. Measured YSI versus reference-case YSI for various air-fuel ratios (λ) and adiabatic flame temperatures (Tad). The reference-case YSI for each fuel is the YSI measured... 425

Figure II.32.5. YSIs for n-alkanes (left panel) and aromatic hydrocarbons (right panel) predicted with computational simulations at pressures from 1-15 atm 426

Figure II.32.6. YSI predictions for 20 Co-Optima Tier 2 and Tier 3 blendstocks using chemical-kinetic-based simulations 426

Figure III.1.1. Single-cylinder engine RCCI results with diesel fuel as the DI fuel and PRF80 as the premixed fuel for varying injection timings as an example result of a fuel reactivity... 432

Figure III.1.2. An experimental comparison between diesel-reformate RCCI and diesel-gasoline RCCI collected on a fully instrumented, single-cylinder research engine 433

Figure III.1.3. Cut-plane equivalence ratio distribution and temperature distribution from the CFD simulations of single-fuel RCCI combustion with diesel and its reformate 433

Figure III.2.1. CCV comparison between two ignition systems, each over two operating conditions 437

Figure III.2.2. CCV comparison between two ignition systems, each over two operating conditions 437

Figure III.2.3. Images are binarized and compared to adjacent pixels and relation to the ignited spark image, to assess if they are valid images or noise 437

Figure III.2.4. Validation of automated to manual ignition processing techniques over two different types of ignition systems (evaluated at the same operating point and condition) 438

Figure III.2.5. Hencken burner test set-up for igniter plasma imaging under atmospheric conditions 439

Figure III.2.6. Image of corona plasma during the 350μs duration corona generation phase is visible in the UV band with the Hencken burner, while the combustion flame front can be... 439

Figure III.3.1. Trending PM concentrations for all tests with OEM aftertreatment average PMmg/m³ 445

Figure III.4.1. Dedicated EGR engine configuration 447

Figure III.4.2. Updated engine configuration for dedicated cylinder testing 448

Figure III.5.1. Knock-limited phasing of propane and isooctane at two compression ratios and various intake temperatures 451

Figure III.5.2. Custom high-efficiency research engine with 1.5:1 stroke-to-bore ratio and high compression ratio 451

Figure III.5.3. Enthalpy ratio of reformed products to initial products in the synthetic exhaust flow reactor for three different engine conditions 452

Figure III.5.4. Modifed gasoline direct injection pump with bleed port for propane 453

Figure III.5.5. (a) Spark-to-CA10 combustion duration and (b) COV of IMEP for both gasoline and propane as a function of the cam overlap at 2,000rpm and 4 bar brake mean effective... 453

Figure III.5.6. Combustion efficiency for both gasoline and propane as a function of the cam overlap at 2,000rpm and 4 bar BMEP 454

Figure III.6.1. Bench-scale fuel injector flow rig for flow studies through the OEM piezoelectric fuel injector from a Ford 6.7-L diesel engine 457

Figure III.7.1. Kinetic intensity vs. average speed for the NREL package delivery cycle vs. various standard drive cycles 462

Figure III.7.2. Schematic of the GT-SUITE injection system model developed for the 4.3-L engine 463

Figure III.7.3. Representative pressure oscillations in Injector Bank #1 of the injection system of the 4.3-L engine for indolene (gasoline) operation at a nominal inlet pressure of 200 bar 464

Figure IV.1.1. 2018 CLEERS Workshop registrations by type of organization 468

Figure IV.1.2. 2018 CLEERS Workshop presentation topics 468

Figure IV.1.3. (a) Core sample temperatures, (b) outlet NO concentration, (c) outlet CO concentration, and (d) outlet CO₂ concentration measured during a synthetic exhaust flow reactor... 469

Figure IV.1.4. Impact of changing CO concentration on NO uptake and release measured in a synthetic exhaust flow reactor over a Pd-exchanged ZSM-5 PNA core sample. NO concentration... 471

Figure IV.1.5. Schematic of NO storage and release on a Pd-exchanged ZSM-5 PNA 471

Figure IV.2.1. NOx conversions as a function of temperature during standard SCR for selected Cu, H/SSZ-13 and Cu, Na/SSZ-13 degreened catalysts. Reactant feed contains 360ppm... 475

Figure IV.2.2. PNA performance of 1wt% and 1.9wt% Pd/SSZ-13 with Si/Al = 6. NOx adsorption at 100°C for 10 min (after 10min bypass) followed with temperature-programmed desorption... 476

Figure IV.2.3. Low-Temperature Three-Way Catalyst Test Protocol, the third protocol prepared by the LTAT group of the ACEC Tech Team and released to the technical community at... 477

Figure IV.2.4. CO oxidation light-off performance showing excellent low-temperature activity and stability of the hydrothermally treated Pt/CeO₂ catalyst under exhaust conditions... 477

Figure IV.2.5. Pore size distributions in samples from three axial regions in a commercial SCR-filter 478

Figure IV.3.1. (a) 2% Pt catalysts supported on a 20% SiO₂/80% Al₂O₃ support from Solvay show exceptionally low THC light-off behavior in the fresh state using the U.S. DRIVE... 481

Figure IV.3.2. Varying the Si content of the primarily alumina support material results in a decrease in THC light-off temperature in the 10-30% Si content range. One of the samples... 482

Figure IV.3.3. (a) In the degreened state, the Pd/ZSM-5 stores significant concentrations of C₃H6, C10H22, and NO following the U.S. DRIVE trapping protocol. (b) Upon aging at 800°C,... 482

Figure IV.3.4. (a) Fully hydrothermally aging the oxidation catalysts according to the U.S. DRIVE protocol results in significant losses in activity. (b) Including a hydrothermally aged trap... 483

Figure IV.3.5. Summary of T50 and T90 light-off temperatures at each of the different aging steps for the dual-bed Pd/ZSM-5 and oxidation catalyst combination 483

Figure IV.3.6. (a) Dual-bed configuration with Pd/ZSM-5 HC-trap in front of the oxidation catalysts while heating at a rate of 40°C/min. Significant HC quantities are trapped at low... 484

Figure IV.4.1. Tail-pipe emissions from a lean gasoline engine equipped with a passive SCR system over a pseudo-transient drive cycle using different operating strategies 488

Figure IV.4.2. (a) H₂ production from CO + H₂O and (b) NH₃ production from NO + CO + H₂O using catalysts with different Ce loading. Blue lines are catalyst activities at 350°C, yellow... 489

Figure IV.4.3. NH₃ production from NO + H₂ + H₂O (left), NH₃ production from NO + CO + H₂O (middle), and H₂ production from CO + H₂O (right) on Malibu-1 (top) and ORNL-1... 489

Figure IV.5.1. SCR-onset conversion infections for (a) a commercial Cu-CHA SCR catalyst and (b) a model Cu-Beta SCR catalyst 492

Figure IV.5.2. Conceptual model of Cu-SCR CI origin 493

Figure IV.5.3. Five-Step Protocol for characterizing onset transients for the individual and combined Cu-redox half cycles 494

Figure IV.5.4. (a) Variation in r-ratio (rRHC/rOHC) with temperature; (b) variation in N₂ CI with increasing r-ratio 494

Figure IV.5.5. Kinetic model results for varying NO (50, 100, 150, 200, 250ppm) at constant 200ppm NH₃ for (a) N₂, (b) NO CI, and (c) rRHC and rOHC. (d) Spatiotemporal CI distribution... 495

Figure IV.5.6. Kinetic-model predictions of the (a) NO, NH3 and N₂ onset transients for fast SCR, and (b) corresponding N₂-specific NO and NH₃ transients 495

Figure IV.6.1. Comparison of controlling pore size distributions for the C1 and A2 substrates found using a custom-built CFP system 499

Figure IV.6.2. Reconstructed microstructure (left), flow solution (center), and simulated concentration field (right) for 50-nm particles passing through a C1 filter wall 499

Figure IV.6.3. Comparison between measured initial penetration of 50-nm particles through various filter materials and predictions made by microscale lattice-Boltzmann simulations 500

Figure IV.6.4. Comparison between measured initial penetration of 50-nm particles through various filter materials and predictions made by a modifed spherical unit collector model 500

Figure IV.6.5. Size-resolved filtration efficiency predicted by a classical spherical unit collector model and by a new constricted tube collector model compared to experimental data for... 501

Figure IV.7.1. NOx uptake and release for 1wt% Pd/SSZ-13 with Si/Al = 6, 12, and 30, before and after hydrothermal aging 505

Figure IV.7.2. Transmission electron microscopy images of 1wt% Pd/SSZ-13 with Si/Al = 6 after hydrothermal aging 505

Figure IV.7.3. 27Al NMR spectra of fresh and hydrothermally aged 1wt% Pd/SSZ-13 with Si/Al = 6, 12, and 30 506

Figure IV.7.4. Methane conversion as a function of time on stream for 3wt% Pd/SSZ-13 with Si/Al = 6, 12, 24, and 36 506

Figure IV.7.5. Comparison of 3wt% Pd/SSZ-13 (Si/Al = 36) with Pd/Al₂O₃ 507

Figure IV.8.1. SCR standard (NO only) performance of Cu-SSZ-13 and Cu-SSZ-13 + 10wt% ZrO₂ catalyst 510

Figure IV.8.2. NH₃ oxidation (by O₂, i.e., without NOx) activity of Cu-SSZ-13 and Cu-SSZ-13 + 10wt% ZrO₂ at varying proximity of SCR and SCO catalyst phases 511

Figure IV.8.3. TPR behavior of Cu-SSZ-13 + 10wt% ZrO₂ catalyst compared to that of Cu-CHA alone and ZrO₂ alone 512

Figure IV.8.4. Pathway towards improved NOx reduction performance by a surface-active NOx species 512

Figure IV.8.5. Thermally induced ion-exchange aging mechanism of Cu-CHA + Ba/ZrO₂ binary catalyst 513

Figure IV.9.1. Engine and instrumentation schematic showing dilution system and instruments used in the experimental study 516

Figure IV.9.2. Engine-out PSDs for the lean homogenous combustion mode at the four different engine conditions for each of the seven fuels tested. From top left to bottom right: LH1,... 518

Figure IV.9.3. Correlation between PN and the PMI for the lean homogenous combustion mode at the 2,000pm 7 bar BMEP engine condition with the E50 fuel omitted 518

Figure IV.9.4. PN 〉 23nm filtration efficiency as a function of non-oxidized soot mass loading on the GPF per filter substrate volume 519

Figure V.1.1. Project schedule and phasing (Volvo) 521

Figure V.1.2. Baseline Model Year 2009 vehicle (left) and VEV3 test mule (right) at rest stop during fuel economy test (Volvo) 522

Figure V.1.3. Synthetic exhaust gas flow reactor rig at Oak Ridge National Laboratory (Volvo) 524

Figure V.2.1. Mule vehicle WHR system integration. This system will be applied in stages in late 2019, finishing in 2020 530

Figure V.3.1. Effects of cylinder deactivation on BSFC 533

Figure V.3.2. Turbocharger optimization 534

Figure V.3.3. Advanced fuel injection strategy simulation 534

Figure V.3.4. Schematic of intent ORC system 535

Figure V.3.5. Effects of oil temperature on BTE of high-temperature and baseline pistons 535

Figure V.3.6. Efficiency breakdown and loss comparison at peak BTE for diesel, EEE, and E85 with a new injector 536

Figure V.4.1. Phases of SuperTruck 2 project 538

Figure V.4.2. ST2 collaboration model 539

Figure V.4.3. ST2 updated validation methodology 540

Figure V.4.4. Detroit roadmap to reach the 55% BTE target 541

Figure V.4.5. Daimler Trucks North America updated pathway to reach 115% freight efficiency target 542

Figure V.4.6. A-Sample impacted bills of material 542

Figure V.5.1. Route selection to evaluate SuperTruck technologies 545

Figure V.5.2. Forecasted freight efficiency 546

Figure V.5.3. Aerodynamic development (computational fluid dynamics) 547

Figure V.5.4. Vehicle velocity vs. grade to determine power requirement 548

Figure V.5.5. Engine roadmap to 55% BTE 549

Figure V.5.6. Simulation results of in-cylinder combustion and port flow optimization 550

Figure V.6.1. Gen 3X GDCI engine 553

Figure V.6.2. Preliminary load sweep at 1,500rpm for the Gen 3X GDCI engine 554

Figure V.6.3. Initial BSFC contour map for Gen 3X GDCI engine 555

Figure V.6.4. Engine compartment of completed Gen 3 vehicle 555

Figure V.6.5. Simulated vehicle fuel economy results for the 2.2L Gen 3X GDCI engine with various powertrain technologies compared to baseline engine and full hybrid engines (Argonne) 556

Figure V.7.1. Fuel consumption improvement achieved during SCE calibration optimization on final MCE hardware set 560

Figure V.7.2. First steady-state engine (fixed-drive supercharger) and dyno installation 561

Figure V.7.3. BSFC reduction of LMC engine running in SMC mode with external EGR, naturally aspirated key points 561

Figure V.7.4. Cycle fuel economy predictions based on weighted test points 562

Figure V.7.5. Transient dyno hardware package 563

Figure V.8.1. Revised path to 55% BTE from project's demonstrated peak BTE 567

Figure V.9.1. UDDS and HWFET drive cycle test results 569

Figure V.9.2. Efficiency gain of LTC 570

Figure V.9.3. Effects of fuel mass injected for second injection 570

Figure V.9.4. Coefficient of variation of IMEP at 50V/1,000μs, 55V/1,000μs, and 60V/1,500μs 571

Figure V.9.5. Coefficient of variation of IMEP and crank angle at 50% mass fraction burned for voltage from 50V to 70V 571

Figure V.9.6. Igniter-by-igniter variation 571

Figure V.10.1. Locking mechanism friction comparison 575

Figure V.10.2. Final engine build DRFF hardware 575

Figure V.10.3. GEN 5 solenoid force vs. stroke throughout durability test 576

Figure V.10.4. Complete DSF cylinder head assembly on test stand 576

Figure V.10.5. Actuator driver module 577

Figure V.10.6. Baseline engine installed in dyno cell 577

Figure V.11.1. Surface roughness is greatly improved by filtering spheres for size and using a top layer of small-diameter spheres sintered under pressure; this results in a more robust... 580

Figure V.11.2. Valve prototypes with microshell TBC applied to the face were fabricated by (1) machining stainless steel valves with a pocket in the face, (2) filling the pocket with nickel... 581

Figure V.11.3. Exhaust port liner coating process begins with (1) plating a core with solid nickel to form a shell once the core is dissolved, (2) spraying the shell with a TBC microshell... 581

Figure V.11.4. Exhaust temperatures in the center (left) and near the wall (right) at the exhaust port exit 581

Figure V.11.5. Pre-and post-test conditions of Generation 2 coated valves vs. Generation 1 582

Figure VI.1.1. HP PTWA coating shows friction benefits with DLC, PVD, and nitride ring face coatings 586

Figure VI.1.2. Cranktrain friction as a function of engine speed at 100°C oil temperature for various HP PTWA coatings 586

Figure VI.1.3. Cranktrain friction as a function of engine speed at 100°C oil temperature for various piston ring face coatings with HP PTWA 3 coating 587

Figure VI.1.4. Cranktrain friction as a function of engine speed at 100°C oil temperature for various piston ring face coatings with HP PTWA 3 coating 587

Figure VI.1.5. Motored engine friction tests showed HP PTWA coatings offer friction benefit 588

Figure VII.1.1. High-resolution C1s XPS analysis of the functionalized GO fillers before the reduction 595

Figure VII.1.2. High-resolution C1s XPS analysis of the functionalized GO fillers after the reduction 595

Figure VII.1.3. High-resolution N1s XPS analysis of the functionalized GO fillers (a) before and (b) after the reduction 595

Figure VII.1.4. SBS samples filled with rGO at several weight contents 596

Figure VII.1.5. Scanning electron microscopy images of the abraded surface of (a) unfilled SBS and (b) SBS-GO samples after 100 abrading cycles 596

Figure VII.2.1. Example of sampling data for determination of diffusion coefficients 601