ORNL is managed by UT-Battelle for the US Department of Energy
Stretch Efficiency for Combustion Engines: Exploiting New Combustion RegimesProject ID: ACS015
James P. Szybist (P.I.), Josh A. Pihl, Melanie Debusk, Daniel W. Brookshear, and Yan Chang (student)Oak Ridge National Laboratory
Galen B. Fisher (subcontractor)University of Michigan
June 20th, 2018
DOE Management TeamGurpreet Singh and Mike Weismiller
This presentation does not contain any proprietary, confidential, or otherwise restricted information.
2 ACS015_Szybist_2018_o
Project OverviewProject OverviewRelevanceMilestonesApproachAccomplishments Reviewer CommentsCollaborationsFuture WorkSummary
• FY17: $300k• FY18: $300k(~ 0.5 FTE + materials)
Budget BarriersACEC Roadmap, Topic Area 1
• Thermal management (efficient low-cost waste-heat recovery…)
• Increase EGR dilution tolerance
• Reduce content, cost, and complexity of engines while increasing efficiency
Timeline
• Part of ORNL’s FY17-FY19lab call
• New lab call beginning FY19, proposing continuing work
• Builds on prior Stretch Efficiency research program at ORNL
• Focus on thermochemical recuperation in 2011
Collaborators• Ford – Providing technical input• Caterpillar – Providing technical input• FCA – Providing technical input • AEC working group led by SNL
– Industry feedback• Aramco Services – Technical collaboration• ANSYS (formerly Reaction design) – CFD
model development• Umicore – Catalyst coatings
Universities• University of Michigan -
Galen Fisher• University of Michigan –
Yan Chang
National Labs• SNL - Isaac Ekoto
3 ACS015_Szybist_2018_o
Relevance: Decreased Petroleum Consumption through Higher Engine Efficiency
Project OverviewRelevanceMilestonesApproachAccomplishments Reviewer CommentsCollaborationsFuture WorkSummary• Increase IC engine efficiency with an approach centered on
thermodynamic of engine processes and minimizing losses
• Thermal management (efficient low-cost waste-heat recovery)– This project is investigating the feasibility of waste-heat
recovery through thermochemical recuperation (TCR)
• Increase EGR dilution tolerance– EGR-loop catalytic reforming produces a H2 and CO mixture
capable of extending EGR dilution tolerance
– EGR dilution tolerance approaching 50% demonstrated in 2017 AMR
• Reduce content, cost, and complexity of engines while increasing efficiency
– Relative to lean-burn options for high efficiency, the stoichiometric approach here simplifies emissions controlcost and complexity
Overall Project Goal
ACEC Roadmap, Area 1, Research Priorities Addressed Exhaust
Intake
Cata
lyst
Note: Schematic represents engine flow paths and is not intended to
represent instrumentation or controls
4 ACS015_Szybist_2018_o
This Project has Two Tracked Milestones for FY18
Second Quarter, FY2018
Perform a coarse engine map of engine operation (efficiency and emissions) with the catalytic reforming strategy from near-idle to boosted engine operating conditions.
Status: Milestone not completed on time. Delayed to Q4.
Project OverviewRelevanceMilestonesApproachAccomplishments Reviewer CommentsCollaborationsFuture WorkSummary
Fourth Quarter, FY2018
Complete an assessment of the Rh catalyst sulfur deactivation and the ability to regenerate the catalyst at engine-relevant conditions.
Status: On-track.
5 ACS015_Szybist_2018_o
Synthetic Exhaust Flow Reactor Used to Guide EngineInvestigations
• Steam and partial oxidation reforming investigated in an automated synthetic exhaust flow reactor for application in an EGR-loop reforming strategy on an SI engine
• Pre-commercial catalyst formulation from Umicore– 2 wt% Rh supported on Al2O3 and coated onto a zirconia-mullite
substrate of 400 cells per square inch
• Identifies catalyst boundary conditions for efficient reforming, including thermochemical recuperation
– Engine operated to mimic the catalyst boundary conditions– Excellent transferability has been demonstrated
Project OverviewRelevanceMilestonesApproachAccomplishments Reviewer CommentsCollaborationsFuture WorkSummary
Primary EGR Feed14% CO2, 13% H2O, 73% N2
Fuel AdditionControls Steam/Carbon (S/C)
Air AdditionControls Oxygen/Carbon (O/C)
6 ACS015_Szybist_2018_o
Implementation in an EGR Reforming Loop Requires Oxygen at the Reforming Catalyst
• Lean combustion by one cylinder to feed oxygen to reforming catalyst– Allows stoichiometric exhaust for 3-way exhaust TWC compatibility– Allows lean cylinder to be at a higher MAP, match load
• Fuel provided to catalyst through post injection event
Lean Combustion
Post Injection
Catalyst Feed (Fuel + O2)
0
500
1000
1500
2000
2500
0
5
10
15
20
-360 -180 0 180 360
Cylin
der P
ress
ure
[ kPa
]
Fuel Injection Comm
and [ A ]
Crank Angle
Exhaust
Intake
Cata
lyst
Project OverviewRelevanceMilestonesApproachAccomplishments Reviewer CommentsCollaborationsFuture WorkSummary
7 ACS015_Szybist_2018_o
Implementation in an EGR Reforming Loop Requires Oxygen at the Reforming Catalyst
• Lean combustion by one cylinder to feed oxygen to reforming catalyst– Allows stoichiometric exhaust for 3-way exhaust TWC compatibility– Allows lean cylinder to be at a higher MAP, match load
• Fuel provided to catalyst through post injection event
• In FY18, moved to in-pipe catalyst fueling usingwater-cooled PFI injector
– Alleviates cylinder-to-cylinder load balancing
– Alleviates oil dilution concerns
Project OverviewRelevanceMilestonesApproachAccomplishments Reviewer CommentsCollaborationsFuture WorkSummary
Exhaust
IntakeCa
taly
ticRe
form
erEG
R Co
oler
Intake Plenum
StaticMixer
In-Pipe Catalyst Fueling (cooled)
8 ACS015_Szybist_2018_o
Three Studies Published in Energy & Fuels Detailing the Catalytic Reforming from Bench Flow Reactor to Full Engine System
• Results contained in these manuscripts presented at an early stage in 2017 DOE merit review
• Significant effort in analysis and reporting in the last year to publish manuscripts
• Combined, these tell a comprehensive story of developing efficient reforming conditions for engines
Paper 1. Reforming using synthetic exhaust flow reactor.
• Reforming performance and energy balances• Impact of different fuel compositions
Paper 2. Catalyst performance on an engine with real exhaust
• Reformate yield with real engine boundary conditions• Thermal conditions in catalyst and impact of water-gas
shift reaction
Paper 3. Full engine performance in multi-cylinder engine
• Brake thermal efficiency gain• Impact of reformate on combustion processes
• Comparison with conventional EGR
Accomplishments
9 ACS015_Szybist_2018_o
Efficient Reforming with Oxygen Present Requires Very Rich Conditions (Φ~7.0). Max H2 Production ≠ Max Efficiency.
Efficient reforming and TCR requires Φ > 6. Energy
balance dependent on inlet T.
Highest H2 at Φ = 5-6, corresponded with significant water consumption (evidence
of steam reforming reactions).
3 4 5 6 7 8 9 106
8
10
12
14
16
18
20
22
24
O2 C
atal
yst F
low
[ g/
min
]
Catalyst Equivalence Ratio
29.5
30.0
30.5
31.0
31.5
32.0
32.5
Brake Efficiency [ % ]
2/3 Engine Fueling
1/3 Engine Fueling
Total Engine Fueling
1.0 0.8 0.6 0.5 0.4 0.3O/C Ratio
Best brake efficiency occurred at minimum reforming oxygen concentration that produced
high H2 concentration.
Synthetic Exhaust Gas Flow Reactor Reforming Study
Full-Sized Catalyst Performance Study On-Engine
Multi-Cylinder Engine Performance Study with
Reforming Strategy
Accomplishments
10 ACS015_Szybist_2018_o
Significant Progress Made on Load Range Expansion, but Experimental Events Resulted in Equipment Failure. Goal Not Yet Met.
• Initial catalytic reforming engine investigation at 2000 rpm, 4 bar BMEP– Post-injection catalyst fueling strategy to efficient reforming boundary conditions– Hydrogen production of >15% catalyst-out, EGR tolerance >45%, fuel consumption decrease of 8%
– Results presented at the 2017 AMR
0
5
10
15
20
1000 1500 2000 2500 3000
Planned Load Expansion, Winter 2018Initial Load Expansion, Fall 2017Baseline, AMR 2017
Brak
e M
ean
Effe
ctiv
e Pr
essu
re [
bar ]
Engine Speed [ rpm ]
Lean Combustion
Post Injection
Catalyst Feed (Fuel + O2)
0
500
1000
1500
2000
2500
0
5
10
15
20
-360 -180 0 180 360
Cylin
der P
ress
ure
[ kPa
]
Fuel Injection Comm
and [ A ]
Crank Angle
3 4 5 6 7 8 9 106
8
10
12
14
16
18
20
22
24
O2
Cat
alys
t Flo
w [
g/m
in ]
Catalyst Equivalence Ratio
0.0
2.5
5.0
7.5
10.0
12.5
15.0Hydrogen_pct
2/3 Engine Fueling
Total Engine Fueling
1/3 Engine Fueling
27
28
29
30
31
32
33
45 50 55 60 65 70
Catalytic ReformingConventional EGRBr
ake
Effic
ienc
y [%
]
Intake Manifold Pressure [ kPa ]
Accomplishments
11 ACS015_Szybist_2018_o
0
5
10
15
20
1000 1500 2000 2500 3000
Planned Load Expansion, Winter 2018Initial Load Expansion, Fall 2017Baseline, AMR 2017
Brak
e M
ean
Effe
ctiv
e Pr
essu
re [
bar ]
Engine Speed [ rpm ]
Significant Progress Made on Load Range Expansion, but Experimental Events Resulted in Equipment Failure. Goal Not Yet Met.
• Preliminary load range expansion accomplished Fall of 2017 – Naturally aspirated load limit encountered at approximately
6 bar BMEP because of high dilution– Included nine thermocouples inside of catalyst to determine axial temperature profile
– Unable to balance load between cylinders – particularly at the 1500 rpm, 0.7 bar BMEP point
• Hardware upgrades required for additional load range expansion
CatalystInlet T
CatalystMidbed T
CatalystOutlet T
Approximate Naturally Aspirated Load Limit
Accomplishments
12 ACS015_Szybist_2018_o
400
500
600
700
800
-2 0 2 4 6 8 10 12
2500 rpm, 6.0 bar BMEP2000 rpm, 4.0 bar BMEP1500 rpm, 0.8 bar BMEP
Tem
pera
ture
[ o C
]
Axial Distance [ cm from catalyst face ]
Axial Temperature Profiles for Preliminary Load Range Expansion Provide Confidence that Further Expansion is Possible
Temperature rise at catalyst mid-bed
Temperature drop of 40-50 °C for 2000, 4
bar, 2500, 6 bar
1500 rpm, 0.8 bar condition is
thermally neutral
Catalyst Monolith 1
Catalyst Monolith 2
Engine OutN2, CO2, H2O,
CxHy, O2
Reformer OutH2, CO, CH4, CO2,N2, H2O
Accomplishments
13 ACS015_Szybist_2018_o
400
500
600
700
800
-2 0 2 4 6 8 10 12
2500 rpm, 6.0 bar BMEP2000 rpm, 4.0 bar BMEP1500 rpm, 0.8 bar BMEP
Tem
pera
ture
[ o C
]
Axial Distance [ cm from catalyst face ]
Axial Temperature Profiles for Preliminary Load Range Expansion Provide Confidence that Further Expansion is Possible
Catalyst Monolith 1
Catalyst Monolith 2
Engine OutN2, CO2, H2O,
CxHy, O2
Reformer OutH2, CO, CH4, CO2,N2, H2O
190°C temperature difference pre-cat
Peak temperature near catalyst face (oxidation)
• Only 75°C difference in peak temperature• Peak temperature controlled by oxygen
present at catalyst inlet
Majority of the catalyst volume is
used for endothermic steam reforming and
water-gas shift reactions
Accomplishments
14 ACS015_Szybist_2018_o
27
28
29
30
31
32
Baseline Reforming
Brak
e Ef
ficie
ncy [
% ]
2000 RPM, 4 bar BMEP
10
11
12
13
14
15
Baseline Reforming
Brak
e Ef
ficie
ncy [
% ]
1500 RPM, 0.8 bar BMEP
30
31
32
33
34
35
Baseline Reforming
Brak
e Ef
ficie
ncy [
% ]
2500 RPM, 6 bar BMEP
At Preliminary Load Range Expansion Points, Significant Increases in Brake Efficiency Realized While Maintaining Stability
0%
1%
2%
3%
4%
5%
Baseline Reforming
COV
of IM
EP [
% ]
2000 RPM, 4 bar BMEP
0%
1%
2%
3%
4%
5%
Baseline Reforming
COV
of IM
EP [
% ]
1500 RPM, 0.8 bar BMEP
0%
1%
2%
3%
4%
5%
Baseline Reforming
COV
of IM
EP [
% ]
2500 RPM, 6.0 bar BMEP
Accomplishments
15 ACS015_Szybist_2018_o
Light Load Condition Disallowed Load Balancing Between Cylinder due to Fuel in Trapped Residuals
10
11
12
13
14
15
Baseline Reforming
Brak
e Ef
ficie
ncy [
% ]
1500 RPM, 0.8 bar BMEP
0%
1%
2%
3%
4%
5%
Baseline Reforming
COV
of IM
EP [
% ]
1500 RPM, 0.8 bar BMEP
0
50
100
150
200
250
300
350
Cyl 1 Cyl 2 Cyl 3 Cyl 4
IMEP
[ kP
a ]
Individual Cylinder IMEP at 1500 rpm, 0.8 bar BMEP
• Due to low exhaust temperature, reforming relied more strongly on POx reactions
– High concentration of O2 for high exotherm
– Entirety of fuel for cyl 1-3 sent through reforming catalyst
• This results in a lot of fuel in cyl 4 trapped residuals– IMEP in cyl 4 could not be decreased sufficiently
– Load could not be balanced across cylinders
• Oil dilution concerns with post-injection at higher loads
• Desire to move away from post-injection in cylinder 4
Accomplishments
16 ACS015_Szybist_2018_o
Experimental Modifications Implemented to Enable Boosted Operation and In-Pipe Reforming Catalyst Fueling
Accomplishments Water-cooled housing for urea doser,
fits Bosch PFI injector
Boosted Operation Required Higher Capacity EGR Cooler (DDC Series 60)
Exhaust
Cata
lytic
Refo
rmer
EGR
Cool
er
Intake Plenum
StaticMixer
In-Pipe Catalyst Fueling (cooled)
Mass Flow Controller
Mass Flow Controller
Rupture Disc
Backpressure Valve
Bank of Mass Flow Controllers to Simulate Boost
Backpressure Valve and Rupture Disc for Realistic
Turbo Boundary Conditions
17 ACS015_Szybist_2018_o
0
5
10
15
20
1000 1500 2000 2500 3000
Planned Load Expansion, Winter 2018Initial Load Expansion, Fall 2017Baseline, AMR 2017
Brak
e M
ean
Effe
ctiv
e Pr
essu
re [
bar ]
Engine Speed [ rpm ]
Significant Progress Made on Load Range Expansion, but Experimental Events Resulted in Equipment Failure. Goal Not Yet Met.
• Planned load expansion points with modified engine hardware aimed for boosted operation – Initial experiments at 2000 rpm 4 bar BMEP and 2500 rpm 6 bar BMEP were conducted successfully
– Prior to going to boosted operation, cooling water connection in cell failed (~20 gal facility water in ~ 2 minutes)
– Rushed shutdown led to overtemperature in catalyst, reached ~1400°C before thermocouple failure
– Overtemperature occurred in EGR cooler immediately downstream of catalyst, melting a braze
– EGR air system filled with water, including reforming catalyst
Accomplishments
Despite severe overtemperature, zirconia mullite maintained structural integrity (i.e., no melted
catalyst pieces recirculated to engine intake)
18 ACS015_Szybist_2018_o
1
10
6 10-5 0.0001 0.0005
Pres
sure
[ ba
r ]
Volume [ m3 ]
2017 AMR Reviewer Question: Could Efficiency be Higher with Lean Combustion?
• Experimental measurements show that the polytropic coefficient is higher for the stoichiometric cylinders than the lean cylinder
• What the heck is going on here???
Accomplishments
1
10
6 10-5 0.0001 0.0005
Pres
sure
[ ba
r ]
Volume [ m3 ]
Cylinder 4λ = 1.2
PolyC=1.32
Cylinder 1-3λ = 1.0
PolyC=1.40
19 ACS015_Szybist_2018_o
Gamma Increases with Dilution Because Concentration of Fuel Decreases; Reforming Benefit 2x Lean (λ = 1.4) for Reactants!
Accomplishments
1.24
1.26
1.28
1.3
1.32
1.34
1.36
1.38
1.4
300 475 650 825 1000
Gam
ma
Temperature [ K ]
1.24
1.26
1.28
1.3
1.32
1.34
1.36
1.38
1.4
300 475 650 825 1000
Gam
ma
Temperature [ K ]
1.24
1.26
1.28
1.3
1.32
1.34
1.36
1.38
1.4
300 475 650 825 1000
Gam
ma
Temperature [ K ]
1.24
1.26
1.28
1.3
1.32
1.34
1.36
1.38
1.4
300 475 650 825 1000
Gam
ma
Temperature [ K ]
Stoichiometric
Lean (λ = 1.4)
40% EGR
Reformate Mix
Stoichiometric Combustion: C8H18 = 1.7 mol%C8H18 + 12.5 (O2 + 3.76 N2) → 8 CO2 + 9 H2O + 47 N2
Lean Combustion (λ = 1.4): C8H18 = 1.2 mol%C8H18 + 17.5 (O2 + 3.76 N2) → 8 CO2 + 9 H2O + 5 O2 + 65.8 N2
EGR (40% EGR): C8H18 = 1.2 mol%C8H18 + 12.5 (O2 + 3.76 N2) + 2.98 CO2 + 3.35 H2O + 17.48 N2 → 10.98 CO2 + 12.35 H2O + 64.48 N2
Steam ReformingC8H18 + 2.98 CO2 + 3.35 H2O + 17.48 N2 → 0.582 C8H18 + 2.98 CO2 + 17.48 N2 + 7.11 H2 + 3.35 CO
Combustion with Reformate: C8H18 = 0.6 mol%0.582 C8H18 + 2.98 CO2 + 17.48 N2 + 7.11 H2 + 3.35 CO + 12.5 (O2 + 3.76 N2) →10.98 CO2 + 12.35 H2O + 64.48 N2
20 ACS015_Szybist_2018_o
0.40.420.440.460.48
0.50.520.54
Cycle Efficiency
Thermodynamic Closed-Cycle Analysis Shows that Stoichiometric Efficiency with Reformate is Competitive with Lean Combustion • Chemkin used to evaluate closed cycle thermodynamics under adiabatic conditions
– Weibe function used to impose common deflagration event for all conditions– All cases starting with the same iso-octane fuel energy
1.31.311.321.331.341.351.36
Poly C
1.231.241.251.261.271.281.29
Poly E
0.95
1
1.05
1.1
1.15
1.2
Normalized Work
• Despite same composition, reforming case has lower Poly E
• Attributable to higher flame temperature for reformate
• Lower molar expansion ratio• Lower exergy/enthalpy ratio
Includes TCR
• Which LHV should we use???• Higher efficiency is iso-octane LHV• Lower efficiency is mixture LHV
after reforming
Accomplishments
21 ACS015_Szybist_2018_o
Related Project Aiming to Apply Same Reforming Strategy to Propane
• Propane has higher H/C ratio than gasoline, making reforming thermodynamics more favorable
• Starting project with a flow-reactor investigation tocharacterize propane performance
– Investigating a variety of inlet temperature/ space velocity
• Reforming thermodynamics become more favorable athigher load (higher temperature) conditions
Accomplishments
Regular Grade Gasoline
Propane
Fuel H/C Ratio 1.97 3.0
Steam Reforming
H2/CO Ratio1.99 2.33
POx ReformingH2/CO Ratio 0.99 1.33
3 4 5 6 7 8 9 10
0.02
0.03
0.04
0.05
Mol
e Fr
actio
n O
2 in R
efor
mer
Fee
d
Reformer Equivalence Ratio
0.70
0.75
0.80
0.85
0.90
0.95
1.00
1.05
1.10
1.15
1.20
1.25Enthalpy RatioSV = 3300 h-1, Tin = 450oC
Replicates 1500 RPM, 0.7 bar Replicates 2000 RPM, 4.0 bar
3 4 5 6 7 8 9 10
0.02
0.03
0.04
0.05
SV = 8,500 h-1, Tin = 580oC
Mol
e Fr
actio
n O
2 in R
efor
mer
Fee
d
Reformer Equivalence Ratio
0.70
0.75
0.80
0.85
0.90
0.95
1.00
1.05
1.10
1.15
1.20
1.25Enthalpy Ratio
Replicates 2000 RPM, 14.0 bar
3 4 5 6 7 8 9 10
0.02
0.03
0.04
0.05
SV = 23,500 h-1, Tin = 800oC
Mol
e Fr
actio
n O
2 in R
efor
mer
Fee
d
Reformer Equivalence Ratio
0.70
0.75
0.80
0.85
0.90
0.95
1.00
1.05
1.10
1.15
1.20
1.25Enthalpy Ratio
22 ACS015_Szybist_2018_o
Four Reviewers Evaluated this Project in 2017 Overall Positive Comments with Room for Improvement
Several reviewers commented on the need for expanded speed/load range
We have made significant progress in this direction (1500 rpm, 0.8 bar BMEP;2500 rpm, 6.0 bar BMEP). We are working towards higher load underboosted operating conditions and expect to have initial results by the endof FY18.
Two comments on emission barriers/opportunities, including cold-start
The focus on stoichiometric operation is to maintain compatibility with conventional 3-way catalysis. The low engine-out NOx emissions may havebenefits under transient conditions. However, significant progress needs to be made before any transient testing. This is a stoichiometric SI engine, so cold-start could be done in a conventional manner (i.e., without reforming).
Collaborators listed, but no particular collaborator input is attributed to specific collaborators
Input from OEM partners has been to investigate light load (tip-out) and high load operation, with attention to combustion stability and emissions.
One reviewer requested additional information about the catalyst, and requested that we report the effects of different catalyst compositions
This catalyst was originally developed by Delphi for reforming purposes, and this project is building on prior industry expertise. Further catalyst development or testing different catalyst formulations is outside of the current project scope.
Reviewer Comments
23 ACS015_Szybist_2018_o
Collaborations
• OEM Collaborations: one-on-one discussions, discussions of implementation barriers, feedback on results and future plans
– Ford– Caterpillar– FCA
• Umicore – Providing pre-production Rh-based catalysts
• ANSYS (formerly Reaction Design) – CFD model development and technical assistance
• University of Michigan: Yan Chang is a UM student working on the project at ORNL for 2016, advised by Stani Bohac and André Boehman
• AEC Working Group bi-annual meetings– Mechanism for industry feedback
• University of Michigan: Galen Fisher advising on catalyst formulation and operating conditions through subcontract
• Related funds-in project with Aramco Services Co.
• Sandia National Laboratories: Historical collaboration with Isaac Ekoto (and Dick Steeper). Projects diverged this year, but technical discussions continue.
• Project direction from 2010 USCAR Colloquiumhttp://feerc.ornl.gov/pdfs/Stretch_Report_ORNL-TM2010-265_final.pdf
Project OverviewRelevanceMilestonesApproachAccomplishmentsReviewer CommentsCollaborationsFuture WorkSummary
24 ACS015_Szybist_2018_o
Project OverviewRelevanceMilestonesApproachAccomplishmentsReviewer CommentsCollaborationsFuture Work Summary
Remaining Barriers and Future Work
This technology has only been demonstrated for a limited number of speed/load conditions. Engines are required to operate over a wider range of conditions.
Remaining Barrier 1 Corresponding Future WorkContinue to experimentally expand the speed-loadoperating regime to both lighter loads and higher loads.Use flow reactor experiments to guide efficient reforming boundary conditions
Remaining Barrier 2 Corresponding Future WorkUnclear how much fuel-borne sulfur will limit the applicability of this technology.
Characterize the extent of sulfur deactivation at multiple speed-load operating conditions. If necessary, develop techniques to regenerate the catalyst with minimum fuel penalty.
Remaining Barrier 3 Corresponding Future WorkCombination of system thermodynamics, combustion processes, and energy losses is not fully understood.
Utilize 0-D thermodynamic modeling, 0-D kinetic modeling, 1-D gas exchange modeling, and 3D CFD modeling to provide a better understanding of the thermodynamics, reforming processes, and combustion processes.
Any proposed future work is subject to change based on funding level
25 ACS015_Szybist_2018_o
Summary
Addressing research priorities called-out in the ACEC Roadmap for topic area 1: a) low cost waste heat recovery, b) increased EGR tolerance, c) reduced complexity
Collaborations
Accomplishments
Approach: Experimental and Modeling Efforts Grounded in Thermodynamics
Relevance
Future Work
• Synthetic exhaust flow reactor experiments define boundary conditions for high η reforming, TCR• Operate engine to achieve efficient reforming boundary conditions in catalyst, pursue highest η• 0-D thermodynamic through CFD modeling used to understand results, guide next steps
• Published three articles in Energy & Fuels documenting reforming strategy from flow reactor to full scale multi-cylinder engine experiments
• Expanded speed-load range lower (1500 rpm, 0.8 bar BMEP) and higher (2500 rpm, 6.0 bar BMEP)• Performed thermodynamic analysis of reformate on closed-cycle efficiency
• Ford, FCA, Caterpillar, Aramco Services Co., Umicore, Ansys, University of Michigan
• Speed load expansion, sulfur tolerance, phenomenological modeling
Project OverviewRelevanceMilestonesApproachAccomplishmentsReviewer Comments CollaborationsFuture WorkSummary
Any proposed future work is subject to change based on funding level
27 ACS015_Szybist_2018_o
Hypothesis: Exergy/enthalpy ratio is related to the molar expansion ratio
Molar Expansion Ratio ≡ (moles products)/(moles reactants)
• Molar expansion ratio is dependent on fuel type
• The molar change during combustion is not accounted for in the LHV measurement or the enthalpy of reaction
• Change in the number of moles is accounted for in the entropy term, so it is included in exergy of reaction
• Current study is limited to stoichiometric combustion with air to maximize fuel differences in molar expansion ratio
• Molar expansion ratio approaches unity with increasing dilution (lower equivalence ratio or higher EGR)
CH4 + 2 (O2 + 3.76 N2) ⟶ CO2 + 2 H2O + 7.52 N2 CH3OH + 1.5 (O2 + 3.76 N2) ⟶ CO2 + 2 H2O + 5.64 N2 CO + 0.5 (O2 + 3.76 N2) ⟶ CO2 + 1.88 N2
nproduct/nreactant = 1.00 nproduct/nreactant = 1.21
nproduct/nreactant = 0.85
Backup (1/5)
Szybist, J.P., K. Chakravathy, C.S. Daw. Analysis of the Impact of Selected Fuel Thermochemical Properties on Internal Combustion Engine Efficiency. Energy & Fuels, 2012, vol 26(5), pp. 2798-2810.
28 ACS015_Szybist_2018_o
Molar expansion ratio determines the extent of residual pressure available to perform work
CH4 + 2(O2 + 3.76 N2)
Q
Pfinal = 1 atmNo residual work potential
CH3OH + 1.5(O2 + 3.76 N2)
Q
Pfinal > 1 atmPositive residual work potential
CO + 0.5(O2 + 3.76 N2)
Q
Pfinal < 1 atmAtmosphiericwork potential
Constant volume reactant chambers, Initial T = 100 C, Initial P = 1 atm
Final T = Initial T = 100 C
Backup (2/5)
29 ACS015_Szybist_2018_o
Ideal Cycle Efficiency Analysis Shows Contribution of γ to Different Portions of the Cycle
• Ideally you would want γ to be different for different parts of the cycle
– Low γ for compression and expansion
– High γ for heat addition
• This type of analysis is removed from reality and can produce efficiency > 100%
• In reality, γ is coupled throughout the cycle– Relationship of γ for different parts of the cycle changes
with stoichiometric, lean, and reforming cases
– More work is necessary to fully understand the tradeoffs
Backup (3/5)
30 ACS015_Szybist_2018_o
Synthetic Exhaust Flow Reactor Demonstrates Sulfur Effect is Significant but Reversible
• 2 ppm SO2 introduced immediately upstream of reforming catalyst (represents fuel with 30 ppm S by mass)
• Steady conditions for 60 minutes show stable concentrations and temperatures
• Introduction of 2 ppm SO2 causes increase in mid- and post-catalyst positions, accompanied by a reduction in reformate
• Near immediate recovery of reforming process when sulfur is removed
No Sulfur 2 ppm Sulfur No Sulfur No Sulfur 2 ppm Sulfur No Sulfur
Backup (4/5)
31 ACS015_Szybist_2018_o
Preliminary Experiments at 1500 rpm, 0.8 bar BMEP Show Similar Trends as Flow Reactor but on Longer Timescales using 30 ppm S Fuel
720
725
730
735
740
745
750
-20 0 20 40 60 80
Pea
k T
in C
atal
yst [
C ]
Relative Time [ min ]
Fueling with S-containing fuel
0
2
4
6
8
10
12
14
16
-20 0 20 40 60 80
Hyd
roge
n [ p
ct ]
Relative Time [ min ]
Fueling with S-containing fuel
0
1
2
3
4
5
6
7
8
-20 0 20 40 60 80
H2O
[ pc
t ]
Relative Time [ min ]
Fueling with S-containing fuel
0
5
10
15
20
25
-20 0 20 40 60 80
Ave
rage
CA
50 P
hasi
ng
Relative Time [ min ]
Fueling with S-containing fuel
Peak T in Catalyst Increases 10-15 deg C
Hydrogen production decreases
Catalyst-out water increases(water consumption decreases)
CA50 Retards
Sulfur poising to be revisited in more complete study later in FY18
Backup (5/5)
Top Related