Top Banner
Oxidation Characteristics of Soot from a Gasoline Direct-Injection (GDI) Engine Seungmok Choi, HeeJe Seong, Kyeong O. Lee Transportation Technology R&D Center Argonne National Laboratory 2014 DOE CLEERS Workshop Dearborn, Michigan, USA April 30, 2014
22

Oxidation Characteristics of Soot from a Gasoline Direct-Injection (GDI) Engine · 2017. 6. 2. · GDI engine regime Conventional diesel engine regime High lube oil consumption Low

Jan 27, 2021

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
  • Oxidation Characteristics of Soot

    from a Gasoline Direct-Injection (GDI) Engine

    Seungmok Choi, HeeJe Seong, Kyeong O. Lee

    Transportation Technology R&D Center

    Argonne National Laboratory

    2014 DOE CLEERS Workshop

    Dearborn, Michigan, USA

    April 30, 2014

  • Background and Objective

    Background

    – Higher PN of GDI engines: gasoline particulate filter (GPF)

    – Most of GDI particulate emissions are formed during cold start (warm-up) and

    transient periods.

    • For GPF regeneration, the oxidation characteristics of cold soot are important.

    – Soot oxidation reactivity: Printex U (flame soot) < Diesel soot < GDI soot

    – Diesel soot oxidation reactivity depends on engine conditions (speed, load,

    EGR, inj. timing): changes in carbon nanostructure and chemical properties

    (organic fractions, SFG)

    – Few studies about GDI soot oxidation characteristics.

    Objectives

    – Investigating GDI soot oxidation characteristics in relations to engine operating

    conditions and TWC effects.

    – Proposing a kinetic correlation relevant to GDI soot which can be used for

    simulation of GPF regeneration

    2

  • Experimental setup with 2.4L 4-cylinder NA GDI

    engine – homogenous/stoichiometric charge strategy

    3

    -540 -480 -420 -360 -300 -240 -180 -120 -60 0 60 120 180

    Crank angle [°ATDC]

    IntakeExhaust

    Singleinjection

    Sparktiming

  • 0

    5

    10

    15

    20

    25

    30

    35

    0 1 2 3 4

    Soo

    t m

    ass

    [μg/

    cyc]

    Ash fraction [mass %]

    1250rpm-25%

    1500rpm-50%

    3000rpm-50%

    Cold idle

    Ash fraction is much higher in GDI soot than in diesel

    soot, primarily due to lower soot mass emissions of

    the GDI engine

    Ash fraction (in mass %) in engine-out soot will vary as function of soot mass emission and lube oil consumption.

    – Ash fraction in soot tends to increase with lower soot mass calibration.

    4

    Adv (330)

    Adv (330) – postTWC

    Rtd (190) Adv (330)

    Rtd (190)

    Adv (330) – postTWC

    Default Adv (330) Adv (330) – postTWC

    GDI soot at advanced or retarded IT shows similar levels of soot mass and ash fraction (order of 0.1%) to diesel soot.

    GDI soot with default calibration contains over 3% of ash due to very low soot mass.

    TWC decreases total soot mass, but increases ash fraction in soot.

    At cold start, GDI soot lies in diesel-like regime with high soot mass and moves to the GDI engine regime as the engine warms up.

    GDI engine regime

    Conventional diesel engine regime

    High lube oil consumption

    Low lube oil consumption

  • GDI soot oxidation reactivity is significantly enhanced

    with increased ash fraction in soot

    Catalytic effect of ash is one of the driving factors that enhances oxidation reactivity of GDI soot.

    5

    0

    20

    40

    60

    80

    100

    0 50 100 150 200

    Printex U

    LD Diesel_Ash 0.36%GDI_Ash 0.10%GDI_Ash 0.58%GDI_Ash 1.35%GDI_Ash 3.36%GDI_Ash 17.27%

    m/m

    0 [

    %]

    Time [min.]

    Increase of

    ash fraction

    0

    20

    40

    60

    80

    100

    120

    0 5 10 15 20

    50% Conv.

    90% Conv.

    Tim

    e [

    min

    ]

    Ash fraction [%]

    PU (90% Conv.)

    PU (50% Conv.)

    Default inj. Timing (Low soot, high ash)

    Advanced inj. Timing (High soot, low ash)

    TGA isothermal oxidation (600 °C, 8% O2, pre-treated by N2)

  • 6

    0

    0.02

    0.04

    0.06

    0.08

    0.1

    0.12

    0.14

    0 0.2 0.4 0.6 0.8 1

    Printex U

    LD Diesel_A0.33%

    GDI_Hot Steady_A0.10%

    GDI_Hot Steady_A0.58%

    GDI_Hot Steady_A3.36%

    m

    /min

    s [

    (mg

    /min

    )/m

    g]

    Conversion ()

    Specific soot oxidation rates

    GDI soot has unique oxidation characteristics different

    from flame or diesel soot

    A

    C

    B

    The oxidation rates of GDI soot are significantly promoted with increase of ash fractions.

    Without ash, oxidation reactivity of GDI soot is lower than that of Printex U.

    GDI soot shows three-staged oxidation which differs from Printex U/diesel soot.

    – Initial stage (A): higher oxidation rate due to additional oxidation of soluble organics and weakly bonded carbons (WBC)

    – Intermediate stage (B): lower oxidation rate after completion of SOF and WBC oxidation (sole carbon oxidation)

    – Final stage (C): soot oxidation rate becomes higher, because of additional catalytic effects of ash by higher ash-to-soot ratio in the remaining sample.

  • TWC decreases SOF/WBC and increases the ash fraction

    in GDI soot, and improves oxidation reactivity in overall

    7

    0

    0.05

    0.1

    0.15

    0.2

    0 0.2 0.4 0.6 0.8 1

    Hot Steady_Eout_ash0.58%

    Hot Steady_TWCout_ash1.35%

    m

    /min

    s [

    (mg

    /min

    )/m

    g]

    Conversion ()

    Engine-out

    TWC-out

    Ash

    SOF/WBC

    Three reasons of ash fraction increase after TWC

    – Catalytic oxidation of organics, WBC, and carbon soot in the TWC

    – Physical loss (attachment on the TWC wall) of soot particles

    • Particles loss was measured by SMPS.

    – Separation of catalyst supporting materials: SEM-EDS data

    • Higher fractions of Mg, Al, and Si were measured from TWC-out ashes.

    Oxidation reactivity of TWC-out soot is enhanced by the increase in ash fraction.

    Specific soot oxidation rates

  • Cold condition GDI soot: ash plays less catalytic roles,

    resulting in lower oxidation reactivity

    Low ash (0.1%) soot represents intrinsic GDI soot oxidation reactivity. (negligible ash effects)

    The oxidation rates of cold GDI soot in the intermediate stage is close to intrinsic GDI soot oxidation rate.

    8

    Hypothesis 1: GDI soot seems to have an “Intrinsic carbon oxidation reactivity” which is unchanged at different engine conditions.

    Hypothesis 2: Ash in GDI soot may have different forms, offering different levels of catalytic effects, depending on hot and cold engine conditions.

    Specific soot oxidation rates

    With similar ash fractions, the oxidation rate is much slower for cold idle soot than hot steady state soot.

    0.01

    0.02

    0.03

    0.04

    0.05

    0.06

    0.07

    0.08

    0 0.2 0.4 0.6 0.8 1

    GDI_Hot Steady_A0.10%

    GDI_Cold Idle_A0.50%

    GDI_Hot Steady_A0.58%

    m

    /min

    s [

    (mg

    /min

    )/m

    g]

    Conversion ()

  • Hypothesis 1: HR-TEM and Raman spectroscopy revealed that

    GDI soot nanostructures are well defined and unchanged by

    engine conditions: “Constant intrinsic carbon soot reactivity”

    9

    0

    0.2

    0.4

    0.6

    0.8

    1

    800 1000 1200 1400 1600 1800 2000

    1800rpm-15%1800rpm-50%1800rpm-75%

    No

    rmali

    ze

    d In

    ten

    sit

    y

    Raman shift (cm-1

    )0

    0.2

    0.4

    0.6

    0.8

    1

    1000 1500 2000

    GDI: cold idle

    GDI: 1250rpm-25% load

    GDI: 1500rpm-50% loadN

    orm

    ali

    zed

    in

    ten

    sit

    y

    Raman shift (cm-1

    )

    Diesel soot nanostructures change significantly with engine conditions. Different carbon soot reactivity

    GDI soot nanostructures do not depend on engine conditions. Constant carbon soot reactivity

    2°CA advanced Inj.

    2°CA Retarded Inj.

    Cold idle

    1500 rpm-50%

    [TEM images: Yehliu, Combustion and Flame, 2013]

    ST

    SLδ ~ 0.1 mmT ~ Tad ~ 2700K

    Unburned mixture

    (λ=1)

    Burned gas

    [Source: Dec, SAE 970873, 1997]

    Conventional diesel engine: spray combustion GDI engine (NA, λ=1): premixed flame propagation

    Flame structures and temperatures of diesel spray combustion strongly depend on ambient P & T at injection, dilution (EGR), and swirl.

    Turbulence increases reaction surface area with highly wrinkled and convoluted flame sheet. Flame sheet structure (thickness) and temperature do not change much at different engine conditions.

  • Hypothesis 2: Three different states of ashes are

    proposed for GDI soot oxidation

    Combustion-derived ash precursor (Ash_C)

    – Metallic oxide nano-particles generated during in-cylinder combustion and soot formation processes.

    – Tight-contact with soot particles in nano-scale, offering strong catalytic effects on soot cake oxidation.

    – Converted to oxidation-derived ash by sintering during soot cake oxidation.

    Unburned oil-derived ash precursor (Ash_U)

    – Unburned oil additives (e.g., ZDDP or calcium sulfonate) discharged to exhaust in cold engine condition.

    Nearly no catalytic effects.

    – Converted to oxidation-derived ash by oxidation and sintering during soot cake oxidation.

    Oxidation-derived ash (Ash_O)

    – Metallic oxide micron-particles generated from the ash precursors with soot cake oxidation.

    – Loose-contact with soot particles, offering weak catalytic effects on soot cake oxidation.

    – Contribution increases at the final soot oxidation stage with increase in ash-to-soot ratio.

    10

    Soot primary particle

    Ash_C

    Ash_U

    Ash_O

    Exhaust emissions Soot cake

  • A global GDI soot oxidation mechanism is proposed

    which includes the effects of organics/WBC and ash

    Assumptions in mechanism

    – Constant intrinsic carbon soot oxidation reactivity

    – Three states of ashes

    – Conversion of ash precursors to Ash_O at the same rate as soot conversion

    11

    0

    0.01

    0.02

    0.03

    0.04

    0.05

    0.06

    0.07

    0.08

    0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

    m

    /min

    s [

    (mg

    /min

    )/m

    g]

    Conversion ()

    Measured oxidation rate

    Carbon oxidation shifted up by

    Ash_C

    Carbon oxidation assisted by Ash_O

    Intrinsic carbon soot oxidation

    Oxidation of SOF/WBC

    Carbon oxidation assisted by Ash_C

    0

    0.01

    0.02

    0.03

    0.04

    0.05

    0.2

    0.4

    0.6

    0.8

    1

    0 0.2 0.4 0.6 0.8 1

    Ox

    ida

    tio

    n d

    eri

    ve

    d a

    sh

    co

    nte

    nt

    [mas

    s %

    ]

    Conversion ( )

    0.10%

    0.58%1.35%

    17.3%

  • A modified kinetic correlation for GDI soot is derived

    in consideration of SOF, WBC and ash effects

    12

    𝛼 = 𝑚𝐶𝑜𝑛𝑣/𝑚0

    𝑟 =𝑑𝛼

    𝑑𝑡= 𝐴 × 𝑒−𝐸𝑎/𝑅𝑇 × (1 − 𝛼)𝑛

    𝑟 =𝑑𝛼

    𝑑𝑡= (𝑚𝑆−𝑊,0/𝑚𝑠𝑜𝑜𝑡,0) ∙ 𝑟𝑆−𝑊 + (𝑚𝐶,0/𝑚𝑠𝑜𝑜𝑡,0) ∙ 𝑟𝐶

    𝑟𝑆−𝑊 =𝑑𝛼𝑆−𝑊

    𝑑𝑡= 𝐴𝑆−𝑊 × 𝑒

    −𝐸𝑎,𝑆−𝑊/𝑅𝑇 × (1 − 𝛼𝑆𝑂𝐹)𝑛𝑆−𝑊

    𝑟𝐶 =𝑑𝛼𝐶𝑑𝑡

    = 𝐴𝐶 × 𝑒−𝐸𝑎,𝐶/𝑅𝑇 × (1 − 𝛼𝐶)

    𝑛𝐶+[𝐴𝑠ℎ_𝑂 𝑎𝑠𝑠𝑖𝑠𝑡𝑒𝑑]

    𝐴𝑠ℎ_𝑂 𝑎𝑠𝑠𝑖𝑠𝑡𝑒𝑑 =𝑑𝛼𝐴𝑠ℎ_𝑂 𝑎𝑠𝑠𝑖𝑠𝑡𝑒𝑑

    𝑑𝑡= (1 − 𝛼𝐶) ∙ 𝑎 ∙ exp 𝑏 ∙ 𝑇 ∙ 𝑓𝐴𝑠ℎ_𝑂,𝑖

    Reaction order (n) of soot samples

    The typical kinetic correlation doesn’t hold for GDI soot oxidation.

    – Activation energy (Ea) and reaction order (n) change with conversion, due to the effects of organics/WBC and ash.

    A modified kinetic correlation has been developed for accurate prediction of soot oxidation.

    – The effects of SOF/WBC and ash are taken into account.

    – Soot oxidation rates can be predicted at different engine conditions without changing kinetic parameters.

    0

    0.01

    0.02

    0.03

    0.04

    0.05

    0.06

    0.07

    0.08

    0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

    m

    /m [

    (mg

    /min

    .)/m

    gin

    sta

    nt]

    Conversion ( )

    GDI soot oxidation mechanism

    -11

    -10

    -9

    -8

    -7

    -2.5 -2 -1.5 -1 -0.5 0

    ln(dα

    /dt)

    ln(1-α)

    Printex U

    GDI-1

    GDI-3

    GDI-4

    GDI-5

  • 0

    0.005

    0.01

    0.015

    400 450 500 550 600 650 700

    ExperimentTypical kinetic correlationModified kinetic correlationCarbon oxidation by Ash_CCarbon oxidation by Ash_OSOF/WBC oxidation

    m

    /m0 [

    (mg

    /min

    .)/m

    g]

    Temperature [degC]

    0

    0.01

    0.02

    0.03

    0.04

    0.05

    0.2 0.4 0.6 0.8 1

    ExperimentTypical kinetic correlationModified kinetic correlationCarbon oxidation by Ash_CCarbon oxidation by Ash_OSOF/WBC oxidation

    m

    /min

    s [

    (mg

    /min

    .)/m

    g]

    Conversion ()

    The modified kinetic correlation predicts accurate

    oxidation rate of GDI soot at a wide ranges of

    conversion and temperature.

    13

    Typical kinetic correlation

    Modified kinetic correlation

    Typical kinetic correlation

    Modified kinetic correlation

    Carbon oxidation by Ash_C

    SOF/WBC oxidation Carbon oxidation

    By Ash_O

    Carbon oxidation by Ash_C

    SOF/WBC oxidation

    Carbon oxidation By Ash_O

    Isothermal oxidation (600 °C, 8% O2) Non-isothermal oxidation

    (increased by 1°C/min, 8% O2)

  • Summary

    Major characteristics of GDI soot oxidation

    – High ash fraction: an order of magnitude higher ash fraction than in diesel soot.

    – Catalytic effects of ash: dominantly enhances soot oxidation reactivity.

    – Three-staged oxidation: additional SOF/WBC oxidation at initial, and Ash_O assisted carbon oxidation at final.

    – Constant intrinsic carbon soot oxidation reactivity, independent to engine conditions.

    – TWC effect: enhances soot oxidation reactivity due to increased ash fraction.

    – Lower oxidation reactivity of cold condition soot: unburned oil-derived ash precursor.

    A general GDI soot oxidation mechanism and a modified kinetic correlation have been proposed.

    – The oxidation rates of GDI soot have been accurately predicted at wide ranges of conversion/temperature and engine conditions, without changing kinetic parameters.

    14

  • Acknowledgement

    Funding

    – U.S. DOE Office of Vehicle Technologies

    Industrial partners

    – Hyundai Motor Company

    – Corning Inc.

    15

  • Thank you for your attention!

    Contact Seungmok Choi

    [email protected]

    16

  • Technical support pages

    17

  • Significance of cold and transient condition soot in

    the GDI engine

    18

    0.004

    0.057

    0.022

    0.206

    0

    0.1

    0.2

    0.3

    Hot steady(1500rpm-25%)

    Cold steady(1500rpm-25%)

    Hot transient(Single ramp)

    Cold transient(Single ramp)

    Soo

    t m

    ass

    [mg/

    g Fu

    el]

  • Contributions of SOF/WBC and ashes on GDI soot

    oxidation of different engine conditions

    19

    0

    0.5

    1

    1.5

    2

    GDI1 GDI2 GDI3 GDI4 GDI5

    No

    rmal

    ized

    So

    ot

    Oxi

    dat

    ion

    Rat

    e

    SOF

    Carbon_Ash_O

    Carbon_Ash_C

    Carbon_intrinsic

  • TWC effects on soot mass/particle size distribution

    and ash composition

    20

    SEM-EDS (Compositions of ash)

    0.E+00

    2.E+06

    4.E+06

    6.E+06

    8.E+06

    1.E+07

    10 100 1000

    dN

    /dlo

    gDp

    [#/

    cm3]

    Mobility diameter [nm]

    EGout_1500-25

    TWCout_1500-25

    0.E+00

    2.E+06

    4.E+06

    6.E+06

    8.E+06

    1.E+07

    10 100 1000

    dN

    /dlo

    gDp

    [#/

    cm3]

    Mobility diameter [nm]

    EGout_1500-75

    TWCout_1500-75

    MSS and SMPS (Soot mass and particle size distribution)

    Soot mass: 1.0 0.2 mg/m3

    Soot mass: 2.5 2.2 mg/m3

    1.5%14.3%

    0.5%0.3%0.9%

    22.0%

    3.2%40.7%

    16.8%

    F0.6%

    Na18.5%

    Mg4.3% Al

    1.2%

    Si8.6%

    P25.0%

    S2.0%

    Ca30.8%

    Zn9.0%

    Engine-Out

    TWC-Out

  • Validation of the modified kinetic correlation –

    Isothermal 600˚C, 8% O2

    21

    0

    0.01

    0.02

    0.03

    0.04

    0.05

    0.2 0.4 0.6 0.8 1

    GDI-1

    ExperimentTypical kinetic correlation (10-90% Param.)Modified kinetic correlationCarbon oxidation by Ash_CCarbon oxidation by Ash_OSOF/WBC oxidation

    m

    /min

    sta

    nt

    [(m

    g/m

    in)/

    mg

    ]

    Conversion ( )

    0

    0.01

    0.02

    0.03

    0.04

    0.05

    0.2 0.4 0.6 0.8 1

    GDI-2

    ExperimentTypical kinetic correlation (10-90% Param.)Modified kinetic correlationCarbon oxidation by Ash_CCarbon oxidation by Ash_OSOF/WBC oxidation

    m

    /min

    sta

    nt

    [(m

    g/m

    in)/

    mg

    ]

    Conversion ( )

    0

    0.01

    0.02

    0.03

    0.04

    0.05

    0.2 0.4 0.6 0.8 1

    GDI-3

    ExperimentTypical kinetic correlation (10-90% Param.)Modified kinetic correlationCarbon oxidation by Ash_CCarbon oxidation by Ash_OSOF/WBC oxidation

    m

    /min

    sta

    nt

    [(m

    g/m

    in)/

    mg

    ]

    Conversion ( )

    0

    0.02

    0.04

    0.06

    0.08

    0.1

    0.2 0.4 0.6 0.8 1

    GDI-4

    ExperimentTypical kinetic correlation (10-90% Param.)Modified kinetic correlationCarbon oxidation by Ash_CCarbon oxidation by Ash_OSOF/WBC oxidation

    m

    /min

    sta

    nt

    [(m

    g/m

    in)/

    mg

    ]

    Conversion ( )

    0

    0.02

    0.04

    0.06

    0.08

    0.1

    0.2 0.4 0.6 0.8 1

    GDI-5

    ExperimentTypical kinetic correlation (10-90% Param.)Modified kinetic correlationCarbon oxidation by Ash_CCarbon oxidation by Ash_OSOF/WBC oxidation

    m

    /min

    sta

    nt

    [(m

    g/m

    in)/

    mg

    ]

    Conversion ( )

  • Validation of the modified kinetic correlation –

    Non-isothermal 1°C/min, 8% O2

    22

    0

    0.005

    0.01

    0.015

    400 450 500 550 600 650 700

    GDI-1

    ExperimentTypical kinetic correlation (10-90% Param.)Modified kinetic correlationCarbon oxidation by Ash_CCarbon oxidation by Ash_O

    SOF/WBC oxidation

    m

    /m0 [

    (mg

    /min

    )/m

    g]

    Temperature [degC]

    0

    0.005

    0.01

    0.015

    400 450 500 550 600 650 700

    GDI-2

    ExperimentTypical kinetic correlation (10-90% Param.)Modified kinetic correlationCarbon oxidation by Ash_CCarbon oxidation by Ash_OSOF/WBC oxidation

    m

    /m0 [

    (mg

    /min

    )/m

    g]

    Temperature [degC]

    0

    0.005

    0.01

    0.015

    400 450 500 550 600 650 700

    GDI-3

    ExperimentTypical kinetic correlation (10-90% Param.)Modified kinetic correlationCarbon oxidation by Ash_CCarbon oxidation by Ash_OSOF/WBC oxidation

    m

    /m0 [

    (mg

    /min

    )/m

    g]

    Temperature [degC]

    0

    0.005

    0.01

    0.015

    400 450 500 550 600 650 700

    GDI-4

    ExperimentTypical kinetic correlation (10-90% Param.)Modified kinetic correlationCarbon oxidation by Ash_CCarbon oxidation by Ash_OSOF/WBC oxidation

    m

    /m0 [

    (mg

    /min

    )/m

    g]

    Temperature [degC]

    0

    0.005

    0.01

    0.015

    400 450 500 550 600 650 700

    GDI-5

    ExpSim_10-90 Param.Sim_ModelSim_Carbon_Ash_tSim_Carbon_Ash_lSim_SOF

    m

    /m0 [

    (mg

    /min

    .)/m

    gto

    tal]

    Temperature [degC]