Jacqueline H. Chen Sandia National Laboratories

jhchen@sandia.gov Princeton Combustion Institute

Summer ShoolJune 23-28, 2019

Part 4. DNS Relevant to Compression Ignition Engines and Stationary Reheat

Gas Turbines

Outline

• Compression Ignition Engines:– 2D lifted DME laminar flames spanning NTC regime– 2D DME mixing layer ignition– 3D turbulent autoignition in n-dodecane temporal jet– 3D multi-injection mixing and combustion at diesel

conditions• Reheat Combustion

• Diesel engines will used for trucks and shipping for the foreseeable future

• Regulations for pollutant emissions are stringent

• Engine design spans a largeparameter space, wide operating regimes, complex multi-physics.

• Requires predictive engineering models to aid design.

• Model development requires more and higher-fidelity data.

Background

Chapter 1. Introduction 3

Figure 1.1: Historical reductions in specific NOx and soot emissions for heavyduty diesel engines [10].

This requires detailed observations at real device conditions in order to develop and

validate the required theory and models.

Such observations could in principle be obtained by experiment. However, the con-

ditions within a real engine are extremely challenging to observe. For instance, the

diesel engine operates in an extreme pressure and temperature environment and at

extremely high levels of turbulence which generates a large range of spatial and

temporal scales, making detailed measurements di�cult.

The other option is to use numerical simulation to investigate representative condi-

tions. Since the governing equations of fluid mechanics are known and the chemistry

of combustion can be adequately reproduced for idealised fuels, simulations which

resolve all physical scales may, in principle, be conducted and the results treated

as a numerical-experiment. Such high fidelity simulations are referred to as direct

numerical simulations [11–14] (DNS). With DNS, all scalar and vector information is

Stanton et al. SAE 2013-23-0094

Dec SAE 1997-97-0873

Figure: Historical pollutant reduction with improving technology

Figure: Dec’s model for conventional diesel combustion.

Experiments and conceptual models

• Great progress from community efforts (e.g. see Engine Combustion Network at: http://www.sandia.gov/ecn/

• Mean flame structure, global observables (LOL, 𝜏) well-documented.

• Diagnostics: Schlieren, LII, PLIF, CL, etc – much can be observed.

• Some things can not be measured, most details can not be resolved (small scales at high pressure).

• Open questions remain (flame stabilization mechanism/ignition dynamics, regime diagram)

Chapter 2. Literature review 31

Figure 2.12: Chemiluminescence imaging of diesel combustion from Ref. [73].Isolated pockets are observed upstream of the flame base due to autoignition.

as a two-stage process, since substantial levels of CH2O and CHO were detected.

Observations of LTC prior to the main autoginition event have been observed in

subsequent studies at similar conditions [20, 74–76]. Imaging of the quasi-steady

lifted flame also detected significant LTC upstream of the mean stabilisation loca-

tion. The presence of the LTC was significant as it demonstrated that autoignition

chemistry can exist in advance of the stabilisation location, and is likely to a↵ect

the LOL either by transitioning to the main autoignition event, or by modifying the

upstream propagation of a premixed flame feature.

The role of LTC on diesel flame stabilisation was further investigated by Idicheria

and Pickett [20]. In that study, the oxygen concentration and sooting propensity of

7

ferred to as the low-soot condition), as shown in the LII images in Fig. 5.

The PLIF images in Fig. 5 show two distinct regions for both 10% and 21% O2 conditions. First, similar to the no-soot condition, a signal begins upstream of the lift-off location and then goes away near the hypothesized fuel-rich premixed reaction zone. However, unlike the no-soot condition, a signal reappears downstream of this location and extends over the length of the image re-gion. For example, in the 10% O2, 1000 K condition, the signal disappears at 50 mm and reappears at 60 mm. Likewise, in the 21% O2, 900 K condition, the signal dis-appears at 40 mm and eventually reappears at 50 mm. From the arguments presented in the previous section and the similarity of the upstream signal structure, any LIF signal upstream of the lift-off length is likely formal-dehyde, formed in a cool-flame that precedes high-temperature reaction. Also similar to the no-soot condi-tion results, the fact that the LIF signal disappears com-pletely when moving downstream suggests that formal-dehyde is consumed in a high-temperature, premixed reaction zone.

The reappearance of an LIF signal downstream of the high-temperature reaction zone is most likely caused by LIF of PAH, a precursor to soot formation that occurs further downstream. The fact that an LIF signal disap-pears and then forms again is useful for interpretation of results in several ways. First, it is unlikely that the reap-pearing signal is from formaldehyde because tempera-tures near the jet center increase with increasing axial distance due to the high-temperature, premixed reaction zone and further mixing with hot combustion products from the diffusion flame. Formaldehyde that is con-sumed during the high-temperature reactions would not be expected to form again if the temperature continues to rise. Second, it is also unlikely that PAH molecules

would form, be consumed, and then form again with in-creasing axial distance. By this reasoning, we argue that the continuous LIF signal—starting upstream of the lift-off length and extending to the fuel-rich premixed reac-tion zone—is not caused by LIF of PAH for these condi-tions. The PAH is therefore found only in the region very close to the soot formation region. Comparing the LII and PLIF images, it is clear that there is some signal just upstream of the soot formation region. Further down-stream, where soot is formed, it is likely that the signal is a mixture of PAH fluorescence and LII of soot.

The results above are also helpful in understanding the no-soot condition results shown in Fig. 3. Before observ-ing conditions where PAH and soot formation occur, it was unknown if part of the continuous LIF signal ob-served in Fig. 3 (e.g., near the region labeled “P”) was caused by PAH. Even though there was no soot formed at the condition in Fig. 3, the formation of soot precur-sors (PAH’s) remained a possibility, especially since the internal regions of the fuel jet are surrounded by a diffu-sion flame—showing that these mixtures are fuel rich. By changing the conditions only slightly from the no-soot condition to form a minimal amount of soot, the region of PAH formation is now defined. The absence of any LIF signal downstream of the region of formaldehyde con-sumption indicates that there is no detectable PAH for the no-soot conditions.

Summarizing the general flame structure at this low-soot condition: a cool-flame (formaldehyde) is formed up-stream of the lift-off length and is consumed at a high-temperature reaction zone. PAH is formed downstream of the high-temperature reaction zone, indicating that formaldehyde and PAH are found in different regions of the fuel jet. We discuss the spatial relationship between formaldehyde and PAH species further in the TSL mod-eling section of the paper.

10% O2, 1000 K

0 10 20 30 40 50 60 70 80

1050

-5-10

LII

HCHO/PAH

OH*

Distance from injector [mm]

P

P

0 10 20 30 40 50 60 70 80

1050

-5-10

LII

HCHO/PAH

OH*

21% O2, 900 K

Distance from injector [mm]

PP

Figure 5. LII, formaldehyde/PAH PLIF and chemiluminescence images for low-soot conditions. Ambient conditions: 10% O2, 1000 K (left) and 21% O2, 900 K (right).

Downloaded from SAE International by University of New South Wales, Tuesday, April 28, 2015Observations

Validation data Conceptual models

Pickett et al, SAE 2005-01-3843 Idicheria, Pickett, SAE 2006-01-3434

ignition delay is around twice as long as the experimental value at the Spray A condition, which complicates the comparison of time-resolved soot results later.

Lift-off LengthFive institutions participating in the ECN have measured the quasi-steady lift-off length at the Spray A baseline condition, with two institutions contributing data at lower ambient temperature and only SNL providing measurements at 1000 K and above. Figure 3 displays the experimental and simulated quasi-steady lift-off lengths as a function of ambient temperature under the same conditions as Fig. 2. For the ambient temperature conditions at 900 K and below, the standard error among the different contributing institutions is provided. For 1000 K and above, the standard error is derived from two independent SNL experimental campaigns. The liftoff length results for the Spray A baseline case are also tabulated in the figure for reference. The ANL result at 900 K is the average of 15 LES realizations, while for the other ambient temperatures only a single realization was performed. The tabulated ANL result for Spray A also includes the standard error.

Figure 3. Experimental and simulated quasi-steady lift-off lengths (H) for Spray A and its parametric variants in ambient temperature.

In comparing the simulated lift-off lengths with the experimental results, we observe good agreement from the POL and UNSW tPDF models, while the UNSW well-mixed model diverges from the experiment at 800 K and 1000 K. The ETHZ lift-off length also diverges at these two temperatures; however, in contrast to the UNSW well-mixed model, the lift-off length at 800 K is instead under-predicted by ETHZ. The ANL lift-off length trend is nearly linear, with a shorter lift-off length than the experiment at 800 K and a longer value at 900 K and above; however, this might be due to the insufficient number of LES realizations for temperatures other than the Spray A case. Although the standard error of the lift-off length for the 15 LES realizations is quite small, fluctuations in lift-off length extend down to 17 mm and up to 24 mm during the transient simulation [23]. More realizations are necessary to build up sufficient statistics for a representative trend at this and other conditions; however, the high computational cost of these simulations is prohibitive. Finally, it is interesting to note that the UNSW well-mixed model matches ETHZ at 900 K and 1000 K even though different mechanisms were used. Clearly, a more in-depth analysis of

the modeling parameters influencing ignition delay and lift-off length among the various models is warranted and should be the subject of future ECN work.

Soot Optical Properties and TEM Imaging for Soot MorphologyKnowledge of the in-situ soot optical properties and morphology is necessary to reduce uncertainty when making optical measurements of soot volume fraction (fv) in flames. For LII measurements calibrated by laser extinction or extinction imaging measurements as will be presented below, the dimensionless extinction coefficient (ke) is a critical parameter in relating the measured optical attenuation (I/I0) to fv through Bouguer’s Law. Because ke is not only dependent on particle refractive index, but also morphology, care must be taken to account for the effect of soot particle aggregation on the scattering-to-absorption ratio (αsa). Provided the appropriate particle and aggregate parameters can be determined, the influence of aggregation can be captured by the Rayleigh-Debye-Gans approximation for fractal aggregates (RDG-FA). The information on particle size and morphology necessary to derive ke via RDG-FA calculations includes the complex refractive index (m), primary particle diameter (dp), fractal prefactor (kf), fractal dimension (Df), and aggregate size (Np). This section provides a brief explanation of the soot refractive index selected for use by the ECN along with a review of TEM imaging results from ECN participants and others providing the necessary information to determine ke in high-pressure spray flames.

Numerous studies have investigated the refractive index of flame-formed soot as well as other types of carbonaceous particles as reviewed by Smyth and Shaddix [46] and Bond and Bergstrom [47]. While the refractive index applied for extinction measurements can be dependent on a variety of parameters including the wavelength of the incident light and soot composition (i.e., carbon-to-hydrogen ratio), the ECN community has adopted the refractive index proposed by Williams et al. [48] of 1.75-1.03i. Although Williams et al. indicated that this refractive index was specific to measurements at 635 nm, we apply this value as the standard across measurements in the visible spectrum to simplify the comparison of ECN soot measurements performed by different institutions and with different incident wavelengths. The consequences of choosing a constant refractive index for the analysis of soot in spray flames is discussed later in the context of the ke derived from RDG-FA calcualtions. More information on this subject can be found in Manin et al. [11].

With regard to primary particle size, TEM analysis published in Cenker et al. [14] showed a count median diameter of 9.1 nm for soot sampled 60-mm from the injector orifice under the Spray A condition. While this was the only spatial location sampled for Spray A, TEM analysis was also performed on samples extracted at three axial locations, namely, 36 mm, 45 mm, and 60 mm from the 21% O2 parametric variant of Spray A. Although the higher ambient oxygen concentration is known to impact the soot onset timing as well as the rate and location of soot burnout, these data provide relevant insight into the axial evolution of particle morphology. Representative TEM images from the soot formation and oxidation regions as well as the particle-size histograms from soot sampled at all three locations are shown in Fig. 4. For this condition, the 36-mm location is slightly

Skeen et al / SAE Int. J. Engines / Volume 9, Issue 2 (June 2016)888

Downloaded from SAE International by University of New South Wales, Wednesday, May 25, 2016

Skeen et al. SAE 2016-01-0734

LES and RANS

Figure 5. Top two rows: Time-series of fv (IFPEN, LII) and KL (SNL, DBIEI) for Spray A (1.5-ms injection duration). The dashed lines designate the location selected for the radial cross section profiles shown in Fig. 6. Bottom five rows: Time-series of fv as simulated by ANL, ETHZ, POL, UNSW, and UW. Note that the UNSW tPDF

and WM results are presented in a single image.

With the exception of the earliest timing shown in Fig. 6, the experimental radial cross sections of fv from IFPEN and SNL are remarkably similar both in shape and in amplitude. This is in spite of the requirement, discussed above, that the projected KL data be interpolated for multiple viewing angles before tomographic reconstruction. The larger fv measured by DBIEI at 0.9 ms may be an indication that this diagnostic is more sensitive to small quantities of soot because the attenuation is path-integrated. Nevertheless, further work should be performed in a single combustion vessel for confirmation.

As a second means of vetting the IFPEN and SNL Spray A soot datasets, a comparison of the total soot mass measured by the two diagnostics as a function of time is presented in the top panel of

Fig. 7. For reference, we have included SNL data acquired in 2012 with the original DBIEI setup using 406-nm and 519-nm incident light. Recall that the IFPEN data were acquired as a single shot at a specific instant in time for several repeated injection events, while the DBIEI data were acquired at 30-45 kHz using high-speed imaging. With the exception of the soot mass determined from the measurements with 406-nm incident light, the IFPEN and SNL data again show reasonably good agreement. The larger soot mass measured with 406-nm incident light may be related to molecular absorption of large PAH and/or a different dispersion exponent precluding the use of a constant refractive index. Manin et al. [11] discussed this in more detail.

Skeen et al / SAE Int. J. Engines / Volume 9, Issue 2 (June 2016) 891

Downloaded from SAE International by University of New South Wales, Wednesday, May 25, 2016

Skeen et al. SAE 2016-01-0734 (2016) [5]

Pei et al. CNF 168 (2016) [6]

Figure: Ignition transient. TPDF modelling of Spray A.

Figure: Comparison of models against experiment. Review of ECN progress on soot modelling (SibenduSom et al. 2016).

• LES and RANS needed for industry• Performance is good for global observables, less good for fine details• Difficult to test and discriminate between models due to sparse (and unresolved) data• Motivation for DNS – augment existing knowledge with fully-resolved information. Care

needed!

Background on Diesel Spray Flame LES/RANS Simulations

• Diesel spray flames are typicalnon-premixed flames

• Critical features of diesel sprayflames: Spray properties, Ignitiondelay, Flame lift-off, Emissions, …

• Challenges in diesel flamesimulations

LES of a Spray A flame

– Strong turbulence– Fully transient processes– Premixed flame fronts are involved

• Most current combustion models arespecific to premixed or non-premixedflames

• Objective: to create accurate combustion models for LES and RANS applicable to both non-premixed & premixed flames in the presence of autoignition (low- and high-temp.)

Sibendu Som et al.

Diesel Flame Condition: Spray A

Parameter Quantity

Fueln-dodecane

(54 species skeletal mechanism)

Nozzle outlet diameter 90 µm

Discharge coefficient 0.89

Fuel injection pressure 1500 bar

Fuel injection temperature 363 K

Injection duration 1.5 ms

Ambient gas temperature 900 K

Ambient gas density 22.8 kg/m3

Ambient oxygen concentration 15 %

LES computational domain 60 x 45 x 45 mm

Spray A, n-Dodecane

http://www.sandia.gov/ecn/

Target experiment apparatus

Importance of Premixed Flames and Ignition in Diesel Spray Flames

• Ignition delay is controlled by the combined effect ofautoignition kinetics of a given fuel together with the mixinghistory

• Lift-off length is affected by the local premixed combustion modedue to competition between autoignition and premixed flamepropagation

• Need to be accounted for in diesel spray flame modeling

Importance of Premixed Flames and Ignition in Diesel Spray Flames

Regime Classification for ExothermicReaction Fronts in Stratified Mixtures

Deflagration (Premixed Flames) Versus Spontaneous Ignition (Zeldovich 1980)

• Mixtures can contain a spatial gradient of ignition delay time:– Spatial gradient of temperature– Spatial gradient of concentration– Spatial gradient of reactivity (mixing of two fuels

with different ignition properties)

• Large spatial gradients in ignition delay lead todeflagration– Balance between reaction and transport

• Small gradients in ignition delay lead to spontaneousignition– Reaction dominates over transport

• Turbulent mixing can modify scalar gradients

How to make a DNS diesel-relevant?• The actual diesel condition may never be feasible with DNS

– ReJET ~ 5e5, variable and extremely complex fuel, multi-phase behaviors, radiation, sparse validation data (…and many more limits!)

• Need to target facets of the problem, play to the strength of DNS– Identify useful but achievable questions of interest that can not be answered

by other means– Contrive simplest possible DNS to conduct investigation

QuestionWhat role does NTC and two-stage ignition have on the ignition and flame-

stabilization processes?

ConfigurationsTargeting high pressure and NTC behaviour

1. Lifted laminar flames 2. Mixing layer ignition 3. Turbulent jet ignition

Lifted laminar flamesOverview

• Conventional diesel combustion resembles quasi-steady lifted flames.

• Stabilization mechanism unclear, but has important consequences.

• Goal: Investigate flame structure and stabilization mechanism across range of diesel-relevant temperatures (span NTC).

• Fuel is DME, features NTC, two-stage ignition. Similar cetane number to diesel.

• 30 species reduced mech, validated.ConditionsØ P = 40 atmØ TOX = 700à1500 KØ Fuel: DME 70% N2 30%.Ø Oxidiser: O2 21% N2 79%. 𝜉 0

1

𝑈

𝑆𝑦𝑚𝑚𝑒𝑡𝑟𝑖𝑐𝑂𝑢𝑡𝑓𝑙𝑜𝑤

𝑂𝑢𝑡𝑓𝑙𝑜𝑤

𝐹𝑢𝑒𝑙𝐴𝑖𝑟

𝜎

Figure: Initial and boundary conditions.

Krisman, 2016

Lifted laminar flamesStabilization transition

Krisman et al. PROCI 35 (2015)

Figure: OH budget analysis reveals gradual flame to ignition transition (Krisman et al. 2015)

• Complex transition. One or two upstream branches observed (tetrabrachial, pentabrachial).

• Transition smooth, wide range of hybrid structures.• Flame speeds (non-turbulent) O(10) m/s, comparable

with stabilization location in diesel flame simulations

Chapter 4. Lifted laminar flames 62

Table 4.1: Values of variable parameters for each case.

Case Tox [K] U [ms�1]1 700 1.252 900 3.503 1100 5.754 1300 10.005 1500 45.00

near-constant lift-o↵ length. Table 1 lists the oxidiser temperature and inlet velocity

for each case.

The chemical mechanism is a 30 species reduced mechanism described in Ref. [117]

and introduced in section 3.1.5.1. As earlier discussed in section 3.1.5.1, the mecha-

nism was obtained by a reduction of the mechanism developed by Zhao et al. [118]

using reduction methods outlined by Lu and Law [134]. The accuracy of the re-

duced mechanism was validated against the detailed mechanism in Ref. [117]. A

mixture-averaged molecular transport model was employed [14].

4.3 Grid convergence

To assess the chemical length scale, a stabilised, lifted, one-dimensional simulation

of a stoichiometric and premixed DME flame was conducted at a pressure of 40

atmospheres and a temperature of 900 K. These conditions were selected as repre-

sentative for the two-dimensional configurations implemented in Chapters 4 and 5.

Figure 4.1 shows the results for several key variables. The results show that a resolu-

tion of one micron is su�cient to correctly resolve the reaction zone. The resolution

of 1 micron implies an acoustic time step of at most tCFL = 1 ns, much smaller than

the smallest chemical time scale for the DME mechanism [117], tCHEM = 5 ns. This

implies that the time step size for DME simulations is limited by the acoustic CFL

constraint, rather than the chemical time scale.

Table: Parametric variation across cases.

Edge flame propagation Hybrid Autoignition

Increasing T and inlet velocity

Lifted laminar flamesUpstream branches

• Upstream branches due to both first and second stage of auto ignition.

• Location of upstream branches correlated to 0D ignition delay times.

• LTC radical species clearly mark ‘cool-flame’ branch.

• Formaldehyde upstream of stabilization point does not necessarily indicate LTC.

Figure: Zero-D ignition delay times [Krisman et al. 2015]

Figure: Chemical structure of polybrachial flames (Krisman et al. 2015)

Mixing layer ignition - overview

• Aim – observe ignition transient at same conditions where polybrachialflames observed.

• Configuration – 2D, no-shear, isotropic pseudo-turbulent mixing layer. Da=0.4.

• Thermochemical state – identical to 900 K case from lifted laminar study. Figure: Initial and boundary conditions

Krisman et al. PROCI 36 (2016)

Figure: Snapshots of HRR multi-stage, multi-mode ignition process

Mixing layer ignition - low temperature chemistry

Figure: Autoignition to flame transition for LTC [Krisman et al. 2016]

Figure: Cool-flame advances ignition, shifts to richer mixtures [Krisman et al. 2016]

Figure: Cool-flame propagates in composition space [Krisman et al. 2016]

• LTC initiates similar to 0D case.

• Rapidly develops diffusive ’cool flame’ nature.

• Propagates into rich mixtures and advances ignition.

• HTC ignition very different to 0D case.

DNS of a Turbulent Autoigniting n-Dodecane temporal jet at 25 Bar

Giulio Borghesi1, Alex Krisman1, Tianfeng Lu2 and Jackie Chen1

1Combustion Research Facility, Sandia National Laboratories2University of Connecticut

Ketohydroperoxide mass fraction

Borghesi et al. Combustion and Flame, 2018

Background and Objective

• Low-temperature combustion (LTC) aims at increasing fuel efficiency and reducing emissions

• Under LTC conditions, combustion occurs in a mixed mode and in multiple ignition stages

• Ignition is now very sensitive to the fuel chemistry, especially to the low temperature reactions branch

Question: How does transport and low-temperature chemistry affect ignition in low-temperature diesel combustion?

time

Background on high-temperature ignition

• Homogeneous PSR calculations show the existence of x value, xm, where ignition delay has a minimum;

• Flamelet simulations show the ignition delay increases with scalar dissipation N until a critical value is reached;

• In practical systems, ignition occurs at locations close to xm where N is low. The ignition delay is longer than in PSR;

• Question: which features of high-T ignition carry over to low-T ignition?

us to project with some confidence homogeneous calculations (thatare done with modest computational cost) to inhomogeneoussituations.

If the fuel is distributed in the oxidizer in a manner that globalmixing may occur quickly and the extrema of x shrink so that thenominal xMR (from one of the methods discussed above) ceases toexist in the flow, the evolution of temperature in x space can besomewhat different than the above picture [95]. Nevertheless, thefundamental finding that a competition between the hightemperature at low x and the high concentrations at higher x leadsto a preferred mixture fraction and that this value does not changemuch with flow configuration, remains valid.

A situationmore relevant to HCCI combustion, with autoignitionachieved by compression, has also been studied. DNSwith one-stepchemistry has shown that if the fuel and air have the sametemperature, autoignition happens around the stoichiometricmixture fraction [107], although notmany details were included forthe temperature in x space. A 2-D simulation using a detailedscheme for hydrogen and including temperature non-uniformitiesbut uniform composition, showed the difference between localautoignition and deflagration following a nearby autoignition site[14]; this different behaviour has also been observed experimen-tally [108]. Earlier, Hasegawa et al. [109] reported similar localisedautoignition in a fully-premixed gas with superimposed tempera-ture fluctuations with a 2-D simulation and one-step chemistry.Strictly speaking, these should be considered as premixedcombustion problems.

There is very limited experimental evidence on the autoignitionsite structure in mixture fraction space. The ambient pressureexperiment by Cabra et al. at the University of Berkeley [110]reported point Raman–Rayleigh and Laser-Induced Fluorescence(LIF) measurements in a nitrogen-diluted hydrogen jet at about100 m/s issuing from a 4.57 mm pipe into a co-flow of hot productsfrom a lean hydrogen flame. The co-flow velocity was low (about3.5 m/s) compared to the jet velocity, so the turbulence was createdby shear. The co-flow was at 1045 K and had an oxygen molefraction of 0.15. In the experiment, no extra ignition source wasavailable, but a lifted flame was evident, and hence this problem isa proper turbulent non-premixed autoignition problem. Plots ofsimultaneously-measured temperature and mixture fraction showlarge scatter, with an overall gradual transition from the inertmixing line to the fully-burnt line over an axial distance of manyjet diameters. The data suggest that autoignition (i.e. high

temperature) kernels are located at mixture fractions leaner thanxst, although a quantification of xMR was not attempted. Later datafrom a methane jet [111], with the co-flow temperature now at1350 K, also showed that the highest temperature across mixturefraction space evolves from a very lean x to the stoichiometric withdownstream distance (Fig. 9), in qualitative agreement with theshapes from the examples of laminar flamelet simulations shown inFig. 4 and the DNS (Fig. 6). In the methane case, it looks as if theinitial temperature increment occurs at very lean mixture fractions,consistent with homogeneous autoignition calculations of sign inthat paper that show a minimum at an equivalence ratio of 0.05,corresponding to xMR¼ 0.01 for the jet dilutions used.4 The devel-opment between these lean mixture fractions and the stoichio-metric, when examined in terms of conditional averages, seemsgradual (see also Fig. 5 of Ref. [82] for transient flamelet methanecalculations at 40 bar demonstrating the same trend). Theseexperiments have been the subject of various modelling efforts,which will be discussed later. Velocity data are also available [112].Fuels other than H2 and CH4 must also be examined in thisconfiguration, perhaps following the hierarchy of the laminar flameautoignition studies, and the effect of pressure must be studied.Development of appropriate diagnostics that can measure themixture fraction for such fuels would also be very fruitful.

2.2.2. Low scalar dissipationThe DNS shows that not all regions with x¼ xMR autoignite

simultaneously [11]. Hence the autoignition process is ‘‘spotty’’ andwe can talk about local autoignition kernels. Such kernels havebeen visualised in virtually all the DNS performed so far, with eitherone-step or complex chemistry, in turbulent non-premixed flows.

In accordance with our expectation from laminar flow auto-ignition, it turns out that the regions that autoignite first are thosethat have low scalar dissipation. This was first revealed with one-

Fig. 7. Conditionally-averaged reaction rate against mixture fraction from one-stepDNS. The times at which the conditional averages were made, normalized by theresulting autoignition time for this simulation (with autoignition defined as the firstappearance of a burning spot), are 0.67 (open circles), 0.86 (filled circles) and 0.99(triangles). Reprinted from Ref. [11], with permission from Elsevier.

0 0.02 0.04 0.06 0.08 0.1

mixture fraction Z [-]

1

2

3

4

5

6

7

8

igni

tion

del

ay t i

gn [m

s]

Li et al.Del Alamo et al.O’Conaire et al.Yetter et al.

920

925

930

935

940

945

T [K

]

Fig. 8. Calculations of autoignition time in homogeneous hydrogen-air mixtures at1 bar with various detailed mechanisms: Li et al. [104], Yetter et al. [80], Del Alamoet al. [105], and O’Conaire et al. [106]. The hydrogen mass fraction in the fuel stream isY1,0¼ 0.13 (the rest in N2), T2,0¼ 945 K, T1,0¼ 720 K. The initial species mass fractionsand temperature (dashed line; right axis) are functions of the mixture fraction cor-responding to frozen mixing. sref is the minimum value (that depends on the scheme)and xMR the corresponding mixture fraction. Calculations performed by Frouzakis andKerkemeier from ETH Zurich. Reproduced from [102].

4 Care is needed with the exact determination of autoignition times for such verylean mixtures that result in a very small temperature increment from the unburntto the burnt state.

E. Mastorakos / Progress in Energy and Combustion Science 35 (2009) 57–9768

The same observations have been found in turbulent situations,which constitutes our first insight into the autoignition of turbulentnon-premixed combustion, and has come virtually completelyfrom DNS with various configurations, codes, chemistries andconditions. For most studies, T1,0< T2,0, so autoignition is accom-plished as in the laminar canonical flows discussed in the previoussub-section. The first flow pattern studied was the temporal mixinglayer (Fig. 1b). Initially, a straight interface separated the fuel fromthe air and homogeneous isotropic turbulence was introduced.Limited results were also obtained from slabs of cold fuel exposedto hot air from either side [11] and small ‘‘blobs’’ of fuel in air [93].The DNS code used for these simulations was NTMIX, developed atCERFACS and IFP in France, and was a fully-compressible code withone-step chemistry simulating two-dimensional turbulent flows.No forcing was used and hence the turbulence was decaying. Thestoichiometry of methane was used in these simulations. A typicaldistribution of the normalized temperature rise above the initial(inert) value Tin(x), defined as b¼ [T" Tin(x)]/[Tb(xst)" Tin(xst)] withTb(xst) the burnt temperature at the stoichiometricmixture fraction,is reproduced from Ref. [11] in Fig. 6. It is evident that ignitionhappens suddenly and that it takes a substantial time for the flameto acquire the expected shape for complete non-premixedcombustion. In more detail, Fig. 7 shows the reaction rate, condi-tionally-averaged on the mixture fraction for various times duringthe induction period. It is evident that the reaction rate peaks at a xbetween 0.10 and 0.15, slightly shifting to higher values with time.The value predicted by the analysis of Linan and Crespo [10] forthese conditions (of T1,0, T2,0 and Tact; these are the key parametersthat affect the location of autoignition in mixture fraction space inthe analysis with one-step chemistry) is about 0.12. By reference toFig. 7, we may call most reactive mixture fraction, xMR, the value of xwhere the reaction rate becomes amaximum.We see that xMR is notequal to the stoichiometric mixture fraction (0.055 for methane).The simulations also showed an insensitivity of xMR to the turbulenttimescale, lengthscale, and initial mixing layer thickness.

The point raised above, i.e. that autoignition occurs away fromstoichiometry and always around the same mixture fraction thatdepends little on mixing rates and mixture fraction patterns, hasbeen confirmed with a wide range of additional DNS results withsimplified and complex chemistry, all of which reveal a similarstructure of the autoignition sites in mixture fraction space whenthe fuel is colder than the air. Reported work includes simulations

with one-step chemistry in 2-D [94] and 3-D [95] with initialmixture fraction distributions from a turbulent spectrum ratherthan a straight fuel-air interface, simulations with hydrogendetailed chemistry in 2-D [81,96,97], and with a four-step heptanemechanism in 2-D [98] and 3-D [99]. In all these, the turbulence isdecaying. Problems with flow include the supersonic mixing layer(in 2-D) [100], jets of hydrogen in slow co-flow of heated air (in3-D) [101], and jets of hydrogen in fast co-flowing turbulent heatedair (in 2-D) [102]. An important difference of the detailed chemistryresults compared to the one-step results is that the location of thepeak of the conditionally-averaged heat release rate may shift bya large amount towards the end of the induction time, and that thisshift depends on the scalar dissipation. However, this shift may ormay not be significant, depending on the fuel, the chemicalmechanism used and the conditions (e.g. T2,0 and T1,0; the effect ofpressure on xMR has not been studied explicitly yet). We can alsoobserve in Fig. 7 that the reaction zone has significant thickness inmixture fraction space. This may have been due to the relativelylow activation energy used in the one-step chemical scheme usedfor that simulation, but the fact that pre-ignition reaction zones arerelatively wide is borne out by detailed-chemistry DNS and laminarflame simulations as well.

Is there a way that we can pre-compute xMR? Three separatemethods may be used, all giving results that are reasonably similar.First, a series of homogeneous reactor calculations can be per-formed, with initial conditions of species and temperature corre-sponding to frozen mixing between the cold fuel and the hot air,which can hence be fully parameterised by the mixture fraction x.Plotting the autoignition time sign from each of these vs. the cor-responding x then results in a curve with a minimum. At very leanmixtures, the temperature is high but there is little fuel. As xincreases, the temperature decreases but the fuel concentrationincreases. The competition of these two effects results in anoptimum composition, giving the fastest autoignition time. Theminimum autoignition time, called the reference autoignition timesref, can be used as a characteristic timescale of the autoignition ofthe particular non-premixed flow corresponding to no mixing. Themixture fraction at which this minimum occurs can be consideredas a surrogate for xMR.

Fig. 8 shows an example of such simulations for hydrogen, forthe conditions of the experiment of Markides and Mastorakos[103]. For these simulations, four different chemical mechanismshave been used [80,104–106]. Note that with detailed chemistryand variable cp, the initial distributions of species mass fractionsand the enthalpy will still be linear functions of x, but thetemperature will not. Unity Le is assumed. It is clear thata minimum exists; in this case and using, say, the mechanism ofO’Conaire et al. [106], we would conclude that xMR¼ 0.04 andsref¼ 1.05 ms. Fig. 8 also shows a sensitivity of the predictions to thedetailed mechanism. Despite the expectation that detailed mech-anisms of hydrogen are probably the best known, differences areevident, although some convergence may be observed, with themore recent mechanisms producing increasingly similar results.

Second, wemay use the asymptotic analysis results from Ref. [9]for constant-strain problems (also summarised in Ref. [49]) or fromRefs. [10,90] for unsteady problems. Note that these are derived forone-step chemistry, but with a proper estimate of the activationenergy they are applicable to real fuels as well (at least ina temperature range where the autoignition of the fuel behaves inan Arrhenius manner). Finally, we may perform laminar mixinglayer simulations and examine the locationwhere the reaction ratepeaks or autoignition occurs, e.g. from graphs as in Fig. 4. Thisapproach can fully incorporate differential diffusion effects andsome effects of strain [81,88].

For the conditions of Ref. [11], xMR from the first method abovegave a value about 0.11, the asymptotic analysis gave 0.12, while the

1 10 100 1000

peak scalar dissipation rate N0 [1/s]

1.0

1.5

2.0

2.5

3.0

igni

tion

del

ay [

ms]

detailed mechanism − VODPKreduced mechanism − CHEMEQ2reduced mechanism − VODPK

Fig. 5. Calculations of autoignition time of heptane in an unsteady flamelet betweenhot air and cold fuel with unity Lewis numbers under constant scalar dissipation withreduced and detailed mechanisms. Reprinted from Ref. [84], with permission fromElsevier. The potential effect on the results of the numerical solver is also evident.

E. Mastorakos / Progress in Energy and Combustion Science 35 (2009) 57–9766

Figure: ignition delay time in PSR and in nonpremixed flamelet simulation1[1] E. Mastorakos, PECS (2009), pp. 57-97

Low Temperature Diesel Combustion Experiments – Engine Combustion Network

Skeen et al. PROCI 35 (2015) Dahms et al. PROCI 36 (2017)

DNS Configuration and Physical Parameters• Pressure: 25 bar

• Air stream: 15% XO2+85% XN2, T=960 K

• Fuel stream: n-dodecane at 𝝃=0.3, T=450 K

• Kinetics: 35-species non-stiff reduced (Lu)

• Fuel jet velocity: 21 m/s, Rej = 7000, Ret ~ 950

• Code and cost: S3D Legion, 60M CPUh

• Setup:

– 3 billion grids

– 3 microns spatial grid resolution

– Dimensions: 3.6 mm x 14.0 mm x 3.0 mm

– 1 ms of physical time with 4 ns timestepsto observe ignition and propagation of burning flames throughout the domain

– BCs: X and Z periodic, Y NSCBC outflowsFigure: H2O2 mass fraction at t=0.17 ms after start of reactions

Homogeneous Multi-Stage Autoignition

Temporal evolution of selected reactive scalars

Homogeneous ignition delay time

1st Stage, Low-T

2nd Stage, High-T

𝝃st

3-4X

Dynamics of 2-Stage Ndodecane Ignition in a Jet at Diesel Conditions

Rendering by Chris Ye, Min Shih, Franz Sauer, and Kwan-Liu Ma

H2O2 mass fractionKetohydroperoxide and T,K (>1150K)

Conditional statistics reveal ignition dynamics

Temperature

H2O2

Ketohydro-peroxide

time

Conditional mean temperature, KET, and H2O2 reveal cool flame propagation and spontaneous ignition

assisted by turbulent diffusion

Temperature KET H2O2

Low N

High N

Low scalar Dissipation

High scalar Dissipation

For low scalar dissipation, cool flames propagate as a wave towards rich conditions while for high dissipation, turbulent diffusion of heat and mass leads to spontaneous ignition where low-T ignition is impeded whereas high-T ignition is accelerated

time

Propagation mechanism for low-T reaction front

Figure: Fraction of reactive low-T fronts propagating as a flame

Low-T fronts propagate through diffusively supported flame

Sketch of reaction / diffusion balance along normal for KET for flame and ignition kernel

timee

Flames + spontaneous ignition

FlamesFlames

𝑫𝒂𝒌𝒆𝒕 =𝒎𝒂𝒙 (𝒎𝒂𝒈 𝝎𝒌𝒆𝒕 )𝒎𝒔𝒙 (𝒎𝒂𝒈 𝒅𝒊𝒇𝒇𝒌𝒆𝒕 )

< 5 is a cool flame

Effect of Scalar Dissipation Rate on Low Temperature Ignition

time

temp KET H2O2 CDF of 𝑥

timeInc 𝑥

Turbulent versus homogeneous ignition

Low-T and high-T ignition in jet can be faster and than in a PSR !

Low T

High T0.01

0.5

Cool flame propagation &

Turbulent diffusion

Borghesi et al. 2018, Borghesi and Mastorakos, 2015, and Krisman et al. 2017

Conclusions• Low-temperature reactions create the conditions for high-temperature

ignition to occur faster than under homogeneous conditions;• Low-temperature front appears to propagate through a diffusively

supported cool flame;

• High scalar dissipation appears to delay low-temperature ignition; however, it leads to faster ignition at very rich mixture conditions;

• High-T ignition starts at conditions richer-than-homogeneous conditions (x=0.16 compared to x=0.12). Edge flames are seen to form around xst. High-T flame ignites mainly by propagation of rich premixed flames following hot ignition to xst.

DOE Exascale Computing Project (ECP) willachieve capable exascale machines in 2021-2023

Software Technology

Scalable and productive

software stack

Application DevelopmentScience and

missionapplications

Hardware TechnologyHardware technologyelements

Exascale Systems

Integrated exascale

supercomputersapplicati

Correctness Visualization DataAnalysis

AppliCo-Design

evelopment environmenandruntimes

ToolsProgrammingmodels,

d t, Math libraries andFrameworks

System Software, resourcemanagement threading,

scheduling, monitoring,andcontrol

Memory andBurst

buffer

Data management I/O and file

system

Res

ilien

ce

Wor

kflo

ws

Node OS,runtimes

Hardwareinterface

ECP’s work encompasses applications, system software, hardwaretechnologies and architectures, and workforce development

From Paul Messina’s ASCAC talk April 19, 2017

ECP application: transforming combustion science and technology through exascale simulation (Pele)

Mechanism Generator (RMG)

Fuel Conditions

Rate Rules Specific Rates

Mechanism

Reaction(s) Theoretical Rate

Constants (EStokTP)

Uncertainty

Chemical Sim.

Observables Sensitivities

Chem. Sim. Acceptable yes no

Mechanism Reduction DRGASA

CFD Simulations of RCCI

Engine Sim. Acceptable yes no

DONE

1st Iteration No Yes

UQ range

Method

Dakota

Automated Mechanism Generation

PeleC and PeleLM: Block-structured adaptive mesh refinement, multi-physics: spray, soot, and radiation, real gas, complex geometry

S3D: multi-block compressible reacting DNS multi-physics validation: spray, soot, radiation

Effects of reactivity stratification at:

• high pressure• high turbulence • fuel blends

on:• ignition delay • combustion

rates • emissions

Direct Numerical Simulation of Multi-Injection Mixing and Combustion at Compression Ignition

Engine ConditionsM. Rietha, M. Dayb, T. Luc, M Kweond, J. Temmed,

and J.H. Chena

aSandia National LaboratoriesbLawrence Berkeley National Laboratory

cUniversity of ConnecticutdUS Army

32/17

xxxxxxxx

Objectives

1st inj.Pilot Dwell

2nd inj.Main

t ime

Inje

ctio

n

rate

Sandia Spray A experiment (www.crf.gov)

Skeen et al., JSTOR, 2015

• Main injection sees very different conditions compared to pilot injection• How do mixing and different thermo-chemical conditions affect ignition

of the main injection? (e.g. what determines whether cool flame or high-temperature products are produced by pilot)

33/17

xxxxxxxx

• Multi-injection is a strategy used in DI-CI engines to improve emissions and noise while max. fuel economy

Two Cases

Ground transportationHigh pressure (60atm), moderate temperature (900K) Exhaust gas recirculationPilot 0.5 ms, dwell 0.5 ms, main 0.5 msImprove mixture formation, reduceemissions

High-altitude operation (UAV)Moderate pressure (10atm), low temperature (750K) No exhaust gas recirculationPilot 0.208 ms, dwell 0.992 ms, main 1.138 msReduce signature (IR, smoke), improve ignition reliability, improve operation with various fuels

Conditions lead to significantly different ignition processes34/17

Images: www.challenger.com, http://insideunmannedsystems.com

xxxxxxxx

Physical Parameters for Multi-Injection Pulsed Jet DNS

Ground conditionsPurely gaseous n-dodecane/air jet @T=470K, Z=0.45 Ambient conditions: 900K, 15% O2, 85% N2, 60atm Conditions adapted from Dalakoti et al. (ICDERS, 2017.)→ downscaled ECN Spray A, keeping Da constant (Re≈19,000, Dajet≈0.02)Multi-injection: 0.5 ms pilot, 0.5 ms dwell, 0.5 ms main

UAVPurely gaseous n-dodecane/air jet @T=450K, Z=0.37 Ambient conditions: 750K, 21% O2, 79% N2, 10atmMulti-injection: Pilot 0.208 ms, dwell 0.992 ms, main 1.138 ms

35/17

xxxxxxxx

Ground and UAV operation –multi-stage homogeneous ignition

1000

1500

T[K

]

Z = 0.0417

0.00

0.25

0.50

2000 0.75

HR

R[J

/m3 /

s]

⇥ 1011

0.000

0.005

0.010

0.015

Y[-]

OC12H23OOH H2O2 ·3OH·5CH2O

Low-temperature ignition

High-temperatureignition

Exemplary ignition sequence

.

0.000 0.005 0.015 0.0200.010time t [s]

0.02

0.04

0.06

0.08

0.10

0.12

0.14

Mix

ture

frac

tion Z

[-]

T [K]

750

1000

1250

1500

1750

2000

2250

2500

0.0000 0.0002 0.0004 0.0006 0.0008 0.0010time t [s]

0.02

0.04

0.06

0.08

0.10

0.12

0.14

Mix

ture

frac

tion Z

[-]

T [K]

800

1000

1800

1600

1400

1200

2000

2200Ground UAV

Low

T

Hig

hT

Low

T

Hig

hT

36/17

Stoich. mix frac.

Stoich. mix frac

time time

xxxxxxxx

• PeleLM – low-Mach adaptive mesh refinement code based on AMReX• Spectral deferred correction scheme for fluid dynamics-chemistry coupling• Multi-physics: soot, radiation, sprays, embedded boundary for geometry• Code is open-source at https://amrex-combustion.github.io/• Resolution ~1.25 micron required for rich premixed cool flames,

currently 5 micron for full multi-injection run• Size of simulation: ~1B cells (O(100)B cells full run without AMR)• 35 species reduced n-dodecane mechanism (Yao et al., 2017;

Borghesi et al., 2018)

37

PeleLM Code & Numerical Setup

Emmett et al., arxiv, 2018. E. Motheau AMReX gallery.

Multi-injection ignition sequence (ground)

38

Mixture fractions track fluid from individual injections (two additional transport equations)

Low temperature species

High-temperature species

*Movie does not show full domain

Homogeneous n-dodecane ignition (ground)

39

• N-dodecane exhibits two-stage ignition with minimum ignition delay at ‘preferred mixture fraction’ for each stage1

• H2O2 is a marker for low temperature combustion; OH is a marker for high temperature combustion

• Low temperature ignition shifts toward richer mixtures (by almost a factor of 3, has a much shorter

ignition delay while hot ignition remains the same)

Temperature H2O2 OH

xxxxxxxx

Oxidizer now consists of an equilibrated lean mixture with Z=0.01

Temperature H2O2 OH

Low-T High-T

Stoich. Mix frac

Conditional statistics (ground)

40

0.0 0.2 0.4Zpilot

500

1000

1500

2000

T[K

]

0.55⌧inj

0.0 0.2 0.4Zpilot

500

1000

1500

2000

T[K

]

0.91⌧inj

0.0 0.2 0.4Zpilot

500

1000

1500

2000

T[K

]

2.09⌧inj

0.0 0.2 0.4Zmain

500

1000

1500

2000

T[K

]

2.56⌧inj

0.0 0.2 0.4Zmain

500

1000

1500

2000

T[K

]

2.93⌧inj

0.0 0.2 0.4Zpilot

0.000

0.001

0.002

0.003

0.004

YH

2O

2

0.0 0.2 0.4Zpilot

0.000

0.001

0.002

0.003

0.004

YH

2O

2

0.0 0.2 0.4Zpilot

0.000

0.001

0.002

0.003

0.004

YH

2O

2

0.0 0.2 0.4Zmain

0.000

0.001

0.002

0.003

0.004

YH

2O

2

0.0 0.2 0.4Zmain

0.000

0.001

0.002

0.003

0.004

YH

2O

2

0.0 0.2 0.4Zpilot

0.0000

0.0002

0.0004

0.0006

0.0008

YOH

0.0 0.2 0.4Zpilot

0.0000

0.0002

0.0004

0.0006

0.0008

YOH

0.0 0.2 0.4Zpilot

0.0000

0.0002

0.0004

0.0006

0.0008

YOH

0.0 0.2 0.4Zmain

0.0000

0.0002

0.0004

0.0006

0.0008

YOH

0.0 0.2 0.4Zmain

0.0000

0.0002

0.0004

0.0006

0.0008

YOH

Tem

pera

ture

H 2O

2O

H

Mid 1st

injection

End 1st

injectionMid 2nd

injectionEnd 2nd

injectionStart 2nd

injection

vs. Zpilot vs. Zmain

Effect of mixing on ignition

41

0.0 0.2 0.4Zpilot

500

600

700

800

900

1000

T[K

]

0.55⌧inj

0.0 0.2 0.4Zpilot

500

1000

1500

2000

T[K

]

0.91⌧inj

0.0 0.2 0.4Zpilot

0.000000

0.000002

0.000004

0.000006

0.000008

YOH

0.55⌧inj

0.0 0.2 0.4Zpilot

0.0000

0.0002

0.0004

0.0006

0.0008

YOH

0.91⌧inj

P (�pilot) 0.25

0.25 < P (�pilot) 0.50

0.50 < P (�pilot) 0.75

0.75 < P (�pilot) 1.00

0.0 0.2 0.4Zmain

750

1000

1250

1500

1750

2000

T[K

]

2.56⌧inj

0.0 0.2 0.4Zmain

1000

1250

1500

1750

2000

T[K

]

2.93⌧inj

0.0 0.2 0.4Zmain

0.0000

0.0002

0.0004

0.0006

YOH

2.56⌧inj

0.0 0.2 0.4Zmain

0.0000

0.0002

0.0004

0.0006

0.0008

YOH

2.93⌧inj

P (��cross) 0.50

0.50 < P (��cross) 1.00

P (�+cross) 0.50

0.50 < P (�+cross) 1.00

�pilot = 2D(@Zpilot/@xj)2

�cross = 2D(@Zpilot/@xj · @Zmain/@xj)

Bins

bas

ed o

n pi

lot s

cala

r di

ssip

atio

n ra

teBi

ns b

ased

on

cros

s sca

lar

diss

ipat

ion

rate

0.0 0.2 0.4Zpilot

500

600

700

800

900

1000

T[K

]

0.55⌧inj

0.0 0.2 0.4Zpilot

500

1000

1500

2000

T[K

]

0.91⌧inj

0.0 0.2 0.4Zpilot

0.000000

0.000002

0.000004

0.000006

0.000008

YOH

0.55⌧inj

0.0 0.2 0.4Zpilot

0.0000

0.0002

0.0004

0.0006

0.0008

YOH

0.91⌧inj

P (�pilot) 0.25

0.25 < P (�pilot) 0.50

0.50 < P (�pilot) 0.75

0.75 < P (�pilot) 1.00

0.0 0.2 0.4Zmain

750

1000

1250

1500

1750

2000

T[K

]

2.56⌧inj

0.0 0.2 0.4Zmain

1000

1250

1500

1750

2000

T[K

]

2.93⌧inj

0.0 0.2 0.4Zmain

0.0000

0.0002

0.0004

0.0006

YOH

2.56⌧inj

0.0 0.2 0.4Zmain

0.0000

0.0002

0.0004

0.0006

0.0008

YOH

2.93⌧inj

P (��cross) 0.50

0.50 < P (��cross) 1.00

P (�+cross) 0.50

0.50 < P (�+cross) 1.00

0.0 0.2 0.4Zpilot

500

600

700

800

900

1000

T[K

]

0.55⌧inj

0.0 0.2 0.4Zpilot

500

1000

1500

2000

T[K

]

0.91⌧inj

0.0 0.2 0.4Zpilot

0.000000

0.000002

0.000004

0.000006

0.000008YOH

0.55⌧inj

0.0 0.2 0.4Zpilot

0.0000

0.0002

0.0004

0.0006

0.0008

YOH

0.91⌧inj

P (�pilot) 0.25

0.25 < P (�pilot) 0.50

0.50 < P (�pilot) 0.75

0.75 < P (�pilot) 1.00

0.0 0.2 0.4Zmain

750

1000

1250

1500

1750

2000

T[K

]

2.56⌧inj

0.0 0.2 0.4Zmain

1000

1250

1500

1750

2000

T[K

]

2.93⌧inj

0.0 0.2 0.4Zmain

0.0000

0.0002

0.0004

0.0006

YOH

2.56⌧inj

0.0 0.2 0.4Zmain

0.0000

0.0002

0.0004

0.0006

0.0008YOH

2.93⌧inj

P (��cross) 0.50

0.50 < P (��cross) 1.00

P (�+cross) 0.50

0.50 < P (�+cross) 1.00

inc 𝛘

𝛘+

Mid 1st

injection

End 1st

injection

Mid 2nd

injectionEnd 2nd

injection

𝛘-

ZmainZpilot

Z

x

Effect of mixing on ignition

42

0.0 0.2 0.4Zpilot

�0.4

�0.3

�0.2

�0.1

0.0

0.1

⇢ Yp,�

pilot

0.55⌧inj0.91⌧inj

0.0 0.2 0.4Zmain

�0.8

�0.6

�0.4

�0.2

0.0

⇢ Yp,�

� cross

2.56⌧inj2.93⌧inj

1st injection consistent with results from Borghesi et al., Comb. Flame, 2018.

1st injectionCorrelation between

progress variable and pilot scalar dissipation rate

2nd injectionCorrelation between

progress variable and cross scalar dissipation rate

Only negative values of cross scalar dissipation rate taken into account

ZmainZpilot

Z

x

Conclusions (Ground)

43

• First- and second stage pilot ignition consistent with previous numerical studies

• Accelerated ignition for main injection observed consistent with experiments

• Strong mixing inhibits ignition of first injection, promotes ignition of second injection

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 0.212 ms

8/17

Pilotmixturefraction

Mainmixturefraction

Temp (K) KET CH2O H2O2 OH

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 0.484 ms

8/17

Pilotmixturefraction

Mainmixturefraction

Temp (K) KET CH2O H2O2 OH

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 0.824 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 1.256 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 1.505 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 1.786 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 2.066 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 2.348 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 2.773 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 3.575 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 3.778 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 3.994 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 4.228 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 4.471 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 4.735 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 4.986 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 5.277 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 5.623 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 5.981 ms

8/17

Ignition sequence UAV

−0 . 0 0 4 0 .000 0 .004x [m]

0 .000

0 .005

0 .010

0 .015

0 .020

0 .025

0 .030

z[m

]

Zpilot [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

Zmain [-]

0 .0 0 .3

−0 . 0 0 4 0 .000 0 .004x [m]

T [K]

1000 2000

−0 . 0 0 4 0 .000 0 .004x [m]

YOC1 2H 2 3OOH [-]

0 .00

−0 . 0 0 4 0 .000 0 .004x [m]

YCH 2O

0 .05 0 .00 0 .01

−0 . 0 0 4 0 .000 0 .004x [m]

YH 2O2 [-]

0 .000 0 .003

−0 . 0 0 4 0 .000 0 .004x [m]

YOH [-]

0 .000 0 .005

t = 6.386 ms

8/17

For UAV conditions pilot accelerates ignition

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

0 .0 0 0

0 .0 0 5

0 .0 1 0

0 .0 1 5

0 .0 2 0

0 .0 2 5

0 .0 3 0

z[m

]

Zpilot [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

Zma i n [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

T [K]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YOC1 2H2 3OOH [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YCH 2O

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YH2O2 [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YOH [-]

0 .0 0 0 .0 2 0 .0 0 .1 1 0 0 0 2 0 0 0 0 .0 0 0 .0 5 0 .0 0 0 .0 1 0 .0 0 0 0 .0 0 3 0 .0 0 0 0 .0 0 5

t = 3.591 ms

No pilot (t=3.6ms after start of main injection)

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

0 .0 0 0

0 .0 0 5

0 .0 1 0

0 .0 1 5

0 .0 2 0

0 .0 2 5

0 .0 3 0

z[m

]

Zpilot [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

Zma i n [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

T [K]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YOC1 2H2 3OOH [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YCH 2O

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YH2O2 [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YOH [-]

t = 4.735 ms

0 .0 0 .3 0 .0 0 .3 1 0 0 0 2 0 0 0 0 .0 0 0 . 0 5 0 . 0 0 0 .0 1 0 .0 0 0 0 . 0 0 3 0 . 0 0 0 0 .0 0 5

9/17

Pilot (t=3.6ms after start of main injection)

→ clear acceleration ofthe ignition process

Zmain Zpilot T (K) KET CH2O H2O2 OH

For UAV conditions pilot accelerates ignition

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

0 .0 0 0

0 .0 0 5

0 .0 1 0

0 .0 1 5

0 .0 2 0

0 .0 2 5

0 .0 3 0

z[m

]

Zpilot [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

Zma i n [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

T [K]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YOC1 2H2 3OOH [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YCH 2O

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YH2O2 [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YOH [-]

0 .0 0 0 .0 2 0 .0 0 .1 1 0 0 0 2 0 0 0 0 .0 0 0 .0 5 0 .0 0 0 .0 1 0 .0 0 0 0 .0 0 3 0 .0 0 0 0 .0 0 5

t = 4.601 ms

No pilot (t=4.6ms after start of main injection)

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

0 .0 0 0

0 .0 0 5

0 .0 1 0

0 .0 1 5

0 .0 2 0

0 .0 2 5

0 .0 3 0

z[m

]

Zpilot [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

Zma i n [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

T [K]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YOC1 2H2 3OOH [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YCH 2O

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YH2O2 [-]

− 0 . 0 0 4 0 . 0 0 0 0 . 0 0 4

x [m]

YOH [-]

t = 5.802 ms

0 .0 0 .3 0 .0 0 .3 1 0 0 0 2 0 0 0 0 .0 0 0 . 0 5 0 . 0 0 0 .0 1 0 .0 0 0 0 . 0 0 3 0 . 0 0 0 0 .0 0 5

10/17

Pilot (t=4.6ms after start of main injection)

→ clear acceleration ofthe ignition process

Zmain Zpilot T (K) KET CH2O H2O2 OH

Joint PDFs of mixture fractions show pilot and main fluid is well mixed

0.00 0.08 0.100.02 0.04 0.06

Zmain

0.000

0.005

0.010

0.015

0.020

0.025

0.030Z p

ilot

t=3.58 msρcorr=0.72cov=6.94e-06

101

102

103

104

jpdf

0.00 0.100.02 0.04 0.06 0.08

Zmain

0.000

0.005

0.010

0.015

0.020

0.025

0.030

Z pilo

t

t=4.74 msρcorr=0.83cov=6.17e-06

101

102

103

104

105

jpdf

0.00 0.02 0.100.04 0.06 0.08

Zmain

0.000

0.005

0.010

0.015

0.020

0.025

0.030

Z pilo

t

t=4.17 msρcorr=0.78cov=6.17e-06

101

102

103

104

105

jpdf

0.00 0.100.02 0.04 0.06 0.08

Zmain

0.000

0.005

0.010

0.015

0.020

0.025

0.030

Z pilo

t

t=5.80 msρcorr=0.86cov=5.44e-06

101

102

103

104

105

jpdf

Pilot and main fluid well-mixed at time of low-/high-temperature ignition (bottom two plots)

11/17

xxxxxxxx

Mixing behavior – scalar dissipation rates

−0.005 0.000 0.005x [m]

0.000

0.005

0.010

0.015

0.020

z[m

]

1 0 − 4 1 0 − 2 100 102

−0.005 0.000 0.005x [m]

1 0 − 4 1 0 − 2 100 102

−0.005 0.000 0.005x [m]

−0.005 0.000 0.005x [m]

Scalar dissipation rates at t=2.3ms and t=3.6ms (before low-T ignition)χZpilot χZmain. s ign(χ cross) × log(χ cross). χcross

−0.005 0.000 0.005x [m]

0.000

0.005

0.010

0.015

0.020

z[m

]

χZp i l o t

1 0 − 4 1 0 − 3 1 0 − 2 1 0 − 1 100

−0.005 0.000 0.005x [m]

χZm a i n

10−140−130−120− 1100 101 102

−0.005 0.000 0.005x [m]

2 1 0 1s i g n ( χ c r o s s ) × l o g 1 0 [ m a x ( 1 , | χ c r o s s | ) ]

− 1 0

−0.005 0.000 0.005x [m]

200 100 0χcross

1 −10 0 10

• Early time: Mostly negative cross scalar dissipation rates

• Later time: Mostly positive cross scalar dissipation rates

m ainZpilot Z

z

x

𝟀cross < 0

z

x

𝟀cross > 0

12/17

xxxxxxxx

Mixing behavior – scalar dissipation rate (SDR) PDFs

− 5 50log(χpilot)

10− 6

10− 4

10− 5

100

10− 1

10− 2

10− 3

pdf[

-]

− 5 50log(χmain)

10− 6

10− 1

10− 2

10− 3

10− 4

10− 5

100

pdf[

-]

100− 100 χ

0cross

10− 1

10− 2

10− 3

10− 4

10− 5

10− 6

100

pdf[

-]

2.35 ms3.58 ms5.80 mslog-normal

From left to right: Pilot SDR, main SDR, cross SDR

Pilot and main SDR nearly log-normal (Hawkes et al., PCI,2007)Cross SDR has stretched exponential tails (similar: Chaudhuri et al., C&F, 2017)

13/17

xxxxxxxx

---

Conditional averages of ketohydroperoxide, OC12H23OOH

0.00 0.02 0.04 0.06 0.08 0.10

Zmain

0.000

0.005

0.010

0.015

0.020

0.025

Z pilo

t

0.000

0.002

0.004

0.006

0.008

Y OC 1

2H23

OO

H

0.000.000

0.005

0.010

0.015

0.020

0.025

0.030

Z pilo

t

t=4.17 ms

0.000

0.005

0.010

0.015

0.020

0.025

0.030

YO

C 12H

23O

OH

0.00 0.02 0.04 0.06 0.08 0.10

Zmain

0.000

0.005

0.010

0.015

0.020

0.025

0.030 0.010 0.030

Z pilo

t

t=2.35 ms t=2.77 ms

0.000

0.002

0.004

0.006

0.008

0.010

Y OC 1

2H23

OO

H

0.02 0.04 0.06 0.08 0.10 0.00

Zmain

0.000

0.005

0.010

0.015

0.020

0.025

0.030

Z pilo

tt=4.74 ms

0.000

0.005

0.010

0.015

0.020

0.025

0.030

YO

C 12H

23O

OH

0.00 0.02 0.04 0.06 0.08 0.10

Zmain

0.000

0.005

0.010

0.015

0.020

0.025

0.030

Z pilo

t

t=3.58 ms

0.000

0.002

0.004

0.006

0.008

0.010

Y OC 1

2H23

OO

H

0.02 0.04 0.06 0.08 0.10 0.00

Zmain0.100.02 0.04 0.06 0.08

Zmain

0.000

0.005

0.010

0.015

0.020

0.025

0.030

Z pilo

t

t=5.80 ms

0.000

0.005

0.010

0.015

0.020

0.025

0.030

YO

C 12H

23O

OH

14/17

xxxxxxxxFormation starts in regions with pilot fluid, moves into richer conditions Change of maximum range between top and bottom row (bottom factor 3 higher) Bottomrow: before main jet low-T ign., start of low-T ign., start of high-T ign.

Conditional averages of OC12H23OOH

0.00 0.02 0.04 0.08 0.100.06Zsum

0.004

0.003

0.002

0.001

0.000

0.005

0.006

YO

C 12H

23O

OH

multi single

OC12H23OOH compared at the same time, 3.59 s with respect to start of simulations. Pilot is able to provide OC12H23OOH to rich regions.

15/17

xxxxxxxx

Conclusions (UAV case)

• Different conditions lead to very different ignitionbehavior• Pilot injection does not ignite prior to mixing with main injection for UAV

• Clear differences between cases with and without pilot

• Pilot mixes well with fluid from main injection

• Pilot supplies low-temperature species that enhance ignition of the main injection

16/17

zzzzzzzzz

Direct Numerical Simulation of flame stabilization assisted by auto-ignition at reheat conditions

Konduri Adityaa, Andrea Gruberb, Mirko Bothienc and Jacqueline H. Chena

aCombustion Research Facility, Sandia National Laboratories, Livermore, CA, USAbSINTEF Energy Research, Trondheim, Norway

cAnsaldo Energia, Baden, Switzerland

Acknowledgements: Basic Energy Sciences, Office of Science, DOE

OLCF, NERSC, Norwegian CCS Research Centre (NCCS)Aditya et al. PCI 2018

Staged gas turbine combustion

• Originally developed by ABB for high efficiency, load flexibility and low emissions

• Recently improved and simplified (reduced cost) for the H-class GT36

• First (premix) combustion stage based on flame propagation• Second (sequential) combustion stage based on auto-ignition

Reheat burner

DNS of idealized reheat burner configuration from Ansaldo EnergiaOperating conditions: • Inlet temperature: ∼ 1100 K • Pressure: ∼ 20 atm

Scaled conditions: • Mean inlet temperature: 1100 K• Pressure: 1 atm• Fuel: hydrogen

Objective: • Understand the flame stabilization • Identify the modes of combustion • Quantify the role of autoignition

Simulation details

• Chemical mechanism: 9 species hydrogen-air (Li et al., 2004)• Inflow composition: premixed H2 + O2 + N2 + H2O (φ = 0.35) • Ubulk = 200m/s, uʹ = 20m/s, Tinlet = 1100K, Twall = 750K• Inflow profile: feed from DNS of a fully developed channel flow,Reb ∼13, 000, Reτ= 378 (viscous length scale)

Feed data sampling plane

Two combustion configurations are observed:

• Design state: mainly auto-ignition in the combustion chamber

• Intermittent auto-ignition state: ignition in mixing section

Iso-surfaces of vorticity magnitude colored by temperature

Enstrophy conditioned on temperature, and heat release rate (red)

Design combustion state

Combustion modes:• Autoignition along center-line

• Flame propagation near corners

• HO2: indicative of chain branching

Combustion mode: OH budget analysis

• Auto-ignition: balance between advection and reaction • Flame propagation: balance between diffusion and reaction

Intermittent auto-ignition state

• Early auto-ignition in the mixing section

• Ignition kernel advectsdownstream

• Occurs intermittently

Chemical Explosive Mode Analysis

• α = φs /φω : ratio of the projected non-chemical source term and the projected chemical source term (C. Xu et al., PROCI 2018 )

Three mode are identified:• Assisted-ignition (α > 1): diffusion significantly promotes reaction• Auto-ignition (−1 < α < 1): chemistry plays a dominant role• Extinction zone (α < −1): diffusion dominates chemistry and suppresses ignition

Intermittent auto-ignition state

Compression wave

Auto-ignition Contours of heat release

• Local rise in pressure due to intermittent constructive interference pattern

• Increases local temperature by 20-30 K

• High reactivity of hydrogen

• Decrease in ignition delay time (30%)

Contours of instantaneous dilatation

Conclusions• Performed DNS of a reheat burner at scaled conditions • Two states of hydrogen/air combustion have been observed:

• design state: flame propagation and auto-ignition in the combustor• intermittent auto-ignition in mixing section

• Premature auto-ignition arises due to pressure (and following temperature) rise in mixing section

• Quantified the contribution of different modes towards heat release using chemical explosive mode analysis (CEMA)

• Future work:• characterize the unstable flame behavior and the conditions

leading to it • find the inlet conditions for statistically stationary reheat flame• perform 2D and 3D simulations with varying fuel composition and

its stratification