This action might not be possible to undo. Are you sure you want to continue?
COMPUTER MODELING OF SOOT FORMATION COMPARING FREE RADICAL AND IONIC MECHANISMS (A Collaborative Project between AeroChem and Penn State)
Final Technical Report
of the Penn State Efforts (Grant No. AFOSR 88-0072) Covering the period from December 1. 1987 to November 30, 1990
AIR FORCE OFFICE OF SCIENTIFIC RESEARCH
by Michael Frenklach and Hal Wang Fuel Science Program Department of Materials Science and Engineering The Pennsylvania State University University Park PA 16802
January 22, 1991
SECURITY CLASSIPICATiON OF tIS AGE
REPORT DOCUMENTATION -PAGE
a. REPORT SECURITY CLASSIFICATION,
3. DISTRIBUTION/AVAILABILITY OF REPORT SHEDLEApproved 2b.DECASSFICTIO10WNGAOIG SC'4EULEdistribution 2b.OECASSFICTIOOONGRDIN
2a. SECURITY CLASSIFICATION AUTHORITY
for public release;
4. PERFORMING ORGANIZATION REPORT NUMBER($)
ORGANIZATION REPORT NUMER(S)
6a. NAME OF PERFORMING ORGANIZATION
6b. OFFICE SYMBOL (11
7a. NAME OF MONITORING ORGANIZATION
Pennsylvania State University
6C. ADDRESS (CitY, State, and ZIP Code)
7b. ADDRESS (Cty, State, and ZIP Code)
University Park, PA
8 b. OFFICE SYMBO0L
Building 410, Bolling AFB DC
S&. NAME OF FUNDING I SPONSORING
St. ADDRESS (Cfty State, and ZIP CO*e)
9. PROCUREMENT INSTRUMENT IDENTIFICATION NUMBER
10. SOURCE OF FUNDING NUMBERS EF LEMENT NO0.
Building 410, Bolling 20332-6448
11. TITLE (Intclude SwwftY Can ca0tmo)
(U) Computer Modeling of Soot Formation Comparing Free Radical and Ionic Mechanisms
12.PESO~l. UTORS)Michael 13a. TYPE OF RIEPORT
14. DATE OF REPORT ()eaIrMoncthay)
1S. PAGE COUNT
13b. TIME COVERED
16. SUPPLEMENTARY NOTATION
FRM12-1-87 Toll -30
January 22, 1991
COSATI COIDES FIEL GRUP SI'GOUP
18. SUBJECT TERMS (Contiu on reyern If necemy and idenff by block number)
Soot Formation, Computer Modeling, Ionic Mechanism
eAewohman Penn State, aimed at isatnirpr o ofentm sta development of a quantitative physicochemical model of soot formation. The Penn State accomplishments during the past three-year period are: The AeroChem ion-reaction mechanism was run with a flame code, and the computational results indicate that the formation of polycyclic aromatic hydrocarbolis, the precursors to soot, via the ionic reaction pathway is slower than via the pathway involving neutral species; A computer code for modeling large species and soot particle formation and growth was developed; Cur neutral reaction mechanism was updated and tested; and A detailed computer simulation of soot particle nucleation and growth
in laminar premixed flames was successfully performed using our updated neutral reaction
echanism -- for the first time soot particle inception could be modeled from first principles, starting with fuel decomposition and going all the way to predicting, in quantitative agreement with experiment, soot particle properties.
20, DISTRIBUTION/ AVAII.AIIUTY OF ABSTRACT M UNCLASSIFIIOIUNUIMITED (3 SAME AS RPT. 22a, NAME OF RESPONSIBLE1 INDIVIDUAL
21. ABSTRACT SECURITY CLASSIFICATION __OTIC USERS
P'~jLgflIr.dY#!1d Codej chIL~A'RKBOL 16
Previous edition ate obsoete.
Julian M Tishkoff 00 Form 1473, JUN
_SECURITY CLASSIFICATION OF T41S PAGE-
...................... 20 PROFESSIONAL PERSONNEL ................................ RESULTS ........ 4 WORK PROPOSED ............... 21 INVENTIONS .................................................. ..................................................................... 21 REFERENCES...r Speul................ 4 4 5 Neutral Reaction Mechanism .......................................................d......TABLE OF CONTENTS INTRODUCTION ......a 3 ........... Ion Reaction Mechanism.................. 22 DTIC TAi Just ite los DistrIbuti AvailablIlty oede 3 ) glut Avail ................ 19 20 PRESENTATIONS AND PUBLICATIONS .. 11 Flame Simulation of Soot Particle Nucleation and Growth............... FUTURE WORK DIRECTION .......................................................... 15 PRINCIPAL CONCLUSIONS AND RESULTS ...
To update the Penn State neutral-reaction mechanism for the formation of polycyclic aromatic hydrocarbons (PAH).INTRODUCTION This is a final report on collaborative study between AeroChem Research Laboratories (principal investigator . the computational results were analyzed. a computer code for modeling large species and soot particle formation and growth was developed. F.Dr. To undertake flame simulation and reaction-path analysis using the ion-reaction mechanism provided to us by AeroChem. 5. M. The specific objectives of the proposed 3-year study were: 1) to delineate the relative importance of radical and ionic mechanisms of soot nucleation in flames. Frenklach). Only the work performed at Penn State is reported here. To perform computer simulation of soot particle formation in flames with the code developed. that of AeroChem is reported in their parallel final technical report. To develop a numerical algorithm and a computer code for modeling the formation and growth of large-sized species and soot particles. 4. we completed most of the planned work: several ion reaction mechanisms provided to us by AeroChem were "cleaned-up" and run with a flame code.Dr. 2. The ultimate objective of this program was to develop a quantitative physical/chemical model of the mechanism of soot formation that is consistent with available experimental data. RESULTS During the three-year period of the project. Calcote) and the Pennsylvania State University (principal investigator . our neutral reaction mechanism was updated and tested. WORK PROPOSED The general objectives proposed for the Penn State efforts were: 1. and a detailed computer simulation of soot particle nucleation and 4 . To analyze the competition for the formation of soot and soot precursors between the ionic and neutral reaction mechanisms. H. 3. and 2) to determine the optimum chemical and physical model of the total soot formation process.
As there is no single flame reported for which there are measurements available of all the pertinent chemical species. as it is for the first time soot particle inception could be modeled from first principles. had to be chosen for the computer simulations. and a significant number of steps in the principal reaction sequence was found to be reversible. starting with fuel decomposition and going all the way to predicting. The premise of such testing is founded on our previous result that the major neutral-reaction pathways for PAH formation under shock-tube oxidation1 and laminar premixed flame2 conditions are very similar to each other. ie. The computer simulations indicated that the forward and reverse reactions of these steps are tightly balanced. Frst is the relatively low rate of ion production. 7 ignoring the kinetic-thermodynamic coupling controlling the rate of the principal reaction pathway. There are two basic factors identified by the sensitivity and rate analysis to be responsible for the low rate of PAH production via the ionic mechanism tested. a "well-studied" laminar premixed flame.4 The detailed account of the shock-tube calculations is given in our first annual reporL5 The basic results of the shock-tube simulations indicated that the ionic mechanism produces polycyclic aromatic hydrocarbons at significantly lower rate than does the mechanism composed of only reactions of neutral species. two close flames were chosen to 5 . soot particle properties. cannot possibly be valid. The latter part is a particularly significant accomplishment. 3. For this purpose.2. The initial ionic mechanism provided to us by AeroChem has been reported and discussed in the AeroChem reports. The second factor limiting the rate of PAH production via the ionic mechanism is the reversibility of some of the principal reaction steps. in quantitative agreement with experiment. for which the experimental data on ionic species concetitrations are available.growth in laminar premixed flames was successfully performed using our updated neutral reaction mechanism. these reactions are in partial equilibrium. Ion Reaction Mechanism We began our work by testing the general performance of an ion-molecule reaction mechanism under shock-tube conditions. similarly to the case for the neutral-reaction mechanism. 6 This reduces dramaticaily the net mass flux and thus limits the rate of PAH production.. Among other things. An extensive sensitivity and rate analysis found no conceivable adjustments in the parameter values that would affect this conclusion. it implies that an oversimplified analysis presented recently. The following is a detailed account for the work performed and the results abtained. We next examined the competition between the ionic and neutral mechanisms for PAH production at flame conditions.
as far as the ions-vs-neutrals competition is concerned. In the computer simulation as opposed to the experiment. The agreement is worse for these species compared to the agreement seen for the major species in Figure 1. that of Delfau and Vovelle 9 referred to as Flame 12 . taking into account the uncertainty in the input reaction data and in the experimental measurements most of them are computed within an order of magnitude of the experimental numbers. Since.. and quality of experimental measurements allow. cold gas velocity 50 cm/s. The stoichiometry. are very - similar to each other (compare.was chosen to test the predictions of the present simulations for the major neutral species. thsoe obtained with the most recent version of the AeroChem mechanism. the accuracy of numerical techniques. the reported temperature profiles of Flames II and 12 are slightly different from each other. Examination of Figure 1 indicates that the concentrations of major species are predicted well . 10 Several reaction mechanisms provided to us by AeroChem have been tested and analyzed during the course of the study. Also. 2 The reasons for this are discussed in the work cited. the main flame . the following conclusion. whereas that of Calcote and Keil had 3 % argon. however. In this sense.g. Temperature profile is a critical input parameter for the flame simulations and even small differences in the given profile data cause significant changes in computed profiles of minor (e.as good as the present state of the art in the knowledge of chemical reactions.represent a "well-studied" flame.9 % C2H 2 .44. PAHs) species. for example. As the primary focus of this study is on the ionic species. pressure 20 torr. their rates and thermochemistr. that the peak values of the ion concentrations are predicted reasonably well. The mixture used by Delfau and Vovelle did not contain any argon. the data in present Figures 1 through 4 with the data reported in Figures 5 through 8 in our first annual report5 ). Calcote and co-workers did not measure the concentrations of neutral species in this flame. however. Some of the results of these flame simulations. Although the analysis of the results obtained in the flame simulations is not entirely completed. are presented in Figures 1 through 4. The comparison between the computed and experimental concentrations of ions is given in Figures 2 and 3. Similar behavior was seen in computer simulations of an acetylene flame with the neutral-reaction mechanism.is the flame for which ion concentration profiles were measured by Calcote and co-workers: 8 52. a second flame.0). pressure and cold gas velocity of this flame are exactly the same as those of Flame I1. It is important to notice. similar to to those reached based on the shock-tube 6 .1 % 02 -Ar (0 = 3. the results obtained with different versions of the AeroChem ionic mechanism. the positions of the concentration peaks are shifted towards the burner and the decay in the concentration profiles is much steeper. The flame computations were performed using the Sandia burner code.referred to as Flame I1 .
U.1 p a UOI~UZJ ~jOJAf (x) o o o 0IV~ c~i 0 ~4j U 'I' U ________________________________ - d d o uo!1ouJJ ~jOJA4 7 0 00 . - 0 I. a 0 I eq eq 0 l~a a c~ '~ M A U a 'a ' (~4 a -.
-~ US + .f 0 G * 0 UOQUJ 2TOw 8 t0 .
C '*U31 - -4 .dw -4) Uor .
! * 4 4~ 4%*% S. 4. I o * CIA 2 o - o UO!~OuiJ o ~T 0N o 0 j 10 . 44 4 ~E~x 0 44~ * * 6)..4 4. . 44 * * * 4 4. * I e~i 0 * * I 4.. 44 8 * . 44 4 . * * *. 9. I.4 4.4 o 0 .4 * I. . 'I 4* * * * I..4 * 4. * 4 4 . 4.U U.
However. For instance. reaction channels of the present ionic mechanism are most likely to greatly overestimate the growth rate of polycyclic aromatic species in flame environments. This means. As can be seen from this comparison. also note that the computed concentrations of ions are significantly larger than the experimental counterparts). more importantly. some available (although limited) information indicates to the opposite. In our initial efforts.simulation results. and reversibility of the reaction steps responsible for the formation of high- molecular weight species. Neutral Reaction Mechanism Accurate prediction of polycyclic aromatic hydrocarbons in flames is critical for modeling of soot formation in hydrocarbon combustion. 2 Now. that the rate coefficients and. Another shortcoming of the present ionic mechanism is the lack of knowledge on the microscopic mechanism of many ion-molecule reaction in the AeroChem scheme. it is unlikely that many individual reaction steps will proceed as written and with the rate coefficient values assigned to them. It appears that even a more conceivable reaction step of the AeroChem mechanism has a measurable potential energy barrier. when the acetyleneaddition mechanism we have advocated for the last seven years is well tested under various experimental conditions and by different research groups. the production of aromatics via the neutrals is much faster than that via ions (the difference between the two cases increases with the switching off oxidation reactions in computer simulations with the neutral-reaction mechanism. where the concentrations of selected aromatics computed with the ionic mechanism are compared to those computed with the neutral-reaction mechanism at the same flame conditions. Most of these ion reactions are written on the premises'1 that ionmolecule reactions have no potential-energy barriers and that ionic species isomerize fast. This result is illustrated in Fig.4. we performed quantum-chemical calculations on a key ion-molecule step. The factors responsible for this are the same as those identified in the analysis of the shock-tube results: a relatively low rate of chemi-ionization as compared to that of radical production at the same conditions. in a low-pressure laminar premixed acetylene flame. the objective was a proof-ofa-concept computer simulation of PAH production in high-temperature hydrocarbon pyrolysis. has emerged: the ionic mechanism produces polycycdic aromatic hydrocarbons at significantly lower rates than does the neutral-reaction mechanism. the objective is to improve the 11 . A sensitivity and rate analysis performed so far found no conceivable adjustments in the parameter values that would change this principal conclusion. In fact. 6 and later.
13 Harris et al.5% C2 H2 . We chose to model the PAH profiles in three well-studied laminar premixerl flames.6% C2 H2 . 2 1 respectively.2 and (c) revised acetylene oxidation mechanism of Hwang et al.5% C2H4 .48. The thermodynamic data were taken from Stein and Fahr. and acetylene-oxygen atom reactions C2H2 + OH -.17. The reaction mechanism for the former part was composed of data taken primarily from several sources: (a) our recently optimized methane oxidation mechanism.14 These flames were computed using the Sandia burner code. 15 The transport coefficients obtained in this manner agreed well with experimental data available for several compounds. 10 The Lennard-Jones parameters of PAH molecules were calculated using empirical correlations and available physical properties.22 The formation of the first aromatic ring followed the basic reaction scheme of Frenklach and Warnatz. responsible for the main flame structures. 17 The reaction mechanism is composed of two parts: the small molecule reactions. 2 n-C4H3 + C2 H2 -+ phenyl C4 4 + H n-C4H5 + C2H2 -4 12 benzene + H . 18 (b) revised mechanism of Frenklach and Warnatz.products C2H2 + 0 -* products were taken from Miller and Melius. Flame conditions for modeling with neutral-species mechanism No. 20 and Mahmud and Fontijn. and reactions describing the formation and growth of PAI-s.mechanism to a point that it is capable of quantitative predictions. whose conditions are summarized in the following table: Table 1. 16 except for the C--H group additivity which was based on the recent results of Green et al.8 Reference MIT group 12 760 Bockhorn et al.9% 02 P (torr) 20 90 Cold gas velocity (cm/s) 50 20 7.21. N1 N2 N3 Composition in Argon 46. 19 The main revisions included: the rate coefficients of acetylene-hydroxyl. The rate coefficient for reaction C22 + C2H3 -+ was taken from Fahr and Stein. Initial results of suca efforts are reported below.4% 02 16.5% 0 2 23.
d 6CO A 00 00 0. the agreement between the modeling predictions and experimental data is good for major species. 2 The important result here is that the near-quantitative accuracy was obtained for PAHs with a single reaction mechanism for all the three flames simultaneously.0 0. 0. still within their experimental uncertainties.however with the rate coefficients taken from the work of Westmoreland et al.6 10-4 Bemm a 0O0 0 0 0. As can be seen in these figures. The major reaction pathways.2 3 who used a QRRK methodology of estimating chemical activation processes involved in these reactions.5 1. Revision of rate coefficients for main reaction classes of PAH growth Reaction aryl-H + H -. but certainly much closer .C0'---10 ---- .aryl + H2 aryl + C2H2 aryl + 02 -+ Prototype reaction benzene + H -. Computed vs. experimental major species concentrations of Flame Ni. The agreement for PAILs is not exactly quantitative.0 0 1 2 3 4 x (cm) Figure 5. experimental major aromatic species concentrations of Flame NI. their roles in PAH growth. x (cm) Figure 6.phenyl + H2 phenyl + C2 H2 phenyl--acetylene + H phenyl + 02 -+ products aryl-C2H + H Reference Kiefer et a. These results were obtained with the rate coefficients for several small-molecule reactions adjusted by a factor of 2 to 3. Computed vs. Table 2.2 4 Fahr and Stein22 Lin and Lin25 products The computational results for the three flames are presented in Figures 5 to 10. 13 .than reported by us previously. 2 0.5 2.0 1. The assignment of the rate coefficients for the main reaction classes responsible for PAH growth are summarized in Table 2.in both shape and absolute values .. and the sensitivity information are essentially the same as reported and discussed in the previous work.
21io4 0.5 X (cm) Figure 8.1 0.0 1. experimental major species concentrations of Flame N3. * .1 14 .5 2.4 x (cm) Figure 9.0 2. experimental minor species concentration of Flame N3. Measured mole fractons of diacetylene and selected PAils of Flame N2.2 0.~ 10-3~ ~ Diaetylene3acctye Benz. Computed vs. 0.0 10 0 Co 10'1 ~ 10. 100.0 2.010- i8 100.5 1.5 X(CM) Figure 7.ene 10 -4.3 0.5 \ ~ ~ Ethynylaenaphihaleae10 1.0 0.4 x (cm) Figure 10.5 2.2 0.0 0. 0.0 0. 0.6 5 10 io0 1ez enaphdialne Naph)J n A6 V 93 lo . Computed vs.0 1.2' CgH p 1 0-i .3 0. Computed mole fractons of diacetylene and PAHs of Flame N2.
and so on. it was assumed that the coagulation dynamics is in the free-molecular regime. the corresponding radical.. (1) where Al represents an aromatic species containing I fused rings. and its subsequent growth to a prescribed size. the forming clusters were assumed to be "solid phase" and allowed to add and lose mass by surface reactions CMo-H + H -'-CM + H2 Csm. The coalescence reactions were treated as ireversible having sticking coefficients of unity. The PAH species formed in (1) are then allowed to coagulate.that is. comprised of replicating-type growth of PAHs beyond the prescribed size. (2) (3) (4) (5) (6) where Cum-H represents an arm-chair site on the soot particle surface and C. The computational model consists of three logical parts: (1) initial PAH formation. 28 was used in calculations of collision frequencies. Beginning with the dirers. formation of the first aromatic ring. nucleation...27 This method provides a mathematically rigorous description of the growth process to an infinite size PAIL which can be schematically represented as Al -+ A1+1 A+ 2 -**" -- A. (I) planar PAH growth.. we performed detailed modeling of soot particle nucleation and growth in several laminar premixed flames. and (1H) spherical particle formation and growth. small particles. the dimers.o. based on the results of Harris and Kennedy.. This mechanism is adopted based on the postulate29 -3 1 that the 15 . +l.2. The initial part of the particle dynamics model.-) collide with each other forming dimers. all the A's (i = 1.Flame Simulation of Soot Particle Nucleation and Growth Using the neutral reaction mechanism described in the previous section.et-H + H Cwooe + 02 -+ products CM-H + OH -+ products. + H Cm-H 47otm + C2H2 -+ C. consisting of coagulation of PAI-s formed in part (II) followed by the growth of the resulting particles by coagulation and surface reactions. A size-independent enhancement factor of 2. As the focus of this work was on very young. in turn. collide with Aiforming trimers or with other dimers forming tetramers..6 using the technique of chemical lumping. describes the planar growth of PAHs via the H-abstraction/Cf2-addition reaction sequence 2. which includes a detailed chemical kinetic description of acetylene pyrolysis and oxidation.26 .
one computer run took from 20 to 50 s on an IBM 3090/600S main-frame computer. 11. assuming A4 (pyrene or acepyrene) to be the Al species which initiates the planar growth of PAHs and the formation of particles.Argon. particle inception part of the flames. 11 and 12 that the model predictions are in relatively close agreement with experiment for the initial. the computed and experimental results begin to deviate from one another with increasing flame height. Estimation of the rate coefficients of the heterogeneous reactions was based on analogous gas-phase reactions of one-ring aromatics.34 The numerical integration of the moment equations is extremely economical. 1 of Bockhorn and co-workers 32 .H-abstraction/C2H2-addition (HACA) reaction sequence 6 is responsible for high-temperature growth of all forms of carbonaceous materials. was assumed to proceed with the collision efficiency of 0. based on the results of Neoh et aL33 The particle dynamics .17. 16 . This is demonstrated in Fig. benzene and phenyl. Wieschnowsky et al.4 % C2 H2 19. Oxidation by OH. of Harris and co-workers.13.Argon. It was also comted that PAH condensation on the particle surface is sufficiently fast to significantly affect the PAH profiles in the post-flame zone.25.flame No. 35 Both flames were computed with the same kinetic model. 26 . The closure of the differential equations for the PSDF moments is accomplished by interpolation between the moments.32 also suggested that the HACA reactionsZ 6 "offer a key to the understanding of a number of phenomena" observed in their flame study. where it is shown that using a narrower (experimental-like 13) profile for A4 improves the agreement between the model and experiment for Flame B.6 % 02 . and Flame H. Figures 11 and 12 present the results computed using the model discussed above for two of the chosen flames: Flame B .9 % 02. coagulation and surface reactions described above . It can be seen in Figs. reaction (6).5% C2I4 .the evolution of soot particles undergoing simultaneous nucleation. This method does not require the assumption of a particle size distribution function (PSDF). parts (11) and (HI) of our model. respectively. 34). Changing Al affects strongly the rate of particle nucleation. One factor identified as a cause of this disagreement is the shape of the PAH profiles. however. pressure 90 tor. for instance.was modeled by a method of moments (Method II of Ref.an atmospheric 16. The reference gaseous reactions for heterogeneous reactions (2)-(5) were the corresponding gas-phase reactions in Table (2).
0 0' 41 0 0 10-0 0 2 4 8 8 Height above Burner (cm) Fig. (b)particle surface area (c)average particle diamnewe (circles .experimental remults? 2 solid lines . Comparison of model. 20 40 60 a 0 0 00 0.model prediction.Time (ins) 0 3 . duhjed lin . 36 17 . (d)specific noe of soot particle oxidation by 02 (solid line .model pedicton. predctions and experimental data for Flame B: (a) particle number density.5 0 5.0 00.from the expression of Nagle and Strickland-Constable.0 __________________ S7.1 0. 11.model prediction using a experimenital-like profile for A4]. dotted lines .
experimental resuts? 5 solid lines model prediction.Height above Burner (mm) 15 10 A OL 3. circles . d .5 4.particle number density.5 0 V0 z0 5-0 0 010 b 0 ~15 0 100 V4 5 500 C 00 10-3 0 1 2 3 4 Time beyond Reaction Zone (ins) ig.particle radius. .0 3. (d)specific surface growth rase coefficient. 12. .average rP. r . Comparison of model predition mid experimental data for Flame H: (a)-c) N .0 4.average particle diameter. a9.
in quantitative agreement with experiment. composed of randomly oriented PAH oligomers. like PAH coagulation is yet to be investigated. The model also predicts the classical structure of soot particles: 39 a less dense particle core. 12d) and the predicted rate of soot oxidation by 02 agrees well with the expression of Nagle and Strickland-Constable 36 (Fig. I ld).(6). contrary to Harris and Weiner. Our model describes the surface processes in terms of elementary reactions of active sites. 32 who observed that the surface growth rate is independent of the surface area of soot particles. initiated and controlled by PAH coalescence into dimers. and a more dense concentrically-arranged particle shell. 19 .As expected from the dynamics of Smoluchowski coagulation. A detailed computer simulation of soot particle nucleation and growth in laminar premixed flames was successfully performed using our updated neutral reaction mechanism . Particle inception is primarily determined by PAH coagulation. their coagulation into clusters and cluster coagulation. Whether ions play role in other soot-forming processes. Our model resolves the controversy brought up recently by Bockhorn and co-wokers. soot particle pIMpties. Our model predicts what is essentially the classical picture of soot particle inception: 38 formation of PAHs. A computer code for modeling large species and soot particle formation and growth was developed. the computed net surface growth rate is computed to be in close agreement with that determined by Harris and Weiner 32 (Fig. starting with fuel decomposition and going all the way to predicting.3 7 who postulated that the rate of surface growth is fint order in particle surface area. PRINCIPAL CONCLUSIONS AND RESULTS The principal conclusion of this study is that the formation of polycyclic aromatic precursors to soot in hydrocarbon flames is dominated by the reactions of neutral species and not ionic ones. Hence.for the first time soot particle inception could be modeled from first principles. the computed rate of nucleation is balanced by the rate of coagulation throughout the particle inception zone. The density of active sites is determined by the gas-phase environment by the competition among reactions (2). Also. the dissimilarity in surface growth behavior can be explained by the difference in the dynamics of active sites. and surface growth due to chemical and physical processes.
October 1989." Frenklach. 20 .Invited seminar at the University of Cincinnati. 1989.. "Understading Carbon: From Soot in Flames to Synthetic Diamond to Interstellar Dust. Japan. Poster presentation at the Twenty-Second International Symposium on Combustion. Academy of Sciences GOttingen. culminated in quantitative prediction of soot particle properties in several laminar premixed flames. H. "Kinetics and Mechanism of Formation of Carbonaceous Materials. 1989. "Detailed Mechanism and Modeling of Soot Formation. Invited seminar at SUNY at Binghamton." Frenklach. in Extended Abstracts of the Conference on Mechanism of Non-Unimfm Combustion. 5." Frenklach. ESNATIONS AND PUBICATIONS 1. the direction for the future research. March 29-30. M." the Commission on Condensation. 115. M. "Particulaft Fomation in Hydrocarbon Combustion. 3. M. Prepared comment at the workshop on "Current Problems in Soot Formation during Combustion. FRG. and Frenklach. February 2.. "Modeling of PAH Profiles in Premixed Flames. September 13.. Department of Chemical and Nuclear Engineering.FUTURE WORK DIRECTION Following the successful accomplishments in soot formation model development. 4. 1989." the Commission on Condensation. Washington. New 2.. M. October 12. Especially the Mechanism of Soot Formation. 8. 1990. "Formation of PCA.." Frenklach. Tokyo Institute of Technology. M." is to extend the modeling efforts to computer simulation of soot particle phenomena in other flames of hydrocarbon fuels along with further refinement of the underlying reaction mechanism of soot formation." Frenklach." Frenklach. M. Invited seminar at the University of Tokyo. August 14-19. FRG.. 1988. 1990. "Detailed Modeling of Soot Particles Nucleation and Growth. "Surface Growth.. Albany." FRenklach. 1990.. Department of Chemistry. Academy of Sciences Gottingen. New York. 6. p. Prepared statement at the workshop on "Current Problems in Soot Formation during Combustion. under the new AFOSR-sponsored project entitled "Development of Predictive Reaction Models of Soot Formation. March 29-30. Especially the Mechanism of Soot Formation. M. Japan. Paper presented at the Fall Technical Meeting of the Eastern States Section of the Combustion Institute." Wang. 7. Paper 12. York. M. Seattle.
Washington. The Combustion Institute. Michael Frenklach -Principal investigator Mr. 11. Reinhold. Feigelson. Oran and J. 21 .9. M. 1991. PROFESSIONAL PERSONNEL Dr. in Numerical Approaches to Combustion Modeling (E. M. in Mobil Workshop on Chemical Reactions in Complex Systems. 12. H. E D. S. D. Science. FL. "Detailed Modeling of Soot Particle Nucleation and Growth. 1991. Am. Aeronautics Astronautics Prog.. submitted.). in press. A. Wang. Twenty-Thir Symposium Internatonal on Combustion. M. W.. in press." Morgan.. "A New Mechanism for the Formation of Meteoritic Kerogen-Uike Material." Frenklach. P. in press. Inst. Hai Wang Graduate student INVENTIONS None. M. 10... Eds.." Frenklach." Frenklach. Scr. and Wang. and Frenklach... Boris.C. "Reduction of Chemical Reaction Models. "Development of Chemical Reaction Models.
12. Combust Flame 74.. 17. 79 (1986). J. p. 2.. F. 5160 (1989). Frenklach. and Stein. 131 (1988). A. 50.. L. Grcar. and Miller. "Experimental and Theoretical Analysis of the Oxidation and Growth Chemistry in a Fuel-Rich Acetylene Flame. 1 (1984). Prausnitz. M.. and Goldsmith. A. Bockhorn. E. J. R. 16. Twentieth Symposium (International) on Combustion. Jr.. C. 3714 (1985). C.. and Gill. A. 94.-L. Reid. J. J. "Optimization and Analysis of Large Chemical Kinetic Mechanisms Using the Solution Mapping Method . F. 887. M. Clary. M... J. 14.. 265 (1987). 87. The propertes of Gases and Liquids McOraw-Hill New York. AeroChem TP-482.. R. C. Jr.D.. and Gill.. G. Technol. and Gill. Frenklach. thesis. M.. Frenklach. T.. F. H. December 1985. D. p. Fetting. December 1988." Twenty-Third Fall Technical Meeting of the Eastern Section of the Combustion Institute. J. Sci. 215 (1981). Yuan.. R. J.." Paper presented at the Twenty-First Symposium (International) on Combustion. Calcote. L W. S. Smooke. 7.REFERENCES 1. E. Phys. Weiner. MIT. E. and Reh.. and Field. D... Westmoreland. Chem.. Blint. W. A..Combustion of Methane. S. J. and Keil.. Pitsburgh. Fetting. M. 4. Phys 91. F. Munich. Techool. 51. H. and Stein. Combust Sci. C. Ber. 18. M. G.. . M. 5.. Technol. R. 1033. Heddrich. and Sherwood... J. AeroChem TP-486. R. The Combustion Institute. 13. D. 1988. 15.. J. Wang.. 10. J. The Combustion Institute. Kinsey. Combust.. "A Foran Program for Modeling Steady Laminar One-Dimensional Premixed Flames. Green. F. H. H. 22 . Bockhorn. Combust Flame 42. Gardiner. Gardiner. Frenklach. and Fahr. R. CH. H. "Formation of High Molecular Hydrocarbons in Sooting Premixed Hydrocarbon-Oxygen Flames. AFOSR 88-0072. L E.. Combust Sci. Calcote.. FRG. W. 1067 (1983). 41. P. S. Chem. 8. Annual Report. R... Delfau. P." submitted for publication. R. and Warnatz.. K. Calcote.. Calcote. M.. SAND85-8240. S. Kee. August 1986. J. Twenty. Annual Report. Phys. 1990. M. L. Calcote. and Wenz. W. H. D. F... Harris. Paper 124. J. H.. F. Stein. 678. "Formation of Large Species in Sooting Flames by the Free Radical and Ionic Mechanisms. F.-F 't Symposium (Internatinal)on Combustion. 1975. and Rabinowitz. Vovelle. 11. M. W. 1986. p. 6. March 1989. T. Poster 1." Ph.. Clary. W. 3.. 9. Bunsenges." Sandia Report No. November 1989. Frenklach.. 1985. H. zeam. Annual Report.
A. and Stein. p. Nagle... m. S. 5142 (19 89). p. Harris.. Dean.F. Eds. B. S. The Combustion Institute. Weiner. Kiefer. 1989. Twent. 26. J. 1989. Eds..-Y. S. S. Y.). 443 (1988). J. 1962. C. G. Combust Flame 67. 252 (1987)..C. A.. P. C. Mahmud. M. Carbon 20. Eds.. 154. Chem. Pittsburgh. J. R. 93. 20. 1990. J.103 1. Twenty-Second Symposium Combustion. Gg. and Harris.. 3. Geithersburg. and Wei. B.. 8171 (1990). Warnatz and W.. TecimoL 59. 229 (1981). M.. (Internatonal) on 22.. M. H.Appl. A. 1918 (1987). A. 261. 198 1. 65 (1986). 29. Donnet.. 25. L J.Phy& 65. 33. 23 . 1989. M. Ping. 2013 (1985). J. and Ashcraft. Eng Sc. W.. U. and Hidaka.. C. "A Unifying Picture of Gas-Phase Formation and Growth of PAH. 39. J Phy. S.. F.. and Melius. 266 (1982). NASA Conference Publication 3061. Chem. Hwang. p. Smith. 1987. L. 32. Plenum.. B. J Phys.. Diamond and Graphite. 36. MD. F. 3. and Fetting. 38. Neob. Fahr.Colloid Interice Sd. TechnoL 32. 7. Frenklach. and Wagner. 89. Jr. E. 28. p. Twentieth Fail Technical Meeting of the Eastern States Section of the Combustion Institute. L M... 31. and Fontijn. Howard. p. M. Enffff Combust Sd. M.1023." in Carbon in the Galaxy: Studes From Earth and Space QJ. in Complex Chemical Reaction Systems. R.). 1987. F. Paper 7-1. H. J. M.. Wieschnowsky.. &.. 2. Frenklach. Proceedings of the Fift Confenc on Carbon. J. M.Matwr.. Harris. Twenty-Second Symposium (International) on Combusion. London. Springer-Verlag: Berlin. Miller. J.y Second Symposium (Internatonal)on Combustion. DeFrees... H. 27. Siegla and G. S. A. 21. and Longwell.. H.. 37.. The Combustion Institute. 3. 259. K. 133 (1988).). 10.. 343. Patel. 24. and Kennedy.L. 1. Howard. and Sarofim.W. M. Pergamon. B. Combust Sci. Haynes. Tarter.. C. M. The Combustion Institute. Jiger.19. Lin.Re&.L C. P. 91. Phys. p. E. C. K. A. New York. and Lin. M. Gardiner. Chem. Bockhorn. Mizerka. 23. in Partculat Carbon: Formation During Combustion (D. Mathematical Modelling anid Simulation (J. J. 3. 1843 (1985). Westmoreland. Chem. M. A. Frenlach. Frenklach. K.. p. Frenkach. Combust Sci. 118. Combust Flame 64. Frenklach. 34. 65 (1987). and Weiner. November 2-5. S. Soot.. Frenklach. and Spear. 35. Pittsburgh. 267 (1983). L. C. and Strickland-Constable.. Hanris S. J. C. Chang and D. 30.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue listening from where you left off, or restart the preview.