Figu re 7。  Corr elat ion betwee n app ar ent an d comput ed soot accumu lat ion rat es。

Tab le 5。 Reac tio n Scheme and Kine tic P aramete rs Emp loyed  in the  Soot Oxidation

th ese res ults, it can be concluded that th e activat ion energy of 40 k J/mol provides an optimum linear fit  with

reaction rat e law

activat ion energy Ei (k J/mol)

frequency factor Ai (mol‚K/m2‚s)

a very good corr elat ion coefficie nt of 0。989 。 This ap- proach ill ustrat es th e potent ial of combining measur e- ment s with modeling towar d providing at least a  rough

C/O2 k1  ) A1 T e-E1/RT 125 2。8 × 10-2

C/NO2 k2  ) A2 T e-E3/RT 40 5。0 × 10-1

Tab le 6。  Kine tic  P aramete rs and  Corr elation

Coe ffici en ts  in Sensitivity   Ana lysis  of Soot  + NO2

Reac tio n

indicat ion of th e govern ing kinetics at real exhau st conditions。

Conc lud ing Rema rks

In th e prese nt ed work, engine expe riment s togeth er

activat ion

energy Ei

(k J/mol)

frequency

factor Ai

(mol‚K/m2‚s) fitt ing equat iona

corr elat ion coefficie nt R2

with math emat ical modeling ha ve bee n employed to- war d un derstan ding an d describi ng th e reaction phe- nomena involving NO2 un der realistic conditions。 The followed app roach ha s a num ber  of  advanta ges  an d dra wbacks compar ed to usua l laborat ory practice。

In th e case of reaction stu dies with synth etic gas,  it is very difficult to simu lat e exactly th e reaction condi- tions in regar ds to th e composition of th e real diesel exhau st。 Moreover, th e reactor param eters, including reside nce  time,  soot  composition,  an d  packing condi-

a x: app ar ent car bon consum ption (g/h)。 y: comput ed car bon consum ption (g/h)。

perform ed th e complete set of simu lat ions using eight alternat ive  sets of kinetic constant s, as shown in Table

6。 These sets ha ve bee n selected in order to tes t in tota l eight different activat ion energies ran ging from 10 to 80 k J/mol。 For each activat ion energy E, th e preexpo- nent ial factor A ha s bee n selected so that th e product Ae-E/RT is th e sam e for T ) 350 °C。 In th is way, all sets of kinetic param eters will be equivalent at th is temper- atur e only。 It is easy to show that higher activat ion energies will yiel d higher rat es at T > 350 °C an d lower rat es at T < 350 °C。 Exactly th e opposite will hold for lower activat ion energies。

Figur e 8 prese nt s a compar ison of th e app ar ent an d th e comput ed gross soot consum ption rat es due to th e C + NO2 reaction for four different sets of kinetic param eters。 The appar ent rat e is inferr ed by adding th e net appar ent consum ption rat e (as described above) with th e ra w engine soot emissions rat e。 The ideal fitt ing line (y ) x) is also give n in th ese gra phs for compar ison with th e respe ctive fitt ing curves to facilitat e judgment of th e optimum fit。 Additiona lly, Table 6 prese nt s th e linear fitt ing laws obta ined with different param eters set for activat ion energies ran ging from 10 to 80 k J/mol, together with the respe ctive corr elation coefficie nt。 From

tions, ar e rar ely rep rese ntat ive of th e real-world con di- tions in th e part iculat e filter。 Modeling of th e reaction kinetics in laborat ory con ditions is easier an d stra ight- forwar d by plott ing Arrh enius cur ves, but th e kinetic param eters cann ot  be  directly  used  in  a complete part iculat e filter simu lat ion model。