2
Cu-zeoli e NH3-SCR ca alys s o NOx emo al in
he combined NSR–SCR echnology
Unai De La To e, Beña Pe eda-Ayo, Juan R. González-Velasco*
Depa amen o de Ingenie ía Química, Facul ad de Ciencia y Tecnología, Uni e sidad del País
Vasco, UPV/EHU, Campus de Leioa, P. O. Box 644, ES-48080 Bilbao, Bizkaia, Spain
KEYWORDS: SCR, NSR, NSR-SCR, Diesel engine, Cu-zeoli e
*Co esponding au ho : [email p o ec ed]
This is he accep ed manusc ip o he a icle ha appea ed in inal o m in Chemical Enginee ing Jou nal 207/208 : 10–17
(2012), which has been published in inal o m a h ps://doi.o g/10.1016/j.cej.2012.06.092. © 2012 Else ie unde CC BY-NC-
ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/)
3
ABSTRACT
The challenge o e ficien NOx emo al om diesel and lean-bu n engine exhaus gas by
combining NSR and SCR ca alys is s udied. Se e al Cu exchanged zeoli es ha e been
p epa ed, a ying he p epa a ion me hod (ion exchange and imp egna ion), he coppe con en
(1–6%) and he zeoli e (BETA and ZSM5). The p epa ed ca alys s ha e been cha ac e ized,
and acidi y, su ace a ea, c ys allini y and me al educ- ibili y ha e been compa ed. SCR
expe imen s unde 750 ppm NO, 750 ppm NH3 and 9.5% O2 (A o bal- ance) disc imina ed
low coppe loading, p epa ed by ion exchange ca alys (Z-IE-1.4 and B-IE-2.1) as he mos
ac i e o NOx con e sion (>95%) in ample empe a u e ange (280–450 ºC). These ac i e
SCR ca alys s we e placed downs eam a monoli h NSR P –BaO/Al O ca alys , unning unde
cycled lean– ich condi ions, and he imp o emen on NOx emo al and selec i i y o only N2
we e de e mined. In an ample ange o empe a u e, om 200 o 400 ºC, NOx con e sion was
inc eased in mo e han 30%, also no ably inc easing he p oduc ion o ni ogen, and educing
p oduc ion o ammonia and N2O below 3% and 2%, espec i ely, when compa ing he
combined NSR–SCR configu a ion e sus he single NSR ca alys .
4
1. INTRODUCTION
I is now well ecognized ha he use o diesel and lean bu n engines dec eases he uel
consump ion and he eby educes he CO2 emissions. Howe e , con en ional h ee way
ca alys s (TWCs) a e no capable o educe ni ogen oxides (NOx), due o he excess o oxygen
in he en i onmen . In he las decade, wo main ap- p oaches owa ds NOx educ ion ha e
been p oposed: he NOx s o age and educ ion (NSR) echnology and he NOx selec i e
ca aly ic educ ion (SCR).
SCR was o iginally de eloped o s a iona y emission sou ces, mainly powe plan s [1].
Howe e , i soon u ned ou o be a p omising echnology o he NOx emo al in au omobile
applica ions as well [2]. In 2005 i was in oduced o comme cial hea y-du y ehicles in
Eu ope, and mo e ecen ly also o passenge ca s [3]. The NH3-SCR con e e needs an
ex e nal sou ce o he selec i e educing agen , e.g. u ea. The u ea solu ion is injec ed in a con-
olled way in o he exhaus line, whe e i is he mally decom- posed in o NH3 and CO2. The
ammonia hen eac s selec i ely wi h NOx unde lean (oxidizing) condi ions, gi ing N2 as he
final p oduc [4,5]. Non-noble me als like Cu, Fe and Ce suppo ed ZSM5 and BETA, a e
among he mos ac i e ca alys s o he u ea/NH3-SCR p ocess [6–9].
The NSR ca alys s (also called lean NOx aps, LNT) consis o a co die i e monoli h
washcoa ed wi h a po ous alumina on which an alkali-ea h oxide (e.g. BaO) and a noble me al
(P ) a e deposi ed [10,11]. These ca alys s ope a e al e na i ely unde lean and ich condi ions
[12]. Du ing he lean pe iod, when oxygen is in excess, he pla inum oxidizes NO o a mix u e
o NOx (NO + NO2), which is adso bed (s o ed) on Ba as a ious species (ni i e, ni a e).
Be o e an unaccep able amoun o NOx slips h ough he ca alys , he engine swi ches o ich
condi ion ( educing) o a sho pe iod whe e he s o ed NOx a e eleased and educed in o N2
o e P . Di e en ypes o educing agen s such as hyd oca bon, CO and H2 ha e been used in
NSR ca alys s udies [13], and hyd ogen has been ound o be he mos e ec i e.
5
On a comme cial NSR sys em, he e iciency o ans o ming he emi ed NOx o N2 should
be as high as possible. This is ema kable, as he P i sel is selec i e o he o ma ion o N2
om NO and H2 only in a na ow ange o NO o H2 a io. No a e al. [14] concluded ha he
ammonia o ma ion o e P /Ba/Al was dependen on he amoun o s o ed NOx, empe a u e
and hyd ogen concen a ion. The educ ion by H2 o ni a es s o ed p oceeds acco ding o a
wo-s ep mechanism in which he fi s s ep is he as eac ion o hyd ogen wi h ni a es
p oducing ammonia, ollowed by he slowe eac ion o he la e wi h ni a e species leading
selec i i y o N2 [15–17]. Acco dingly a good uning o he ope a ing condi ions o bo h he
adso p ion a e and he educ ion phases can d i e selec i ely o N2 and/o NH3 [18,19].
Consequen ly, LNTs gene a e NH3 du ing he uel- ich pu ge pe iod, and SCR ca alys s,
especially hose based on zeoli es s o e significan quan i ies o NH3 unde eac ion condi ions
[20]. Hence, combining he LNT wi h a downs eam zeoli e SCR ca alys o e s a po en ial
means o cap u ing NH3 gene a ed by he LNT and using i o con e NOx ha slips h ough
he NSR ca alys . We a e e e - ing o hese sys ems as combined NSR–SCR echnology.
Co bos e al. [7,21] showed ha he NOx emo al e iciency can be g ea ly imp o ed unde
lean– ich a mosphe e i a NSR model ca alys is physically mixed wi h CuZSM-5; his e ec
was asc ibed o an inc ease in he o ma ion o NCO species, hei o ma ion being p omo ed
by Cu/ZSM-5 ca alys s. The con ol o NOx s o age and educ ion in LNTs o designing
combined NSR–SCR sys ems has been epo ed elsewhe e [22]. We p oposed he use o N2/
NH3 p oduc ion su aces in esponse o ope a ional a iables, including empe a u e and H2
concen a ion du ing he ich pe iod, o un e icien ly a combined NSR–SCR sys em wi h Fe-
BETA zeoli e ca alys placed downs eam a P –BaO/Al2O3 monoli h. Howe e , only a 2% Fe-
BETA ca alys was es ed in [22].
Cu2+ ion-exchanged ZSM5 (Cu-ZSM5) zeoli es we e fi s showing high NO decomposi ion
a es and NOx SCR ac i i ies [23,24]. Mo e ecen ly, Cu2+-exchanged be a zeoli es (Cu-BETA)
6
ha e been shown o ha e good ac i i y in he NH3-SCR o NOx, and me al-ex- changed be a
zeoli es a e gene ally ound o ha e be e hyd o he mal s abili y han simila ZSM5 ca alys s
[25].
In his pape , se e al Cu exchanged ZSM5 and BETA zeoli e ca alys s ha e been p epa ed
wi h coppe loadings be ween 1 and 6 w .% and hei NH3-SCR beha io has been compa ed
in ela ion o physico-chemical p ope ies, including physical s uc u e, edox p ope ies and
acidi y. The powde ca alys s we e placed down- s eam o a P –BaO/Al2O3 monoli h
(p e iously syn hesized and cha ac e ized elsewhe e [10]) and he significan imp o emen in
NOx emo al e iciency o N2 wi hou p ac ical NH3 slip h ough he combined NSR–SCR
sys em, unning unde cycled lean– ich a mosphe e, is demons a ed. The H2 concen a ion
du ing he ich pe iod o he NSR cycle should be adequa ely uned o he equi ed
in e media e p oduc ion o ammonia.
2. EXPERIMENTAL
2.1. Ca alys s p epa a ion
The SCR ca alys s consis ed o Cu-suppo ed zeoli es. F esh zeoli es we e supplied by
Zeolys In e na ional, namely CP414E (BETA, Si/Al = 25) and CBV5524G (ZSM5, Si/Al =
50). Zeoli es we e fi s calcined a 550 ºC o 4 h o ge he p o onic o m. The ca alys s we e
p epa ed by wo di e en con en ional p ocedu es, namely ion exchange (IE) and
imp egna ion (IM). Me al ion exchange was ca ied ou by dissol ing he equi ed amoun o
Cu(COOCH3)2 (Pan eac, 98%) in wa e . Then, 12 g H-ZSM5 o H-BETA we e added o
1.5 l o his solu ion and i was s i ed o 24 h a 65 ºC. The ion exchanged samples we e
hen fil e ed, washed wice in deionised wa e , d ied du ing all nigh and cal- cined a 550 ºC o
4 h.
7
On he o he hand, he imp egna ion me hod consis ed in adding slowly he equi ed amoun
o he p ecu so dissol ed in wa e (1.5 w .%) a 40 ºC and 3 mm Hg o some g ams o H-be a o
H- ZSM5, unde con inuous o a ion un il he sol en was e apo a ed. The samples we e d ied
and la e calcined a 550 ºC o 4 h.
The ac ual amoun o Cu in he p epa ed ca alys s was de e mined by ICP-AES om he solid
sample. All he ca alys s we e hen pelle ized, c ushed and sie ed o 0.3–0.5 mm o a oid mass
ans e limi a ions, which was checked in some p e ious expe imen s ca ied ou wi h di e en
pa icle sizes. In o de o cha ac e ize he SCR ca alys , se e al echniques we e employed, such
as BET su ace a ea analysis, H2-TPR, NH3-TPD, and XRD. The p epa ed ca alys s, wi h he
p ecu so dissolu ion concen a ion and he ac ual coppe con en , a e shown in Table 1.
The P –BaO/Al2O3 NSR monoli h ca alys was p epa ed acco ding o ou p e iously epo ed
p ocedu e [10]. In summa y, a co die i e monoli h, 20 mm in leng h and diame e , wi h a cell
densi y o 400 cells pe squa e inch and a wall hickness o 150 lm was washcoa ed wi h c-
alumina (163 m2 g-1 a e s abiliza ion a 700 ºC, 4 h) by se e al imme sions o he monoli h in o
he alumina slu y un il app ox. 1g Al2O3 was deposi ed in he monoli h s uc u e. The
inco po a ion o pla inum was ca ied ou by adso p ion om e aammine pla inum (II) ni a e
solu ion and he excess o liquid emaining in he channels was blown ou wi h comp essed ai .
A e calcina ion in ai (500 ºC, 4 h) and subsequen educ ion o he me allic phase in a 5%
H2/N2 s eam (500 ºC, 1 h), he ba ium was inco po a ed by imme sion o he monoli h in a
ba ium ace a e solu ion. Finally, he ca alys was calcined again (500 ºC, 4 h).
8
Table 1. Cha ac e is ics o he p epa ed ca alys s.
Suppo
Si/Al
Me al inco po a ion
me hodology
Cu ini ial conc. (ppm)
Exchanged amoun (%)
Ca alys con en Cu (w .%)
Nomina ion
BETA
25
Ion exchange
160
75
1.5
B-IE-1.5
Ion exchange
320
53
2.1
B-IE-2.1
Ion exchange
640
36
2.9
B-IE-2.9
Ion exchange
960
38
4.5
B-IE-4.5
Ion exchange
2000
23
5.8
B-IE-5.8
Imp egna ion
–
–
1.3
B-IM-1.3
ZSM-5
50
Ion exchange
160
70
1.4
Z-IE-1.4
Ion exchange
320
40
1.6
Z-IE-1.6
Ion exchange
640
33
2.6
Z-IE-2.6
Ion exchange
960
28
3.4
Z-IE-3.4
Ion exchange
2000
20
4.9
Z-IE-4.9
Imp egna ion
–
–
1.2
Z-IM-1.2
9
2.2. Ca alys s cha ac e iza ion
2.2.1. Ammonia empe a u e p og ammed deso p ion (NH3-TPD)
Eigh y millig ams sample (hyd a e s a e) was placed in a U- shaped qua z eac o connec ed
o a Mic ome i ics Au oChem 2910 ins umen . The sample was p e ea ed in ni ogen flow a
550 ºC o 15 min, cooled down o 100 ºC and ea ed wi h helium o 60 min. Then, he sample
was lushed wi h 10% NH3/He un il sa u a ion and he TPD was s a ed using helium as ca ie
gas (500 ml/min, STP). The ma e ial was hea ed o 550 ºC a he a e o 10 ºC/min, while he
NH3 deso p ion was con inuously moni o ed wi h a TCD de ec o . The amoun o ammonia
deso bed a some gi en empe a u e ange was aken as he acid si e concen a ion, whe eas
he empe a u e ange a which mos o he ammonia was deso bed indica ed he acid s eng h
dis ibu ion.
2.2.2. Su ace a ea
The BET su ace a eas o he zeoli e samples we e de e mined by N2 adso p ion–deso p ion
a -196 ºC using a Mic ome i ics ASAP 2020 equipmen .
2.2.3. XR di ac ion
The change in c ys alline s uc u e o he Cu modi ied zeoli e samples we e analyzed by XRD
(Philips PW1710 di ac ome e ). The samples we e inely g ound and we e subjec ed o Cu
Ka adia ion in con inuous scan mode om 5º o 80º o 2h wi h 0.02º pe second sampling
in e al. PANaly ical X’pe HighSco e speci ic so wa e was used o da a ea men . JCPDS
da abase was used o con i m he spec um.
2.2.4.
Hyd ogen empe a u e p og ammed educ ion (H
2
-TPR)
Reducibili y o Cu in he ca alys was in es iga ed by empe a u e-p og ammed educ ion
(TPR) using H2. The sample was p e- ea ed in 30 ml/min o 10% O2/He mix u e gas flow a
550 ºC o 45 min and hen cooled down o 30 ºC and flushed wi h helium o 60 min. Then
samples we e hea ed om oom empe a u e o 600 ºC wi h 10 ºC/min amp in a 60 ml/min o
10
5% H2/A mix u e gas flow. The wa e o med du ing educ ion wi h H2 was apped using a
cold ap and he hyd ogen consump ion was con inuously moni o ed wi h a TCD de ec o .
2.3. Ac i i y es s
2.3.1. SCR expe imen s
The SCR expe imen s we e pe o med in a downflow s ainless s eel eac o . The eac o ube,
wi h 1 g o 0.03–0.05 mm pelle ized Cu-zeoli e SCR ca alys inside, was loca ed in o a 3-zone
ube u nace. The empe a u e was measu ed by a he mocouple a he op o he ca alys bed.
The eac ion empe a u e was a ied om 100 o 500 ºC. The composi ion o he eed gas
mix u e was 750 ppm NO, 750 ppm NH3 and 9.5% O2 using A as he balance gas. Gases we e
ed ia mass flow con olle s and he o al flow a e was se a 3000 ml min-1, which
co esponded o a space eloci y (GHSV) o 90,000 h-1. P e ious expe imen s made wi h
GHSV o 22,500 and 45,000 h-1 achie ed almos 100% con e sion in a wide ange o
empe a u es (220–460 ºC) wi h all he ca alys s p epa ed, which made di icul compa ison o
beha io and elec ion o he bes candida e o he double NSR–SCR con igu a ion. The NO,
NO2, NH3 and N2O concen a ion a he eac o exi we e moni o ed e e y 40 ºC, once he
analysis has been s abilized o a leas 10 min, by online FTIR mul igas analyze (MKS 2030).
The NOx and NH3 con e sions we e calcula ed as
in ou
NO NO
NO in
NO
100
FF
X
F
−
= ×
(1)
33
3
3
in ou
NH NH
NH in
NH
100
FF
X
F
−
= ×
(2)
and he N2, N2O, and NO2 selec i i ies we e calcula ed as
2
2
33
ou
N
Nin in
NO NO NH NH
2F
S
FX F X
=+
(3)
17
Acidi y o he zeoli e is one o he impo an pa ame e ha de e mines he ex en o NOx
educ ion wi h ammonia o e zeoli e based ca alys [33]. The acidi y o all p epa ed ca alys s
was de e - mined by NH3-TPD and he alues a e gi en in Table 3. All he samples exhibi ed
wo majo deso p ion peaks; o ZSM5 based ca alys s he i s deso p ion peak was si ua ed in
he ange 175– 210 ºC co esponding o weak acid si es, whe eas he second deso p ion peak,
co esponding o s ong acid si es, was de ec ed in he ange 220–370 ºC. Fo BETA based
ca alys , he low empe a u e deso p ion peak was coinciden wi h ha obse ed o ZSM5, bu
he high empe a u e deso p ion peak was obse ed in a much mo e na ow window, i.e. 250–
280 ºC. The high empe a u e deso p ion peak, o s ong acid si es, can be ela ed o he
p esence o B øns ed acid si es in zeoli e ca alys s [34–36]. The quan i y o deso bed ammonia
du ing TPD expe imen can illus a e he numbe o acid si es in he sample, which gene ally
inc eases wi h he coppe con en , he s ong acidi y in mo e ex ension. I can be also obse ed
ha he amoun o deso bed ammonia a lowe empe a u e is in gene al lowe han he
deso bed ammonia a highe empe a u es, which means a supe io i y o B øns ed acid si es in
compa ison wi h Lewis acid si es.
18
Table 3. Acidi y o Cu-zeoli e ca alys s, de e mined by NH3-TPD.
Sample
Weak acidi y
S ong acidi y
To al
µmol NH
3
g-1
T (ºC)
µmol NH
3
g-1
T (ºC)
µmol NH
3
g-1
H-ZSM5
183
197
237
371
420
Z-IE-1.4
161
210
256
246
417
Z-IE-1.6
122
197
295
220
417
Z-IE-2.6
46
188
380
274
426
Z-IE-3.4
94
183
359
269
453
Z-IE-4.9
176
200
379
235
555
Z-IM-1.2
50
185
248
273
298
H-BETA
388
183
151
278
539
B-IE-1.5
304
183
307
264
611
B-IE-2.1
251
210
395
258
646
B-IE-2.9
205
194
464
255
669
B-IE-4.5
312
174
444
266
756
B-IE-5.8
328
191
746
257
1074
B-IM-1.3
174
182
433
257
607
Fo ZSM5 suppo ed ca alys s, he o al acidi y g ows g adually wi h he coppe con en , om
420 µmol NH3 (g ca .)-1 co esponding o he ba e zeoli e o 555 µmol NH3 (g ca .)-1 o Z-IE-
4.9, when he coppe was inco po a ed by L.I.E. On he o he hand, he inco po a ion o coppe
by imp egna ion educed he o al acidi y o he ca alys o 298 µmol NH3 (g ca .)-1 o Z-IM-
19
1.2, p obably due o he blockage o he zeoli e po es by coppe agg ega es which quan i y is
expec ed o be less o liquid ion exchange ca alys s. Simila ends can be obse ed o Cu-
BETA ca alys s.
The inco po a ion o coope o he zeoli e suppo also a ec ed no ably he su ace a ea o
he ca alys s. The esh ZSM5 zeoli e p esen ed a su ace a ea o 398 m2 g-1 which g adually
dec eased wi h inc easing coope loading. In ac , when 4.9% o Cu (Z-IE-4.9) was
inco po a ed h ough liquid ion exchange o he zeoli e, he su ace a ea dec eased o 350 m2
g-1 (Table 4). On he o he hand, he inco po a ion o Cu h ough we ness imp egna ion led o
a much la ge dec ease in he exposed su ace a ea (298 m2 g-1, Z- IM-1.2), which was ela ed
o he p esence o coope agg ega es blocking he po es. In he case o BETA suppo ed
ca alys s, a simila end was obse ed. The su ace a ea o he ba e zeoli e (BETA, 585 m2 g-1)
g adually dec eased o 459 m2 g-1 (B-IE-5.8) when Cu was inco po a ed by liquid ion exchange,
whe eas he su ace a ea o he B-IM-1.3 was educed o 428 m2 g-1 when Cu was added by
imp egna ion.
Table 4. Su ace a ea o he p epa ed SCR ca alys s
ZSM5 samples
Su ace a ea (m2 g-1)
BETA samples
Su ace a ea (m2 g-1)
H-ZSM5
398
H-BETA
585
Z-IE-1.4
385
B-IE-1.5
562
Z-IE-1.6
380
B-IE-2.1
493
Z-IE-2.6
361
B-IE-2.9
481
Z-IE-3.4
374
B-IE-4.5
463
Z-IE-4.9
350
B-IE-5.8
459
Z-IM-1.2
298
B-IM-1.3
427
20
3.2. Ammonia SCR ac i i y es s
The NOx selec i e ca aly ic educ ion ac i i y (SCR) es s o he p epa ed ca alys s was
ca ied ou unde a eeds eam wi h he ol- lowing composi ion: 750 ppm NO, 750 ppm NH3,
9% O2 and A o balance. The o al flow a e was se a 3000 ml min-1, which co esponded o
a space eloci y (GHSV) o 90,000 h-1. This GHSV, high- e han used in p ac ice, allowed
bes compa ing he beha io o he p epa ed ca alys s, as makes he e ec o empe a u e mo e
ele an . The NO, NO2 NH3 and N2O concen a ions a he eac o exi we e moni o ed om
100 o 500 ºC, e e y 40 ºC, once he analysis had been s abilized o 10 min.
Fig. 2a and b shows he con e sions o NOx and NH3 as a unc- ion o he eac ion
empe a u e. Fo each suppo , ZSM5 and BETA, h ee ca alys s we e chosen as ep esen a i e
o low, in e media e and high Cu con en , p epa ed by ion exchange. The coppe low con en
ca alys s p epa ed by imp egna ion a e also included in Fig. 2. The con e sion ends a e
ypical o NOx SCR eac ions, eaching a maximum in ac i i y o an in e media e empe a u e
[37,38]. Fo he samples s udied, he NOx con e sion maximum was eached be ween 350 and
450 ºC o BETA suppo ed ca alys s and be ween 250 and 350 ºC o ZSM5 suppo ed
ca alys s. The highe ac i i y o ZSM-5 based ca alys s a lowe empe a u es is p obably
ela ed wi h he highe educibili y o Cu a low empe a u e obse ed in TPR expe imen s.
21
Fig. 2. Con e sion o NOx (filled symbols) and NH3 (emp y symbols) o (a) BETA and (b)
ZSM5 suppo ed ca alys s o SCR eac ion. Coppe con en : ■ low IE, in e media e IE,
high IE and low IM.
Table 5 shows he maximum NOx con e sion o each ca alys and he empe a u e a which
i has been eached ( h ee i s columns). In he case o BETA suppo ed ca alys s, B-IE-2.1
showed he maximum NOx con e sion o 95.8% a 420 ºC. Fo ca alys s wi h highe coppe
loadings he maximum NOx con e sion dec eased o 87.3% and 82.7% o B-IE-4.5 and B-IE-
5.8, espec i ely, which in- s ead we e achie ed a lowe empe a u e, 340 ºC. I is sugges ed
ha he inc ease o Cu loading p omo es he oxida ion o NO o NO2 a lowe empe a u e,
which ac i a es he as SCR eac ion (2NH3 + NO + NO2 → 2N2 + 3H2O), and consequen ly
shi s he maximum NOx con e sion o lowe empe a u e.
22
Table 5. Maximum NO con e sion and co esponding NH3 con e sion and N2/N2O a his empe a u e.
Sample
Maximum NO
x
con e sion
Tempe a u e window o X
NO
> 70%
X
NO
(%)
T (ºC)
X
NH3
(%)
S
N2
(%)
S
N2O
(%)
B-IE-2.1
95.8
420
98.6
98.4
1.3
280–520 ºC (240)
B-IE-4.5
87.3
340
99.2
95.5
4.5
275–440 ºC (165)
B-IE-5.8
82.7
340
98.4
93.5
6.5
250–395 ºC (145)
B-IM-1.3
82.7
380
99.8
97.6
2.4
320–450 ºC (130)
Z-IE-1.4
97.9
340
99.7
99.5
0.5
280–440 ºC (160)
Z-IE-3.4
93.6
340
99.3
96.1
3.9
235–425 ºC (190)
Z-IE-4.9
89.3
300
99.7
94.9
5.1
205–375 ºC (170)
Z-IM-1.2
75.2
300
98.9
98.6
1.1
260–355 ºC (75)
23
The ou h column in Table 5 shows ha ammonia con e sion eached almos 100% o NOx
maximum con e sion. Only N2 and N2O (no NO2) we e de ec ed a he eac o exi as deduced
om selec i i ies shown in columns i h and six h, which close he mole balance wi h N2 and
N2O. I can be no ed ha highe coppe loadings enhance selec i i y o N2O, bu always
main ained below 6.5%. The las column in Table 5 indica es he ampli ude o he empe a u e
windows o main aining NOx con e sion highe han 70%. The wide ampli ude was achie ed
wi h low coppe con en (B-IE-2.1) and i was dec eased wi h he coppe loading. Thus, i can
be concluded ha among he BETA ca alys s p epa ed by ion exchange, he low Cu con en B-
IE-2.1ca alys esul ed he mos ac i e as i combines he highes NOx con e sion and he wide
empe a u e window. Al hough he inc ease in he coppe con en has a bene icial e ec on he
ac i i y a low empe a u e, howe e he maximum NOx con e sion and empe a u e window
ampli ude a e signi ican ly educed due o some ine ec i e use o coppe . The ca alys B-IM-
1.3, p epa ed by imp egna ion, achie ed he lowes maximum con e sion and he na owes
empe a u e window, as i was expec ed due o he low Cu dispe sion ob ained by his
p epa a ion me hod.
In he case o ZSM5 suppo ed ca alys s p epa ed by ion exchange, as he coppe con en
inc eases he NOx con e sion cu e shi ed o lowe empe a u es whe eas he ca alys achie ed
lowe maximum NOx con e sions. This end is simila o ha p e iously desc ibed o BETA
ca alys s; howe e , all Cu-ZSM5 ca alys s p esen ed e y simila ampli udes o empe a u e
windows. Thus, he bes ZSM5 ca alys has o be chosen only based on he maxi- mum NOx
con e sion, esul ing in he low Cu loading Z-IE-1.4. Again, he imp egna ed Z-IM-1.2 can be
dis ega ded due o i s wo se beha io , wi h 75.2% o maximum NOx con e sion and na ow
empe a u e window ampli ude o 75 ºC.
3.3. NSR–SCR ac i i y es s
24
The mos ac i e ca alys o SCR eac ion, B-IE-2.1 and Z-IE-1.4, we e used o accomplish
he double NSR–SCRcon igu a ion, by adap ing he SCR ca alys downs eam he NSR
ca alys . The monoli h NSR ca alys used has been he same o all expe imen s [10]. The
expe imen s in he double sys em we e made wi h he ollowing eeds eam composi ion: 750
ppm NO, 9.5% O2 in A du ing he lean pe iod (150 s); and 750 ppm NO, 4% H2 in A du ing
he ich pe iod (20 s). Concen a ions o N2/NH3/N2O we e moni o ed a he in e media e
posi ion, i.e. a e NSR bu be o e SCR, whe e p oduc ion o ammonia as eac an o
subsequen SCR should be impo an , and also hey we e moni o ed a he exi o he double
NSR–SCR eac o .
Fig. 3 shows he e olu ion o NOx con e sion and he NH3 and N2 p oduc ion wi h
empe a u e, o he simple NSR con igu a ion and o he combined NSR–SCR con igu a ion,
he la e wi h B-IE-2.1 and Z-IE-1.4 SCR ca alys s. I can be obse ed ha he addi ion o an
SCR ca alys downs eam he NSR ca alys imp o es signi ican ly he NOx con e sion, mainly
o empe a u es be ween 200 and 250 ºC. The maximum NOx con e sion achie ed wi h he
single NSR esul ed in 50%, and i was inc eased up o 76% wi h he double NSR–SCR
con igu a ion, independen ly o he SCR ca alys . In conco dance wi h he SCR ac i i y
p e iously analyzed, he Z-IE-1.4 ca alys made he double NSR–SCR con igu a ion mo e
ac i e a lowe empe a u e (200 ºC) han he B-IE-2.1 (250 ºC). I is wo hy o no e ha he
combined NSR–SCR sys em achie ed e icien NOx educ ion a lowe empe a u es han NSR
o SCR single sys ems wi h he aid i in e media e ammonia o med du ing he ich pe iod o
he NSR p ocess in he SCR placed downs eam, bu no need o ex e nal NH3 eed as in he
case o he single SCR sys em.
25
Fig. 3. (a) NOx con e sion, (b) NH3 p oduc ion, and (c) N2 p oduc ion o single NSR and
double NSR–SCR con igu a ions, om 150 o 400 °C. (Z o ZSM5, B o BETA).
The p oduc dis ibu ion a he exi o he eac o is ma kedly in luenced by he con igu a ion
used (Table 6). The N2O p oduc ion was always below 3% and e en dec eases wi h
empe a u e below 1%. The double NSR–SCR con igu a ion dec eases he NH3 p oduc ion
26
(Fig. 3b) in a o o highe N2 p oduc ion (Fig. 3c), in he whole s udied empe a u e ange.
This beha io imp o emen can be explained om da a shown in Fig. 4.
Fig. 4. NOx and NH3 concen a ion p ofiles, and N2 MS-signal p ofiles o single NSR and
combined NSR–SCR con igu a ions, a 200 ºC.
33
[15]
I. No a, L. Lie i, P. Fo za i, Mechanis ic aspec s o he educ ion o s o ed NOx o e
P –Ba/Al2O3 lean NOx ap sys ems, Ca al. Today 136 (2008) 128.
[16]
S.S. Mulla, S.S. Chaugule, A. Yeze e s, N.W. Cu ie , W.N. Delgass, F.H. Ribei o,
Regene a ion mechanisms o P /BaO/Al2O3 lean NOx ap ca alys wi h H2, Ca al.
Today 136 (2008) 136.
[17]
B. Pe eda-Ayo, J.R. González-Velasco, R. Bu ch, C. Ha dac e, S. Chansai,
Regene a ion mechanisms o a lean NOx ap (LNT) ca alys in he p esence o NO
in es iga ed using iso ope labeling echniques, J. Ca al. 285 (2012) 177.
[18]
B. Pe eda-Ayo, D. Du aiswami, J.J. Delgado, R. López-Fonseca, J.J. Cal ino, S.
Be nal, J.R. González-Velasco, Tunning ope a ional condi ions o e ficien NOx
s o age and educ ion o e P –Ba/Al2O3 monoli h ca alys , Appl. Ca al. B: En i on. 96
(2010) 329.
[19]
B. Pe eda-Ayo, D. Du aiswami, J.A. González-Ma cos, J.R. González-Velasco,
Pe o mance o NOx s o age- educ ion ca alys in he empe a u e– educ an
concen a ion domain by esponse su ace me hodology, Chem. Eng. J. 169 (2011) 58.
[20]
G. Guo, J. Wa ne , G. Ca a aio, D. Dobson, E. Badillo, C. Lambe , The De elopmen
o Ad anced U ea-SCR Sys ems o Tie 2 Bin 5 and Beyond Diesel Vehicles, SAE
Technical Pape 2010-01-1183.
[21]
E.C. Co bos, M. Haneda, X. Cou ois, P. Ma eco , D. Dup ez, H. Hamada, NOx
aba emen o lean-bu n engines unde lean– ich a mosphe e o e mixed NSR–SCR
ca alys s: influences o he addi ion o a SCR ca alys and o he ope a ional condi ions,
Appl. Ca al. A: Gen. 365 (2009) 187.
34
[22]
B. Pe eda-Ayo, D. Di aka , J.R. González-Velasco, Con ol o NOx s o age and
educ ion in NSR bed o designing combined NSR–SCR sys ems, Ca al. Today 172
(2011) 66.
[23]
M. Iwamo o, H. Yahi o, Y. To ikai, T. Yoshioka, N. Mizuno, No el p epa a ion
me hod o highly coppe ion-exchanged ZSM5 zeoli es and hei ca aly ic ac i i ies o
NO decomposi ion, Chem. Le . 11 (1990) 1967.
[24]
A. G ossale, I. No a, E. T onconi, D. Cha e jjee, M. Weibel, The chemis y o he
NO/NO2–NH3 ‘‘ as ’’ SCR eac ion o e Fe-ZSM5 in es iga ed by ansien eac ion
analysis, J. Ca al. 256 (2008) 312.
[25]
S. B andenbe ge , O. K öche , A. Tissle , R. Al ho , The s a e o he a in he
selec i e ca aly ic educ ion o NOx by ammonia using me al-exchanged zeoli e
ca alys s, Ca al. Re . Sci. Eng. 50 (2008) 492.
[26]
G. Cen i, S. Pe a hone , Na u e o ac i e species in coope -based ca alys s and hei
chemis y o ans o ma ion o ni ogen oxides, Appl. Ca al. A: Gen. 132 (1995) 179.
[27]
C. To e-Ab eu, M.F. Ribei o, C. Hen iques, G. Delahay, Cha ac e isa ion o CuMFI
ca alys s by empe a u e p og ammed deso p ion o NO and empe a u e p og ammed
educ ion. E ec o he zeoli e Si/Al a io and coppe loading, Appl. Ca al. B. 12 (1997)
249.
[28]
R. Bulánek, B. Wich e lo á, Z. Sobalik, J. Tichy´ , Reducibili y and oxida ion ac i i y
o Cu ions in zeoli es: e ec o Cu ion coo dina ion and zeoli e amewo k
composi ion, Appl. Ca al. B 31 (2001) 13.
35
[29]
J.Y. Yan, G.D. Lei, W.M.H. Sach le , H.H. Kung, The ac i a ion o Cu/ZSM5 ca alys s
o lean NOx educ ion: cha ac e iza ion o changes o Cu s a es and zeoli e suppo , J.
Ca al. 161 (1996) 43.
[30]
J.Y. Yan, W.M.H. Sach le , H.H. Kung, E ec o Cu loading and addi ion o modifie s
on he s abili y o Cu/ZSM5 in lean NOx educ ion ca alysis, Ca al. Today 33 (1997)
279.
[31]
B. Moden, J.M. Donohue, W.E. Co mie , H.-X. Li, E ec o Cu-loading and s uc u e
on he ac i i y o Cu-exchanged zeoli es o NH3-SCR, in: P oc. 4 h In . FEZA
Con e ence, 2008, pp. 1219–1222.
[32]
P. Da Cos a, B. Moden, G.D. Mei zne , D.K. Lee, E. Iglesia, Spec oscopic and
chemical cha ac e iza ion o ac i e and inac i e Cu species in NO decomposi ion
ca alys s based on Cu-ZSM5, Phys. Chem. Chem. Phys. 4 (2002) 4590.
[33]
A. Sul ana, T. Nanba, M. Sasaki, M. Haneda, K. Suzuki, H. Hamada, Selec i e ca aly ic
educ ion o NOx wi h NH3 o e di e en coppe exchanged zeoli es in he p esence o
decane, Ca al. Today 164 (2011) 495.
[34]
N.A.S. Amin, S. Ammasi, Dual-bed ca aly ic sys em o di ec con e sion o me hane
o liquid hyd oca bons, J. Na . Gas Chem. 15 (2006) 191.
[35]
C. Cos a, I.P. Dzikh, J.M. Lopes, F. Lemos, F.R. Ribei o, Ac i i y–acidi y ela ionship
in zeoli e ZSM-5. Applica ion o B öns ed- ype equa ions, J. Mol. Ca al. A: Chem. 154
(2000) 193.
36
[36]
D.W. Fickel, E. D’Addioa, J.A. Lau e bacha, R.F. Lobo, The ammonia selec i e
ca aly ic educ ion ac i i y o coppe -exchanged small-po e zeoli es, Appl. Ca al. B 102
(2011) 441.
[37]
J.H. Kwak, D. T an, S.D. Bu on, J. Szanyi, J.H. Lee, C.H.F. Peden, E ec s o
hyd o he mal aging on NH3–SCR eac ion o e Cu/zeoli es, J. Ca al. 287 (2012) 203.
[38]
J. Li, H. Changa, L. Maa, J. Haoa, R.T. Yang, Low- empe a u e selec i e ca aly ic
educ ion o NOx wi h NH3 o e me al oxide and zeoli e ca alys s – a e iew, Ca al.
Today 175 (2011) 147.
[39]
U. Elizundia, R. López-Fonseca, I. Landa, M.A. Gu ié ez-O iz, J.R. González-
Velasco, FTIR s udy o NOx s o age mechanisms o e P –BaO/Al2O3 ca alys s. E ec
o he P –BaO in e ac ion, Topics Ca al. 42–43 (2007) 37.
