scieee Science in your language
[en] (orig)

m-Xylene Isomerization over IWW Zeolite with a Three-Modal Pore Structure: The Effect of Crystal Morphology

Author: Shamma, Emad; Seidlová, Alica; Samanta, Subhajyoti; Mazur, Michal; Opanasenko, Maksym; Shamzhy, Mariya
Publisher: Zenodo
DOI: 10.5281/zenodo.17657664
Source: https://zenodo.org/records/17657664/files/Manuscript.pdf
m-xylene isome iza ion o e IWW zeoli e wi h a h ee-modal po e
s uc u e: The e ec o c ys al mo phology
Emad Shamma, Alica Seidlo ´
a , Subhajyo i Saman a , Michal Mazu , Maksym Opanasenko,
Ma iya Shamzhy
*
Depa men o Physical and Mac omolecula Chemis y, Facul y o Science, Cha les Uni e si y, Hla o a 2030/8, P ague 12843, Czech Republic
ARTICLE INFO
Keywo ds:
Zeoli e
Xylene isome iza ion
IWW
Ge manosilica e
C ys al mo phology
Ac i e si e
Shape-selec i i y
ABSTRACT
Xylene isome iza ion is a key zeoli e-ca alyzed pe ochemical p ocess o he p oduc ion o p-xylene, a highly
demanded in e media e in he polyme indus y. While MFI- ype zeoli es a e widely used in indus y as shape-
selec i e ca alys s, xylene isome iza ion also se es as a benchma k eac ion o e alua ing he shape-selec i i y
o new zeoli e ca alys s in ela ion o hei po osi y. In his s udy, IWW zeoli e wi h a h ee-modal po e ne wo k
o isola ed 8- and 12- ing po es in e sec ed by 10- ing channels was in es iga ed o m-xylene isome iza ion, wi h
a ocus on how bo h he mul idimensional po e sys em and c ys al mo phology a ec ca aly ic pe o mance.
IWW zeoli es we e syn hesized as ge manosilica es wi h pla ele -like and needle-like c ys als, unc ionalized by
pos -syn he ic Ge- o-Al subs i u ion, and es ed in gas-phase m-xylene isome iza ion in compa ison wi h e e -
ence zeoli e ca alys s con aining unimodal 8-, 10-, o 12- ing channels. Compa ed o MFI wi h a simila con-
cen a ion o acid si es, Al-IWW ca alys s exhibi ed highe m-xylene con e sion ac oss a wide ange o WHSV
alues (4.4 – 40 h
−1
), ega dless o c ys al mo phology. Needle-like Al-IWW c ys als achie ed pa a-selec i i y
compa able o ha o MFI, while pla ele -like Al-IWW ou pe o med MFI in p-xylene yield a sho con ac imes
(WHSV =40 h
−1
). STEM analysis con i med ha he 10- ing channels a e aligned along he leng h o he needle-
like c ys als, p omo ing shape-selec i e p-xylene o ma ion. In con as , he 12- ing channels unning along he
ex ended dimension o he pla ele -like c ys als acili a e he di usion o eac an and p oduc molecules o and
om he ac i e si es. All in all, he in eg a ion o mul i-sized po es and unable c ys al mo phology in IWW
zeoli es may o e a p omising s a egy o balancing selec i i y and ac i i y in p-xylene syn hesis.
1. In oduc ion
Xylene isome iza ion is an essen ial pe ochemical p ocess o he
selec i e p oduc ion o p-xylene, a key building block in he polyme s
indus y. The highly demanded pa a-isome ypically cons i u es only
25 % o he xylene mix u e ob ained a e hyd o ea ing C9 a oma ics in
e ine ies. To inc ease he p-xylene con en , he mix u e unde goes se-
lec i e isome iza ion o e solid acid ca alys s, ypically aluminosilica e
zeoli es, wi h app op ia ely sized channel en ances and oids which
es ic p oduc dis ibu ion. Zeoli e ca alys s a e widely used in in-
dus y o xylene isome iza ion due o molecula di usion con ol inside
hei mic opo es, abundance o acid si es, and long- e m s abili y.
Indus ially used shape-selec i e zeoli e ca alys s (e.g., MFI, MOR, EUO)
a o he p e e en ial o ma ion o p-xylene and play a key ole in m-
xylene ans o ma ion, which in ol es a complex eac ion ne wo k,
including (Fig. 1): (A) monomolecula o bimolecula isome iza ion, ha
enables he in e con e sion o xylene isome s; (B) monomolecula
dealkyla ion, ha yields oluene and ligh hyd oca bons h ough he
emo al o a me hyl g oup om m-xylene; and (C) bimolecula disp o-
po iona ion, ha p oduces a mix u e o oluene and ime hylbenzenes
(TMBs) h ough me hyl g oup ans e be ween wo a oma ic molecules
[1−3].
MFI zeoli e wi h a unimodal 3D sys em o 10- ing po es is he mos
widely used ca alys o m-xylene isome iza ion due o i s high pa a-
selec i i y. I con ains wo ypes o in e sec ing channels: s aigh
channels (0.53 ×0.56 nm) and sinusoidal channels (0.51 ×0.55 nm),
wi h a channel in e sec ion o 0.86 nm. These s uc u al ea u es a e key
o bo h p oduc and ansi ion-s a e shape-selec i i y. The kine ic
diame e o p-xylene (~0.58 nm) is smalle han hose o m- and o-xylene
(~0.68 nm), acili a ing i s p e e en ial di usion h ough he MFI po es
[4]. Adso p ion s udies wi h pu e xylene isome s show a di usi i y
end o pa a >o ho >me a in MFI, wi h p-xylene exhibi ing he highes
* Co esponding au ho .
E-mail add ess: [email p o ec ed] (M. Shamzhy).
Con en s lis s a ailable a ScienceDi ec
Ca alysis Today
jou nal homepage: www.else ie .com/loca e/ca od
h ps://doi.o g/10.1016/j.ca od.2025.115563
Recei ed 27 June 2025; Recei ed in e ised o m 20 Augus 2025; Accep ed 8 Sep embe 2025
Ca alysis Today 462 (2026) 115563
A ailable online 10 Sep embe 2025
0920-5861/© 2025 The Au ho (s). Published by Else ie B.V. This is an open access a icle unde he CC BY license ( h p://c ea i ecommons.o g/licenses/by/4.0/ ).
di usion coe icien . Mo eo e , he ansi ion s a e complex o he
me a- o-o ho isome iza ion is bulkie (0.67 nm) han ha o me a--
o-pa a isome iza ion (0.62 nm), u he a o ing p-xylene o ma ion
due o lowe s e ic cons ain s. As a esul , MFI zeoli e ypically deli e s
a pa a/o ho xylene a io exceeding 2, and a low
disp opo iona ion- o-isome iza ion (Dis/Iso) a io o ~0.01, wi h
unimolecula pa hways domina ing he eac ion mechanism [3−6].
As such, xylene isome iza ion also se es as a benchma k eac ion o
assessing he shape-selec i i y o new zeoli e ca alys s in ela ion o hei
po osi y [1]. In his way, p-xylene selec i i y o e MFI zeoli e has been
u he enhanced by ailo ing c ys al mo phology. Fo ins ance,
chain-like c ys als wi h an ex ended b-axis showed a 10 % inc ease in
p-xylene selec i i y compa ed o con en ional co in-shaped c ys als.
Howe e , he na ow po es o MFI limi molecula di usion and
accessibili y o ac i e si es, which can p omo e undesi ed dealkyla ion
eac ions and educe o e all p-xylene yield [4,7,8]. These limi a ions
ha e p omp ed he de elopmen o al e na i e zeoli e ca alys s.
One such example is EUO zeoli e, designed o indus ial m-xylene
isome iza ion. I ea u es 1D 10- ing channels and special 12- ing side
pocke s, achie ing highe pa a/o ho a io han MFI (2 s. 1.35) unde
simila condi ions [9]. In combina ion wi h o he componen s, EUO
o ms he basis o he comme cial Opa is® ca alys amily, which is
widely used o he isome iza ion o e hylbenzene and m-xylene [8].
Hong and Jones [10,11] compa ed he pe o mance o zeoli es wi h
unimodal 1D 10- ing and 12- ing po e sys ems o ha o MFI. The 1D
10- ing MTT zeoli e exhibi ed a highe pa a/o ho a io (3.4) han he 3D
10- ing MFI (2.5) a 20 % con e sion. In con as , inc easing po e size,
as in 1D 12- ing AFI and 3D 12- ing IFR, esul ed in a pa a/o ho a io
nea ly wo imes lowe han ha o MFI (app oaching ~1). Mo eo e ,
he Dis/Iso a io was lowe o MTT compa ed o AFI and IFR, due o he
s onge es ic ion o bimolecula eac ions in he na owe 10- ing
po es.
MOR zeoli e wi h bimodal 2D sys em o 8- and 12- ing channels
shows highe m-xylene con e sion han 3D 10- ing MFI. I s la ge 12-
ing po es enable easie di usion bu also p omo e bimolecula
disp opo iona ion and ansalkyla ion eac ions, inc easing he o ma-
ion o oluene and ime hylbenzenes (TMBs), and he eby educing p-
xylene selec i i y. In con as , MFS zeoli e, wi h a 2D ne wo k o 8- and
10- ing po es, e ec i ely supp esses bulky byp oduc o ma ion and
shows a Dis/Iso a io oughly hal ha o MFI (~0.005 s. 0.01) [10,12].
Mo e ecen ly, m-xylene isome iza ion has been employed o e alua e
he shape-selec i i y o new UTL-de i ed iso e icula zeoli e ca alys s
wi h ailo ed mul imodal po osi y, in compa ison o MFI [13]. Zeoli e
PCR wi h 8- and 10- ing po es achie ed highe pa a/o ho a ios (4.5)
han MFI (1.4).
O e all, he li e a u e su ey sugges s ha designing zeoli e ca alys s
wi h a combina ion o di e en po e sizes ep esen s a p omising
s a egy o op imizing bo h shape-selec i i y and di usion con ol in
xylene isome iza ion. Beyond po e size, c ys al mo phology-induced
shape-selec i i y, a ising om a ia ions in c ys al size and shape, has
been demons a ed o impac eac ion pa hways in bo h xylene isom-
e iza ion and ela ed acid-ca alyzed ans o ma ions. The IWW zeoli e, a
h ee-modal ge manosilica e wi h a sys em o 8-, 10-, and 12- ing po es
and unable c ys al mo phologies, o e s a unique oppo uni y o in e-
g a e po e a chi ec u e and c ys al mo phology e ec s [14,15]. I s
10- ing channels a e associa ed wi h p omo ing monomolecula e-
ac ions and high p-xylene selec i i y, while he 12- ing po es can acil-
i a e he di usion o eac ing molecules.
Ou s udy in es iga es he unde -explo ed zeoli e IWW as a p om-
ising ca alys o m-xylene isome iza ion, while p oposing s a egies o
balance p oduc selec i i y and ca aly ic ac i i y. The pe o mance o
IWW was sys ema ically compa ed wi h zeoli es con aining unimodal 8-,
10-, o 12- ing po e sys ems, allowing us o isola e he e ec s o po e
connec i i y and size, wi h MFI se ing as he indus ial benchma k.
This compa ison unde sco es he ole o mul i-sized mic opo e ne wo k
in di ec ing ac i i y and selec i i y o IWW ca alys s. Fu he mo e, we
demons a e ha uning he c ys al mo phology o IWW o e s an
e ec i e app oach o op imize he ade-o be ween p-xylene selec i i y
and yield. Collec i ely, ou indings sugges ha in eg a ing mul i-sized
po es wi h unable mo phology in IWW zeoli es ep esen s a p omising
s a egy o balance selec i i y and ac i i y in p-xylene syn hesis.
2. Expe imen al pa
The zeoli es in es iga ed in his s udy a e lis ed in Table 1. They
include IWW zeoli es ea u ing a h ee-modal po e sys em wi h 8-, 10-,
and 12- ing channels and ei he needle- (Al-IWW-needles) o pla ele -
like (Al-IWW-pla ele s) c ys al mo phologies, as well as e e ence zeo-
li es wi h unimodal po e sys ems (i.e., ha ing uni o m sizes o he
po es), ei he non-in e sec ing (1D) o in e sec ing (3D). *BEA (CP811E)
and MFI (CBV8014) zeoli es we e p o ided by Zeolys in ammonium
o m. Al-IWW-needles and Al-IWW-pla ele s ca alys s we e syn hesized
Fig. 1. Scheme o he mechanisms o m-xylene ans o ma ion: (A) monomolecula isome iza ion, (B) monomolecula dealkyla ion, and (C) bimolecula
disp opo iona ion.
E. Shamma e al.
Ca alysis Today 462 (2026) 115563
2
Table 1
Zeoli es s udied in his wo k.
E. Shamma e al.
Ca alysis Today 462 (2026) 115563
3
by he pos -syn hesis Al- o -Ge subs i u ion in hyd o he mally syn he-
sized ge manosilica e zeoli es, IWW-needles and IWW-pla ele s,
espec i ely. O he zeoli es unde s udy we e p epa ed in aluminosili-
ca e o m by hyd o he mal c ys alliza ion.
2.1. Syn hesis o zeoli e ca alys s
CHA
3D 8- ing
zeoli e was syn hesized ollowing he p ocedu e e-
po ed in he p e iously published wo k [16], using N,N,
N- ime hyl-1-adaman ammonium hyd oxide (TCI, 25 % in H
2
O) as
SDA and eac ion mix u e wi h he mola composi ion o 1 SiO
2
: 0.35
Na
2
O: 0.05 Al
2
O
3
: 0.35 SDA: 18 H
2
O. 1.6 g o sodium hyd oxide (VWR,
99.2 %) was i s dissol ed in he SDA hyd oxide solu ion (32.96 g o
SDA in 0.77 g o wa e ). A e he comple e dissolu ion, 0.22 g o
aluminium hyd oxide (Ac os O ganics, ex a pu e) was added, and he
mix u e was le o s i o 30 min. Las ly, 16.67 g o colloidal silica
HS-40 (Sigma Ald ich, 40 % in H
2
O) was added d opwise o he mix u e,
and he eac ion mix u e was le o s i o 90 min a oom empe a u e.
0.10 g o CHA p e iously syn hesized acco ding o Zones [17] was
added o he mix u e as seeds. The eac ion mix u e was ans e ed in o
a 90 mL Te lon-lined au ocla e and le o c ys allize o 12 days in an
o en unde o a ion (60 pm) a 160 ◦C. The solid p oduc was isola ed
by il a ion, washed wi h dis illed wa e and d ied o e nigh a 60 ◦C.
The syn hesis yielded 3.0 g o CHA zeoli e. The zeoli e was calcined a
550 ◦C o 5 h wi h a hea ing a e o 1 ◦C/min unde ai low.
ESV
1D 8- ing
zeoli e was syn hesised ollowing he p ocedu e e-
po ed by B. J. Campbell e al. [18], using N,N-dime hylpipe idinium
hyd oxide as SDA, which was p epa ed by adding dime hylamine o 1,
5-dib omopen ane (1:1) in e hanol unde e lux and hen he b omide
sal ob ained was il e ed, washed wi h e hanol, and d ied. The b omide
aqueous solu ion was ion exchanged wi h Ambe sep® 900(OH) anion
exchange esin (Sigma Ald ich, 0.8 mmol o solid pe 1 g o esin) o he
hyd oxide o m o he SDA. The mola composi ion o he eac ion
mix u e was 1.0 SiO
2
: 0.3 Na
2
O: 0.04 Al
2
O
3
: 0.2 SDA: 40 H
2
O. In a
Te lon beake , 22.48 g o he hyd oxide SDA was s i ed wi h 7.04 g o
wa e , and hen 1.03 g o aluminium sul a e hexadecahyd a e (Lechne ,
99 %) was added o he solu ion. The mix u e was le o s i o 30 min
o dissol e he aluminium sal . Then, sodium silica e solu ion (37 % in
H
2
O, 18.80 g) was added d opwise, and he eac ion mix u e was le o
s i o 1 h a oom empe a u e. The mix u e was hen ans e ed in o a
Te lon-lined 90 mL au ocla e. The mix u e was c ys allized o 14 days
in an o en unde o a ion (60 pm) a 170 ◦C. The solid p oduc was
isola ed by il a ion, washed wi h dis illed wa e and d ied a 60 ◦C.
The syn hesis yielded 2.7 g o ESV zeoli e. The zeoli e was calcined a
680 ◦C o 5 h wi h a hea ing a e o 1 ◦C/min unde ai low.
TON
1D 10- ing
zeoli e was syn hesised ollowing a modi ied syn hesis
p ocess epo ed by K. Hayasaka e al. [19], using a eac ion mix u e
wi h he mola composi ion o 112 SiO
2
: 10 K
2
O: 1 Al
2
O
3
: 24.5 SDA:
3265 H
2
O, whe e 1,6-diaminohexane (Al a Aese , 98 %) was used as he
SDA. In a Te lon beake unde s i ing, 0.69 g o he SDA was dissol ed
in 11.50 g o dis illed wa e . A e a comple e dissolu ion o he SDA,
0.33 g o po assium hyd oxide (Lechne , PA) and 0.15 g o aluminium
sul a e hexadecahyd a e we e added, and he mix u e was le o s i o
dissol e all componen s. Subsequen ly, 4.0 g o colloidal silica AS-40
(Sigma Ald ich, 40 % in H
2
O) was added d opwise. The eac ion
mix u e was le o s i o 90 min a oom empe a u e. A e wa ds, he
mix u e was ans e ed o a Te lon-lined 25 mL au ocla e. The mix u e
was c ys allized in 5 days in an o en unde o a ion (60 pm) a 160 ◦C.
The solid p oduc was isola ed by cen i uga ion, washed wi h dis illed
wa e and d ied a 60 ◦C. The syn hesis yielded 3.1 g o TON zeoli e. The
zeoli e was calcined a 550 ◦C o 5 h wi h a hea ing a e o 1 ◦C/min
unde ai low.
AFI
1D 12- ing
zeoli e was syn hesized acco ding o Re . [20] using N,
N,N- ime hyl-1-adaman ammonium hyd oxide as SDA and eac ion
mix u e wi h he mola composi ion o 30 SiO
2
: 3 NaOH: 1 Al(OH)
3
: 3
SDA: 954 H
2
O. In a Te lon beake unde s i ing, 7.78 g o he SDA was
added o 47.00 g o dis illed wa e , and hen sodium hyd oxide (0.38 g)
and subsequen ly aluminium hyd oxide (0.24 g) we e added o he so-
lu ion. The mix u e was le o s i o 30 min. Las ly, 5.50 g o
CAB-O-SIL M5 umed silica (Ac os O ganics) was added unde igo ous
s i ing, and he eac ion mix u e was le o s i o 90 min a oom
empe a u e un il i became homogenous. The eac ion mix u e was
hen ans e ed in o a 90 mL Te lon-lined au ocla e and le o c ys-
allize o 8 days in an o en unde o a ion (60 pm) a 160 ◦C. The solid
p oduc was isola ed by cen i uga ion, washed wi h dis illed wa e and
d ied a 60 ◦C. The syn hesis yielded 2.2 g o AFI zeoli e. The zeoli e was
calcined a 550 ◦C wi h a hea ing a e o 1 ◦C/min unde ai low o 5 h.
The ca aly ically ac i e H- o m o CHA
3D 8- ing
, ESV
1D 8- ing
, TON
1D
10- ing
and AFI
1D 12- ing
aluminosilica es we e ob ained by ion-exchange
wi h ammonium ni a e (Sigma Ald ich, ≥99 %) and hen calcina ion.
The zeoli e powde s we e mixed wi h 1 M ammonium ni a e solu ion (1
g/100 mL). This mix u e was le o s i o 8 h a oom empe a u e
be o e il a ion. The ion-exchange p ocess was epea ed i e imes.
A e wa ds, he samples we e collec ed, d ied a 60 ◦C, and calcined a
350 ◦C o 3 h wi h a hea ing a e o 1 ◦C/min unde ai low.
IWW-needles and IWW-pla ele s zeoli es we e syn hesized as
ge manosilica es using 1,5-bis-me hylpy olidinium pen ane as SDA.
The p epa a ion o he SDA is desc ibed elsewhe e [14]. The mola
composi ion o he eac ion mix u e was a ied as ollows: (y) SiO
2
:
(1-y) GeO
2
: 0.25 SDA: 15 H
2
O, whe e y=0.90 esul ed in c ys alliza-
ion o IWW-needles and y=0.66 esul ed in he o ma ion o
IWW-pla ele s. Fo he IWW-pla ele s, 0.78 g o he ge manium sou ce
(GeO
2
, Sigma Ald ich, 99.99 %) was i s dissol ed in 4.40 g o he SDA
solu ion, con aining 3.37 g o wa e , and hen 3.18 g o TEOS (The mo
Scien i ic, 98 %) was added o he mix u e as he silicon sou ce. To
syn hesize he IWW-needles, 0.21 g o GeO
2
was dissol ed in he
aqueous solu ion o 4.40 g o he SDA, con aining 3.37 g o wa e , be o e
adding 4.10 g o TEOS (The mo Scien i ic, 98 %) o he mix u e. 70 mg
o a p e iously p epa ed IWW-needles sample was added o he syn-
hesis mix u e as seeds. The mix u e was le o s i o enough ime o
e apo a e he alcohol be o e cha ging he inal gel in o a 25 mL au o-
cla e. The syn hesis con inued o en days a 175 ◦C unde o a ion wi h
a speed o 20 pm. La e , he syn hesis was quenched, and he solid
p oduc was collec ed by cen i uga ion, washed wi h wa e and d ied a
60 ◦C. The desc ibed syn hesis p ocedu e yielded 1.4 g o IWW-pla e-
le s and 1.6 g o IWW-needles. The as-syn hesized IWW samples we e
calcined a 550 ◦C o 5 h wi h a hea ing a e o 1 ◦C/min unde ai low.
Al-IWW-needles and Al-IWW-pla ele s we e p epa ed by pos -
syn hesis Al- o -Ge subs i u ion in IWW-needles and IWW-pla ele s ze-
oli es acco ding o Re . [21]. Fo ha , he pa en ge manosilica e was
ea ed wi h 1 M aluminium ni a e (The mo Scien i ic, 98 %) solu ion
(1 g/100 mL) a 96 ℃ o ou days. The solid p oduc was collec ed by
il a ion, washed wi h excess wa e and 0.1 M HCl, and hen d ied a
60 ◦C o one day. The d y solid was calcined a 450 ◦C o 4 h.
2.2. Cha ac e iza ion
Powde XRD pa e ns we e collec ed on a B uke D8 Ad ance
di ac ome e equipped wi h a LYNXEYE XE-T de ec o and a Cu-K
α
adia ion gene a o (λ=1.54 Å). Zeoli e samples we e homogenized
using a mo a and pes le and loaded in o plas ic holde s o measu e-
men s. Di ac og ams we e ob ained in he 2θ ange be ween 3 ◦and
40 ◦.
Physiso p ion measu emen s we e pe o med on a Mic ome ics
3Flex olume ic Su ace A ea Analyze using 0.1 g o he ca alys . Each
sample was ou gassed a 110 ◦C o 1 h, wi h a hea ing a e o 1 ◦C/min,
using he Mic ome ics Sma Vac P ep sys em. Subsequen ly, he sam-
ple was ac i a ed a 250 ◦C o 8 h wi h he same hea ing a e. A o al o
40 da a poin s wi h an equilib a ion in e al o 180 s we e collec ed o
each ni ogen adso p ion iso he m a –196 ◦C (77 K) in he p/p◦ ange
be ween 0 and 1. The mic opo e olume and ex e nal su ace a ea we e
ob ained using he -plo me hod [22,23]. The o al po e olume was
E. Shamma e al.
Ca alysis Today 462 (2026) 115563
4
calcula ed a p/p◦=0.95.
SEM images we e ob ained using JEOL JSM-IT800 and a The mo
Fishe Scien i ic Scios 2 DualBeam FIB-SEM mic oscopes using second-
a y and backsca e ed elec on de ec o s. The sample was g ound and
sp ead o e ca bon ape. Images we e acqui ed a an accele a ing
ol age o 1–3 kV, a beam cu en o 10–50 pA, and a wo king dis ance
up o 10 mm. The size o c ys als was es ima ed by measu ing and
a e aging he dimensions o a leas se en c ys als using he ImageJ
so wa e.
Elemen al composi ion analysis was conduc ed by EDS using he FEI
Quan a 200 F X- ay de ec o in eg a ed in o he SEM sys em. Da a
collec ion was pe o med a an accele a ing ol age o 15 kV, a beam
cu en o 10 nA, and a wo king dis ance o 10 mm.
In si u FTIR spec oscopic measu emen s we e pe o med using a
Nicole iS50 spec ome e equipped wi h a DTGS de ec o in T ans-
mi ance mode. Spec a we e collec ed in he ange 4000–400 cm
−1
wi h a esolu ion o 4 cm
−1
wi h 64 scans. The sample was homogenized
by g inding and o med in o a sel -suppo ed wa e wi h a densi y be-
ween 8 and 12 mg/cm
2
. The wa e s we e ac i a ed unde high acuum
(≈6×10
−4
To ) a 450 ◦C o 4 h wi h a hea ing a e o 5 ◦C/min.
T ideu e oace oni ile (AN), py idine (Py) and 2,6-di- e -bu ylpy idine
(DTBPy) we e used as p obe molecules and we e degassed by eezing-
pump- haw cycles p io o adso p ion o e ac i a ed sample wa e s.
AN was adso bed a oom empe a u e a a pa ial p essu e o 3.5 To
o 20 min. Subsequen ly, i was deso bed a oom empe a u e o
20 min. Py was adso bed a 150 ◦C a a pa ial p essu e o 3.5 To o
20 min. A e wa ds, i was deso bed a 150 ◦C, 250 ◦C, 350 ◦C and
450 ◦C o 20 min. DTBPy was adso bed a 150 ◦C a he equilib ium
apo p essu e o he p obe molecule o 20 min and hen deso bed a
he same empe a u e o 60 min. Spec a we e ob ained (i) be o e
ac i a ion, (ii) a e ac i a ion, (iii) a e adso p ion o he p obe
molecule and (i ) a e deso p ion o he physiso bed p obe molecule.
Concen a ions o B øns ed (BAS) and Lewis (LAS) acid si es we e
calcula ed by in eg a ing he in ensi ies o speci ic IR bands. Fo AN,
bands a 2298 cm
−1
, 2310 cm
−1
and 2326 cm
−1
we e analysed. Mola
abso p ion coe icien s used o e alua ion o acid si e concen a ions
we e:
ε
(2298 cm
−1
, BAS) =2.05 cm/
μ
mol and
ε
(2310, 2326 cm
−1
,
weak and s ong LAS) =3.60 cm/
μ
mol [24]. Fo py idine, bands a
1455 cm
−1
and 1546 cm
−1
we e analysed and he ollowing mola ab-
so p ion coe icien s we e used:
ε
(1455 cm
−1
, LAS) =1.71 cm/
μ
mol;
ε
(1546 cm
−1
, BAS in 10- ing) =1.09 cm/
μ
mol and
ε
(1546 cm
−1
, BAS in
12- ing) =1.12 cm/
μ
mol [25]. Fo DTBPy, he band a 1615 cm
−1
was
analysed wi h he co esponding in eg a ed mola abso p ion coe icien
ε
(1615 cm
−1
, BAS) =5.3 cm
2
/
μ
mol [26].
2.3. Ca aly ic e alua ion
Gas-phase m-xylene isome iza ion expe imen s we e ca ied ou on a
Mic ome i ics Mic oac i i y E i mic o low expe imen al uni equipped
wi h a s ainless-s eel ixed-bed eac o (in e nal diame e 11 mm). P io
o he expe imen , zeoli e powde s we e p essed, g ained, and sie ed o
ob ain pa icle sizes be ween 200 and 500
μ
m. The g anula ed zeoli e
ca alys s (0.22 – 2.00 g depending on weigh hou space eloci y
(WHSV) alues, Table S1) we e dilu ed wi h ca bo undum (pa icle size
500
μ
m) o each a consis en eac o bed olume ac oss all samples.
The eac ion was pe o med unde a mosphe ic p essu e and a
350 ◦C using N
2
as he ca ie gas. Be o e he eac ions, he ca alys was
ac i a ed a 450 ◦C o 90 min unde an ai low o 200 mL/min. Sub-
sequen ly, he eac o was lushed wi h N
2
and cooled down o he e-
ac ion empe a u e. Liquid m-xylene was pumped om a ese oi o an
e apo a o , wi h a low o 0.17 mL/min, using an HPLC pump. Gaseous
m-xylene was blended wi h N
2,
and his mix u e was ed o he ca alys
bed. The low a es o N
2
and m-xylene we e main ained a 8.77 g.h
−1
and 170 mL.min
−1
, espec i ely. To p e en condensa ion o he eac-
an and p oduc s, connec ing lines and al es we e hea ed o 145 ◦C.
Typically, he du a ion o he expe imen was 215 min o ime-on-
s eam (T-O-S). The long- e m pe o mance o Al-IWW-pla ele s and
Al-IWW-needles ca alys s was s udied o 24 h TOS.
Samples we e collec ed a 30-minu e in e als, wi h he ze o h
sample aken a 5 min T-O-S. Samples o he eac ion mix u e we e
analysed using an online-connec ed Agilen 8890 Gas Ch oma og aph
(GC) equipped wi h a hea ed 6-po sampling al e, VF-WAXms column
(30 m ×0.25 mm ×1.00 µm) and an FID de ec o using N
2
as ca ie gas
and Ai /H
2
as lame igni o . FID signals (a ea) we e used o he ca a-
ly ic ac i i y e alua ion.
The alues o WHSV we e uned in he ange o 4.4–40h
−1
by a ying
he amoun o he zeoli e ca alys (Table S1) and calcula ed using Eq.
(1):
WHSV =
ρ
m−xylene •Fm−xylene
mzeoli e
(1)
whe e
ρ
m-xylene
is he densi y o m-xylene, [g/mL]; F
m-xylene
is he low
a e o m-xylene, [mL/h]; and m
zeoli e
is he mass o he zeoli e ca alys in
he eac o bed, [g], espec i ely.
Ca bon balances we e de e mined om GC-FID analyses. The FID
esponse was no malized o ca bon a om numbe , assuming nea -
cons an sensi i i y pe ca bon. The mola low was ob ained om he
GC peak a ea, mul iplied by i s numbe o ca bon a oms o yield he
ca bon con ibu ion. The o e all ca bon balance was hen calcula ed as
he a io o o al ca bon in he e luen o he ca bon ed wi h m-xylene.
In all ca aly ic es s, he balance was 92–99 %, excep in he cases
(ca bon balances o 85–86 %, o *BEA and AFI zeoli es, Table S1)
whe e amewo k-dependen coke o ma ion led o pa ial ca bon loss.
Con e sion o m-xylene, X
m-xylene
, was calcula ed using Eq. (2):
Xm−xylene =1−Am−xylene
∑Ai,no m.
•100 % (2)
whe e A
m-xylene
is he in eg a ed a ea o he GC signal o m-xylene
and A
i, no m.
is he in eg a ed a ea o he GC signal o e e y compound in
he eac ion mix u e, no malized o he mola esponse o m-xylene
based on he p oduc ca bon numbe .
Yield o a p oduc , Y
p
, was de e mined using Eq. (3):
Yp=Ap,no m.
∑Ai,no m.
•100 % (3)
whe e A
p, no m
is he in eg a ed a ea o he GC signal o a p oduc
and A
i, no m.
is he in eg a ed a ea o he GC signal o e e y compound in
he eac ion mix u e, no malized o he mola esponse o m-xylene
based on p oduc ca bon numbe .
p-Xylene selec i i y, S
p-xylene
, was de e mined using Eq. (4):
Sp−xylene =Yp−xylene
Xm−xylene
•100 % (4)
whe e Y
p-xylene
is he yield o p-xylene and X
m-xylene
is he con e sion
o m-xylene.
The a io o disp opo iona ion o isome iza ion was de e mined
using Eq. (5):
Dis/Iso =YDis.
YIso.
=Y oluene +YTMB
Yp−xylene +Yo−xylene
(5)
whe e Y
oluene
is he yield o oluene, Y
TMB
is he yield o TMB, Y
p-
xylene
is he yield o p-xylene, and Y
o-xylene
is he yield o o-xylene.
3. Resul s and discussion
3.1. S uc u e, ex u e, and acidi y o he zeoli e ca alys s
Powde XRD pa e ns e ealed he s uc u al iden i y and phase
pu i y o he s udied zeoli es, bo h e e ence zeoli e ca alys s wi h
unimodal sys em o 8-, 10- o 12- ing channels (Fig. S1) and mul imodal-
po e IWW samples. Fig. 2-A p esen s he di ac og ams o he calcined
E. Shamma e al.
Ca alysis Today 462 (2026) 115563
5

IWW samples wi h peak posi ions iden ical o he e e ence pa e ns
om [27]. The b oade di ac ion lines in he IWW-needles sample s em
om he smalle pa icle size ( ide in a) and sugges a dis inc c ys al
mo phology o he sample. The dis inc ea u es obse ed in he XRD
pa e ns o IWW zeoli es wi h di e en c ys al mo phologies a e p e-
sen ed in Fig. S2 o he Suppo ing In o ma ion and discussed he ein.
The PXRD pa e ns o he pa en IWW samples a e Al inco po a ion
main ained he posi ions o cha ac e is ic e lexes bu showed a dec ease
in hei in ensi y. The esul sugges s he p ese a ion o he s uc u e
o de ing o IWW zeoli e a a dec ease in he amewo k densi y due o
he non-equi alen subs i u ion o Ge o Al, as p e iously epo ed in
Re . [21,28,29].
The c ys al mo phology o he p epa ed zeoli es was examined by
scanning elec on mic oscopy (SEM) imaging (Fig. S3 and Fig. 3).
Zeoli e IWW was ob ained in wo dis inc c ys al mo phologies,
depending on he Si/Ge a io in he syn hesis mix u e (Fig. 3). While he
IWW-pla ele s zeoli e syn hesized om a syn hesis gel wi h Si/Ge =2
showed small pla ele -shaped c ys als (0.43 ×0.21 ×<0.10
μ
m, Table 2)
s acked on op o each o he , he IWW-needles sample ob ained om a
syn hesis mix u e wi h Si/Ge =10 demons a ed piles o e y hin
needle-shaped c ys als (0.48 ×<0.01 ×<0.10
μ
m).
SEM images we e eco ded a a ious scales o con i m he uni o -
mi y o he samples and o show pa icle dimensions. Fig. 3B,C compa es
he size and dimensionali y o IWW pla ele s and needles a he same
scale: he needles appea as hinne , elonga ed 2D pa icles, whe eas he
pla ele s exhibi la ge 3D g ow h. A highe magni ica ion (Fig. 3D),
indi idual needles a e clea ly isible, con i ming he absence o a la ge
ac ion o dense, bulky agg ega es and subs an ia ing he needle-like
mo phology o he sample.
SEM images o he Al-IWW samples a e aluminium inco po a ion
indica e ha IWW-pla ele s and IWW-needles p ese ed c ys al
mo phology, and he e was no dis o ion o agmen a ion o he c ys als
(Fig. 3- E, F).
Tex u al p ope ies o he zeoli e samples we e de e mined by N
2
physiso p ion (Fig. S4 and 2-B). IWW ge manosilica es display ype-I
iso he m, which con i ms hei mic opo ous na u e (Fig. 2-B). Fo
IWW-needles and Al-IWW-needles samples, a s eep inc ease in adso bed
olume a p/p◦>0.8 was obse ed, which likely e lec s an in e c ys-
alline adso p ion due o he loosely packed, small, needle-like c ys als.
In con as , his ea u e is absen in he iso he m o IWW-pla ele s
samples, which consis s o la ge , mo e densely packed c ys als ha
minimize in e pa icle oids.
A mino inc ease in he mic opo e olume in IWW samples upon he
Ge- o -Al subs i u ion, was obse ed independen ly o c ys al
mo phology (Table 2). This inc ease is consis en wi h p e ious epo s
and is a ibu ed o he non-equi alen eplacemen o Ge by Al, which
sligh ly inc eases he amewo k oid olume [29]. Chemical analysis
suppo s his in e p e a ion as he pa en IWW ge manosilica es ha e
Si/Ge a ios o 4.7 (IWW-pla ele s) and 12.5 (IWW-needles), while he
alumina ed samples show lowe Ge con en wi h Si/Ge a ios o 21 and
31, and Si/Al a ios o 12 and 14 o Al-IWW-pla ele s and
Al-IWW-needles, espec i ely (Table 2).
Va ia ion in he Si/Ge a io no only in luenced he mo phology o
he IWW c ys als bu also di ec ed hei c ys al g ow h. Scanning
ansmission elec on mic oscopy (STEM) e ealed he spa ial a ange-
men o pe pendicula 12- and 10- ing channels wi hin indi idual
c ys als (Fig. 4). IWW-pla ele s sample exhibi s 10- ing channels pene-
a ing he la ges c ys al plane (i.e., 0.43 ×0.21
μ
m, Fig. 4-A). When
o ien ed along he (210) plane, hese pla ele s display well-de ined
c ys alline IWW laye s wi h a cha ac e is ic d-spacing o 1.2 nm,
consis en wi h e e ence [30]. Unlike he pla ele -like c ys als, each
c ys al o IWW-needles sample g ows along he 10- ing channels, while
sho e 12- ing channels ex end om he op and bo om su aces
(Fig. 4-B). These indings a e essen ial when e alua ing he ole o po e
size and c ys al mo phology in m-xylene isome iza ion (Sec ion 3.2).
The concen a ion o acid si es, bo h BAS and LAS, was assessed by
FTIR spec oscopy o adso bed basic p obe molecules o a ying kine ic
diame e s (Fig. S5): AN (~0.38 nm), Py (~0.54 nm), and DTBPy (~0.79
nm). AN was used o quan i y he o al acid si e concen a ion, as his
smalle p obe molecule can access he acid cen es in all 8-, 10- and 12-
ing po es. Py has a size compa able o ha o he a ge ed p oduc in he
xylene isome iza ion eac ion, p-xylene, and was used o p obe acid si es
ele an o ca alysis. DTBPy was employed o assess acid si es loca ed on
he ex e nal su ace o zeoli es wi h unimodal 8- and 10- ing po e sys-
ems, as i s bigge size p e en s di usion in o <12- ing mic opo es.
The concen a ion o acid si es accessible o AN anges om 0.06 o
0.70 mmol/g ac oss he samples, wi h BAS (ac i e si es in xylene
isome iza ion [1,31,32]) ac ions be ween 39 % and 88 %, indica ing
e iden Lewis acidi y. No ably, he numbe o acid si es accessible o
AN was lowe han he o al Al con en , as e lec ed by he ΣAN/Al
(Table 3), which anges om 0.04 o 0.80. This disc epancy is
commonly obse ed in li e a u e; o ins ance, he esul s o Zholobenko
e al. sugges (BAS +LAS)/Al a ios be ween 0.47 and 0.82 o a ious
zeoli es [25]. In ou s udy, his a io was signi ican ly lowe o zeoli es
wi h 1D unimodal po es (0.04 o ESV
1D 8- ing
, 0.28 o TON
1D 10- ing
,
0.16 o AFI
1D 12- ing
) han ha o zeoli es wi h 3D unimodal channel
sys ems (0.58 o CHA
3D 8- ing
, 0.80 o MFI
3D 10- ing
, 0.72 o *BEA
3D
12- ing
) and mul imodal-po e IWW ca alys s (0.51 o bo h
Al-IWW-needles and Al-IWW-pla ele s).
As we can exclude he inaccessibili y o BAS o AN ( he
Fig. 2. A) PXRD pa e ns and B) N
2
physiso p ion iso he ms o IWW samples: IWW-needles, Al-IWW-needles, IWW-pla ele s, and Al-IWW-pla ele s.
E. Shamma e al.
Ca alysis Today 462 (2026) 115563
6
Fig. 3. SEM images o IWW wi h wo dis inc c ys al mo phologies: A, B) IWW-pla ele s a low and high magni ica ion, C, D) IWW-needles a low and high
magni ica ion, E) Al-IWW-pla ele s, and F) Al-IWW-needles.
Table 2
C ys al size, elemen al composi ion and ex u al p ope ies o he IWW samples
a
.
Ca alys Si/Al Si/Ge C ys al size,
μ
mV
mic
(cm
3
/g) V
o
(cm
3
/g) S
ex
(m
2
/g)
IWW-pla ele s *n.d. 4.7 0.43 ×0.21 ×<0.10 0.17 0.24 81
IWW-needles *n.d. 12.5 0.48 ×<0.10 ×<0.10 0.16 0.39 107
Al-IWW-pla ele s 12 21 0.43 ×0.21 ×<0.10 0.24 0.32 70
Al-IWW-needles 14 31 0.48 ×<0.10 ×<0.10 0.17 0.41 110
*n.d.- no de ec ed.
a
In o ma ion on he physicochemical cha ac e is ics o he zeoli es wi h unimodal po e sys ems is p o ided in Table S2, Suppo ing In o ma ion.
E. Shamma e al.
Ca alysis Today 462 (2026) 115563
7
disappea ance o 3610 cm
−1
band o b idging ≡Si-(OH)-Al≡g oups
a e AN adso p ion is obse ed o all he samples), he esul s a e mos
p obably ela ed o a ia ions in he spa ial dis ibu ion and coo dina-
ion en i onmen o aluminium, which a ec he s eng h and accessi-
bili y o LASs, ende ing some inac i e o unde ec able by AN unde he
applied condi ions.
Py accessed he same numbe o acid si es as AN in zeoli es AFI
1D 12-
ing
and *BEA
3D 12- ing
wi h he unimodal 12- ing po es, and in he
mul imodal Al-IWW samples o bo h mo phologies, while no acid si es
we e accessible o Py in zeoli es ESV
1D 8- ing
and CHA
3D 8- ing
wi h
Fig. 4. STEM images o : A) IWW-pla ele s and B) IWW-needles.
Table 3
Concen a ion o BASs and LASs in Al-IWW ca alys s and e e ence zeoli es wi h unimodal po e sys ems, as de e mined by FTIR spec oscopy o adso bed p obe
molecules o a iable sizes.
Ca alys c(Al)
a
(mmol/g)
c(AN)
b
(mmol/g)
c(Py)
b
(mmol/g)
c(DTBPy)
b
(mmol/g)
BAS LAS Σ Σ
AN
/Al BAS LAS Σ Σ
Py
/Al BAS BAS
DTBPy
/ Σ
BAS
ESV
1D 8- ing
1.54 0.04 0.02 0.06 0.04 *n.d. *n.d. *n.d. *n.d. *n.d. *n.d.
CHA
3D 8- ing
1.20 0.47 0.23 0.70 0.58 n.d. n.d. n.d. n.d. 0.03 0.06
TON
1D 10- ing
0.60 0.15 0.02 0.17 0.28 0.29 0.02 0.31 0.52 0.01 0.02
MFI
3D 10- ing
0.46 0.25 0.12 0.37 0.80 0.26 0.11 0.36 0.78 0.03 0.12
AFI
1D 12- ing
0.64 0.05 0.06 0.11 0.17 0.08 0.07 0.15 0.23 0.04 0.67
*BEA
3D 12- ing
0.43 0.14 0.17 0.31 0.72 0.20 0.15 0.35 0.81 0.19 1
Al-IWW-pla ele s 1.20 0.30 0.31 0.61 0.51 0.29 0.23 0.53 0.44 0.30 1
Al-IWW-needles 1.12 0.22 0.36 0.57 0.51 0.32 0.27 0.59 0.53 0.23 1
a
– based on EDS esul s.
b
– based on in si u FTIR spec oscopic esul s. * n.d.- no de ec ed
E. Shamma e al.
Ca alysis Today 462 (2026) 115563
8
unimodal 8- ing channels (Table 3).
In u n, based on DTBPy adso p ion, all 8- and 10- ing zeoli es wi h
unimodal po e sys ems showed negligible concen a ion o ex e nal
BASs (0.01 – 0.03 mmol/g), ep esen ing <12 % o he o al B øns ed
acidi y de e mined wi h AN. On he o he hand, all BAS de ec ed by
smalle p obe molecules we e also accessible o DTBPy in he 3D 12- ing
*BEA
3D 12- ing
zeoli e and he mul imodal Al-IWW zeoli es. A clea di -
e ence in acid si e accessibili y o DTBPy was obse ed be ween 12- ing
zeoli es wi h 1D and 3D po e sys ems: only 67 % o BAS we e accessible
o DTBPy in AFI
1D 12- ing
compa ed o 100 % in *BEA
3D 12- ing
.
The comp ehensi e acidi y cha ac e iza ion o he s udied zeoli es
e eals wo impo an ends:
1) The o al BAS concen a ions a y widely ac oss he zeoli es
(0.04–0.47 mmol/g). Mul imodal-po e Al-IWW ca alys s exhibi BAS
concen a ions compa able o he benchma k MFI
3D 10- ing
zeoli e
(0.22–0.30 mmol/g), while 1D zeoli es such as ESV
1D 8- ing
and AFI
1D
12- ing
showed much lowe alues (0.04–0.05 mmol/g). This a ia-
ion in ac i e si e numbe s, along wi h he di e ences in po e size
and dimensionali y, is expec ed o s ongly in luence m-xylene con-
e sion in he isome iza ion eac ion.
2) The concen a ion o ex e nal BAS is negligible in 8- and 10- ing
zeoli es (<0.03 mmol/g; <12 % o o al BAS), sugges ing a mini-
mal ole in ca alysis.
O e all, acco ding o hei basic cha ac e is ics, he p epa ed se ies
o zeoli es (Table 1) cons i u es a ep esen a i e and well-de ined se o
ca alys s sui able o add essing he ole o he mul i-size po e a chi-
ec u e and c ys al mo phology o IWW zeoli e ca alys s in he acid-
ca alyzed ans o ma ion o m-xylene.
3.2. Ca aly ic pe o mance in m-xylene isome iza ion
The ca aly ic pe o mance o he Al-IWW and e e ence zeoli es was
analyzed based on he m-xylene con e sion, p-xylene selec i i y and
yield, Dis/Iso, and p-xylene/o-xylene a ios. MFI wi h unimodal 3D
sys em o 10- ing po es was used as he co e e e ence o compa ison
because o i s indus ial signi icance.
A di ec compa ison be ween zeoli es lis ed in Table 1 a simila
WHSV does no clea ly e eal he e ec o po e size o channel a chi-
ec u e on key ca aly ic pe o mance cha ac e is ics in m-xylene isom-
e iza ion. The key pa ame e s o he ca aly ic pe o mance (i.e., p-xylene
selec i i y, p-xylene/o-xylene and Dis/Iso a ios) a e s ongly dependen
on he con e sion and a y widely among he s udied ca alys s due o
signi ican di e ences in BAS concen a ions (0.03–0.47 mmol/g,
Table 3). Speci ically, a compa able WHSV (4.4 – 8.8 h
−1
), m-xylene
con e sion anged om 0 ( o ESV
1D 8- ing
) o 48 % ( o MFI
3D 10- ing
,
Fig. S6-A). The e o e, o concen a e on he e ec o po e s uc u e, he
alues o WHSV we e uned, yielding a na owe ange o con e sion
be ween 20 % and 35 %, o be e compa ison (Fig. S6-B).
Based on he analysis o he ca aly ic pe o mance o zeoli es wi h
unimodal po e sys ems unde op imized condi ions and aking in o ac-
coun po e size-ca aly ic pe o mance co ela ions known om he
li e a u e, he in luence o po e size o zeoli e ca alys s on he selec i i y
in m-xylene isome iza ion can be summa ized as ollows:
•Al hough he 8- ing po es o ESV
1D 8- ing
and CHA
3D 8- ing
a e
gene ally conside ed oo small o m-xylene (0.73 nm) o en e [5,33,
34], ESV
1D 8- ing
shows minimal ac i i y (1–2 % con e sion),
whe eas CHA
3D 8- ing
achie es 58 % p-xylene selec i i y a 20 %
con e sion. Gi en limi ed ex e nal BAS (Table 3), he pe o mance o
CHA
3D 8- ing
can be a ibu ed o i s dynamic amewo k lexibili y,
enabling molecula di usion and p omo ing monomolecula isom-
e iza ion, as suppo ed by a low Dis/Iso (0.04) and high p/o-xylene
a io (1.5), unlike ESV
1D 8- ing
, whose igid po es and sca ce acces-
sible BAS esul in poo ac i i y (Fig. S6-A; Fig. S7). 8- ing zeoli es (e.
g., CHA
3D 8- ing
) we e unexpec edly ac i e and selec i e, achie ing
up o 58 % p-xylene selec i i y wi h negligible side eac ions.
•Bo h 1D and 3D 10- ing zeoli es, TON
1D 10- ing
and MFI
3D 10- ing
,
showed he highes p-xylene selec i i y o 63 % (in he ange o
20–35 % con e sion) and he lowes Dis/Iso (0.01), con i ming
monomolecula isome iza ion p edominance, hough only MFI
3D 10-
ing
main ains s able ac i i y wi hou deac i a ion.
•Despi e lowe BAS (0.14 s. 0.25 mmol/g) and sho e con ac ime,
3D 12- ing *BEA
3D 12- ing
shows simila m-xylene con e sion o
MFI
3D 10- ing
bu he lowes p-xylene selec i i y (44 %) due o i s la ge
po es p omo ing bimolecula pa hways, as e idenced by he highes
Dis/Iso (~0.30). In con as , 1D 12- ing AFI
1D 12- ing
, wi h compa-
able po e size, su passes *BEA
3D 12- ing
in selec i i y (51 % s. 44 %)
a e 35 min, ising o 60 % wi h ime as coke o ma ion likely na -
ows po es and shi s he eac ion owa d monomolecula isome i-
za ion (Dis/Iso d ops om 0.18 o 0.11). 12- ing zeoli es (e.g.,
*BEA
3D 12- ing
) showed he lowes p-xylene selec i i y due o he
p e alence o bimolecula side- eac ions, such as disp opo iona ion.
When compa ing he ca alys s o in e es (IWW) wi h he well-
es ablished s anda d (MFI), bo h pla ele - and needle-like Al-IWW
samples consis en ly showed g ea e m-xylene con e sion han MFI
ac oss all he es ed con ac imes (Fig. 5).
A long con ac ime (WHSV 4.4 h
−1
), Al-IWW-pla ele s and Al-IWW-
needles exhibi ed simila con e sions (~65 %), no ably exceeding ha
o MFI zeoli e (~47 %). Gi en hei compa able BAS concen a ions
(0.26–0.32 mmol/g, Table 2), he highe con e sion o Al-IWW ca alys s
can be a ibu ed o he enhanced molecula anspo h ough hei 12-
ing po e sys em. Dec easing he con ac ime (WHSV inc ease 4.4 →20
→40 h
−1
) esul ed in a s eady decline in m-xylene con e sion in MFI (47
→ 35 → 32 %) and Al-IWW-needles (65 → 55 → 44 %) (Fig. 5) in ele-
ance wi h he common explana ions: (i) sho e esidence ime limi s
he ex en o he eac ion, (ii) di usion cons ain s hinde access o
ac i e si es, and (iii) insu icien ime o each equilib ium lowe s
obse ed con e sion [10,35,36].
No ably, in compa ison o Al-IWW-needles and MFI, Al-IWW-
pla ele s main ained m-xylene con e sion in he ange 65-55 % ac oss
he same WHSV ange, sugges ing mo e s able pe o mance unde
educed con ac ime (Fig. 5). Fu he insigh s in o he long- e m s a-
bili y o Al-IWW-pla ele s and Al-IWW-needles we e ob ained om
ca aly ic es s ex ended o 24 h on s eam (Fig. S8). The esul s sugges
ha Al-IWW-pla ele s deac i a e mo e slowly han Al-IWW-needles,
wi h deac i a ion cons an s o 0.020 and 0.025 h
−1
(conside ing he
1s o de deac i a ion model). This s abili y may heo e ically o igina e
om (i) highe in insic ac i i y o BAS in Al-IWW-pla ele s leading o
as e eac ion a es; (ii) highe accessibili y o BASs leading o hei
be e u iliza ion a high h oughpu due o mo e a ou able po e
a angemen , minimizing in e nal di usion limi a ions. Hypo hesis (i)
was uled ou by py idine-FTIR he modeso p ion esul s (Fig. S9),
which showed simila acid s eng h dis ibu ions in IWW ca alys s o
bo h s udied mo phologies. In con as , hypo hesis (ii) was in line wi h
FTIR-DTBPy esul s, showing highe concen a ions o accessible BAS in
Al-IWW-pla ele s (Table 3). Fu he mo e, STEM esul s e eal he po e
a angemen in he Al-IWW samples, which may esul in mo e e icien
mass anspo in he Al-IWW-pla ele s. Indeed, he pla ele -shaped
c ys als (0.43 ×0.21 ×<0.10 µm) o e mo e balanced po e o ien a-
ion and di usion pa hs compa ed o he needle-shaped Al-IWW c ys-
als, whe e 10- ing channels un along he long edge o a c ys al (~0.48
µm), and 12- ing po es a e con ined o much sho e dimensions (<0.1
µm). This aniso opy in he needle mo phology likely es ic s di usion
and con ibu es o he obse ed decline in con e sion a highe WHSV.
While p o iding bene i s o m-xylene con e sion a sho con ac
ime, he pla ele -like mo phology o Al-IWW was less e icien o
achie ing high pa a-selec i i y. Mo phology- ela ed di e ences in p-
xylene selec i i y be ween he Al-IWW ca alys s a e appa en in Fig. 6-A.
A simila con e sion le els (44–55 %), MFI ou pe o ms bo h Al-IWW
E. Shamma e al.
Ca alysis Today 462 (2026) 115563
9