Spec o-kine ics o he me hanol o hyd oca bons eac ion combining
online p oduc analysis wi h UV– is and FTIR spec oscopies h oughou
he space ime e olu ion
José Valecillos
a,
⇑
, Hec o Vicen e
a
, Ana G. Gayubo
a
, And és T. Aguayo
a
, Ped o Cas año
a,b,
⇑
a
Depa men o Chemical Enginee ing, Uni e si y o he Basque Coun y (UPV/EHU), P.O. Box 644, Bilbao 48080, Spain
b
Mul iscale Reac ion Enginee ing KAUST Ca alysis Cen e (KCC), King Abdullah Uni e si y o Science and Technology (KAUST), Thuwal 23955-6900, Saudi A abia
a icle in o
A icle his o y:
Recei ed 22 No embe 2021
Re ised 3 Feb ua y 2022
Accep ed 23 Feb ua y 2022
A ailable online 28 Feb ua y 2022
Keywo ds:
Me hanol o hyd oca bons (MTH)
ZSM-5 (MFI) zeoli e
In si u spec oscopy
Reac ion kine ics
Coke deac i a ion
abs ac
The well-s udied me hanol o hyd oca bons eac ion o e a ZSM-5 zeoli e ca alys has been used o
de elop a spec o-kine ic app oach o ob ain an o e all eac ion mechanism in ol ing bo h e ained spe-
cies and gas-phase p oduc s. We combined wo in si u spec oscopic echniques (ul a iole – isible and
Fou ie - ans o m in a ed spec oscopies) wi h online p oduc analysis o ob ain he ime- and space
ime- esol ed e olu ion o he en i e eac ion media. A ZSM-5 zeoli e ca alys was es ed in wo comme -
cial spec oscopic cells a 400 °C using di e en space imes (di e en inle low a es). Speci ically, ou
wo k ocusses on he e ec o he space ime (key pa ame e in any kine ic s udy) and how o une o he
pa ame e s such as pa ial p essu e o me hanol o esol e, om he spec oscopic and gas-phase poin s
o iew, he mechanisms o eac ion and deac i a ion. Ou app oach ein o ces he p e ious in e p e a-
ion o hese wo combined ne wo ks in he selec ed eac ion, hus, p o ing ha he spec o-kine ic
app oach is a obus me hodology o simul aneously build o e all eac ion and deac i a ion mechanisms.
Ó2022 The Au ho (s). Published by Else ie Inc. This is an open access a icle unde he CC BY-NC-ND
license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
1. In oduc ion
The ca aly ic me hanol o hyd oca bons (MTH) eac ion is he
basis o he de elopmen o a obus al e na i e p ocess o he
p oduc ion o gasoline, ligh ole ins (e hylene, p opylene, and
bu ylenes), and a oma ics [1–6]. The mos -used ca alys s a e based
on medium-po e ZSM-5 zeoli es and small-po e SAPO-34 zeo ypes
[3,6–10] o he de elopmen o he me hanol- o-p opylene (MTP)
and me hanol- o-ole in (MTO) p ocess echnologies, espec i ely.
The ZSM-5 ca alys s a e mo e s able han SAPO-34 o simila
small-po e ca alys s (such as SAPO-18) because he ZSM-5 chan-
nels acili a e he di usion o species ha a e bo h eac ion
in e media es and coke p ecu so s, signi ican ly educing he coke
o ma ion and deac i a ion a es [11,12]. In addi ion, SAPO-34 and
simila small-po e ca alys s a e ema kably selec i e o ligh ole-
ins because o hei shape selec i i y [12,13], albei hey unde go
as e deac i a ion ha can be con olled by inc easing he wa e
concen a ion in he eac ion medium [14]. Howe e , he p ope
balance o acid p ope ies o ZSM-5 ca alys s inc eases he selec i -
i y o ligh ole ins, pa icula ly p opylene, wi h he su plus ad an-
age o hei s abili y p o ided by hei channel s uc u es and
balanced acidi y [10,15].
The unce ain y o he eac ion mechanism and he apid ca a-
lys deac i a ion ha e been key que ies o gene a ing hund eds o
esea ch s udies, wi h in e es ing esul s cla i ying hese issues,
summa ized in a ious e iews [1,6,16–18]. In he mos gene al
sense, he cu en b oade consensus is ha he MTH eac ion p o-
ceeds h ough a mechanism comp ising ini ia ion, au oca alysis,
and deac i a ion pe iods. In he ini ia ion pe iod, me hanol and
dime hyl e he a e adso bed on he acid si es o ming su ace
me hoxy species ha eac o yield oxygena ed in e media es.
These in e media es a e decomposed in o ole ins ha also eac ,
p o iding a oma ics h ough oligome iza ion, cycliza ion, and
hyd ogen ans e eac ions. Once he e is an incipien o ma ion
o ole ins and a oma ics, he au oca aly ic pe iod s a s wi h he
h ps://doi.o g/10.1016/j.jca .2022.02.021
0021-9517/Ó2022 The Au ho (s). Published by Else ie Inc.
This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
Abb e ia ions: FTIR, Fou ie - ans o m in a ed; GC, gas ch oma og aph; MTH,
me hanol o hyd oca bons; MTO, me hanol o ole ins; MTP, me hanol o p opylene;
MS, mass spec ome e ; UV– is, ul a iole - isible Va iables; F
M0
, weigh o
ca bon-based mola low a e o me hanol in he eed; F
W0
, weigh low a e o
wa e in he eed; P
M0
, me hanol pa ial p essu e in he eed; S
i
, ca bon-based
p oduc iselec i i y; , ime on s eam; T, empe a u e; W, ca alys weigh ; X,
ca bon-based ac ional con e sion o oxygena es; Y
i
, ca bon-based p oduc iyield.
⇑
Co esponding au ho s a : Depa men o Chemical Enginee ing, Uni e si y o
he Basque Coun y (UPV/EHU), P.O. Box 644, Bilbao 48080, Spain.
E-mail add esses: [email p o ec ed] (J. Valecillos), ped o.cas ano@kaus .
edu.sa (P. Cas año).
Jou nal o Ca alysis 408 (2022) 115–127
Con en s lis s a ailable a ScienceDi ec
Jou nal o Ca alysis
jou nal homepage: www.else ie .com/loca e/jca
buildup o ca ionic ole ins and a oma ics, cons i u ing a hyd oca -
bon pool ha ac s as a coca alys wi h he acid si es. The hyd oca -
bon pool species unde go me hyla ion, oligome iza ion, alkyla ion,
c acking, dealkyla ion, cycliza ion, and hyd ogen ans e eac ions
esul ing in mo e ole ins and a oma ics. The deac i a ion pe iod
s a s wi h a p o ound deg ada ion o hyd oca bon pool species,
pa icula ly a oma ics, esul ing in coke deposi s ha block he di -
usion o eac an s and p oduc s. Mos o his knowledge acquisi-
ion has been possible wi h he analysis o e ained species in
he ca alys , ollowing he p oposal o a ‘‘ca bon pool” ha se es
as an in e media e in o ming gaseous p oduc s and coke [19].
F om an expe imen al iewpoin , he in es iga ion o e ained
species comp ises ex si u and in si u analyses using a ious ech-
niques [12]. The ex si u analysis consis s o conduc ing he eac ion
in a con en ional eac o and eco e ing he spen ca alys o ana-
lyze he e ained species using he mog a ime ic analysis ( ypi-
cally applying empe a u e-p og ammed oxida ion), soluble
species ex ac ion, mic oscopies, spec oscopies, and o he su ace
cha ac e iza ion echniques. The mog a ime ic analysis is
ex ended o quan i y he amoun o e ained species o coke
[17,20], o en p o iding in o ma ion on hei na u e o loca ion,
suppo ed by o he echniques [21–23]. Likewise, he ex ac ion
o soluble species (disco e ed in he Guisne g oup [24]) is o en
a ou ine analysis echnique in he s udy o he chemis y o
e ained species in he MTH eac ion [17,25]. Al hough he ex si u
app oach p o ides p ecise kine ic da a based on he p oduc anal-
ysis in he e luen s eam, he e ained species is analyzed a di -
e en condi ions han he eac ion condi ions. This implies ha
he na u e o he e ained species may ha e been modi ied; he e-
o e, he esul s o he ex si u analysis o e ained species may be
cau iously in e p e ed.
The in si u analysis consis s o ca ying ou he eac ion in a es-
sel ha allows he simul aneous cha ac e iza ion o he ca alys
su ace using no mal he mog a ime ic analyses, mic oscopies,
o spec oscopies. The mos common echniques o he in si u
analysis o e ained species include se e al in si u spec oscopic
me hodologies based on Fou ie - ans o m in a ed (FTIR), ul a i-
ole – isible (UV– is), and
13
C solid-s a e nuclea magne ic eso-
nance (NMR) spec oscopies [26–32]. These echniques allow
de e mining he chemical na u e o he hyd oca bon e ained spe-
cies in he ca alys by he spec oscopic signa u e. In ecen yea s,
he Weckhuysen g oup has published se e al s udies co e ing he
MTH eac ion wi h di e en ca alys opologies o p ope ies [33–
38] o unde di e en eac ion condi ions [14,39,40], ocusing hei
a en ion on he analysis o e ained species. These expe imen al
app oaches sol e he p oblem o analyzing e ained species in con-
di ions di e en om hose o he eac ion, bu ob aining app op i-
a e kine ic da a is a challenge. In ecen yea s, his in ol es he
ex ended use o comme cial spec oscopic cells, acili a ing he
labo o designing (homemade) spec oscopic cells [27,41] whose
pe o mance may di e om ha o a model eac o .
Despi e he issue ela ed o he pe o mance o spec oscopic
cells as model eac o s, in si u spec oscopic me hodologies p o-
ide aluable spec o-kine ic analyses because spec oscopic da a
can be ob ained h ough ime on s eam o ime- esol ed spec-
oscopy [41,42]. The epo ed spec o-kine ic analyses o he
MTH eac ion usually consis o analyzing he e olu ion wi h ime
on s eam o spec oscopic bands ela ed o e ained species, o en
hyd oca bon pool o coke species. Mo es e al. [34] analyzed he
ime on s eam e olu ion o he 415 nm band in he UV– is spec-
um (a ibu able o ions o polyme hylbezenes) a di e en eac-
ion empe a u es. Upon i ing o a i s -o de kine ic model, hey
ob ained A henius plo s ha in e p e he e ec o empe a u e
on he deg ada ion o hese species in o coke in ZSM-5 ca alys s.
Goe ze e al. [36] analyzed he ime on s eam e olu ion o he
1000 nm band in he UV– is spec um (a ibu able o coke) a
h ee posi ions along he bed o wo ZSM-5 ca alys s and co e-
la ed his wi h he me hanol con e sion, inding ha seconda y
coke ( o med om ole ins) g ows mo e slowly in a less acidic
ZSM-5 ca alys . Simila ly, Bo odina e al. [39,40] analyzed he ime
on s eam e olu ion o a ious UV– is bands ela ed o he
e ained species in SAPO-34 and SSZ-13 a a iable empe a u es.
They co ela ed hese analyses wi h he deac i a ion kine ics o
de e mine he na u e o he ac i e and deac i a ing species.
In ou p e ious wo ks [11,12,15], we used in si u FTIR and UV–
is spec oscopies o assess he o ma ion kine ics o e ained spe-
cies in he MTH eac ion on a ious ca alys s. The app oach o he
spec o-kine ic analysis consis ed o analyzing he ime on s eam
e olu ion o he mos ep esen a i e bands in he FTIR and UV– is
spec a wi h he con e sion o oxygena es. The e olu ion o he
1570 and 1616 cm
1
bands in he FTIR spec um du ing he MTH
eac ion on he ZSM-5 ca alys p o ides kine ic da a ela ed o
he o ma ion o ac i e species, whe eas ha o he 1481 cm
1
band p o ides kine ic da a ela ed o he o ma ion o deac i a ing
species. Likewise, he ime on s eam e olu ion o he 400 nm (ac-
i es species) and 500 o 800 nm (coke) bands in he UV– is spec-
um p o ides spec o-kine ic da a on o ming se e al e ained
species. The esul s we e consis en wi h he ac ha he o ma-
ion o e ained species is slowe on less acidic ZSM-5 ca alys s,
making he o ma ion a es dependen on he concen a ion o
B øns ed acid si es. Howe e , he mos ele an a iable o he
kine ic s udies o low eac o s is he space ime (wi h mul iple
de ini ions in he ield o he e ogeneous ca alysis) [43], and he e
is a lack o s udies ela ed o spec o-kine ic analyses applied o
he MTH eac ion in he li e a u e in which he space ime is he
main s udy a iable.
This wo k aims o de elop a obus spec o-kine ic me hodol-
ogy analyzing he e ec o ime and space ime in he well-
known MTH eac ion on a ZSM-5 ca alys using comme cial spec-
oscopic cells. The use o comme cial cells acili a es he expe i-
men al wo k, p o iding apid ca aly ic es s ob aining as much
expe imen al da a (gas phase p oduc and su ace species analy-
ses) as possible. To his end, we used wo comme cial spec o-
scopic cells (designed o in si u FTIR and UV– is spec ome ies)
wi h online p oduc analysis (mass spec ome y). The ime- and
space ime- esol ed mass-spec ome y esul s we e co ec ed o
ep esen he en i e gas-phase eac ion media, calib a ing he
esul s wi h hose o gas ch oma og aphy. The spec oscopic cell
o FTIR spec oscopy is used in he ansmission mode, whe eas
ha o UV– is spec oscopy is used in he di use e lec ance
mode. The esul s ob ained using di e en space imes, o he eac-
ion media composi ion (mass spec ome y) a e co ela ed o he
chemis y o he su ace species (FTIR and UV– is spec ome y
and e ained species ex ac ion) in e ms o hei ime e olu ion.
These co ela ions ede elop and e i y he MTH eac ion mecha-
nisms on he ZSM-5 zeoli e ca alys , and hey may be applied o
o he eac ions and ca alys s, uning se e al pa ame e s, including
pa ial p essu e (as demons a ed in his wo k).
2. Expe imen
2.1. Ca alys p epa a ion and cha ac e iza ion
We p epa ed he ca alys by mixing 50 w % o ZSM-5 zeoli e
wi h 30 w % o pseudo-boehmi e and 20 w % o
a
-alumina, ol-
lowed by d ying a oom empe a u e o 24 h and 110 °C o
24 h, c ushing and sie ing a 0.125 o 0.300 mm, and calcining a
550 °C o 3 h. The ZSM-5 zeoli e is a comme cial NH
4
ZSM-5 zeoli e
(Zeolys In e na ional, CBV8014, SiO
2
/Al
2
O
3
mola a io = 80),
which is calcined a 575 °C, as desc ibed in a p e ious wo k [15]
o ob ain he acid (p o onic) o m: HZSM-5. We cha ac e ized he
José Valecillos, H. Vicen e, A.G. Gayubo e al. Jou nal o Ca alysis 408 (2022) 115–127
116
ca alys using con en ional echniques [15], including x- ay pho o-
elec on spec oscopy, x- ay di ac ion, N
2
physiso p ion, NH
3
adso p ion and empe a u e-p og ammed deso p ion (NH
3
-TPD),
FTIR spec oscopy, and py idine adso p ion moni o ed wi h FTIR
spec oscopy.
Table 1 summa izes he main ca alys p ope ies. The B unaue ,
Emme , and Telle (BET) speci ic su ace a ea (S
BET
) is wi hin he
expec ed alue o a ZSM-5 zeoli e (400 m
2
g
1
), whe eas i
dec eases o he ca alys due o he p esence o alumina phases
wi h a less po ous speci ic su ace a ea. Likewise, he zeoli e has
a mo e mic opo ous speci ic su ace a ea (S
mic o
) han he ca alys
due o he absence o mic opo es in he alumina phases. The o al
acidi y de e mined wi h NH
3
adso p ion indica es ha he zeoli e
is mo e acidic han he ca alys because o he dilu ion o he zeo-
li e wi h alumina phases ha ha e a low concen a ion o acid
si es. Addi ionally, he s eng h o he acid si es de e mined wi h
NH
3
-TPD indica es ha he concen a ions o s ong acid si es
(C
SAS
) and weak acid si es (C
WAS
) dec ease o he ca alys compa ed
wi h he zeoli e, which is ela ed o he dilu ion e ec . Howe e ,
he concen a ion o medium-s eng h acid si es (C
MAS
) is sligh ly
highe o he ca alys , indica ing ha he alumina phases (pa ic-
ula ly hose o pseudo-boehmi e) p o ide his acidi y ea u e. The
concen a ion o B øns ed acid si es (C
BAS
) and Lewis acid si es
(C
LAS
) is compa able be ween he zeoli e and ca alys , conside ing
ha he alues a e lowe o he ca alys because o he dilu ion
o he zeoli e wi h alumina phases.
2.2. Expe imen s in spec oscopic cells
We conduc ed he MTH eac ion in spec oscopic cells a con-
s an condi ions and a iable space imes o moni o he changes
on he ca alys su ace using FTIR o UV– is spec oscopy. The con-
di ions o he MTH eac ion we e empe a u e (T) = 400 °C, o al
p essu e (P) = 1 ba , me hanol pa ial p essu e (P
M0
) = 0.04 o
0.16 ba , me hanol low a e (F
M0
) = 0.94 o 60 mmol h
1
, ca alys
weigh (W) = 0.012 o 0.048 g, and space ime (W/F
M0
) = 0.2, 0.4,
0.8, 1.6, 3.2, 6.4, o 12.8 g h mol
1
. We used wo comme cial spec-
oscopic cells: (i) a Specac high-p essu e, high- empe a u e cham-
be coupled wi h an FTIR spec ome e (The mo Scien i ic, Nicole
6700), and (ii) a Linkam s age (THMS600) coupled wi h a UV– is
spec ome e (Jasco, V-780) wi h a specially adap ed compa men
(Jasco, ARN-915i) o he cell. Fig. 1 p esen s a schema ic ep esen-
a ion o he expe imen al se up o using comme cial spec o-
scopic cells as eac ion sys ems. The FTIR spec oscopic cell
wo ks in he ansmission mode, equi ing hin samples, whe eas
he UV– is spec oscopic cell wo ks in he di use e lec ance mode
using an in eg a ing sphe e placed on he cell window. The eed
consis ed o N
2
wi h me hanol apo ob ained by lowing N
2
h ough a sa u a o essel con aining liquid me hanol a oom
empe a u e o imme sed in an ice ba h. We calcula ed he me ha-
nol concen a ion using he he modynamic equilib ium da a a
oom empe a u e o 1 °C ( he measu ed empe a u e in he liquid
me hanol when he sa u a o is imme sed in an ice ba h).
To analyze he gaseous e luen , we used a mass spec ome e
(MS) (P ei e Vacuum, OmniS a GSD 320O Se ies) con inuously
measu ing he m/zsignals o 16, 18, 27, 29, 31, 41, 43, 45, 55, 56,
57, 78, and 91. Addi ionally, we analyzed he gaseous e luen o
wo expe imen s by sampling e e y 20 min and measu ing he
gas composi ion in a mic o-gas ch oma og aph (GC) (Va ian,
CP4900) desc ibed in p e ious wo k [15] because he GC analysis
p o ides a be e iden i ica ion and quan i ica ion o componen s.
We co ela ed bo h analyses o he gaseous e luen and used he
NIST da abase [44] o ob ain he s anda d mass spec um o he
expec ed MTH eac ion p oduc s, inding he ollowing:
The maximum abundance o ligh ole ins (p opene and bu enes)
is a m/z= 41. The analysis o e hene was excluded because i
has a maximum abundance a m/z= 28, coinciding wi h ni o-
gen used as an ine gas.
The maximum abundance o o he alipha ic compounds wi h
ou o mo e ca bon a oms is a m/z= 43 o 57 o pa a ins
and 55 o ole ins. The analysis o e hane and p opane was
excluded because hey ha e a maximum abundance a m/
z= 28 o 29, coinciding wi h ni ogen used as an ine gas.
The maximum abundance o a oma ic compounds is a m/z=78
o 91.
The maximum abundance o me hanol is a m/z= 31 and o
dime hyl e he is a m/z= 45, being bo h eac i e oxygena es
easily dis inguishable om hyd oca bons.
Al hough he MS analysis does no p o ide an accu a e p oduc
analysis, we app oached some p oduc g oups by g ouping m/zsig-
nals: oxygena es (O) summing he in ensi y o m/z= 31 and 45
(I
O
=I
31
+2I
45
), ligh ole ins (LO) aking he in ensi y o m/z=41
(I
LO
=3I
41
), hea y alipha ics (HA) summing he in ensi y o m/
z= 43, 55, and 57 (I
HA
=4I
43
+4I
55
+4I
57
), and a oma ics (BTX) sum-
ming he in ensi y o m/z= 78 and 91 (I
BTX
=6I
78
+7I
91
). Then, con-
side ing ha he eac ed moles o oxygena es a e equal o he
o med moles o p oduc s in he gaseous e luen , we calcula ed
he con e sion o oxygena es based on he MS analysis (X
MS
)as
ollows:
X
MS
¼I
LO
þI
HA
þI
BTX
I
O
þI
LO
þI
HA
þI
BTX
ð1Þ
and he yield o a p oduc g oup ibased on he MS analysis (Y
i
MS
)
is as ollows:
Y
MS
i
¼I
i
I
O
þI
LO
þI
HA
þI
BTX
ð2Þ
These amoun s (X
MS
and Y
i
MS
) we e compa ed wi h hose
ob ained using he GC da a (Xand Y
i
) o he wo expe imen s,
and using O igin 8.5, we ound ha he bes co ela ion is as
ollows:
X¼
aX
MS
b
c
b
þX
MS
b
ð3Þ
Y
i
¼
a
i
Y
MS
i
b
i
c
i
ðÞ
b
i
þY
MS
i
b
i
ð4Þ
whe e a,b,c,a
i
,b
i
, and c
i
a e he co ela ion pa ame e s calcu-
la ed o each p oduc g oup (Table S1 in he Suppo ing In o ma-
ion p o ides hese calcula ed pa ame e s). Thus, hese
co ela ions allow es ima ing he con e sion (X) and yield o p o-
duc g oups (Y
i
) om he MS da a.
A ypical expe imen consis ed o p epa ing a ca alys sample
by p essing ca alys powde in o a hin disk by applying 10 o
Table 1
Main ca alys p ope ies.
P ope y Zeoli e Ca alys
S
BET
(m
2
g
1
) 434 290
S
mic o
(m
2
g
1
) 366 130
To al acidi y (mmol g
1
) 0.33 0.21
C
WAS
(mmol g
1
) 0.023 0.012
C
MAS
(mmol g
1
) 0.061 0.074
C
SAS
(mmol g
1
) 0.26 0.13
C
BAS
(mmol g
1
) 0.32 0.13
C
LAS
(mmol g
1
) 0.060 0.027
José Valecillos, H. Vicen e, A.G. Gayubo e al. Jou nal o Ca alysis 408 (2022) 115–127
117
p essu e in a Specac manual hyd aulic p ess. The p epa ed ca alys
sample was placed in he spec oscopic cell and subjec ed o he -
mal ea men a 550 °CinN
2
low o 1 h o emo e con aminan s.
A e wa d, he ca alys was cooled o he eac ion empe a u e,
and a e e ence spec um was collec ed (FTIR o UV– is spec um).
The eac ion es s a ed by con inuously eeding me hanol while
collec ing di e en ial FTIR o UV– is spec a (by sub ac ing he
co esponding e e ence spec um). Al e na i ely, we also con-
duc ed expe imen s in he FTIR spec oscopic cell by discon inu-
ously eeding me hanol wi h al e na ed lushing pe iods (pulses
o me hanol eed). In his case, he me hanol was inpu in o he cell
o a de e mined ime, and hen he eed was swi ched o an N
2
low o sweep weakly adso bed o gaseous species om he ca a-
lys su ace and collec a spec um. We collec ed FTIR spec a e e y
2 min in he con inuous mode, wi h a measu emen ange o 1300
o 4000 cm
1
, a esolu ion o 4 cm
1
, and 100 scans. We collec ed
he UV– is spec a e e y 35 s in he con inuous mode, wi h a mea-
su emen ange o 250 o 850 nm a a scan speed o 4000 nm min
1
,
a UV– is esponse o 0.24 s, and a da a in e al o 2 nm.
Addi ionally, we analyzed he e ained species a e each expe -
imen using ex ac ion wi h dichlo ome hane ( he p ocedu e is
desc ibed in p e ious publica ions [11,12,15]). B ie ly, he p oce-
du e consis s o (1) dissol ing he spen ca alys disk wi h HF
(Me ck, 40%) in a Te lon con aine o 1 h, (2) neu alizing wi h
NaOH, (3) ex ac ing species wi h 3 cm
3
o dichlo ome hane
(Sigma-Ald ich, 99.9%), and (4) allowing he o ganic and aqueous
phases o sepa a e. The o ganic phase was analyzed using a GC
wi h an MS (Shimadzu, GCMS-QP2010S), and he componen s
we e iden i ied and semi-quan i ied (peak in eg a ion).
3. Resul s
3.1. Kine ic pe o mance
The kine ic pe o mance o he cells was assessed using a gas-
eous e luen analysis wi h an MS e i ied wi h a GC. Fo his, we
calcula ed he con e sion and yield o p oduc g oups using he
MS da a as desc ibed in he expe imen sec ion (Sec ion 2).
Figs. 2a-b illus a e he ime on s eam e olu ion o he con e sion
a a iable space imes o he MTH eac ion in he spec oscopic
cells. In gene al, he con e sion le els inc ease wi h he inc ease
in he space ime o bo h eac ion sys ems, which is expec ed
o he MTH eac ion [15,45–48]. Addi ionally, he expe imen al
da a e eal ha he con e sion p og essi ely dec eases wi h
inc easing imes on s eam, indica ing ca alys deac i a ion. Pa ic-
ula ly, expe imen s in he FTIR spec oscopic cell demons a e
comple e ca alys deac i a ion a low space imes (0.2 and
0.4 g h mol
1
), whe eas hose in he UV– is cell do no exhibi such
le els o ca alys deac i a ion in 180 min on s eam. Thus, using
high space imes imp o es he ca alys s abili y in ag eemen wi h
Fig. 1. Schema ic ep esen a ion o he expe imen al se up o he eac ion sys em using comme cial spec oscopic cells.
José Valecillos, H. Vicen e, A.G. Gayubo e al. Jou nal o Ca alysis 408 (2022) 115–127
118
p o iding slowe eloci ies o he p og essi e e olu ion o ac i e
and deac i a ing species h ough he ca alys bed in he MTH eac-
ion. The eason o he slowe deac i a ion in he UV– is cell is
ha his cell seems o exhibi a beha io in which he me hanol
concen a ion on he ca alys su ace ises mo e slowly han in
he FTIR cell. Thus, he beha io o he o me sys em is close o
a mixed low eac o , whe eas ha o he la e is close o a plug
low eac o .
Figs. 2c-d p esen he e olu ion wi h he con e sion o he yield
o he main p oduc g oups a a iable space imes o he MTH
eac ion in he spec oscopic cells. The iden i ica ion o p oduc s
is app oxima e, as explained in he expe imen sec ion (Sec ion 2),
because he MS can iden i y agmen s o compounds and hese
agmen s a e common o se e al compounds ha a e ypical
MTH eac ion p oduc s. Howe e , he ends in Figs. 2c-d ag ee
wi h hose ypically expec ed o he MTH eac ion [15]. The yield
o ligh ole ins inc eases as e wi h inc easing con e sions (i.e., by
changing space ime) desc ibing a con ex cu e, whe eas he yield
o hea y alipha ics and a oma ics slowly inc eases wi h inc easing
con e sions (by changing space ime) desc ibing a conca e cu e.
This beha io desc ibes he ole o p oduc s in he eac ion, poin -
ing ou ha ligh ole ins a e eac ion in e media es and hea y
alipha ics and a oma ics a e inal p oduc s, in ag eemen wi h
he global mechanism o he MTH eac ion [18]. Thus, ligh ole ins
a e p ima ily o med a a high a e (low con e sions). When he e
is a high concen a ion o ligh ole ins in he eac ion medium, hey
unde go oligome iza ion, me hyla ion, cycliza ion, and hyd ogen
ans e eac ions yielding alipha ics and a oma ics a inc easing
a es wi h inc easing con e sions.
These esul s e idence a easonable kine ic beha io o he
spec oscopic cells o he MTH eac ion, b inging he oppo uni y
o explo e mo e aspec s o his eac ion using hese eac ion
sys ems. Thus, he ollowing sec ions ocus on s udying he
in si u o ma ion o species on he ca alys su ace (also e e ed
as su ace species) using FTIR and UV– is spec oscopies. The con-
di ions used in hese expe imen s esul in he ca alys unde going
deac i a ion a 0.2 and 0.4 g h mol
1
in he FTIR spec oscopic cell,
which is an in e es ing expe imen al scena io o s udy he species
in ol ed in ca alys deac i a ion.
3.2. In si u FTIR spec oscopy
Fig. 3 depic s he ime on s eam e olu ion o he di e en ial
FTIR spec a ( he FTIR spec um o he esh ca alys is sub ac ed)
o he ca alys du ing he expe imen s a he space imes o 0.2 and
1.6 g h mol
1
ca ied ou in he FTIR spec oscopic cell wi h a con-
inuous me hanol eed. Figu e S1 in he Suppo ing In o ma ion
p esen s he FTIR spec a e olu ion o o he space imes. Exposing
he ca alys o me hanol esul s in se e al FTIR bands ela ed o
hyd oca bons in he 1300 o 3200 cm
1
egion, which a e ypical
‘‘signa u es” o he MTH eac ion [12,15,49]. Al hough he compo-
si ion o he ca alys is 50 w % o zeoli e ( he es co esponds o
alumina phases), he spec oscopic signa u e is simila o ha
ob ained o expe imen s using pu e zeoli e ca alys s, which
implies ha he alumina phases do no a ec he na u e o he spe-
cies o med on he agglome a ed ca alys . Likewise, we es ed a
blank composed o 30 w % o boehmi e and 70 w % o
a
-alumina
(zeoli e is eplaced wi h
a
-alumina) in a ixed-bed eac o a he
same eac ion condi ions used in his wo k, and we obse ed ha
only he me hanol dehyd a ion ook place yielding dime hyl e he
and wa e wi h no e idence o hyd oca bons o ma ion. The bands
in he 1300 o 1460 cm
1
egion may be associa ed wi h he bend-
ing ib a ion o C-H bonds, hose in he 1400 o 1700 cm
1
egion
wi h he s e ching ib a ion o C = C bonds in he ing o a oma ics
Fig. 2. Time on s eam e olu ion o he con e sion in a (a) FTIR spec oscopic cell and (b) UV– is spec oscopic cell, and e olu ion wi h con e sion o he p oduc yield in (c)
an FTIR spec oscopic cell and (d) UV– is spec oscopic cell. Reac ion condi ions: T= 400 °C, P
M0
= 0.16 ba , W/F
M0
= 0.2–1.6 g h mol
1
.
José Valecillos, H. Vicen e, A.G. Gayubo e al. Jou nal o Ca alysis 408 (2022) 115–127
119
o in ole ins, and hose in he 2800 o 3100 cm
1
egion wi h he
s e ching ib a ion o C-H bonds [28,50–54]. Fu he mo e, he
3200–4000 cm
1
egion bands p o ide in o ma ion on he O-H
bonds in a ious species, such as adso bed me hanol and wa e
(posi i e bands a abou 3530 cm
1
), and he zeoli e acid si es
exhibi nega i e bands in he 3550–3800 cm
1
egion.
We pe o med a mul i a ia e analysis o he FTIR band assigna-
ion o he bands associa ed wi h he ex ac ed species in a p e i-
ous wo k [12]. Based on he s abili y and beha io o he bands, we
classi ied he mos ele an bands o he spec a. The bands a 2950
and 2844 cm
1
a e assigned o –CH
3
and –CH
2
g oups, espec i ely,
in oxygena ed o non-oxygena ed alipha ic and a oma ic species
ha may be on he ca alys su ace o gas phase. Howe e , as seen
in Fig. 3, o he se e al bands also appea in he 2800–3100 cm
1
egion, e idencing he complexi y o he eac ion medium and
making di icul o iden i y speci ic species (su ace and gas phase
species). The band a 1616 cm
1
is assigned o monocyclic a o-
ma ic and ole inic species, p obably in a ca ionic o m being he
ac i e hyd oca bon pool species [54,55]. The band a 1570 cm
1
is assigned o monocyclic and polycyclic a oma ic species, and
he band a 1481 cm
1
is assigned o linea polycyclic a oma ic
species. We e i ied he p esence o monocyclic a oma ics (xyle-
nes, ime hylbenzenes ( iMBs), e ame hylbenzenes ( e aMBs),
pen ame hylbenzenes (pen aMBs), and hexame hylbenzenes (hex-
aMBs)) by ex ac ing soluble species in he spen ca alys samples
(Fig. 4), which a e ini ially ela ed o he 1616 and 1570 cm
1
bands (Table S2). The abundance o soluble monocyclic a oma ic
species inc eases wi h inc easing space imes, and he dis ibu ion
o componen s changes om being cen e ed a pen aMB/hexaMB
a 0.2 g h mol
1
o being cen e ed a e aMB a 1.6 g h mol
1
.
These obse a ions a e based on he ac i i y s a e o he ca alys ,
which is se e ely deac i a ed a low space imes and is ac i e a
high space imes (Fig. 2a). Thus, an ac i e ca alys has a high con-
cen a ion o monocyclic a oma ic species wi h a composi ion cen-
e ed a he mos ac i e species (e.g., e aMB [17]). In con as , a
deac i a ed ca alys has a low concen a ion o monocyclic a o-
ma ic species wi h a composi ion cen e ed a he less ac i e spe-
cies (e.g., hexaMB, as an inac i e species in ZSM-5 ca alys s [17]).
Fig. 5 illus a es he ime on s eam e olu ion o he maximum
in ensi y o he 2950, 1616, 1570, and 1481 cm
1
bands a a iable
space imes. The in ensi y o he 2950 and 1570 cm
1
was di ec ly
aken om each spec um, whe eas ha o he 1616 and
1481 cm
1
bands was co ec ed wi h a baseline as shown in Fig-
u es S2 and S3. The eason o his co ec ion is ha hese bands
appea on he ail o he 1570 and 1456 cm
1
bands, like a shoul-
de , and he e o e hei in ensi y is masked. Addi ionally, Figu e S2
also e idences ha he 1481 cm
1
band is p one o appea when
he ca alys loses ac i i y whose condi ion is a o ed a dec easing
space imes and p olonged imes on s eam, making e iden his
band is he signa u e o he o ma ion o deac i a ing species.
Thus, he spec o-kine ics indica e ha he 2950 cm
1
band apidly
inc eases a he beginning o he eac ion and eaches a maximum
alue ha emains s able du ing he en i e eac ion. The maximum
alue depends on he space ime, which is highe as he space ime
dec eases. Because he FTIR in ensi y can be di ec ly ela ed o he
species concen a ion on he sample h ough Lambe –Bee ’s law,
his obse a ion also indica es ha he concen a ion o species
whose bonds ib a e a 2950 cm
1
is highe as he space ime
dec eases. In ac , i is expec ed ha me hoxy species show ib a-
ion ea u es a equencies in he ange o 2800–3000 cm
1
,as
epo ed by Saepu ahman e al. [50] who s udied he FTIR ea u es
o me hanol adso bed on zeoli es. Likewise, hyd oca bons also
show impo an ib a ion ea u es a hese equencies and i is
di icul o make a p ecise di e ence be ween me hoxy and hyd o-
ca bon species in he eac ion, and i is e en mo e complex when
wo king wi h a con inuous me hanol eed because he gas phase
species seem o show s ong ib a ion ea u es a hese equencies
[12]. Albei hese issues, he 2950 cm
1
band may be also associ-
a ed wi h eac i e species (me hoxides, alkoxides, ole ins) because
he species concen a ion is heo e ically expec ed o dec ease
wi h inc easing space imes, leading o high con e sions ( e i ied
by analyzing he gaseous e luen in Fig. 2a).
The 1616 cm
1
band in ensi y also ises e y apidly a he
beginning o he eac ion (Fig. 3c), eaching a maximum e idenced
in he expe imen s a low space imes. This beha io indica es ha
hese species a e in e media es because hei concen a ion
eaches a maximum and p og essi ely dec eases wi h ca alys
deac i a ion (Fig. 2a). These species a e pa o he hyd oca bon
Fig. 3. Time on s eam e olu ion o he di e en ial FTIR spec a o he HZSM-5
ca alys in he MTH eac ion wi h con inuous me hanol eed a space imes o (a)
0.20 and (b) 1.6 g h mol
1
. Condi ions: T= 400 °C, P
M0
= 0.16 ba .
Fig. 4. Dis ibu ion o he componen s o soluble species in he spen ca alys
samples in he expe imen s in he FTIR spec oscopic cell a e 180 min on s eam.
José Valecillos, H. Vicen e, A.G. Gayubo e al. Jou nal o Ca alysis 408 (2022) 115–127
120
pool in he MTH eac ion, p ima ily comp ising monocyclic a o-
ma ics p obably in ca ionic o ms [54,55]. This is u he suppo ed
by he disappea ance o soluble species a p olonged imes on
s eam and low space imes, coinciding wi h he beha io o his
band. P e ious wo ks [12,15] demons a ed ha monocyclic a o-
ma ic species a e con e ed in o coke (mo e condensed a oma ic
s uc u es) as he ca alys unde goes deac i a ion. On his ca alys ,
he p esence o abso bed ole ins as in e media e species seems o
be negligible compa ed o o he small-po e ca alys s [12,56]. How-
e e , examining he spec a, bands a 1590, 1390, and 1377 cm
1
a e he i s o ise a a sho ime on s eam, indica ing he o ma-
ion o adso bed ole ins [57], bu hei in ensi y is apidly a enu-
a ed, esul ing in bands a 1616 and 1570 cm
1
. Thus, he
majo i y o he hyd oca bon pool species would be a oma ic spe-
cies du ing he eac ion.
The 1570 cm
1
band in ensi y also e ol es wi h he ime on
s eam, eaching a maximum bu main aining a high alue a he
end o he expe imen s (Fig. 3b). The pa ial a enua ion o his
band in ensi y may indica e he con ibu ion o in e media e
monocyclic a oma ic species (also showing ib a ion ea u es a
1616 cm
1
) ha disappea upon ca alys deac i a ion, whe eas
he es o he species may be associa ed wi h polycyclic a oma ics.
I should be men ioned ha he in ensi y o he 1455 and
1510 cm
1
bands (and ac ually he in ensi y o he 1400–
1600 cm
1
egion) also dec eases a p olonged ime on s eam
and, pa icula ly, a low space ime (when he ca alys loses ac i -
i y), which can be clea ly seen in Figu es S2 and S3 in he Suppo -
ing In o ma ion. This con i ms ha ac i e species has ib a ional
ea u es in his whole ange o he FTIR spec um, o e lapping wi h
he ib a ional ea u es o o he s able species (e.g. polycyclic a o-
ma ics). This is also e iden when analyzing he expec ed ib a-
ional ea u es o di e en monocyclic a oma ic species ound
(Table S2), exhibi ing equencies in he 1400–1700 cm
1
ange
in an unp edic able way. Thus, he 1616 cm
1
is mo e ep esen a-
i e o hese ac i e species because i is isola ed ( he e a e no o he
o e lapping bands) and i s appea ing and disappea ing beha io is
clea ly seen.
Based on Raman spec oscopy (complemen a y o FTIR spec-
oscopy), se e al ca bon s uc u es ypically exhibi bands
be ween 1300 and 1700 cm
1
[56,58–62]. Fo ins ance, an appea -
ing and disappea ing band a 1602–1606 cm
1
has been ela ed o
ca ionic species (e.g. benzenium species) [61], which can be analo-
gous o he 1616 cm
1
band obse ed in his wo k. Likewise, a
band a 1510 cm
1
has been ela ed o cyclopen enyl ca ions in
he FTIR spec um [55]. These indings epo ed in he li e a u e
s eng hens he band assigna ion in his wo k, and makes e iden
ha he appea ing and disappea ing beha io (ba ely epo ed in
he li e a u e) o he 1616 cm
1
band and some bands in he
1400–1600 cm
1
egion co esponds o ib a ional ea u es o
ac i e hyd oca bon pool species (ca ionic o ms).
The 1481 cm
1
band in ensi y ises when he ca alys unde -
goes se e e deac i a ion (p olonged ime on s eam and low space
imes) and when he in e media e species dec ease (Fig. 3a), which
can be clea ly seen in Figu es S2 and S3 in he Suppo ing In o ma-
ion. Thus, he species wi h bonds ib a ing a 1481 cm
1
a e asso-
cia ed wi h deac i a ing species. Acco ding o Lee e al. [63], hese
deac i a ing species may be associa ed wi h linea monocyclic a o-
ma ics g owing in he zeoli e channels. Addi ionally, Rojo-Gama
e al. [56] ha e assigned a Raman band a 1485 cm
1
o luo ene,
suppo ing ha linea polycyclic a oma ics can show ib a ional
ea u es a his equencies.
When analyzing he band e olu ions o e he space ime (di -
e en ac i i y s a es o he ca alys ), we obse ed ha all e ained
species e ol e as e a low space imes. This obse a ion also
Fig. 5. Time on s eam e olu ion o he in ensi y o FTIR bands o (a) 1481, (b) 1570, (c) 1616, and (d) 2950 cm
1
in he MTH eac ion wi h a con inuous me hanol eed a
a iable space imes. Condi ions: T= 400 °C, P
M0
= 0.16 ba , W/F
M0
= 0.20–1.6 g h mol
1
.
José Valecillos, H. Vicen e, A.G. Gayubo e al. Jou nal o Ca alysis 408 (2022) 115–127
121
applies o o ming alipha ic species (2950 cm
1
) and hea ie poly-
cyclic a oma ic species (1481 and 1570 cm
1
). We also obse e a
mo e apid con e sion o ac i e in o deac i a ing species
(1616 cm
1
) wi h dec easing space imes, which is also e idenced
when analyzing he e olu ion o soluble species o e space ime:
he mos ac i e monocyclic a oma ic species o he MTH eac ion
disappea as e a low space imes acco ding o he ca alys ac i -
i y s a e.
In his way, he esul s imply ha alipha ic and me hoxy/alkoxy
species (2950 cm
1
) and he mos ac i e hyd oca bon pool species
(ole ins and monocyclic a oma ic hyd oca bons, 1616 cm
1
) a e
gene a ed e y quickly o e he su ace. The la e species g oup
(ac i e) deg ades p og essi ely o polycyclic a oma ic species, as
obse ed in he e olu ion o he 1616, 1570, and 1481 cm
1
bands.
As he eac ion e ol es wi h ime on s eam, he in e media e
monocyclic a oma ic species in he mic opo es polyme ize in o
hea y polycyclic a oma ic species (1481 cm
1
), causing ca alys
deac i a ion by blocking access o acid si es on he mic opo ous
su ace.
3.3. In si u UV– is spec oscopy
Figu e S4 in he Suppo ing In o ma ion p esen s he ime on
s eam e olu ion o he di e en ial UV– is spec a ( he UV– is
spec um o he esh ca alys is sub ac ed) o expe imen s wi h
di e en space imes (0.2, 0.4, 0.8, and 1.6 g h mol
1
) conduc ed in
he UV– is spec oscopic cell wi h a con inuous me hanol eed. As
obse ed, se e al bands occu in he UV (250 o 330 nm) and is-
ible (330 o 850 nm) egions. The i s bands appea a ound 380 o
400 nm, and hen o he bands appea a abou 500, 604, and
704 nm. These bands ha e been ex ensi ely discussed in he li e -
a u e [11,15,33,34,36,37] and co espond o ca ions o monocyclic
a oma ics, monoenes, o polyenes (380 o 400 nm) and ca ionic o
neu al polycyclic a oma ics (500 o 800 nm). Figu e S5 p esen s
he ime on s eam e olu ion o he in ensi y o selec ed UV– is
bands. The spec o-kine ic da a e eal ha he e olu ion o hese
bands p o ides limi ed kine ic in o ma ion because he in ensi y
apidly eaches a s able alue h ough ime on s eam, which
occu s because his spec oscopic echnique is limi ed by he colo
o he ca alys sample. The as da kening o he ca alys sample
p o okes a apid sa u a ion o he spec a as e idenced by he
b oad backg ound abso p ion ac oss he whole isible egion. This
simila obse a ion was epo ed by Mo es e al. [34] o he
me hanol con e sion on a ZSM-5 ca alys a high empe a u es
(abo e 400 °C).
To add ess his p oblem, we a ied o he eac ion condi ions,
such as he me hanol pa ial p essu e and space ime o ob ain
mo e incipien eac ion condi ions. Dec easing hese a iables
slows he eac ion a es o all kine ic s eps occu ing in he MTH
eac ion wi hou signi ican ly changing he eac ion mechanisms,
so ha a slowe o ma ion o e ained species and, he e o e, an
a enua ion on he spec a sa u a ion is expec ed. Fig. 6 compa es
he kine ics o he MTH eac ion a he wo me hanol pa ial p es-
su es in he UV– is spec oscopic cell. The e olu ion wi h space
ime o he con e sion o oxygena es (Fig. 6a) is slowe as he
me hanol pa ial p essu e dec eases, equi ing highe space imes
o each high con e sions. Likewise, he ime on s eam e olu ion
o he con e sion (Fig. 6b) is highly s able a a low me hanol pa ial
p essu e, p o iding insigni ican e idence o ca alys deac i a ion
compa ed wi h expe imen s a a high me hanol pa ial p essu e
(Fig. 2b). The p oduc dis ibu ion a a low me hanol pa ial p es-
su e (Fig. 6c) is compa able wi h ha a a high me hanol pa ial
p essu e (Fig. 2d), indica ing ha his a iable has a poo e ec
on changing he eac ion mechanisms bu has a s ong in luence
on he eac ion a e.
Fig. 7 depic s he co esponding UV– is spec a o expe imen s
wi h a low me hanol pa ial p essu e and space imes o 0.8 and
6.4 g h mol
1
. Addi ionally, Figu e S6 in he Suppo ing In o ma-
ion p esen s he spec al e olu ion o o he space imes. The
spec al e olu ion indica es ha hese eac ion condi ions signi i-
can ly a enua e he spec al sa u a ion due o slowing he da ken-
ing o he ca alys sample, and his a enua ion is e en mo e
a o able a high space imes. The UV– is bands a e he same as
hose obse ed o he expe imen s a a high me hanol pa ial
p essu e; he e o e, he band assigna ion is analogous. We e i ied
he p esence o monocyclic a oma ic species (xylenes, iMB, e -
aMB, pen aMB, and hexaMB) by ex ac ing soluble species in he
spen ca alys samples (Fig. 8), which a e obse ed a 400 nm in
he spec oscopic signa u e. The concen a ion o hese species
inc eases wi h dec easing space imes, whe eas he dis ibu ion
changes om being cen e ed a hexaMB a low space imes o
Fig. 6. Kine ic pe o mance o he UV– is spec oscopic cell a a iable me hanol
pa ial p essu e: (a) maximum con e sion cu es a a me hanol pa ial p essu e o
0.050 and 0.16 ba , (b) ime on s eam e olu ion o he con e sion a 0.050 ba , (c)
e olu ion wi h con e sion o he p oduc yield a 0.050 ba . Condi ions: T= 400 °C,
P
M0
= 0.050 o 0.16 ba , W/F
M0
= 0.2 o 1.6 g h mol
1
.
José Valecillos, H. Vicen e, A.G. Gayubo e al. Jou nal o Ca alysis 408 (2022) 115–127
122
being cen e ed a pen aMB a high space imes. In hese expe i-
men s, he ca alys was ac i e a all es ed space imes (as shown
in Fig. 6b wi h no e idence o deac i a ion).
Fig. 9 e eals he ime on s eam e olu ion o he in ensi y o
he selec ed UV– is bands o hese eac ion condi ions. The
spec o-kine ic da a indica e ha hese eac ion condi ions signi -
ican ly slow he in ensi y e olu ion o he bands, con olling he
apid spec al sa u a ion. The da a e eal ha he monocyclic a o-
ma ic species (a ound 400 nm) a e he i s o g ow, ollowed by
o he polycyclic a oma ic species obse ed a high wa eleng hs
(604 and 704 nm).
In all expe imen s conduc ed in he UV– is spec oscopic cell,
we only obse ed he o ma ion dynamics o species wi hou e i-
dence o deg ada ion indica ed by he disappea ance o bands
compa ed o he obse a ions in he expe imen s in he FTIR spec-
oscopic cell. We in e ha we canno obse e he deg ada ion o
ac i e species in hese expe imen s because he ca alys did no
unde go deac i a ion (Figs. 2b and 6b). We epo ed simila obse -
a ions when compa ing di e en ca alys s using a simila expe -
imen al se up [11,15]. Howe e , Bo odina e al. [39,40] and
Goe ze e al. [64] epo ed he disappea ance o ac i e species
de ec ed using in si u UV– is spec oscopy in small-po e ca alys s
du ing deac i a ion. The di e ence be ween conduc ing expe i-
men s wi h medium-po e ca alys s, such as in his wo k, elies
on o ming la ge polycyclic a oma ic species a high empe a u es
ha lead o a apid da kening o he ca alys sample and he sa u-
a ion o he spec a. Al hough his e ec can be con olled by
ope a ing a less se e e eac ion condi ions, he deac i a ion pe -
iod is e en mo e delayed. Thus, he obse a ions in his wo k using
a medium-po e ca alys a e limi ed o he o ma ion o e ained
species due o eac ion condi ions ha do no allow he obse a-
ion o se e e ca alys deac i a ion.
When analyzing he band e olu ion o e space ime, we
obse e ha all e ained species e ol e as e a dec easing space
imes (simila o he obse a ions om he FTIR spec a da a).
Ano he peculia kine ic obse a ion is he sigmoidal p o ile o
he UV– is band e olu ions (Fig. 9), which may be associa ed wi h
he ypical kine ic pe iods o induc ion and au oca alysis o he
MTH eac ion, as epo ed in o he wo k using in si u spec oscopic
app oaches [14,35]. Fi s , he slow inc ease o he band in ensi y
indica es he induc ion pe iod, which is longe as he space ime
inc eases. Second, he au oca aly ic pe iod s a s when he band
in ensi y inc eases apidly, eaching a s able alue, which seems
o be slowe as space ime inc eases. The obse ed da a a e om
he induc ion and au oca aly ic pe iod du ing a ansien s a e go -
e ned by he inc ease in he eac i e species concen a ion on he
ca alys su ace om ze o o a maximum alue. Thus, he use o
low space imes inc eases he con ac eloci y (in e se o space
ime), esul ing in a apid inc ease in he eac i e species concen-
a ion and he apid o ma ion o e ained species on he ca alys
su ace. In con as , using high space imes dec eases he con ac
eloci y, causing a slowe inc ease in he eac i e species concen-
a ion and he o ma ion o e ained species on he ca alys
su ace.
The kine ic analysis o he e ained species on he ca alys su -
ace using FTIR and UV– is spec oscopies is use ul o ob ain mech-
anis ic in o ma ion when hese species pa icipa e in he eac ion
ne wo k, such as in he case o he MTH eac ion. Using he kine ic
in o ma ion om bo h echniques, we ske ched a eac ion mecha-
nism based on he iden i ied e ained species (Fig. 10). Thus, he
o ma ion o eac i e species om oxygena es (band a
2950 cm
1
, albei his band is no unambiguously assigned because
i may also include he con ibu ion o –CH
3
g oups in o he alipha-
ic and a oma ic species) on he ca alys su ace o ms ac i e spe-
cies o he hyd oca bon pool (1616 cm
1
o 400 nm) e ained in he
ca alys mic opo es. These species o m small ole ins, pa a ins and
a oma ics (m/z= 43, 55, 78...) ha di use in o he gas phase as
p oduc s and o m la ge polycyclic a oma ic species (1570 cm
1
o 500 o 700 nm) deposi ed on he ca alys mesopo ous su ace
wi h a poo deac i a ing e ec . E en ually, he ac i e species poly-
me ize in o linea polycyclic a oma ics (1481 cm
1
) in he ca alys
mic opo ous su ace, blocking he mic opo es and se e ely deple -
ing he ca aly ic ac i i y. As discussed, some o hese mechanisms
ha e been s udied in de ail in o he wo ks. We in end o cons uc
an o e all kine ic eac ion ne wo k and deac i a ion om he
expe imen s conduc ed wi h he spec o-kine ic app oach.
Fig. 7. Time on s eam e olu ion o he di e en ial UV– is spec a o he HZSM-5
ca alys in he MTH eac ion wi h con inuous me hanol eed a space imes o (a)
0.80 and (b) 6.4 g h mol
1
. Condi ions: T= 400 °C, P
M0
= 0.16 ba .
Fig. 8. Dis ibu ion o he componen s o soluble species ound in he spen ca alys
samples in he expe imen s in he UV– is spec oscopic cell a 0.05 ba a e
120 min on s eam.
José Valecillos, H. Vicen e, A.G. Gayubo e al. Jou nal o Ca alysis 408 (2022) 115–127
123
