Chemosphe e 300 (2022) 134499
A ailable online 4 Ap il 2022
0045-6535/© 2022 The Au ho s. Published by Else ie L d. 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/).
Composi e was e ecycling: P edic i e simula ion o he py olysis apou s
and gases upg ading p ocess in Aspen plus
A. Se as-Malillos
*
, E. Acha, A. Lopez-U ionaba enechea, B.B. Pe ez-Ma inez, B.M. Caballe o
Chemical and En i onmen al Enginee ing Depa men , Facul y o Enginee ing o Bilbao, Uni e si y o he Basque Coun y (UPV/EHU), Plaza Ingenie o To es Que edo,
1, 48013, Bilbao, Spain
HIGHLIGHTS GRAPHICAL ABSTRACT
•Kine ic mul i- eac ion model is p o-
posed o py olysis apou s and gases
upg ading.
•Simula ions con i m he educ ion o he
numbe o compounds a he ou le
s eam.
•95 ol% o syngas is he modynamically
p edic ed a 900 ◦C.
•Kine ic p edic ion o H
2
(g) a 900 ◦C
has a 3.8 ol% absolu e e o .
ARTICLE INFO
Handling Edi o : De ek Mui
Keywo ds:
Composi e was e ecycling
Ca bon ibe ein o ced polyme
Epoxy esin alo isa ion
P edic i e p ocess modelling
Aspen plus
Ci cula economy
ABSTRACT
Was e gene a ion is one o he g ea es p oblems o p esen imes, and he ecycling o ca bon ib e ein o ced
composi es one big challenge o ace. Cu en ly, no esin alo isa ion is done in he mal ib e ecycling me hods.
Howe e , when py olysis is used, addi ional aluable compounds (syngas o H
2
- ich gas) could be ob ained by
upg ading he gene a ed apou s and gases. This wo k p esen s he he modynamic and kine ic mul i- eac ion
modelling o he py olysis apou s and gases upg ading p ocess in Aspen Plus so wa e. These models o ecas
he heo e ical and in-be ween scena io o a he mal upg ading p ocess o an expe imen ally cha ac e ised a-
pou s and gases s eam (a blend o hi y- i e compounds). Indeed, he in luence o empe a u e
(500 ◦C–1200 ◦C) and p essu e (ΔP =0, 1 and 2 ba ) ope a ing pa ame e s a e analysed in he ou le compo-
si ion, esidence ime and possible eac ion mechanisms occu ing. Valida ion o he kine ic model has been done
compa ing p edic ed ou le composi ion wi h expe imen al da a (a 700 ◦C and 900 ◦C wi h ΔP =0 ba ) o H
2
(g), CO (g), CO
2
(g), CH
4
(g), H
2
O ( ) and C (s). Kine ic and expe imen al esul s show he same endency wi h
empe a u e, alida ing he model o u he esea ch. Good kine ic i is ob ained o H
2
(g) (absolu e e o : 0.5
w % a cons an empe a u e and 0.3 w % a a iable empe a u e) and H
2
O ( ) shows he highes e o a
a iable T (8.8 w %). Bo h simula ion and expe imen al esul s e ol e owa ds simple , less oxic and highe
gene a ion o hyd ogen- ich gas wi h inc easing ope a ing empe a u e and p essu e.
* Co esponding au ho .
E-mail add ess: [email p o ec ed] (A. Se as-Malillos).
Con en s lis s a ailable a ScienceDi ec
Chemosphe e
jou nal homepage: www.else ie .com/loca e/chemosphe e
h ps://doi.o g/10.1016/j.chemosphe e.2022.134499
Recei ed 25 Decembe 2021; Recei ed in e ised o m 25 Feb ua y 2022; Accep ed 31 Ma ch 2022
Chemosphe e 300 (2022) 134499
2
1. In oduc ion
The e is no denial ha ci cula economy is an e ec i e s a egy o
educe was e. The challenge is o implemen his concep wi h complex
was e, such as composi e ma e ials. The ou s anding ma e ial p ope ies
o ca bon ib e ein o ced polyme s (CFRP) make hem o be widely used
in many indus ies such as de ence, ae onau ic, ene gy and au omo i e.
So a , abou he 23% o he cu en ca bon ib e (CF) demand co e-
sponds o he ae onau ic sec o (Zhang e al., 2020). This indus y has
p og essi ely inc eased he use o composi es in he manu ac u ing o
planes, ha ing eached 50 and 52 w % in he Boeing 737 (Hale, 2012)
and Ai bus A350 (Ma sh, 2010) plane models, espec i ely. Mo eo e ,
ca bon ib e is also consolida ing i s p esence in he au omo i e sec o ,
due o con inuous imp o emen s in he au oma ion o manu ac u ing
p ocesses. In his case, he d i ing o ce is he ad an age p o ided by he
e y ligh weigh o he composi es, ha ing a di ec e ec in he educ-
ion o emissions in ehicles. As a esul , composi e was e gene a ed
du ing he manu ac u ing p ocesses, as well as a he end o hei li e-
span, will con inue inc easing in he coming yea s. In ac , when he
planes manu ac u ed be ween 2010 and 2020 each hei end o li e, he
ae onau ic sec o is expec ed o p oduce 9540 o CFRP was e/yea
be ween 2020 and 2035 and 23,360 o CFRP was e/yea in he pe iod
2035–2050 (Le eu e e al., 2017). The e o e, he e is an u gen need o
implemen a ci cula economy model o eco e ca bon ib es om
was e, and ein oduce hem as seconda y aw ma e ials in
ib e-consuming sec o s, al hough cu en ly he e is no speci ic egula-
ion o his was e.
In ecen yea s, signi ican esea ch is being done owa ds he
eclama ion o ca bon ib es om was e composi es. Among he exis ing
ecycling p ocesses, he mal ones, such as py olysis, ha e he highes
echnology eadiness le el (TRL), ollowed by sol olysis and biological
deg ada ion. Nowadays, he exis ing composi e was e ecycling com-
panies, such as Gen 2 Ca bon, Ka bo ek RCF and Reciclalia (Bueno
Lopez e al., 2017) among o he s, ocus only on he ca bon ib e ecla-
ma ion and do no alo ise he esin, which e ol es in o a complex
condensa e ha would p obably be classi ied as haza dous was e ac-
co ding o i s composi ion. A pa en ed esea ch ca ied ou by he au-
ho s o his wo k shows ha aluable chemical compounds, such as
syngas o hyd ogen, could be eco e ed om he apou s and gases
gene a ed in he py olysis p ocess o was e composi e i an app op ia e
he mal-ca aly ic ea men is implemen ed o upg ade hem (L´
opez
U ionaba enechea e al., 2016). Howe e , i is s ill needed o wo k on
he up-scale o his p ocess p io o i s in eg a ion in he exis ing was e
composi e ecycling p ocesses. Among o he s, p ocess-modelling ools
a e equi ed o b idge he gap be ween labo a o y and indus ial
implemen a ion. The scien i ic wo k equi ed o implemen he esin
alo isa ion p ocess in ca bon ib e ecycling indus ies would be a
signi ican b eak h ough, as i would ans o m a haza dous was e in o a
sa e and saleable by-p oduc . Ne e heless, he success ul eac o up-
scale would depend on he adequa e con igu a ion, sizing and choice o
he p ocess ope a ing condi ions. In his sense, he e is one majo limi-
a ion wi h size speci ic pa ame e s, as op imum ope a ing condi ions a
labo a o y-scale would no necessa ily be alid o a pilo o indus ial
scale. E en hough eac o simula ion may no p o ide highly p ecise
p edic ion o i s pe o mance, i could gi e quali a i e guidance o
design and ope a ion, meaning i can assis in he scale-up design om
one success ully ope a ing size o ano he (Basu, 2018).
The added alue o he upg ading p ocess s udied in his wo k is he
gene a ion o hyd ogen- ich gas om a complex apou s and gases
mix u e. The e o e, he gene a ion o his gas will be he key pa ame e
indica o in he upscale design p ocess, and hence, in he simula ion
p ocess. The s a egy adop ed o his wo k is o model he chemical
ans o ma ion occu ing du ing he esin alo isa ion p ocess wi h
Aspen Plus so wa e (Aspen Technology Inc., 2021). Ou le gas s eam
composi ion is p edic ed in unc ion o inle s eam composi ion, eac o
empe a u e, p essu e and basic geome ic dimensions. The simula ion
is ed and alida ed wi h expe imen al own da a. The signi icance o
de eloping such a p edic i e model is wo old: I) o gain a be e un-
de s anding o he eac ions occu ing du ing he upg ading p ocess and
he in luence o p ocess pa ame e s ( empe a u e and p essu e) in he
composi ion o ou le apou s and gases, and II) i s a ailabili y o
in eg a ion wi h luid dynamics and hea ans e simula ion models o
e alua e he impac o p ocess dimensions a he ou le gas composi ion
in a subsequen esea ch s age.
Un il now he goal o he majo i y o he simula ion esea ch wo ks,
ela ed o composi e ma e ials, has been o be e unde s and and p e-
dic he esponse o CFRP ma e ials in he p esence o i e (G ange e al.,
2018; McKinnon e al., 2017; S olia o e al., 2014; T ancha d e al.,
2015, 2017a, b, c). Howe e , he emphasis o he kine ic models p o-
posed by hese esea ch g oups lays on he weigh loss, mo e han in he
speci ic chemical decomposi ion eac ions. No wi hs anding he ele-
ance o exis ing he mal deg ada ion simula ion s udies o epoxy
based composi es, no p edic i e modelling has been ound in he li e -
a u e in e ms o chemical compounds gene a ion. Nei he was ound on
he p oduc s gene a ed du ing seconda y c acking ha unde go py ol-
ysis apou s and gases when a subsequen he mal deg ada ion p ocess
is applied, as is he case o he echnology pa en ed by he au ho s.
Mo eo e , e en hough many py olysis simula ion wo ks ha e been
ca ied ou using Aspen Plus so wa e o biomass eeds ock, no
modelling conce ning CFRP py olysis o he mal ea men o he py-
olysis apou s and gases was ound. Al hough hose wo ks a e ocused
on biomass, adop ed modelling s a egies could be o in e es . Indeed,
acco ding o a ecen e iew (Mu lu and Zeng, 2020) and o he s udies
(Gao e al., 2021; Salisu e al., 2021), he mos common adop ed s a-
egies and assump ions include: 1) equilib ium and kine ic app oaches,
2) s eady s a e egime, 3) iso he mal homogeneous empe a u e and
p essu e p o ile, 4) no p essu e d ops, 5) no a conside a ion o ep-
esen a ion by model compounds, and 6) model alida ion based on
li e a u e da a in e ms o ob ained syn-gas composi ion. Indeed,
ega ding py olysis apou s upg ading p ocess, A. Ahmed e al. (2015)
e iewed he p incipal a o ma ion and emo al (by e o ming o
c acking) modelling s a egies in biomass he mal ea men using
Aspen Plus so wa e. In he e iewed wo ks (Dama zis e al., 2012; De
Kam e al., 2009; F ancois e al., 2013; Hannula and Ku kela, 2012;
Nilsson e al., 2012; Panopoulos e al., 2006; Sadhukhan e al., 2010), a
is ep esen ed by model compounds (naph halene, oluene, benzene and
phenol) a ying om one o ou species depending on he s udy.
The p esen wo k aims o de elop a py olysis apou s and gases
upg ading model in Aspen Plus by using expe imen al own da a o
alida e i s applicabili y in he up-scale p ocess op imisa ion design o
an epoxy based CFRP was e. Mo e p ecisely, he modynamic and kine ic
eac o models a e used ed by expe imen al da a gene a ed a labo a-
o y scale, and cons an and a iable empe a u e and p essu e d op
scena ios a e s udied. Mo eo e , apou s and gases he mal upg ading is
modelled kine ically wi h a simpli ied i een compound s eam,
including h ee model-compounds o ep esen oxygen, sulphu and
ni ogen a oma ics. Model alida ion is done compa ing da a ela ed o
he gene a ion endencies (in w %) o main gaseous species (H
2
(g), CO
(g), CO
2
(g), CH
4
(g)), wa e apou and solid ca bon o expe imen al
and simula ion esul s. Absolu e e o (di e ence be ween expe imen al
and simula ed esul s) o he p edic ed esul s is also calcula ed.
2. Ma e ials and me hods
2.1. Expe imen al da a o he simula ion
The yield and composi ion o he apou s, gases and solid ca bon a
he inle and ou le o he apou s and gases ea men eac o was
needed o un he he modynamic and kine ic simula ions. On he one
hand, o de e mine he inle composi ion, py olysis o 100 g o an
expi ed ca bon ib e ein o ced epoxy p e-p eg composi e (T800/M21)
was ca ied ou in a non-s i ed ank eac o by hea ing he sample a
A. Se as-Malillos e al.
Chemosphe e 300 (2022) 134499
3
3 ◦C min
−1
up o 500 ◦C in he absence o any ca ie gas. In such
condi ions 20.2 g o apou s +gases and 79.8 g o solid ( eclaimed
ca bon ib es 66 g and cha 13.8 g, assumed as 100 w % solid ca bon)
we e ob ained. The composi ion o apou s (collec ed as liquids) and
gases we e analysed by gas ch oma og aphy (GC AGILENT 6890, Uni ed
S a es) coupled wi h mass spec ome e de ec o (MS AGILENT 5973,
Uni ed S a es) and gas ch oma og aphy (AGILENT 7890A, Uni ed
S a es) coupled wi h he mal conduc i i y de ec o (TCD) and lame
ioniza ion de ec o (FID), espec i ely. The composi ion o he gases was
calib a ed using a s anda d e ine y gas mix u e, while in he case o
liquids (condensed apou s), a ea pe cen age ob ained a he GC-MS
was assumed o equal weigh pe cen age, due o he complexi y o he
composi ion. In addi ion o he men ioned analy ical echniques, Fou ie
ans o m in a ed (FTIR) DX-4000 model gas analyse (Gasme ,
Finland) was also used o quali a i ely iden i y ace compounds.
Fu he de ails o he py olysis labo a o y ins alla ion a e desc ibed
elsewhe e (Lopez-U ionaba enechea e al., 2021).
On he o he hand, he composi ion o apou s, gases and solid
ca bon a he ou le o he ea men eac o was de e mined by ca ying
ou expe imen s whe e py olysis o he expi ed p e-p eg ook place a
he same condi ions desc ibed abo e and he gene a ed py olysis a-
pou s and gases ascending by na u al con ec ion passed h ough a ixed
bed ubula eac o ( ea men eac o ), o be upg aded a 700 ◦C and
900 ◦C. Ma e ial o he ixed bed ea men eac o is s ainless s eel
(AISI-309), dimensions a e 0.6 m long and a 0.02 m inne diame e and
i is elec ically and concen ically hea ed. The eac o was illed wi h
e ac o y b ick ma e ial o 0.5–1 mm pa icle diame e as solid bed.
The composi ion o apou s (condensed in o liquids) and gases
e ol ing om he ea men eac o was analysed by he same ch o-
ma og aphic me hods desc ibed p e iously. Some chemicals we e no
well iden i ied by he MS lib a y when analysing liquids coming om
apou s condensa ion composi ion de e mina ion. To eed he simula-
ion, his pe cen age o uniden i ied compounds a bo h, inle (12 w %)
and ou le (16 w % a 700 ◦C and 4 w % a 900 ◦C), was p opo ionally
dis ibu ed among he well-iden i ied compounds o speci y he
composi ion o he s eams unde s udy. Addi ionally, six compounds
ound in apou s (1,3-benzo hiazole, 1-me hyl-3-phenoxybenzene,
phenylsul anybenzene, 6H-benzo[c]ch omene, 5-me hyl-1H-indole,
2,3-dime hyl-1H-indole) we e excluded om he simula ion due o he
p oblems hey pose in Aspen Plus 11 blocks because o missing p op-
e ies and/o inadequa ely es ima ed p ope ies. Fig. 1 shows he
apou s, gases and solid ca bon composi ion (in w % and ol%) used as
inle and ou le (a 700 ◦C and 900 ◦C ope a ing empe a u es) s eams
o be employed and compa ed wi h he ea men eac o simula ions.
The exac weigh pe cen ages o each compound and expe imen al
condi ion shown in Fig. 1 a e included in he Supplemen a y In o ma-
ion (SI) documen , in Table S1. T aces o sulphu dioxide, ni ous oxide,
ammonia, hyd ochlo ic acid, hexane and o maldehyde we e also
iden i ied in gas phase by FTIR, bu due o lack o quan i ica ion hey
we e no included in he simula ion. Finally, he ac ion o solid ca bon
gene a ed du ing he expe imen s and he ac ion o non- eco e able
condensed apou s along he walls o he eac o s is also included in
Fig. 1. Schema ic ep esen a ion o expe imen ally gene a ed ac ions
and yield calcula ion appea in Figu e S1-S6 and Table S2-S3.
Apa om apou s and gases, ca bon coming om ca boniza ion
eac ions (assumed as 100% solid ca bon) is also a by-p oduc o he
upg ading p ocess and i is o med and e ained in he pa icles o ming
he solid bed o he ea men eac o . Consequen ly, he solid bed ma-
e ial was cha ac e ised in esh and a e -expe imen condi ions. In he
las case, samples ex ac ed om h ee zones o he ea men eac o
we e analysed: low, medium and high. Cha ac e iza ion included
di e en analysis: measu ing weigh gain a e expe imen s, weigh loss
by calcina ion a 500 ◦C, p oxima e analysis (humidi y, ola ile ma e ,
ixed ca bon and ashes) using LECO TGA-701 (Uni ed S a es) and ul i-
ma e analysis using LECO T ueSpec CHNS au oma ic analyse (Uni ed
S a es). The weigh gain o he ixed bed ma e ial ascended o 3.3 g a
700 ◦C and 2.9 g a 900 ◦C, espec i ely, and an inc ease in w % was
app ecia ed o he ou h analysed elemen s, CHNS. See Table S4,
Table S5, Table S6 and Table S7 in SI o u he de ails abou he esul s
o ixed bed ma e ial cha ac e iza ion.
A las , he kine ic modelling in Aspen Plus o e s he possibili y o
es ablishing a empe a u e g adien along he leng h o he ea men
eac o . Wi h ha objec i e, empe a u e dis ibu ion along he eac o
was measu ed wi h a 600 mm leng h he mocouple (ex ac ing i 1 cm
each ime om he uppe pa o he eac o ) using wa e apou (0.02
mol/min ed h ough a Gilson pump) as ep esen a i e luid o he py-
olysis apou s o he wo s udied se poin empe a u es a s eady s a e
condi ions. The majo i y (cen al pa ) o he eac o leng h was a
cons an empe a u e, while he inle and ou le sec ions showed a
empe a u e g adien which was highes a 900 ◦C and lowes a 700 ◦C.
De ails on empe a u e dis ibu ion a e shown in Figu e S7.
Fig. 1. Vapou s and gases composi ion, solid ca bon and condensed a a he inle /ou le o he ea men eac o in WT% and ol%.
A. Se as-Malillos e al.
Chemosphe e 300 (2022) 134499
4
2.2. P ocess simula ion
The py olysis apou s and gases upg ading p ocess is ep esen ed in
Aspen Plus as indica ed in “s ep 2 upg ading” in he lowshee diag am
shown in Fig. 2, whe e he concep ualiza ion o an in eg al was e
composi e ecycling p ocess is shown. Fo he simula ion in Aspen Plus
11, p ope y me hod selec ed was Soa e-Redlich-Kwong-Kabadi-
Danne (SRK-KD), which is adequa e o wa e and hyd oca bon mix-
u es in he p esence o ligh gases (see de ails o he model and he
equa ion o s a e in he SI documen in pages 8–9). Rega ding model
assump ions, simula ion egime was se as s eady s a e.
Inle s eam o he PYRO eac o , whe e he py olysis o he expi ed
p e-p eg (PRE-PREG) is ca ied ou , was se a a mass a e o 100 g/h,
equal o he mass in oduced in he labo a o y expe imen s pe es un
(g/ es ), and he ou le s eam (PYRO-OUT) comp ised he gene a ed
py olysis apou s and gases (VAP +GAS) a a a e o 20.2 g/h and he
py olysed solid (PYRO-SOL) a a a e o 79.8 g/h (66 g/h o eclaimed
ca bon ib e and 13.8 g/h o cha de ined as solid ca bon), which a e he
yields ob ained in he lab-scale py olysis expe imen s (see Table S2).
The he modynamically p edic ed apou s and gases composi ion
ob ained a he ou le o he ea men eac o (V +G-OUT1) was
simula ed by a he modynamic model eac o RGibbs (RTHERM), whose
calcula ions a e based on he minimiza ion o Gibbs ee ene gy o he
chemical sys em in ol ed. Consequen ly, i is no he chemical eac ions
ha need o be de ined bu he chemical sys em which, in his case, is he
one composed by he chemicals expe imen ally iden i ied in bo h he
eac o inle and ou le s eams. A wide ope a ing empe a u e ange
(500 ◦C–1200 ◦C) was simula ed, which emb aces he expe imen al
ones, a 1 ba (absolu e p essu e) o e alua e possible endencies in
unc ion o empe a u e. Simula ion was pe o med on he basis o
chemical and phase equilib ium calcula ions.
Conce ning he global he mal decomposi ion kine ic modelling, he
apou s and gases ou le composi ion (V +G-OUT2) including solid
ca bon was p edic ed based on he eac o model RPlug (RKIN). This
la e app oach was mean o p o ide a mo e ealis ic scena io as eac o
eal dimensions (see sec ion 2.1) and a ious global he mal decompo-
si ion and o he ypical kine ic eac ions o he py olysis en i onmen
(Table 1) o a simpli ied solid ca bon, apou s and gases s eam, we e
speci ied. The compila ion o hese kine ic eac ions we e chosen by he
au ho s o his wo k, as ep esen a i e o he he mal ea men p e-
sen ed and ex ac ed om li e a u e, whe e he eac ion s oichiome y
and kine ic pa ame e s in o ma ion (ac i a ion ene gy, equency ac o
and eac ion o de ) was a ailable. Rega ding he ope a ing empe a u e
(700 ◦C, 800 ◦C and 900 ◦C), he simula ed empe a u e ange ocused
on p e ious own esea ch expe imen s (Gas elu O azua, 2020) and he
simula ions we e de ined conside ing wo scena ios. Fi s ly, i was
assumed he ideal si ua ion whe e he empe a u e along he whole
ea men eac o was cons an and, secondly, a mo e ealis ic scena io
was simula ed by in eg a ing he expe imen ally measu ed g adien
empe a u e along he ix bed ea men eac o (see sec ion 2.1). Be-
sides, he in luence o p essu e was also analysed o de e mine wha he
e ec o a possible gene a ion o p essu e d op in he ea men eac o
caused by he ixed bed ma e ial could be. On he basis o own expe i-
men al obse a ion in he labo a o y-scale pilo plan , h ee di e en
inle p essu es we e de ined (1 ba , 2 ba and 3 ba ), which could be
close o eali y, while ou le p essu e was main ained a 1 ba . The e-
o e, h ee p essu e d op scena ios we e s udied: ΔP =0 ba , ΔP =1 ba
and ΔP =2 ba .
As a as he au ho s a e conce ned, in he li e a u e he e is no
kine ic da a ela ed o he he mal decomposi ion o each o he chem-
ical compounds iden i ied in he apou s and gases mix u e (VAP +
GAS) s eam. This da a esea ch p ocess has mainly, bu no exclusi ely,
been done accessing he NIST Chemical Kine ics Da abase (Manion
e al., 2013). Due o his ac , a simpli ied i een compound s eam was
de ined (V +G-IN2) as ep esen a i e o he whole apou s, gases and
solid ca bon composi ion: solid ca bon (40.6 w %), wa e (15.2 w %),
benzene (1.02 w %), oluene (3.05 w %), py ole (14.4 w %), phenol
(22.5 w %), p opane (0.09 w %), p op-1-ene (0.16 w %), e hane (0.24
w %), e hene (0.20 w %), me hane (0.46 w %), ca bon dioxide (0.29 w
%), ca bon monoxide (0.87 w %), hyd ogen sulphide (0.89 w %) and
hyd ogen (0.03 w %). Indeed, h ee model-compounds (py ole, phenol
and oluene) ep esen ed he ni ogenous, oxygena ed a oma ic and
sulphu a oma ic compounds, espec i ely.
I should be no ed ha py ole was no p esen among he expe i-
men ally iden i ied species, bu i was selec ed as a ep esen a i e
compound o ni ogenous a oma ics in he absence o kine ic in o -
ma ion ega ding he ni ogenous compounds iden i ied in he pe -
o med expe imen al wo k. Thus, he p epa ed model assumes ha he
c acking mechanism o py ole is ep esen a i e o he ni ogenous
species p esen in he eeding s eam, which could di e o a ce ain
ex en om eali y. Likewise, he kine ic pa ame e s o he he mal
decomposi ion o any o he expe imen ally obse ed a oma ic sulphu
Fig. 2. Flowshee o he p ocess modelled in Aspen Plus 11.
A. Se as-Malillos e al.
Chemosphe e 300 (2022) 134499
5
Table 1
The mal decomposi ion kine ic modelled eac ions, a e o he eac ions and kine ic pa ame e s.
The mal decomposi ion kine ic eac ion - kmol⋅m
−3
⋅s
−1
, a m⋅s
−1
E
a
(kJ⋅ kmol
−1
) A (s
−1
, m, kmol, a m) Re
R-1 C
6
H
6
(g) +H
2
O (g) →
3 C (s) +2⋅CH
4
(g) +CO (g)
k⋅C
1.3[C6H6]
⋅C
0.2[H2O]
443,000 4.00 ⋅ 10
16
Abdelouahed e al., (2012)
R-2 C
7
H
8
(g) +H
2
(g) →
C
6
H
6
(g) +CH
4
(g)
k⋅C
1[C7H8]
⋅C
0.5[H2]
247,000 1.04 ⋅ 10
12
Abdelouahed e al., (2012)
R-3 C
4
H
4
NH (g) →
CH
4
(g) +2 C (s)
a
+HCN (g)
k⋅C
1[C4H4NH]
311,000 7.19 ⋅ 10
13
B uinsma e al., (1988)
R-4 C
6
H
6
O (g) →
CO (g) +0.4⋅C
10
H
8
(g) +0.15⋅C
6
H
6
(g) +0.1⋅CH
4
(g) +0.75⋅H
2
(g)
k⋅C
1[C6H6O]
100,000 1.00 ⋅10
7
Abdelouahed e al., (2012)
R-5 C
3
H
8
(g) →
C
3
H
6
(g) +H
2
(g)
k ⋅ C
1[C3H8]
234,000 1.27 ⋅ 10
12
Benson, (1967)
R-6 C
3
H
6
(g) →
C
3
H
4
(g) +H
2
(g)
k ⋅ C
1[C3H6]
297,000 5.01 ⋅ 10
12
Ba b´
e e al., (1996)
R-7 C
2
H
6
(g) →
C
2
H
4
(g) +H
2
(g)
k ⋅ C
2[C3H8]
343,000 1.15 ⋅ 10
−7
B odsky e al., (1960)
R-8 C
2
H
4
(g) →
C
2
H
2
(g) +H
2
(g)
k ⋅ C
1[C2H4]
318,000 1.8 ⋅ 10
13
Towell and Ma in, (1961)
R-9 CH
4
(g) →
C (s) +2H
2
(g)
k⋅C
1[CH4]
370,000 6.60 ⋅ 10
13
Roda e al., (2009)
R-10 CO (g) +H
2
O (g) →
CO
2
(g) +H
2
(g)
k⋅C
1[CO]
⋅C
1[H2O]
102,400 1.35 ⋅ 10
5
Abdelouahed e al., (2012)
CO
2
(g) +H
2
(g) →
CO (g) +H
2
O (g)
k⋅C
1[CO2]
⋅C
0.5[H2]
318,000 1.20 ⋅ 10
10
Abdelouahed e al., (2012)
R-11 C
10
H
8
(g) →
9 C (s) +(1/6)C
6
H
6
(g) +(7/2)H
2
(g)
k⋅C
1.6[C10H8]
⋅C
−0.5[H2]
350,000 3.40 ⋅ 10
14
Abdelouahed e al., (2012)
R-12 C (s) +β⋅H
2
O ( ) →
(2-β)⋅CO
2
(g) +(β-1)⋅CO (g) +β⋅H
2
(g)
Choice: β =1.5
k⋅P
1[H2O]
240,000
b
18412.1 Gao e al., (2021)
a
Hypo hesis: 2 C (s) gene a ion ins ead o C
2
(g).
b
Ea has been adjus ed in o de o be e adjus he eac ion mechanism o expe imen al obse a ions.
A. Se as-Malillos e al.
Chemosphe e 300 (2022) 134499
6
compounds we e no ound. In his case, a di e en s a egy was
implemen ed. Taking in o accoun he s udy ca ied ou by Xu e al.
(2004) i was assumed ha a oma ic sulphu compounds c ack in o
oluene and sulphu adical species. This is he eason why, in he p e-
sen wo k, he weigh pe cen age associa ed o hose sulphu ous com-
pounds was included in he g oup ep esen ed by he model compound
oluene. Finally, wa e weigh pe cen age in he he modynamic simu-
la ion (Table S8) does no exac ly equal he one indica ed he e, because
in he kine ic model compounds we e p opo ionally edis ibu ed only
due o he exclusion o he weigh pe cen age associa ed wi h he un-
known species. The e o e, he de ined mul i- eac ion sys em consis s o
wel e powe -law kine ic eac ions, shown in Table 1, associa ed o he
compounds p esen in he simpli ied s eam (R-1 o R-10 and R-12) and
one in e media e specie (R-11) gene a ed as p oduc s o he o me R-4
eac ion. No e ha H
2
O ( ) and H
2
(g) we e also included as pa o
eac ions R-1, R-2, R-10 and R-12. In his Table 1 he powe law (k =A ⋅
exp(-E
a
/RT)) kine ic pa ame e s and he a e o he eac ions (- ) a e also
gi en.
3. Resul s and discussion
3.1. The modynamic model
The modynamic p edic ion model esul s a empe a u es be ween
500 and 1200 ◦C a e included in Fig. 3, whe e he ou le composi ion o
apou s, gases and solid ca bon o he ea men eac o is shown in
weigh and olume pe cen age. The weigh pe cen age esul s show he
dis ibu ion o he inal p oduc s, including he solid ca bon ha would
be o med in he eac o . The olume pe cen age esul s, on he o he
hand, show he composi ion o wha would be he ou le apou and gas
s eam, which ob iously does no include he solid ca bon. Nume ic
esul s a e included in Supplemen a y In o ma ion (Table S8 and S9).
The main highligh is ha al hough a e y complex mix u e o apou s
and gases en e ed he eac o (speci ied in Table S8 and S9 and depic ed
in Fig. 3 as “Inle ”), he mix u e simpli ied in o C (s), N
2
(g), H
2
(g), CO
2
(g), CO (g), CH
4
(g), H
2
S (g) and H
2
O ( ), e en a he lowes empe a-
u e. This means ha he haza dousness ela ed o he inle composi ion
o he ea men eac o would be educed o he minimum. I could be
app ecia ed ha when empe a u e inc eased, he amoun o C (s), H
2
O
( ), CO
2
(g) and CH
4
(g) dec eased, whe eas CO (g) and H
2
(g) inc eased.
The dec ease in wa e con en means ha he gases gene a ed a 900 ◦C,
and abo e, would no longe con ain condensa e, as he pa ial p essu e
o wa e ( he only condensable subs ance) in his gas mix u e would be
lowe han i s apou p essu e a oom empe a u e. Addi ionally, i
seems ha he sys em s abilized abo e 1000 ◦C, wi h no u he big
di e ence in composi ion, comp ising 31/66 ol% o CO (g)/H
2
(g)
wi h aces o N
2
(g) and H
2
S (g), which a e he minimum ee ene gy
chemicals o he amilies o ni ogenous and sulphu compounds,
espec i ely, be ween hose de ined o he simula ion. Rega ding he
compa ison be ween hese esul s and he expe imen ally ob ained ones
(a 700 ◦C and 900 ◦C), expe imen ally highe amoun o o ganic com-
pounds we e ob ained, wha gi es an idea o he di e ence be ween he
eal ope a ion and he he modynamic equilib ium. In any case, he
same end was also obse ed in he expe imen al esul s wi h ega d o
Fig. 3. Py olysis apou s and gases ou le composi ion esul s o The modynamic simula ion a empe a u es be ween 500
◦C and 1200 ◦C and 1 ba absolu e
p essu e: a) ol% and b) w %.
A. Se as-Malillos e al.
Chemosphe e 300 (2022) 134499
7
he in luence o empe a u e. A 900 ◦C, highe amoun s o H
2
(g) and
CO (g) we e obse ed o he de imen o he o ganic and wa e con en
compa ed o hose egis e ed a 700 ◦C. Commen s ela ed o he C (s)
alues a e discussed in he nex sec ion, combined wi h kine ic simula-
ion esul s.
Based on he eac ion scheme ypically used o ep esen was e
gasi ica ion p ocesses (A ena, 2012), some ex apola ion could be done
o py olysis eac ion mechanisms in o de o y o in e he possible
eac ion mechanism lying behind he he modynamic esul s ob ained.
Taking in o conside a ion he complex na u e o he py olysis apou s
and gases inle composi ion and he high empe a u e o he upg ading
p ocess in he ea men eac o , i could be expec ed ha endo he mic
decomposi ion eac ions o hyd oca bons pC
x
H
y
→ qC
n
H
m
+ H
2
and
a s C
n
H
m
→ nC +(m/2)H
2
we e occu ing. This ac implies ha
simple compounds we e gene a ed due o b eaking o complex o ganic
molecules, as well as ca boniza ion and dehyd ogena ion eac ions.
Indeed, a he inle o he upg ading p ocess, minimal H
2
(g) was p e-
sen , and he e o e, he highe he empe a u e he mo e hese equilib-
ium eac ions we e shi ed o he p oduc s due o hei endo he mic
na u e. Fu he mo e, he p esence o wa e in he py olysis apou s and
gases inle s eam enables an addi ional pa h o he gene a ion o H
2
(g)
and CO (g) h ough s eam e o ming eac ion C
n
H
m
+nH
2
O ←→ nCO +
(n +m/2)H
2
. E en hough ca bon monoxide was p esen a he inle
s eam, i migh p e ail he endo he mic na u e o his la e eac ion so
as o educe wa e apou in a ou o CO (g) and H
2
(g) p oduc ion.
Acco ding o he obse ed endency wi h empe a u e inc ease, whe e C
(s), H
2
O ( ), CO
2
(g) and CH
4
(g) dec ease, addi ional eac ions such as
me hane s eam e o ming CH
4
+H
2
O ←→ CO +3H
2
, d y e o ming
C
n
H
m
+nCO
2
←→ 2nCO +(m/2)H
2
and Boudoua d C +CO
2
←→ 2CO
could also be occu ing. Finally, ega ding he ni ogen and sulphu
he e oa oms p esen in he inle s eam, i seems ha hey e ol e in o
hei simples o ms, H
2
S (g) and N
2
(g), no ma e he ope a ing
empe a u e.
3.2. Kine ic mul i- eac ion model
The ou le composi ion o apou s, gases and solid ca bon in olume
and weigh pe cen age o he ixed bed ea men eac o kine ic
simula ion is shown in Fig. 4. Nume ic esul s a e included in SI
(Table S10 o S15). Mo e p ecisely, esul s ob ained o h ee ope a ing
empe a u es (700 ◦C, 800 ◦C and 900 ◦C), conside ing hem as bo h
cons an , along he ea men eac o (T
c
), and a iable, applying he
expe imen ally measu ed empe a u e p o ile (T
ble
), a e shown. In
addi ion, his empe a u e s udy was analised o h ee simula ed p es-
su e d op scena ios (ΔP =0 ba , 1 ba and 2 ba ). I is necessa y o
emembe ha in he kine ic case, he compounds ha will appea as
pa o he p oduc s a e only hose included in Table 1 as a consequence
o he lis o eac ions de ined. This means ha he simula o could
p edic he occu ence o some subs ances ha we e no obse ed
expe imen ally.
As also obse ed in he he modynamic p edic ion, he highe he
ope a ing empe a u e he simple he gene a ed p oduc s o a speci ic
p essu e d op. Indeed, less wa e , oluene, py ole, and solid ca bon
we e gene a ed and mo e benzene, hyd ogen cyanide, ace ylene,
me hane, ca bon dioxide, ca bon monoxide and hyd ogen we e p o-
duced a bo h s udied empe a u e scena ios (cons an and a iable),
Fig. 4. Kine ic mul i- eac ion model Simula ion ou le Py olysis apou s and gases composi ion in ol% ( op) and in w % (down) in unc ion o empe a u e
(cons an and a iable a 700
◦C, 800 ◦C and 900 ◦C) and p essu e DROP (0 ba , 1 ba AND 2 ba ).
A. Se as-Malillos e al.
Chemosphe e 300 (2022) 134499
8
al hough he e ec was maximized a he cons an . This can be due o
he ac ha chemical compounds we e kep o longe ime a high
empe a u e inside he eac o . Also, a cons an quan i y o naph halene
was p edic ed o all he s udied condi ions, meaning ha based on he
kine ic pa ame e s associa ed o he he mal deg ada ion eac ion R-11,
his compound emains s able. Rega ding he p edic ion o HCN (g)
gene a ion, i was no expe imen ally de ec ed. The e o e, u he
esea ch would ocus on checking i s p esence as a he mal deg ada ion
p oduc o ni ogenous a oma ic compounds. Likewise, C
2
H
2
(g) is
kine ically p edic ed o be gene a ed and de ec ed expe imen ally by
GC-TCD/FID a 900 ◦C, al hough no quan i ica ion was possible. Finally,
e en hough p edic ions could co espond o eali y, i could also
happen ha HCN (g) and C
2
H
2
(g) would decompose in o simple spe-
cies such as C (s), N
2
(g) and H
2
(g) as p edic ed by he he modynamic
model. Howe e , he kine ic mul i- eac ion model would no be able o
p edic he gene a ion o compounds ha a e no al eady p esen in he
conside ed kine ic eac ions as p oduc s.
Rela ed o he e ec o he p essu e d op along he ea men eac o
(ΔP =0 ba , ΔP =1 ba and ΔP =2 ba ), equi alen end o ha
desc ibed o a empe a u e inc ease a cons an p essu e was obse ed.
Besides, he highe he p essu e-d op, he highe he esidence ime
(Table S16). Indeed, longe ime inside he eac o seemed o be di ec ly
ela ed wi h a highe amoun o hea y o ganic compounds e ol ing in o
ligh e ones. In his sense, he e ec o empe a u e would p o oke he
opposi e e ec on he esidence ime. A empe a u e inc ease would
mean he amoun o gene a ed moles inc eases as endo he mic eac ions
a e kine ically and he modynamically a ou ed (see de ined eac ion
s oichiome y in Table 1). The e o e, as he eac o olume emained
cons an , he gene a ed p oduc s would pass h ough he eac o as e .
Rega ding he in luence o he empe a u e g adien along he ixed
bed ea men eac o , based on he p o ile o each compound along he
ea men eac o (Figu e S8), wi h a iable empe a u e he compounds
s a ed eac ing a inne posi ions and s opped ea lie oo, compa ed o
ope a ions a cons an empe a u e. No only he empe a u e is
esponsible o his di e ence, bu also he concen a ion a which
compounds a e p esen and a e a ailable o each o he .
Following he kine ic mul i- eac ion model esul s o he p incipal
compounds a e analysed one by one. These esul s a e discussed
compa ing hem wi h he p e ious he modynamic simula ion esul s
and he expe imen al da a. The expe imen al and kine ic esul s com-
pa ison is depic ed in Figu e S9 and absolu e e o o he compounds
conside ed o model alida ion a e gi en in Table S17.
Solid ca bon: a ΔP =0 ba esul s show dec easing endency wi h a
empe a u e inc ease a 700 ◦C and 900 ◦C, espec i ely. The mody-
namic (62.7 w % and 59.9 w %), kine ic (T ble: 39.5 w % and 35.7 w
%) and expe imen al (42.1 w % and 37.1 w %). Howe e , kine ically o
he T c scena io, i could be assumed as p ac ically cons an (38.6 w %
and 38.8 w %), al hough when p essu e d op was inc eased (e.g. ΔP =2
ba ), i sligh ly inc eased (33 w % and 35 w %). This ac could be
explained because a highe p essu e d ops and empe a u es, i seems
he e is no enough wa e apou a ailable (0 w % and 0 w %) o
con inue ca ying ou he ca bon gasi ica ion eac ion (see R-12 in
Table 1) and, he e o e, ca boniza ion eac ions p e ail in he o e all
e ec . Rega ding expe imen al esul s, he p esence o ca bonaceous
deposi ion o e he ixed bed ma e ial was de e mined om he ixed
bed ma e ial cha ac e iza ion esul s and he gene a ed cha in he ba ch
eac o om expe imen al yield calcula ions. Acco ding o he analy ical
esul s and expe imen al obse a ions, compounds such as o ganic a -
oma ics, solid ca bon, sulphu compounds and wa e apou , could
deposi o e he ixed bed ma e ial du ing he py olysis apou s and
gases upg ading p ocess.
Wa e apou : he modynamic (4.5 w % and 0.5 w %), kine ic (T c :
10.3 w % and 0.0 w %; T ble: 12.7 w % and 0.0 w %) and expe imen al
(3.9 w % and 3.7 w %) esul s show dec easing endency wi h a em-
pe a u e inc ease (a 700 ◦C and 900 ◦C, espec i ely) a ΔP =0 ba . I
seems ha ca bon gasi ica ion wi h s eam had signi ican esponsibili y
on wa e consump ion, as a consequence o he solid ca bon p esen in
he eac o s. Indeed, he highe he empe a u e he highe he
consumed solid ca bon and wa e apou quan i y. Howe e , expe i-
men ally no such big di e ence was obse ed be ween 700 ◦C and
900 ◦C compa ed o he simula ion models. This o e es ima ion could be
explained by he s eady s a e model assump ion, as expe imen ally
could happen ha pa o he gene a ed wa e exi s he eac o p io o
in e ac wi h o he compounds.
Naph halene and benzene: in his case he modynamic simula ion and
expe imen al esul s appea consis en as none o hem show naph ha-
lene o benzene as p oduc s, while kine ic model esul s show ha once
gene a ed i migh be ha d o decompose hem due o hei he mal
s abili y. In his ega d, al hough none o hese compounds we e
de ec ed expe imen ally a 700 ◦C and 900 ◦C, u he expe imen al
esea ch could be needed as he ac ion labelled as non- eco e able
condensed apou s (Figu e S2 and Figu e S5) could co espond wi h hose
hea ie compounds.
Hyd ogen gas: he modynamic (3.9 w % and 4.8 w %), kine ic (T c :
0.9 w % and 2.1 w %; T ble: 0.7 w % and 2.1 w %) and expe imen al
(0.4 w % and 2.0 w %) esul s show inc easing endency wi h a em-
pe a u e inc ease (a 700 ◦C and 900 ◦C, espec i ely) a ΔP =0 ba .
Indeed, kine ic p edic ions i e y well he expe imen al da a. Kine i-
cally, he gene a ion o hyd ogen was minimal (0.7 w %) a he lowes
empe a u e (700 ◦C, simula ed as a iable) and ze o p essu e d op,
while maxim (2.2 w %) a he highes empe a u e (900 ◦C, simula ed as
cons an ) and a p essu e d op o 2 ba .
Ca bon monoxide, ca bon dioxide and me hane: he modynamically
CO inc eased while CO
2
and CH
4
dec eased. Howe e , kine ically and
expe imen ally, he wo la e also showed inc easing endency wi h
empe a u e. Kine ic esul s could be explained due o he ac ha no
d y e o ming o me hane eac ion (CH
4
(g) +CO
2
(g) → 2CO (g) +2H
2
(g)) was conside ed in he kine ic model. Whe eas expe imen ally, i
could occu ha in he absence o an adequa e ca alys , his eac ion
could be ha dly occu ing. The e o e, esul s a ΔP =0 ba and a 700 ◦C
and 900 ◦C, espec i ely o CO (g), CO
2
(g) and CH
4
(g) a e consis en
(CO: kine ic T c : 9.9 w % and 15.5 w %; kine ic T ble: 8.7 w % and
16.2 w %; expe imen al: 4.9 w % and 9.9 w %. CO
2
(g): kine ic T c : 4.6
w % and 12.9 w %; kine ic T ble: 2.4 w % and 13.4 w %; expe imen al:
7.7 w % and 9.2 w %. CH
4
(g): kine ic T c : 1.0 w % and 4.8 w %; kine ic
T ble: 5.0 w %; expe imen al: 4.1 w % and 6.2 w %).
Ni ogen gas: he modynamic p edic ion sugges ed he gene a ion o
N
2
(g) while kine ic model did no because none o he decomposi ion
eac ions included i as inal p oduc . In his ega d, based on he con-
sul ed li e a u e, ypical decomposi ion species o ni ogenous a oma ic
compounds seemed o be ela ed o HCN (g), NH
3
(g) and ni ogen ox-
ides (N
2
O (g), NO
x
(g)) among o he s. Abou expe imen al esul s, aces
o N
2
O (g) and NH
3
(g) we e iden i ied (sec ion 2.1), which is consis en
wi h he he mal deg ada ion analysis ca ied ou by P. T ancha d o a
ca bon- ein o ced epoxy lamina e (T ancha d e al., 2017a). While
ega ding HCN (g) compound, u he expe imen al esea ch is equi ed
in o de o de e mine i his compound is being p oduced.
Hyd ogen sulphide gas: he modynamic p edic ion sugges ed he
gene a ion o H
2
S (g) while kine ic model did no because none o he
decomposi ion eac ions included i as inal p oduc , due o he easons
s a ed in sec ion 2.2. Rela ed o expe imen al esul s i seemed ha H
2
S
(g) dec eased wi h inc easing empe a u e (6.5 w % a 700 ◦C and 2.6 w
% a 900 ◦C), which is consis en wi h he highe amoun o sulphu
deposi ed along he ixed bed ma e ial a 900 ◦C compa ed o 700 ◦C
(see Table S7 in SI).
4. Conclusions
An equilib ium and kine ic mul i- eac ion model was de eloped in
Aspen Plus so wa e o p edic he gene a ion o gaseous ou le s eam in
unc ion o p ocess pa ame e s (T, P and basic eac o dimensions) o
he upg ading p ocess o he apou s and gases gene a ed du ing he
A. Se as-Malillos e al.
Chemosphe e 300 (2022) 134499
9
ecycling p ocess o an expi ed epoxy p e-imp egna ed composi e om
he ae onau ics indus y by py olysis. Resul s show ha he sys em ends
o o m ligh e chemical compounds i enough ime and empe a u e is
p o ided. E en hough he modynamically he p ocess is dis a ou ed by
a p essu e inc ease, kine ically, he a e o he eac ion is p opo ional o
he pa ial p essu e o he compounds, and i appea s ha his e ec
p e ails. Kine ic mul i- eac ion model is consis en wi h expe imen al
da a a 700 ◦C and 900 ◦C a ΔP =0 ba ega ding he gene a ion en-
dency (in w %) o he gaseous ou come compounds and wa e apou .
Solid ca bon shows sligh di e ences ha could be explained due o he
as e wa e deple ion. Indeed, H
2
(g), CO (g), CO
2
(g) and CH
4
(g) in-
c ease while wa e apou dec eases. The lowes absolu e e o is ob-
ained o H
2
(g) (0.5 w % a T c and 0.3 w % a T ble), while H
2
O ( )
(6.4 w % a T c and 8.8 w % a T ble) showed he highes . I is
concluded ha he de eloped kine ic model se s he base case scena io
o assis in a subsequen p ocess upscale op imisa ion design, in e ms o
maximising hyd ogen gas p oduc ion, aking in o accoun luid dy-
namics and hea ans e phenomena.
Au ho con ibu ion
Se as-Malillos A.: concep ualiza ion, Me hodology, So wa e, Vali-
da ion, Fo mal analysis, In es iga ion, W i ing – o iginal d a , isual-
isa ion. Acha E.: concep ualiza ion, Me hodology, In es iga ion,
Resou ces, W i ing – e iew & edi ing, Supe ision, P ojec adminis-
a ion, Funding acquisi ion. Lopez-U ionaba enechea A.: concep ual-
iza ion, Me hodology, Resou ces, W i ing – e iew & edi ing,
Supe ision, P ojec adminis a ion, Funding acquisi ion. Pe ez-
Ma inez B.B.: Valida ion, In es iga ion. Caballe o B.M.: Resou ces,
W i ing – e iew & edi ing, Supe ision, P ojec adminis a ion, Fund-
ing acquisi ion.
Decla a ion o compe ing in e es
The au ho s decla e ha hey ha e no known compe ing inancial
in e es s o pe sonal ela ionships ha could ha e appea ed o in luence
he wo k epo ed in his pape .
Acknowledgemen s
The au ho s wan o hank he Minis y o Science and Inno a ion o
Spain (Re . PID2019-110770RB-I00) and he Basque Go e nmen
(Re . KK-2020/00107, ELKARTEK p og am) o he unding o ca y ou
he in es iga ion. The au ho s also hank he inancing g an ed o he
“Sus ainable P ocess Enginee ing” esea ch g oup o he 2016–2021
pe iod (Basque Go e nmen , Re . IT993-16) and a e g a e ul o I˜
naki
Múgica om Su Medioambien e (SUMA Soluciones Medioambien ales,
S.L.) o he echnical suppo p o ided.
Appendix A. Supplemen a y da a
Supplemen a y da a o his a icle can be ound online a h ps://doi.
o g/10.1016/j.chemosphe e.2022.134499.
Re e ences
Abdelouahed, L., Au hie , O., Mau iel, G., Co iou, J.P., Ve die , G., Du ou , A., 2012.
De ailed modeling o biomass gasi ica ion in dual luidized bed eac o s unde aspen
plus. Ene gy Fuel. 26, 3840–3855. h ps://doi.o g/10.1021/e 300411k.
Ahmed, A.M.A., Salmia on, A., Choong, T.S.Y., Wan Azlina, W.A.K.G., 2015. Re iew o
kine ic and equilib ium concep s o biomass a modeling by using Aspen Plus.
Renew. Sus ain. Ene gy Re . 52, 1623–1644. h ps://doi.o g/10.1016/j.
se .2015.07.125.
A ena, U., 2012. P ocess and echnological aspec s o municipal solid was e gasi ica ion.
A e iew. Was e Manag. 32, 625–639. h ps://doi.o g/10.1016/j.
wasman.2011.09.025.
Aspen Technology Inc., 2021. Aspen Plus.
Ba b´
e, P., Ma in, R., Pe in, D., Scacchi, G., 1996. Kine ics and modeling o he he mal
eac ion o p opene a 800 K. Pa I. Pu e p opene. In . J. Chem. Kine . 28, 829–847.
h ps://doi.o g/10.1002/(SICI)1097-4601(1996)28:11<829::AID-KIN5>3.3.CO;2-
O.
Basu, P., 2018. Biomass Gasi ica ion, Py olysis and To e ac ion P ac ical Design and
Theo y, hi d ed. Joe Hay on, London.
Benson, A.M., 1967. Py olysis o p opane in a shock ube. AIChE J. 13, 903–908. h ps://
doi.o g/10.1002/aic.690130517.
B odsky, A.M., Kalinenko, R.A., La o sky, K.P., 1960. 863. The p inciples go e ning
high- empe a u e e hane c acking. J. Chem. Soc. 4443–4454. h ps://doi.o g/
10.1039/j 9600004443.
B uinsma, O.S.L., T omp, P.J.J., Nol ing de, H.J.J.S., Moulijn, J.A., Ins i u e o Chemical
Tehnology (Ams e dam), 1988. Gas phase py olysis o coal- ela ed a oma ic
compounds in a coiled ube low eac o . 2. He e ocyclic compounds, hei benzo
and dibenzo de i a i es. Fuel 67, 334–340. h ps://doi.o g/10.1016/0016-2361(88)
90315-8.
Bueno Lopez, A., Lozano Cas ell´
o, D., Pe ucho Sanchez, F., 2017. WO 2017/021574 A1.
P ocedimien o de ecupe aci´
on de ib as ino g´
anicas a empe a u a ambien e en
ma e iales compues os ib a- esina. pc .
Dama zis, T., Michailos, S., Zabanio ou, A., 2012. Ene ge ic assessmen o a combined
hea and powe in eg a ed biomass gasi ica ion-in e nal combus ion engine sys em
by using Aspen Plus®. Fuel P ocess. Technol. 95, 37–44. h ps://doi.o g/10.1016/j.
up oc.2011.11.010.
De Kam, M.J., Vance Mo ey, R., Ti any, D.G., 2009. Biomass in eg a ed gasi ica ion
combined cycle o hea and powe a e hanol plan s. Ene gy Con e s. Manag. 50,
1682–1690. h ps://doi.o g/10.1016/j.enconman.2009.03.031.
F ancois, J., Abdelouahed, L., Mau iel, G., Pa isson, F., Mi gaux, O., Rogaume, C.,
Rogaume, Y., Feid , M., Du ou , A., 2013. De ailed p ocess modeling o a wood
gasi ica ion combined hea and powe plan . Biomass Bioene gy 51, 68–82. h ps://
doi.o g/10.1016/j.biombioe.2013.01.004.
Gao, N., Chen, C., Magdzia z, A., Zhang, L., Quan, C., 2021. Modeling and simula ion o
pine sawdus gasi ica ion conside ing gas mix u e e lux. J. Anal. Appl. Py olysis
155. h ps://doi.o g/10.1016/j.jaap.2021.105094.
Gas elu O azua, N., 2020. Ka bono Zun zezko Ma e ial Konposa uen Pi olisi Bidezko
Bi ziklapena. Uni e si y o he Basque Coun y (Enginee ing School o Bilbao).
G ange, N., Tadini, P., Che ehouna, K., Gascoin, N., Reynaud, I., Sena e, S., 2018.
De e mina ion o he mophysical p ope ies o ca bon- ein o ced polyme -based
composi es up o 1000 ◦C. The mochim. Ac a 659, 157–165. h ps://doi.o g/
10.1016/j. ca.2017.11.014.
Hale, J., 2012. Boeing 787 om g ound up. Boeing Come . Ae omagazine 9.
Hannula, I., Ku kela, E., 2012. A pa ame ic modelling s udy o p essu ised s eam/O 2-
blown luidised-bed gasi ica ion o wood wi h ca aly ic e o ming. Biomass
Bioene gy 38, 58–67. h ps://doi.o g/10.1016/j.biombioe.2011.02.045.
Le eu e, A., Ga nie , S., Jacquemin, L., Pillain, B., Sonnemann, G., 2017. An icipa ing
in-use s ocks o ca bon ibe ein o ced polyme s and ela ed was e lows gene a ed
by he comme cial ae onau ical sec o un il 2050. Resou . Conse . Recycl. 125,
264–272. h ps://doi.o g/10.1016/j. escon ec.2017.06.023.
Lopez-U ionaba enechea, A., Gas elu, N., Acha, E., Caballe o, B.M., de Ma co, I., 2021.
P oduc ion o hyd ogen- ich gases in he ecycling p ocess o esidual ca bon ibe
ein o ced polyme s by py olysis. Was e Manag. 128, 73–82. h ps://doi.o g/
10.1016/j.wasman.2021.04.044.
L´
opez U ionaba enechea, A., De Ma co Rod íguez, I., Caballe o, B.M., Gas elu
O azua, N., He n´
andez S´
ainz, A., Ad ados L´
opez de Vi˜
nasp e, A., Sola I azabal, J.,
2016. WO 2016/135359 A1. Me hod o T ea ing Vapou s Gene a ed du ing he
P ocess o Reco e ing Ca bon Fib es om Composi es by Py olysis.
Manion, Huie, Le in, Bu gess J , O kin, Tsang, McGi e n, Hudgens, Knyaze , A kinson,
Chai, Te eza, Lin, Allison, Malla d, Wes ley, He on, Hampson, F izzell, 2013. NIST
Chemical Kine ics Da abase. NIST S anda d Re e ence Da abase 17, Ve sion 7.0
(Web Ve sion), Release 1.6.8, Da a e sion 2015.09. Na ional Ins i u e o S anda ds
and Technology, Gai he sbu g, Ma yland, 20899-8320. Web add ess: h ps://
kine ics.nis .go /kine ics/.
Ma sh, G., 2010. Ai bus A350 XWB upda e. Rein o c Plas 54, 20–24. h ps://doi.o g/
10.1016/S0034-3617(10)70212-5.
McKinnon, M.B., Ding, Y., S olia o , S.I., C owley, S., Lyon, R.E., 2017. Py olysis model
o a ca bon ibe /epoxy s uc u al ae ospace composi e. J. Fi e Sci. 35, 36–61.
h ps://doi.o g/10.1177/0734904116679422.
Mu lu, ¨
O.Ç., Zeng, T., 2020. Challenges and oppo uni ies o modeling biomass
gasi ica ion in aspen plus: a e iew. Chem. Eng. Technol. 43, 1674–1689. h ps://
doi.o g/10.1002/cea .202000068.
Nilsson, S., G´
omez-Ba ea, A., Fuen es-Cano, D., Olle o, P., 2012. Gasi ica ion o biomass
and was e in a s aged luidized bed gasi ie : modeling and compa ison wi h one-
s age uni s. Fuel 97, 730–740. h ps://doi.o g/10.1016/j. uel.2012.02.044.
Panopoulos, K.D., F yda, L.E., Ka l, J., Poulou, S., Kaka as, E., 2006. High empe a u e
solid oxide uel cell in eg a ed wi h no el allo he mal biomass gasi ica ion. Pa I:
modelling and easibili y s udy. J. Powe Sou ces 159, 570–585. h ps://doi.o g/
10.1016/j.jpowsou .2005.12.024.
Roda , S., Abanades, S., Couli´
e, J., Flaman , G., 2009. Kine ic modelling o me hane
decomposi ion in a ubula sola eac o . Chem. Eng. J. 146, 120–127. h ps://doi.
o g/10.1016/j.cej.2008.09.008.
Sadhukhan, J., Zhao, Y., Shah, N., B andon, N.P., 2010. Pe o mance analysis o
in eg a ed biomass gasi ica ion uel cell (BGFC) and biomass gasi ica ion combined
cycle (BGCC) sys ems. Chem. Eng. Sci. 65, 1942–1954. h ps://doi.o g/10.1016/j.
ces.2009.11.022.
Salisu, J., Gao, N., Quan, C., 2021. Techno-economic assessmen o Co-gasi ica ion o ice
husk and plas ic was e as an o -g id powe sou ce o small scale ice milling - an
aspen plus model. J. Anal. Appl. Py olysis 158, 105157. h ps://doi.o g/10.1016/j.
jaap.2021.105157.
A. Se as-Malillos e al.
