Assessing the economic and environmental sustainability of bio-olefins: the case of 1,3-butadiene production from bioethanol
Abstract
There is a growing interest in the production of biobutadiene from bioethanol, but it is not possible to conclude from the literature which conversion process, either by one or two reaction steps, is preferred since there is a lack of works comparing their economic and environmental performances. The aim of this work is to perform that comparison so it can serve as a guide for decision making by future investors as well as set future research needs that can improve the technology. Assuming production in Brazil from sugarcane ethanol, technoeconomic and lifecycle assessments were performed from process simulations, and the reliability of the results was studied with uncertainty analyses. The comparison reveals that the one-step process should be preferred because of its better economic and environmental performance. Biobutadiene from ethanol is not cost competitive against naphtha-derived butadiene (probability of positive net present value is 11–17% for one-step process scenarios and 5% for the two-step process scenarios) but producing biobutadiene from ethanol in Brazil leads to significant reductions in emissions of CO2 compared to naphtha-derived butadiene (102–103% and 7.6–52.4% for one- and two-step process scenarios, respectively). A critical issue for the cost competitiveness of biobutadiene is that the CO2 saved can be sold in an international carbon emission trading market. In that case, the one-step processes would be much more favored. Future research is needed concerning the development of highly selective two-step catalysts operating in less energy-demanding conditions so that two-step processes can compete with one-step processes.
Full text
Jou nal o Cleane P oduc ion 374 (2022) 133963
A ailable online 5 Sep embe 2022
0959-6526/© 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/).
Assessing he economic and en i onmen al sus ainabili y o bio-ole ins: The
case o 1,3-bu adiene p oduc ion om bioe hanol
C.E. Cab e a Camacho
a
,
b
, A.L. Villanue a Pe ales
a
,
*
, Be nab´
e Alonso-Fa i˜
nas
a
,
F. Vidal-Ba e o
a
, Ped o Olle o
a
a
Depa amen o de Ingenie ía Química y Ambien al, Escuela T´
ecnica Supe io de Ingenie ía, Uni e sidad de Se illa, Camino de los Descub imien os s/n. 41092 Se illa,
Spain
b
Depa men o Bio echnology and Biomedicine, Technical Uni e si y o Denma k, Søl o s Plads, 2800 Kgs. Lyngby, Denma k
ARTICLE INFO
Handling Edi o : Ka hleen A iso
Keywo ds:
Bioe hanol
Bu adiene
Li ecycle assessmen
Economic assessmen
One-s ep p ocess
Two-s ep p ocess
ABSTRACT
The e is a g owing in e es in he p oduc ion o biobu adiene om bioe hanol, bu i is no possible o conclude
om he li e a u e which con e sion p ocess, ei he by one o wo eac ion s eps, is p e e ed since he e is a lack
o wo ks compa ing hei economic and en i onmen al pe o mances. The aim o his wo k is o pe o m ha
compa ison so i can se e as a guide o decision making by u u e in es o s as well as se u u e esea ch needs
ha can imp o e he echnology. Assuming p oduc ion in B azil om suga cane e hanol, echnoeconomic and
li ecycle assessmen s we e pe o med om p ocess simula ions, and he eliabili y o he esul s was s udied wi h
unce ain y analyses. The compa ison e eals ha he one-s ep p ocess should be p e e ed because o i s be e
economic and en i onmen al pe o mance. Biobu adiene om e hanol is no cos compe i i e agains naph ha-
de i ed bu adiene (p obabili y o posi i e ne p esen alue is 11–17% o one-s ep p ocess scena ios and 5% o
he wo-s ep p ocess scena ios) bu p oducing biobu adiene om e hanol in B azil leads o signi ican educ ions
in emissions o CO
2
compa ed o naph ha-de i ed bu adiene (102–103% and 7.6–52.4% o one- and wo-s ep
p ocess scena ios, espec i ely). A c i ical issue o he cos compe i i eness o biobu adiene is ha he CO
2
sa ed can be sold in an in e na ional ca bon emission ading ma ke . In ha case, he one-s ep p ocesses would
be much mo e a o ed. Fu u e esea ch is needed conce ning he de elopmen o highly selec i e wo-s ep
ca alys s ope a ing in less ene gy-demanding condi ions so ha wo-s ep p ocesses can compe e wi h one-s ep
p ocesses.
1. In oduc ion
In he pe ochemical sec o , 1,3-bu adiene (1,3-BD) is one o he
mos impo an ole ins, along wi h e hylene and p opylene, and i is a
building block o p oducing elas ome s, esins, and syn he ic ubbe s
wi h a la ge a ie y o uses such as i es o he au omo i e indus y,
sealan s, clo hing, and plas ic casings o elec onic p oduc s (S a is a,
2017). Mos 1,3-BD (95%) is p oduced as a byp oduc o e hylene
manu ac u ing h ough naph ha s eam c acking, while a small ac ion
o 1,3-BD is p oduced delibe a ely by he dehyd ogena ion o n-bu ene
and n-bu ane (Angelici e al., 2013).
The shi om oil-de i ed eeds ocks (Ren e al., 2008) o enewable
eeds ocks o 1,3-BD p oduc ion is appealing o educe he en i on-
men al impac (Panahi e al., 2019). This has b ough a en ion again o
he ca aly ic p oduc ion o 1,3-BD om bioe hanol, a echnology ha
was abandoned in he 1960s a e cheape naph ha-de i ed 1,3-BD
became a ailable. Two indus ial p oduc ion p ocesses we e de el-
oped: he one-s ep and he wo-s ep eac ion p ocess. In bo h p ocesses,
he same eac ion pa hway is ollowed o he con e sion o bioe hanol
o 1,3-BD, bu in he wo-s ep p ocess, he con e sion is pe o med in
wo sepa a e s ages: in he i s eac o , ace aldehyde is p oduced by
e hanol dehyd ogena ion (Eq. (1)), and in he second eac o , he
mix u e o ace aldehyde and e hanol is con e ed in o 1,3-BD (Eq. (2)).
The o e all eac ion (Eq. (3)) indica es a s oichiome ic mass yield o
0.587 kg 1,3-BD pe kg o e hanol. Lowe 1,3-BD yields han he s oi-
chiome ic yield a e ac ually achie ed since side p oduc s a e o med
(Pomalaza e al., 2020), o example die hyl e he , e hylene, n-bu anol,
bu enes, and hea y p oduc s (C
6+
).
C
2
H
6
O→C
2
H
4
O +H
2
Eq. 1
C
2
H
6
O +C
2
H
4
O →C
4
H
6
+2H
2
O Eq. 2
* Co esponding au ho .
E-mail add ess: [email p o ec ed] (A.L. Villanue a Pe ales).
Con en s lis s a ailable a ScienceDi ec
Jou nal o Cleane P oduc ion
jou nal homepage: www.else ie .com/loca e/jclep o
h ps://doi.o g/10.1016/j.jclep o.2022.133963
Recei ed 14 Ma ch 2022; Recei ed in e ised o m 20 Augus 2022; Accep ed 30 Augus 2022
Jou nal o Cleane P oduc ion 374 (2022) 133963
2
2 C
2
H
6
O → C
4
H
6
+2H
2
O +H
2
Eq. 3
The con e sion in wo s eps allows achie ing high selec i i y by
using a ailo ed ca alys and choosing op imal eac ion condi ions o
each con e sion s ep. I canno be concluded om he li e a u e
(Rodge s e al., 2022) whe he he wo-s ep p ocess should be p e e ed
o e he one-s ep p ocess. The eason is ha all published wo ks excep
one (Cespi e al., 2016) ha e s udied only one o he wo p ocesses, and i
is di icul o compa e hei esul s since hei assump ions di e sub-
s an ially (Table 1) ega ding he ollowing: (i) s a ing aw ma e ial (C
6
suga s, ce eal c ops, biomass esidues, and black liquo ); (ii) p ocess
design decisions such as inal use o byp oduc s (as uel in he plan o
sold as chemicals) and powe and hea in eg a ion o he plan (as a
s and-alone plan wi h i s own CHP sys em o as a p ocessing a ea wi hin
a bio e ine y wi h a sha ed CHP sys em); (iii) plan loca ion and p o-
duc ion capaci y ( anging om 24 o 200 k /y); and (i ) da abases and
me hodologies o echnoeconomic and li ecycle assessmen s (TEA and
LCA, espec i ely).
Only one s udy (Cespi e al., 2016) compa ed bo h p ocesses, and he
au ho s concluded ha he wo-s ep p ocess was less economically and
en i onmen ally sus ainable han he one-s ep p ocess. This conclusion
mus be in e p e ed cau iously because he au ho s decided ha un-
con e ed eac an s om he eac o we e bu n as uels a he han
eco e ed and ecycled o he eac o as usual. This dis a o ed he
wo-s ep p ocess, whose consump ion o e hanol pe on o 1,3-BD was
wice he consump ion o he one-s ep p ocess. This is coun e in ui i e
since in hei s udy he wo-s ep ca alys was mo e selec i e o 1,3-BD.
As e hanol was he main con ibu o o p oduc ion cos s and en i on-
men al bu dens, he wo-s ep p ocess was ound less sus ainable.
The no el y o he p esen wo k is ha i igo ously compa es he
en i onmen al and economic pe o mances o one and wo-s ep p o-
cesses by applying he same me hodology and assump ions in he TEA
and LCA o a ai compa ison, wi h he aim o ill he gap iden i ied in
he li e a u e and p o ide insigh in o wha p ocess should be p e e ed.
This is he i s ime ha a s a is ical compa ison o he economic and
en i onmen al pe o mances o he one- and wo-s ep p ocesses has
been ca ied ou . The indings o his wo k can se e as a guide o he
decision making o u u e in es o s by e ealing he ela i e ad an ages
and disad an ages o each p ocess, and hey se u u e esea ch needs so
his echnology can again each comme cial s a us.
2. Ma e ials and me hods
Fou scena ios we e s udied o compa e he p oduc ion o 1,3-BD
om bioe hanol by one (B1 and B2 scena ios) and wo eac ion s eps
(B3 and B4 scena ios). In he B1 and B3 scena ios he one- and wo-s ep
p ocesses, espec i ely, we e concep ually designed om he pe o -
mance da a o one- (Cabello Gonz´
alez e al., 2019) and wo-s ep (Cabello
Gonz´
alez e al., 2022) ca alys s wi h high yield selec ed om he li e -
a u e. Pe o mance da a we e ob ained om ca aly ic es s in ou lab-
o a o y in which he e ec o eac ion p oduc s and e hanol impu i ies
in he eac o eed we e accoun ed o o model he eac o pe o mance
mo e ealis ically and achie e accu a e TEA and LCA esul s. To e al-
ua e he impac o ca alys selec i i y in he TEA and LCA, an addi ional
scena io was conside ed o each p ocess (B2 and B4 o he one- and
wo-s ep p ocess, espec i ely) by using pe o mance da a o o he
highly selec i e ca alys s om he li e a u e (Sec ion 2.2). In all sce-
na ios he p ocesses we e designed o a p oduc ion capaci y o 200 k /y
o 1,3-BD. The simula ion o he p ocesses p o ided he mass and ene gy
balances o he TEA and LCA. In he case o he p oduc ion p ocess o 1,
3-BD om naph ha, he buil -in model in he Ecoin en (V3) da abase
was used o ex ac ha in o ma ion o he LCA (Ecoin en Cen e,
2018).
The design and simula ion o he one-s ep p ocess was p esen ed in a
p e ious wo k by Cab e a Camacho e al. (2020) om which he ene gy
and mass balances o scena ios B1 and B2 we e aken. The design o he
wo-s ep p ocess (Sec ion 2.1) and he mass and ene gy balance o sce-
na ios B3 and B4 (Sec ion 3.1) a e p esen ed in his wo k. The me h-
odologies and assump ions o he TEA and LCA applied o bo h p ocesses
a e desc ibed in Sec ions 2.3 and 2.4. S ochas ic analyses we e ca ied
ou in he TEA and LCA o assess he impac o unce ain y in he inpu
da a ega ding he eliabili y o he esul s. Fo he s udy, B azil was
chosen as he plan loca ion since based on ou p e ious wo k on he
one-s ep p ocess, his loca ion esul ed in he bes en i onmen al pe -
o mance compa ed o o he majo e hanol-p oduce egions (Eu ope
and he US).
2.1. Desc ip ion o he wo-s ep p ocess
The hie a chical me hod o Douglas (Dimian, 2003) was ollowed o
he concep ual design o he o e all p ocess, while he me hod o Ba -
nicki and Fai (Ba nicki and Fai , 1990, 1992) was speci ically ollowed
o he design o he sepa a ion sec ion. The plan is assumed o be s and
alone, bu i is adjacen o a pe ochemical complex o which i can sell
he p oduced 1,3-BD and cop oduc s. A simpli ied block diag am o he
esul ing wo-s ep p ocess is shown in Fig. 1, and i is b ie ly desc ibed in
his subsec ion. A mo e de ailed low diag am and desc ip ion o he
p ocess a e p o ided in Appendix B.
An e hanol ecycle s eam ha has been pu i ied bu s ill con ains
mainly wa e (7.5 w %) as impu i y is apo ized and ed o he dehy-
d ogena ion u nace eac o ( eac ion and sepa a ion 1, Fig. 1). The
condensable species (mos ly ace aldehyde, wa e , and un eac ed
e hanol) in he eac o e luen a e sepa a ed om hyd ogen based on a
la ge di e ence in ola ili y. Hyd ogen wi h high pu i y (99.9 mol%) is
eco e ed and sold as a byp oduc . The condensable ac ion om he
dehyd ogena ion eac o e luen is mixed wi h an ace aldehyde ecycle
s eam, esul ing in an aqueous e hanol/ace aldehyde mix u e (7.5 w %
wa e ) ha cons i u es he eed o he second u nace eac o . The
condensable ac ion (un eac ed e hanol and ace aldehyde, wa e ,
ace one, bu anal, die hyl e he , and hea y compounds) and gases (bu-
enes, p opylene, e hylene, and 1,3-BD) om he second eac o a e
sepa a ed based on la ge di e ences in ola ili y. The s eam o gases is
sc ubbed wi h esh e hanol (7.5 w % wa e , e hanol eed o eac ion
and sepa a ion 2, Fig. 1) o eco e 1,3-BD. The e hanol con aining he
eco e ed 1,3-BD is mixed wi h he condensable ac ion o he second
eac o , which con ains a subs an ial amoun o 1,3-BD. This mix u e is
sepa a ed by dis illa ion o ob ain c ude bu adiene wi h a ela i ely high
Abb e ia ions
1,3-BD 1,3-bu adiene
CAPEX Capi al expendi u e
CED Cumula i e ene gy demand
CHP Combined hea and powe
DMF Dime hyl o mamide
ETB E hanol o bu adiene
GWP100 Global wa ming po en ial o e 100 yea s
HEN Hea exchange ne wo k
HPS High-p essu e s eam
IRR In e nal a e o e u n
LCA Li ecycle assessmen
LPS Low-p essu e s eam
MBSP Minimum bu adiene selling p ice
MPS Medium-p essu e s eam
NPV Ne p esen alue
OPEX Ope a ing expendi u e
ROI Re u n on in es men
TEA Techno-economic assessmen
WC Wa e consump ion
C.E. Cab e a Camacho e al.
Jou nal o Cleane P oduc ion 374 (2022) 133963
3
Table 1
Published wo ks on TEA and LCA o he e hanol- o-bu adiene (ETB) p ocesses.
Wo k Rou e S a ing aw ma e ial Plan
loca ion
P ocess design and modelling TEA LCA Main conclusions
Cespi e al.
(2016)
One-
and
wo-
s ep
Anhyd ous i s -
gene a ion e hanol om
egional c ops
US, Eu ope,
B azil
Ca alys pe o mance da a om pa en s.
Simple p ocess modelling by s oichiome ic
balances. Combus ion o uncon e ed
eac an s and byp oduc s.
De e minis ic. Rela i e economic
assessmen based on economic
index ( a io o BD p ice/cos o
aw ma e ial and u ili ies).
P oduc ion capaci y unknown.
De e minis ic. C adle o ga e. GWP,
WC, and CED impac s. Compa ison
wi h naph ha-c acking ou e. LCA
modelling wi h SimaP o so wa e
and ecoin en da abase.
Be e en i onmen al pe o mance o
one-s ep p ocess han wo-s ep p ocess
and naph ha-c acking ou e o all
plan loca ions.
Shylesh
e al.
(2016)
One-
s ep
Anhyd ous e hanol om
he US (co n-g ain o
co n-s o e e hanol) o
impo ed om B azil
(suga cane e hanol)
US Ca alys pe o mance da a om ca aly ic
es s by au ho s. P ocess modelling in Aspen
Plus. Plan as add-on o exis ing bioe hanol
plan . Byp oduc s a e sold.
Economic assessmen no
pe o med. P oduc ion capaci y:
200 k /y bu adiene.
De e minis ic. C adle o g a e. Only
GWP impac . Compa ison wi h
naph ha-c acking ou e. LCA
so wa e no disclosed. Emission
ac o s aken om li e a u e.
Only bu adiene de i ed om co n-
s o e e hanol o suga cane e hanol
has lowe GWP han naph ha-de i ed
bu adiene.
Fa zad e al.
(2017)
Two-
s ep
Azeo opic e hanol om
suga cane bagasse and
ha es ing esidue
Sou h A ica Ca alys pe o mance da a om indus ial
ca alys s. P ocess modelling in Aspen Plus.
Simula ion o bioe hanol and bu adiene
plan wi h common CHP sys em. Two
scena ios: (i) sel -su icien ene gy plan and
(ii) impo o coal. Byp oduc s a e bu n .
S ochas ic. P o i abili y calcula ed
by cash low analysis. P oduc ion
capaci y: 30–37 k /y bu adiene.
De e minis ic. C adle o ga e. GWP,
CED, and 10 mo e impac s. GWP
compa ed o naph ha-c acking
ou e. LCA modelling wi h SimaP o
so wa e.
Ve y low p obabili y o be p o i able.
Signi ican GHG educ ion compa ed
o naph ha-de i ed bu adiene.
Moncada
e al.
(2018)
One-
s ep
Azeo opic e hanol om
C6 suga s
Ne he lands Ca alys pe o mance da a om expe s.
P ocess modelling in Aspen Plus.
Bio e ine y co-p oducing bu adiene and
cap olac am wi h common CHP sys em.
Byp oduc s a e sold.
De e minis ic. P o i abili y
calcula ed by cash low analysis.
P oduc ion capaci y: 24 k /y
bu adiene.
No pe o med. Bu adiene p oduc ion om C6 suga s
is no p o i able o a e age ma ke
p ices.
Cab e a
Camacho
e al.
(2020)
One-
s ep
Azeo opic i s -
gene a ion e hanol om
egional c ops
US, Eu ope,
B azil
Ca alys pe o mance model de eloped
om ca aly ic es s by au ho s o accoun
o impu i ies in e hanol. P ocess modelling
in Aspen Plus. Two scena ios: (i) low and (ii)
high ca alys selec i i y. S and-alone plan .
Byp oduc s a e sold.
De e minis ic. Cos calcula ions
o a gene ic loca ion based on
in e na ional p ices. P o i abili y
calcula ed by cash low analysis.
P oduc ion capaci y: 200 k /y
bu adiene.
S ochas ic. C adle o ga e. GWP,
WC, and CED impac s. Compa ison
wi h naph ha-c acking ou e. LCA
modelling wi h SimaP o so wa e,
ecoin en da abase, and li e a u e
su ey.
E hanol-de i ed bu adiene is no
p o i able o a e age ma ke p ices.
B azil is he bes loca ion in e ms o
GWP and WC, and he only one wi h
GWP educ ion ela i e o naph ha-
de i ed bu adiene.
Dimian
e al.
(2021)
Two-
s ep
Azeo opic e hanol.
Feeds ock o e hanol
p oduc ion unknown.
– Ca alys kine ic models om li e a u e o
i s and second s ep. P ocess modelling in
Aspen Plus. S and-alone plan . Al e na i e
sepa a ion ain. Byp oduc s a e sold.
De e minis ic. P o i abili y
calcula ed by cash low analysis.
P oduc ion capaci y: 91 k /y
bu adiene.
No pe o med E hanol-de i ed bu adiene is
p o i able o a e age ma ke p ices.
Rodge s
e al.
(2022)
One-
s ep
E hanol om syngas
(supe c i ical wa e
gasi ica ion o woody
biomass o black liquo )
China Ca alys pe o mance da a o second s ep
aken om he li e a u e. P ocess modelling
in Aspen Hysys. Two ou es o e hanol: (i)
e men a ion o syngas om black liquo
and (ii) ca aly ic con e sion o syngas om
woody biomass. Common CHP sys em o
e hanol and bu adiene p oduc ion.
Byp oduc s a e sold.
S ochas ic. P o i abili y calcula ed
by cash low analysis. P oduc ion
capaci y: 9–31 k /y bu adiene.
De e minis ic. C adle o ga e. Only
GWP impac . Compa ison wi h
naph ha-c acking ou e. LCA
modelling wi h SimaP o so wa e.
Only syngas e men a ion ou e can be
p o i able bu wi h low p obabili y.
Bo h ou es wi h GWP educ ion
ela i e o naph ha-de i ed bu adiene.
This wo k One-
and
wo-
s ep
Azeo opic i s -
gene a ion e hanol om
suga cane
B azil Ca alys pe o mance om ca aly ic es s by
au ho s whe e impu i ies in e hanol a e
accoun ed o . P ocess modelling in Aspen
Plus. Two scena ios o each ou e: (i) low
and (ii) high ca alys selec i i y. S and-
alone plan . Byp oduc s a e sold.
S ochas ic. Cos calcula ions o
B azil. P o i abili y calcula ed by
cash low analysis. P oduc ion
capaci y: 200 k /y bu adiene.
S ochas ic. C adle o ga e. GWP,
WC, and CED impac s. Compa ison
wi h naph ha-c acking ou e. LCA
modelling wi h SimaP o so wa e,
ecoin en da abase, and li e a u e
su ey.
One-s ep p ocess has highe
p obabili y han wo-s ep p ocess o be
p o i able. Expec ed en i onmen al
impac o wo-s ep p ocess is highe
han one-s ep p ocess in all impac
ca ego ies.
C.E. Cab e a Camacho e al.
Jou nal o Cleane P oduc ion 374 (2022) 133963
4
ace aldehyde con en and a c ude e hanol/ace aldehyde mix u e ( e-
ac ion and sepa a ion 2, Fig. 1).
In he pu i ica ion ain o 1,3-BD, c ude bu adiene is sc ubbed wi h
wa e (pu i ied wa e ecycle o eac ion and sepa a ion 2, Fig. 1) o
emo e ace aldehyde, and hen bu enes a e sepa a ed by ex ac i e
dis illa ion wi h dime hyl o mamide (DMF), esul ing in highly pu e
1,3-BD (99.9 w %; bu adiene pu i ica ion, Fig. 1). The c ude e hanol/
ace aldehyde mix u e con ains hea y compounds ha mus i s be
emo ed because o he wise hey would hinde he la e eco e y o
e hanol and ace aldehyde by dis illa ion (Das illung e al., 2016). These
hea y compounds a e ex ac ed using an o ganic sol en ( emo al o
hea y compounds, Fig. 1). The emo ed hea y compounds a e bu n in
he plan as uel, while he clean ace aldehyde/e hanol mix u e is sen o
a ain o dis illa ion columns whe e wa e and emaining byp oduc s
a e emo ed om ace aldehyde and e hanol (sepa a ion o pola com-
pounds, Fig. 1). Se e al s eams esul om all hese sepa a ions: i) a
s eam o pu i ied e hanol (7.5 w % wa e ), a ac ion o which is ed o
he i s eac o and he es used as sol en in he sc ubbe s; ii) a pu-
i ied ace aldehyde s eam ha is ed o he second u nace eac o ; iii) a
mix u e o mino byp oduc s ha is bu n in he plan as uel ( esidual
uel s eam); i ) a pu i ied wa e s eam ha is ecycled and used as a
mass sepa a ion agen ; and ) a wa e s eam pollu ed wi h o ganic
byp oduc s ha is ea ed in a Fen on p ocess (WWT; Iboukhoule e al.,
2016).
A s eam om he sepa a ion ain a e he second eac o comp ises
p opylene and e hylene (gases) ha a e sepa a ed (sepa a ion o ligh
gases, Fig. 1). Reco e ed e hylene, p opylene, and bu enes a e sold as
cop oduc s.
2.2. Modelling o he wo-s ep p ocess
The modelling o he wo-s ep p ocess was ca ied ou wi h Aspen
Plus V8.8. Fo he simula ion o each sec ion o he plan , sui able
he modynamic me hods we e selec ed. The SRK equa ion o s a e was
chosen o he sepa a ion o e hylene and p opylene (sepa a ion o ligh
gases), he Wilson me hod o ex ac i e dis illa ion wi h DMF in he
bu adiene sepa a ion sec ion, and he NRTL me hod o simula ion o
he es o he plan . Valida ion and calib a ion o hese he modynamic
models we e ca ied ou in a p e ious wo k (Cab e a Camacho e al.,
2020). Rigo ous models in Aspen Plus based on equilib ium s ages
(RadF ac and Ex ac models) we e used o simula e sepa a ion p ocess
uni s such as dis illa ion columns, abso be s, and ex ac ion columns.
Fo modelling he dehyd ogena ion eac o , expe imen al da a om
e hanol dehyd ogena ion (Eq. (1)) o e a con en ional Cu/SiO
2
ca alys
(Klein e al., 2016) we e used o es ima e he eac ion condi ions o a
a ge e hanol con e sion ( oo no e g, Table 2), so he mixing o he
ace aldehyde and e hanol om he i s eac o and he ace aldehyde
ecycled o he second eac o esul ed in he desi ed e hanol/ace-
aldehyde a io in he eed o he second eac o . The dehyd ogena ion
eac o was modelled in Aspen Plus as a s ochiome ic eac o (RS oic
model) conside ing only he e hanol dehyd ogena ion eac ion (Eq. (1))
since no side eac ions we e obse ed based on he expe imen al da a
(Klein e al., 2016).
The ca aly ic eac o o he second eac ion s ep was modelled in
Aspen Plus as a yield eac o (RYield model) based on he expe imen al
pe -pass con e sion o ace aldehyde and e hanol and ca bon selec i i y
o p oduc s o he wo selec ed ca alys s. In scena io B3, he pe o -
mance da a o a Ta/SBA-15 ca alys om expe imen s conduc ed in ou
labo a o y wi h aqueous e hanol/ace aldehyde mix u es as eed we e
employed (Cabello Gonz´
alez e al., 2022). When choosing he eac ion
condi ions o his ca alys (space eloci y, eac ion empe a u e, e ha-
nol/ace aldehyde mole a io, and wa e mass con en in he eac o
eed), wo design decisions we e made: (i) he wa e con en in he
ace aldehyde/e hanol mix u e should be 7.5 w % o a oid he cos ly
sepa a ion o he wa e -e hanol azeo ope (Cab e a Camacho e al.,
2020) and (ii) he e hanol/ace aldehyde mole a io should be low
Fig. 1. Block diag am o he wo-s ep p ocess.
C.E. Cab e a Camacho e al.
Jou nal o Cleane P oduc ion 374 (2022) 133963
5
enough o a oid a la ge e hanol ecycle s eam in he p ocess bu high
enough o achie e a high selec i i y o 1,3-BD in he second eac ion
s ep. In Table 2, he selec ed ope a ing condi ions a e shown o which
he highes expe imen al selec i i y o 1,3-BD was ob ained (~77%) a a
low e hanol/ace aldehyde mole a io (1.7 compa ed o 2.7 o he in-
dus ial p ocess; Ky iienko e al., 2016). In scena io B4, he impac o
using a ca alys mo e selec i e o 1,3-BD on he pe o mance o he
p ocess was examined. One o he mos selec i e wo-s ep ca alys s e-
po ed in he li e a u e, a TaSiBEA zeoli e (Ky iienko e al., 2016), was
selec ed (Table 2) ha exhibi s highe 1,3-BD selec i i y han he
Ta-SBA-15 ca alys (~87% compa ed o ~77%) bu a he expense o a
highe e hanol/ace aldehyde mole a io (2.2 compa ed o 1.7).
Fo he sake o la e compa ison be ween he one- and wo-s ep
p ocesses, he pe o mance o he ca alys s conside ed in he one-s ep
p ocess scena ios (B1 and B2) a e shown in Table 2.
As in ou p e ious s udy on he one-s ep p ocess, he p ocess plan
was hea in eg a ed. The hea exchange ne wo k (HEN) was op imally
designed using he ool Aspen Ene gy Analysis, which conside s he
capi al and ope a ing cos s associa ed o he HEN. App op ia e hea ing
and cooling u ili ies we e chosen o sa is y he ene gy demand o he
plan a di e en empe a u e le els. In he u nace eac o s, na u al gas
was bu n o p o ide hea . In he Suppo ing In o ma ion sec ion (Ap-
pendix C), mo e de ails a e p o ided on he hea in eg a ion and
esul ing HEN o he wo-s ep p ocess.
2.3. Economic assessmen
The me hodology o he economic calcula ions is desc ibed in ou
p e ious wo k on he one-s ep p ocess (Cab e a Camacho e al., 2020),
and o he sake o he eade is p esen ed again in Appendix A. The
minimum bu adiene selling p ice (MBSP,
€
/ ) was chosen as an eco-
nomic indica o o compa ing he economics o each scena io, and i is
de ined as he selling p ice o 1,3-bu adiene o eco e he capi al in-
es men wi h a 10% a e o e u n a he end o he plan li e. The MBSP
was calcula ed om a cash low analysis based on he economic as-
sump ions shown in Table A4. The capi al expendi u es (CAPEX;
Table A1), ope a ing expendi u es (OPEX; Table A2), and e enue om
byp oduc s (Table A3) we e calcula ed using economic in o ma ion o
B azil as he plan loca ion whene e possible. A sensi i i y analysis was
pe o med using a o nado cha o de e mine wha economic pa ame-
e s had he la ges in luence on he MBSP o each scena io. These
economic pa ame e s we e he p ice o e hanol and he cos o na u al
gas. An unce ain y analysis using he Mon e Ca lo me hod implemen ed
in a sel -made cos sp eadshee was pe o med by simul aneously sam-
pling he ma ke p ice o bioe hanol, na u al gas, and bu adiene om
his o ical da a o de e mine he p obabili y o each scena io being
p o i able (ne p esen alue [NPV] >0). Fo ha pu pose, his o ical
da a o he ma ke p ice o azeo opic suga cane bioe hanol (CEPEA,
2022) and indus ial na u al gas (Agˆ
encia Nacional do Pe ´
oleo, n.d.;
COGEN.com, n.d.) in B azil as well as he in e na ional ma ke p ice o
bu adiene (IHS Ma ki , 2019; Rodge s e al., 2022) we e used. These
his o ical da a a e shown in Appendix D. The Mon e Ca lo simula ion
was un 9000 imes o each scena io o ensu e a ce ain s a is ical ac-
cu acy in he es ima ion o he NPV dis ibu ion (Appendix D).
2.4. Li ecycle assessmen
The 100-yea global wa ming po en ial (GWP), wa e consump ion
(WC), and cumula i e ene gy demand (CED) we e assessed in he p e-
sen pape o 1,3-BD p oduc ion om suga cane bioe hanol in B azil ia
he wo-s ep p ocess (scena ios B3 and B4) and compa ed wi h hose
en i onmen al impac s p e iously ob ained o he one-s ep p ocess in
B azil (scena ios B1 and B2) and he naph ha-c acking p ocess (Cab e a
Camacho e al., 2020). The GWP and WC impac ca ego ies we e
selec ed because igh ing clima e wa ming is one o he main goals when
adop ing bio-based indus ies, and wa e is a aluable and sca ce na u al
esou ce. The CED impac ca ego y was selec ed because i is a use ul
indica o o he o e all en i onmen al impac o a p oduc (Huijb eg s
e al., 2006).
Al hough he me hodology o he LCA was ho oughly desc ibed in
ou p e ious s udy on he one-s ep p ocess (Cab e a Camacho e al.,
2020), a summa y is p o ided o he sake o he eade . Following he
guidelines o he ISO 14040/44 s anda ds (ISO, 2006a, 2006b), an
a ibu ional LCA was pe o med using SimaP o so wa e e sion 8.5.2.0
(P ´
e Consul an s, 2017), he Ecoin en (V3) da abase (Ecoin en Cen e,
2018), he ReCiPe 2016 midpoin cha ac e iza ion me hod (V 1.02), and
he CED me hod (V 1.10) o li ecycle modelling, backg ound da a o
in en o y, en i onmen al impac , and CED calcula ion. One o p o-
duced 1,3-BD was chosen as he unc ional uni . A c adle- o-ga e
app oach was applied. Backg ound in en o y da a om he Ecoin en
(V3) da abase (Ecoin en Cen e, 2018) we e used o hose consum-
ables and u ili ies needed in he 1,3-BD p oduc ion p ocess, including
he e hanol supply chain. Fo eg ound in en o y da a we e sou ced om
he ene gy and mass balances om he simula ions ( epo ed in Cab e a
Camacho e al., 2020, o cases B1 and B2 and in Table A7 o his
manusc ip o scena ios B3 and B4). In he case o he GWP, he end o
li e o 1,3-BD was conside ed (Ins i u e o Sc ap Recycling Indus ies
[ISRI], 2009) as sugges ed by Shylesh e al. (2016). The backg ound da a
o p oduce 1,3-BD om naph ha we e aken om he Ecoin en V3
da abase (Ecoin en Cen e, 2018).
Table 2
Ca alys ope a ing condi ions and pe o mance o each scena io.
P ocess One-s ep Two-s ep
Scena io B1
B2
B3
g
B4
g
Ca alys H –Zn H –Zn Ta-SBA-15 TaSiBEA
Ope a ing condi ions
Tempe a u e (◦C) 380 360 350 350
Wa e in eac o eed (w %) 7.5 7.5 7.5 7.5
h
WHSV
a
(1/h) 1.12 0.64 1.63 0.79
E hanol con e sion (%) 81.9 87.1 49.3 38.6
b
Ace aldehyde con e sion (%) – – 70.8 52.4
b
E hanol/ace aldehyde (mole a io) – – 1.7 2.2
Ca bon selec i i y (%)
E hylene 9.9 14.6 3.2 3.0
P opylene 2.9 3.3 2.0 1.0
1-bu ene 2.7 2.1 1.5 2.0
e
1,3-BD 58.8 69.5 77.4 87.5
2- ans-bu ene 2.7 2.1 1.5 2.0
e
Ace aldehyde 8.1 4.1 – –
Die hyl e he 1.3 2.6 2.5 1.0
e
Ace one 1.4 0.5 2.4
d
1.0
e
N-bu anol 0.2 0.1 – –
N-bu anal 3.6 1.1 3.6
d
1.0
e
Hea y compounds 8.4 0.0 5.9
c
1.5
c,e
1,3-BD ca bon yield (%) 48.2 60.5 44.3 37.5
a
Weigh hou ly space eloci y (WHSV, 1/h) is de ined as he a io o mass
low o ace aldehyde and/o e hanol (kg/h)/load o ca alys (kg).
b
Pe sonal communica ion wi h D . Ky iienko (June 15, 2019).
c
Hea y compounds modelled as die hoxye hane.
d
Nume ous oxygena ed compounds obse ed in he expe imen s a e lumped
in o n-bu anal and ace one.
e
The “o he s” ac ion epo ed by Ky iienko e al. (2016) was assumed o be a
mix u e o bu enes and die hyl e he , while he c o onaldehyde ac ion was
assumed o be a mix u e o oxygena ed compounds (ace one and n-bu anal) and
hea y compounds.
F om Cab e a Camacho e al. (2020)
g
Es ima ed ope a ion condi ions o he dehyd ogena ion eac o a e T =
225 ◦C, P =1.5 ba , and WSHV~0.237 1/h o a a ge e hanol con e sion o
44.4% and 42.1% o B3 and B4 scena ios, espec i ely.
h
I is assumed ha ca alys pe o mance is no a ec ed by wa e a 350 ◦C, as
obse ed o he Ta-SBA-15 ca alys .
C.E. Cab e a Camacho e al.
Jou nal o Cleane P oduc ion 374 (2022) 133963
6
3. Resul s and discussion
3.1. O e all mass and ene gy balances
The o e all mass and ene gy balances o B3 and B4 scena ios along
wi h hose o B1 and B2 a e shown in Table 3 o compa ison. P ocess
s eam ables o scena ios B3 and B4 can be ound in Appendix B. As
expec ed, he o e all 1,3-BD p oduc i i y (B1 =0.334; B2 =0.400; B3
=0.443; B4 =0.508 1,3-BD/ anhyd ous e hanol) inc eases wi h
ca alys selec i i y o 1,3-BD. A la ge 1,3-BD p oduc i i y means a
lowe amoun o aw e hanol and a lowe gene a ion o aluable
byp oduc s such as e hylene, p opylene, and bu enes. A dec ease in he
gene a ion o esidual uel s eams (comp ised by ligh oxygena es and
hea y compounds) om scena io B1 o B4 would also be expec ed, bu i
inc eases om scena io B2 o B3 because o he g ea e agg ega e
selec i i y o ligh oxygena es and hea y compounds o scena io B3
(14.4%) compa ed o scena io B2 (8.4%; Table 2). Al hough he gen-
e a ion o hese esidual uel s eams is unwan ed, hey help educe he
hea ing u ili y consump ion o he p ocess by bu ning hem in a s eam
boile .
The ene gy balances show ha he o al consump ion o hea ing and
cooling u ili ies o he wo-s ep p ocess scena ios is highe han ha o
he one-s ep p ocess scena ios (B1 =29.1; B2 =25.5; B3 =51.7; B4 =
59.3 GJ/ 1,3-BD). The main eason o his is he ope a ion wi h a la ge
excess o e hanol in he second eac ion s ep o he wo-s ep p ocess,
which en ails a la ge consump ion o na u al gas in he u nace eac o s
and g ea e hea ing (MPS and LPS) and cooling wa e demands in he
sepa a ion ain o eco e and ecycle e hanol. These e ec s p e ail
o e he educ ion in na u al gas consump ion o he u nace eac o s
and u ili y demand o he sepa a ion ain due o he lowe quan i y o
aw e hanol o be p ocessed and byp oduc s o be sepa a ed, espec-
i ely. The g ea e 1,3-BD p oduc i i y o he wo-s ep p ocesses is a he
expense o highe ene gy consump ion han he one-s ep p ocesses. Fo
wo-s ep p ocesses, a highe o e all 1,3-BD p oduc i i y migh imply a
la ge o al ene gy-speci ic consump ion, as obse ed by compa ing
scena ios B3 and B4 (Table 3). The la ge 1,3-BD selec i i y o scena io
B4 is achie ed a he expense o ope a ing a a la ge e hanol/
ace aldehyde a io han scena io B3. In scena ios B3 o B4, he opposing
e ec s o 1,3-BD selec i i y and he e hanol/ace aldehyde a io on
u ili y consump ion coun e balance each o he , and he lowe ene gy
consump ion o scena io B3 wi h espec o scena io B4 is explained by
he la ge amoun o s eam gene a ed om he ene ge ic alo iza ion o
he esidual uel s eams.
To summa ize, he o e all 1,3-BD p oduc i i y o wo-s ep p ocesses
is la ge han ha o one-s ep p ocesses due o he highe 1,3-B selec-
i i y o wo-s ep ca alys s, esul ing in an impo an educ ion in
e hanol consump ion. This highe p oduc i i y is a he cos o ope a ing
wi h la ge e hanol ecycles, which d ama ically inc eases speci ic en-
e gy consump ion. The di e ence in e hanol and ene gy consump ion
will signi ican ly de e mine he ela i e economic and en i onmen al
pe o mance o he p ocesses, as shown in Sec ions 3.2 and 3.3.
A compa ison o he o e all mass and ene gy balances o published
wo ks on ETB p ocesses is shown in Table 4. The p oduc i i y o 1,3-BD
e lec s he mass con e sion e iciency o e hanol in o bu adiene, while
he o al u ili y consump ion indica es how much he mal ene gy mus
be pu in o play. The conclusion eached in he p esen s udy ha he
speci ic ene gy demand is la ge o wo-s ep p ocesses han one-s ep
p ocesses canno be deduced om he published wo ks. The main
eason o his is he dispa i y in he decisions made in hose wo ks
ega ding hea in eg a ion. In some wo ks, byp oduc s a e combus ed in
he CHP sys em o he p ocess, and/o he p ocess is hea in eg a ed wi h
an adjacen e hanol plan . Bo h decisions dec ease he u ili y demand, so
a compa ison wi h wo ks in which hose decisions a e no made is no
ai ; indeed, opposing esul s a e ob ained. Fo ins ance, Moncada e al.
(2018) and Cespi e al. (2016) epo ed one- and wo-s ep p ocesses,
espec i ely, wi h close 1,3-BD p oduc i i y in which he speci ic u ili y
demand o he wo-s ep p ocess is lowe , while jus he opposi e occu s
when compa ing he one- and wo-s ep p ocesses epo ed by Cespi e al.
(2016) and Dimian e al. (2021), espec i ely.
3.2. Economic assessmen
Fig. 2 depic s a b eakdown o he MBSP o all scena ios (OPEX and
CAPEX a e shown in Table A5). The MBSPs o he one-s ep p ocess
Table 3
Resul s o he o e all mass and ene gy balances o he one- and wo-s ep p ocess scena ios.
Scena ios B1 B2 B3 B4
I em Uni uni / BD uni / BD uni / BD uni / BD
Inpu s To al 3.800 3.125 2.462 2.137
Raw ma e ial E hanol (93 w %) 3.220 2.690 2.425 2.116
sol en s Oc ane 0.003 0.001 0.036 0.020
Wa e 0.575 0.432 0.000 0.000
DMF 0.001 0.001 0.001 0.001
Ou pu s To al 3.800 3.125 2.461 2.136
P oduc s 1,3-BD 1.000 1.000 1.000 1.000
Bu enes 0.097 0.062 0.040 0.048
E hylene 0.155 0.197 0.035 0.029
P opylene 0.054 0.049 0.015 0.003
Ace aldehyde 0.127 0.000 0.000 0.000
Hyd ogen 0.047 0.042 0.040 0.029
To boile Residual uel s eams 0.460 0.235 0.381 0.139
To ea men Was ewa e 1.861 1.539 0.950 0.888
U ili y Demand
a
To al GJ 29.109 25.483 51.729 59.347
Hea ing To al GJ 9.416 10.781 21.585 28.177
HPS GJ 0.025 3.326 4.106 4.100
MPS GJ 0.225 0.093 0.787 0.753
LPS GJ 1.325 0.775 5.094 11.644
Na u al gas GJ 7.841 6.588 11.598 11.680
Cooling To al GJ 19.693 14.702 30.144 31.170
Cooling wa e GJ 15.926 11.323 28.845 28.478
Chilled wa e GJ 0.227 0.000 0.008 0.107
Re ige an 1 GJ 3.347 3.194 1.291 2.585
Re ige an 2 GJ 0.193 0.186 0.000 0.000
Elec ici y kWh 500.7 344.6 656.6 564.8
a
P ope ies and empe a u es o each u ili y a e p o ided in Table C1. To al u ili y demand excludes elec ici y.
C.E. Cab e a Camacho e al.
Jou nal o Cleane P oduc ion 374 (2022) 133963
7
scena ios a e lowe han hose o he wo-s ep p ocesses, wi h he lowes
MBSP o scena io B2 (2006
€
/ ). This is ema kable since he 1,3-BD
p oduc i i y o he one-s ep p ocess scena ios is lowe . Fig. 2 shows
ha he MBSP is mos ly de e mined by he OPEX since he impac o he
CAPEX on he MBSP h ough e u n on in es men and dep ecia ion is
ela i ely low (5.7, 5.3, 6.9, and 6.2% o he MBSP om B1 o B4 sce-
na ios). E hanol accoun s o he highes ope a ing cos (78.1, 75.3,
63.6, and 57.6% o he MBSP om B1 o B4 scena ios), ollowed by
ene gy cos s (u ili ies and na u al gas; 7.1, 10.6, 16.9, and 23.8% o he
MBSP om B1 o B4 scena ios). The la ge 1,3-BD p oduc i i y o he
wo-s ep scena ios means lowe e hanol cos s han wi h one-s ep sce-
na ios, bu his cos educ ion is mo e han o se by he la ge ene gy
cos s, esul ing in highe MBSP han one-s ep p ocess scena ios. The
MBSP o he wo-s ep scena ios a e almos equal. The educ ion in
e hanol cos o he highe p oduc i i y o he B4 scena io is almos
coun e balanced by he inc ease in ene gy cos s because o ope a ing a
a la ge e hanol/ace aldehyde mole a io han he B3 scena io. In one-
s ep p ocess scena ios, he la ge e hanol cos s a e compensa ed by a
lowe ene gy cos and highe e enues om byp oduc s, he la e a
consequence o he lowe selec i i y o 1,3-BD, esul ing in lowe MBSP
han he wo-s ep scena ios. Fo he design capaci y chosen in his wo k
(200 k 1,3-BD/y), he CAPEX luc ua es om 0.9 o 1.1 M
€
/k 1,3-BD
o one-s ep p ocesses and 1.0–1.4 M
€
/k 1,3-BD o wo-s ep p ocesses.
A sensi i i y analysis o he MBSP o each scena io was pe o med by
changing ±20% he nominal alues o he economic pa ame e s ha
ha e he la ges in luence on he MBSP: CAPEX and he ma ke p ice o
e hanol, na u al gas, and byp oduc s (Fig. 3). As expec ed, he sensi i i y
o he MBSP o he p ice o e hanol is subs an ial, and i is la ge o one-
s ep p ocess scena ios since he p oduc ion o 1 o 1,3-BD equi es a
g ea e amoun o e hanol, while g ea e sensi i i y o he na u al gas
p ice in wo-s ep p ocess scena ios is explained by hei g ea e speci ic
ene gy consump ion ( he na u al gas p ice in luences he cos s o hea -
ing and cooling u ili ies acco ding o he co ela ions used o hei
es ima ion; Table A2). The sensi i i y o MBSP o he p ices o byp od-
uc s, assuming a simul aneous change in all o hem, is impo an only
o one-s ep p ocess scena ios. Changes in CAPEX ha e he smalles
impac on he MSBP o all scena ios.
The nominal MBSP o all scena ios (Fig. 2) is g ea e han he a e age
in e na ional ma ke p ice o 1,3-BD in he las decade (1423
€
/ ; IHS
Ma ki , 2019; Rodge s e al., 2022), so he scena ios a e no expec ed o
be p o i able. Since he economics o one- and wo-s ep p ocess sce-
na ios is highly dependen on he p ices o e hanol and na u al gas, and
hey a y o e ime, an unce ain y analysis was pe o med as explained
in Sec ion 2.3 o e alua e he p obabili y ha he scena ios a e p o i -
able. The esul s show (Fig. 4) ha he p obabili y ha any scena io is
p o i able (NPV >0) is low (B1: 11.1%, B2: 17%, B3:5.5%, and B4:
5.5%). The one-s ep p ocess scena ios ha e a highe p obabili y o being
p o i able, wi h he highes p obabili y o scena io B2.
The MBSPs epo ed on p e ious TEAs on ETB p ocesses a e shown in
Table 5. All MBSPs excep one (case Ace -BD; Rodge s e al., 2022) a e
g ea e han he a e age ma ke p ice o bu adiene in he las decade, so
he ETB ou e, ei he by one- o wo-s ep eac ion, is no expec ed o be
p o i able. This was con i med by wo ks (Table 5, las column) in which
unce ain y analyses we e pe o med by a ying he mos in luencing
economic pa ame e s based on his o ical da a. The unce ain y analyses
e ealed ha he p obabili y o being p o i able o one-s ep p ocesses,
al hough low, is highe han ha o wo-s ep p ocesses.
3.3. Li ecycle assessmen
In his sec ion, h ee en i onmen al impac s (GWP100, WC, and
CED) o he p oduc ion o 1,3-BD om suga cane bioe hanol (scena ios
B1 o B4) in B azil a e compa ed o hose o he naph ha s eam-c acking
ou e.
Because e hanol is he main con ibu o o he h ee analyzed impac
Table 4
Summa y o o e all mass and ene gy balances o published wo ks on ETB p ocesses.
P ocess Wo k Scena io P oduc i i y ( BD/
e hanol)
To al u ili y demand
g
(GJ/
BD)
Byp oduc ene gy
alo iza ion
b
In eg a ion adjacen
plan
c
One-s ep Cespi e al. (2016) – 0.511 11.4 Yes No
Shylesh e al. (2016) – 0.550 6.2
a
No Yes
Moncada e al. (2018)
h
Case I 0.279 11.5 No Yes
Case I 0.395 6.9 No Yes
Cab e a Camacho e al.
(2020)
B1 0.334 29.1 No No
B2 0.400 25.5 No No
Rodge s e al. (2022)
e
– – – Yes Yes
Two-
s ep
Cespi e al. (2016) – 0.281 8.8 Yes No
Fa zad e al. (2017)
d
BD-c 0.331 11 Yes Yes
Dimian e al. (2021) – 0.495 15.8 No No
This wo k B3 0.443 51.7 No No
B4 0.508 59.3 No No
a
Only consump ion o hea ing u ili ies was epo ed.
b
Byp oduc s a e no sold bu combus ed o p o ide hea o he p ocess.
c
The ETB plan is hea in eg a ed wi h an adjacen e hanol plan .
d
Coal bu ning scena io (BD-c) in which coal is bu n in a CHP sys em o sa is y he ene gy de ici .
e
P oduc i i y and u ili y demand a e no a ailable because o e all mass and ene gy balances we e no epo ed.
g
Sum o hea ing and cooling u ili y demand, elec ici y excluded.
h
Case I is he base case, while in Case II be e ca alys pe o mance is assumed.
Fig. 2. B eakdown o minimum bu adiene selling p ice o all scena ios. The
igu e o e each ba is he ne MBSP when byp oduc e enues a e accoun ed
o . Nume ical alues can be ound in Table A6.
C.E. Cab e a Camacho e al.
Jou nal o Cleane P oduc ion 374 (2022) 133963
8
ca ego ies, a calcula ion o unce ain ies in he impac ac o s ela ed o
he backg ound in en o y da a o e hanol p oduc ion om suga cane in
B azil was conside ed based on he li e a u e e iew unde aken by he
au ho s (Cab e a Camacho e al., 2020). Fo each impac ca ego y, he
sco es ob ained by using backg ound da a om he Ecoin en (V3)
da abase a e discussed i s , ollowed by he unce ain y analysis esul s.
3.3.1. Global wa ming po en ial
The sco e o his impac ca ego y is shown o each scena io in Fig. 5.
B2 is he scena io wi h he lowes GWP100 sco e, ollowed by he B3, B1,
and B4 scena ios, wi h 111, 143, 152, and 160% highe sco e han ha
o he naph ha-based p oduc ion o 1,3-BD. The GWP100 sco es o he
scena ios would be e en highe i he end o li e o 1,3-BD we e no
conside ed.
E hanol p oduc ion is he main con ibu o o GWP100, accoun ing
o 84% (B1), 81% (B2), 70% (B3), and 59% (B4) o he o al impac in
each scena io. The con ibu ion o e hanol o GWP100 dec eases wi h
ca alys selec i i y o 1,3-BD as lowe aw e hanol is consumed, bu
ewe c edi s om byp oduc subs i u ion a e ob ained. Mos o hese
c edi s come om hyd ogen (53–60% o one-s ep scena ios and
80–82% o wo-s ep scena ios) and e hylene (15–25% o one-s ep
scena ios and 6–7% o wo-s ep scena ios). As 1,3-BD ca alys
Fig. 3. Sensi i i y o MBSP o each scena io o CAPEX and he ma ke p ices o e hanol, na u al gas, and byp oduc s. Solid ba : +20% change. S iped ba :
−20% change.
Fig. 4. Resul s o economic unce ain y analysis. P obabili y o achie ing a
NPV lowe han a nume ical alue when conside ing simul aneous a ia ions in
he ma ke p ice o bioe hanol, na u al gas, and bu adiene based on his o i-
cal da a.
Table 5
Summa y o TEA s udies on ETB p ocesses.
P ocess Wo k Scena io P oduc ion Capaci y (k /y) Requi ed IRR (%) MBSP (
€
/ )
a
Unce ain y analysis
e
One-s ep Moncada e al. (2018)
b
Case I 24 10 4980 No
Case II 34 10 3883 No
This wo k B1 200 10 2247 Yes (11.1%)
B2 200 10 2006 Yes (17%)
Rodge s e al. (2022)
d
Ace -BD 9 10 1367 Yes (19%)
E h-BD 36 10 1954 Yes (1%)
Two-s ep Fa zad e al. (2017)
c
BD-b 30 10 3068 Yes (0%)
BD-c 37 10 2766 Yes (0%)
Dimian e al. (2021) – 91 ROI =13% 1660 No
This wo k B3 200 10 2427 Yes (5.5%)
B4 200 10 2380 Yes (5.5%)
a
P ices ha e been adjus ed o in la ion o 2022 (mul iplica ion by 1.16, 1.14, and 1.09 o he epo ed MBSPs; Fa zad e al., 2017; Moncada e al., 2018; Dimian
e al., 2021).
b
Case I is he base case, while in Case II, be e ca alys pe o mance is assumed.
c
Bd-b is a sel -su icien ene gy scena io in which a ac ion o biomass eed is bu n in he CHP sys em o sa is y he ene gy de ici o he p ocess, while in he BD-c
scena io, impo ed coal is bu n ins ead.
d
In he Ace -BD scena io, black liquo is gasi ied, syngas e men ed o ace aldehyde, hen hyd ogena ed o e hanol, which is con e ed o bu adiene. In he E h-BD
scena io, pulpwood is gasi ied, syngas con e ed in o mixed alcohols, and sepa a ed e hanol con e ed in o bu adiene.
e
Numbe s shown in b acke s is he p obabili y o NPV>0.
C.E. Cab e a Camacho e al.
Jou nal o Cleane P oduc ion 374 (2022) 133963
9
selec i i y inc eases om scena io B1 o B4, he dec ease in e hanol
impac on GWP100 ou weighs he loss o c edi s by he educed p o-
duc ion o byp oduc s. Fo he wo-s ep p ocess scena ios, he bene i s o
high 1,3-BD selec i i y a e exceeded by he nega i e impac o he high
ene gy (s eam and na u al gas) consump ion, esul ing in la ge
GWP100 sco es han he bes one-s ep p ocess scena io (B2).
The ange o expec ed GWP100 sco es o each scena io due o un-
ce ain y in he impac alue o B azilian suga cane-based e hanol is
shown in Fig. 6. The median GWP100 sco es om he unce ain y
analysis a e much lowe han he sco es shown in Fig. 5, he eason
being ha in he Ecoin en V3 da abase, he con ibu ion o he land-use
change on he GWP impac alue o B azilian e hanol is qui e la ge,
esul ing in an impac alue i e imes he median alue ound in he
li e a u e (Cab e a Camacho e al., 2020). Only o he one-s ep p ocess
scena ios is a educ ion in emissions o CO
2
almos ce ain wi h espec
o he naph ha-c acking ou e since he whole in e al o expec ed
GWP100 sco es o scena ios B1 and B2 is lowe han he GWP100 sco e
o he ossil ou e. Ne ca bon seques a ion o he one-s ep p ocess
scena ios is e y likely since he median alue o hei GWP100 sco e is
nega i e (−73 and −52 kg CO
2
eq/ 1,3-BD o B1 and B2 scena ios,
espec i ely). Based on CO
2
emissions, one-s ep p ocesses should be
p e e ed o e wo-s ep p ocesses, whose en i onmen al impac is
wo sened by CO
2
emissions ela ed o he high consump ion o
ossil-de i ed u ili ies.
A compa ison o ou esul s on CO
2
emissions wi h o he published
wo ks is di icul due o he di e ences in p ocess design, assessmen
me hodology, assump ions, and sou ces o backg ound in en o y, as
discussed in he in oduc ion sec ion. A common conclusion o all LCA
s udies is ha he main con ibu o o CO
2
emissions is he e hanol
supply chain. F om Table 6, i can be deduced ha (i) he ETB ou e can
achie e subs an ial educ ion in CO
2
emissions ela ed o he naph ha-
c acking ou e and (ii) he use o second-gene a ion e hanol esul s in
lowe CO
2
emissions han i s -gene a ion e hanol. Only wi h B azilian
suga cane can e hanol CO
2
emissions be as low as hose wi h second-
gene a ion e hanol. Since he second-gene a ion e hanol indus y is
no ye well de eloped, B azil is a sui able loca ion o p oduc ion o 1,3-
BD om e hanol a indus ial scale when conside ing e hanol a ail-
abili y and CO
2
emissions.
3.3.2. Wa e consump ion
All bio-scena ios (B1 o B4) show much g ea e sco es han he
naph ha-based p ocess (Fig. 7), mainly due o he i iga ion wa e used
in he cul i a ion o suga cane o e hanol p oduc ion, which anges
om 96 o 98% o he o al WC o B1 and B2 scena ios and 95–97% o
B3 and B4 scena ios. In his ca ego y, lowe e hanol consump ion pe o
1,3-BD is di ec ly ansla ed in o a educ ion in WC, esul ing in 34%
(B4), 25% (B3), and 18% (B2) educ ions o hese scena ios wi h
espec o he highes sco e scena io (B1, wi h 964 m
3
/ 1,3-BD). Two-
s ep p ocesses a e p e e ed o e one-s ep p ocesses in he WC ca e-
go y. This esul is no in line wi h ha o Cespi e al. (2016), who e-
po ed be e WC sco es o he one-s ep p ocess han he wo-s ep
p ocess owing o he lowe 1,3-BD p oduc i i y hey calcula ed o he
wo-s ep p ocess (Table 6).
The unce ain y analysis (Fig. 8) shows a simila pic u e o single
sco es based on he Ecoin en V3 da abase (Fig. 7), wi h lowe median
WC alues o he wo-s ep p ocess scena ios, al hough he o e lapping
o he in e als o WC sco e be ween he di e en scena ios is signi i-
can . I can be concluded ha i is likely ha he wo-s ep p ocess can
achie e a sligh ly be e pe o mance han he one-s ep p ocess in e ms
o WC due o i s lowe e hanol consump ion.
3.3.3. Cumula i e ene gy demand
Fig. 9 shows ha he CED o he 1,3-BD p oduc ion om bioe hanol
is 202–258% g ea e han ha om oil-de i ed naph ha. The supply
chain o e hanol demands mos o he ene gy (75–91%), and his pe -
cen age is la ge o he one-s ep p ocess scena ios due o hei lowe
1,3-BD p oduc i i y. The c edi s om byp oduc s, he second mos
impo an con ibu ion o one-s ep p ocess scena ios, help educe he
ene gy demand o hose scena ios by ~12%, while o he wo-s ep
scena ios, he educ ion is less signi ican (~6%). The non- enewable
ene gy associa ed wi h he impo ed elec ici y, na u al gas, and s eam
has a la ge impac on he wo-s ep p ocess scena ios (16 and 23% o B3
and B4, espec i ely), while i is no subs an ial in one-s ep p ocess
scena ios (5 and 8% o B1 and B2, espec i ely). O e all, he changes in
he con ibu ions o e hanol and u ili y consump ion as well as
byp oduc s when mo ing be ween scena ios cancel each o he ou , and
he di e ence in he CED sco e among scena ios is low, excep o sce-
na io B1, wi h he la ges CED sco e in which he e hanol impac is
dominan due o he low 1,3-BD p oduc i i y. A common ea u e o all
he scena ios is ha mos o he ene gy consumed is om enewable
esou ces, in ag eemen wi h Cespi e al. (2016) and Fa zad e al. (2017),
and ha he deple ion o non- enewable esou ces is lowe han p o-
ducing 1,3-BD om oil-de i ed naph ha, pa icula ly in one-s ep
p ocesses.
The unce ain y analysis shows ha he median CED sco e o each
scena io is close o i s single CED sco e (Fig. 10) because he p edic ed
Fig. 5. GWP100 sco es o each scena io conside ing B azil as he plan loca-
ion and he end o li e o 1,3-BD as s y ene-bu adiene ubbe o au omo i e
i es (Shylesh e al., 2016). CO
2
emissions om bu ning na u al gas a e
accoun ed o as di ec p ocess emissions (bu adiene ca ego y). The da a o
backg ound in en o y we e aken om Ecoin en (V3) da abase. “O he s”
comp ises sol en s, ca alys s, adso ben s, and disposal o esidues.
Fig. 6. Range o expec ed GWP100 sco es o each scena io calcula ed using
impac alues o B azilian suga cane-based e hanol om he li e a u e (Cab e a
Camacho e al., 2020). The median sco e o each case is shown in each box.
The CO
2
emission o he naph ha-c acking ou e is shown as a e e ence line.
C.E. Cab e a Camacho e al.