Cleane En i onmen al Sys ems 11 (2023) 100139
A ailable online 11 Sep embe 2023
2666-7894/© 2023 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/).
E alua ion o aba emen op ions o educe o maldehyde emissions in
ehicle assembly pain shops using he Li e Cycle me hodology
Daniel G anade o
a
,
b
,
*
, Aida Ga cia-Mu˜
noz
b
, Rena e Adam
b
, F ancisco Omil
a
,
Gume sindo Feijoo
a
a
CRETUS, Depa men o Chemical Enginee ing, School o Enginee ing, Uni e sidade de San iago de Compos ela, 15782, San iago de Compos ela, Spain
b
En i onmen al Compliance and Sus ainabili y, S ellan is, 65423, Rüsselsheim am Main, Ge many
ARTICLE INFO
Keywo ds:
Vehicle pain shop
Fo maldehyde emissions
Recupe a i e he mal oxida ion
Regene a i e he mal oxida ion
Li e cycle assessmen (LCA)
Eco-e iciency
ABSTRACT
Recupe a i e he mal oxidize s, consis ing o a single combus ion chambe whe e ola ile o ganic compounds
a e oxidized, and egene a i e he mal oxidize s comp ising se e al ce amic beds whe e he oxida ion akes
place, a e he mos common aba emen echnologies applied in ehicle pain shops o educe o maldehyde
emissions. In his wo k, a “c adle- o-g a e” Li e Cycle Assessmen and an eco-e iciency analysis we e ca ied ou
o a eal pain shop o compa e hese wo aba emen echnologies and iden i y he mos en i onmen ally sus-
ainable op ion. The esul s show ha he egene a i e oxidize leads o a dec ease o he human oxici y impac
ca ego y om 1329 kg 1,4-DB eq. in he ini ial si ua ion wi hou aba emen o 1284 kg 1,4-DB eq., while an
al e na i e wi h ecupe a i e oxidize s achie es a signi ican ly highe educ ion o 1176 kg 1,4-DB eq. Consid-
e ing he mos ele an selec ed impac ca ego ies, he esul s demons a e ha he ecupe a i e oxidize s cause
a educ ion om he ini ial si ua ion o 2.6% o he no malized index, whe eas he egene a i e oxidize implies
a aise o 3.1%. This indica es ha he ins alla ion o ecupe a i e oxidize s is he mos en i onmen ally sus-
ainable al e na i e om he wo in es iga ed echnologies. Ne e heless, he eco-e iciency analysis con i ms
ha he cos s o he ecupe a i e oxidize s op ion a e 2.2 imes highe .
1. In oduc ion
Vehicle manu ac u ing is one o he mos complex indus ial
manu ac u ing p ocesses because i equi es la ge spaces as well as he
sequen ial and coo dina ed ope a ion o mul iple echnologies (Bysko
e al., 2020; Giampie i e al., 2022). Pain ing ope a ions ha e been
iden i ied as he s ages wi h he g ea es en i onmen al impac s o he
en i e p oduc ion p ocess (VDI guideline 3455:2013-08; Ono e e al.,
2020). The high demand o ene gy and esou ces, such as na u al gas
and aw ma e ials o he p oduc ion o coa ings and pain s, wa e
consump ion along wi h he gene a ion o was e and was ewa e a e he
mos signi ican en i onmen al aspec s associa ed wi h he pain ing
ope a ions o ehicle assembly plan s (STS BREF, 2020). In pa icula ,
when i comes o iden i ying he mos pollu ing compounds in his ype
o acili y, he en i onmen al impac is ma ked by he emission o ol-
a ile o ganic compounds (VOCs), which ep esen he mos ele an
di ec emissions. They a e de i ed om sp ay boo hs, d ying o ens and
cleaning o equipmen wi h o ganic sol en s (Ri e a and Reyes-Ca illo,
2014, 2016; Ou e al., 2022). The Eu opean Union (EU) ca manu ac-
u e s emi ed in 2021 an a e age o 2.2 kg/ca and a o al o 25.7
housand ons o VOCs due o hei ehicle pain ing ac i i ies (ACEA,
2022), which co esponds o 2.5% o he o al VOCs emi ed in Eu ope
(Ge man En i onmen Agency, 2023).
An exhaus i e analysis o he pain s and coa ings used in a ehicle
pain ing plan allows he iden i ica ion o complex mix u es o mula ed
om nume ous componen s (Aka uah e al., 2016). In gene al, o ganic
sol en s, wa e , esins, plas icize s, dyes and pigmen s a e he basic
componen s o pain s, being he binde (i.e. he esins) he p edominan
cons i uen de ining he pain ’s cha ac e is ics (McMahon e al., 2023).
Addi ionally, new de elopmen s in au omo i e pain s include he use o
nanopa icles as ille s (Nayane de Quei oz e al., 2022). The sol en
con en o colo basecoa s is in he ange o 12–17% in wa e -based and
55–82% in sol en -based pain s, while clea coa s ypically ha e a sol-
en con en up o 50% (STS BREF, 2020). One o he mos commonly
used binding agen s in indus ial coa ings a e amino esins and, mainly,
hose de i ed om melamine (Pizzi and Ibeh, 2022). Melamine esins
used in au omo i e coa ings a e p oduced by he eac ion o melamine
* Co esponding au ho . CRETUS, Depa men o Chemical Enginee ing, School o Enginee ing, Uni e sidade de San iago de Compos ela, 15782, San iago de
Compos ela, Spain.
E-mail add ess: [email p o ec ed] (D. G anade o).
Con en s lis s a ailable a ScienceDi ec
Cleane En i onmen al Sys ems
jou nal homepage: www.jou nals.else ie .com/cleane -en i onmen al-sys ems
h ps://doi.o g/10.1016/j.cesys.2023.100139
Recei ed 17 July 2023; Recei ed in e ised o m 25 Augus 2023; Accep ed 7 Sep embe 2023
Cleane En i onmen al Sys ems 11 (2023) 100139
2
and o maldehyde (Po h, 2008). Pain manu ac u e s indica e melamine
esin con en s a ying om 0.1% o 18%, depending on he ype o
coa ing. O he amino esins used in coa ings, such as hose de i ed om
u ea o (me h)ac ylamide, a e also ob ained by he eac ion o di e en
chemicals wi h o maldehyde (Pizzi and Ibeh, 2022). Thus, o malde-
hyde is p esen in au omo i e coa ings as a esidue associa ed wi h he
esin manu ac u ing p ocess (wi h a concen a ion below 0.1%, as can
be aken om he ma e ial Sa e y Da ashee s) bu also as a componen
bound in melamine polyme s.
Conce ning he iden i ica ion o o maldehyde sou ces, he emission
o small quan i ies o his compound in o he a mosphe e p esen in he
exhaus gas low o he sp ay boo h has been quan i ied a e y low
le els (<1 mg/m
3
) (Kim e al., 2011). Howe e , also he bound o m-
aldehyde om melamine esins is eleased om inside he d ying
chambe s, which ope a e a ele a ed empe a u es ypically be ween
140 ◦C and 190 ◦C (Sal hamme e al., 2010; Sal hamme , 2019);
he e o e, depending on he con igu a ion o he coa ing line and he
e iciency o he end-o -p ocess aba emen equipmen connec ed o he
d ying o ens, he o maldehyde emission can be signi ican (So els
e al., 2017). A p esen , no all ehicle pain ing acili ies loca ed in
Eu ope ha e aba emen equipmen ins alled in he d ying o ens (STS
BREF, 2020). This depends on he speci ic legal equi emen s o each
coun y, hei speci ic se -up and he subsequen e o i ing s eps om
he plan assembly o he p esen ime. Besides, he use o aba emen
equipmen can be connec ed o he ype o pain s used in he pain shop.
Fo ins ance, in 2016, 43% o he EU pain shops we e es ima ed o apply
sol en -based p ime o basecoa s, o a combina ion o bo h. These pain
shops a e mo e likely o ha e ai emissions aba emen equipmen (STS
BREF, 2020).
In 2015, o maldehyde was eclassi ied as a subs ance wi h ca ci-
nogenic po en ial (ca ego y 1B) and mu agenic po en ial (ca ego y 2) in
he amendmen s o he so-called CLP Regula ion on classi ica ion,
labelling and packaging o subs ances and mix u es (Regula ion (EC) No
1272/2008; Regula ion (EU) No 605/2014; Regula ion (EU) No
2015/491). Wi h his change, all EU indus ies a ec ed by he Indus ial
Emissions Di ec i e 2010/75/EU shall, as a as possible, eplace
o maldehyde wi h less ha m ul subs ances. I o maldehyde emissions
canno be a oided, wha is he case o pain ing ope a ions in he
au omo i e indus y due o he use o he melamine esins, ope a o s
need o cope wi h signi ican ly mo e s ingen Emission Limi Values
(Di ec i e, 2010/75/EU).
Di e en echnologies a e a ailable o he emo al o VOCs om
exhaus gases. These include biological ea men , ecupe a i e he mal
oxida ion, egene a i e he mal oxida ion and ca aly ic oxida ion
(Mulholland and Dye , 1999; Be enjian e al., 2012; Toma is e al., 2019;
Li e al., 2023). In he au omo i e indus y, o he ea men o VOCs in
ehicle pain shops, egene a i e he mal oxida ion and ecupe a i e
he mal oxida ion, he la e some imes in combina ion wi h ca aly ic
oxida ion, ha e been commonly applied (VDI guideline 3455:2013-08;
STS BREF, 2020). Bo h echnologies ha e p o en o be e icien in VOC
des uc ion wi h e iciencies highe han 95% (STS BREF, 2020). The
in oduc ion o ca aly ic oxida ion o ecupe a i e oxidize s helped
educing he high ope a ing empe a u es om 700 ◦C–740 ◦C o 400 ◦C
necessa y o ob ain he desi ed aba emen e iciency, wi h he conse-
quen educ ion in na u al gas consump ion (Yang e al., 2019; STS
BREF, 2020; B umme e al., 2022). Howe e , his echnique is no in
use in ehicle pain shops because o he high main enance equi emen s
o he ca alys (STS BREF, 2020).
Gi en ha bo h egene a i e and ecupe a i e he mal oxida ion
p esen impo an di e ences, he selec ion be ween one o hem will
depend on a ious aspec s such as he VOC le els in he inle gas, he
low a es o be ea ed, he spaces a ailable in he pain shop, and he
in es men es ic ions o he pain shop ope a o (Bo wanka e al.,
2012). In he selec ion o he mos sui able echnology, i is also c ucial
o conside he en i onmen al impac o bo h echnologies om a li e
cycle pe spec i e in o de o sea ch o he mos sus ainable ea men
me hod and ake i in o accoun in he decision-making p ocess (Kl¨
op e
and G ahl, 2014).
Li e Cycle Assessmen (LCA) is a powe ul ool o analyze he en i-
onmen al impac s o a p oduc o p ocess (ISO 14040:2006; ISO
14044:2006). The e a e se e al LCA s udies ela ed o en i onmen al
impac s in he au omo i e indus y (Lopes Sil a e al., 2018; Geble
e al., 2020), bu mos o hem a e ocused on he ehicle i sel , new
engines, elec ic ca s and au omo i e li hium-ion ba e ies (He nandez
e al., 2017; Bou e and Guiche , 2022; Winjobi e al., 2022; Guo e al.,
2023). To he bes o ou knowledge, he e a e ha dly any s udies
a ailable in he scien i ic li e a u e on LCA associa ed wi h he ehicle
pain ing p ocess. Papasa a e al. (2001) s udied di e en coa ing ma-
e ials commonly used in au omo i e pain ing, ocusing on he use o
powde pain s as an al e na i e o sol en -bo ne pain s wi h he
ad an age o no con aining sol en s and being easily eusable. Ri e a
and Reyes-Ca illo (2016) e alua ed he o e all en i onmen al chal-
lenges o a pain shop, wi h special emphasis on me al and plas ic su -
aces. The a eas wi h he g ea es con ibu ion o en i onmen al impac
we e iden i ied and po en ial p ocess imp o emen al e na i es ocused
on educing ene gy consump ion we e p oposed. Bianco e al. (2020)
Abb e ia ions
ALO Ag icul u al Land Occupa ion
CC Clima e Change
COD Chemical Oxygen Demand
COT To al O ganic Ca bon
DNPH 2,4-Dini ophenylhyd azine
EU Eu opean Union
FD Fossil Deple ion
FE F eshwa e Eu ophica ion
FET F eshwa e Eco oxici y
HPLC High-Pe o mance Liquid Ch oma og aphy
HT Human Toxici y
IR Ionizing Radia ion
LCA Li e Cycle Assessmen
LCI Li e Cycle In en o y
LCIA Li e Cycle Impac Assessmen
MD Me al Deple ion
ME Ma ine Eu ophica ion
MET Ma ine Eco oxici y
MSDS Ma e ial Sa e y Da ashee
NLT Na u al Land T ans o ma ion
OD Ozone Deple ion
PMF Pa icula Ma e Fo ma ion
POF Pho ochemical Oxidan Fo ma ion
RTO Regene a i e The mal Oxidize
S1 Sys em 1
SS1 Subsys em 1
SS2 Subsys em 2
SS3 Subsys em 3
STS Su ace T ea men Using O ganic Sol en s
TA Te es ial Acidi ica ion
TET Te es ial Eco oxici y
ULO U ban Land Occupa ion
VOC Vola ile O ganic Compound
WD Wa e Deple ion
WWTP Was ewa e T ea men Plan
D. G anade o e al.
Cleane En i onmen al Sys ems 11 (2023) 100139
3
p o ided a li ecycle in en o y and LCA o an I alian au omo i e pain ing
p ocess, iden i ying he high ene gy consump ion, he di ec emissions
o VOCs and he was e p oduc ion and ea men as he ho poin s in
se e al analyzed impac ca ego ies. O e all, i is likely ha he sca ci y
o a icles published on his opic is associa ed wi h con iden iali y is-
sues, wi h he consequen di icul y in collec ing eliable da a.
Beside he echnical e alua ion o al e na i e aba emen echnolo-
gies also an economic assessmen o po en ial solu ions is o impo ance
o decision making. Eco-e iciency assessmen is a quan i a i e ool o
s udy he en i onmen al impac s o a sys em in conjunc ion wi h i s
economic alue (ISO 14045:2012) and helps en i ies o make conscious
decisions (To eg ossa e al., 2018; Ramí ez-Melga ejo e al., 2021).
Thus, eco-e iciency analysis es ablishes a connec ion be ween he
ecological impac o a sys em and i s economic dimension (Desli e al.,
2021). Al hough his iden i ies he solu ion wi h he mos economic
alue and he lowes possible ecological impac , only a ew
eco-e iciency s udies a e a ailable in he li e a u e. Recen examples
sha e esea ch esul s on was ewa e ea men plan s (WWTPs) (Sal-
a-Ga ido e al., 2023); howe e , publica ions desc ibing an
eco-e iciency analysis o ai emissions ea men equipmen in he
indus y could no be ound.
The objec i e o his wo k is o ad ance wi h eal da a on he en i-
onmen al analysis o wo di e en echnologies o he educ ion o
o maldehyde emissions om he opcoa d ying o ens o a eal ehicle
pain shop loca ed in Eu ope. A “c adle- o-g a e” LCA was ca ied ou
o he ehicle pain ing p ocess in he ini ial si ua ion o he pain shop
and in wo p oposed scena ios co esponding o he wo di e en
aba emen sys ems. As an essen ial s ep a e he LCA, an eco-e iciency
analysis was pe o med. The ele an cos da a needed we e ob ained by
an economic e alua ion o he wo aba emen sys ems. This ype o
analysis is no a ailable in he li e a u e and i aims a demons a ing
ha he LCA me hodology and eco-e iciency analysis can be used by
ehicle manu ac u ing companies and pain shop ope a o s o e alua e
di e en p ocess al e na i es and a o he mos en i onmen ally sus-
ainable echnologies.
2. Ma e ials and me hods
2.1. Desc ip ion o he pain shop
This s udy e alua es a eal pain shop in a ehicle manu ac u ing
plan loca ed in Eu ope. The ehicle pain ing p ocess consis s o a
sequence o dipping p ocesses as well as he applica ion o pain s and
coa ing ma e ials using a omiza ion equipmen (S ei be ge and D¨
ossel,
2008). The di e en laye s applied p o ide chemical and co osion
p o ec ion, wea he and sc a ch esis ance, as well as colo and isual
cha ac e is ics (Aka uah e al., 2016).
Be o e he main coa ing laye s a e applied o he ehicle body, a
se ies o p e- ea men s ages a e ca ied ou o clean and condi ion he
body o he nex pain ing s eps (Debna h, 2013). The i s s age is he
p e-cleaning, ollowed by wo o h ee deg easing phases. These s eps
emo e oil, g eases and lub ican s om he p e ious manu ac u ing
p ocesses. A e cleaning, an ac i a ion, a phospha ing, a passi a ion
and, inally, a ious insing and d ainage s ages a e applied o inc ease
he co osion esis ance o he ehicle body and p epa e i o he nex
coa ing p ocesses (Giampie i e al., 2020).
A e he p e- ea men , he i s main coa ing p ocess is he elec-
ocoa ing, in which he ehicle body is coa ed in a ba h by applying he
ca apho e ic pain , p o iding addi ional co osion esis ance (Giampie i
e al., 2020). In his elec ochemical p ocess, he pain is deposi ed on
he me al by passing elec ic cu en h ough he body, causing pain o
deposi on he su ace in a uni o m p o ec ion laye h oughou he
whole body (Ma de and Goodwin, 2023).
Once he elec ocoa ing has been comple ed, addi ional coa ing
p ocesses a e applied o wa e p oo sealing o body and weld seams and
o p o ec he unde body om damage wi h an elas ic op laye . A e
sealing and unde body coa ing, he p ime su ace , he basecoa and
he clea coa pain s a e applied in sp ay boo hs o di e en pu poses.
The successi e applica ion o he basecoa and he clea coa is
commonly e e ed o as he opcoa . While he p ime p o ides he
ehicle body wi h u he co osion esis ance and p o ec ion agains
s one chip, he basecoa and he clea coa a e mainly applied o add he
equi ed colo and gloss e ec s o he ehicle. Mo eo e , he basecoa
and he clea coa gi e du abili y and p o ec ion agains sc a ches and
chemical agg essions (Ju ge z, 1995). Be ween he applica ion o he
di e en laye s om elec odeposi ion o he opcoa each laye is cu ed
o d ied be o e he applica ion o he subsequen coa ing. This happens
in d ying o ens which a e ypically connec ed o emissions con ol
equipmen o elimina e VOCs and haza dous subs ances eleased om
he o ens (STS BREF, 2020).
Addi ional waxing, sealing and epai ing ope a ions on he d ied
opcoa comple e he p ocess and ensu e ha he desi able quali y and
co osion p o ec ion equi emen s a e me (Giampie i e al., 2020).
The pain shop unde s udy has wo pa allel ba hs o he elec o-
coa ing p ocess wi h wo d ying o ens pe ba h. To elimina e VOC
emissions and unpleasan odo s gene a ed in he o ens, each o en is
connec ed o wo indi idual ecupe a i e he mal oxidize s. The oxi-
dize s a e ope a ed on na u al gas as uel, and he excess hea o he
oxidize s is used o hea he o en by means o hea exchange s. As a
esul o he oxida ion p ocess, ni ogen oxides (NOx) and ca bon
monoxide (CO) a e emi ed in o he a mosphe e h ough he aba emen
ins alla ions exhaus sys ems (STS BREF, 2020).
Fo he p ime su ace applica ion, he pain shop has a line wi h
se e al pain boo hs in which a coa ing o wa e bo ne p ime is applied
o he ehicle bodies using obo s. In o de o d y all ehicle bodies a e
he p iming p ocess, he line has h ee o ens which, simila o he
elec ocoa ing p ocess, a e connec ed o he aba emen ins alla ions, i.e.
ecupe a i e he mal oxidize s, wi h he aim o ea ing he VOC emis-
sions (including o maldehyde) p oduced du ing he d ying p ocess.
P ime coa ing is ollowed by he inishing p ocesses, which consis
o he applica ion o a basecoa and a clea coa ha a e applied in
successi e p ocesses sepa a ed by a sho lash o a ea. The pain shop
s udied uses sol en -based basecoa and clea coa and has ou coa ing
lines wi h se e al boo hs whe e he pain s a e applied by obo s. A he
end o each clea coa line, wo o ens a e ins alled in o de o d y he
pain ed ehicle bodies. Unlike he elec ocoa ing and p ime p ocesses,
he o ens a e no connec ed o aba emen ins alla ions. As a esul , VOC
emissions, including he o maldehyde eleased in he o ens, a e
emi ed in o he a mosphe e. Pa o he exhaus ai om he basecoa
pain boo hs is ea ed in wo ecupe a i e he mal oxidize s. In a p e-
ious s age, he ola ile o ganic compounds p esen in he exhaus gas
a e pa ially emo ed in an ac i a ed ca bon uni (Be enjian e al.,
2012).
P oposed aba emen scena ios. To emo e o educe o maldehyde
emissions om he opcoa (basecoa and clea coa ) o ens, wo al e -
na i e ea men me hods co esponding o wo di e en aba emen
echnologies we e e alua ed om bo h economic and en i onmen al
iewpoin s.
The i s al e na i e was he ins alla ion o a egene a i e he mal
oxidize (RTO) ha can ea all was e gases om he opcoa o ens. In
egene a i e he mal oxida ion, o ganic sol en s a e oxidized a a em-
pe a u e be ween 800 ◦C and 850 ◦C (VDI guideline 3455:2013-08). The
aw gas en e s a i s bed illed wi h ce amic medium and is p ehea ed
be o e en e ing he combus ion chambe . The i s bed is cooled down
while he gas is hea ed. A e lea ing he combus ion chambe , he ho
gas passes h ough a second bed, eleasing he ene gy o he ce amic
packaging. The clean gas is hen cooled down and eleased in o he a -
mosphe e. The p ocess is ca ied ou al e na ely so ha in he nex cycle,
he inle gas is p ehea ed in he second bed and he i s bed is whe e he
ho gas is cooled. Using ce amic ma e ials, he hea is s o ed and mos o
he ene gy p oduced by combus ion is eused in he sys em i sel wi hou
he need o addi ional hea exchange s (VDI guideline 3455:2013-08).
D. G anade o e al.
Cleane En i onmen al Sys ems 11 (2023) 100139
4
The numbe o ce amic beds in an RTO a ies and ollows a ce ain
sequence o p ehea ing and cooling s eps aking place al e na ely in he
di e en beds. The des uc ion e iciency o he egene a i e he mal
oxida ion is 95% (STS BREF, 2020). Fo he pain shop s udied, a 5-bed
RTO was assumed o he ea men o he was e gas olumes om he
opcoa o ens.
The second al e na i e echnology conside ed was he ins alla ion o
ecupe a i e he mal oxidize s. In ecupe a i e he mal oxida ion, VOCs
a e oxidized in a simple oxida ion chambe by chemical eac ion wi h
oxygen p esen in he exhaus ai (STS BREF, 2020). The chambe has an
auxilia y bu ne unning on na u al gas in which he was e gas is hea ed
up o 700 ◦C–740 ◦C, leading o an almos comple e oxida ion, i.e. a
des uc ion e iciency close o 100% o he con ained o ganic com-
pounds (VDI guideline 3455:2013-08; STS BREF, 2020). Ene gy is sa ed
by p ehea ing he aw gas wi h he ho lue gas in an in e media e hea
exchange . Addi ionally, he ene gy excess o he clean gas is used o
hea he d ying o ens. Fo his s udy, eigh oxidize s a e assumed o ea
he was e gas olume om he opcoa o ens.
2.2. LCA me hodology
An a ibu ional LCA o he pain ing p ocess in he pain shop being
s udied was pe o med om a “c adle- o-g a e” pe spec i e ollowing
he equi emen s and guidelines o he ISO 14040:2006 and ISO
14044:2006.
Goal de ini ion and s a egy. The goal o his LCA was o e alua e
he al e a ion o he en i onmen al impac o he pain ing p ocess when
in oducing wo di e en aba emen sys ems o he educ ion o
o maldehyde emissions, allowing he compa ison o bo h echnologies
and suppo ing he selec ion o he mos sui able sys em du ing he
planning p ocess. This LCA is si ua ed a he mic o s a egic le el o he
ehicle manu ac u ing plan since he s ep o educe he emissions o
o maldehyde is one o he nume ous ac ions o inc ease i s en i on-
men al pe o mance and suppo co po a e en i onmen al s a egies.
Func ional uni . All he inpu s and ou pu s o he pain ing p ocess
a e ela ed o he unc ional uni . The ehicle pain shop p ocesses
360,000 ehicles pe yea , which co esponds o abou 75% o i s o al
capaci y. The unc ional uni selec ed was 1 h o ope a ion o each sub-
p ocess in es iga ed.
Sys em bounda ies. F om he en i e pain ing p ocess elec ocoa ing,
p ime and opcoa we e e alua ed as hey a e he mos ele an sub-
p ocesses in e ms o a mosphe ic emissions and, in pa icula , o mal-
dehyde eleases. The s udy conside ed only he ope a ion o he equip-
men while he cons uc ion o he ins alla ions was le ou o he LCA.
The en i onmen al impac o p e- ea men , sealing, waxing and
epai ing ope a ions was conside ed no ele an and hus is no
included in he scope o his wo k (VDI guideline 3455:2013-08).
The scope o he s udy is isualized by he sys em bounda ies shown
in Fig. 1. Addi ionally, he igu e illus a es he inpu s and ou pu s o he
sub-p ocesses wi hin scope as well as hei backg ound and o eg ound
sys ems, used as he basis o elabo a e he Li e Cycle In en o y (LCI).
As inpu s o he pain ing ope a ions na u al gas, elec ici y, pu ge
sol en s, wa e and elec ocoa ing and pain ing ma e ials we e consid-
e ed. The backg ound p ocesses a e hose ela ed o he p oduc ion and
supply o ma e ials and ene gy necessa y o ope a e he ins alla ions:
•Ex ac ion and p oduc ion o Na u al Gas: Na u al gas is needed o un
he bu ne s in o de o hea he o ens and o ope a e he aba emen
equipmen . I is p o ided by he na u al gas supplie o he
manu ac u ing plan . I s ex ac ion and p oduc ion p ocesses cause
en i onmen al impac s, such as emissions o ai , wa e deple ion and
was ewa e gene a ion (Shamoon e al., 2022).
Fig. 1. Scheme o he sys em bounda ies and backg ound and o eg ound p ocesses o he pain shop unde e alua ion.
D. G anade o e al.
Cleane En i onmen al Sys ems 11 (2023) 100139
5
•P oduc ion o elec ici y: Elec ici y is necessa y o ope a e he pain
shop ins alla ions, such as obo s, ai supply o sp ay boo hs, com-
p essed ai and ligh ing. The con ibu ion o each ene gy sou ce o
he elec ici y gene a ion was aken om he coun y mix whe e he
pain shop is loca ed.
•P oduc ion o Elec ocoa ing, P ime and Topcoa ma e ials and pain s:
The p oduc ion o he a ious ma e ials and pain s used in he
pain ing p ocesses con o ms a signi ican backg ound p ocess o his
sys em. I implies aw ma e ial ex ac ion, he use o ene gy and
wa e o he p oduc ion p ocess, di ec emissions, was ewa e and
was e gene a ion (da Sil a e al., 2016; Nai K e al., 2021; Shi e al.,
2023).
•P oduc ion o Sol en s: Simila o he p oduc ion o ma e ials and
pain s, he manu ac u ing o sol en s used o iscosi y adjus men o
he pain s as well as o cleaning o pain ing equipmen has associ-
a ed en i onmen al impac s.
•Pu i ica ion and supply o indus ial and DI (deionized) wa e : Indus ial
and deionized wa e is used o ca y ou he elec ocoa ing p ocess
and o wash ou he o e sp ay om he sp ay boo hs. Wa e coming
om he i e is p ocessed and pu i ied o be used in he pain shop
in a se ies o decan a ion, deca boniza ion and il a ion p ocesses
and, in he case o deionized wa e , a deioniza ion p ocess. The
deioniza ion is pe o med in an ion-exchange esin sys em.
As ou pu s om he pain ing ope a ions he s udy e lec s ai emis-
sions, was e and was ewa e , whe eo was ewa e is ea ed onsi e in a
WWTP be o e discha ge. In mo e de ail, he o eg ound p ocesses a e:
•Emissions in o he a mosphe e o VOC, dus and o maldehyde om
pain boo hs, o ens and ai emissions aba emen equipmen and CO,
CO
2
and NOx om bu ne s and aba emen equipmen .
•Release o was ewa e pollu an s in concen a ions below gi en
h esholds, i.e. COD, COT, Zinc, I on, Phospho us, Oil & g eases,
suspended solids and Fluo ides, a e he ea men in an onsi e
WWTP discha ged in o he i e . Was ewa e ea men is included
in he sys em bounda ies o his wo k.
•The gene a ion o haza dous solid was e and o he pain was e which
lea e he si e o be ea ed in an o -si e was e ea men acili y.
Was e ea men is ou side o he sys em bounda ies o his s udy.
Da a collec ion and da a quali y. Fo he elabo a ion o he LCI a
p ocess-based bo om-up app oach was ollowed. Real consump ion and
emissions da a o inpu s and ou pu s o he pain ing p ocess we e
di ec ly ob ained om he pain shop. Ma e ials consump ion da a a e
based on logis ical and inancial da abases, as well as in o ma ion ob-
ained om he ope a ional con ol sys ems o he pain acili y. The
composi ion o he ma e ials is a ailable h ough he ma e ial Sa e y
Da a Shee s (Regula ion (EC) No 1907/2006) and addi ional in o ma-
ion p o ided by he supplie s. Pain s used in he opcoa p ocess di e
in composi ion and sol en con en . In o de o elabo a e he LCI o he
opcoa p ocess, an exhaus i e e alua ion o all ma e ials was pe o med
o c ea e se e al clus e s ep esen a i e o he a ie y o p oduc s used.
The consump ion o na u al gas and elec ici y was ob ained om
me e s ins alled in he pain shop, while he alloca ion o ene gy con-
sump ion by sub-p ocess was made using bibliog aphic alues om he
STS Bes A ailable Techniques Re e ence Documen .
VOC emissions we e calcula ed using annual mass balances in
acco dance wi h he ele an Eu opean legisla ion (Di ec i e,
2010/75/EU). Da a on he es o he a mosphe ic emissions and
was ewa e pa ame e s we e ob ained om ac ual measu emen cam-
paigns and ins alled s anda d moni o ing equipmen commonly
accep ed in indus y (VDI guideline 3455:2013-08; STS BREF, 2020). In
cases whe e con inuous moni o ing esul s o da a o he whole yea
we e no a ailable, he a e age esul s ob ained du ing he moni o ing
pe iods we e used. Fo maldehyde concen a ions in emission sou ces
we e de e mined using he DNPH me hod (VDI 3862 Pa 2:2000-12),
whose p o ocol s a es ha om a known olume o gas samples wi h-
d awn using glass impinge s con aining a solu ion o 2,4-dini ophenyl-
hyd azine (DNPH), he samples a e ans e ed o da k glass ials o
HPLC analysis. The emission loads o he di e en pollu an s measu ed
in each emission sou ce o he analyzed sub-p ocesses we e summed up
o calcula e he o al emissions o each pollu an o he elabo a ion o
he LCI.
Finally, he was e da a we e collec ed om he mon hly was e
ans e eco ds o he pain shop o he di e en iden i ied was e
s eams.
2.3. Assessmen me hodology and impac ca ego ies
A Li e Cycle Impac Assessmen (LCIA) was conduc ed using com-
me cial so wa e SimaP o .8.2 wi h Eu opean ReCiPe Midpoin V1.12
me hodology. This me hod is scien i ically sound, easy o use and
in e p e , and in e na ionally accep ed. The me hod ocuses on en i-
onmen al impac and damage, dis inguishing 18 di e en impac ca -
ego ies (Gelde mann and Ren z, 2005; ILCD, 2010a; ILCD, 2010b;
Goedkoop e al., 2013).
A e he selec ion o he impac ca ego ies and he classi ica ion o
he in en o y esul s as he ini ial s eps o he LCIA, he cha ac e iza ion
and no maliza ion s eps de ined in he ISO 14040 and ISO 14044 s an-
da ds we e ollowed o ca y ou he impac assessmen . The cha ac-
e iza ion o he sys em in i s ini ial si ua ion wi hou aba emen was
pe o med in o de o iden i y he sub-p ocess wi h he g ea es en i-
onmen al impac . In addi ion, a cha ac e iza ion o each indi idual sub-
p ocess, i.e. elec ocoa ing, p ime and opcoa , was conduc ed o ob ain
he inpu s and ou pu s wi h he highes con ibu ion o he en i on-
men al impac .
The nex s ep was he cha ac e iza ion o he sys em wi h he wo
al e na i e ai emissions aba emen echnologies. In o de o simpli y
he analysis, his s ep was conduc ed exclusi ely o he opcoa p ocess
since he applica ion o he aba emen sys ems was only conside ed o
his sub-p ocess. A e he cha ac e iza ion, he esul s o he impac
ca ego ies we e no malized o he sys em in he ini ial si ua ion and
wi h he wo aba emen echnologies o allow a be e compa ison.
Fo he compa ison o he no malized da a, in a i s analysis, all 18
impac ca ego ies included in he me hodology we e aken in o accoun .
In subsequen s ages, he mos ele an impac ca ego ies we e selec ed
on he basis o hese wo c i e ia:
•All impac ca ego ies whose ele an con ibu ion o he o al en i-
onmen al impac was highe han 2%. Eigh ca ego ies could be
iden i ied by e alua ing he no maliza ion esul s. These eigh ele-
an impac ca ego ies con ibu e o 95.5% o he o al en i on-
men al impac .
•Human oxici y (HT) exclusi ely, since he educ ion o human
oxici y h ough he educ ion o o maldehyde emissions due o i s
cance -causing cha ac e is ic is he objec i e o he aba emen
equipmen in his s udy.
Finally, o he pu pose o compa ing he wo p oposed scena ios
wi h he ini ial si ua ion and iden i y he mos en i onmen ally a o -
able solu ion, he no malized index conside ing only he HT impac
ca ego y and he sum o he no maliza ion alues o he selec ed impac
ca ego ies as desc ibed abo e we e ep esen ed o each scena io.
2.4. Me hodology o economic e alua ion and eco-e iciency analysis
As a p e equisi e o he eco-e iciency analysis, an ini ial s udy o he
wo aba emen sys ems was pe o med, including an economic e alua-
ion o ob ain all necessa y economic da a. The capi al cos s o he
ins alla ion o he wo sys ems we e eques ed om equipmen manu-
ac u e s o he pain shop s udied. The ope a ional and main enance
cos s da a we e a ailable om o he company plan s wi h simila
D. G anade o e al.
Cleane En i onmen al Sys ems 11 (2023) 100139
6
equipmen . These cos s we e calcula ed o he whole li espan o he
ins alla ions (RTO =25 yea s; Recup. Oxidize s =15 yea s) and he
annual inc ease in he cos o na u al gas and spa e pa s o equipmen
was aken in o accoun using he a e age in la ion a es o ecen yea s
o he coun y whe e he pain shop is loca ed. Finally, he o al cos s
we e no malized o he unc ional uni o allow he compa ison be ween
he wo sys ems.
Ul ima ely, o b ing he economic pe spec i e in o he compa ison,
an eco-e iciency analysis was conduc ed based on he c i e ia p e-
sc ibed in he ISO 14045:2012.
The eco-e iciency analysis was pe o med wi h he ocus se on he
opcoa p ocess, o which he aba emen equipmen was planned. The
no maliza ion esul s p e iously ob ained in he LCA we e used o he
eco-e iciency e alua ion. Two di e en indica o s we e selec ed and
ep esen ed agains he o al cos pe hou :
•Va ia ion o he no malized index, exp essed in pe cen age, consid-
e ing he sum o he impac ca ego ies wi h a ele an con ibu ion o
he o al en i onmen al impac g ea e han 2% in he no maliza ion
esul s.
•Va ia ion o he no malized index, exp essed in pe cen age, o he HT
impac ca ego y.
I mus be highligh ed ha , e en hough he LCA did no conside he
cons uc ion o he pain ing ins alla ions bu only he ope a ion o he
equipmen , he cos s o he ins alla ion o he wo planned aba emen
sys ems could no be excluded. They a e he majo cos ac o in he o al
calcula ed cos s o he wo sys ems and a e necessa y o achie e a
ep esen a i e compa ison. Besides, all o he pain ing equipmen and
ins alla ions we e al eady a ailable in he pain shop a he ime o his
s udy and no addi ional in es men would be necessa y.
3. Resul s and discussion
3.1. Ini ial e alua ion o he aba emen sys ems
An ini ial e alua ion o he wo di e en aba emen sys ems was
pe o med p io o he LCA. This in o ma ion is equi ed o be e un-
de s and he di e ences be ween he wo sys ems and, along wi h he
esul s ob ained in he LCA and eco-e iciency analysis, suppo he
decision-making p ocess. The ad an ages and d awbacks o bo h ech-
nologies can be summa ized as ollows:
•As a as na u al gas consump ion is conce ned, he ins alla ion o
eigh ecupe a i e he mal oxidize s was conside ed much mo e
con enien since he excess hea can be eused o hea he opcoa
d ying o ens, wi h he consequen elimina ion o exis ing bu ne s
and he concomi an sa ings o na u al gas.
•Ini ial in es men cos s a e much highe o he eigh ecupe a i e
he mal oxidize s wi h hei connec ed hea exchange s han o he
RTO. Main enance cos s a e highe o he RTO, mainly due o he
egula exchange o he ce amic packaging. Gene ally, he ins alla-
ion o he RTO is mo e ad an ageous om an economic poin o
iew.
•Fo he ins alla ion o ecupe a i e oxidize s, longe p oduc ion
shu down pe iods a e needed, which would ha e an impac on he
p oduc ion olumes o he plan . An RTO can be buil in pa allel o
he p oduc ion and only se e al weeks o p oduc ion shu down a e
needed o conclude he ins alla ion. The e o e, he ime ame o he
ins alla ion o an RTO is conside ably sho e han he ime equi ed
o he ins alla ion o he ecupe a i e oxidize s.
•The RTO is independen o he opcoa pain ing lines and o ens, so
ha in he e en o b eakdowns, p oduc ion can con inue wi hou
ehicle losses and p oduc ion delays. Since he excess hea om he
ecupe a i e he mal oxidize s would be used o hea he d ying
o ens, d ying and pain ing shall be s opped in case o mal unc ions
and necessa y epai s.
A simpli ied ep esen a ion o he main di e ences, ad an ages and
d awbacks o he wo al e na i e sys ems is shown in Table 1.
3.2. Li e Cycle In en o y
The in en o y da a p esen ed in Table 2 co espond o he s epwise
pain ing p ocess ha akes place in he pain shop analyzed. Da a we e
collec ed o one yea o p oduc ion and all da a e e o he unc ional
uni o 1 h o ope a ion. E en hough he educ ion o o maldehyde
emissions cons i u es he ocus o his wo k, he emissions o all o he ai
and wa e pollu an s ha occu in he ehicle pain shop we e consid-
e ed in o de o elabo a e he li e cycle in en o ies.
To compile he li e cycle in en o ies, da a we e g ouped in o h ee
subsys ems co esponding o he coa ing and pain ing s eps included in
he scope o his wo k. The h ee subsys ems oge he cons i u e an
o e all pain ing p ocess sys em as shown in Table 2. Addi ionally, all
inpu s and ou pu s ha could no be assigned o he di e en subsys ems
we e included in he global pain p ocess sys em o he LCA.
The ins alla ion o he wo di e en aba emen echnologies
compa ed in his wo k o he was e gas s eam led o signi ican a i-
a ions in he in en o y da a. The di e ences a e shown in Table 3.
The da a we e collec ed om he con inuous sampling and moni-
o ing campaigns o wo sis e plan s wi h simila cha ac e is ics and in
which bo h ypes o aba emen echnology a e cu en ly in ope a ion. A
he same ime as he o e all o maldehyde emissions om he pain
acili y we e educed by he aba emen equipmen in he opcoa p o-
cess, a signi ican inc ease was obse ed in he concen a ion o he co-
p oduc s om he oxida ion p ocess, i.e. NOx and CO. I was no ed ha
he emission concen a ions o o maldehyde, NOx and CO om he
aba emen sys em we e e y simila o bo h echnologies. Thus, he
absolu e emissions o he h ee pollu an s in e ms o kilog am pe hou
we e conside ed o be he same in he Li e Cycle In en o y (LCI). The
majo di e ence ound in de eloping he in en o y da a o he wo
di e en sys ems is due o he use o na u al gas. All o he in en o y
pa ame e s o inpu ma e ials, wa e consump ion, quan i ies o was e
and was ewa e gene a ed, and was ewa e emissions we e no a ec ed
due o he ins alla ion o he was e gas aba emen equipmen .
3.3. Li e Cycle Assessmen
Cha ac e iza ion. In he cha ac e iza ion s ep, he ho spo s we e
de ec ed by analyzing he con ibu ion o he di e en subsys ems o he
Table 1
Compa ison o he main cha ac e is ics o he wo al e na i e aba emen sys ems
iden i ied in he ini ial e alua ion o he pain shop o s udy. Au ho s’ own
elabo a ion.
Main Cha ac e is ics 5-bed RTO 8 Recupe a i e The mal
Oxidize s
Aba emen e iciency 95–99% ≈100%
Ini ial in es men cos s Lowe Highe
Main enance cos s Highe Lowe
Na u al gas consump ion Highe Lowe due o hea eco e y
o o en
Ene gy eco e y No possible in his
pain shop
Excess hea o o ens
Abili y o Run O en Wi hou
Aba emen Sys em
Yes No
Space U iliza ion Lowe Highe due o high numbe
o de ices
Accessibili y o Repai o
Pa s Replacemen
Highe Lowe
Time o ins all Du ing p oduc ion Ou side p oduc ion
Li espan Long Li e
Equipmen
I will ha e o be eplaced
du ing li e o o en
D. G anade o e al.
Cleane En i onmen al Sys ems 11 (2023) 100139
7
Table 2
Li e Cycle In en o y (LCI) o he ini ial si ua ion o he pain shop unde e alua ion. All quan i ies a e e e ed o he unc ional uni , i.e. 1 h o ope a ion o each sub-
p ocess.
S1 Pain ing P ocess SS1 Elec ocoa ing SS2 P ime SS3 Topcoa Uni Me hod/Sou ce o da a
INPUTS: om Technosphe e
Ma e ials
Wa e DI 8.39 2.39 m
3
Pain shop consump ion eco ds
Pu ge sol en 73.1 kg Pain shop consump ion eco ds
Indus ial wa e 16.07 m
3
Pain shop consump ion eco ds
Binde : Pain shop consump ion eco ds
O ganic sol en 12.18 kg MSDS
a
Epoxy esins 203 kg Supplie in o ma ion
b
Wa e 190.82 kg Calcula ion
Ca ionic pas e: Pain shop consump ion eco ds
O ganic Sol en 139.09 kg MSDS
Epoxy esins 794.8 kg Supplie in o ma ion
Pigmen 397.4 kg Supplie in o ma ion
Wa e 655.71 kg Calcula ion
P ime Pain s: Pain shop consump ion eco ds
O ganic sol en 4.82 kg MSDS
Polyes e esins 15.41 kg Bibliog aphic esea ch
c
Amino esins 6.74 kg Supplie in o ma ion
F ee o maldehyde 0.1 kg Supplie in o ma ion
Polyu e hane 1.93 kg Supplie in o ma ion
Wa e 38.52 kg Supplie in o ma ion
Pigmen s 28.89 kg Calcula ed
Basecoa Pain s: Pain shop consump ion eco ds
O ganic sol en 85.4 kg MSDS
Polyes e esins 32.2 kg Bibliog aphic esea ch
Amino esins 14 kg Supplie in o ma ion
F ee o maldehyde 0.14 kg Supplie in o ma ion
Pigmen whi e 0.14 kg Calcula ed
Pigmen Black/g ey 4.2 kg Calcula ed
Pigmen Blue/G een 0.84 kg Calcula ed
Addi i es 1.4 kg Supplie in o ma ion
Basecoa Whi e Pain : Pain shop consump ion eco ds
O ganic Sol en s 19.32 kg MSDS
Polyes e Resins 7.14 kg Supplie in o ma ion
Amino esins 2.94 kg Supplie in o ma ion
F ee o maldehyde 0.04 kg Supplie in o ma ion
Pigmen whi e 11.76 kg Supplie in o ma ion
Addi i es 0.84 kg Supplie in o ma ion
Clea coa Lacque : Pain shop consump ion eco ds
O ganic sol en 82.81 kg MSDS
Amino esins 28.73 kg Supplie in o ma ion
Hyd oxyl Resin/Ac ylic 52.39 kg Supplie in o ma ion
F ee o maldehyde 0.17 kg Supplie in o ma ion
Addi i es 5.07 kg Supplie in o ma ion
Sol en s Basecoa ( iscosi y adjus men ): Pain shop consump ion eco ds
Bu anol 7.28 kg MSDS
Bu ylace a e 25.48 kg MSDS
Xylene 53.69 kg MSDS
Sol en o ganic 5.46 kg MSDS
Ene gy
Na u al Gas 4.81 6.23 15.2 MWh Pain shop ene gy eco ds
Elec ici y 2.04 2.35 7.22 MWh Pain shop ene gy eco ds
Cogene a ion 1.38 1.59 4.88 MWh Pain shop ene gy eco ds
OUTPUTS: emissions o he en i onmen
Emissions o ai
d
Dus 3.51 0.05 0.27 10.71 kg Measu ed, Calcula ed
NOx 24.62 4.15 8.09 35.02 kg Measu ed, Calcula ed
CO 7.4 20.6 2 4.3 kg Measu ed, Calcula ed
CO
2
433 1193 n.a.
e
n.a. kg Measu ed, Calcula ed
VOCs 73.13 4.42 1.3 200.57 kg Calcula ed
Fo maldehyde 0.12 0.01 0.01 1.36 kg Measu ed, Calcula ed
Emissions o wa e
COD 3.25 222 g O
2
Measu ed, Calcula ed
COT 0.99 59.8 g Measu ed, Calcula ed
Zinc 0.03 1.7 g Measu ed, Calcula ed
I on 0.03 2.4 g Measu ed, Calcula ed
Phospho us 0.11 6.86 g Measu ed, Calcula ed
Oil & g eases 0.11 7.45 g Measu ed, Calcula ed
Suspended solids 0.59 41 g Measu ed, Calcula ed
Fluo ides n.a. 15.8 g Measu ed, Calcula ed
(con inued on nex page)
D. G anade o e al.
Cleane En i onmen al Sys ems 11 (2023) 100139
8
impac ca ego ies. An ini ial assessmen o he o e all pain ing p ocess
(Fig. 2 (a)), comp ising he inpu s and ou pu s o he elec ocoa ing,
p ime and opcoa p ocesses, as well as all o he inpu s and ou pu s ha
could no be assigned o he h ee subsys ems, shows ha he opcoa
p ocess has he la ges con ibu ion o mos impac ca ego ies. The
smalles e ec o he opcoa p ocess can be no iced in he WD impac
ca ego y wi h 40.1 m
3
in a o al o 246.6 m
3
(16.3%), whils he highes
sha e o his p ocess in an impac ca ego y can be obse ed in POF wi h
251.4 kg NMVOC in a o al o 422.2 kg NMVOC (59.5%). Fu he mo e,
he impac ca ego ies MD =242.1 kg Fe eq. o he opcoa p ocess (a
con ibu ion o 59.4% o he o al o he impac ca ego y), FET =164.4
kg 1,4-DB eq. (57.3%), MET =137.0 kg 1,4-DB eq. (57.3%), NLT =1.48
m
2
(55.3%), OD =0.001 kg CFC-11 eq. (54.1%), ALO =106.9 m
2
a
(53.6%), TET =0.65 kg 1,4-DB eq. (53.5%) and IR =230.0 kBq U235 eq.
(53.2%) a e ema kably a ec ed by he opcoa p ocess as he gi en
pe cen ages indica e.
These esul s can be explained by he ac ha he opcoa is he
p ocess wi h he highes consump ion o ma e ials and p oduce o
emissions, in pa icula he use o sol en -bo ne pain s, highe con-
sump ion o na u al gas due o he use o mo e bu ne s o hea he d ying
o ens han in he o he sub-p ocesses, and he di ec emissions in o he
a mosphe e wi hou ins alled aba emen sys ems. A simila ou come
was obse ed by Bianco e al. (2020), speci ically o he POF impac
ca ego y, who iden i ied he p o ision o hea ing and he di ec emis-
sions o VOCs as ho spo s o he ehicle pain ing p ocess.
Addi ionally, i can also be obse ed ha he elec ocoa ing p ocess
plays an impo an ole in all impac ca ego ies, wi h he g ea es
con ibu ion in he ca ego ies o WD =172.9 m
3
(70.2% con ibu ion o
he o al o he impac ca ego y), PMF =25.2 kg PM10 eq. (51 %), CC =
10224.5 kg CO
2
eq (47.3%), FD =3587.1 kg oil eq. (45.4%) and TA =
49.1 kg SO
2
eq. (42 %), p ima ily due o he ma e ials used in his
p ocess. Subsequen ly, he h ee subsys ems we e analyzed sepa a ely:
Elec ocoa ing p ocess: Fig. 2 (b) shows he esul s o he
cha ac e iza ion o he elec ocoa ing p ocess. Fo mos o he impac
ca ego ies, he ma e ials used in he elec ocoa ing ba hs, i.e. he binde ,
consis ing o o ganic sol en s, epoxy esins and wa e , and he ca ionic
pas e, which is made o o ganic sol en s, epoxy esins, pigmen s and
wa e , make he g ea es con ibu ion. In pa icula , he ca ionic pas e
p esen s he la ges impac s. Fo ins ance, in he NLT and he FE impac
ca ego ies, he ca ionic pas e ep esen s 43.4% (0.24 m
2
) and 59.1%
(0.31 kg P eq.) o he o al impac ca ego y, espec i ely. In o de o
iden i y which componen o he ca ionic pas e is esponsible o his
esul , a cha ac e iza ion o he ca ionic pas e based on i s cons i uen s
was ca ied ou . In his analysis, i was obse ed ha he epoxy esin in
he o mula ion o he ca ionic pas e is he main con ibu o o he
majo i y o he impac ca ego ies, being POF =0.02 kg NMVOC (97.3%
o he o al o he impac ca ego y), FD =1.07 kg oil eq. (97.2%), PMF =
0.009 kg PM10 eq. (96 %), CC =2.71 kg CO
2
eq. (95.8%) and TA =
0.016 kg SO
2
eq. (93 %) he mos a ec ed ca ego ies. This esul can be
mainly explained due o he di ec emissions o ai pollu an s associa ed
wi h he esin p oduc ion p ocess, such as CO
2
om combus ion p o-
cesses (Wilson, 2009).
P ime p ocess: As can be seen in Fig. 2 (c), mos o he impac ca e-
go ies a e highly a ec ed by na u al gas consump ion, gi en i s back-
g ound p ocesses ela ed o i s ex ac ion and p oduc ion, each wi h a
high impac . The di ec emissions o he p ime p ocess a e mo e ele-
an han in he elec ocoa ing and ha e hei main con ibu ion o he
impac ca ego ies o POF =9.5 kg NMVOC (67.1%), ME =0.32 kg N eq.
(61.8%), PMF =1.78 kg PM10 eq. (42.9%) and TA =4.53 kg SO
2
eq.
(36.4%). Wi h ega d o he acidi ica ion and eu ophica ion impac
ca ego ies, he high ele ance o he di ec emissions can be assigned o
he NOx emissions coming om he oxida ion p ocess o he ecupe a-
i e he mal oxidize s ins alled in he p ime o ens, wha con i ms ob-
se a ions made by Bana and Çokaygil (2010).
Topcoa p ocess: The cha ac e iza ion esul s o he opcoa p ocess
a e shown in Fig. 2 (d). To simpli y he in e p e a ion o he esul s, all
c ea ed pain clus e s we e inally g ouped in o an indi idual sys em
namely “To al pain s”. These esul s p esen a simila beha io as o he
p ime p ocess. Di ec emissions conside ably inc ease hei ele ance in
he ollowing impac ca ego ies: POF =236.9 kg NMVOC (94.3%), ME
=1.37 kg N eq. (70.3%) and PMF =7.7 kg PM10 eq. (53.4%). Mo eo e ,
he sha e o he di ec emissions in he TA impac ca ego y aises wi h
19.6 kg SO
2
eq. o 47.8%, 11.4% highe han in he p ime p ocess. This
ai can be a ibu ed o wo easons. Fi s ly, he inc emen o he NOx
emissions in he exhaus ai o he ecupe a i e he mal oxidize s
ins alled in he basecoa sp ay boo hs. In pa icula , he high con ibu-
ion o he di ec emissions o he TA and ME ca ego ies is mani es ly
connec ed o he NOx emissions om he oxida ion p ocess (Bana and
Çokaygil, 2010). Secondly, he inc ease o he VOC emissions om he
opcoa o ens, whe e he sol en -bo ne pain s a e being cu ed wi hou
aba emen equipmen , unlike he p ime p ocess. The high impac o he
di ec emissions o he POF impac ca ego y is easonable since VOCs a e
well-known p ecu so s o pho ochemical smog and oposphe ic ozone,
Table 2 (con inued)
S1 Pain ing P ocess SS1 Elec ocoa ing SS2 P ime SS3 Topcoa Uni Me hod/Sou ce o da a
Was e
Emulsion pain s o incine a ion 4.37 kg Pain shop was e eco ds
Pain o be sepa a ed 68.6 kg Pain shop was e eco ds
Solid haza dous o incine a ion 1.78 kg Pain shop was e eco ds
a
MSDS: Ma e ial Sa e y Da ashee .
b
Supplie In o ma ion e e s o indica i e pe cen ages communica ed by supplie s o he cons i uen s o each pain and ma e ial. The pe cen ages we e used o
calcula e he absolu e amoun s o each cons i uen in he inpu ma e ials.
c
S oye and F ei ag (1998). Bibliog aphic esea ch e e s o indica i e pe cen ages o he cons i uen in pain s and ma e ials.
d
The "emissions o ai " alues we e calcula ed as he sum o he emission loads measu ed a all emission sou ces o he di e en sub-p ocesses o he pain shop. The
measu emen unce ain ies o he me hod a e as ollows: Dus : n.a.; NOx: 15%; CO: 9%; Fo maldehyde: 15%.
e
n.a.: no a ailable.
The "emissions o wa e " alues we e calcula ed by means o he annual a e age concen a ion alues o wo di e en was ewa e s eams and conside ing he o al
olume lows and ope a ing hou s.
Table 3
Di e ences in he LCI o he opcoa p ocess due o he aba emen ins alla ion.
All quan i ies a e e e ed o he unc ional uni , i.e. 1 h o ope a ion o each
sub-p ocess.
Ini ial
Si ua ion
5-bed
RTO
8 Recupe a i e
Oxidize s
Uni
INPUTS: om Technosphe e
Ene gy
Na u al Gas 15.2 17.9 12.7 MWh
OUTPUTS: emissions o he en i onmen
Emissions o ai
NOx 35.0 56.74 56.74 kg
CO 4.27 9.63 9.63 kg
VOCs 201 172 172 kg
Fo maldehyde 1.36 0.48 0.48 kg
D. G anade o e al.
Cleane En i onmen al Sys ems 11 (2023) 100139
9
as also discussed by Finlayson-Pi s and Pi s (2000), Xu e al. (2015) and
Xu e al. (2016).
As shown in Fig. 2 (d), he ele ance o he pain s in mos impac
ca ego ies no ably inc eased in compa ison o he p ime p ocess. The
highes con ibu ions o he opcoa pain s can be obse ed in WD =
25.3 m
3
(63.1%), ALO =56.1 m
2
a (52.4%) and IR =93.0 kBq U235 eq.
(40.4%), whose ele ance can p esumably be aced back o he
manu ac u ing p ocess o he pain s and o hei cons i uen s.
As can be obse ed in he cha ac e iza ion esul s o all he sub-
p ocesses, na u al gas consump ion has a ele an in luence on almos
all impac ca ego ies and he e o e on he o al en i onmen al impac o
he pain ing p ocess. Fo ins ance, in he opcoa p ocess, CC =3897.4
kg CO
2
eq. (50.5%), FD =1338.2 kg oil eq. (43.8%) and NLT =0.55 m
2
(37.5%) a e he mos a ec ed impac ca ego ies. Mo eo e , na u al gas
plays an impo an ole in he OD impac ca ego y wi h a ele an
con ibu ion o 29.0%. The con ibu ions o he na u al gas o he CC, FD
and OD impac ca ego ies we e al eady epo ed in o he s udies (Skone
e al., 2016; Toma is e al., 2019).
P oposed aba emen scena ios: A cha ac e iza ion o he wo p oposed
scena ios conside ing he opcoa p ocess was pe o med and he esul s
compa ed o he ini ial si ua ion. Fo he majo i y o he impac ca e-
go ies, an inc ease can be obse ed when he RTO scena io is compa ed
o he ini ial si ua ion. On he con a y, mos o he ca ego ies dec ease
hei alues when he scena io wi h he ecupe a i e oxidize s is
compa ed o he ini ial si ua ion. The highes di e ences could be ound
o CC (ini ial si ua ion =7715 kg CO
2
eq; RTO =8407 kg CO
2
eq;
Recup. Oxidize s =7082 kg CO
2
eq), FD (ini ial si ua ion =3056 kg oil
eq.; RTO =3294 kg oil eq.; Recup. Oxidize s =2839 kg oil eq.), NLT
(ini ial si ua ion =1.48 m
2
; RTO =1.58 m
2
; Recup. Oxidize s =1.39
m
2
), IR (ini ial si ua ion =230.0 kBq U235 eq.; RTO =243.6 kBq U235
eq.; Recup. Oxidize s =217.6 kBq U235 eq.), OD (ini ial si ua ion =
1.036 ×10
−3
kg CFC-11 eq.; RTO =1.089 ×10
−3
kg CFC-11 eq.; Recup.
Oxidize s =0.987 ×10
−3
kg CFC-11 eq.), and TET (ini ial si ua ion =
0.647 kg 1,4-DB eq.; RTO =0.633 kg 1,4-DB eq.; Recup. Oxidize s =
0.572 kg 1,4-DB eq.). Rega ding he educ ion o he HT, he main
objec i e o he ins alla ion o he aba emen sys em, he esul s ob-
ained o he ini ial si ua ion and he wo scena ios a e: ini ial si ua ion
=1329 kg 1,4-DB eq.; RTO =1284 kg 1,4-DB eq.; Recup. Oxidize s =
1176 kg 1,4-DB eq. A educ ion in he HT ca ego y was achie ed by bo h
scena ios, wi h a dec ease being signi ican ly highe o he ecupe a i e
he mal oxidize s compa ed o he RTO. Fu he mo e, in he impac
ca ego ies ME (ini ial si ua ion =1.94 kg N eq.; RTO =2.81 kg N eq.;
Recup. Oxidize s =2.77 kg N eq.), PMF (ini ial si ua ion =14.46 kg
PM10 eq.; RTO =19.69 kg PM10 eq.; Recup. Oxidize s =18.83 kg PM10
eq.) and TA (ini ial si ua ion =41.0 kg SO
2
eq.; RTO =54.9 kg SO
2
eq.;
Recup. Oxidize s =51.6 kg SO
2
eq.), an inc emen was obse ed o bo h
p oposed sys ems bu a highe inc ease is caused by he RTO.
All in all, he cha ac e iza ion esul s indica e ha , om he wo
p oposed scena ios, he ins alla ion o an RTO is he less bene icial om
he en i onmen al impac pe spec i e. This esul can be explained by
he ac ha he ins alla ion o eigh ecupe a i e oxidize s implies an
o e all educ ion in na u al gas consump ion due o he ene gy eco e y
Fig. 2. Resul s o he LCI cha ac e iza ion. (a) S1: Pain ing p ocess; (b) SS1: Elec ocoa ing p ocess; (c) SS2: P ime p ocess; (d) SS3: Topcoa p ocess. Ac onyms used
in his igu e: CC: Clima e Change; OD: Ozone Deple ion; TA: Te es ial Acidi ica ion; FE: F eshwa e Eu ophica ion; ME: Ma ine Eu ophica ion; HT: Human
Toxici y; POF: Pho ochemical Oxidan Fo ma ion; PMF: Pa icula e Ma e Fo ma ion; TET: Te es ial Eco oxici y; FET: F eshwa e Eco oxici y; MET: Ma ine
Eco oxici y; IR: Ionizing Radia ion; ALO: Ag icul u al Land Occupa ion; ULO: U ban Land Occupa ion; NLT: Na u al Land T ans o ma ion; WD: Wa e Deple ion; MD:
Me al Deple ion; FD: Fossil Deple ion.
D. G anade o e al.
