Cap oic acid p oduc ion is linked o biomass ac i i y du ing
xylose e men a ion
Iglesias-Riob´
o J
*
, Mau icio-Iglesias M , Ca balla M
CRETUS, Depa men o Chemical Enginee ing, Uni e sidade de San iago de Compos ela, 15782 San iago de Compos ela, Spain
HIGHLIGHTS GRAPHICAL ABSTRACT
•Cap oic acid p oduc ion is a ou ed a
low biomass ac i i ies.
•Elec on dono s a e no ully consumed
a high biomass ac i i ies.
•Sequen ial eac o ope a ion leads o
highe cap oic acid yields.
•Sho HRT leads o non-consumed
xylose and no cap oic acid p oduc ion.
•High eed xylose concen a ion does no
a ec cap oic bu os e s bu y ic acid
yield.
ARTICLE INFO
Keywo ds:
CSTR
Hyd aulic e en ion ime
O ganic loading a e
SBR
Solid e en ion ime
ABSTRACT
This s udy aims a add essing inconsis encies in li e a u e ega ding he o ganic loading a e (OLR) and hyd aulic
e en ion ime (HRT) e ec on cap oic acid p oduc ion, using biomass ac i i y as an indica o . Xylose was ully
consumed in he e e ence CSTR and SBR (HRT =1-day; OLR =12 g COD/(L⋅d)), bu di e en cap oic acid yields
(0.02 s 0.11 Cmol/Cmol-s) we e obse ed, which was linked o di e ences in biomass ac i i y (12 s 3.5 g/(g
VSS⋅d)). A HRT 0.5 days, xylose con e sion was incomple e and lac ic acid and e hanol appea ed, educing
cap oic acid p oduc ion. Howe e , inc easing xylose concen a ion in he eeding o 24 g COD/L did no change
he cap oic acid yield (0.12 Cmol/Cmol-s), which was explained by simila biomass ac i i y as in he e e ence
SBR (4.8 g/(g VSS⋅d)). These indings indica e ha he SBR is he op imal con igu a ion, since i allows main-
aining a low biomass ac i i y and he e o e a high cap oic acid yield.
1. In oduc ion
Medium chain ca boxyla es (MCC) a e p omising p oduc s o eco e
o ganic ca bon h ough anae obic e men a ion since hey can be used
as eed addi i es o as p ecu so s o bio uel o biopolyme p oduc ion
(O-Thong e al., 2020; Sca bo ough e al., 2018). One o he mos
common MCC is cap oic acid, which can be ob ained h ough e e se
β-oxida ion o a y acid biosyn hesis (FAB) (Han e al., 2018). Bo h
pa hways consis o a edox p ocess in which a sho chain ca boxyla e
(SCC), such as bu y ic acid (C4), is elonga ed h ough an elec on dono
* Co esponding au ho .
E-mail add ess: [email p o ec ed] (I.-R. J).
Con en s lis s a ailable a ScienceDi ec
Bio esou ce Technology
jou nal homepage: www.else ie .com/loca e/bio ech
h ps://doi.o g/10.1016/j.bio ech.2024.131952
Recei ed 11 June 2024; Recei ed in e ised o m 17 Sep embe 2024; Accep ed 3 Decembe 2024
Bio esou ce Technology 418 (2025) 131952
A ailable online 4 Decembe 2024
0960-8524/© 2024 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 hanol o lac ic acid), which p o ides wo ca bon a oms o he
elonga ion o cap oic acid (C6). The la e means ha , o a ge cap oic
acid p oduc ion in a single-s age mixed-cul u e e men a ion p ocess,
he bio eac o ope a ional condi ions should p o ide he sui able win-
dow o he p oduc ion o he elec on dono and subsequen chain
elonga ion.
Se e al s udies epo ed ha sligh ly acidic condi ions (pH 5–6) we e
he op imal pH ange o he chain elonga ion p ocess as i is he bes
comp omise be ween he modynamics and mic obial compe i ion, as
chain elonga ion is less he modynamically a ou able unde alkaline
condi ions and a neu al pH he me hanogens could ou compe e wi h
chain elonga o s (De G oo e al., 2019), bu he p ope o ganic loading
a e (OLR) o HRT and solid e en ion ime (SRT) ange is s ill unce ain.
Since he elec on dono a ailabili y is one o he limi ing s eps, i s
ex e nal supplemen a ion eme ges as a good al e na i e o enhance
cap oic acid yield. Adding punc ually e hanol in a xylose e men a ion
sys em, he cap oic acid yield ose om 0.20 Cmol/Cmol-s o a ound
0.27 Cmol/Cmol-s (Tang e al., 2022), while cons an lac ic acid sup-
plemen a ion using xylan as ca bon sou ce yielded a ound 0.15 Cmol/
Cmol-s o cap oic acid a di e en OLR alues (Liu e al., 2022). How-
e e , his ex e nal addi ion en ails highe ope a ional cos s, so o he
s udies we e ocused on in-si u elec on dono p oduc ion.
Wi hou an exogenous elec on dono addi ion, cap oic acid (0.05
Cmol/Cmol-s) was only de ec ed in CSTR xylose e men a ion when
inc easing he HRT om 4.5 o 9 days. These esul s co esponded o an
OLR o 1.2 g COD/(L⋅d) (Wang e al., 2023), which is also consis en
wi h Qian e al. (2020), whe e cap oic acid was no de ec ed wi h an
HRT o 5 days (OLR 2.1 g COD/(L⋅d)) and Tang e al., (2022), whe e
cap oa e (0.20 Cmol/Cmol-s) was de ec ed wi h an OLR o 1.8 g COD/
(L⋅d). O e all, i seems ha a low OLR imp o e he cap oic acid
p oduc ion.
Howe e , o he s udies showed ha chain elonga ion s age was
boos ed when inc easing he OLR. Liu e al. (2022), using xylan and
lac ic acid as subs a es, obse ed ha ising he OLR om 2.9 o 11.6 g
COD/L⋅d (by sho ening he HRT om 8 o 2 days) led o he
imp o emen o MCC yields ( om 0.11 o 0.18 Cmol/Cmol-s), which is
consis en wi h Ra ay e al. (2022) e men ing glucose, in which chain
elonga ion p oduc s ose when he OLR inc eased om 12 o 28.8 g
COL/L⋅d (by inc easing glucose concen a ion in he eeding om 10 o
24 g COD/L).
A high OLR could be applied ei he by lowe ing he HRT (hyd aulic
o e load) o by inc easing he subs a e concen a ion in he eac o
eeding (o ganic o e load), and consequen ly, he implica ions a e
di e en . A hyd aulic o e load migh cause a biomass g ow h limi a ion
( eac o washou ) and he e o e a kine ic limi a ion (insu icien ime
o subs a e o in e media es con e sion). On he con a y, an o ganic
o e load migh cause he incomple e subs a e con e sion due o
p oduc inhibi ion o high cellula densi y inhibi ion. The e o e, bo h
o e loads migh d i e o di e ences in mic obial communi y composi-
ion and p oduc selec i i y.
The di e ences obse ed in li e a u e sugges ha he OLR migh be
incomple e, as i only akes in o accoun he subs a e ed and he HRT,
hus igno ing o he a iables ha may be c ucial, such as he SRT o
e en he subs a e eeding p o ile. Since he SRT and HRT a e coupled in
a CSTR eac o due o he pe manen mixing, he solid phase is e ained
he same ime as he liquid phase. In con as , he Sequen ial Ba ch
Reac o (SBR) eme ges as a good ope a ion al e na i e, as i gi es he
oppo uni y o decouple he HRT om he SRT by including a se le
phase in he cycle, allowing o e ain he biomass inside he eac o .
Fu he mo e, his eac o con igu a ion allows o al e na e high and low
subs a e concen a ions in he eac o (Rombou s e al., 2018), which
migh be in e es ing o MCC p oduc ion, since lac ic acid p oduc ion
was epo ed o be boos ed when a high subs a e concen a ion is
a ailable due o he as g ow h a e o he lac ic acid bac e ia
(Rombou s e al., 2020). Lago e al. (2023) epo ed ha decoupling he
SRT (30 days) om he HRT (10, 20 days) in a SBR eac o ea ing a
ca bohyd a e- ich esidue a pH 6 caused he dec ease o cap oic acid
p oduc ion compa ed o he CSTR ope a ion (4.1 % s 25 % w/w) in
a ou o bu y ic and lac ic acid, besides he same OLR (3 g COD/(L⋅d))
was applied in bo h eac o s, hus ein o cing ha he OLR is no he
p ope pa ame e o analyse he load impac on MCC p oduc ion.
The biomass ac i i y p o ides a be e desc ip ion o he eac o
pe o mance han HRT o OLR because i akes in o accoun no only he
load applied o he sys em bu also he biomass yield and he SRT. and
hus i helps o close he gap in unde s anding he e ec o di e en
loads on MCC p oduc ion du ing mixed-cul u e e men a ion. The e o e,
he objec i e o his wo k was o e alua e he in luence o biomass ac-
i i y on xylose e men a ion a ge ing cap oic acid. Fo ha pu pose,
wo eac o con igu a ions (CSTR and SBR) we e assessed a simila HRT
and OLR.
2. Ma e ials and me hods
2.1. Feeds ock desc ip ion
Same eeds ock composi ion was applied in bo h he con inuous and
he sequen ial ba ch eac o . The ca bon sou ce was a s anda d hemi-
cellulose monome (syn he ic xylose (I is Bio ech GMBH, Ge many)),
which was complemen ed wi h mac onu ien s and mic onu ien s (see
Suppl. Ma .). The eeds ock was kep a 4 ◦C o p e en xylose
deg ada ion.
2.2. Con inuous eac o and sequen ial ba ch eac o
A con inuous s i ed ank eac o (CSTR) and a sequen ial ba ch
eac o (SBR) we e ope a ed du ing 156 and 104 days, espec i ely, a
a ying ope a ional condi ions (Table 1). Bo h eac o s we e inocula ed
wi h he same biomass om a mesophilic anae obic sludge diges e (pH
7.4; HRT 54 d).
Two ope a ional s ages can be dis inguished in he con inuous
eac o , acco ding o he HRT applied, o assess he e ec o a hyd aulic
o e load: 1 day (CSTR-Re , OLR o 12 g COD/L⋅d) and 0.5 days (CSTR-
Hyd, OLR o 24 g COD/L⋅d).
The cycle o he SBR comp ised he ollowing ou phases: cha ge
phase (5 min), eac ion phase (du a ion dependen on HRT), se ling
phase (3 min) and discha ge phase (5 min). Th ee ope a ional s ages
we e applied o his eac o . Du ing he i s s age (SBR-Re ), he same
HRT (1 day) and OLR (12 g COD/L⋅d) as in CSTR-Re we e applied in
o de o compa e he wo eac o s. Once he pseudo-s eady s a e was
eached, he eac o was spli in wo o assess a hyd aulic o e load (by
dec easing he HRT o 0.5 days, SBR-Hyd) and an o ganic o e load (by
inc easing xylose concen a ion in he eeds ock o 24 g COD/L, SBR-
O g); inc easing in bo h cases he OLR o 24 g COD/(L⋅d).
Reac o s’pe o mance was moni o ed ia he measu emen o he
chemical oxygen demand (COD), ca boxyla es, soluble mic obial com-
pounds (SMP) (lac a e, succina e, e hanol and o ma e) and xylose
concen a ions wice a week, while he o al and suspended solids con-
cen a ions we e measu ed once a week. Fo he SBR mode, a cycle
cha ac e isa ion was pe o med when pseudo-s a e was eached in each
Table 1
Expe imen al condi ions o he eac o s’ope a ion.
CSTR-
Re
CSTR-
Hyd
SBR-
Re
SBR-
Hyd
SBR-
O g
Tempe a u e (◦C) 37 37 37 37 37
pH 6.0 6.0 6.0 6.0 6.0
Wo king olume (L) 1.0 1.0 1.2 1.2 1.2
Volume Exchange Ra io (%) – – 50 50 50
HRT (d) 1.0 0.5 1.0 0.5 1.0
Xylose eeding
concen a ion (g COD/L)
12 12 12 12 24
OLR (g COD/(L⋅d)) 12 24 12 24 24
I.-R. J e al. Bio esou ce Technology 418 (2025) 131952
2
ope a ional s age (SBR-Re : day 71, SBR-Hyd: day 99, SBR-O g: day 98).
2.3. Analy ical me hods
Con en ional physicochemical pa ame e s we e de e mined ac-
co ding o S anda d Me hods (APHA, 2017). Raw samples we e used o
calcula e he o al (TS and VS) and suspended (TSS, VSS) solids (SM2540
B, D, E) and he o al COD (SM5220C modi ied). Fil e ed samples we e
used o measu e soluble COD (SM5220C), ca boxyla es and SMP.
Ca boxyla es om C2 o C7 we e measu ed by gas ch oma og aphy
(Shimadzu UV-1800) wi h a DB-Wax column om Agilen Technologies
(30 m ×0.250 mm ×0.25 µm). The de ec o empe a u e was se a
300 ◦C while he injec o empe a u e was se a 200 ◦C. The ca ie gas
used was N
2
. The samples we e cen i uga ed and il e ed (0.45 µm), and
hen acidi ied wi h 10 µM o concen a ed H
3
PO
4
(85 %) be o e being
analysed.
SMP (lac ic acid, o mic acid, succinic acid, glyce ol and e hanol)
and xylose we e de e mined h ough high-pe o mance liquid ch oma-
og aphy (HPLC) acco ding o he GLEFG1 me hod wi h an HP 1100 (IR
HP1047A de ec o ). H
2
SO
4
(5 mM) as an isoc a ic eluen was used on he
AMINEX HPX-87H (300 ×7.8 mm) column, which was a 30 ◦C while
he de ec o was a 35 ◦C.
2.4. Calcula ions
The biomass exchange a io (λ) ep esen s he di e ence be ween
biomass concen a ion in he eac o and biomass concen a ion in he
e luen :
λ=X eac o
Xe luen
(1)
Whe e X
eac o
ep esen s he biomass concen a ion wi hin he eac o
(in g VSS/L) and X
e luen
he biomass concen a ion in he e luen (in g
VSS/L). In a CSTR, λis equal o one.
The p oduc and biomass yields indica e he biomass g ow h and
p oduc o ma ion pe xylose consumed, espec i ely, and, conside ing
ha he concen a ion o biomass in he in luen is negligible hey a e
calcula ed as ollows:
Biomass yield(Yx/s)(Cmol biomass
Cmol subs a e)=Cx,e luen
Cxyl,in luen −Cxyl,e luen
(2)
P oduc yield(Yp/s)(Cmol p oduc
Cmol subs a e)=Cp,e luen
Cxyl,in luen −Cxyl,e luen
(3)
whe e C
x,e luen
s ands o he biomass concen a ion in he e luen (in
Cmol/L), C
p,e luen
ep esen s he a ge ed p oduc concen a ion in he
e luen (in Cmol/L), C
xyl,in luen
e lec s o al xylose concen a ion in he
eeding o he eac o (in Cmol/L), C
xyl,e luen
cons i u es he xylose
concen a ion in he eac o e luen (in Cmol/L).
The a e age biomass ac i i y indica es he xylose consumed pe
biomass and pe day:
Biomass ac i i y(g xylose consumed
g VSS⋅d)=1
Yx/s⋅ ⋅HRT⋅λ(4)
whe e is a con e sion ac o be ween he di e en uni s equal o 0.75 g
xylose⋅Cmol-x/(g VSS ⋅Cmol-s).
The maximum biomass ac i i y in he SBR eac o ep esen s he
heo e ical maximum concen a ion o xylose he biomass can consume
pe day, and i is es ima ed as ollows:
Maximum biomass ac i i y(g xylose consumed
g VSS⋅d)=
ν
xyl
Xini ial
(5)
Whe e
ν
xyl
s ands o he a e age xylose consump ion a e (in g xylose/
(L⋅d)), which is calcula ed om he subs a e consump ion p o ile ob-
ained du ing cycle cha ac e isa ions, and X
ini ial
cons i u es he biomass
concen a ion in he eac o a he beginning o he cycle (in g VSS/L),
which is es ima ed as he di e ence be ween he mass o biomass inside
he eac o and he mass o biomass in he e luen di ided by he eac o
olume a he beginning o he eac ion phase (1.2 L).
3. Resul s and discussion
3.1. In luence o eac o con igu a ion in p oduc selec i i y
Du ing he i s s age (HRT 1 day; OLR 12 g COD/(L⋅d), xylose was
comple ely consumed in bo h eac o s (CSTR-Re and SBR-Re ), and a
simila concen a ion o ca boxyla es (7.5–8.5 g COD/L, Fig. 1 a,b) and
biomass in he e luen (0.8–1.0 g VSS/L, Fig. c,d) was ob ained. How-
e e , he biomass concen a ion inside he SBR was 3- old (Fig. 1d) he
biomass concen a ion inside he CSTR (Fig. 1c), inc easing he biomass
exchange a io and decoupling he SRT (3.2 d s 1.0 d) om he HRT.
Consequen ly, he biomass ac i i y in he SBR-Re (3.5 g/(g VSS⋅d),
Fig. 1 ) was much lowe han in CSTR-Re (12 g/(g VSS⋅d), Fig. 1e).
This di e en biomass ac i i y migh explain he p oduc spec a
ob ained in bo h eac o s (Fig. 2). The SBR-Re led o a highe cap oic
acid yield (0.11 s 0.02 Cmol/Cmol-s) (Fig. 2) in de imen o a lowe
bu y ic acid yield (0.19 s 0.28 Cmol/Cmol-s) compa ed o CSTR-Re ,
sugges ing ha lowe biomass ac i i ies os e he chain elonga ion
s age. This ac would be consis en wi h he esul s o Wang e al.,
(2023), who obse ed ha cap oic acid was only p oduced when HRT
was inc eased om 4.5 o 9 days, yielding 0.05 Cmol/Cmol-s wi h a
biomass ac i i y o 1.4 g/(g VSS⋅d). The simila cap oic acid yield ob-
ained in CSTR-Re (0.02 Cmol/Cmol-s) wi h mo e es ic i e condi ions
han s udies a longe HRT and lowe biomass ac i i ies (Qian e al.,
2020; Wang e al., 2023) sugges an easie xylose up ake by he
mic oo ganism conso ia p esen in his s udy.
A cycle cha ac e isa ion was pe o med in SBR-Re wi h he aim o
inding ou how subs a e and p oduc s e ol e du ing he cycle (Fig. 3a).
I can be obse ed ha xylose was apidly consumed (2.9 g COD xylose/
(L⋅h)) and con e ed mainly in o ace ic acid (3.4 g COD/L), bu y ic acid
(3.0 g COD/L) and cap oic acid (2.0 g COD/L). Lac ic acid (max. 1.5 g
COD/L) was p oduced du ing he i s 2 h and subsequen ly consumed
when xylose was no longe p esen in he eac o . A hese condi ions,
he maximum biomass ac i i y was 24 g/(g VSS⋅d) (Table 2), indica ing
ha he sys em was a om being o e loaded. In e es ingly, wo
p oduc p o iles can be dis inguished du ing he cycle: when xylose was
s ill p esen in he medium (0–2 h in e al) and when i was comple ely
consumed (2–12 h in e al). Du ing i s s age, 93 % o o al ace ic acid
p oduc ion was achie ed, sugges ing ha i is mainly p oduced di ec ly
om xylose. Bu y ic acid is p oduced equally in bo h s ages, which could
indica e ha is o med s aigh o wa dly om xylose con e sion, as
epo ed in Rombou s e al., (2018), and also by he lac a e-chain
elonga ion, as demons a ed by B odowski e al., 2022a. Cap oic acid
is mainly p oduced in he second s age (62 %), which leads o he
I.-R. J e al. Bio esou ce Technology 418 (2025) 131952
3
hypo hesis ha i s p oduc ion is dependen on he p io o ma ion o
o he p oduc s (i.e lac ic and bu y ic acid).
3.2. Hyd aulic o e load: does eac o con igu a ion ma e ?
The hyd aulic o e load led o high ins abili y in he CSTR ope a ion
(CSTR-Hyd), since xylose concen a ion in he e luen luc ua ed be-
ween 0 and 8 g COD/L, and he e o e, he ca boxyla es concen a ion
a ied acco dingly om 1 o 8 g COD/L (Fig. 1a). The ope a ion o he
SBR was mo e s able (SBR-Hyd), bu xylose con e sion and ca boxyla e
p oduc ion d opped as well (Fig. 1b). The la e indica es ha he low
HRT applied in bo h eac o s hampe ed subs a e con e sion in o ca -
boxyla es. In e es ingly, he biomass concen a ion inside he SBR
eac o dec eased signi ican ly (Fig. 1d), equalling he biomass con-
cen a ion in he e luen , and hus leading o he con e gence o SRT
and HRT. As a consequence, he biomass ac i i y in CSTR-Hyd and in he
SBR-Hyd we e simila (≈18 g/(g VSS⋅d)).
Despi e he same biomass ac i i y in bo h eac o s, he p oduc
spec a we e qui e di e en . Simila ace ic acid (0.27 s 0.30 Cmol/
Cmol-s), cap oic acid (no de ec ed) and lac ic acid yields (0.19 s 0.20
Cmol/Cmol-s) we e achie ed (Fig. 2), bu di e ences we e no iced in
bu y ic acid (0.21 s 0.07 Cmol/Cmol-s) and e hanol yields (0 s 0.06
Cmol/Cmol-s). As bu y ic acid was epo ed o be p oduced ei he
di ec ly om xylose (Rombou s e al., 2018) o by chain elonga ion o
ace a e (Spi i o e al., 2014), he inc ease o e hanol and lac ic acid
yields a e consis en wi h he d op o bu y ic acid yield and ha cap oic
was no longe p oduced in bo h ope a ions.
The hyd aulic o e load (SBR-Hyd) led o lowe (50 %) and slowe
(0.6 g COD xylose/(L⋅h)) xylose con e sion (Fig. 3b). Howe e , he
maximum biomass ac i i y inc eased un il 36 g/(g VSS⋅d) (Table 2),
Fig. 1. Xylose and ca boxyla es concen a ion, biomass concen a ion and biomass ac i i y du ing CSTR (a, c, e) and SBR (b, d, ) pe o mance.
I.-R. J e al. Bio esou ce Technology 418 (2025) 131952
4
which means he mic obial communi y can cope wi h highe o ganic
loads bu he e was no enough biomass concen a ion inside he eac o
a his HRT alue. Mo eo e , al hough lac ic acid was p esen as an end
p oduc , i s speci ic p oduc ion a e, calcula ed om lac ic acid p o ile
in cycle cha ac e isa ion, dec eased compa ed wi h SBR-Re om 8.7 o
6.9 g/(g VSS⋅d), sugges ing ha hese ope a ional condi ions did no
imp o e he selec i i y owa ds lac ic acid.
Fig. 2. P oduc spec a o di e en ope a ions o ganised om he lowes o he highes biomass ac i i y. Blue colou co esponds o SBR and g een colou o CSTR.
Fig. 3. Subs a e (Xyl) and p oduc (ace ic acid (HAc), bu y ic acid (HBu), cap oic acid (HCa), lac ic acid (HLa) and e hanol (E OH)) p o ile du ing one cycle in SBR-
Re (a) and SBR-Hyd (b).
Table 2
Biomass ac i i y, maximum biomass ac i i y and lac ic acid p oduc ion a e pe
uni biomass du ing one SBR cycle. Columns a e o ganised om he lowes o
highes biomass ac i i y.
Pa ame e SBR-Re SBR-O g SBR-Hyd
Biomass ac i i y (g/g VSS⋅d) 3.5 4.8 17
Maximum biomass ac i i y (g/g VSS⋅d) 24 22 36
Lac ic acid p oduc ion a e (g/g VSS⋅d) 8.7 10 6.9
I.-R. J e al. Bio esou ce Technology 418 (2025) 131952
5
3.3. O ganic o e load in SBR ope a ion (SBR-O g)
The cycle cha ac e isa ions demons a ed ha he SBR mode can
main ain high cap oic acid yields wi h a conside able OLR applied, bu
he sys em is oo sensi i e o SRT dec ease due o he dec ease in
biomass concen a ion inside he eac o . Howe e , i s ill emains un-
clea how he p ocess selec i i y would be a ec ed a highe o ganic
loads wi h enough biomass concen a ion in he eac o . The e o e, in
pa allel o SBR-Hyd, an o ganic o e load (SBR-O g) was assessed by
inc easing he xylose concen a ion in he eac o eeding om 12 o 24
g COD/L a HRT 1 day, hus equalling he OLR o he SBR–Hyd (24 g
COD/(L⋅d)).
Con e sely o SBR-Hyd, he o ganic o e load led o a ise o biomass
concen a ion (Fig. 4) in he eac o (4.5 g VSS/L) and in he e luen
(1.5–2 g VSS/L), co esponding o a SRT o 2.5 days. This SRT change
did no a ec biomass yield (0.12 Cmol x/Cmol-s) o subs a e con e -
sion, as xylose was ully consumed, leading o a biomass ac i i y o 4.8
g/(g VSS⋅d) (Fig. 1 ). Simila p oduc spec a han SBR-Re (Fig. 2) was
epo ed, excep o bu y ic acid yield, which was highe (0.29 s. 0.19
Cmol/Cmol-s).
Cycle cha ac e isa ion e ealed ha he o ganic o e load in SBR-O g
esul ed in a highe xylose consump ion a e (3.8 s 2.9 g COD xylose/
(L⋅h)), bu simila maximum biomass ac i i y (22 g/(g VSS⋅d), Table 2),
hus con i ming ha he sys em can assimila e ele a ed subs a e
concen a ions wi hou a p oduc inhibi ion. Lac ic acid eached a
highe maximum concen a ion (4.4 g COD/L, Fig. 5) han in SBR-Re
(1.5 g COD/L, Fig. 3), due o he speci ic lac ic acid p oduc ion a e
was simila (10 s 8.7 g/g VSS⋅d, Table 2). Li e a u e epo ed ha
lac a e could be oxidised o ace ic acid, con e ed o p opionic acid ia
ac yla e pa hway o used as elec on dono in e e se β-oxida ion
Fig. 4. Biomass concen a ion in he eac o and in he e luen o SBR-Re
and SBR–O g.
Fig. 5. Subs a e (Xyl) and p oduc (ace ic acid (HAc), bu y ic acid (HBu), cap oic acid (HCa), lac ic acid (HLa) and e hanol (E OH)) p o ile du ing one cycle o
SBR-O g.
Table 3
Summa y o main pa ame e s and esul s o CSTR and SBR ope a ions o ganised
om he lowes o he highes biomass ac i i y.
Pa ame e SBR-
Re
SBR-
O g
CSTR-
Re
SBR-
Hyd
CSTR-
Hyd
SRT (d) 3.2 2.3 1.0 0.5 0.5
Biomass exchange a io
(λ)
3.2 2.3 1.0 1.0 1.0
Biomass ac i i y (g/g
VSS⋅d)
3.5 4.8 12 17 18
Ca bon balance (%
Cmol)
98 101 100 103 106
Xylose con e sion (%) 100 100 100 61 66
Y
x/s
(Cmol/Cmol) 0.12 0.12 0.12 0.16 0.15
Y
Ca/s
(Cmol/Cmol) 0.11 0.12 0.02 0.00 0.00
Y
Bu/s
(Cmol/Cmol) 0.19 0.29 0.28 0.07 0.21
Fig. 6. Compa ison o biomass ac i i ies and chain elonga ion p oduc yield.
Black ma ke s (■) s ands o pH 6, blue ma ke s ( ) o pH 5.5, g een ma ke s
() o pH 5.4 and yellow ma ke s ( ) o pH 5.
I.-R. J e al. Bio esou ce Technology 418 (2025) 131952
6
p ocess o he p oduc ion o bu y ic and cap oic acid (B odowski e al.,
2022b; Cand y e al., 2020; Spi i o e al., 2014). I is hypo hesised ha
he highe xylose concen a ion ed in o he sys em led o a highe lac ic
acid concen a ion which inhibi ed ac yla e pa hway and lac a e
oxida ion in a ou o bu y ic acid p oduc ion, specially using mono-
saccha ides as subs a e (De G oo e al., 2020; Jankowska e al., 2018),
as he dec ease in odd ca boxyla es (p opionic and ale ic acid) and
ace ic acid yields (−0.1 Cmol/Cmol-s) is equal o he inc ease in bu y ic
acid yield (+0.1 Cmol/Cmol-s).
3.4. In luence o biomass ac i i y on chain elonga ion s age
Rega dless o eac o con igu a ion and ope a ional condi ions, a
close ela ion be ween biomass ac i i y and cap oic acid yield is
obse ed, as lowe biomass ac i i ies a e associa ed wi h highe cap oic
acid yields (Table 3).
To compa e hese indings wi h li e a u e, biomass ac i i y was
calcula ed o some s udies e men ing xylose o glucose (Qian e al.,
2020; Ra ay e al., 2022; Tang e al., 2022; Wang e al., 2023), and
ela ed o chain elonga ion p oduc yield (Fig. 6). Cap oic acid and
ale ic acid we e conside ed he e as chain elonga ion p oduc s as ale ic
acid canno be p oduced di ec ly by he acidi ica ion o ei he xylose o
glucose. O e all, i is con i med ha he e exis a biomass ac i i y
window in which chain elonga ion p ocess is a ou ed, ega dless o pH
(be ween 5 and 6) and subs a e (glucose o xylose). Ye , he highes
cap oic acid yields we e ob ained a a ious pH, HRT and eac o mode
condi ions, which sugges ha he op imal ope a ional condi ions o
chain elonga ion may a y depending on subs a e cha ac e is ics.
4. Conclusions
The in e play be ween biomass ac i i y and cap oic acid yield in
xylose mixed-cul u e e men a ion was assessed in his s udy. Highe
cap oic acid yields we e ob ained a lowe biomass ac i i ies, ega dless
he OLR applied o he eac o con igu a ion. The hyd aulic o e load
induces eac o ope a ion ins abili y, leading o a highe biomass ac-
i i y and o he appea ance o elec on dono s as end p oduc s. In
con as , SBR demons a ed esilience o o ganic o e load wi hou
al e ing cap oic acid yield. O e all, his s udy b ings mo e ligh o he
in e ac ions among HRT, SRT and OLR o op imise he medium chain
ca boxyla es p oduc ion.
CRediT au ho ship con ibu ion s a emen
Juan Iglesias-Riob´
o: concep ualiza ion, me hodology, in es iga-
ion, w i ing –O iginal d a , W i ing –Re iew &edi ing. Miguel
Mau icio-Iglesias: concep ualiza ion, me hodology, w i ing –O iginal
d a , W i ing –Re iew &edi ing, supe ision. Ma a Ca balla:
concep ualiza ion, me hodology, esou ces, w i ing –O iginal d a ,
W i ing –Re iew &edi ing, supe ision, p ojec adminis a ion, unding
acquisi ion.
Decla a ion o compe ing in e es
The au ho s decla e he ollowing inancial in e es s/pe sonal e-
la ionships which may be conside ed as po en ial compe ing in e es s:
[Ma a Ca balla epo s inancial suppo was p o ided by Eu opean
Regional De elopmen Fund. Juan Iglesias-Riob´
o, Miguel Mau icio
Iglesias and Ma a Ca balla epo inancial suppo was p o ided by
Xun a de Galicia. I he e a e o he au ho s, hey decla e ha hey ha e
no known compe ing inancial in e es s o pe sonal ela ionships ha
could ha e appea ed o in luence he wo k epo ed in his pape ].
Acknowledgemen s
This esea ch was amed in CELL4CHEM p ojec (PCI2021-121989,
ERACoBioTech 3 d call) unded by MICIU/AEI/10.13039/
501100011033 and he Eu opean Union Nex Gene a ionEU/PRTR. The
au ho s belong o a Galician Compe i i e Resea ch G oup (ED431C-
2021/37)
Appendix A. Supplemen a y da a
Supplemen a y da a o his a icle can be ound online a h ps://doi.
o g/10.1016/j.bio ech.2024.131952.
Da a a ailabili y
Da a will be made a ailable on eques .
Re e ences
Apha, 2017. S anda d me hods o he examina ion o wa e and was ewa e . Ame ican
Public Heal h Associa ion, Washing on.
B odowski, F., Łę˙
zyk, M., Gu owska, N., Kabasakal, T., Oleskowicz-Popiel, P., 2022a.
In luence o lac a e o ace a e a io on biological p oduc ion o medium chain
ca boxyla es ia open cul u e e men a ion. Sci. To al En i on. 851. h ps://doi.o g/
10.1016/j.sci o en .2022.158171.
B odowski, F., Łę˙
zyk, M., Gu owska, N., Oleskowicz-Popiel, P., 2022b. E ec o ex e nal
ace a e on lac a e-based ca boxyla e pla o m: Shi ed lac a e o e loading limi and
hyd ogen co-p oduc ion. Sci. To al En i on. 802. h ps://doi.o g/10.1016/j.
sci o en .2021.149885.
Cand y, P., Radi´
c, L., Fa e e, J., Ca ajal-A oyo, J.M., Rabaey, K., Ganigu´
e, R., 2020.
Mildly acidic pH selec s o chain elonga ion o cap oic acid o e al e na i e
pa hways du ing lac ic acid e men a ion. Wa e Res 186. h ps://doi.o g/10.1016/
j.wa es.2020.116396.
De G oo , V., Coma, M., A no , T., Leak, D.J., Lanham, A.B., 2019. Medium chain
ca boxylic acids om complex o ganic eeds ocks by mixed cul u e e men a ion.
Molecules 24. h ps://doi.o g/10.3390/molecules24030398.
De G oo , V., Coma, M., A no , T.C., Leak, D.J., Lanham, A.B., 2020. Adjus ing o ganic
load as a s a egy o di ec single-s age ood was e e men a ion om anae obic
diges ion o chain elonga ion. P ocesses 8, 1–18. h ps://doi.o g/10.3390/
p 8111487.
Han, W., He, P., Shao, L., Lü, F., 2018. Me abolic in e ac ions o a chain elonga ion
mic obiome. Appl. En i on. Mic obiol. 84. h ps://doi.o g/10.1128/AEM.01614-18.
Jankowska, E., Dube , A., Chwialkowska, J., S odolny, M., Oleskowicz-Popiel, P., 2018.
Con e sion o o ganic was e in o ola ile a y acids –The in luence o p ocess
ope a ing pa ame e s. Chem. Eng. J. 345, 395–403. h ps://doi.o g/10.1016/j.
cej.2018.03.180.
Lago, A., G eses, S., Aboudi, K., Mo eno, I., Gonz´
alez-Fe n´
andez, C., 2023. E ec o
decoupling hyd aulic and solid e en ion imes on ca bohyd a e- ich esidue
alo iza ion in o ca boxylic acids. Sci. Rep. 13, 20590. h ps://doi.o g/10.1038/
s41598-023-48097-2.
Liu, B., S ¨
aube , H., Sa ai a, J., Ha ms, H., Sil a, S.G., Kasmanas, J.C., Kleins eube , S.,
Nunes da Rocha, U., 2022. Machine lea ning-assis ed iden i ica ion o bioindica o s
p edic s medium-chain ca boxyla e p oduc ion pe o mance o an anae obic mixed
cul u e. Mic obiome 10. h ps://doi.o g/10.1186/s40168-021-01219-2.
O-Thong, S., Zhu, X., Angelidaki, I., Zhang, S., Luo, G., 2020. Medium chain a y acids
p oduc ion by mic obial chain elonga ion: Recen ad ances, in: Ad ances in
Bioene gy. Else ie Inc., pp. 63–99. h ps://doi.o g/10.1016/bs.aibe.2020.04.002.
Qian, D.K., Geng, Z.Q., Sun, T., Dai, K., Zhang, W., Jianxiong Zeng, R., Zhang, F., 2020.
Cap oa e p oduc ion om xylose by mesophilic mixed cul u e e men a ion.
Bio esou . Technol. 308. h ps://doi.o g/10.1016/j.bio ech.2020.123318.
Ra ay, R., Allegue, T., Fowle , S.J., Rod íguez, J., 2022. Explo ing he limi s o
ca bohyd a e con e sion and p oduc o ma ion in open mixed cul u e e men a ion.
J. En i on. Chem. Eng. 10. h ps://doi.o g/10.1016/j.jece.2022.107513.
Rombou s, J.L., Mos, G., Weissb od , D.G., Klee ebezem, R., Van Loosd ech , M.C.M.,
2018. Di e si y and me abolism o xylose and glucose e men ing mic obial
communi ies in sequencing ba ch o con inuous cul u ing. FEMS Mic obiol. Ecol. 95.
h ps://doi.o g/10.1093/ emsec/ iy233.
Rombou s, J.L., K anendonk, E.M.M., Reguei a, A., Weissb od , D.G., Klee ebezem, R.,
an Loosd ech , M.C.M., 2020. Selec ing o lac ic acid p oducing and u ilising
bac e ia in anae obic en ichmen cul u es. Bio echnol. Bioeng 117, 1281–1293.
h ps://doi.o g/10.1002/bi .27301.
I.-R. J e al. Bio esou ce Technology 418 (2025) 131952
7
Sca bo ough, M.J., Lynch, G., Dickson, M., McGee, M., Donohue, T.J., Nogue a, D.R.,
2018. Inc easing he economic alue o lignocellulosic s illage h ough medium-
chain a y acid p oduc ion. Bio echnol. Bio uels 11. h ps://doi.o g/10.1186/
s13068-018-1193-x.
Spi i o, C.M., Rich e , H., Rabaey, K., S ams, A.J.M., Angenen , L.T., 2014. Chain
elonga ion in anae obic eac o mic obiomes o eco e esou ces om was e. Cu .
Opin. Bio echnol. h ps://doi.o g/10.1016/j.copbio.2014.01.003.
Tang, J., Dai, K., Wang, Q.-T., Zheng, S.-J., Hong, S.-D., Jianxiong Zeng, R., Zhang, F.,
2022. Cap oa e p oduc ion om xylose ia he a y acid biosyn hesis pa hway by
genus Cap oicip oducens domina ed mixed cul u e e men a ion. Bio esou .
Technol. 351, 126978. h ps://doi.o g/10.1016/j.bio ech.2022.126978.
Wang, Q., Dai, K., Tang, J., Hong, S., Zheng, S., Sun, T., Zeng, R.J., Zhang, F., 2023. Low-
empe a u e cap oa e p oduc ion, mic obial di e si y, and me abolic pa hway in
xylose anae obic e men a ion. F on En i on. Sci. Eng. 17. h ps://doi.o g/
10.1007/s11783-023-1637-9.
I.-R. J e al. Bio esou ce Technology 418 (2025) 131952
8
