scieee Science in your language
[en] (orig)

Impact of dissolved sulfide on a hybrid membrane bioreactor treating the effluent of a mainstream up-flow anaerobic sludge blanket

Author: Allegue Martínez, Tomás Francisco; Arias Baño, Adrián; Liñares Lamas, Alberto; Omil Prieto, Francisco; Garrido Fernández, Juan Manuel
Publisher: Royal Society of Chemistry
Year: 2023
DOI: 10.1039/d3ew00404j
Source: https://minerva.usc.es/bitstreams/9f7f4810-0f6c-4b32-86fb-b7896f8875ad/download
En i onmen al
Science
Wa e Resea ch & Technology
PAPER
Ci e his: En i on. Sci.: Wa e Res.
Technol.,2023,9,2733
Recei ed 2nd June 2023,
Accep ed 30 h Augus 2023
DOI: 10.1039/d3ew00404j
sc.li/es-wa e
Impac o dissol ed sul ide on a hyb id memb ane
bio eac o ea ing he e luen o a mains eam
up- low anae obic sludge blanke †
Tomás Allegue, *Ad ián A ias, Albe o Liña es,
F ancisco Omil and Juan Manuel Ga ido
Despi e being oxic o some mic obes in was ewa e ea men , sul ide can also p omo e ni ogen emo al
h ough sul ide-oxidizing bac e ia. This s udy e alua es he dissol ed sul ide impac on a hyb id MBR
ea ing he e luen o a mains eam UASB. A UASB-MBR (176 L) was ed wi h syn he ic domes ic sewage
and ope a ed o 154 days. Two pe iods we e dis inguished, one wi hou (Pe iod I) and one wi h (Pe iod II)
sul ide dissol ed in he UASB e luen . Dissol ed me hane, COD, ni ogen, and o ganic mic opollu an s
(OMP)s emo als accomplished in he MBR du ing bo h pe iods we e compa ed. Ini ially, sul ide inhibi ed
me hane emo al, bu once ully oxidized in o sul a e in he anoxic compa men , he e iciencies
eco e ed o simila le els as wi hou sul ide (>70%). Sul ide addi ions signi ican ly enhanced he MBR
deni i ica ion po en ial h ough sul ide-oxidizing bac e ia, wi h imp o ed emo als in Pe iod II (63.4 TN
L
eed
−1
) compa ed o Pe iod I (40 mg TN L
eed
−1
). Mos o he ni ogen emo al occu ed in he anoxic
compa men o he MBR, howe e , up o 21% o he ni ogen was deni i ied in he ae obic compa men
wi hin he bio ilm ca ie s. Ae obic me hane oxida ion coupled wi h deni i ica ion, he e o ophic
deni i ie s, sul ide oxida ion, and anammox p ocesses we e in ol ed in he ni ogen emo al. COD and
OMPs emo als we e no a ec ed by sul ide.
1. In oduc ion
The up- low anae obic sludge blanke (UASB) echnology is a
popula choice o ea ing municipal was ewa e in wa m
clima es due o i s bene i s, such as ene gy eco e y, and
lowe sludge p oduc ion and ope a ing cos s compa ed o
con en ional ae obic p ocesses. I is widely used in wa m
egions, whe e hund eds o ull-scale UASB eac o s a e
ea ing domes ic sewage.
1
Addi ional pos - ea men is
equi ed due o he poo quali y o UASB e luen s wi h
ega ds o o ganic ma e , solids, ammonium, and dissol ed
me hane. Despi e he high me hane concen a ion in he
biogas,
2
which anges om 70 o 80%, no all he me hane is
p esen in he biogas, and be ween 20 and 60% could be
p esen dissol ed in he e luen .
3
This dissol ed me hane
can easily be eleased in o he en i onmen , leading o i s
con ibu ion o clima e change as me hane is a po en
g eenhouse gas.
4
Thus, he e is an u gen need o me hods
o eco e o emo e he dissol ed me hane.
Unlike con en ional ae obic ea men s, anae obic
p ocesses a e well-known o hei negligible ni ogen
emo al capaci y. Typically, UASB e luen s con ain
concen a ions anging om 30 o 50 mg TN L
−1
.
5
The
p esence o ni ogen in hei e luen s could lead o se ious
en i onmen al p oblems such as eu ophica ion o oxici y in
he ecei ing wa e bodies. In case he ea ed e luen is no
used o eusing pu poses (e.g., i iga ion), ni ogen emo al
mus be he e o e add essed. Ac oss he globe, discha ge
s anda ds a e becoming inc easingly s ingen , wi h limi s
En i on. Sci.: Wa e Res. Technol., 2023, 9, 2733–2744 | 2733This jou nal is © The Royal Socie y o Chemis y 2023
CRETUS, Depa men o Chemical Enginee ing, Uni e sidade de San iago de
Compos ela, 15782 San iago de Compos ela, Galicia, Spain.
E-mail: omas.ma [email protected], ad ian.a [email protected],
albe olinha [email protected], [email protected],
juanmanuel.ga [email protected]
†Elec onic supplemen a y in o ma ion (ESI) a ailable. See DOI: h ps://doi.o g/
10.1039/d3ew00404j
Wa e impac
This s udy assesses he impac o dissol ed sul ide on a no el hyb id MBR (176 L) ea ing he e luen om a mains eam UASB. The key inding o he
s udy is ha dissol ed sul ide no only did no hinde he MBR's deni i ica ion po en ial bu signi ican ly imp o ed i . Addi ionally, sul ide had no e ec
on he emo al o COD, dissol ed CH
4
, o mic opollu an s.
Open Access A icle. Published on 30 Augus 2023. Downloaded on 11/8/2023 9:09:27 AM.
This a icle is licensed unde a
C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence.
View A icle Online
View Jou nal
| View Issue
2734 |En i on. Sci.: Wa e Res. Technol., 2023, 9, 2733–2744 This jou nal is © The Royal Socie y o Chemis y 2023
o o al ni ogen (TN) as low as 10 mg TN L
−1
in sensi i e
a eas o he Eu opean Union (Council Di ec i e 91/271/ EEC).
In B isbane (Aus alia), he limi s a e e en igh e a 3 mg TN
L
−1
.
6
To mee ni ogen discha ge egula ions, domes ic
sewage ea men plan s o en add ex e nal ca bon sou ces,
such as e hanol, me hanol, o ace ic acid, o enhance
con en ional deni i ica ion p ocesses. Howe e , his
app oach has limi a ions, as he high cos o hese ca bon-
based compounds and he esul ing la ge amoun o sludge
p oduc ion make i unsus ainable and expensi e.
Addi ionally, he issue o dissol ed me hane in hei e luen
is no esol ed.
U ilizing dissol ed me hane and ni ogen in UASB e luen
o e s a scope o c ea ing inno a i e pos - ea men
s a egies ha employ me hane as a po en ial elec on dono
o ni ogen emo al h ough biological p ocesses. Gi en he
low C/N a io in UASB e luen s, using me hane as an elec on
dono could be a highly appealing app oach o emo ing
ni ogen om was ewa e wi hou he need o cos ly ex e nal
ca bon sou ces, esul ing in mo e en i onmen ally sus ainable
and e icien sewage ea men p ocesses.
7–11
Va ious mic obiological pa hways exis o he coupling o
me hane oxida ion wi h deni i ica ion p ocesses. In he
ae obic me hane oxida ion coupled o deni i ica ion (AMO-
D), ae obic me hano ophs can con e me hane in o
oxida ion p oduc s (e.g., me hanol), ha can be used by
con en ional deni i ie s (eqn (1) and (2)).
12
The emo al o
dissol ed me hane and ni ogen can also be achie ed
simul aneously unde anae obic condi ions ia ni a e/ni i e-
dependen anae obic me hane oxida ion biop ocesses,
e e ed o as N-damo. This p ocess in ol es wo g oups o
mic oo ganisms: N-damo a chaea and N-damo bac e ia.
N-damo a chaea can oxidize me hane in o ca bon dioxide by
educing ni a e o ni i e,
13
while N-damo bac e ia can
oxidize me hane by educing ni i e in o N
2
.
14
5CH
4
+5O
2
+ 4NO
3
−
+4H
+
→2N
2
+ 12H
2
O + 5CO
2
(1)
3CH
4
+3O
2
+ 4NO
2
−
+4H
+
→2N
2
+8H
2
O + 3CO
2
(2)
Besides me hane, sul ide is p esen dissol ed in e luen s
o anae obic eac o s wi h concen a ions ha could each up
o 97.5 mg S L
−1
.
15
The sul a e ion is educed o hyd ogen
sul ide in sewe age sys ems and anae obic eac o s h ough
sul a e- educing bac e ia (SRB), using o ganic ma e as he
elec on dono . The sul ide concen a ion mainly depends on
he sul a e con en in esh sewage, which could ha e ei he
an an h opic o na u al o igin
16
o be caused by seawa e
in usion in asho e sewe age sys ems.
17
Sul ide is esponsible
o se e al en i onmen al p oblems such as equipmen
co osion, odo nuisances, de e io a ion o he ecei ing
wa e bodies due o i s chemical oxygen demand (COD)
con ibu ion, and inhibi o y e ec on biological p ocesses.
7
Sul ide concen a ions o only 2.6 and 1.2 mg S L
−1
caused
50% inhibi ion (IC
50
) owa ds ammonia (AOB) and ni i e-
oxidizing bac e ia (NOB) ac i i y, espec i ely.
18
In ano he s udy, sul ide concen a ions o only 0.04 and
0.1 mg H
2
S–SL
−1
inhibi ed he N
2
O educ ion capaci y o
con en ional he e o ophic deni i ie s by 50%.
19
The i s
alue was achie ed using biomass unacclima ed o
en i onmen s wi h sul ide, while he la e was a ained using
biomass acclima ed o sul ide, which could explain he
highe ole ance. The inhibi ions le els obse ed we e
s ongly co ela ed wi h he concen a ion o H
2
S a he han
he o al sul ide. In he same s udy, a educ ion o 50% in
he ni i e educ ion capaci y was obse ed a 2 mg H
2
S–SL
−1
o bo h, unacclima ed and acclima ed biomass.
The po en ial impac o sul ide on ce ain
mic oo ganisms could be signi ican in he case o
implemen ing addi ional biological pos - ea men sys ems
o ea anae obic e luen . Howe e , sul ide may also be
u ilized by au o ophic sul u -oxidizing bac e ia (SOB) as an
elec on dono o deni i ica ion, which could enhance he
ni ogen emo al capabili ies o UASB pos - ea men
sys ems. SOB can use educed sul u compounds, such as
sul ide, as elec on dono s o ans o m ni i e/ni a e in o
N
2
(eqn (3) and (4)).
3HS
−
+ 8NO
2
−
+5H
+
→3SO
42−
+4N
2
+4H
2
O (3)
5HS
−
+ 8NO
3
−
+3H
+
→5SO
42−
+4N
2
+4H
2
O (4)
The use o memb ane bio eac o s (MBR)s has been
conside ed by se e al au ho s as an in e es ing app oach o
ea e luen s om mains eam UASB sys ems,
8,10,20,21
o
wo main easons: i) he high quali y o he e luen ( ee o
suspended solids) and ii) o gua an ee he comple e e en ion
o slow-g ow h deni i ie s such as anammox, N-damo
bac e ia, and N-damo a chaea, mic obes o g ea in e es
conside ing he low concen a ion o COD p esen in he
e luen o UASB sys ems o deni i y.
Sil a-Tei a e al.
8
in es iga ed he use o a p e-anoxic/
ae obic MBR sys em (56 L) o minimize he impac o he
e luen om a UASB (120 L) ea ing syn he ic domes ic
sewage a oom empe a u e. The pos - ea men sys em was
composed o a i s mechanically s i ed anoxic compa men
wi h bio ilm ca ie s and a second ae obic memb ane
il a ion compa men . In he la e , ammonium was
oxidized, and suspended biomass/e luen sepa a ion was
ca ied ou . A eci cula ion om he ae obic o he anoxic
compa men was needed o p o ide an elec on accep o o
ni ogen emo al. Ni a e was he main elec on accep o
a ailable, and ni i e was ba ely p esen . Me hane emo al
e iciencies o 80% and ni ogen emo als be ween 10 and 15
mg N L
Feed
−1
we e achie ed in he anoxic compa men o he
MBR.
Using he same echnology and simila ope a ing
condi ions, al hough wi h he p esence o ni i e and ni a e
in he anoxic compa men , Al a ino e al.
10
achie ed much
highe ni ogen emo als in he anoxic compa men , 35 mg
TN L
Feed
−1
. The au ho s sugges ha a highe ni i e p esence
in he anoxic compa men p omo ed he g ow h o
En i onmen al Science: Wa e Resea ch & TechnologyPape
Open Access A icle. Published on 30 Augus 2023. Downloaded on 11/8/2023 9:09:27 AM.
This a icle is licensed unde a
C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence.
View A icle Online
En i on. Sci.: Wa e Res. Technol., 2023, 9, 2733–2744 | 2735This jou nal is © The Royal Socie y o Chemis y 2023
anammox. Ne e heless, me hane emo al e iciencies o
a es we e simila o Sil a-Tei a e al.
8
In he p esen s udy, he impac o dissol ed sul ide on
he pe o mance o a hyb id memb ane bio eac o (MBR)
ea ing he e luen o a mains eam UASB is e alua ed. A
pilo -scale UASB-MBR in eg a ed sys em (176 L) ed wi h low-
s eng h syn he ic sewage and con inuously ope a ed a
ambien empe a u e was employed. This s udy in oduces a
echnological enhancemen o he hyb id MBR when
compa ed o ou p io p o o ype.
8,10
This imp o emen
en ails he inclusion o an ex a ae obic chambe u ilizing
suspended bio ilm ca ie s (Biochip). I s p ima y pu pose is
o acili a e deni i ica ion p ocesses, as demons a ed by
Allegue e al.
9
Two main ope a ing pe iods we e
dis inguished based on he absence o p esence o dissol ed
sul ide in he UASB e luen . In he i s pe iod, no dissol ed
sul ide was p esen in he UASB e luen , while in he second
pe iod, dissol ed sul ide was in oduced by adding sul a e o
he UASB eeding subs a e. The wo pe iods we e compa ed,
wi h a ocus on he elimina ion o o ganic ma e , dissol ed
me hane, ni ogen, and o ganic mic opollu an s (OMP)s in
he MBR pos - ea men sys em. Addi ionally, mic obial
communi ies we e in es iga ed.
2. Me hods
2.1 Reac o se -up
In his s udy, a pilo -scale UASB-MBR in eg a ed sys em o
176 L was ope a ed (Fig. 1). The sys em consis ed o a
me hanogenic UASB eac o (120 L) coupled in se ies o a
hyb id MBR pos - ea men sys em (56 L) wi h h ee di e en
compa men s: i) a i s anoxic (36 L); ii) a second ae obic (8
L), and a hi d ae obic il a ion uni (12 L) wi h a subme ged
hollow- ibe memb ane. The anoxic compa men was mixed
mechanically, while he ae obic and il a ion compa men s
we e mixed pneuma ically using ae a ion. The o ganic ma e
in he UASB is ans o med in o biogas unde anae obic
condi ions, while he simul aneous emo al o me hane and
ni ogen occu s in he anoxic compa men . In he MBR, he
ae obic compa men acili a es he con e sion o
ammonium in o oxidized ni ogen species (NOx), while he
il a ion memb ane compa men gua an ees biomass
e en ion.
The sys em u ilized wo di e en eci cula ion me hods,
including an in e nal eci cula ion om he il a ion
memb ane o he anoxic compa men wi h a ecycling a io
(R) o 3.5 and an ex e nal eci cula ion om he il a ion
memb ane o he ae obic compa men (R= 1). The in e nal
ecycling a io o 3.5 was consis en ly applied o op imize he
emo al o bo h me hane and ni ogen, as demons a ed in a
s udy by Sil a-Tei a e al.
8
in a simila UASB-MBR. A ZW-10
Zenon ul a il a ion memb ane module, ea u ing a po e size
o 0.04 μm and a o al su ace a ea o 0.9 m
2
, was placed in
he il a ion compa men . The memb ane ope a es in cycles
o 7 minu es, wi h a 0.5 min elaxa ion pe iod.
The UASB-MBR sys em combines h ee dis inc edox
en i onmen s (anae obic, anoxic, and ae obic) and h ee
ypes o biomass (g anula in he UASB, adhe ed o ca ie s,
and suspended biomass in he MBR). F esh and ac i e
biomass om a simila pilo plan con igu a ion was used as
inoculum o bo h compa men s o he sys em (UASB and
MBR); he e o e, no s abiliza ion ime was equi ed.
10
In
addi ion o suspended biomass, he g ow h o adhe ed
biomass in he anoxic and ae obic compa men s was
encou aged by adding bio ilm ca ie s called Biochip. The
Biochips ha e an ex e nal su ace a ea o 2174 m
2
m
−3
and a
Fig. 1 The diag am p esen s a compa ison o he p e ious ( op) and new (bo om) designs o he UASB-MBR. The new design is he one used in
he p esen s udy, whe eas he old design was used in p e ious s udies.
8,10
En i onmen al Science: Wa e Resea ch & Technology Pape
Open Access A icle. Published on 30 Augus 2023. Downloaded on 11/8/2023 9:09:27 AM.
This a icle is licensed unde a
C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence.
View A icle Online
2736 |En i on. Sci.: Wa e Res. Technol.,2023,9,2733–2744 This jou nal is © The Royal Socie y o Chemis y 2023
olume pe ca ie pa icle o 4.15 ×10
−4
L, allowing o a
highe ni i ica ion capaci y pe uni olume as he high
su ace a ea acili a es oxygen ans e . Allegue e al.
9
obse ed
imp o ed me hane and ni ogen emo al due o he
accumula ion o me hane oxidize s and anammox bac e ia
adhe ed o he Biochips in he ae obic compa men . Biochips
we e also added o he anoxic compa men o p omo e he
g ow h o anae obic mic obes, such as N-damo, ha migh be
sensi i e o he oxygen in he in e nal eci cula ion s eam.
The appa en olume o Biochips in he anoxic and ae obic
compa men s was 17% and 20%, espec i ely.
In his s udy, modi ica ions we e made o he design o
he UASB-MBR sys em compa ed o he p e ious
p o o ype,
8,10
as shown in Fig. 1. Fi s ly, he p e ious ae obic
compa men , which housed he subme ged memb ane
module, was spli in o wo sepa a e compa men s: he
memb ane il a ion compa men and he ae obic
compa men . The addi ion o an ae obic compa men
allows o he inco po a ion o Biochip ca ie s, which ha e
been shown o imp o e me hane and ni ogen emo al in he
MBR.
9
The objec i e o employing Biochip ca ie s in he
ae obic compa men is o acili a e he p oli e a ion o
deni i ie s and me hane oxidize s wi hin he bio ilm.
Addi ionally, Biochip we e p e e ed o e Le apo ca ie s o
u iliza ion in he anoxic compa men o he MBR. Ea lie
s udies using Le apo demons a ed ha he mic obial
communi ies in he suspended biomass and adhe ed
biomass we e simila , implying ha Le apo does no
encou age he o ma ion o bio ilms wi h dis inc
a ibu es.
10
The UASB was ed wi h low-s eng h syn he ic was ewa e
composed o a concen a ed medium (con aining skimmed
milk, NaHCO
3
,NH
4
Cl, ace elemen s, and OMPs) dilu ed
wi h ap wa e , and wi h he ollowing inal composi ion:
NaHCO
3
(200 mg L
−1
); NH
4
Cl (9.3 mg N L
−1
); ace elemen s
concen a ions and composi ion as indica ed by Sil a-Tei a
e al.;
8
10 μgL
−1
o 15 di e en o ganic mic opollu an s
(OMPs), including endoc ine dis up o s (bisphenol BPA,
es one E1, β-es adiol E2, α-e hinyles adiol EE2), an i-
in lamma o ies (ibup o en IBP, nap oxen NPX, diclo enac
DCF), a disin ec an ( iclosan TCS), an ibio ics
(sul ame hoxazole SMX, ime hop im TMP, e y h omycin
ERY, oxi h omycin ROX), and neu o d ugs (diazepam DZP,
ca bamazepine CBZ, luoxe ine FLX). Collapsible silicone
bags we e used o house he concen a ed medium, and
e ige a ion was employed o ensu e i s p ese a ion.
The s udy was di ided in o wo main ope a ing pe iods,
based on he absence (days 0–70) o p esence (days 71–154)
o sul a e in he eed, e e ed o as Pe iods I and II,
espec i ely. Du ing Pe iod II, 50 mg SO
42−
–SL
−1
was added
o he eed as Na
2
SO
4
.
2.2 Analy ical me hods
S anda d me hods
22
we e used o measu e o al (COD
T
) and
soluble chemical oxygen demand (COD
S
), ni ogen species
(ni i e, ni a e, and ammonium), mixed-liquo o al
(MLTSS), and ola ile suspended solids (MLVSS), pH, and
empe a u e. The me hane dissol ed in he liquid phase
was de e mined aking samples o app oxima ely 300 mL in
wa e igh 500 mL Py ex lasks. Exsol ed me hane o gas
phase was analyzed by Gas Ch oma og aphy (GC), as
desc ibed by Sil a-Tei a e al.
8
A olume ic gas- low me e
MGC-10 (Ri e ®) was used o measu e he biogas
p oduc ion.
The biogas composi ion was analyzed using a gas
ch oma og aph 5890 Se ies II (Hewle -Packa d) wi h a
he mal conduc i i y de ec o and a column o Po apack Q80/
100 2 m ×1/8″(SUPELCO). The column and de ec o
empe a u e we e se a 34 °C and 110 °C, espec i ely, using
helium as he ca ie gas. To al sul ide species (including
H
2
S, HS
−
,S
2−
) in he liquid phase we e measu ed using a
spec opho ome ic sul ide es me hod (Spec oquan ®)
acco ding o APHA 4500. An 861 Ad anced Compac Ion
Ch oma og aph (IC) wi h a Me osep C3-250 column, and an
838 Ad anced Sample P ocesso , we e used o measu e
sul a e ions. In e media y sul u compounds (S
0
,S
2
O
32−
, and
SO
32−
) we e no analyzed.
To p epa e o OMPs analysis, he samples we e
p e il e ed (using AP3004705, Millipo e), p econcen a ed by
solid-phase ex ac ion, and hen p ese ed a 4 °C. The
an ibio ics, neu o d ugs, and ho mones concen a ions we e
de e mined using liquid ch oma og aphy coupled o mass
spec ome y (LC-MS-MS), whe eas gas ch oma og aphy
coupled o mass spec ome y (GC-MS) was u ilized o
measu e he concen a ions o an i-in lamma o ies, TCS, and
BPA. OMPs samplings we e ca ied ou du ing Pe iod I (days
57, 64, and 70) and Pe iod II (141, 145, 147, and 149).
Al a ino e al.
23
p o ided a de ailed desc ip ion o he OMPs
analysis. The mic obial communi ies we e cha ac e ized
h ough 16S RNA gene amplicon sequencing on days 70
(Pe iod I) and 154 (Pe iod II). The DNA ex ac ion and
bioin o ma ic analysis we e conduc ed as indica ed by A ias
e al.
24
3. Resul s and discussion
The eac o was ope a ed o 154 d wi h empe a u es
anging be ween 18 and 22 °C, and wi h a consis en eed
low o 138 ± 15 L d
−1
. The hyd aulic e en ion imes applied
in he UASB and MBR pos - ea men sys em we e 21 ± 2 h
and 12 ± 1 h, espec i ely. In e nal and ex e nal ecycling
a ios (R) o 3.5 and 1, espec i ely, we e applied du ing he
s udy. The pH in he UASB and he anoxic compa men
emained cons an a 7.5 ± 0.5 and 7.6 ± 0.3, espec i ely. The
pH o he ae obic compa men du ing Pe iod II, 8.0 ± 0.1,
was highe han ha obse ed in Pe iod I, 7.7 ± 0.1, p obably
due o a highe deni i ying ac i i y. The MLVSS
concen a ion measu ed in he anoxic compa men we e 5.8
± 0.4 and 4.8 ± 0.3 g MLVSS L
−1
o Pe iods I and II,
espec i ely.
En i onmen al Science: Wa e Resea ch & TechnologyPape
Open Access A icle. Published on 30 Augus 2023. Downloaded on 11/8/2023 9:09:27 AM.
This a icle is licensed unde a
C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence.
View A icle Online
En i on. Sci.: Wa e Res. Technol., 2023, 9, 2733–2744 | 2737This jou nal is © The Royal Socie y o Chemis y 2023
3.1 COD emo al
The a e age COD
T
and COD
S
o he in low we e 1046 ±
176 mg L
−1
and 906 ± 180 mg L
−1
, espec i ely (Fig. 2).
COD emo al e iciencies abo e 98.2% we e obse ed in
he UASB-MBR sys em h oughou he en i e s udy, wi h
mos o he emo al happening in he UASB, 90 ± 4%
(Fig. 2). Analysis o he COD balances showed ha 73%
o he o ganic ma e ed o he UASB was con e ed
in o CH
4
. The biogas p oduced had an a e age
composi ion o 76 ± 4% CH
4
, 15 ± 3% CO
2
, and 8 ±
3% N
2
. Du ing Pe iod II, a ace amoun o H
2
S (0.5%)
was de ec ed in he biogas, which was p oduced by
SRB.
Fig. 2 Va ia ion in he o al chemical oxygen demand (COD
T
) concen a ion in a ious s eams o he UASB-MBR sys em ( op); COD emo al
e iciencies obse ed in he UASB and he UASB-MBR (bo om). The wo di e en ope a ing pe iods, one wi hou sul a e (Pe iod I) added o he
eed and one wi h (Pe iod II), a e sepa a ed by he e ical line.
Fig. 3 Change in dissol ed me hane concen a ion in a ious s eams o he UASB-MBR sys em ( op); me hane emo al a es (MRR)s and me hane
emo al e iciencies pe cen ages (%) obse ed in he anoxic compa men (bo om).
En i onmen al Science: Wa e Resea ch & Technology Pape
Open Access A icle. Published on 30 Augus 2023. Downloaded on 11/8/2023 9:09:27 AM.
This a icle is licensed unde a
C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence.
View A icle Online

2738 |En i on. Sci.: Wa e Res. Technol.,2023,9,2733–2744 This jou nal is © The Royal Socie y o Chemis y 2023
The COD
T
and COD
S
concen a ions in he UASB e luen
we e 103 ± 35 and 47 ± 23 mg L
−1
du ing he en i e s udy,
espec i ely. These alues exclusi ely accoun o he COD
con ibu ion de i ed om o ganic ma e p esen in he
UASB e luen , lea ing ou he COD con ibu ions o
dissol ed me hane and sul ide in his measu emen . The
ac ion o o ganic ma e no emo ed in he UASB eac o
can be used as an elec on dono o deni i y in he p e-
anoxic compa men o he MBR. The MBR pe mea e
con ained only 11.9 ± 9.3 mg COD L
−1
(Pe iod I) and 7.5 ± 3.7
mg COD L
−1
(Pe iod II), wi h concen a ions ne e exceeding
23 mg COD L
−1
du ing he en i e ope a ion. I should be
conside ed ha he COD
T
and COD
S
alues in he pe mea e
a e he same as no suspended solids can go h ough he
ul a il a ion memb ane. The e was no no iceable di e ence
in he pe mea e COD le els be ween Pe iods I and II,
indica ing ha he p esence o sul ide did no ha e an
impac on he COD emo al po en ial o he MBR.
3.2 Dissol ed me hane emo al
The dissol ed me hane concen a ion in he e luen o he
UASB sys em was 24.4 ± 1.7 mg L
−1
(Pe iod I) and 28.4 ± 7.4
mg L
−1
(Pe iod II), making up 11.8 ± 1.1% and 13.9 ± 3.0% o
o al me hane p oduced, espec i ely (Fig. 3). Du ing Pe iod I
(days 42–70), an a e age dissol ed me hane emo al
e iciency o 74.8 ± 7.2% was achie ed in he anoxic
compa men . Howe e , wi h he addi ion o sul a e in o he
eeding subs a e, he emo al e iciency conside ably
dec eased in he ea ly s ages o Pe iod II (days 70–97) o
alues anging be ween 17% and 46%. The ac i i y o
me hane oxidize s was ound o be ad e sely a ec ed by he
ini ial p esence o sul ide dissol ed in he anoxic
compa men . O e ime, as he sul ide was comple ely
oxidized by sul u -oxidizing bac e ia, he me hane emo al
e iciencies eco e ed, eaching high and s able alues om
day 103 un il he end o he s udy, 70.9 ± 5.1% (simila o
Pe iod I). The absence o dissol ed sul ide in he anoxic
compa men o he MBR om day 112 onwa ds was
con i med, bu ea lie dissol ed sul ide concen a ions we e
no measu ed.
Simila me hane emo al a es (MRR)s we e obse ed in
he anoxic compa men o bo h pe iods (Fig. 3), 94.0 ± 7.3
(Pe iod I) and 93.3 ± 23.4 mg CH
4
L
−1
d
−1
(Pe iod II). The
dissol ed me hane ha was no emo ed in he anoxic
compa men could ha e ei he been eleased in o he
a mosphe e h ough ae a ion o oxidized by ae obic
me hano ophs in he subsequen ae obic and/o il a ion
compa men s. Du ing mos o he ope a ion, he
concen a ion o dissol ed me hane in he ae obic
compa men was ei he no de ec ed o insigni ican (Fig. 3).
These esul s a e simila o he ones achie ed in p e ious
s udies ha used he old design o he UASB-MBR sys em
(Fig. 1) and simila expe imen al condi ions.
8,10
Using he
old design, Sil a-Te ia e al.
8
accomplished a maximum MRR
o 195 ± 17 mg CH
4
L
−1
d
−1
, doubling he alues obse ed in
he p esen s udy. The same expe imen al condi ions we e
used, bu highe eed lows we e applied (355 L d
−1
). They
ound ha he MRR was dependen on he eed low a e,
showing he bes esul s wi h he highes eed lows.
Howe e , he me hane emo al e iciency was no in luenced
by he eed low a e, bu by he eci cula ion a io be ween
he ae obic and anoxic compa men s. The lowe eed low
a e used in he p esen s udy could explain he lowe MRRs
ob ained.
Du ing his s udy, i was possible ha some o he
dissol ed me hane was emo ed h ough AMO-D (eqn (1) and
(2)) as dissol ed oxygen was ex e nally eci cula ed om he
memb ane il a ion o he anoxic compa men . Assuming
ha all he eci cula ed oxygen was solely used in AMO-D,
and conside ing eqn (1), me hane emo als up o 19.5 and 27
mg CH
4
L
−1
d
−1
could be expec ed. This accoun s o 21%
and 29% o he o al me hane emo al achie ed du ing
Pe iods I and II, espec i ely. The emaining emo al o
me hane may be a ibu ed o o he me hane oxida ion
p ocesses, such as N-damo o , as sugges ed by Kalyuzhnaya
e al.,
25
o ae obic me hane oxidize s ha con e me hane
in o e men a ion p oduc s unde oxygen-limi ed condi ions.
3.3 To al ni ogen emo al
One o he challenges in ea ing domes ic sewage in
me hanogenic bio eac o s is he high le els o o al ni ogen
(TN) p esen in hei e luen s, mos ly as ammonium. The
limi ed a ailabili y o easily biodeg adable o ganic ma e
exace ba es his issue, making i impe a i e o ind
al e na i e elec on accep o s o imp o e deni i ica ion. In
his sense, he p esence o ei he dissol ed me hane o
sulphu compounds in he was ewa e may be an al e na i e
o ob ain e luen s wi h lowe TN. Howe e , he hyd ogen
sul ide p oduced in he UASB s age may also inhibi he
ac i i y o mic obes in ol ed in elimina ing bo h ni ogen
and dissol ed me hane.
An a e age ammonium concen a ion o 71 ± 19 (Pe iod I)
and 84 ± 10 (Pe iod II) mg NH
4+
–NL
−1
was measu ed in he
UASB (Fig. 4). Al hough low concen a ions o ammonium
we e obse ed in he pe mea e by he end o Pe iod I,
comple e ae obic ammonium oxida ion was only achie ed by
he end o Pe iod II. High ammonium concen a ions we e
de ec ed in he pe mea e a he beginning o Pe iod II (days
85–125), up o 30 mg NH
4+
–NL
−1
. This accumula ion could
be a ibu ed o he p esence o dissol ed sul ide in he UASB
e luen , which a e known o inhibi ni i ie s a low
concen a ions.
18
Ni a e was he majo ammonium oxida ion p oduc ,
and signi ican concen a ion a ia ions we e obse ed in
bo h pe iods (Fig. 4). Du ing Pe iod I, highe
concen a ions o ni a e we e ound in bo h he anoxic
compa men and pe mea e, wi h concen a ions o 23.1 ±
6.7 and 33.2 ± 7.6 mg NO
3
−
–NL
−1
, espec i ely. Howe e , a
he beginning o Pe iod II, he ni a e concen a ion
dec eased signi ican ly, especially in he anoxic
En i onmen al Science: Wa e Resea ch & TechnologyPape
Open Access A icle. Published on 30 Augus 2023. Downloaded on 11/8/2023 9:09:27 AM.
This a icle is licensed unde a
C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence.
View A icle Online
En i on. Sci.: Wa e Res. Technol., 2023, 9, 2733–2744 | 2739This jou nal is © The Royal Socie y o Chemis y 2023
compa men , wi h concen a ions o only 1.9 ± 1.7 mg
NO
3
−
–NL
−1
. This ema kable lowe ni a e concen a ion
could be caused by he highe ni a e deni i ica ion a e o
SOB
26
due o he p esence o sul ide in he e luen o he
anae obic compa men . The obse ed ni a e in he
pe mea e was usually below 15 mg NO
3
−
–NL
−1
, which was
lowe han he concen a ion obse ed in Pe iod I. The TN
emo als in he MBR du ing Pe iod I (days 42–70) and
Pe iod II (days 132–154) we e 40.2 ± 5.0 and 63.4 ± 2.2 mg
TN L
Feed
−1
, espec i ely (Fig. 4).
The inc eased TN emo als in Pe iod II can be a ibu ed
o he p esence o dissol ed sul ide in he UASB e luen (14.2
± 4.0 mg S L
−1
), which could be used by SOB o deni i y. This
implies ha , apa om esidual o ganic ma e and
dissol ed me hane, dissol ed sul ide se ed as an elec on
dono o deni i ica ion, as demons a ed by he
conside able educ ion in ni a e concen a ion compa ed o
Pe iod I. The obse a ion sugges s ha he p esence o
dissol ed sul ide boos ed he deni i ica ion capaci y o he
sys em, a he han inhibi ing i .
I should be highligh ed ha ni ogen emo als we e no
conduc ed only in he anoxic compa men o he MBR, bu
also in he ae obic compa men . This is demons a ed by
he ac ha he a e age TN concen a ion in he pe mea e
was lowe han ha measu ed in he anoxic compa men
(Fig. 4). The di e ences in TN concen a ions be ween hese
wo compa men s we e consis en ly de ec ed. Th ough mass
balances, i was es ima ed ha 21 ± 14% (Pe iod I) and 18 ±
6% (Pe iod II) o he TN emo al in he MBR was conduc ed
in he ae obic compa men . This emo al was likely achie ed
in he bio ilm loca ed in he ae obic compa men , as
epo ed by o he au ho s using a simila sys em.
9
Mos o
he ni ogen was emo ed in he anoxic compa men , whe e
ni ogen emo al a es o 134.1 ± 28.8 (Pe iod I) and 188.5 ±
10.2 mg N L
−1
d
−1
(Pe iod II) we e accomplished.
The TN concen a ion emo ed in his s udy, e en du ing
Pe iod I wi hou he SOB con ibu ion (40.2 ± 5.0 mg TN
L
Feed
−1
), was much highe han ha in p e ious s udies wi h
he old design o he UASB-MBR sys em, wi h 15 and 20 mg
TN L
−1
.
8,21
The enhanced deni i ica ion po en ial obse ed
wi h he new design was a ibu ed o he p esence o bio ilm
ca ie s in he ae obic compa men , whe e di e en
deni i ying mic oo ganisms can h i e.
To es ima e he ni ogen ha could be deni i ied using
he o ganic ma e , dissol ed me hane, and dissol ed sul ide
measu ed in he e luen o he UASB, mass balances we e
conduc ed. Con en ional he e o ophic deni i ica ion
equi es app oxima ely 4.98 o 2.98 g COD g
−1
N, using
ni a e o ni i e as an elec on accep o , espec i ely. By
using s oichiome ic analyses and conside ing ha all he
o ganic ma e p esen in he UASB e luen was used o
deni i y, up o 15.3 (Pe iod I) and 14.1 (Pe iod II) mg NO
3
−
–
NL
eed
−1
, could be emo ed in he MBR sys em. This sugges s
ha he o ganic ma e con en in he UASB e luen was no
high enough o explain he obse ed o al ni ogen emo als.
The me hane dissol ed in he UASB e luen could also ha e
been used o deni i ica ion in he MBR. In AMO-D
p ocesses using me hanol as an in e media y elec on dono ,
0.95 g CH
4
g
−1
NO
3
−
–N is heo e ically equi ed. By using
s oichiome ic eac ions and i all he oxygen consumed in
he anoxic compa men was used only o AMO-D, 3.5
(Pe iod I) and 5.8 mg NO
3
−
–NL
−1
(Pe iod II) could ha e been
emo ed.
Mo eo e , he dissol ed sul ide p esen in he UASB
e luen (14.2 mg L
−1
) could ha e also played a signi ican
ole in he deni i ica ion ac i i y (eqn (3) and (4)). SOB was
ound o ha e ully oxidized he sul ide be ween days 112 and
154, as i was no de ec ed in he anoxic and ae obic
compa men s. Based on eqn (4), i was es ima ed ha 9.9
mg NO
3
−
–NL
−1
could ha e been deni i ied in Pe iod 2 by
SOB. Conside ing ni i e as he elec on dono (eqn (3)), his
alue could inc ease up o 16.5 mg NO
2
−
–NL
−1
. The
combined impac o con en ional he e o ophic
deni i ica ion, AMO-D, and sul ide oxida ion p ocesses could
Fig. 4 Changes in ni ogen species (NH
4+
,NO
2
−
, and NO
3
−
) and o al
ni ogen (TN) concen a ions in di e en compa men s o he UASB-
MBR sys em h oughou he en i e expe imen a ion pe iod. To al
ni ogen (TN) emo al and TN emo al e iciency (%) obse ed in he
MBR pos - ea men sys em du ing he expe imen a ion pe iod.
En i onmen al Science: Wa e Resea ch & Technology Pape
Open Access A icle. Published on 30 Augus 2023. Downloaded on 11/8/2023 9:09:27 AM.
This a icle is licensed unde a
C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence.
View A icle Online
2740 |En i on. Sci.: Wa e Res. Technol.,2023,9,2733–2744 This jou nal is © The Royal Socie y o Chemis y 2023
esul in a o al es ima ed deni i ica ion o 18.8 and 29.8 mg
TN L
−1
o Pe iods I and II, espec i ely. Howe e , his
capaci y was s ill lowe han he ac ual TN emo al obse ed
in Pe iod I (40.2 mg TN L
−1
) and Pe iod II (63.4 mg TN L
−1
).
Using he old UASB-MBR sys em p o o ype, Sil a-Tei a
e al.
8
achie ed TN emo als (10–15 mg N L
Feed
−1
)
signi ican ly lowe compa ed o his s udy. Using he same
echnology as Sil a-Tei a e al.,
8
bu wi h p esence o ni i e
and ni a e in he anoxic compa men , Al a ino e al.
10
achie ed highe ni ogen emo als o 35 mg TN L
Feed
−1
, bu
s ill lowe han wi h he new UASB-MBR p o o ype. This
highligh s he impo ance o he ex a ae obic compa men
wi h suspended bio ilm ca ie s added o he MBR.
In ano he s udy, Pelaz e al.
7
p oposed he use o a ixed
ilm bio eac o illed wi h co uga ed PVC ings o ea he
e luen om an anae obic memb ane bio eac o con aining
me hane, sul ide, low concen a ions o o ganic ma e , and
ni ogen. Tha esea ch p o ed he easibili y o using all
hese compounds as elec on dono s o emo e a ound 73
mg TN L
−1
a 18 °C and 2 h o HRT, a alue which is highe
han in he p esen s udy. Howe e , he quali y o hei
e luen , in e ms o COD, was lowe (58.6 mg COD L
−1
),
which migh be explained by he absence o an ul a il a ion
memb ane in he pos - ea men sys em.
3.4 O ganic mic opollu an s emo al
The OMPs emo al e iciencies we e calcula ed by using mass
balances in bo h, he UASB and he MBR o he in eg a ed
sys em. O e all, high OMPs emo al e iciencies we e
obse ed in he UASB-MBR sys em du ing he en i e
expe imen a ion (Fig. 5), wi h no majo di e ences be ween
Pe iods I and II. This sugges s ha sul a e addi ion did no
a ec he OMPs emo al e iciencies. NPX, SMX, TMP, and
FLX we e mainly emo ed in he UASB sys em wi h anae obic
condi ions, while mos o he BPA, IBP, and E1 emo als we e
obse ed in he MBR pos - ea men sys em. The emo al o
ERY, ROX, E2, EE2, and TCS was obse ed in he UASB and
he MBR, indica ing he key ole o he MBR o ob ain highe
bio ans o ma ion e iciencies o mos o he OMPs and
pa icula ly o hose which we e no a ec ed by he UASB.
CBZ, DZP, and DCF we e ba ely emo ed du ing his s udy,
indica ing he ecalci an na u e o hese OMPs agains
biological deg ada ion. Simila OMPs emo al e iciencies
we e obse ed in ou p e ious s udy using he old UASB-MBR
p o o ype (Fig. 1),
10
and o he s udies using ano he ype o
in eg a ed echnologies combining di e en edox
condi ions.
27
These esul s p o e he obus ness o his
sys em o emo e mos o he OMPs.
SMX, TMP, and NPX a e easily biodeg adable in anae obic
condi ions.
27
This explains why SMX and TMP an ibio ics
we e en i ely emo ed and why NPX, an an i-in lamma o y
d ug, expe ienced signi ican bio ans o ma ion (>90%)
du ing Pe iods I and II. FLX, a lipophilic compound, was also
highly emo ed in he UASB eac o wi h anae obic
condi ions, wi h e iciencies abo e 83% du ing bo h pe iods.
The high emo als obse ed can be explained by he as
so p ion capaci y o his neu o d ug on o he g anula
sludge.
28
Simila emo als could be expec ed o ano he
lipophilic mic opollu an such as TCS, howe e , despi e he
high e iciencies a ained du ing Pe iod I (95.6%), an a e age
alue o only 30% was achie ed o his disin ec an du ing
Pe iod II. The eason o his lowe deg ada ion obse ed in
he las pe iod was no elucida ed. The emaining ac ion o
FLX and TCS p esen in he UASB e luen was almos o ully
deg aded in he subsequen MBR pos - ea men sys em.
IBP and BPA we e mainly emo ed in he MBR sys em
wi h al e na ing anoxic and ae obic condi ions. Mos o he
IBP emo al was conduc ed mos likely in he ae obic
compa men since his an i-in lamma o y is known o be
easily biodeg adable unde ae obic condi ions as epo ed by
Sua ez e al.
29
The highe emo als obse ed du ing Pe iod II
could be explained by he highe ni i ica ion ac i i ies
obse ed in he ae obic compa men o he MBR, compa ed
Fig. 5 O ganic mic opollu an s (OMPs) emo al e iciencies obse ed in he up- low anae obic sludge blanke (UASB) and he memb ane
bio eac o (MBR) du ing he di e en ope a ing pe iods: Pe iod I, wi hou sul a e addi ion in o he eeding subs a e (le ) and Pe iod II, wi h sul a e
addi ion ( igh ).
En i onmen al Science: Wa e Resea ch & TechnologyPape
Open Access A icle. Published on 30 Augus 2023. Downloaded on 11/8/2023 9:09:27 AM.
This a icle is licensed unde a
C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence.
View A icle Online
En i on. Sci.: Wa e Res. Technol., 2023, 9, 2733–2744 | 2741This jou nal is © The Royal Socie y o Chemis y 2023
o Pe iod I. The BPA biodeg ada ion could ha e also occu ed
unde ae obic condi ions h ough ni i ica ion p ocesses as
epo ed by o he au ho s.
30,31
E1 and E2 we e ully emo ed
du ing bo h pe iods. E1 was mos ly emo ed in he MBR,
while E2 was emo ed join ly be ween he UASB and he
MBR. The emo al o BPA, a well-known endoc ine dis up o ,
ook place in he MBR, showing i s ecalci an na u e unde
anae obic condi ions.
3.5 Mic obial communi ies in he MBR
The s udy ocused on bo h he suspended biomass in he
MBR and he bio ilm communi ies ha adhe ed o he
biochip ca ie s in he anoxic and ae obic compa men s by
he end o Pe iod I (day 70) and Pe iod II (day 154). The
dominan bac e ia ound in he MBR sys em du ing bo h
pe iods we e membe s o he Chi inophagaceae and
Sap ospi aceae amilies, as well as he Chlo obi phylum
(Fig. 6; Table S1†).
Sap ospi aceae a e ae obic he e o ophs conside ed
sap ophy ic mic oo ganisms. The ole o Chi inophagaceae
and Chlo obi in he MBR emains unclea . The p esence o
he amily Comamonadaceae (including he e o ophic
deni i ie s), was con i med in all he samples. The p esence
o he Comamonadaceae amily (which includes he e o ophic
deni i ie s) was con i med in all samples and i is sugges ed
ha hey u ilized a ious o ganic compounds as elec on
dono s o deni i ica ion in he MBR.
Ae obic me hano ophs, mainly o he Me hyloccoccaceae
and Me hylophilaceae amilies, we e also ound in all
samples, wi h abundances anging om 1.8 o 8.1%. The
ema kable abundance o hese bac e ia in he adhe ed
biomass o he ae obic compa men du ing bo h pe iods
was de ec ed, wi h abundances o 3.6% and 8.1% in Pe iods I
and II, espec i ely. This ac con i ms ha dissol ed
me hane was elimina ed in bo h compa men s o he MBR.
The p esence o N-damo bac e ia which was de ec ed in an
analog sys em by FISH analysis
8
was no de ec ed in he
p esen s udy o any o he samples.
The p esence o anammox bac e ia (B ocadiaceae), which
can oxidize ammonium by educing ni i e in anoxic
condi ions,
32
was low in Pe iod I (<0.4%) bu highe alues
we e obse ed in he adhe ed biomass in he anoxic (3.7%)
and ae obic compa men s (4.6%) du ing Pe iod II. SOB,
belonging o he Thiobac e ales o de , we e abundan in all
samples analyzed in Pe iod II, wi h abundances anging om
3.6 o 8.5%. This indica es ha sul ide oxida ion and
deni i ica ion p ocesses we e conduc ed in bo h
compa men s o he MBR. The SOB p oli e a ion du ing
Pe iod II is explained by he p esence o sul ide in he UASB
e luen .
Anae obic en i onmen s can be gene a ed in he inne
laye s due o he exis ence o dis inc dissol ed oxygen
concen a ion g adien s wi hin he bio ilm ca ie s. This is a
no ewo hy ea u e o bio ilm sys ems ha could explain he
obse ed ni ogen emo al in he ae obic compa men . The
co-occu ence o ni i ica ion–deni i ica ion p ocesses in he
Biochip unde sco es he impo ance o he supplemen a y
ae obic compa men , which was added o imp o e me hane
and ni ogen emo als compa ed o ea lie s udies.
8,10
Fig. 6 Mic obial di e si y p o iles o communi ies p esen in he MBR in wo di e en ope a ing pe iods: wi h (Pe iod I) and wi hou (Pe iod II)
dissol ed sul ide. The communi y composi ion exp essed as he ela i e abundance o he mos abundan mic obial amilies is shown o each
sample. The analysis was pe o med on he suspended biomass as well as on he bio ilm ca ie s (Biochip) loca ed in he anoxic and ae obic
compa men s.
En i onmen al Science: Wa e Resea ch & Technology Pape
Open Access A icle. Published on 30 Augus 2023. Downloaded on 11/8/2023 9:09:27 AM.
This a icle is licensed unde a
C ea i e Commons A ibu ion-NonComme cial 3.0 Unpo ed Licence.
View A icle Online