Bioelectrocatalyzed reduction of acetic and butyric acids via direct electron transfer using a mixed...

This journal is c The Royal Society of Chemistry 2013 Chem. Commun., 2013, 49, 6495--6497 6495

Cite this: Chem. Commun.,2013,

49, 6495

Bioelectrocatalyzed reduction of acetic and butyric

acids via direct electron transfer using a mixed culture

of sulfate-reducers drives electrosynthesis of alcohols

and acetone†

Mohita Sharma,ab Nabin Aryal,a Priyangshu M. Sarma,bc Karolien Vanbroekhoven,a

Banwari Lal,c Xochitl Dominguez Benetton*a and Deepak Pant*a

Sulfate-reducing bacteria (SRB) developed biocathodes efficient for

reduction of acetic and butyric acids to alcohols and acetone via

direct electron transfer reaching current densities of 160–210 Amÿ2.

Microbial Electro Synthesis (MES) demonstrates the ability of

microbes to use electrodes as electron donors or acceptors

for organic synthesis reactions in electrochemical cells.1,2

Electrochemically-active (EA) microbes on the cathodic surface

can be tweaked to provide an effective alternative to maximize

single/multiple product generation and recovery from waste

streams. Besides, products are released near the electrodes,

which makes feasible their intensified and in situ recovery.

Reduction of CO2 or substrate organics are two means for

bioelectrochemical production at cathodes. Both have been

prospected for the synthesis of valuable organic chemicals.3–9

Biofilms of Sporomusa ovata grown on graphite have been

shown to consume electrons while achieving reduction of

CO2 to acetate, formate and 2-oxobutyrate.5 Later, S. ovata,

Clostridium ljungdahlii, C. aceticum and Moorella thermoacetica

consumed current with concomitant production of organic

acids.7 S. ovata has achieved coulombic efficiencies of 90%

for CO2 to acetate cathodic reduction.10 Recently, CO2 reduction to

glycerol was achieved with Geobacter sulfurreducens.11 On the other

hand, enzyme-mediated systems were earlier developed for

reduction of organics.12,13 Using artificial redox mediators,

reduction of fumarate to succinate was achieved with Actinobacillus

succinogenes and Shewanella oneidensis.14,15 Meanwhile, other

investigations considered electrochemically-generated H2 as a

reducing equivalent to drive fermentation; for instance, reduction

of glucose to boost the production of propionate and butanol.8,9,16,17

Increase of hydrogen partial pressure (HPP) resulted in meta-

bolic shift from acidogenesis to alcohol production.18 Decrease

in biogenic H2 was observed. Steinbusch et al. achieved bioelec-

trochemical ethanol production through mediated acetate

reduction, using mixed cultures.19 Still, immediate technological

challenges for this pioneering alternative for organic synthesis

include development of cheap and robust bioelectrocatalysts,

capable of highly selective product formation, performing at

high current densities, and with long-term stability.7 Moreover,

systems with direct electron transfer (DET) would be ideal, as

excess biomass and waste decrease.1,20

These issues were addressed in the present study by develop-

ment of effective biocathodes on economical carbon-based elec-

trode materials.21 The present research also signifies advancement

beyond the existing state of the art as for the first time a mixed

culture of sulfate reducers converted volatile fatty acids (VFAs) into

alcohols and acetone, via the DET bioelectrochemical route.

Electrochemical H2 formation took place, but this was utilized only

in negligible quantities, thus partaking only a secondary role in

product formation. Low full-mineralization to CO2 was detected.

Nomethane was observed. High cathodic current densities of up to

160–210 A mÿ2 were successfully achieved.

For practicality, experimental steps followed (details in S1,

ESI†) are referred to as Exp., and differentiated by a number.

Exp. 1: chrono-amperometric (CA) experiments consisted of

polarization of half-cell cathodes at a potential of ÿ0.85 V vs.

Ag/AgCl. After inoculating the medium containing VFAs (acetic

and butyric acids) as substrates with SRB, reduction current

progressively increased (Fig. 1a). This was replicated in 14 inde-

pendent reactors, showing comparable results. Two additional

systems were reproduced independently at the laboratory of

co-authors, with the same inoculum source.

The inoculum source, characterized by Fluorescence In Situ

Hybridization (FISH), consisted of groups clearly predominated by

SRB, as they were enriched with a standard medium (API Recom-

mended Practice RP-38): Curvibacter (Fig. 2e–h), Deltaproteo-

bacteria (Fig. 2i–l), Desulfovibrionales, Desulfobacteraceae and

a Separation and Conversion Technologies, VITO - Flemish Institute for

Technological Research, Boeretang 200, 2400 Mol, Belgium. E-mail: [email protected],

[email protected] TERI University, Plot No. 10, Institutional Area, Vasant Kunj, New Delhi, 110070,

Indiac The Energy and Resource Institute (TERI), IHC, Lodhi Road, New Delhi, 110003,

India

† Electronic supplementary information (ESI) available. See DOI: 10.1039/

c3cc42570c

Received 9th April 2013,

Accepted 31st May 2013

DOI: 10.1039/c3cc42570c

www.rsc.org/chemcomm

ChemComm

COMMUNICATION

6496 Chem. Commun., 2013, 49, 6495--6497 This journal is c The Royal Society of Chemistry 2013

Syntrophobacteraceae (Fig. 2m–p). Traces of Archaea (a–d) were

also found, but no methanogenic activity was registered in any

of the experiments.

SRB can participate in electrochemical reactions through diverse

mechanisms.22 For instance, SRB hydrogenases are capable of DET

with a wide variety of materials.23 More recently, in the context of

anaerobic biocorrosion, it was demonstrated that SRB whole cells

are capable of DET from a cathodic site, while the bacterial uptake

of H2 does not influence corrosion potential or current density at

a given potential.24 Thus, consumption of electrogenerated H2

should not affect the quantity of charge consumed by the electro-

chemical reaction.25 In this work, the quantity of charge was always

affected by every consecutive addition and reduction of the sub-

strate (Fig. 1a and b). CVs were regularly recorded at 1 mV sÿ1, at

starting and subsequent times (Fig. 1c and d), corresponding to

specific days as indicated in Fig. 1a and b, respectively. CVs indicate

evolution towards a Nernstian-like behavior as the system becomes

mature.26 In order to verify that the current measured was due to

bioelectrocatalytic reduction of the VFAs and not to the classical

electrochemical proton reduction to H2, several experiments were

carried out. Exp. 2 to 8 are described in S4 (ESI†).

Exp. 1 resulted in the production of an assortment of chemicals,

derived from the bioelectrochemical reduction of acetate and

butyrate. Product formation and substrate consumption for the

reactor corresponding to Fig. 1a are presented in Fig. 3.

Main products detected were ethanol, butanol, methanol, pro-

pionic acid, caproic acid, acetone and propanol from the VFAs used

as substrates (Fig. 3a and b, respectively). Though ethanol was

produced early (up to day 19), it was not observed later (Fig. 3a).

SRB can use ethanol directly as the electron donor;27 which could

be the reason for its absence later. Ethanol accumulation could

make the VFA reduction reactions thermodynamically unfeasible or

inhibit microbial growth. Besides, ethanol oxidation has been

reported to favor the production of butyrate and caproate and the

conversion of propionate to propanol.4 In future studies, the role of

ethanol in overall product formation will be clarified.

The organic compounds produced were detected progressively

with the start of current utilization. Therefore, it is assumed that

all products were the consequence of bioelectrocatalyzed reduction

reactions (Fig. 4). A magnification of Fig. 4 as well as the detailed

Fig. 1 Current density evolution of two different half-cell SRB-enriched bio-

cathodes poised at ÿ0.85 V vs. Ag/AgCl, with consecutive addition of substrates.

(a) Large electrode (27 cm2) (b) small 3 cm2 piece of electrode (a) cut at day 60 of

operation and placed in an independent reactor, maintaining a constant area to

volume ratio. Red arrows indicate cyclic voltammetries (CV). (c) and (d) CV taken

for different times at (a) and (b), respectively, at 1 mV sÿ1. Day 0 in (b)

corresponds to day 60 of (a).

Fig. 2 FISH micrographs of the inoculum source. Probe combinations: (a–d)

ARC915/red/cy3 (Archea) and EUB 338mix/green/fluos (bacteria domain); (e–h)

BET42a/red/cy3 (Betaproteobacteria) and AQS997/green/6-fam (Curvibacter);

(i–l) EUB338I red/cy3 (bacteria domain) and DELTA 495a/green/6-fam (Delta-

proteobacteria); (m–p) SRB385/red/cy3 (Desulfovibrionales) and DSBAC357/

green/6-fam (Desulfobacteraceae and Syntrophobacteraceae). Columns 1 and

2 indicate all green and red probes individually over DAPI stained DNA in the

background, respectively. Columns 3 and 4 show superimposed probes, in the

absence and presence of DAPI staining for DNA in the background, respectively.

An acquisition system (Coolsnap, Roper Scientific Photometrics) connected to an

Axioskop2 epifluorescence microscope (Zeiss, Germany) was used to record

fluorescence signals.

Fig. 3 Concentrations measured for the reactor containing the SRB biocathode

presented in Fig. 1a for: (a) product formation and (b) availability of acetic acid

and butyric acid as substrates. Although not visible in Fig 3b, caproic acid was

produced in very less quantities (B0.02 g Lÿ1).

Fig. 4 SRB bioelectrocatalyzed reduction of acetic and butyric acids into

alcohols, acetone and caproate, controlled via direct electron transfer.

Communication ChemComm

This journal is c The Royal Society of Chemistry 2013 Chem. Commun., 2013, 49, 6495--6497 6497

mechanistic description and electrochemical characterization by

CV and electrochemical impedance spectroscopy (EIS) are pro-

vided in S5 (ESI†). The EIS and CV responses revealed character-

istics of an electrochemical–(bio)chemical process limited by the

relaxation of the electrode coverage by one or several adsorbed

intermediates, which were absent in Exp. 4 to 8. The voltampero-

metric response suggested the presence of membrane-bound

redox enzymes that could be a limiting step in the bioelectrocata-

lytic reduction process,28 reflected as stimulation of cathodic

current.24,29 Furthermore, if direct electron transfer would proceed

through these proteins, this would most likely occur via super-

exchange.28,30 This occurrence would be supported by the features

observed through cyclic voltammetry (ESI,† S5).

For instance, the charge of the transient voltammetric peaks

(area under the peak) scales in magnitude with the square root of

the voltammetric scan rate (at a sufficiently high scan rate of

>20 mV sÿ1), as presented in Fig. S5.4 (ESI†). These peaks result

from the change in the oxidation state of the redox cofactors,

presumptively involved in the adsorption-mediated redox process,

within the SRB electrochemically-active biofilm. Coexistence of

diffusional limitations is also confirmed by the decreasing peak

current as a function of the cycle number at a scan rate of 50mV sÿ1,

whereas the peak current for peak 2 shows a slight increase at

the lower scan rates due to adsorption predominance. These

observations are consistent with superexchange features28 and

are attributed to the membrane-bound redox enzymes.

The validity of these postulates is dependent on whether or not

the referred enzymes are found to be present with high activities

within the SRB mixed culture, as well as on their association as

physiological intermediates in the bioelectrocatalytic process. Still, the

adsorption behavior related to the redox couples found by CV would

play a limiting role in the whole bioelectrocatalytic mechanism.

These results together with experimental results provided in

the ESI† are consistent with previous studies,24,25 suggesting that

the reduction processes for bioelectrosynthesis of alcohols and

acetone were presumably initiated by DET and not by the use of

H2 as the energy carrier. Based on the obtained results, it is

inferred that DET could happen via conductive redox enzymes in

the SRB cells. Deprotonation of weak acids also would be

implicated in the initial stages of this bioelectrocatalytic pro-

cess.25 Future work will focus on elucidating in detail DET and

on the validation of an overall model reaction mechanism.

As described elaborately by Rabaey et al.3 it becomes highly

important for microbial electrochemical technologies to ensure that

the right amount of substrate is provided to the microbes and

conditions are modified in such a way that it promotes specificity of

the production pathway relative to existing conversions and minimizes

cell growth. These conditions were essentially taken into account while

conducting the present experiments by providing acetic and butyric

acids as substrates that ensured no undesirable side reactions (full

mineralization to CO2, methanogenesis). The ultimate purpose of this

study is to valorize highly complex and variable waste streams contain-

ing high organic loads. Since the mixed culture is capable of tolerating

highmolar concentration of VFAs (as high as 0.1M each) and also high

salt concentrations (e.g. 10 g Lÿ1), this consortium is considered highly

robust and efficient for treating such types of streams without pre-

dilution, which would reduce the treatment costs.

In conclusion, DET microbially electrocatalyzed reduction of

acetate and butyrate to mainly alcohols and acetone was achieved

using a highly robust and halotolerant mixed SRB culture. It is

expected that these results will signify major progress in the

valorization of side streams to valuable chemicals.

M.S. acknowledges Indo-Belgian scholarship from the Flemish

Government (Vlaamse Gemeenschap), DST (India) and MICINN

(Spain). We thank C. Porto-Carrero for technical help, C. Gielen

for analytics, P. Jain, Dr P. Dureja and J. Varanasi for their

experimental contributions. We also acknowledge the Group of

Environmental Engineering and Bioprocesses-Department of

Chemical Engineering, Universidad de Santiago de Campostela,

Spain, for providing infrastructure for FISH.

Notes and references

1 D. R. Lovley, Environ. Microbiol. Rep., 2011, 3(1), 27–35.2 X. D. Benetton, S. Sevda, K. Vanbroekhoven and D. Pant, Chem. Soc.

Rev., 2012, 41, 7228–7246.3 K. Rabaey, P. Girguis and L. K. Nielsen, Curr. Opin. Biotechnol., 2011,

22, 1–7.4 M. T. Agler, B. A. Wrenn, S. H. Zinder and L. T. Angenent, Trends

Biotechnol., 2011, 29, 70–78.5 K. P. Nevin, T. L. Woodard, A. E. Franks, Z. M. Summers and

D. R. Lovley, mBio, 2010, 2, e00103–e00110.6 K. Rabaey and R. Rozendal, Nat. Rev. Microbiol., 2010, 8, 706–716.7 K. P. Nevin, S. A. Hensley, A. E. Franks, Z. M. Summers, J. Ou,

T. L. Woodward, O. L. Snoeyenbos-West and D. R. Lovley, Appl.Environ. Microbiol., 2011, 77, 2882–2886.

8 R. Emde and B. Schink, Appl. Environ. Microbiol., 1990, 56, 2771–2776.9 T. S. Kim and B. H. Kim, Biotechnol. Lett., 1988, 1, 123–128.10 Y. Gong, A. Ebrahim, A. M. Fiest, M. Embree, T. Zhang, D. Lovley

and K. Zengler, Environ. Sci. Technol., 2013, 47, 568–573.11 L. Soussan, J. Riess, B. Erable, M. Delia and A. Bergel, Electrochem.

Commun., 2013, 28, 27–30.12 M. Aizawa, R. W. Coughlin and M. Charles, Biotechnol. Bioeng., 1976,

8, 209–215.13 M. Hongo and M. Iwahara, Agric. Biol. Chem., 1979, 43, 2075–2081.14 D. H. Park, M. Laivenieks, M. V. Guettler, M. K. Jain and J. G. Zeikus,

Appl. Environ. Microbiol., 1999, 65, 2912–2917.15 D. E. Ross, J. M. Flynn, D. B. Baron, J. A. Gralnick and D. R. Bond,

PLoS One, 2011, 6(2), e16649, DOI: 10.1371/journal.pone.0016649.16 B. H. Kim, P. Bellows, R. Datta and J. G. Zeikus, Appl. Environ.

Microbiol., 1984, 48, 764–770.17 M. C. A. A. V. Eastern-Jansen, A. T. Heijne, T. I. M. Grootscholten,

K. J. J. Steinbusch, T. H. J. A. Sleutels, H. V. M. Hamlers and C. J. N.Buisman, ACS Sustainable Chem. Eng., 2013, 1, 513–518.

18 K. J. J. Steinbusch, H. V. M. Hamelers and C. J. N. Buisman, WaterRes., 2008, 42, 4059–4066.

19 K. J. J. Steinbusch, H. V. M. Hamelers, J. D. Schaap, C. Kampmanand C. J. N Buisman, Environ. Sci. Technol., 2010, 44, 513–517.

20 K. B. Gregory, D. R. Bond and D. R. Lovley, Environ. Microbiol., 2004,6, 596–604.

21 D. Pant, G. V. Boggaert, C. P. Carrero, L. Diels and K. Vanbroekhoven,Water Sci. Technol., 2011, 63, 2457–2461.

22 H. Castaneda and X. D. Benetton, Corros. Sci., 2008, 50, 1169–1183.23 S. Da Silva, R. Basseguy and A. Bergel, Corrosion, 2002, 02462.24 H. Venzlaff, D. Enning, J. Srinivasan, K. J. J. Mayrhofer, A. W. Hassel,

F. Widdel and M. Stratmann, Corros. Sci., 2013, 66, 88–96.25 S.DaSilva,R.BasseguyandA.Bergel,Electrochim.Acta, 2004,49, 4553–4561.26 R. Rousseau, X. D. Benetton, M. Delia and A. Bergel, Electrochem.

Commun., 2013, 33, 1–4.27 R. K. Thauer, K. Jungermann and K. Decker, Bacteriol. Rev., 1977, 41,

100–180.28 D. R. Bond, S. M. Strycharz-Glaven, L. M. Tender and C. Torres,

ChemSusChem, 2012, 5, 1099–1105.29 E. C. Hatchikian, V. M. Fernandez and R. Cammack, inMicrobiology

and Biochemistry of Strict Anaerobes involved in interspecies hydrogentransfer, ed. J. P. Belaich, M. Bruschi and J. L. Garcia, Springer, NewYork, USA, 1990, vol. 54, pp. 53–73.

30 N. S. Malvankar and D. R. Lovley, ChemSusChem, 2012, 5, 1039–1046.

ChemComm Communication

Bioelectrocatalyzed acetic and butyric acid reduction via direct 1

electron transfer by a mixed culture of sulfate-reducers drives 2

electrosynthesis of alcohols and acetone 3

Mohita Sharma,a,b

Nabin Aryal,a Priyangshu M. Sarma,

b,c Karolien Vanbroekhoven,

a Banwari 4

Lal,c Xochitl Dominguez Benetton*

,a and Deepak Pant*

,a 5

6 a Separation & Conversion Technologies, VITO - Flemish Institute for Technological Research, Boeretang 200, 2400 Mol, Belgium 7 b TERI University, Plot No. 10, Institutional Area, Vasant Kunj, New Delhi 110070, India 8 cThe Energy and Resource Institute (TERI), IHC, Lodhi Road, New Delhi, 110003, India 9

10

11

Electronic Supplementary Material (ESI) for Chemical CommunicationsThis journal is © The Royal Society of Chemistry 2013

S1: Experimental Details 1

1.1 Source of inoculum and media 2

3

The anaerobic mixed consortium used to inoculate the cathodic chamber was taken from a laboratory scale continuous reactor 4

developed by inoculating mixed SRB strains isolated from different petroleum refineries and formation sites across India. 5

The two-liter lab scale reactor was run continuously on a fed batch mode with API RP-38 medium (4 g sodium lactate, 1 g 6

yeast extract, 0.1 g ascorbic acid, 0.2 g MgSO4, 0.01 g K2HPO4, 0.1 g KH2PO4, 0.2 g NH4FeSO4 and 10 NaCl, for 1 L, pH-7

7.4) for the development of a stable SRB consortium. A 10% v/v was taken from the reactor and transferred to the minimal 8

sulfate/sulfide free synthetic feed that was used as electrolyte in the electrochemical cell. The synthetic feed composed of 572 9

mg NH4Cl, 416 mg K2HPO4, 8 mg CaCl2, 96 mg MgCl2·6H2O, 1.98 mg FeCl2·4H2O, 2.37 mg CoCl2·6H2O, 0.59 mg 10

MnCl2·4H2O, 0.034 CuCl2·2H2O, 0.062 mg H3BO3, 0.073 mg Na2MoO4·2H2O, 0.069 mg Na2SeO3, 0.095 mg NiCl2·6H2O, 11

0.055 mg ZnCl2 and 10 g NaCl per liter. Additionally the substrate concentration that was added to the reactors included 0.1 12

M each of acetic and butyric acid. 13

14

1.2 Configuration of electrochemical cell 15

16

The electrochemical cell was placed over a magnetic stirrer and sparged continuously with nitrogen gas to maintain anoxic 17

headspace and all the experiments were operated at 18-22 °C. The working volume of the reactor was 275 ml. A 80-20% 18

carbon-polytetrafluoroethylene (PTFE) mix with 316L grade stainless steel (SS) mesh embedded inside was used as a 19

working electrode. The projected surface area of the working electrode was 9 × 3 cm2 and it was pre-treated by dipping in the 20

synthetic feed for a period of 3 hours at 70 °C in order to fill up all the air spaces present in the porous carbon layers of the 21

electrode and to decrease the overpotential faced during the electrochemical cell start up. It was connected with a 22

Platinum/Iridium (Pt/Ir:- 90/10%) wire to the working electrode of the Biologic potentiostat (VMP3) and Ag/AgCl/3.5M KCl 23

(+199 mV against SHE) was used as reference electrode (Radiometer Analytical).Pt/Ir (90/10%) wire (Alfa Asear) was used 24

as a counter electrode. For the small reactor set up as described in Exp. 3, the same surface area of working electrode to 25

electrochemical cell volume ratio was maintained to scale down the cell operation to 30 ml working volume cell and to 26

monitor its performance in more controlled conditions and observe the subsequent effect on current density and product 27

formation. All the experiments were performed under sterile conditions. 28

29

1.3 Electrochemical measurements 30

31

Electrochemical analysis was conducted using a Biologic multichannel potentiostat (software Easy Lab vs. 10.23). 32

Potentiostatic control was maintained in the bioelectrochemical cell throughout its operation at -0.85 V/Ag/AgCl (3M KCl) 33

with Biologic potentiostat (VMP3) and one point every 15 minutes was recorded. CV with the vertex potentials of 0.6 V to -34

0.9 V were conducted at a scan rate of 1 mVs-1, in order to obtain mechanistic and phenomenological information of the 35

processes occurring in the system. 36

37

1.4 Sampling and analysis methods 38

39

For the analysis of VFA samples, samples were filtered and acidified with 0.5 ml of H2SO4 solution (50%), which were 40

further extracted using diethyl ether method.1 The samples were centrifuged at 1900g (5810R centrifuge, Eppendorf, 41

Hamburg, Germany) for three minutes and the supernatant was transferred to GC vial and analyzed in a gas chromatograph 42

(CE Instruments-Thermoquest) equipped with a Flame Ionization detector (FID) and a ATM,-1000 capillary column (15 m-43

0.53 mm -1.21 m). Carrier gas used was Helium with a constant flow rate of 6 ml min-1. The determination of acetone and 44

other solvents was performed by gas chromatography using AT-WAX capillary column (60 m-0.32 mm-1 m) with FID. D6 45

ethanol was used as internal standard. pH were measured using a pH meter (Knick SE204). 46

For experiments performed for the hermetically closed BES reactor, head space gas analysis was done with a gas 47

chromatograph (Agilent 7890A) equipped with a NUCON SS packed column (length 2 m, ID 2 mm) with He as the carrier 48

gas, at a flow rate of 6.0 ml/min using a thermal conductivity detector. The operating temperature of the injector, the oven 49

and the detector were 50, 100 and 150 °C respectively. The volatile fatty acid (VFA) analysis for this experiment was done 50

using a gas chromatograph (GC Agilent 7890A) using a DB-WAXetr (J&W Scientific) column (30 m X 530 µm X 1µm 51

nominal) with He as the carrier gas, at a flow rate of 1.5 ml/min. The operating temperature of the injector, the oven and the 52

FID detector were 150, 220, and 230 °C respectively. The products were analyzed using High Performance Liquid 53

Chromatography (HPLC, Agilent 1100 series, USA) equipped with Aminex® HPX-87H column (1300 X 7.8 mm, Bio-rad ) 54

with Refractive index detector. (RID) The mobile phase used here was 0.005M H2SO4 at a flow rate of 0.6-0.9 ml/min at 80 55

°C. 56

57

1.5 FISH analysis 58

FISH samples were prepared using the protocol by Daims.2 Freshly inoculated active culture was fixed by using 4% 59

paraformaldehyde solution in PBS. This was followed by immobilization of cells on a microscopic slide and hybridization 60


with 16S rDNA probes at 48 ºC, 0-65% formamide, followed by washing of specimens by wash buffer (48 ºC). Acquisition 1

system (Coolsnap, Roper Scientific Photometrics) connected to Axioskop2 epifluorescence microscope (Zeiss, Germany) 2

was used to record fluorescence signals. 3

References 4

5

1. W. V. Hecke, P. Vandezande, S. Claes, S. Vangeel, H. Beckers, L. Diels and H. D. Wever, Bioresour. Technol., 6

2012, 111, 368-377. 7

2. H. Daims, Cold Spring Harb. Protoc., 2009.4, 8 pp. 8

3. D. Pant, G. V. Bogaert, M. D. Smet, L. Diels and K. Vanbroekhoven, Electrochimica Acta, 2010, 55, 7709-7715. 9

4. F. Zhang, D. Pant, B. Logan, 2011, Biosensors and Bioeelectronics, 30, 49-55. 10

5. Y. Alvarez-Gallego, X. Dominguez-Benetton, D. Pant, L. Diels, K. Vanbroekhoven, I. Genné, P. Vermeiren, 2012, 11

Electrochimica Acta, 82, 415-426. 12

13

14


1

S2: FISH micrographies 2

Maximized FISH micrographies of inoculum source as presented in Figure 2. Probe combinations: (a-d) ARC915/red/cy3 3

(Archea) and EUB 338mix/green/fluos (Bacteria domain); (e-h) BET42a/red/cy3 (Betaproteobacteria) and 4

AQS997/green/6fam (Curvibacter); (i-l) EUB338I red/cy3 (Bacteria domain) and DELTA 495a/green/6fam 5

(Deltaproteobacteria); (m-p) SRB385/red/cy3 (Desulfovibrionales) and DSBAC357/green/6-fam (Desulfobacteraceae and 6

Syntrophobacteraceae). Column1 and 2 indicates all green and red probes individually over DAPI stained DNA in the 7

background, respectively. Columns 3 and 4 show superimposed probes, in the absence and presence of DAPI stain for DNA 8

in the background, respectively. Acquisition system (Coolsnap, Roper Scientific Photometrics) connected to Axioskop2 9

epifluorescence microscope (Zeiss, Germany) was used to record Fluorescence signals. 10

11


S3: Schematic of the reactor showing the number of products formed from the initial substrates added into the reactor. In the 1

order of number of carbon, the products synthesized in the reactor include methanol, ethanol, acetone, propionic acid, lactic 2

acid, propanol, glycerol, butanol and caproic acid 3

4


S4: Experiments carried out to differentiate direct electron uptake from hydrogen mediated bioelectrocatalysis. 1

2

In order to verify that the current measured was due to bioelectrochemical reduction of the VFAs and not to the classical 3

electrochemical proton reduction to H2, several experiments were carried out. The experiments were performed in sterile 4

conditions. 5

Exp. 2 Whenever current was observed to decline in Exp 1, a consecutive addition of substrate was applied. Each time, 6

systematic restoration of the reduction current was observed (repeated 20 times as shown in Fig. 1a/S4.1a). 7

Exp. 3 A section of the EA-biofilm-electrode (Fig. 1a) was cut at day 60 and placed in smaller reactor with fresh medium 8

(Fig. 1b/S4.1b). The ratio between the electrode surface area (3 × 1 cm2) and operational volume (30 ml) was kept constant. 9

As expected, with the EA-biofilm already present over the electrode, almost immediate appearance (~1-2 h) of reduction 10

current was observed for the half-cell biocathode, suggesting that the reduction process was neither accomplished by the 11

planktonic microorganisms nor mediated by a compound present in the bulk exhausted medium (e.g. a microbially-produced 12

mediator). As for Fig. 1(a)/ (S4.1a), current density (Fig. 1(b)/S4.1.b) was also restored after every consecutive addition of 13

substrate (repeated 6 times), reaching current densities as high as 160-210 A/m2. Exp. 1, 2 and 3 resulted in the production of 14

an assortment of chemicals, derived from the bioelectrochemical reduction of acetate and butyrate. 15

Exp. 4 The exhausted medium of reactor corresponding to Fig 1a was collected and placed in an independent reactor with a 16

clean, non-colonized electrode, polarized at -0.85 V. H2 was immediately generated (visible bubbles). Yet, despite 17

consecutive substrate additions, progressive current consumption only happened after 3 days of operation, when EA-biofilm 18

was successfully established. Maximal current density reached for this period (day 0 to 3) was 12.4 A/m2. (See Fig. S4.2) 19

Planktonic SRB were present; however, even if they would be capable of consuming the electrogenerated H2 as energy 20

carrier, the quantity of charge was not affected by their presence or metabolic activity. This experiment was also performed 21

by including a filtering pretreatment (0.22 µm), to remove the bacteria. No significant differences were observed. The 22

constant and low current density that evolved, was attributed to classical H2 evolution. 23

Exp. 5 Control experiments with SRB in the presence of a non-polarized electrode were also considered. Microbial biomass 24

did not show noticeable increase (observed by optical density). 25

Exp. 6 Additional control experiments with SRB in the absence of electrodes were also considered. Suitable electron donor 26

(sodium lactate) and acceptor (acetate and butyrate) were added. Although copious biomass growth was observed, as was 27

also the case of Exp. 5, no alcohols, acetone or other reduced products from acetate or butyrate were detected. 28

Exp. 7 Abiotic and sterile experiments with clean electrode in synthetic feed (Exp. 7a) was also performed. Additionally, 29

control experiment with only the stainless steel mesh (Exp. 7b) current collector (without carbon) as working electrode were 30

also conducted to facilitate H2 evolution alone, but contrary to the work of De Silva et al.,1 only low current densities were 31

reached (~1-8 A/m2) for rather short periods (See Fig S4.2b) even after subsequent substrate additions. None of these resulted 32

in current density improvement correlated to substrate consumption, or in the production of alcohols, acetone or elongated 33

VFAs. 34

Exp. 8 Exp. 1–7 were not adapted to accumulate produced gases in the headspace, as they were continuously flushed with 35

sterile N2 flow-rate. For this reason, independent reactors were set hermetically, without any atmosphere alteration. From the 36

first hours up to 2 days, the composition of the headspace progressed to 92-99% H2. Traces of CO2 (0.2-0.6%) evolved and 37

no methane was produced (See Fig. S4.3). In such experiments, the products detected in Exp. 1–3 were not found. In 38

contrast, traces of glycerol (up to 0.96 g/L) and lactic acid (up to 2.36 g/L) were formed (See Fig. S4.4). Several reported 39

bioelectrochemical conversions are favored in the presence of high HPP provided in the headspace or entrapped on the 40

electrode surface with the EA-biofilm (e.g. butyrate to butanol by mixed cultures of fermentative bacteria).2 The effect of 41

increased HPP resulted here was detrimental for effective bioelectrosynthesis of alcohols and acetone. The influence of HPP 42

on product selectivity will be elucidated in future studies. Equally important, in Exp. 8, accumulation of H2 in the headspace 43

radically affected the reduction current, which was not anymore varying with systematic additions of substrates (See Fig. 44

S4.2). Thus, product formation is possible in the presence of H2 as electron donor, but it was clearly not the main driver for 45

the conversions achieved by SRB in Exp. 1, 2 and 3. 46

Exp. 9 A double chambered BES reactor was also set up, separated by a CMI-7000 cation exchange membrane (Membranes 47

International, USA). The chronoamperogram for this particular reactor set up has been shown in Fig. S4.5. Highest current 48

density achieved in this system was 210 A/m2. Scheme as well as the pictorial representation of this set up has also been 49

provided as Fig. S4.6 and S4.7 respectively. 50

51

References 52

53

1. L. De Silva Muñoz, B. Erable, L. Etcheverry, J. Reiss, R. Basséguy and A. Bergel, Electrochem. Commun, 2010, 54

12, 183-186 55

2. M. T. Agler, B. A. Wrenn, S. H. Zinder and L. T. Angenent, Trends Biotechnol., 2011, 29, 70-78 56

57


1 2

S4.1: As described in Exp. 2and 3, this graph shows representative curves of the current density evolution of two different 3

half-cell reactors poised at -0.85 V vs Ag/AgCl, with consecutive additions of substrates marked in red circles in Fig. S4.1 a 4

and Fig S4.1b respectively. 5

6


1

2 3

S4.2: As described in Exp. 4,7 and 8, this graph shows representative curves of the current density evolution of four different 4

half-cell reactors poised at -0.85 V vs Ag/AgCl, with consecutive additions of substrates marked in red circles. Black, blue 5

and green color represents chronoamperograms that correspond to Exp. 4, 7 and 8 respectively. Exp 4 describes the 6

independent new reactor in which clean non-colonized fresh electrode was place in the exhausted medium of the reactor 7

corresponding to Fig 1(a). For the first three days of operation, the reactor showed maximum current density of 12.4 A/m2 8

but later it started to increase after successful establishment of the biofilm on the electrode surface. Exp.7 discusses the 9

abiotic and sterile control experiments in which clean electrode in synthetic feed (Exp. 7a) and ss mesh (Exp. 7b) were used 10

as working electrode. As seen in the case of Exp.7b current density not more than 8 A/m2 was observed even after 11

subsequent substrate additions. Exp. 8 describes the effect of accumulation of gases inside a hermetically closed BES reactor. 12

As seen in the chronoamperogram, the current density remained stable around 5.7 A/m2 throughout the reactor operation even 13

after subsequent substrate addition. 14

15


1 S4.3: As described in Exp. 8, this graph shows the gas composition inside the hermetically closed reactor. Since the trend did 2

not change beyond this time, the measurements were discontinued 3

4

5 S4.4: As described in Exp. 8, this graph shows the production formation inside the hermetically closed reactor. Since the 6

trend did not change beyond this time, the measurements were discontinued7

8


1

2 S4.5: The chronoamperogram represents the current density evolution of a double-chambered BES reactor separated by a 3

CMI-7000 cation exchange membrane (Membranes International, USA) as described in Exp.9. Highest current density 4

achieved in this system was 210 A/m2. 5

6


S4.6: Scheme of the double chambered bioelectrochemical reactor used for the experiment described above in Exp. 9 1

2


S4.7 Pictorial representation of one of the electrochemical double chambered cell configuration separated by a CMI-7000 1

cation exchange membrane (Membranes International, USA). The performance of this system has been described in Exp. 9. 2

3


S5: Mechanistic considerations of the bioelectrocatalyzed reduction of acetic and butyric acids to alcohols, acetone 1

and caproate, by sulfate-reducing bacteria. 2

The organic compounds identified in solution, other than acetate and butyrate, namely alcohols, acetone and caproate in trace 3

quantities, were detected progressively with the start of current utilization. Therefore, it is assumed that all products were 4

consequence of bioelectrocatalyzed reduction reactions, as shown in Fig. S5.1, in which the electrons are primarily supplied 5

by the cathode to the SRB via direct electron transfer. SRB would be bounded to active sites at the electrode surface, which 6

facilitate electron and proton transport processes. Cathodic deprotonation of the weak carboxylic acids would occur at the 7

cathode and this would boost microbial uptake. Electrochemically or bioelectrochemically generated hydrogen is considered 8

to partake a secondary role in the formation of propionate, caproate, methanol and propanol.1 Moreover, DET controlled 9

reduction of acetate and butyrate would be a critical step for the production of propionate, methanol and propanol. 10

11

12 13

Fig. S5.1. SRB bioelectrocatalyzed reduction of acetic and butyric acids to alcohols, acetone and caproate, controlled via 14

direct electron transfer. 15

16

Several mechanisms through which microorganisms accept electrons from cathodes have been proposed, but it has been 17

recently argued that mechanisms for electron transfer at the cathode to microorganisms may be much different than the better 18

known electron transfer to an anode.2 Although such mechanisms require to be critically evaluated with molecular, 19

biochemical and analytical studies, verification of some essential steps on the cathodic bioelectrocatalysis was accomplished 20

by means of electrochemical characterization methods, from which some conclusions can be drawn. 21

22

The EIS response related to this bioelectrocatalytic reduction process systematically revealed an inductive loop at low 23

frequencies (1 Hz to 1 mHz). A representative diagram is shown in Fig. S5.2. This feature is commonly attributed to 24

electrochemical processes limited by the relaxation of the electrode coverage by one or several adsorbed intermediates, and 25

was absent in Exp. 4 and 7. 26

27

28 Fig. S5.2. Complex frequency response diagram of 3 cm2 SRB-half-cell mature biocathode in fresh culture medium poised at 29

-0.85 V vs Ag/AgCl during first day of operation (Figure 1b). 30


On the other hand, the presence of multiple peaks in the voltamperometric response indicates possible presence of multiple 1

redox couples/cofactors across a range of formal potentials.3 The derivative of current with respect to potential as a function 2

of potential at different scan rates (Fig. S5.3), calculated from the experimental voltammograms (e.g. Fig. 1c, day 60 and day 3

0), exhibited three consistent peaks, with a response characteristic of adsorption intermediates thus confirming the 4

observations made via EIS. When a reactant or product is weakly adsorbed onto the surface, voltammograms generally 5

exhibit an enhancement of peak currents. For instance, the peak current of the forward or backward scans scale up or down, 6

accordingly, as a function of the scan rate in the case of weak adsorption of a reactant or a product, respectively. Strong 7

adsorption is characterized by the appearance of such peaks as compared to a case with no adsorption.4 8

The small peak 1 (Fig. S5.3) recorded at 1, 5 and 50 mV/s, exhibits visible increase in magnitude as a function of the scan 9

rate, for both forward and backward cycles; in fact, at the lower scan rate, such peak is even negligible. This indicates a 10

strong adsorption of the electrochemically-active molecule implicated in both the oxidized and reduced forms, performing at 11

a formal potential of -0.418 V vs Ag/AgCl (-0.213 V vs SHE). Membrane-integral b-type cytochromes are known to have a 12

role as electron carriers at about this potential (-0.215 V vs SHE), during energy conservation processes in H+ dependent 13

acetogens.5. Microorganisms containing such type of enzyme (i.e. Moorella thermoacetica) have been previously reported to 14

be implicated in cathodic bioelectrosynthesis, from carbon dioxide.2 SRB may utilize metabolic schemes that also involve 15

them in electron transfer processes.6; yet, the exact nature of the electron donor an acceptor systems for b-type cytochromes is 16

not entirely clear.5 Whether their presence would be confirmed, the electrode would be potentially acting as electron donor 17

for cytochrome-b enzymes strongly adsorbed to the electrode. 18

19

20

Fig. S5.3. First derivative of the catalytic current associated to cyclic voltammograms of the cathodic SRB biofilm. Scan rates 21

were 1, 5 and 50 mV/s, respectively. Inset figure shows a magnification of the voltammogram at applied potentials from 0.65 22

V to -0.05 V, that are associated to peak 1. 23

24


Conversely, peak 2 decreases as a function of the increase in scan rate, with a trend characteristic of weak adsorption for both 1

the reagent and product. Formal potential of this peak was found at -0.272 V vs Ag/AgCl (-0.067 mV vs SHE). The redox 2

potential of menaquinone has been reported to be -0.074 V vs SHE.1 Menaquinone has been found to be functionally 3

associated with membrane-bound fumarate reductase which is always present in high activities even in SRB which are not 4

grown on fumarate or malate. Such enzyme appears to be a necessary redox mediator for dissimilatory fumarate reduction 5

and it has been found together with b-type cytochromes in SRB species (e.g. Desulfovibrio gigas).7,8 Although fumarate was 6

not provided as substrate, on the likelihood that pyruvate had been formed as partially reduced intracellular intermediate, it 7

could be readily transformed into tri-carboxylic acid cycle intermediates. 8

9

Finally, peak 3 is postulated to be limited by weak reagent and strong product adsorption. The formal potential of the redox 10

species involved is at 0.225 V vs Ag/AgCl (0.430 V vs SHE). This is the case of a “non-standard” plateau in the context of 11

bioelectrochemical systems. However, it has been suggested by several authors that this response could be due to 12

cytochromes interfaced with the electrode.9 13

14

In summary, membrane-bound redox enzymes could be a limiting step on the bioelectrocatalytic reduction process.10 In the 15

absence of the classical electron acceptor for sulfate-reducers, such enzymes may facilitate direct electron transfer reflected 16

as stimulation of cathodic current 11,12 Furthermore, whether direct electron transfer would be mediated by these proteins, 17

such would most likely occur via superexchange.10,13 This occurrence would be supported by the features observed through 18

cyclic voltammetry. For instance, the charge of the transient voltammetric peaks (area under the peak) scales in magnitude 19

with the square root of the voltammetric scan rate (at sufficiently high scan rate >20 mV/s), as presented in Figure S5.4. 20

These peaks result from the change in oxidation state of the redox cofactors, presumptively involved in adsorption-mediated 21

redox process with menaquinone, within the SRB electrochemically-active biofilm. Coexistence of diffusional limitations are 22

also confirmed by the decreasing peak current as a function of the cycle number at a scan rate of 50 mV/s, whereas the peak 23

current for peak 2 shows slight increase at the lower scan rates due to adsorption predominance. These observations are 24

consistent with superexchange features10 and are attributed to the membrane-bound redox enzymes. 25

26 Fig. S5.4. Characteristic response of the charge of transient voltammetric peak, corresponding to peak to on Figure S4.3, at 27

scan rates: 20, 30, 50, 80 and 100 mV/s. 28

29

The validity of these postulates is dependent on whether or not the referred enzymes are found to be present in high activities 30

within the SRB mixed culture, as well as on their association as physiological intermediates in the bioelectrocatalytic process. 31

Yet, the adsorption behavior related to the redox couples found by cyclic voltammetry would play a limiting role in the whole 32

bioelectrocatalytic mechanism. 33

34

Moreover, cell-bound polymeric filaments extending into the extracellular environment were consistently found (Figure 35

S5.5). Although their implication in the electrochemically-active phenomena has not been yet elucidated, ongoing 36

experiments suggest that current densities increase in proportion to their qualitative abundance. Whether such structures 37

would partake a role on direct electron transfer, such would also be consistent to an electron superexchange mechanism.10 as 38

the response of all the voltamperometric peaks found exhibits the same trend. Of course, such hypotheses would require 39

meticulous quantitative corroboration by correlation between the increase in biofilm conductivity, the abundance of these 40

filamentous structures and charge consumption. 41


1

2

S5.5. Electronic photomicrography of cathodic SRB biofilm that electrocatalyzed the reduction of acetic and butyric acids to 3

alcohols, acetone and caproate. Extracellular polymeric filament-like structures were also observed in the SEM 4

micrographies that may or may not have an important role in the overall DET process 5

6

In summary, acetic and butyric acids were provided to an SRB culture which was capable to reduce them to mostly alcohols 7

and acetone and caproate in trace amounts; such process is accomplished via direct electron transfer, limited by a 8

combination of diffusion and specially adsorption phenomena. b-Type cytochromes and menaquinone are two likely types of 9

enzymes which are critical on this bioelectrocatalytic conversions. 10

11

However, more studies will be required to characterize the chemical species involved and ascertain the overall mechanism, 12

especially as not all the biocatalytic transformations steps have been distinguished thus far (i.e. not all the chromatographic 13

and voltamperometric peaks have been resolved). 14

15

References 16

1. R. K. Thauer, K. Jungermann and K. Decker, Bacteriological reviews, 1977, 41, 100-180. 17

2. K. P. Nevin, S. A. Hensley, A. E. Franks, Z. M. Summers, J. Ou, T. L.Woodward, O. L. Snoeyenbos-West and D. R. Lovley, 18

Appl. Environ. Microbiol., 2011, 77, 2882-2886. 19

3. R. M. Snider, S. M. Strycharz-Glaven, S. D. Tsoi, J. S. Erickson and L.M. Tender, PNAS, 2012, 109, 15467-15472. 20

4. D. D. Macdonald, Transient techniques in Electrochemistry, 1977, ISBN: 978-1-4613-4147-5 doi: 10.1007/978-1-4613-4145-1. 21

5. V. Muller, Appl. Environ. Microbiol., 2003, 69, 6345-6353. 22

6. M. Dworkin, S. Falkow, E. Rosenberg, K.-H. Schleifer, E. Stackebrandt, (Eds), The Prokaryotes, A handbook on the Biology of 23

Bacteria, Springer, New York, 2006, pp 893. 24

7. B. Kamlage and M. Blaut, J. Bacteriol, 1992, 174, 3921-3927. 25

8. J, Macy, I, Probst and G. Gottschalk, J. Bacteriol, 1975, 123, 436-442. 26

9. E. LaBelle and D. R. Bond, Cyclic voltammetry for the study of microbial electron transfer at electrodes, 27

In Rabaey K, Angenent L,Schröder U, Keller J , Bioelectrochemical systems: from extracellular electron transfer to 28

biotechnological application. IWA Publishing, London, United Kingdom, 2009, p 137–152. 29

10. D. R. Bond, S. M. Strycharz-Glaven, L. M. Tender and C. Torres, ChemSusChem, 2012, 5, 1099-1105. 30

11. H. Venzlaff, D. Enning, J. Srinivasan, K. J. J. Mayrhofer, A. W. Hassel, F. Widdel and M. Stratmann, Corros. Sci. 2013, 66, 88-31

96. 32

12. E. C. Hatchikian, V. M. Fernandez, R.Cammack, Microbiology and Biochemistry of Strict Anaerobes involved in interspecies 33

hydrogen transfer, ed. J.P. Bélaich, M. Bruschi and J.L. Garcia, Springer, New York, USA, 1990, 54, 53-73. 34

13. N. S. Malvankar and D. R. Lovley, ChemSusChem, 2012, 5, 1039-1046. 35

36

37


S6: Representative GC chromatogram of day 19 of the Exp. 2. The extra peaks in the chromatogram represent products 1

which are still not determined with the present calibration method used for GC analysis. 2

S6: Representative GC chromatogram of day 19 of the Exp 2 The extra peaks in the chromatogram represent products1


Bioelectrocatalyzed reduction of acetic and butyric acids via direct electron transfer using a mixed...

Documents

Transcript of Bioelectrocatalyzed reduction of acetic and butyric acids via direct electron transfer using a mixed...