food and bioproducts processing 9 8 ( 2 0 1 6 ) 196200

Contents lists available at ScienceDirect

Food and Bioproducts Processing

journal homepage: www.elsevier.com/locate/fbp

Production of low or non-alcoholic beer in

microbial fuel cell

Attila Szollosi , Quang Duc Nguyen, Attila Gabor Kovacs,

Attila-Levente Fogarasi, Szilard Kun, Beata Hegyesne-Vecseri
Corvinus University of Budapest, Faculty of Food Science, Department of Brewing and Distilling, Menesi St. 45,
Budapest H-1118, Hungary

a r t i c l e i n f o a b s t r a c t

Article history: Microbial fuel cells (MFC) are bioreactors that are used to generate electricity on the expense
Received 10 July 2015 of organic substrate oxidation. The MFCs operation-mechanism offers the possibility to
Received in revised form 12 use the fuel cell in the beer production technologies to make low or non-alcoholic beers.
December 2015 In this study the effect of MFC anode surface area and riboavin as an electron shuttle
Accepted 14 January 2016 on electric current and on ethanol production in fermentation of wort is demonstrated.
Available online 25 January 2016 The enlargement of anode surface induces an increase of electricity generation (from 4
to 12.5 mA/m2 ) while the ethanol production decreases. Adding riboavin to the wort in
Keywords: the MFC results signicant increase in electricity production, furthermore it decreases the
Brewers yeast ethanol synthesis. Addition of 50 M riboavin enhances the current output to 51 mA/m2
Microbial fuel cell with a 1.5 V/V% decrease of ethanol production, while 100 M results 86 mA/m2 and 2 V/V%
Non-alcoholic beer decrease of ethanol production. These results show the potentiality to brew low or non-
Low-alcoholic beer alcoholic beer in MFC.
Riboavin 2016 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Yeast metabolism

1. Introduction removal of alcohol from regular beer (vacuum distillation,

Microbial fuel cells (MFC) are applied in the wastewater
treatment processes of industrial, agricultural, and munici-
pal wastewaters with the benecial side effect of generating
electric current (Kim et al., 2007). MFC technologies are
already successfully used in semi industrial-scales in a brew-
ery (Queensland, Australia) (Zhou et al., 2013) and in a winery
(Cusick et al., 2011) as part of a wastewater treatment process
(Feng et al., 2008; Wen et al., 2010).
Beer is described as an aromatically sensitive beverage,
which means that even minor modications bring major
changes in the aroma prole of beers. The removal of ethanol
from the beer heavily affects the sensory properties of the
product (Brown and Hammond, 2003). There are two main
strategies to produce alcohol free beer one is using physical
methods, the other is gaining benets from the biologi-
cal processes. The physical methods are based on gentle
dialysis, osmotic distillation, etc.) (Liguori et al., 2015). Bio-
logical approaches are based on limited ethanol formation
during fermentation. The inhibition of ethanol production
can be achieved by using special yeast strains (Selecky et al.,
2008), and also using continuous fermentation with immo-
bilized (entrapped, enclosed) yeasts (Pilkington et al., 1998),
etc. With these biological methods the sensory properties
of the low (or non) alcohol beer differ less from the alco-
holic homologue than the physically treated beer, such beers
are usually off-avored, because of the aroma-loss (Branyik
et al., 2012). Several studies reported the capability of extra-
cellular electron transport of yeast cells in the presence of
different mediator molecules like methylene blue, neutral red,
riboavin, etc. in MFCs (Babanova et al., 2011; Rawson et al.,
2012). In MFC the reduced NADH molecules are re-oxidized by
the reduction of mediators thus the electron transport chain
terminates (Prasad et al., 2007). The electrons are transported

Corresponding author. Tel.: +36 14826247.
E-mail address: attila.szollosi@uni-corvinus.hu (A. Szollosi).
http://dx.doi.org/10.1016/j.fbp.2016.01.012
0960-3085/ 2016 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
 food and bioproducts processing 9 8 ( 2 0 1 6 ) 196200
through the microbial cell wall, and the cycle terminates by
the oxidation of mediators on the fuel cell anode surface.
Therefore the alcoholic fermentation pathway is deactivated
in the absence of NADH, and the formation of ethanol is
limited (Searle and Kirsop, 1979). In addition many Saccha-
romyces cerevisiae strains can metabolize ethanol as carbon
source in sugar and nutrient limited and aerated conditions
(Kaliterna et al., 1995). The anode of MFC also serves as elec-
tron acceptor this way enhancing the conversion of ethanol
to CO2 through gluconeogenesis. In this study a new fer-
mentation method is demonstrated for low alcoholic beer
production. The hopped sweet beer-wort samples were fer-
mented in MFC systems and the effect of MFC redox conditions
on the yeast carbohydrate metabolism; ethanol production
and electricity production were evaluated. The effect of the
size of the MFC anode surface, and the effect of the non-toxic
extracellular electron mediator, riboavin on the electric per-
formance and the ethanol production of yeast were analyzed
as well. Using MFC for wort fermentation gives us new pos-
sibilities to make low or non-alcoholic beer while generating
electricity during the fermentation.
2. Materials and methods
2.1. Inoculum preparation
In the experiment S. cerevisiae WS-120 (Hefebank Weichen-
stephan, Mnchen) a bottom-fermenting (lager) yeast was
used as fermentative microorganism in the microbial fuel
cell. For propagation of the yeast strain YEPD (Yeast
extractPeptoneDextrose) broth was used at 30 C.
2.2. Composition of wort
The wort for the experiments was made of pilsner malt with-
out any adjunct. The all-malt wort was brewed with infused
mashing that was followed by the mash separation and hop
boiling steps (EBC). The original extract content of the beer
was 12 B, with bitterness of 25 IBU.
2.3. Construction and operation conditions of MFCs
In the experiment, dual-chamber type of MFCs were applied.
In every case approx. 107 CFU/mL yeast cell was used to inoc-
ulate the anode chamber. Riboavin was added to the hopped
wort, as a non-toxic extracellular mediator in concentrations
of 0 M, 50 M and 100 M. Potassium ferrocyianide, 0.1 M,
was used as catholyte in the MFC. The working volume of
anode and cathode chambers was 12 mL separately, and they
were separated with proton-exchange membrane (Naon 117).
Graphite plate, projected surface area of 12 cm2 (smaller size)
and 24 cm2 (larger size), was used as anode in the MFC sett-
ings. Graphite plate was used as cathode in each MFC, having
a projected surface of 24 cm2 . The MFCs were operated sepa-
rately in batch mode, no pre-inoculation was used. After each
experiment the MFCs were emptied, cleaned and sterilized.
Electrodes were connected to an external resistance (500 )
and parallel connected to a digital multimeter. The generated
voltage on the external resistance was measured. Using Ohms
law (I = V/R) the electric current was calculated from the regis-
tered voltage values. The running temperature was 15 C in the
metabolic analysis, in other cases 30 C. The pH of the anode
and cathode chamber was neutral (pH 7.0 0.5). Each exper-
imental run was carried out simultaneously, in triplicates in
197
three independent MFC units to conrm the reproducibility
and reliability of the results.
2.4. Control experiment
As a control sample, the same type of wort was fermented with
the same yeast strain on 30 C for 3 days in the MFC device,
however the two electrodes (anode and cathode) were not con-
nected to each other. All other conditions were the same as
other MFC settings.
2.5. Fermentable sugar concentration measurement
Fermentable sugars of the wort were detected by a HPLC
system (Surveyor, Thermo Scientic, San Jose, USA) with an
Hi-Plex H column (Agilent, Santa Clara, USA). The mobile
phase was 5 mM H2 SO4 solution. The ow rate for elution
was 0.6 mL/min at 45 C. The measurement time was 25 min at
constant ow rate. The carbohydrates were detected by refrac-
tometric index (Surveyor RI Plus Detector, Thermo Scientic,
San Jose, USA). All chemicals for the standards were HPLC
grade of purity and used without further purication.
2.6. Alcohol content measurement
Two different methods were used to determine ethanol con-
tent of the samples. In case of metabolism analysis the ethanol
concentration was measured by HPLC (same set up as at car-
bohydrate measurement) with RI detector. In other cases the
alcohol content of the beer was measured with Anton Paar
4500 M Alcolyzer system. After 3 days of operation ethanol
concentration of the anolyte of MFC was measured. In case
of the control samples the fermented wort samples were cen-
trifuged to remove separate sediments from the uids and cold
bath-ultrasound treated to remove the produced CO2 .
2.7. Statistical analysis
Statistical analysis was carried out using the Statistica 8 (Stat-
soft, USA) software. One-way analysis of variance (ANOVA)
for repeated measures was applied to the data obtained from
the result of ethanol measurements. Results from treatments
showing signicant overall changes were subjected to post-
hoc Tukeys test with signicance for = 0.05 (p-value shows
the possibility of making statistical rst type error, n is the
size of the sample, is the level of signicance).
3. Results and discussion
3.1. Yeast metabolism during the MFC treatment
Wort in the anode chamber of three MFC unit was inoculated
with S. cerevisiae yeast. The three independent MFC units were
run on 15 C for 2 weeks which is a typical fermentation time
in beer fermentation. The sugar consumption throughout
the metabolism of the yeast was evaluated by simultaneous
monitoring of fermentable carbohydrate concentration (tak-
ing samples in every 8 h), ethanol concentration and electricity
generation. The measured parameters are shown on Fig. 1.
This kinetics of current generation is in accordance with
the main metabolic pathway regulations of fermentative
sugars in yeasts (Pasteur-effect, Crabtree-effect). The perfor-
mance of electricity production in yeast-based MFC can be
signicantly inuenced by the substrate concentration. At the
198 food and bioproducts processing 9 8 ( 2 0 1 6 ) 196200
Fig. 1 Change of fermentable carbohydrate concentration,
ethanol concentration and generated currentdensity
during fermentation in microbial fuel cell.
early stages of the fermentation both ethanol and electric-
ity production was negligible however the fermentable sugars
concentration decreased. This is a typical phenomenon in
the lag phase of the yeast; the residual oxygen in the anode
chamber also has a relevant effect on the yeast metabolism,
too much oxygen leads to yeast cell multiplication (propa-
gation) while no oxygen can cause limited lipid synthesis
which further limits the formation of the cell organs. Initially
the concentrations of the fermentable carbohydrates were
high enough to inhibit the aerobic respiration (Crabtree effect)
therefore ethanol production is preferred by the yeast cells. As
the concentration of fermentable carbohydrates reached the
critical value (approx. 45 g/100 mL) electricity started to be
produced. Furthermore the intensity of electric current gen-
eration increased while the ethanol content of the broth is
reduced at the same time.
In a previous study (Jafary et al., 2013) researchers experi-
enced similar metabolic characteristics when they evaluated
the effect of substrate load on the performance of MFC using
S. cerevisiae as a bio-catalyst; using high concentrations of sub-
strates they observed an inverse relationship between the MFC
electrical performance and the substrate load. The research
group suggested this phenomena is probably due to substrate
inhibition on the electricity production in the MFC. This sug-
gestion is in accordance with the theory of Crabtree-effect,
because the fermentable carbohydrates were above the critical
concentration. In the beginning of the fermentation the reduc-
ing sugar content of the wort is too high, therefore the yeast
produces ethanol from the carbohydrates even if the redox
milieu is more suitable for the terminal oxidation (Liu et al.,
2013). When fermentable sugar concentration falls under the
critical value yeast cells use the anode of the MFC as an
electron acceptor, this provides adequate redox potential for
simulated aerobic breathing instead of ethanol synthesis. The
repression of ethanol production can be measured through the
electric current output.
3.2. Effect of electrode surface on electricity and
ethanol production
For evaluation of the effect of anode surface size on ethanol
and electricity production two different sized (12 and 24 cm2 )
graphite anode plates were applied in MFC systems. We found
that the anode surface size inuences the ethanol production
of the yeast. The effect on ethanol production of anode surface
is shown on Fig. 2.
Fig. 2 Effect of anode enlargement on ethanol production
to the third day of fermentation. The anode size of the MFC
setting was 12.5 and 24 cm2 . Control samples were made in
MFC where the electrodes were not connected.
The enlargement of the anode surface area led to the
reduction of the ethanol synthesis. The ethanol concentra-
tion decreased from 2.9 to 2.3 V/V% on the third day of the
fermentation; this is not signicant difference but it indicates
a tendency. Comparing with the control sample, the alcohol
content change between the 24 cm2 anode surface MFC was
more than 1.2 V/V% on the third day of the fermentation. The
differences were statistically tested by one-way ANOVA (post
hoc Tukey-test). The result of the evaluation showed signi-
cant difference between the control setting and the connected
MFC brewed samples. The difference was relevant at larger
anode surface, in case of 24 cm2 the p-value was 0.000643 (n = 9,
= 0.05), while in case of 12 cm2 the p-value was only 0.00132
(n = 9, = 0.05). The samples which were brewed in connected
electrode MFC did not differ from each other signicantly
(p = 0.459, n = 9, = 0.05).
According to the aforementioned information, during the
fermentation in MFC ethanol is only produced until the fer-
mentable sugar content drops to the critical level. In case of
sugar deciency the yeast utilizes ethanol as carbon source
in presence of O2 or other electron acceptors (MFC elec-
trode), therefore the produced ethanol is metabolized to CO2
(Raghavulu et al., 2011).
The enlargement of the anode surface increased the elec-
tric performance of MFC. Directly after inoculation of the
anodic chamber the electric current production was not
detectable, but after the lag-phase (approx. from the 10th hour
of the experiment) the electric output increased constantly
in both cases (12 cm2 and 24 cm2 ). As the 24 cm2 anode sur-
face sized MFC reached its maximum current production the
voltage values dropped rapidly, supposedly due to the sub-
strate diminution. Despite of these characteristics the smaller
(12 cm2 ) anode surface showed relevantly slower voltage rise
and fall as well. The effect of electric current generation of the
size of anode surface is shown on Fig. 3.
When the anode surface of MFC was 12 cm2 the maximal
generated current-density was only 5 mA/m2 while in case of
24 cm2 anode 12.5 mA/m2 maximal currentdensity was regis-
tered. This phenomenon was expected due to previous studies
 food and bioproducts processing 9 8 ( 2 0 1 6 ) 196200
Fig. 3 Effect of anode surface on the electric current
generation. Two anode size (12.5 and 24 cm2 ) were applied
in MFC.
that demonstrate the need of direct contact of yeast and the
anode surface to successfully pass on the produced electrons
(Prasad et al., 2007). Larger anode surfaces increase the possi-
bility of yeast cell attachment to the electrode.
3.3. Effect of riboavin on electricity and ethanol
production
The use of electron mediators in MFC make possible to create
indirect connection between microbe and anode. In this study,
riboavin was applied as a natural, non-toxic electron shut-
tle. For demonstration of the effect of riboavin addition into
the anolyte on ethanol and electricity production different
riboavin concentrations were applied (0 M, 50 M, 100 M).
At each sample the addition of the redox shuttle molecules
inuenced the ethanol production of the yeast. The effect of
riboavin addition on ethanol production is shown on Fig. 4.
Supplying the wort (anolyte) with 50 or 100 M riboavin
resulted relevant reduction in the alcohol content of the sam-
ples. 50 M and 100 M riboavin addition caused 1.52 and
2.0 V/V% ethanol concentration decrease, respectively on the
third day of fermentation. In order to reveal the signicant
Fig. 4 Effect of riboavin addition on the ethanol
production on the third day of fermentation, with MFC
anode surface of 24 cm2 . The mediator content (riboavin)
of the MFC setting was 0, 50 and 100 M. Control samples
were made in MFC where the electrodes were unconnected
without mediator.
199
Table 1 Result of ANOVA (post hoc Tukey-test); effect of
MFC treatment with different riboavin concentrations
on ethanol production.
Comparison (M) p-Value
Control0 0.001095
Control50 0.000274
Control100 0.000237
050 0.092265
0100 0.008981
50100 0.377574
Control samples were fermented in MFC where the electrodes are
not connected; MFC settings were supplemented with 0, 50 and
100 M riboavin (n = 12, = 0.05).
Fig. 5 Effect of riboavin addition on the electric current
generation. The mediator content (riboavin) of the MFC
setting was 0 (without riboavin), 50 and 100 M.
differences between the different settings one-way ANOVA
(Tukey-test) was performed on the ethanol concentration
results. The results of the statistical evaluation is listed in
Table 1. The control samples were beer-wort that was brewed
and fermented in MFC where the electrodes were not con-
nected. The control samples signicantly differed from all of
the samples.
The characteristic of electric current production is the
same in all cases. The electric current is produced as long
as the carbon sources hold. The addition of riboavin signi-
cantly inuenced the electric current generation and also the
ethanol production. The effect on electric current generation
of riboavin addition is shown on Fig. 5.
The addition of 50 M riboavin resulted approximately
5-folds higher current production (from 12.5 to 51 mA/m2 ),
while 100 M riboavin resulted about 7-folds higher (from
12.5 to 86.4 mA/m2 ) current production. The addition of elec-
tron shuttle molecules give similar results as if the electrode
surface was enlarged (Najafpour et al., 2011; Schroder, 2007).
Although in the present of these molecules the electrons from
the re-oxidation of NADH can be transported to the anode of
MFC, and the electric efciency is increased while the ethanol
production is inhibited, the high concentration of fermentable
sugars turns the metabolism of the yeast to ethanol synthe-
sis. The ethanol content of the beer can be decreased by the
reduction of the concentration of fermentable sugars, increase
of the mediator concentration or extend the time of fermenta-
tion to force the yeast to metabolize ethanol as carbon source.
4. Conclusions
Fermentation of beer-wort in MFC resulted in a decreased
ethanol production in the samples, and in generated electric-
ity, simultaneously. The enlargement of anode surface and the
200 food and bioproducts processing 9 8 ( 2 0 1 6 ) 196200
addition of riboavin as electron shuttle molecule increased
the electric current generation and further decreased the
ethanol production. Although in this study only low alcoholic
beer was produced in the MFC, with the proper enlargement
of anode surface and/or the addition of proper amount of elec-
tron shuttles non-alcoholic beer may be achieved, as well.
Acknowledgements
This work was supported by National Development Agency
(TMOP-4.2.1./B-09/1-KMR-2010-0005, TMOP-4.2.2/B-
10/1-2010-0023 and TECH 09-A3-2009-0194) and MKI Plexi
Kft.
References
Babanova, S., Hubenova, Y., Mitov, M., 2011. Inuence of articial
mediators on yeast-based fuel cell performance. J. Biosci.
Bioeng. 112, 379387.
Branyik, T., Silva, D.P., Baszczynski, M., Lehnert, R., Silva, J., 2012.
A review of methods of low alcohol and alcohol-free beer
production. J. Food Eng. 108, 493506.
Brown, A.K., Hammond, J.R.M., 2003. Flavour control in
small-scale beer fermentations. Food Bioprod. Process. 81,
4049.
Cusick, R.D., Bryan, B., Parker, D.S., Merrill, M.D., Mehanna, M.,
Kiely, P.D., Liu, G.L., Logan, B.E., 2011. Performance of a
pilot-scale continuous ow microbial electrolysis cell fed
winery wastewater. Appl. Microbiol. Biotechnol. 89, 20532063.
Feng, Y., Wang, X., Logan, B.E., Lee, H., 2008. Brewery wastewater
treatment using air-cathode microbial fuel cells. Appl.
Microbiol. Biotechnol. 78, 873880.
Jafary, T., Ghoreyshi, A.A., Najafpour, G.D., Fatemi, S.,
Rahimnejad, M., 2013. Investigation on performance of
microbial fuel cells based on carbon sources and kinetic
models. Int. J. Energy Res. 37, 15391549.
Kaliterna, J., Weusthuis, R.A., Castrillo, J.I., Vandijken, J.P., Pronk,
J.T., 1995. Coordination of sucrose uptake and respiration in
the yeast debaryomyces-yamadae. Microbiology (UK) 141,
15671574.
Kim, B.H., Chang, I.S., Gadd, G.M., 2007. Challenges in microbial
fuel cell development and operation. Appl. Microbiol.
Biotechnol. 76, 485494.
Liguori, L., De Francesco, G., Russo, P., Perretti, G., Albanese, D., Di
Matteo, M., 2015. Production and characterization of
alcohol-free beer by membrane process. Food Bioprod.
Process. 94, 158168.
Liu, C.-G., Xue, C., Lin, Y.-H., Bai, F.-W., 2013. Redox potential
control and applications in microaerobic and anaerobic
fermentations. Biotechnol. Adv. 31, 257265.
Najafpour, G., Rahimnejad, M., Ghoreyshi, A., 2011. The
enhancement of a microbial fuel cell for electrical output
using mediators and oxidizing agents. Energy Sources A:
Recover. Util. Environ. Eff. 33, 22392248.
Pilkington, P.H., Margaritis, A., Mensour, N.A., Russell, I., 1998.
Fundamentals of immobilised yeast cells for continuous beer
fermentation: a review. J. Inst. Brew. 104, 1931.
Prasad, D., Arun, S., Murugesan, A., Padmanaban, S.,
Satyanarayanan, R.S., Berchmans, S., Yegnaraman, V., 2007.
Direct electron transfer with yeast cells and construction of a
mediatorless microbial fuel cell. Biosens. Bioelectron. 22,
26042610.
Raghavulu, S.V., Goud, R.K., Sarma, P.N., Mohan, S.V., 2011.
Saccharomyces cerevisiae as anodic biocatalyst for power
generation in biofuel cell: inuence of redox condition and
substrate load. Bioresour. Technol. 102, 27512757.
Rawson, F.J., Gross, A.J., Garrett, D.J., Downard, A.J., Baronian,
K.H.R., 2012. Mediated electrochemical detection of electron
transfer from the outer surface of the cell wall of
Saccharomyces cerevisiae. Electrochem. Commun. 15, 8587.
Schroder, U., 2007. Anodic electron transfer mechanisms in
microbial fuel cells and their energy efciency. Phys. Chem.
Chem. Phys. 9, 26192629.
Searle, B.A., Kirsop, B.H., 1979. Sugar utilization by a brewing
yeast in relation to the growth and maintenance phases of
metabolism. J. Inst. Brew. 85, 342345.
Selecky, R., Smogrovicova, D., Sulo, P., 2008. Beer with reduced
ethanol content produced using Saccharomyces cerevisiae
yeasts decient in various tricarboxylic acid cycle enzymes. J.
Inst. Brew. 114, 97101.
Wen, Q., Wu, Y., Zhao, L.X., Sun, Q., 2010. Production of electricity
from the treatment of continuous brewery wastewater using a
microbial fuel cell. Fuel 89, 13811385.
Zhou, M.H., Wang, H.Y., Hassett, D.J., Gu, T.Y., 2013. Recent
advances in microbial fuel cells (MFCs) and microbial
electrolysis cells (MECs) for wastewater treatment, bioenergy
and bioproducts. J. Chem. Technol. Biotechnol. 88, 508518.