Simultaneous Production of Bioethanol and Bioelectricity in a Membrane Less Single Chambered Yeast Fuel Cell by Saccharomyces Cerevisiae and Pichia Fermentans

Production of bioethanol and bioelectricity is a promising approach through microbial electrochemical technology. Sugars are metabolized by yeast to produces ethanol, CO 2 and energy. Surplus electrons produced during the fermentation can be transferred through the circuit to generate electricity in a Microbial fuel cell (MFC). In the present study, a membrane less single chambered microbial fuel cell was developed for simultaneous production of bioethanol and bioelectricity. Pichia fermentans along with a well-known ethanol producing yeast Saccharomyces cerevisiae was allowed to ferment glucose. S. cerevisiae demonstrated maximum open circuit voltage (OCV) 0.287 ± 0.009 V and power density 4.473 mW m − 2 on 15th day, with a maximum ethanol yield of 5.6% (v/v) on 12th day. P. fermentans demonstrated a maximum OCV of 0.318 ± 0.0039 V and power density of 8.299 mW m − 2 on 15th day with ethanol yield of 4.7 % (v/v) on 12th day. Coulombic eciency (CE) increased gradually from 0.002– 0.471 % and 0.012–0.089 % in the case of S. cerevisiae and P. fermentans, respectively, during 15 days of experiment. Thus, the result indicated that Single chambered fuel cell can be explored for its potential applications for ethanol production along with clean energy generation.


Introduction
Fossil fuels are naturally occurring carbon or hydrocarbon fuel such as peat and coal while natural gases are formed by the decay of plants or animals. The products formed by fossil fuels are essential for our living and required for the production of materials like plastics, medicines, life-saving devices, gasoline, fuel, heating oil, while natural gas to be used as fuel and generate electricity. In spite of being several usages, increasing price and negative environmental impact of fossil fuels have increased the concerns related to its future availability and use. Several potential alternative fuels including biodiesel, methanol, hydrogen, natural gas, bio-oil, bio-char, lique ed petroleum gas (LPG), Fischer-Tropsch fuel, p-series, bioelectricity, and solar fuels (Aditiya et al. 2016; Baig et al. 2019) have been developed. Moreover, bioethanol is also produced from various agricultural raw materials such as sucrose-containing feedstocks, lignocellulosic materials and starch materials (Bušić et al. 2018). Since the conversion of lignocellulosic biomass into ethanol is di cult due to the fact of being resistant in nature to be degraded (Liu et al. 2019). Bioethanol is a promising liquid fuel that is expected to have promising future.
The increasing interest in bioethanol production started since 1980s and hence has been considered as an alternative fuel in many countries. It helps to reduce CO 2 emission up to 80% as compared to using the petrol, thus encourages a healthier environment for the future. They are recyclable and contribute to sustainability. Similarly, production of bioelectricity in microbial fuel cell (MFC) as biotechnological system is another developing green approach towards renewable energy from sustainable development (Paul et al. 2018; Kumar et al. 2019). Bioelectricity is produced through microbial catalytic activity using various organic sources. In MFC, anode accepts electrons from microbial catabolic activity through oxidation and reduction processes. In general, it is evident that 2 mol of ethanol is produced from 1 mol of glucose (glucose → Pyruvate → acetaldehyde → ethanol) through NADH-dependant enzyme and alcohol dehydrogenase. This reaction generates two ATP molecules, two H + ions and two electrons.
These electrons are stored in cells in the form of NADH. An increasing ratio of NADH to NAD + could help to generate voltage output (Geng et al. 2020). Thus the generated electrons from NADH/NAD + redox cycle may be used in MFC system (Walker and Stewart 2016). Therefore, it is possible that yeast metabolic activity, energy production and its conversion into heat could be e ciently harvested as electricity through a combined approach of MFC during sugar fermentation and ethanol production.
Yeast could potentially be considered as an ideal model organism for MFC applications due to its nonpathogenic nature and being able to utilize a wide-ranging substrate. The most commonly used biocatalyst for fermenting the sugars into ethanol is Saccharomyces cerevisiae, but its use is restricted as it cannot ferment xylose and other 5-carbon sugars present in lignocellulosic material. On the other hand, xylose-fermenting yeasts e.g. Pichia stipitis, Candida parapsilosis and Candida shehatae (Mohd Azhar et al. 2017) may ferment both 5-carbon as well as 6-carbon sugars. Pichia fermentans has also been reported to produce xylitol using non-detoxi ed xylose rich pre-hydrolysate from sugarcane bagasse (Prabhu et al. 2020).
Recently published reports suggests that non-pathogenic Pichia fermentans could be a potential yeast to be used in a microbial fuel cell due to its possible exoelectrogenic property (Pal and Sharma 2019). The signaling molecule produced by one cell and sensed by another to induce oriented growth is considered for cell-cell communication (Stephens and Bentley 2020). Most bacterial communities are embedded in structured extracellular polymeric substances (EPS) to survive in a harsh environment (Flemming and Wuertz 2019). Electroactive bio lms play an important role in bioelectrochemical system via various electron transfer mechanism. Yeasts may also produce bio lm to enhance electron transfer mechanism and bio lm dynamics in microbial fuel cells (Speranza et al. 2020).
Most of the work related to MFC has been focused on reactor design, proton exchange membrane, electrolyte development and modi cation of electrode design and materials to increase electricity production (Christwardana et al. 2019). There are several advantages in MFC as it can be operated in fed batch, continuous or batch mode whether single chambered or double chambered, with membrane or membrane less etc. This paper presents an approach to explore MFC for bioethanol production with simultaneous generation of electricity. The study further evaluates the production of ethanol in MFC by P. fermentans and compared its e ciency with a well-known sugar fermenting yeast S. cerevisiae under a batch type operation in a single chambered Microbial fuel cell. The electrochemical data was generated by calculating current density, power density, output voltage along with the glucose consumption for maximum ethanol yield and fermentation e ciency.

Microorganisms
Pichia fermentans was procured from the Microbial Type Culture collection (MTCC 189) Chandigarh, India, and cultured aerobically in yeast extract and glucose (YG) agar and maintained. And Saccharomyces cerevisiae was purchased from local market and maintained. Both the yeasts, P. fermentans and S. cerevisiae was cultured for 24 hours at 30°C enriched with yeast extract 0.25% (w/v) and Glucose 10% (w/v) in a 1L solution. The same broth medium was used for inoculum preparation to be inoculated in the MFC.

Single Chambered MFC setup
A typical mediator less membrane less single chambered MFC was constructed using glass asks (with working volume 100 mL). Bow shaped Carbon bers (100 cm length, 7 mm diameter) were used as anode and circular stainless steel wire (as a mesh) (100 cm length, 0.05 mm diameter) as cathode (Nam et al. 2018). Both the electrodes were sterilized with ethanol, rinsed with autoclaved distilled water, and treated under UV radiation for 20 min, then dried under aseptic conditions. The asks contained production medium (Glucose 10% (w/v), enriched with yeast extract 0.25% (w/v) was sterilized and then sterilized electrodes were xed at a vertical distance of 3 cm using rubber corks as plug to maintain anaerobic condition for fermentation. The asks were inoculated with either of the yeast (10 % (v/v) and incubated at 30°C for 15 days. All experiments were performed in triplicates and repeated along with suitable controls. The sample (2 ml) was taken out for glucose and ethanol estimations.

Electrochemical calculations
The open circuit voltage (OCV) and output voltage was recorded across different external resistor ranging from 1000Ω to 820KΩ via a digital multimeter. For the preparation of polarization curve the external resistance (Rex) was varied at time intervals. The current (I) was calculated by using Ohm's law V = IR, where V represents cell voltage value and R represents external resistance value. Power was calculated according to P = VR whereas, Current density (j) and power density (p) were calculated using electrode (anode) area (2.2 cm 2 ). The Coulombic e ciency (CE) was calculated using CE = CE x ×100/C TH , CEx is the experimental value of total coulombs. The theoretical value of coulombs (C TH ) was calculated using C TH = FnMV, where F stands for faraday constant (96,485C per mole of electrons), n represents number of electrons produced per glucose unit consumed during fermentation, M is the glucose concentration and V represents the reaction volume (L). Internal resistance (R int ) of the cell was calculated as R int = R (E/V-1) where R represents external resistance, E represents OCV (voltage without any resistance), and V represents output voltage with resistors (Pal and Sharma 2020). All data presented in the manuscript are average of triplicates along with the standard error bars.

Glucose estimation
Concentration of glucose in the cell setup was routinely measured by dinitrosalicylic acid (DNS) method. The quanti cation was performed according to the step-by-step method (Pal and Sharma 2019). DNSA reagent was prepared by dissolving 2g of DNSA and 60g of sodium-potassium tartaric acid in 160 mL of 0.5N NaOH. It was cooled to room temperature and diluted to 200mL with the help of distilled water. Then 1 mL of DNSA reagent was pipette out in a test tube containing 0.5 mL of sample (1g/mL) and kept at 100°C for 5 min. After cooling, 1.5 mL of distilled water was added to the same test tube to stop further reaction and absorbance was measured at 540 nm using UV-VIS spectrophotometer. The glucose concentration was calculated from the standard curve of D-glucose (0.1mg-1g/mL), and results were expressed as mg glucose equivalent (GE) per mL sample. The ethanol fermentation was con rmed by using High performance liquid chromatography.

Ethanol estimation
All parameters were performed with an Agilent HPLC 1260 II INFINITY with autosampler and C18 column (5 µm; 25 x 0.46cm). All the outputs were processed through Agilent Chemstation-Open lab software. All samples were drawn out and recorded on daily basis. The sample was centrifuged at 8000 rpm for 15 min to collect for further process. The total runtime was 6 min with peaks detected with a UV detector. Water was used as a mobile phase with 4% acetone added in an optimized parameter. The standard ow rate was set at 1mL/min with injection volume of 1µL and detection was monitored at 235 nm and 25°C temperature. Standard ethanol solutions were prepared in water ranging from 0.1% -10% (v/v).
Ethanol absorbs less UV compared to high UV absorbing background, which result in the reverse positive or negative peak polarity. Direct estimation of ethanol as a negative peak from samples was con rmed by reversed phase-HPLC and performed same according to (Nisha et al. 2016). In every run, the three peaks were corresponding to acetone, ethanol or water singly or in combination were detected. Acetone, pure water and ethanol samples were injected separately for separate veri cation of retention times into the HPLC system in separate runs. The quantitative analysis of ethanol was calculated from the formula-Concentration of unknown sample = Unknown sample area / standard solution area × concentration of standard solution Where peak areas are in arbitrary units and peak heights and area in milli absorbance units (mAU).
2.6 EPS production P. fermentans and S. cerevisiae were grown separately in 10mL sterilized Yeast extract and Glucose (YG) broth and incubated for 15 days at 30°C. The samples were taken out from both the cultures at altered time intervals (day 1, day 5, 10 and 15) to estimate EPS. Ten mL of cell culture was centrifuged simultaneously at 8000 rpm for 15 min. The carbohydrate in EPS was analysed by phenol/sulphuric acid method and glucose was used as a standard (Pal and Sharma 2019). Protein was estimated by Lowry method and bovine serum albumin was used as a standard (Zhang et al. 2019).

Fourier transform infrared spectroscopy
Functional groups in EPS were detected by Fourier transform infrared (FTIR) spectroscopy. A drop of EPS solution from both the samples was used for study on a glass slide as a dry lm.

Scanning electron microscopy
The visual appearance of P. fermentans and S. cerevisiae was evaluated on anodic surface at 1000X magni cation. The sample was prepared by using electrode material from both the cell cultures at different time interval (day 1, 5, 10 and 15) and was air dried on a glass slide.

Electrochemical response
Electrochemical performance of both the MFC setup containing either P. fermentans or S. cerevisiae, gradually increased after 24 h of inoculation, which was almost stable upto 15 days of the incubation.
Current density also increased gradually from 4.848 mA m − 2 (1st day) to 57.348 mA m − 2 (15th day) ( Fig. 2c and d). Power density of the cell also followed similar pattern with maximum power density of 8.299 mW m − 2 was recorded on 15th day. For S. cerevisiae maximum recorded OCV was 0.287 ± 0.009 V on 15th day (Fig. 1a) with increasing current density from 1.66 mA m − 2 on 1st day to 43.63 mA m − 2 on 15th day ( Fig. 2a and b). Power density was also recorded with same increasing pattern out of which maximum is 4.47 mW m − 2 on 15th day (Fig. 2a).
The internal resistance (R int ) was very high against different R ex from 1000 Ω-82 KΩ for both the organisms on 1st day which drastically decreases from 2nd day onwards ( Fig. 3a and b). Combination of carbon bre anode and stainless steel cathode has been explored well in the setup. Thus indicates the present setup was intact without any physical damage or corrosion and can be reused e ciently after sterilization. MFC appears to be an ecological approach to achieve the cost-effective electricity generation. The maximum power density for P. fermentans was 8.299 mW m − 2 /100mL while the current density was 57.348 mA m − 2 /100mL on 15th day and 0.318 ± 0.0039 V power output for the same day. From past years, Pichia has been explored in the eld of MFC such as Pichia pastoris, Pichia stipitis, Pichia kudriavzevii etc., for power generation with genetic modi cations and less interest was focused on e cient bioethanol formation (Pal and Sharma 2019). Whereas the S. cerevisiae was measured at power density of 4.473 mW m − 2 /100mL on 15th day while the current density was 43.636 mA m − 2 /100mL on 15th day and power output was of 0.287 ± 0.0094 V.
Amongst them P. fermentans was resulted in better power, current density and OCV than S. cerevisiae. This is because of low resistance and large surface area of carbon brush electrode. Makes them ideal features as anodes from small as well as large scale applications of MFCs (Yuan et al. 2020a). Yuan et al recently published a report on simultaneous power generation system and bioethanol formation. In this, they used S. cerevisiae as model organism for dual and single chambered MFC setup with or without Mediator for power generation and ethanol production. This study resulted in a high power output (5.2 ± 0.5 W/m 3 ) and high ethanol yield (92.5 ± 2%). There are many factors affecting electron transport such as cell metabolism, pH value, mediator type, yeast cell growth, substrate. But excessive addition of MB for electron transfer could affect the activity of yeast and lasts in unbalancing and blocking of electron transmission.

Glucose consumption and ethanol production
The quanti able reducing sugar content was carried out using Dinitrosalicylic acid (DNSA) method. Basic principle involves for the method is interaction of an alkaline solution of DNSA with reducing sugars such as glucose and fructose wherein the aldehyde group of reducing sugar is oxidized to the carboxylic acid and the 3-nitro group (NOO − ). The change in colour was measured at 540 nm (Jain et al. 2020). The variation of the orange-red color index was the indication of presence of reducing sugar to high/low extent.
The sugar consumption and ethanol production by these yeasts was studied in 100 mL single chambered cell setup in sugar medium containing 10% (Glucose) 100 mL Erlenmeyer asks, supplemented with 0.25% yeast extract and incubated as described above. The glucose content was calculated from the standardization curve of D-glucose (y = 6.5333x + 0.0613, R² = 0.9909) and expressed as mg/mL. The glucose was e ciently consumed during initial days of incubation period; about 15-20% of glucose was consumed in one day. After 24 h, the glucose concentration reduced from 10% (w/v) to 7.8 ± 0.0004% (w/v) in the reactors inoculated with P. fermentans and 8.4 ± 0.0003% (w/v) for S. cerevisiae, while the glucose concentration was only 0.2 ± 0.00002% (w/v) at the end of the experiment. A detectable amount 0.025% (v/v) of ethanol was observed on day 1 for P. fermentans, which was maximum on 12th day 4.7% (v/v) and remain constant on 13th day and decreased gradually. As, Pichia utilizes hexoses slowly than pentose sugar as a carbon source and ethanol formation during fermentation process (Tahir and Mezori 2020).
Whereas, in the case of S. cerevisiae the ethanol concentration measured on day 1 was 0.03 % (v/v) with remaining glucose concentration of 8.4 % (w/v). S. cerevisiae produced maximum ethanol on 12th day 5.6 % (v/v) and the remaining glucose concentration was left with 0.23 % (w/v) at the end of experiment ( Fig. 4a and b). It shows alcohol production by yeast cells P. fermentans and S. cerevisiae on day wise fermentation analysis.
The 24 h old cells subjected to fermentation in a 10 % glucose solution (10g/100mL) and have ability to ferment glucose to ethanol. The cell setup of P. fermentans showed maximum 4.7 % (v/v) ethanol production of theoretical yield (6.41 % v/v for 10 % glucose) on day 12 and continued to remain constant upto day 13 and gradually decreased on consecutive days (Nandal et al. 2020). Since 1 mol of glucose produces 2 mol of ethanol and 2CO 2 . This infers that the power generation in MFC setup is independent of fermentation process. In point of fact, ethanol and power generation occur concurrently with the process of glucose metabolization through yeast. It been reported from several studies that under anaerobic conditions, 1 mol of glucose can be converted to give 2 mol of ethanol and 2 mol of electrons. It is evident that the under anaerobic conditions, yeast-MFC cannot utilize all the electrons from complete oxidation of glucose. Another major reason is the direct electron transfer by a yeast cell is very limited, as compared to Shewanella and Geobacter (Christwardana et al. 2018). Therefore, this study focused on the simultaneous production of both bioethanol and electricity generation so that the substrate can be fully utilized for signi cant bene t.
Based on the stoichiometric information, glucose concentrations were theoretically converted to electron amounts. It was used to calculate the amount of electrons passed through the MFC circuit during glucose oxidation (Flimban et al. 2019). The CE According to the formula for CE, CE = (CE X × 100/C Th ), one can calculate how many electrons were involved in electron transfer through the MFC external circuit (Yuan et al. 2020b). Over 24 h, the CE of the proposed MFC setup for P. fermentans was 0.012% and increased upto 0.89% gradually along with the increase in the current density (Fig. 1b). Whereas, the CE e ciency after 24 h for the S. cerevisiae was 0.002% and increased to 0. Still there is lot to unravel about the e cient mechanism behind the electron transfer and different substrates conversion to energy generation requires further consideration.

Effect of pH
In industrial ethanol production, yeast tolerates wide range of pH, thus making the whole process less susceptible to contamination. It was observed that higher acidic conditions produces larger amount of ethanol (Mohd Azhar et al. 2017). At the time of yeast fuel cell setup for P. fermentans and S. cerevisiae, the initial pH was maintained at 7. In the cells inoculated with P. fermentans, the pH decreases to 6.5 after 24 h and here ethanol was measured is 0.3% (v/v), which was further decreased to 5 and was constant from day 4 to day 7 and during these days the concentration of ethanol ranged from 0.8-2.3% (v/v). A decrease in pH and increase in ethanol yield was observed during further incubation. In S. cerevisiae after 24 h the pH decreased upto 6.6 and a detectable volume of ethanol was recorded 0.3% (v/v). The pH decreased upto 5 during further 9 days of incubation and increasing ethanol concentration upto 5.6% (v/v). As per the results obtained P. fermentans produced maximum ethanol 4.7% on 12th day with pH 4.3, while S. cerevisiae produced maximum ethanol 5.6% on 12th day of incubation with 4.8 pH ( Figure  S1). The yeast S. cerevisiae strains are considered to be pillars of fermentation industry since then dominated ethanol fermentation due to their low pH tolerance for ethanol formation, organic acids, and low oxygen availability. It is evident that in P. fermentans the correlated effect of respiratory and fermentative pathways supports growth and product formation. This yeast ferments glucose or xylose under oxygen-limited conditions (Kwak et al. 2019).

EPS production
EPS production was observed in both the yeast P. fermentans and S. cerevisiae along with their growth and colonization on anode surface. Growing bio lm were observed under scanning electron microscopic images (Fig. 5).
The bio lm was developed on the anode surface as well as on the top of the medium. A signi cant correlation between EPS production and yeast growth was observed as analysed on different days (day 1, 5, 10 and 15). The cells grew rapidly along with the biomass accumulation on different time intervals and resulted in gradual increase in biomass, protein, EPS and carbohydrate ( Fig. 6a and b).
Both the yeast P. fermentans and S. cerevisiae produced EPS along with a dense bio lm on carbon bre anode, which showed e cient direct electron transfer. However, its EPS may have boosted the electron transfer via the indirect mechanism (Pal and Sharma 2019). Evident studies suggested that the combination of yeast attached anode improves electron transfer directly creating synergistic effect (Chung et al. 2016). EPS composed of polysaccharides, extracellular DNA, glycoproteins, glycolipids and proteins. This EPS plays some signi cant roles such as microbial cell to cell communication, protection from external and specially extracellular electron transfer (Flemming et al. 2016). It is evident that presence of carbohydrate in EPS carry out speci c functions in mat formation.
The protective glycocalyx acts as a mediator in yeast for electron transfer and may be involved in oxidation and reduction reactions. The presence of EPS matrix was con rmed by infrared (IR) spectrum. After 24 h (day 1), the EPS production for Saccharomyces cerevisiae and Pichia fermentans was minimum but increased gradually during further incubation on day 5, day 10 and day 15. The spectrum of puri ed EPS showed numerous peaks from 3585 − 502 cm ¹ ( Figure S2a

Conclusion
In this study a system for simultaneous power generation and ethanol production was developed. The study demonstrated ethanol production by Pichia fermentans and Saccharomyces cerevisiae, which reached upto 4.7% and 5.6% v/v of the theoretical yield (6.41% (v/v)) respectively for 10% glucose in yeast microbial fuel cell. Analysis of EPS showed the presence of polysaccharides, protein having several functional groups like C = O, -CONH 2 , −CH 3 and OH, of P. fermentans and S. cerevisiae. Another important aspect of the present study is the application of P. fermentans for ethanol production in a fuel cell, which was not explored yet. P. fermentans turned out as an e cient yeast for further microbial fuel cell application.

Declarations
Ethics approval and consent to participate: Not Applicable

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.