The possibility of utilizing rainwater as a sustainable anolyte in an air-cathode microbial fuel cell is investigated in this study. The results indicate that the proposed microbial fuel cell can work within a wide temperature range (from 0 to 30 oC), and under aerobic or anaerobic conditions. However, the rainwater season has a distinct impact. Under anaerobic conditions, the summer rainwater achieves a promised open circuit potential of 553±2 mV without addition of nutrients at the ambient temperature, while addition of nutrients leads to increase the cell voltage to 763±3 and 588±2 mV at 30 oC and ambient temperature, respectively. The maximum open circuit potential for the winter rainwater (492±1.5 mV) is obtained when the reactor is exposed to the air (aerobic conditions) at ambient temperature. Furthermore, the winter rainwater microbial fuel cell generates a maximum power output of 7±0.1 mWm-2 at a corresponding current density value of 44±0.7 mAm-2 at 30 oC. While, at the ambient temperature, the maximum output power is obtained with the summer rainwater (7.2±0.1 mWm-2 at 26±0.5 mAm-2). Moreover, investigation of the bacterial diversity indicates that lactobacillus sp. is the dominant electroactive genus in the summer rainwater, while in the winter rainwater, Staphylococcus sp. is the main electroactive bacteria. The cyclic voltammetry analysis confirms that the electrons are delivered directly from the bacterial biofilm to the anode surface and without mediators. Overall, the study opens a new avenue for utilizing a novel sustainable type of microbial fuel cell derived by rainwater.

2.1. Rainwater samples

The samples were collected in April and December for summer and winter rainwater samples, respectively. A 500 ml of each rainwater sample was collected in a sterilised glass bottle at elevation of 59 meters above sea level in Jeonju, Jeollabuk-do province, Republic of Korea. A 200 ml from each sample (immediately after collecting) was transferred to 1 L sterilised bottle for chemical and microbiological analyses. These samples were transported in a chilled-cold box to the laboratory and processed within 8 h of collection. The average corresponding electrical conductivities of the collecting samples were 61 and 24 µS for the summer and winter rainwaters, respectively.

2.2. Anolyte preparation

The rainwater (without pretreatment) was used as anolyte without additives, and with Nutrient Broth media (1:1 ratio v/v) which contains (for 1 L), 15g of peptone (Samchun Pure Chemical Co., Ltd.), 3g of yeast extract (Samchun Pure Chemical Co., Ltd.), 1g of glucose (Junsei Chemical Co., Ltd.) and 6g of sodium chloride (Samchun Pure Chemical Co., Ltd.), for comparison purposes.

2.3. MFC Construction and operation

A single chamber air-cathode MFC made of transparent polyacrylic was used. 4 cm between the anode and the cathode was maintained. An Ag/AgCl reference electrode is placed in the anode compartment to measure the potential of the electrodes. Cation exchange membrane (CEM, CMI-7000, Membrane International Inc., NJ, USA) was used for ions exchange. The membrane was firstly treated by dipping in a H2SO4 solution (0.5 M) for 18 h, and then kept in a distilled water. The cathode was 2.4 cm × 2.4 cm Pt-loaded (0.5 mgcm-2) carbon cloth (EC-20-5, Electro Chem, Inc., USA). The cathode material was in a direct contact with the natural air. A carbon felt (2.4 cm × 2.4 cm, 3.18 mm, Alfa Aesar) was used as an anode. The power calculation was normalized for the anode projected surface area (6.25 cm2). The current collectors were two 0.1cm-thick high corrosion resistance stainless steel plates. The cell was assembled as described in our previous work 25 (add yours) as follows: the anode was attached to anode current collector, this assembly was placed at one side of the cell. Then, at the other side, the cathode was sandwiched between the cation exchange membrane and cathode current collector. The assembled MFC was sterilised by UV (CHC LAB Co., Ltd, Korea), while the culture media and the glassware adopted in this study were autoclaved at 121 °C for 20 min before operation with a steam steriliser (Autoclave, AC-60, Hanyang Scientific Equipment Co. Ltd, Korea).

To evaluate the performance, the RMFC was driven with pure rainwater as anolyte, and with mixture of rainwater and Nutrient broth media (1:1 ratio v/v). Furthermore, the aerobic and anaerobic conditions were examined, the bacteria was cultivated under anaerobic conditions by blocking the access of oxygen 26. For the aerobic conditions, the MFC was in direct contact to the ambient air through an open access of 10 mm 27. Moreover, the temperature effect was investigated, where the RMFC was operated at the ambient temperature (with lowest temperature degree around 0 to 5 oC, from 10 to 31 of January) and at 30 oC for comparison purposes.

2.4. Electrochemical Characterization

The MFCs electrochemical performances were evaluated using HA-151A potentiostat (HA-151A POTENTIO STAT/GALVANOSTAT, Japan). The cyclic voltammetry analysis (CV) was carried out on the assembled cell using a three-electrode setup in which the anode was assigned as a working electrode (WE), and the cathode and Ag/AgCl as a counter (CE) and reference (RE) electrodes, respectively. The used potential window was from 1.0 to -0.4 V at a scan rate of 1 mVs-1. The open circuit potential (OCP) values were recorded using GL220 midi-logger.  After OCP stabilisation, the cell circuit was closed and the power curves were obtained by a linear sweep voltammetry from the maximum OCP to a zero-voltage at a scan rate of 1 mVs-1 with a two-electrode mode, where the cathode was connected as WE and the anode as both CE and RE 28, 29. The power (P) was obtained as P =IV. The morphology of the used anodes was investigated by the scanning electron microscopy (SEM Hitachi S-7400, Japan). The electrodes were dried at room temperature (20 ± 2 oC) in sterilised Petri dishes and then used for the analysis.

2.5. Microbial Community Analyses

At the end of the experiment (after 22 days of operation), the RMFC was dismantled, and the anode was abstracted from the MFC in a sterilized laminar cabinet (CHC LAB Co., Ltd, Korea). The biofilm was sampled from both anode surface and anolyte, and then cultured in Nutrient Agar media (Sigma-Aldrich, Inc.). The bacterial growth was subsequent isolated in a separate culture using the same media according to the growth morphology characteristic. Finally, the strains were grown overnight and processed to extract the total genomic DNA. The total genomic DNA was isolated using DNA extraction kit protocol (QIAGene, Hilden, Germany) with a little modifications as described previously 30. Briefly, a 1 ml from all isolated cells were harvested in a 1.5 ml microcentrifuge tube, after centrifuging for 5 min at 8000 g. Then, the bacterial pellet re-suspended in 180 µl buffer T1 and mixing well, following by adding 25 µl proteinase K and vigorously homogenized by vortex and incubated at 56 oC for 1-3 h with shaking during incubation. The samples after that were vortexed and a volume of 200 µl of buffer B3 was added and totally mixed by vigorous vortexing. A 210 µl of ethanol (100%) was added to the samples and vigorously mixed with vortex and then each sample was applied to the Nucleospin Tissue Column and placed into a collection tube and centrifuged for 1 min at 11000 g. After discarding the flow-through from the previous step, a 500 µl of buffer BW was added following by centrifuge for 1 min at 11000 g. The Nucleospin Tissue Columns were placed in a new collection tube, 600-µl buffer B5 were added and centrifuged for 1 min at 11000 g. The samples were centrifuged again for 1 min at 11000 g. After that, the Nucleospin Tissue Columns were shifted to a 1.5 ml microcentrifuge tube and incubated for 1 min at room temperature after a 100 µl of prewarmed buffer BE (70 oC) was added. Finally, the previous tubes were centrifuged for 1 min at 11000 g and the flow-through was kept at -4 oC for the following analysis. The DNA purity ratio was determined using 260 and 280 nm absorbance ratio. The 16S rRNA gene was amplified using forward primer 27f (5′-AGAGTTTGATCCTGGCTCAG-3′) and reverse primer 1492r (5′-GGTTACCTTGTTACGACTT-3′) by polymerase chain reaction (PCR) as previously described 31. Besides that, the total bacterial count was carried out using Nutrient agar media (Sigma-Aldrich, Inc.) 32 before as well as after utilization in the MFC for a comparison purpose.

Origin program should be used to access the raw data.