Preliminary verification of the electricity generation performance in coated-MFC and uncoated-MFC
Fig. 1a showed that 5 cycles of K.quasipneumoniae sp. 203-inoculated MFCs were observed, a total of 5 cycles, each cycle lasted approx 120 hours. Due to the consumption of the substrate sodium citrate in the anolyte, the output voltage will start to decline, and it is necessary to replace the anolyte. The average maximum output voltage in coated-MFC and uncoated-MFC were detected with 621 mV and 327 mV, respectively. Although the average maximum output voltage of coated-MFC was 300 mV lower than that of uncoated-MFC, it can be seen from Fig. 1b that coated-MFC still showed a certain electrochemical performance, coated-MFC reached 40.26 mW / m2 at a current density of 770.9 mA / m2 and the uncoated-MFC reached 90.69 mW / m2 at a current density of 1224.49 mA / m2. The difference in the electricity generation performance between coated-MFC and uncoated-MFC was probably due to the microfiltration membrane covered in anode, which inhibited the growth of EAB on the anode. Therefore, we speculated that K.quasipneumoniae sp. 203 can also perform EET through the biofilm mechanism. C. Yuvraj et al. also indicated that K.quasipneumoniae can directly transfer electrons to the anode without any external mediator, and the increase in electrochemical performance is directly proportional to the electroactive biofilm formed on the electrode surface[18]. Shewanella and Geobacter are considered model exogenous electrons, and are known to be able to perform direct extracellular electron transitions through outer membrane redox proteins[19]. According to previous studies, the common feature of EAB with this EET ability is the presence of many polyheme c- type cytochromes ( MH- cytC ) in their genome[20].
Fig. 1c presented that the internal resistance of the MFCs, the semicircles in the high frequency region and the straight lines in the low frequency region represent the charge transfer resistance ( Rct ) and diffusion resistance ( W1) of the MFCs, respectively. The Rct and W1 of coated-MFC were 78.20 Ω and 55.10 Ω, the Rct and W1 of uncoated-MFC were 6.84 Ω and 30.20 Ω, the Rct and W1 of coated-MFC were both higher than uncoated-MFC. After the operation of MFCs, the anode biofilm continued to grew and generate electrons until a complete biofilm was formed on the anode surface, and the electricity generation performance of the system reached a stable state. For EAB, the phospholipid bilayer structure of the cell membrane acts as a capacitor, and the electron shuttle ( or redox mediator ) generated endogenously on the cell membrane acts as an electrochemical active site. The surface of the coated-MFC anode was covered with a microfiltration membrane. It was difficult for microbial cells to adhere to the smooth surface of the microfiltration membrane, and the electrons generated by it cannot be transmitted to the anode surface over a long distance without electron mediators[7]. It is generally believed that, within a certain period of time, the thickness of the anode biofilm and the content of the redox mediator are negatively related to the internal resistance[21]. MFCs performance metrics were summarized in Table 1 and the data indicated that the output power of MFCs is inversely proportional to the internal resistance.
Fig. 1d showed the electron transfer mechanism and catalytic efficiency during the stable stage of 1st and 3rd operation of the MFCs. The CV curve revealed that no significant redox peaks were observed in 1st cycle of coated-MFC and uncoated-MFC. After 3rd cycle of operation, the existence of a reversible redox process in both MFCs, but the peak of coated-MFC was significantly lower than that of uncoated-MFC, and more than one pair of redox peaks of uncoated-MFC can be observed. In 3rd cycle of coated-MFC, a redox peak that was not detected in the 1st operation was observed. Therefore, it can be inferred that the electrochemical activity of coated-MFC due to the self-excreted electron mediators lead to the mechanism of electron transfer from bacterium to the anode. A similar result was also obtained from the analysis by Deng et al[8]. Interestingly, uncoated-MFC showed more intense redox activity than coated-MFC, indicating that uncoated-MFC can also transfer electrons through other EET mechanisms in addition to generating electron mediators. M. Amirul Islam et al. also observed a similar redox peak from the MFCs inoculated with Klebsiella.variicola[7]. Consequently, the appearance of more intense redox peaks indicated that the mature and effective biofilm was formed on the anode surface, which shortened the diffusion distance of extracellular electron transfer between EAB and the anode.
Table 1 Comparison of electricity generation performance and internal resistance in coated-MFC and uncoated-MFC
|
Power density
( mW / m2 )
|
Current density
( mA / m2 )
|
Internal resistance ( Ω )
|
Rohm
|
Rct
|
W1
|
CPE1
|
CPE2
|
coated- MFC
|
40.26
|
770.97
|
23.54
|
78.20
|
55.10
|
1.76
|
2.74
|
uncoated-MFC
|
90.69
|
1224.49
|
7.41
|
6.84
|
30.20
|
7.20
|
8.61
|
Effect of microfiltration membrane on the growth of cells
We were interested in whether the microfiltration membrane affects cell growth, so we measured the biomass of the anolyte and anode. The biomass of the microbial population can be indirectly calculated by measuring the protein content[22]. During the electricity generation process of MFCs, the amounts of EAB in anode biofilm and anolyte suspension in coated-MFC and uncoated-MFC as assessed by the protein contents were compared ( Fig. 2 ). We found that the protein content is basically the same in coated-MFC and uncoated-MFC, and the presence of microfiltration membrane has little effect on the anolyte suspension biomass ( Fig. 2a ). Moreover, we found that the biomass of uncoated-MFC anode biofilm was much higher than that of the coated-MFC, and the average protein content by more than 5 times ( Fig. 2b ). The existence of the microfiltration membrane only affected the biomass of the anode biofilm, and had little effect on the biomass in the anolyte. Therefore, it can be considered that the difference in the electricity generation performance between coated-MFC and uncoated-MFC was due to the extremely little biomass of the anode biofilm.
Effect of anode biofilm on electricity generation performance of MFCs
Fig. 3 revealed that the formation of the biofilm on anode electrode surface by SEM. Results clear showed that almost no EAB were attached to the carbon paper surface in coated-MFC ( Fig. 3a-c ) . This further verified the existing results, the extremely low protein content of coated-MFC anode biofilm ( Fig. 2b ) . However, the adsorption capacity of rod-shaped EAB at the anode increased with time in uncoated-MFC ( Fig. 3d-f ) . It can be seen that an incomplete biofilm is formed on the electrode surface due to the adsorption of EAB in the 1st operation of uncoated-MFC ( Fig. 3d ) . It was not until the third operation of uncoated-MFC that we observed that the microorganisms attached to the surface of the carbon paper began to secrete an enveloping matrix consisting mainly of polysaccharides and proteins, known as EPS. According to the research by Wu et al. , EPS can accumulate group-sensing effect signaling molecules, extracellular enzymes, and bacterial secondary metabolites, providing a place for microbial to exchange information[23]. Kim et al. tested the impedance of the biofilm formation process in the early adhesion stage of P. aeruginosa PAO1 and found that early adhesion of microorganisms on the anode would lead to a reduction in electrical resistance[24]. In combination with the above electrochemical performance results, we considered that microorganisms clusters form complete biofilms with metabolic activity, thereby exhibiting higher power generation performance ( Fig. 3e and Fig. 1a ) . However, in the 5th cycle, several layers of biofilm were adsorbed on the anode of uncoated-MFC which decreased the electrochemical performance ( Fig. 3f ) . This result was consistent with the result of protein content of coated-MFC anode biofilm ( Fig. 2b ). In the 3rd cycle of uncoated-MFC, the output voltage ( 621 mV ) and the protein content anode biofilm ( 2.00 gprot / L) reached their peak values at the same time ( Fig. 1a ). And then the protein content decreased with the output voltage - time. This may be related to the EAB proliferation rate of anode biofilm.
In addition, the changes in biofilm viability were examined over time using fluorescent staining to distinguish live versus dead cells ( Fig. 4 and Fig. S1 ). As previously described, the presence of microfiltration membrane maked it almost impossible to observe the presence of living and dead cells ( Fig. S1a - c ). The fluorescence intensity of dead cells was much less than that of living cells ( Fig. 4 ) . On the contrary, as the biofilm grew, cells existed in two states ( live and dead ) in uncoated-MFC ( Fig. S1d - f ). When the maximum output voltage reached 621 mV in the 3rd cycle, we observed that the entire biofilm of K.quasipneumoniae sp.203 was alive, with very few dead cells ( Fig.1a and Fig. S1e ). As the biofilm grew, the fluorescence intensity of red dead cells increased from 30.49 to 64.95, the increased fluorescence intensity of dead cells can help explain that the output voltage gradually decreases after reaching the peak ( Fig. 4, Fig. S1e and Fig. S1f ). Although we did not analyze the location and spatial structure of live and dead cells in biofilm, previous studies suggested that living cells can only exist in the outer layer of thick biofilm, which may be due to the availability of anode electrolyte substrates[25, 26]. Furthermore, the accumulation of dead cells ( more than active cells ) at the bottom of the biofilm over time will not exert redox activity. Thus this increased the intrinsic resistance of MFCs.
Effect of electron mediators on electricity generation performance of MFCs
The model strains for EET mechanism were Geobacter sulfurreducens and Shewanella oneidensis. G.sulfurreducens can also synthesize a small amount of flavin compound, but it can only binds to the outer membrane protein and cannot be released as an electron mediator, like S.oneidensis. Therefore, it is generally believed that G.sulfurreducens undergoed short-range electron transfer by direct contact with extracellular electron acceptors[5]. Combined with this study, it was found that when a new anode medium was replaced at the end of each cycle, the output voltage decreased significantly and rose slowly, even over 24 hours, which was in sharp contrast to G.sulfurreducens[27].
Therefore, compared with the CV curve of coated-MFC, we found that a pair of redox peak that appeared in the range of - 0.4 ~ 0 V in the CV curve of the supernatant did not appear in the MFCs ( Fig. 1d and Fig. 5a ). According to the report by Stefano Freguia et al, it may be due to EAB actively removing mediators in an oxidized state[28]. In order to better detect the electron mediators, we conduct CV detection on the anode supernatant. As shown in Fig. 5a, we found that more than one pair of redox peaks appeared in the anode supernatant. By comparing the standard library and consulting related references, it is inferred that the possible quinone electron mediators secreted by K.quasipneumoniae sp.203 were 2,6-di-tert-butyl-p-benzoquinone ( 2,6-DTBBQ ) , 2,6-Di-tert-butylphenol ( 2,6-DTBHQ ), 1,4-dihydroxy-2-naphthoic acid ( DHNA ) and 2-amino-3carboxy-1, 4-naphthoquinone( ACNQ ) ( Fig. 5b, Table S1 and S2 )[14, 29, 30].
Herein, combined with the existing research ideas of this experiment, MFCs constructed by K.quasipneumoniae sp. 203 was used as the research object, upon the addition of 10µM 2,6-DTBBQ, 2,6-DTBHQ and DHNA to the reactor respectively, and the results showed that the voltage of coated-MFC and uncoated-MFC increased immediately, except for the addition DHNA of MFCs ( Fig. 6a ) . Although the voltage of the coated-MFC was lower than that of the uncoated-MFC, the voltage reached the highest output voltage within 40 hour. The sharp rise in the voltage proved that the 2,6- DTBBQ and 2,6- DTBHQ can function as electron mediators in MFCs, thus facilitating the electron transfer from EAB to electrode. This speculation was similar to the results of Deng et al[31]. HPLC-MS detected a low content of ACNQ, so we did not add ACNQ to the reactors.
We also performed CV to emamine the redox state of MFCs after the addition of electron mediators ( Fig. 6b ) . The redox peaks appeared in coated-MFC and uncoated-MFC after adding 2,6-DTBBQ and 2,6-DTBHQ. Notably, the pair of redox peak at 0 ~ 0.8V were attributed to 2,6-DTBBQ, and the pair of redox peak at - 0.4 ~ 0V were assigned to 2,6-DTBHQ. This result almost corresponded to the CV curve of the supernatant ( Fig. 5a ) . In contrast, no redox peaks were found in the addition DHNA of MFCs. Combined with the time-voltage curve, we also found that after adding DHNA, the voltage of the MFC did not show a significant upward trend. Therefore, we speculate that DHNA was not a redox metabolite of K.quasipneumoniae sp.203.
Combining the time-voltage curve and CV curve, we found that 2,6-DTBBQ and 2,6-DTBHQ had high electrocatalytic activity toward the redox reaction of K.quasipneumoniae sp.203-inoculated MFCs. Moreover, the oxidation and reduction peaks at the range of - 0.4 ~ 0 V correspond to 2,6-DTBBQ, 2,6-DTBHQ, this result was similar to that of L.Z. Zeng. et al. in detecting the electrocatalytic activity of Klebsiella pneumoniae on 2,6-DTBBQ[32]. This contributes to the electron transfer between the EAB and the MFCs anode and also contribute to the power density of the MFCs.
According to previous studies, the electron mediators-mediated extracellular electron transport mechanism is a circulation mechanism. The oxidized mediators are converted into the reduced mediators after being coupled intracellularly with the reduction products on the respiratory chain[33]. Then reduced mediators are discharged out of the extracellular, and the electrons are transferred to the electrode to be oxidized. It is speculated that the electron mediator secreted by K.quasipneumoniae sp.203 can be reused. In the stable growth stage of anode EAB in MFCs, the addition of electron mediators can effectively improve the electricity generation performance. Quinones can be used as electron mediators, mainly because of their quinone group with electron transfer function. Based on the existing research of this experiment, it is considered that the transfer mechanism of the electron mediators in coated-MFC and uncoated-MFC is: the quinone compounds secreted by EAB were obtained electrons and reduced into hydroquinones, and then electrons were transferred to the anode, and hydroquinones were oxidized to benzoquinones. The main role of quinones as electron mediators is due to their quinone group with electron transfer function. The group is circulated intracellularly in three states: oxidized, semiquinone radical, reduced or hydroquinone. The transition of the electron mediator from the oxidized state to the reduced state is accomplished by quinone reductases under the action of flavin adenine dinucleotide ( FAD )[29]. And the studies by Itzel Ramos et al. has shown that the redox mediators secreted by EAB can also promote the formation of biofilms[34].