Biology Reference
In-Depth Information
To increase the current, we used 100 mM ferricyanide at the cathode as the electrolyte
due to the higher concentration of the electron acceptor in the cathode solution [4].
Under this condition,
S. oneidensis
MR-1 produced a current density of 5.54 ± 0.43
mA/m
2
(mean ± SE, n = 5) 100 min after loading microbes into the device (Figure 3B).
Over time, the current density gradually dropped, but remained higher than that of the
negative control wells containing medium only. The electricity generation profi les of
spatially distinct wells measured at multiple time points during 15 hr of operation dif-
fered by less than 8%, demonstrating that individual wells of the MFC array displayed
comparable performance characteristics.
Performances of the same microbial culture were also compared in different MFC
arrays. Two arrays with the same confi guration were tested with the same microbial
culture (OD
600
= 0.8) and showed current density of 2.94 ± 0.16 mA/m
2
(mean ± SE,
n = 8) (Figure 3C). Thus the MFC arrays showed chip-to-chip variances of less than
5.4%. Performances of the same chip with microbial cultures of the same concentra-
tion (OD
600
= 0.8) prepared on different days were also examined (Figure 3D). The
current densities on two different days were 3.05 ± 0.18 mA/m
2
(mean ± SE, n = 16),
showing a 5.9% variance. Therefore, the MFC array provided a platform for reproduc-
ible experimentation.
Encouraged by the performance and reproducibility of the MFC array, we exam-
ined whether the device could be employed to quickly screen environmental microbes
for individual isolates that display enhanced electrochemical activities. A schemat-
ic representation of the screening process is shown in Figure 4A. We pre-screened
~12,000 microbes derived from environmental (water and soil) samples on solid me-
dium containing Reaction Black 5, an azo dye that indicates electrochemically active
organisms (Figure 4B,C) [39]. The 16S rDNA sequencing analysis of 26 hits obtained
from the pre-screening plates revealed that the majority of the isolates (n = 10) were
members of the Bacilli and γ-proteobacteria classes (Table 1). We then exploited the
MFC arrays to characterize the electrochemical activities of several isolates. One iso-
late 7Ca reproducibly showed 266% higher power output than the
S. oneidensis
MR-1
reference strain in both the primary screening (Figure 4E) and the secondary confi r-
mation with more replicates in the MFC arrays (Figure 4F). Phylogenetic analysis
demonstrated that 7Ca was most closely related to
Shewanella putrefaciens
IR-1 (98%
sequence similarity) and
Shewanella
sp. MR-7 (98% sequence similarity) (Figure
4D). The high power generation capability of 7Ca was further validated in 24-hr trials
in a conventional H-type MFC system (Figure 4G) [6]. In our specifi c conventional
MFC confi guration, the maximum current density of 7Ca was 169.00 ± 10.60 mA/m
2
,
which was 217% higher than the current density (78.00 ± 7.30 mA/m
2
) generated by
the
S. oneidensis
MR-1 control. The maximum power density of 7Ca was also 233%
higher than this reference strain. Although we used gold as the anode material in the
MFC array and carbon cloth as the anode material in the H-type MFC system, the
power density increases of 266% in the MFC array and 233% increase in the H-type
MFC system showed that fi ndings in our MFC array system can be translated to larger
scale conventional systems. Thus, insights garnered using gold anodes in miniaturized
MFCs can be transferred to conventional MFC formats using carbon anodes.
