Fig. 3 (a) OCP decay relations for self-discharge at MnO2 nanosheet
electrode under a series of anodic and cathodic potential sweeps.
(b) Nyquist plots of impedance spectra at different applied potentials;
inset: equivalent circuit.
Fig. 2 (a) Transformation in Raman spectra of MnO2 under anodic
potential sweep; inset: frequency shift of the n1 and n2 stretching modes
as a function of applied potential. (b) Schematic illustration showing
the reversible hexagonal–monoclinic phase transition.
charging up to targeted potential values followed by
termination of the charging to record the decay of the open-
circuit potential (OCP). The resulting data, plotted as OCP vs.
log t in Fig. 3a, clearly reveals that OCP decays to 0.4 and 0.56 V
under anodic and cathodic charging-potential, respectively.
Interestingly, the behaviors of OCP decay seem to be confined
at two specific potentials under anodic and cathodic charging-
polarization. Even at the same charging-potential, the trend of
OCP decay displayed an obvious difference when different
directions of polarization were applied. The two confined
potentials were found to be located at the potential edge of
the individual phases of hexagonal and monoclinic MnO2. The
confinement of phase retention can thus be understood on this
basis because activated energy is still needed to overcome the
hexagonal–monoclinic phase transition during OCP decay.
Hence, the phase transformation plays a significant role in
the self-discharge process of ECs.
The n1 Raman band at 625–650 cmÀ1 can be attributed to the
symmetric stretching vibration (Mn–O) of the MnO6 groups,
and the n2 band at 575–585 cmÀ1 is attributed to the (Mn–O)
stretching in the basal plane of the MnO6 sheet.15 The n1 and
n2 Raman bands also confirm the birnessite-type MnO2 with
layered structure of the samples.
More detailed Raman spectra evolution is illustrated in
Fig. 2a, showing clear features during the anodic potential
sweep. At 0 V, only a broad n2 band at 578 cmÀ1 with a
shoulder n1 band at 622 cmÀ1, and a weak band at 480 cmÀ1
were observed. As the potential increased to 1 V, all these three
Raman bands became sharper, whereas the n2 band slightly
shifted to low wavenumber and grew in intensity. Meanwhile,
with increasing potential from 0.5 to 1 V, two new bands
gradually appeared at 650 and 500 cmÀ1, and became stronger in
intensity. The inset of Fig. 2a shows the frequency shift of the two
high wavenumber bands, n1 and n2, as a function of the applied
potential. The variation in n1 displayed clearly a stepped
frequency shift and the coexistence of two bands at 0.5 V. This
indicates that two different structural phases appeared at
different sides of the potential and coexisted in the potential
range of 0.4–0.6 V. During the anodic potential sweep, the step
increase of the n1 stretching mode resulted from the increase of
distance between two successive layers of birnessite-type MnO2
due to the removal of Na+ ions from the interlayer, which are
replaced by water molecules.15 As for the shift of the n2 band,
Na+ ion incorporation during the cathodic potential sweep
accompanied reduction of Mn4+ and shorter Mn–O chemical
bonds of the MnO6 layer, which caused local lattice distortion by
a shift of the distance of one MnO6 layer with regard to the next.
Hence, the shrinkage of interlayer spacing and local lattice
distortion of the MnO6 layer upon Na+ ion intercalation can
be deduced, corroborating with the structure transition from the
hexagonal phase to the monoclinic phase, as shown in Fig. 2b.
Furthermore, the dramatically enhanced intensity of the vibration
mode with increasing potential indicated preferential orientation
corresponding to the alignment of MnO6 layers parallel to the ab
plane, and a well-ordered hexagonal phase.
Since the electron storage in MnO2 nanosheets is associated
with the insertion of Na+ between the layers, a probe into the
Na+ intercalation process using the impedance technique is
essential. Fig. 3b shows Nyquist plots of impedance spectra
measured at various potential values and were analyzed by
using the CNLS fitting method based on an equivalent
circuit,16 as shown in the inset of Fig. 3b. Those fitting values
of Rct, RW, td and Cps are listed in Table 1. It is worth noting
that as the applied potential decreased, the values of Rct and
RW increased, but the value of td decreased. Clearly, as
shrinkage of interlayer spacing occurs, Na+ diffusion
obviously becomes difficult. At the same time, the available
sites for Na+ within the interlayer space become less with
applied potential. This also hinders the diffusion of subsequent
Na+ and thus results in further increase of the value of Rct and
RW. Meanwhile, the gradually increased resistances cause the
decrease of td, indicating a shortening Na+ diffusion path in
MnO2 nanosheets.
To our knowledge, this is the first observation of the
reversible hexagonal–monoclinic phase transformation in bir-
nessite-type MnO2 electrode upon potential cycling in aqueous
Na2SO4 electrolyte. This structural evolution during Na+
insertion results in a shift of the distance of the MnO6 layer
to the next and decrease in interlayer spacing between the
sheets. Meanwhile, the Na+ insertion would balance the
valence of reduced Mn accompanying the storage of electrons.
Furthermore, this structural transformation not only constrains
Self-discharge experiments of the MnO2 nanosheet electrode
were conducted in 0.1 M Na2SO4 by applying constant-current
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 1252–1254 1253
Hsu, Yu-Kuei
