Layered manganates from soft-templates: Preparation, characterization and enhanced dye demethylation capabilities

Article abstract of DOI:10.1039/c2jm32583g

A soft-template method for the synthesis of layered manganate (K _xMnO₂, birnessites) compounds with a relatively good control at the nanoscale regime on the number of octahedral layers (thickness) is described. The soft-template method is based on the stabilization of reverse micelles at moderate temperatures and the optimization of parameters to modulate the size of the templates while still preserving their stability during the formation of layered manganates. Degradation experiments of a model system (methylene blue) using the birnessites here reported have shown efficient demethylation in the absence of oxidants for aqueous suspensions made of non-dried birnessites containing the fewest number of layers. The observed strong thickness and aggregation dependence (i.e. dependence on exposed surfaces) for demethylation explains why particulate birnessites have only been considered for dye sequestration and not as an efficient demethylation agent in the absence of oxidants. Finally, when compared to electrodeposited thin films, the higher loading capabilities of the colloidal suspensions here prepared have made it possible to demethylate larger amounts of methylene blue at a similar time.

Full text of DOI:10.1039/c2jm32583g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 17718
www.rsc.org/materials
PAPER
Layered manganates from soft-templates: preparation, characterization and
enhanced dye demethylation capabilities
*
Pedro Tartaj
Received 24th April 2012, Accepted 4th July 2012
DOI: 10.1039/c2jm32583g
A soft-template method for the synthesis of layered manganate (KxMnO2, birnessites) compounds with
a relatively good control at the nanoscale regime on the number of octahedral layers (thickness) is
described. The soft-template method is based on the stabilization of reverse micelles at moderate
temperatures and the optimization of parameters to modulate the size of the templates while still
preserving their stability during the formation of layered manganates. Degradation experiments of a
model system (methylene blue) using the birnessites here reported have shown efficient demethylation
in the absence of oxidants for aqueous suspensions made of non-dried birnessites containing the fewest
number of layers. The observed strong thickness and aggregation dependence (i.e. dependence on
exposed surfaces) for demethylation explains why particulate birnessites have only been considered for
dye sequestration and not as an efficient demethylation agent in the absence of oxidants. Finally, when
compared to electrodeposited thin films, the higher loading capabilities of the colloidal suspensions
here prepared have made it possible to demethylate larger amounts of methylene blue at a similar time.
shifts make N-demethylation easily to follow by UV/Vis spec-
1. Introduction
trophotometry. Therefore, from a practical point of view, the use
of MB assures relatively rapid checking for demethylation and
reliable comparisons between materials (i.e. demethylation of
MB can be considered a model system). MB in the presence of
manganese oxides/hydroxides is usually degraded in combina-
tion with oxidants (H2O2, tert-butyl hydroperoxide, etc.).6–9
Recent studies on electrodeposited layered manganate birnessite
(KxMnO2) thin films, however, have shown that MB demethy-
lation can take place even in the absence of oxidants at relatively
short times (3 h).7 Showing the same effect in colloidal birnessites
could open the possibility of moderately and highly active
material loadings for demethylation of MB. Therefore, we
understand that implementing a method to synthesize particulate
birnessites with a relatively good control at the nanoscale regime
on the number of octahedral layers (thickness) together with a
simple, fast, yet robust method of checking degradation capa-
bilities (methylene blue demethylation) has practical and
fundamental interest. Demethylation of a cationic dye is of
practical interest while the relatively good control at the nano-
scale regime in a layered compound can shed some light on the
different factors that cause the lower performance of particulate
materials when compared to thin films.
The redox properties of layered transition-metal oxides have
been routinely used for environmental and energy applica-
tions.1–3 These oxides, however, when operating at highly or even
moderately active material loads (say for example when pro-
cessed in particulate form) have low capabilities associated with
diffusion/contact problems. Indeed, layered transition-metal
oxides when processed as thin films can approach theoretical
capabilities especially when redox processes are involved,4,5 but
thin films’ applicability is limited by low active material loadings.
Hereafter we report a soft-template method based on the Igepal/
organic solvent/water system for the controlled synthesis of
layered manganates (KxMnO2, birnessite). The method allows a
good control at the nanoscale regime on the number of octahe-
dral layers (thickness) without altering chemical and basal
spacing. We show, for a model system (methylene blue), that
aqueous suspensions containing the birnessites here prepared can
demethylate larger amounts of this colorant compared to the
homologous electrodeposited thin films.
Among organic pollutants, organic dyes used in a broad
variety of applications represent a real environmental problem
and particularly methylene blue (MB) is the most important
basic cationic dye. There is consensus in the literature that
N-demethylation of auxochromic alkylamine groups is involved
in the first stages of MB degradation.6,7 Discoloration and blue
There are many scientific reports that describe synthetic routes
to produce K-birnessites.1,10–14 Precipitation routes usually
involve not only the oxidation of aqueous Mn2+ cations but also
the oxidation of amorphous solid precursors via the redox
ꢀ
reaction between Mn2+ and MnO4
.
Sol–gel synthesis
1,16–18
Instituto de Ciencia de Materiales de Madrid (CSIC), Campus
Universitario de Cantoblanco28049, Madrid, Spain. E-mail: ptartaj@
icmm.csic.es
usually involves various sugars and other organic polyalcohols as
well as further heating to develop good crystallinity and remove
17718 | J. Mater. Chem., 2012, 22, 17718–17723
This journal is ª The Royal Society of Chemistry 2012

View Article Online
the organic additives.15 Recently, K-birnessites were prepared by
dispersing monosheets in solutions containing K+ cations.13,19 To
our knowledge none of these methods are able to produce a
satisfactory control on the number of manganate octahedral
layers together with a simultaneous control on chemical
composition and basal spacing for colloidal K-birnessites. Here
it is worth noting that we are not interested in methods that
involve the preparation of birnessite in the presence of large
cations. MB is a cationic dye, and there are numerous pieces of
evidence that surface mechanisms are involved in the oxidative
degradation of organic compounds by manganese oxides.16
Large cations can compete with MB for the adsorption sites
lowering demethylation efficiencies (as better described in the
Results and discussion section, lower efficiencies were detected
by using a cationic polyelectrolyte).
surfactants (CO-210; CO-520; CA-520; and CO-720), Igepal
CO-520 at a 0.2 M concentration was used. Heptane and decane
are known to favor lamellar-like structures; however, Mn salt
solubility via reverse micelle formation in these pure solvents or
in cyclohexane–decane mixtures was very low, and therefore a
mixture of cyclohexane–heptane was used. The aqueous volume
was set to 3% as the temperature range for microemulsion
formation (single phases) and solubility of Mn salts were maxi-
mized. For a 3 vol% of aqueous component, this temperature
ꢁ
range extends from 60 to 35 C for the highest amount of Mn
salts that can be dissolved (ꢂ0.07 M which is equivalent to 2.3 M
in the aqueous component). For higher aqueous contents the
temperature range for stability of microemulsions is reduced
because high temperature destabilization takes place. At lower
aqueous contents the solubility of Mn salts is low.
As mentioned above a soft-template method based on the
Igepal/organic solvent/water system has been selected in this
work for the controlled synthesis of colloidal K-birnessites.
Microemulsions in the system Igepal/cyclohexane/water have
been routinely used to produce isotropic structures.17 Less
frequently, however, is to exploit the fact that Igepal can form
lamellar-like structures especially in some solvents such as
heptane and decane.18 Furthermore, destabilization (say for
example by the addition of an excess of water or by lowering the
temperature) can go through the formation of interconnected
lamellar-like structures.19 Soft lamellar structures are flexible and
can easily accommodate the growth of isotropic structures.19,20
However, we are of the opinion that they will adapt better if the
material itself has a tendency to grow in a lamellar structure as is
the case of K-birnessites.
2.3. Soft-template assisted synthesis of birnessites
We find it mandatory to remark that for security and technical
issues (bubbling of H2O2), we have carried out all the experiments
in 250 or 500 mL closed beakers (50/100 mL of solution, respec-
tively) and H2O2 concentration was restricted to 0.12 M for a
KOH concentration of 0.47 M (concentrations referred to the total
content). Hydrogen peroxide in acid medium can even reduce
permanganate (MnO4ꢀ) to Mn2+ cations (i.e. the acidity
provided by the H2O2 itself makes solutions containing Mn2+
salts stable). However, H2O2 in basic medium (i.e. after addition
of KOH) decomposes into water and gaseous oxygen (bubbling).
In a typical experiment 50/100 mL (100 mL for the lowest Mn
concentrations) of a mixture containing the organic solvent
(cyclohexane–heptane, 50 vol%), Igepal Co-520 (0.2 M), the
aqueous component (3 vol%) and the reactants were prepared at
room temperature under strong magnetic stirring in closed
2. Experimental section
ꢁ
2.1. Chemicals
beakers. Then to assure equilibrium at 45 C, the Mn salt was
ꢁ
dissolved at 55 C/15 min and the temperature was lowered to
All chemical reactants (purity, >99%) were purchased from
Sigma Aldrich. Mn(NO3)2$4H2O, H2O2 (50%), NH4S2O7, and
KOH were used as precursors. Igepal CO-520 was used as a
surfactant. Cyclohexane/heptane/decane dehydrated with
zeolites were used as the continuous oil phase. Methylene blue,
KH2PO4 and NaH2PO4 were used for demethylation studies.
45 ꢁC in 30 min. In order to avoid Mn3O4 formation in the
samples with the higher Mn concentrations (0.05 and 0.065 M),
NH4S2O7 was added once the microemulsion reached 45 ꢁC.
NH4S2O7 dissolves during the 30 min that the microemulsion
remains at this temperature before KOH addition. After 30 min
ꢁ
at 45 C, the appropriate volume of a KOH 14 M solution was
rapidly added (note that the container was kept open during
addition of the KOH solution for security reasons associated with
bubbling). After 24 h the suspension was centrifuged at 1000 g for
10 minutes to discard the sediment. The supernatant was desta-
bilized with isopropanol, centrifuged and washed (stirring 30
min) several times (twice with EtOH/water and 3 times with
water) and redispersed in water for storage. An aliquot of these
suspensions was centrifuged and dried for chemical analyses or
for some methylene blue degradation studies involving dried
samples.
2.2. Soft-template optimization
The optimization of conditions to obtain adequate soft-
templates was carried out by temperature-dependent dynamic
light scattering experiments (Nanosizer ZS, Malvern Instru-
ments). Mixing of the components was carried out at room
temperature under magnetic stirring in closed beakers. Then the
ꢁ
mixtures were brought to 60 C to assure the dissolution of the
inorganic components. An aliquot was transferred for tempera-
ture-dependent dynamic light scattering experiments (the
temperature of the cell compartment was 60 ꢁC). Changes in
hydrodynamic sizes with temperature were thus registered on
cooling. For some conditions the destabilization of micro-
2.4. Sample characterization
ꢁ
emulsions was observed at 60 C. If so, the same protocol was
Temperature-dependent dynamic light scattering (DLS) experi-
ments were carried out in a Nanosizer ZS (Malvern Instruments).
X-Ray diffraction (XRD) patterns were collected from 5 to 50ꢁ
(2q) by using a Bruker D8 Advance instrument with CuKa
applied lowering the maximum temperature in 5 ꢁC steps.
Adequate conditions were found if the organic solvent contained
cyclohexane–heptane mixtures (50 vol%, checked in a range that
changes in a 25% step) and among four different Igepal
radiation (l
¼
0.15406 nm) and
a
SOLX detector
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 17718–17723 | 17719

View Article Online
(postdiscriminator of fluorescence) operating at 40 kV and
30 mA. The crystal domain size was determined from the X-ray
profiles following the Scherrer equation by using the DIFRAC-
PLUS EVA software (BRUKER AXS). For estimation of the
full width at half maximum the patterns were registered 5
different times. Basal spacing was determined using the same
software and Si powder was used as an internal standard to
correct the position of the peaks. For crystal size and basal
spacing determination, the XRD patterns were registered with
20 s of accumulation time in the regions where the birnessite
(001) peaks appear. The morphology of the obtained samples
was studied with a JEOL transmission electron microscope
working at 200 keV. Selective area diffraction patterns were
taken using the same instrument. The chemical composition of
samples was determined by ICP (Perkin Elmer Optima 2100 DV)
and thermal analysis (Seiko EXSTAR 6300). For MB degrada-
tion studies the UV/Vis spectra were registered in a Perkin Elmer
Lambda 35.
for micelles prepared in the absence of Mn (Fig. 1, Table 1). This
result suggests that for Mn concentrations below about 0.025 M
the micelle size reflects the contribution of both the aqueous
content itself and the contribution of the dissolved NH4S2O7 salt.
Based on the characteristics of the soft-templates at 45 ꢁC
(Fig. 1 and Table 1) our initial assumptions to try to attain a
relatively good control at the nanoscale regime on the number of
manganate octahedral layers (thickness) were: (a) thickness
primarily increases with micelle size except for the lowest Mn
concentrations (templating effect) – strictly speaking with the
aqueous micellar cavity size, which we define as the hydrody-
namic size minus twice the surfactant thickness (1–1.5 nm is
about the thickness of the surfactant layer)21 and (b) large
aqueous micellar cavities can produce thicker birnessites but also
thinner ones. This hypothesis suggests that the reactant
concentrations also modulate the birnessite thickness. Based on
nucleation theory, we assumed that higher oxidation rates (high
oxidant/Mn ratios) could lead to a reduction in thickness asso-
ciated with faster reactions. Addition of KOH (see Experimental
section for technical details) under these guidelines made it
possible to modulate the birnessite thickness from 3.0 to 8.1 nm
(Table 1, Fig. 2A) without altering the chemical composition
(K0.3MnO2$0.6H2O) and basal spacing (0.72–0.73 nm).
3. Results and discussion
3.1. Preparation and characterization of samples
Fig. 1 shows the evolution of micelle size as a function of Mn
concentration and temperature for optimal conditions (see
Experimental section for technical details). For K-birnessite
synthesis, setting a constant temperature for the reaction
increases the probability to produce these materials with similar
chemical and basal spacing (the range of thermodynamic
stability of hydrated forms such as birnessites is strongly
dependent on temperature). Besides, we are dealing with very
sensitive systems. Taking into consideration these two factors we
set the working temperature at 45 ꢁC. At this temperature
significant differences in the micelle size are observed (thicknesses
are thus prone to change) while still working in the middle of the
The colloidal nature (particle sizes with an average of around
150 nm) of the non-dried birnessites was assessed by dynamic
light scattering (Fig. 2B). BET (Brunauer–Emmett–Teller)
surface area determination and scanning electron microscopy
characterization were discarded as drying of the material may
alter the surface area and microstructure (referred to as the
assembly of primary units in different morphologies). In fact, as
we better describe below demethylation did not take place if
experiments were made in dried samples, which clearly indicates
aggregation after drying (i.e. less active surface area exposed). In
any case, the transmission electron microscopy (TEM) in dried
samples clearly shows us the presence of a layered compound in
the form of a birnessite aggregate. TEM shows an aggregate with
the typical corrugation associated with ultrathin sheet-like
materials, that is, with a layered compound (Fig. 2C). The
disordered arrangement of the aggregate formed by thin nano-
sheets (polycrystal-like assembly) gives a ring-type diffraction
pattern, which can be indexed within the birnessite structure
(Fig. 2D).
ꢁ
stability window (Fig. 1, Table 1). At 45 C, the size of reverse
micelles increases with the increase in Mn concentration except
for the lower Mn concentrations where such variations are not
significant (Fig. 1, Table 1). In fact, similar values were obtained
3.2. Analysis of experimental errors
An analysis of errors is mandatory as we are dealing with rela-
tively sensitive synthesis routes. Fig. 1 for example indicates that
at 45 ꢁC errors can be larger for high Mn concentrations (Dsize/
DT increases). Errors, however, are minimized by the layered
structure of the material itself especially as thickness decreases.
We cannot produce birnessites with a non-integer number of
layers (i.e. layered structures introduce quantization). As the
thickness decreases, we come close to the minimum number of
layers that a birnessite can contain, and polydispersity (referred
to the possible presence of birnessites with different number of
layers) is low. Errors were not detected at small thicknesses, and
the maximum error was limited to around a bilayer (Table 1). As
we will better describe later, the better demethylation capabilities
are found in the samples with the smallest thicknesses (no errors,
Fig. 1 Evolutionof hydrodynamic size as a functionof Mn concentration
and temperature (cyclohexane–heptane, 50 vol%; Igepal CO-520, 0.2 M;
aqueous volume ¼ 3%; NH4S2O7 ¼ 0.02 M). The curve was registered on
cooling to assure equilibrium conditions. Note the split in the y axis to
better display size differences at the working temperature (45 ꢁC).
17720 | J. Mater. Chem., 2012, 22, 17718–17723
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Table 1 Reactant concentrations (mol Lꢀ1 referred to the total volume) used to prepare birnessites between 5 and 12 layers. Micelle and aqueous
micellar cavity hydrodynamic sizes at 45 ꢁC. As mentioned in the main text the micellar cavity is the hydrodynamic size minus twice the surfactant
thickness (2–3 nm). Crystal sizes, basal spacing and number of octahedral layers (1 + (crystal size/basal spacing)). Errors in the number of layers were
derived from five or ten experiments (if errors were detected during the first 5 experiments we carried out 5 additional experiments)
Micelle size
at 45 ꢁC (nm)
Micellar cavity
at 45 ꢁC (nm)
Crystal
size (nm)
Basal
spacing (nm)
Layers
number
[Mn]
[H2O2]
[KOH]
[NH4S2O7]
0.0125
0.025
0.025
0.05
0.12
0.12
0.053
0.053
0.044
0.47
0.47
0.47
0.35
0.42
0.02
0.02
0.02
0.025
0.025
9
9
9
17
25
6–7
6–7
6–7
14–15
22–23
3.0
3.1
4.3
5.8
8.1
0.73
0.73
0.73
0.73
0.72
5 ꢃ 0
5 ꢃ 0
7 ꢃ 0
9 ꢃ 2
12 ꢃ 2
0.065
solubility is limited to around 0.025 M (0.8 M in the aqueous
component).
3.3. Methylene blue degradation
Discoloration experiments as a function of time were carried out
to check for the thickness dependence of MB demethylation.
Unless the contrary is stated the experiments were carried out
using a KH2PO4 buffer (pH ¼ 7). We, thus, assure a constant pH
and avoid possible deintercalation associated with K+ concen-
tration gradients. Prior to carrying out absorbance experiments
that involve solution/solid separation (centrifugation/filtering), a
possible power law dependence of absorbance with wavelength
was checked in birnessite suspensions for regions where only
scattering contributes to the total attenuation coefficient. We
found an inverse-square dependence (Fig. 3) that allowed
carrying out measurements at short times (subtraction of scat-
tering) without the problems associated with centrifugation (slow
processing, and separation of adsorption and discoloration
Fig. 2 (A) X-Ray diffraction patterns show only diffraction peaks
associated with birnessite. The zoom shows the different broadness
(thickness) of the (001) reflection (20 s of accumulation time) associated
with sizes from 3.0 (the wider peak) to 8.1 nm (the thinner peak). (B)
Hydrodynamic size of birnessites when still dispersed in the organic
medium (just before isopropanol addition, see Experimental section for
technical details). (C) and (D) Typical TEM pictures and SAED patterns
of samples.
thus, assure confidence in the measurements). Furthermore,
errors were exploited to our own benefit as they allowed checking
the demethylation efficiency for some conditions while limiting
the number of experiments. This was possible as differences from
different batches caused only differences in thickness (no chem-
ical or basal spacing changes). For example, this was the case
for 9 and 11 layer samples (Mn concentration of 0.05 M, see
Table 1). We must point out that production of birnessites was
not as straightforward as we had anticipated. Specifically, we
found lower and upper limits for the formation of birnessites.
Slow oxidation (low oxidating/Mn ratios) gives Mn3O4 while
substantial bubbling marks an upper limit for the amount of
H2O2 and KOH that can be added. Replacement of H2O2 by
NH4S2O7 to avoid substantial bubbling is not possible as its
Fig. 3 (A) and (B) show the lꢀ2 scattering dependence of absorbance for
aqueous suspensions containing only birnessites and MB/birnessites
(variations of 0.03 in scattering are within the experimental error). While
Fig. 3A shows that the lꢀ2 dependence extends well beyond 664 nm
(maximum for MB absorption), Fig. 3B shows the same linear depen-
dence when MB is added (registered in the region where no absorbance of
MB is detected). Scattering at 664 nm was therefore extrapolated from
the linear region. The experiments were carried out for a solid concen-
tration of 0.8 mg mLꢀ1 and a MB concentration of 2.5 ꢄ 10ꢀ4 M. The
suspensions were diluted 1 : 35 before registering the UV/Vis spectrum.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 17718–17723 | 17721

View Article Online
effects) or filtration (we observed systematic dye retention in
filters). Subtraction was possible as the absorbance maximum for
MB (664 nm) is located at a region where the total attenuation
coefficient of birnessites is only the result of scattering.
confirmed by the blue shift associated with this process (Fig. 4B),
and the formation of the fully demethylated form of MB, which
is called thionin, for the most efficient birnessites (Fig. 4C).
Assuming that adsorption rates are high and charge compensa-
tion coming from MB is significant (we are dealing with few-layer
birnessites) there might be a possible increase in discoloration
efficiency if K+ gradients were established. Experimentally, this
theory can be checked by simply replacing a KH2PO4 by a
NaH2PO4 buffer while keeping the pH (pH ¼ 7) constant. The
result is that a significant increase in efficiency was observed
(Fig. 4A). This result brings to our attention that care must be
taken when registering and comparing discoloration efficiencies
in ultrathin birnessites. In any case, when compared to thin
films,7 the high loading capabilities of concentrated aqueous
suspensions containing birnessites make it possible to discolor
10-fold larger amounts of MB (2.5 ꢄ 10ꢀ4 vs. 3.1 ꢄ 10ꢀ5 in thin
films) at similar times (3 h). It is important to remark that we
limited the solid concentration of birnessites to values able to be
monitorized by visible spectrophotometry with adequate preci-
sion. Of course if we had prepared more concentrated suspen-
sions larger amounts of MB would have been demethylated or
the time taking to demethylate the same amount would have
been shorter.
Fig. 4A clearly shows that the discoloration efficiency strongly
increases with the decrease in thickness in a margin of 7 octa-
hedral layers. Fig. 4A also shows that drying leads to a total loss
of efficiency due to the formation of irreversible aggregates
during drying (see for example TEM in Fig. 2C). We also observe
in Fig. 4A a loss of efficiency in experiments involving a 5 layer
sample in which a cationic polyelectrolyte (PDDA, poly-
diallyldimethylammonium chloride) was added. This result
indicates that large cations cannot be present in solution if
degradation of MB is required. As mentioned in the Introduction
there exists substantial evidence that surface mechanisms are
involved in the oxidative degradation of organic compounds by
manganese oxides; therefore large cations compete with MB for
all of the adsorption sites and demethylation efficiency decreases.
The implication of this result is that methods to control the
thickness based on monosheets which are obtained via exfolia-
tion using large cations,10,22,23 followed by electrostatically driven
layer-by-layer growth using cationic polyelectrolytes,24 must not
be applied if MB demethylation is required.
Finally, and only with the aim of making some comparison
with the mechanism proposed for demethylation in the absence
of oxidants for electrodeposited thin films, some kinetic param-
eters were derived from the samples that show sufficient number
of data at the shortest times to get a statistically reliable fit and
statistically reliable differences between samples (Fig. 4D). A
pseudo-first order process was detected for the samples with
thicknesses changing from 9 to 12 layers with a pseudo-first order
constant value increasing as expected from 12 to 9 layers (0.6 ꢄ
10ꢀ3 and 3.5 ꢄ 10ꢀ3 minꢀ1). This pseudo-first order process
along with the absence of Mn cations in solution after MB
degradation, the above-mentioned strong dependence with
parameters related with available surface sites, and the formation
of thionin point to the mechanism proposed for electrodeposited
3.4. Mechanism of methylene blue degradation and comparison
with electrodeposited thin films
The strong dependence of discoloration with thickness, aggre-
gation and the presence of bulky cations clearly points to a
mechanism in which MB is only adsorbed onto the external
surface (i.e. not intercalated in the birnessites). The absence of
changes in the position of diffraction peaks further confirmed
this mechanism. This strong dependence with parameters related
to exposed surfaces such as thickness and aggregation can
explain why particulate birnessites have only been considered for
dye sequestration purposes and not as an efficient demethylating
agent in the absence of oxidants.8,9 Demethylation of MB was
Fig. 4 (A) Discoloration efficiency ((Ao ꢀ A) ꢄ 100/Ao) as a function of time and the number of layers (5 to 12 layers) for K-birnessites. Three extra 5
layer samples are included (dried, PDDA 10ꢀ5 M, and NaH2PO4 buffer of pH ¼ 7). The error bar represents the average of 5 different preparations. The
error bar was not estimated in the dried sample since milling is far from assuring homogeneity at the level of the non-dried preparations. (B) Typical
evolution of absorbance with time (birnessite the ‘‘substrate’’ is also included). (C) Normalized absorbance for methylene blue (MB) and MB after full
demethylation (thionin formation) produced by the most efficient of the birnessites (spectrum taken after centrifugation, drying and further dissolution).
(D) Pseudo-first order kinetic plot of Ln[MB] as a function of time for samples with 9 and 12 layers. All the experiments were carried out for a solid
concentration of 0.8 mg mLꢀ1 and a MB concentration of 2.5 ꢄ 10ꢀ4 M. The suspensions were diluted 1 : 35 before registering the UV/Vis spectrum.
17722 | J. Mater. Chem., 2012, 22, 17718–17723
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Fig. 5 Schematic representation of the mechanism proposed for complete demethylation of methylene blue (MB) to give thionin (TH) in the presence of
colloidal birnessites and in the absence of oxidants. As explained in the main text the mechanism is similar to that reported for electrodeposited thin
films. The strong dependence with parameters related to available surface sites (number of octahedral layers per particle) is clearly visualized.
thin films for MB demethylation.7 Briefly, adsorption of the MB
S. M. Auerbacth, K. A. Carrado and P. K. Dutta, Marcel Dekker,
New York, 2004, ch. 9, pp. 475–508.
is followed by demethylation induced by reduction of Mn(IV) to
2 R. Ma and T. Sasaki, Adv. Mater., 2010, 22, 5082.
3 A. Takagaki, C. Tagusagawa, S. Hayashi, M. Hara and K. Domen,
give Mn(III) species. These Mn(III) species can either form a
complex with the organic moieties or be reoxidized to Mn(IV) to
proceed with the demethylation process (Fig. 5).
Energy Environ. Sci., 2010, 3, 82.
4 X. Lang, A. Hirata, T. Fujita and M. Chen, Nat. Nanotechnol., 2011,
6, 232.
5 W. Yan, T. Ayvazian, J. Kim, Y. Liu, K. C. Donavan, W. Xing,
Y. Yang, J. C. Hemminger and R. M. Penner, ACS Nano, 2011, 5,
8275.
Conclusions
6 T. Sriskandakumar, N. Opembe, C.-H. Chen, A. Morey,
C. Kingondu and S. L. Suib, J. Phys. Chem. A, 2009, 113, 1523.
7 M. Zaied, S. Peulon, N. Bellakahl, B. Desmazieres and A. Chause,
Appl. Catal., B, 2011, 101, 441.
8 M. A. Al-Ghouti, Y. S. Al-Degs, M. A. M. Khraisheh, M. N. Ahmad
and S. J. Allen, J. Environ. Manage., 2009, 90, 3520.
9 L. Zhang, Y. Nie, C. Hu and X. Hu, J. Hazard. Mater., 2011, 190,
780.
10 K. Kai, Y. Yoshida, H. Kageyama, G. Saito, T. Ishigaki,
Y. Furukawa and J. Kawamata, J. Am. Chem. Soc., 2008, 130, 15938.
11 S. Stroes-Gascoyne, Adsorption behaviour of delta-manganese
dioxide in relation to its use as a resin in trace metal speciation
studies, Open access dissertations and theses. paper 3899, 1983.
12 D.-Y. Sung, I. Y. Kim, T. W. Kim, M.-S. Song and S.-J. Hwang, J.
Phys. Chem. C, 2011, 115, 13171.
13 O. Prieto, M. Del Arco and V. Rives, J. Mater. Sci., 2003, 38,
2815.
14 M.-S. Song, K. M. Lee, Y. R. Lee, I. Y. Kim, T. W. Kim,
J. L. Gunjakar and S.-J. Hwang, J. Phys. Chem. C, 2010, 114,
22134.
15 S. Ching, J. A. Landrigan and M. L. Jorgensen, Chem. Mater., 1995,
7, 1604.
16 A. T. Stone and J. J. Morgan, Environ. Sci. Technol., 1984, 18, 617.
17 D. G. Shchukin and G. B. Sukhorukov, Adv. Mater., 2004, 16, 671.
18 J. C. Ravey, M. Buzier and C. Picot, J. Colloid Interface Sci., 1984, 97,
9.
We have described a soft-template method for the synthesis of
layered manganate (KxMnO2, birnessites) compounds. The
method allows a relatively good control at the nanoscale regime
on the number of octahedral layers (thickness) in a layered
compound of interest in energy and environmental applications.
Visual inspection (Fig. 5) and spectroscopic monitoring (Fig. 4A)
make clear that the birnessites here obtained can efficiently
demethylate MB and that there is a strong dependence of this
demethylation with thickness and aggregation (i.e. with exposed
surfaces). This strong dependence among other things explains
why particulate birnessites have only been considered for dye
sequestration purposes and not as an efficient discoloration
agent in the absence of oxidants. It is quite interesting to remark
that experimental errors in this method are minimized (especially
in the most efficient birnessites) by the layered structure of the
material. Basically, layered structures cannot contain a non-
integer number of layers; in other words, layered structures
introduce quantization effects.
Acknowledgements
19 C.-L. Chang and H. S. Fogler, Langmuir, 1997, 13, 3295.
20 P. Tartaj and C. J. Serna, J. Am. Chem. Soc., 2003, 125, 15754.
21 J. L. Lemyre, S. Lamarre, A. Beaupre and A. M. Ritcey, Langmuir,
2010, 26, 10524.
The financial support from the Ministerio of Economia and
Competitividad (MAT2008-03224) is gratefully acknowledged.
22 Y. Omomo, T. Sasaki, L. Z. Wang and M. Watanabe, J. Am. Chem.
Soc., 2003, 125, 3568.
23 Y. Oaki and H. Imai, Angew. Chem., Int. Ed., 2007, 46, 4951.
24 L. Z. Wang, Y. Omomo, N. Sakai, K. Fukuda, I. Nakai, Y. Ebina,
K. Takada, M. Watanabe and T. Sasaki, Chem. Mater., 2003, 15,
2873.
References
1 J. Liu, J. P. Durand, L. Espinal, L. J. Garces, S. Gomez, Y.-C. Son,
J. Villegas and S. L. Suib, Layered Manganese Oxides: Synthesis,
Properties, and Applications, in Handbook of Layered Materials, ed.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 17718–17723 | 17723