Complete Formaldehyde Removal over 3D Structured Na1.1Mn4O8@Mn5O8 Biphasic-Crystals

Article abstract of DOI:10.1002/cctc.202000449

With the increasing concerns on the formaldehyde (HCHO) hazard, the efficient oxidation of HCHO over affordable and harmless catalyst is of significant practical interest. Herein, the intermediate state of Mn₅O₈?Na_1.1Mn₄O₈ (Na?Mn₅O₈) biphasic-crystal catalyst, which features Na_1.1Mn₄O₈-nanosheet@ Mn₅O₈-nanorod interconnected 3D structure, is successfully fabricated through the anaerobic calcination of manganite followed by alkali treatment. The resultant Na?Mn₅O₈ exhibits 100 % conversion of HCHO (150 ppm and 60000 h^?1) and reliable stability (50 h) at 90 °C under demanding water-free feed conditions, which is superior to that of the reference samples of Mn₅O₈, Na_1.1Mn₄O₈, MnO₂ and Na?MnO₂. The experimental results demonstrate that the abundant structure defects (oxygen vacancies) on Na?Mn₅O₈, which may locate along the phase interfaces / shared-edges of Mn₅O₈ nanorods and Na_1.1Mn₄O₈ nanosheets, significantly enhance the active oxygen species and finally contribute to the high activity for HCHO oxidation. Furthermore, the 3D structure of Na?Mn₅O₈ favors mass transfer and fully exposing active sites for the HCHO adsorption and conversion. These findings reveal a way to enhance the catalytic activity of metal oxides via controlling the intermediate state of heteropical crystals.

Full text of DOI:10.1002/cctc.202000449

Full Papers
doi.org/10.1002/cctc.202000449
ChemCatChem
1
2
3
4
5
Complete Formaldehyde Removal over 3D Structured
Na1.1Mn4O8@Mn5O8 Biphasic-Crystals
Chunlei Zhang,[a, b] Guojuan Liu,[b] Ping Wu,[b] Gaofeng Zeng,*[b, c] and Yuhan Sun[b, c]
6
7
8
9
With the increasing concerns on the formaldehyde (HCHO)
hazard, the efficient oxidation of HCHO over affordable and
harmless catalyst is of significant practical interest. Herein, the
intermediate state of Mn5O8À Na1.1Mn4O8 (NaÀ Mn5O8) biphasic-
crystal catalyst, which features Na1.1Mn4O8-nanosheet@ Mn5O8-
nanorod interconnected 3D structure, is successfully fabricated
through the anaerobic calcination of manganite followed by
alkali treatment. The resultant NaÀ Mn5O8 exhibits 100% con-
version of HCHO (150 ppm and 60000 hÀ 1) and reliable stability
Na1.1Mn4O8, MnO2 and NaÀ MnO2. The experimental results
demonstrate that the abundant structure defects (oxygen
vacancies) on NaÀ Mn5O8, which may locate along the phase
interfaces / shared-edges of Mn5O8 nanorods and Na1.1Mn4O8
nanosheets, significantly enhance the active oxygen species
and finally contribute to the high activity for HCHO oxidation.
Furthermore, the 3D structure of NaÀ Mn5O8 favors mass transfer
and fully exposing active sites for the HCHO adsorption and
conversion. These findings reveal a way to enhance the catalytic
activity of metal oxides via controlling the intermediate state of
heteropical crystals.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
°
(50 h) at 90 C under demanding water-free feed conditions,
which is superior to that of the reference samples of Mn5O8,
1. Introduction
mental-friendly materials, were widely studied due to their
excellent reactivity in many catalytic oxidation reactions e.g.
ethyl acetate and CO oxidation.[12] Compared with the noble
metal based catalysts, however, MnOx is still challenged by the
relatively low reactivity for HCHO oxidation.
Formaldehyde (HCHO),
a major pollutant among volatile
organic compounds and one of the 1st group human carci-
nogens, released from modern building materials, furniture and
consumables poses a serious threat to our living quality and
health.[1] It is urgent to develop appropriate technologies for
efficient HCHO removal. Catalytic oxidation of HCHO to
harmless CO2 and H2O is the most promising approach for the
HCHO elimination due to its continuous and facile operation.[2]
Although the noble metal-based catalysts, such as Pt/PbO,[3] Pt/
ZrO2,[4] Au/FeOx[5] and Pd/TiO2,[6] exhibited complete oxidation
of HCHO even at room temperature, the unaffordable cost
limits their practical application in large scale. Thus, there are
many interests in developing highly efficient HCHO oxidation
MnOx could compose by a wide range of stoichiometry of
Mn and O in the formation of various crystal phases like β-
MnO2, γ-MnO2, α-Mn2O3, γ-Mn2O3 and α-Mn3O4 due to the
special electronic structure of Mn 3d54s2.[13] To improve the
reactivity of MnOx, many efforts have been made by investigat-
ing the effects of valance state, crystal structure, defects and
heteroatoms on the HCHO oxidation. Zhang et al. found that δ-
MnO2 is more active than α-, β- and γ-MnO2 due to its special
2D layered structure and high lattice oxygen migration rate.[14]
Rong et al. demonstrated that 3D structure MnO2 has high
reactivity in HCHO conversion owing to the increased amount
of active sites and the intensified mass transfer over the
catalysts.[15] Furthermore, the heteroatoms also play an impor-
tant role on the reactivity of MnOx. With doping Ce atoms into
MnO2 lattice, the surface adsorbed oxygen species and the
migration rate of surface lattice oxygen were improved, led to
an enhanced HCHO oxidation performance.[1b,16] Xu et al.
introduced K+ ions into the lattice of manganese oxide to form
d-sp hybrid orbitals, which improved the activity of lattice
oxygen and the surface adsorbed oxygen species in the HCHO
oxidation.[17] On the other hand, it is well known that the mixed
valency catalysts have higher reactivity in the catalytic oxidation
because that the high-low valences cycling benefits to the
oxidation of reagent and recovery of catalyst.[11] However, MnOx
studied for HCHO oxidation mostly focus on MnO2 with Mn(IV)
in the catalysts. Thus, it is highly attractive to improve the
catalytic performance by developing Mn mixed valency cata-
lysts. Mn5O8 has the chemical formula of Mn(II)2Mn(IV)3O8,
possessing a mixed valency of Mn. It has a similar crystal
catalysts using the transition metal oxides such as MnO2,[7]
[10]
Co3O4,[8] CeO2,[1b] MnO2À CeO2,[1b,9] TiO2
and FeOx.[11] Among
them, manganese oxide (MnOx) catalysts, a serial of environ-
[a] C. Zhang
Institute of Nanochemistry and Nanobiology
School of Environmental and Chemical Engineering
Shanghai University
99 Shangda Road
Shanghai 200444 (P. R. China)
[b] C. Zhang, G. Liu, P. Wu, Prof. G. Zeng, Dr. Y. Sun
CAS Key Laboratory of Low-carbon Conversion Science and Engineering
Shanghai Advanced Research Institute
Chinese Academy of Sciences
100 Haike Road
Shanghai 201210 (P. R. China)
E-mail: zenggf@sari.ac.cn
[c] Prof. G. Zeng, Dr. Y. Sun
School of Chemical Sciences
University of Chinese Academy of Sciences
19 A Yuquan Road
Beijing 100049 (P. R. China)
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/cctc.202000449
structure to Cd2Mn3O8 and Ca2Mn3O8 which is
a highly
ChemCatChem 2020, 12, 1–12
1
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202000449
ChemCatChem
asymmetric monoclinic structure. In addition, Mn5O8 has a
layered structure composed of a [Mn3O8]4À anionic layer with a
MnO6 octahedron and a Mn2+ cationic layer, which would be
easy for the heteroatom intercalation.[18] According to the
aforementioned effects of valency, crystal structure and heter-
oatom on the reactivity of MnO2 based catalysts, these special
features of Mn5O8 signs a great potential in catalytic oxidation
of HCHO.
In the present work, we designed and demonstrated a facile
strategy to fabricate 3D Mn5O8À Na1.1Mn4O8 mixed-crystalline
catalyst, comprised of Na1.1Mn4O8-nanosheet@ Mn5O8-nanorod
hybrid structure and large and accessible surface area, through
the anaerobic calcination of manganite followed by alkali
treatment. The resultant co-crystal catalysts exhibited excellent
activity and reliable stability for the complete oxidation of
HCHO. The structure, surface chemistry, redox properties and
oxidation performance of the intermediate phase of
Mn5O8À Na1.1Mn4O8, pure phases of Mn5O8 and Na1.1Mn4O8 as
well as the reference samples of MnO2 and NaÀ MnO2 were
investigated in detail. Concomitantly, the role of the unique
features of Mn5O8À Na1.1Mn4O8 on HCHO oxidation mechanism
was also evaluated.
intercalate into the [Mn3O8]4À /Mn2+ layered structure of Mn5O8
in compared with MnO2.
1
2
3
4
5
6
7
8
9
The morphology and microstructure of these samples were
measured by scanning electron microscopy (SEM) and trans-
mission electron microscopy (TEM), as shown in Figure 2. As
MnOOH prefers to grow along the [0 0 1] direction, the
precursor MnOOH exhibited a tetragonal nanorod structure
with a high aspect ratio (Figure 2A and B). Under the high
resolution, MnOOH displayed the ordered microstructure with
an interplanar spacing of 0.263 nm (Figure 2C), which can be
assigned to MnOOH (0 2 0). The nanorod morphology of
MnOOH was well remained for the obtained Mn5O8 (Figure 2D
and E). The interplanar spacings are 0.486 and 0.501 nm,
matching the (2 0 0) and (À 1 1 0) facets of Mn5O8, respectively
(Figure 2F). Interestingly, the NaÀ Mn5O8 biphasic-crystals display
two kinds of morphology, in which the ultrathin 2D nanosheets
formed along the edges of Mn5O8 nanorods to build an
interconnected 3D structure (Figure 2G). As the single nanorod
shown in Figure 2I, the ultrathin nanosheet sprouted along the
longitudinal direction of Mn5O8 nanorod. This epitaxiaLgrowth
nanosheet was proved as Na1.1Mn4O8 due to the interplanar
spacing of 0.242 nm is ascribed to Na1.1Mn4O8 (2 0 À 2). For the
Na1.1Mn4O8 sample, the nanorod structure disappeared and only
wrinkling nanosheets were observed (Figure 2J). The nanosheet
of Na1.1Mn4O8 displayed a scroll structure with the nanorod as
an axis (Figure 2K). The unchanged interplanar spacing
(0.242 nm) proves the formation of Na1.1Mn4O8 pure phase. This
is consistent with the XRD analysis. During the phase transition
from Mn5O8 to Na1.1Mn4O8, it is reasonable that a part of Mn5O8
nanorod was firstly dissolved in the alkali solution and then
rearranged and epitaxially grew on the defect surface of Mn5O8
until the whole Mn5O8 nanorod was used up. Generally, high
index plane contains the highly discordant electron distribution,
which leads to the high reactivity.[19] Herein, the higher index
planes exposed on NaÀ Mn5O8 may benefit to its activity in the
HCHO oxidation. In addition, catalytic reaction active site tends
to be the structural defects, such as edges and corners. The
intermediate phase of NaÀ Mn5O8 potentially have higher
defects density due to the uncomplete phase transition and
thus high reactivity in HCHO oxidation. On the other hand, the
phase transition from 1D nanorod of Mn5O8 to 2D nanosheet of
Na1.1Mn4O8 and 3D nano-sheet/rod significantly increased the
specific surface areas from 27.6 to 50.8 and 47.1 m2 gÀ 1,
respectively (Table 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
2. Results and Discussion
2.1. Structure and texture of MnOx based catalysts
Figure 1A illustrates the preparation procedure of manganese
oxides. Mn5O8 and MnO2 (reference sample) nanorods could be
obtained by heating the precursor of manganite (MnOOH) in
inert atmosphere and air, respectively. In sequence, the
NaÀ Mn5O8 biphasic-crystals (i.e. Mn5O8À Na1.1Mn4O8) and pure
phase of Na1.1Mn4O8 are prepared by the different strength of
alkali treatments on Mn5O8. The structure transformation of the
manganese oxide species in preparation was monitored by X-
ray diffractometry (XRD), as shown in Figure 1B. The un-
calcinated precursor is assigned to the manganite i.e. Mn3+
O
(OH) species (JCPDS 41-1379), which transferred to Mn5O8
(JCPDS 39-1218) upon the calcination in N2. After the 1st alkali
treatment on Mn5O8, interestingly, additional two sharp and
°
intense peaks at 12.5 and 25.2 , ascribed to Na1.1Mn4O8 ·3H2O
(JCPDS 43–1456), were observed. It suggests that partial Mn5O8
transferred to Na1.1Mn4O8 ·3H2O, resulting in Mn5O8À Na1.1Mn4O8
(NaÀ Mn5O8) biphasic-crystals. Further treating the NaÀ Mn5O8
biphasic-crystals with harsher conditions, the residual Mn5O8 in
NaÀ Mn5O8 biphasic-crystals were completely convert into the
pure phase Na1.1Mn4O8, as only Na1.1Mn4O8 pattern was
observed.
Table 1. Physicochemical properties of Mn5O8, MnO2 and their derivatives.
Samples
SBET
XPS
Na
As control, β-MnO2 (JCPDS 24-0735) could be obtained by
calcining MnOOH in air. In addition, β-MnO2 kept stable upon
the alkali treatment because of no visual structure changes
were observed from the XRD pattern of alkali treated β-MnO2
(Figure S1). The structure analysis indicates that the intermedi-
ate phase between pure Mn5O8 and Na1.1Mn4O8 can be
fabricated through an appropriate alkali treatment of Mn5O8.
On the other hand, it also reveals that Na ions are easily
[m2/g]
Mn/O
Oads/Olatt
Mn4+/Mn2+
AOS
[at.%]
Mn5O8
27.6
47.1
50.8
10.7
14.1
0
0.50
0.25
0.32
0.36
0.27
0.50
0.61
0.43
0.21
0.39
0.6
1.7
2.8
3.4
3.7
3.9
3.9
NaÀ Mn5O8
Na1.1Mn4O8
MnO2
NaÀ MnO2
5.1
10.7
0
[a]
–
[a]
–
[a]
2.1
–
[a] Only Mn (IV) was indicated by the XPS results.
ChemCatChem 2020, 12, 1–12
www.chemcatchem.org
2
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202000449
ChemCatChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 1. (A) Scheme of MnOx preparation; (B) XRD patterns of MnOOH, Mn5O8, NaÀ Mn5O8 mixed-crystal and Na1.1Mn4O8.
As comparison, the as-prepared MnO2 also exhibited a
nanorod morphology (Figure S2). The interplanar spacing,
0.318 nm, was ascribed to β-MnO2 (1 1 0), in line with the XRD
measurements. Unlike the morphology change for alkali treated
Mn5O8, NaÀ MnO2 remained the nanorod morphology of MnO2
and the almost same interplanar spacing of 0.316 nm, which
confirms the structure stability of MnO2 in the alkali treatment.
2.2. Surface chemistry of MnOx catalysts
The surface chemical states of catalysts were elucidated by X-
ray photoelectron spectroscopy (XPS). The Mn2p spectrum of
Mn5O8 shows typically two components of Mn 2p3/2 and Mn
2p1/2, in which the binding energies 642.9/654.7 eV and 641.0/
652.8 eV could be ascribed to Mn (IV) and Mn (II), respectively
(Figure S3). In comparison, the Mn2p peaks of NaÀ Mn5O8 and
Na1.1Mn4O8 shifted to higher binding energies, indicating a
more oxidic state of Mn and an increased content of Mn(IV).[20]
ChemCatChem 2020, 12, 1–12
www.chemcatchem.org
3
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202000449
ChemCatChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 2. SEM (A, D, G, J) and TEM (B, C, E, F, H, I, K, L) images of MnOOH (A–C), Mn5O8 (D–F), NaÀ Mn5O8 (G–I) and Na1.1Mn4O8 (J–L).
Similarly, the Mn 3 s XPS also displayed the oxidation state of
surface Mn atoms. Two peaks appear in the Mn 3 s spectra
because of the coupling of Mn 3 s electron with Mn 3d electron
giving rise to parallel or antiparallel spin configuration.[21] The
magnitude of the splitting energy of Mn 3 s peaks increase
approximately linearly as the Mn oxidation state decreases,
which can be described as the empirical Equation 1:[22]
treatment due to the stable structure of MnO2 (Table 1 &
Figure S4).
The O 1s XPS spectrum of Mn5O8 can be deconvoluted into
three peaks at 529.78 eV, 531.03 eV and 532.88 eV, correspond-
ing to lattice oxygen (Olatt), surface adsorbed oxygen (Oads
)
À
species (such as O2 and OÀ groups) and adsorbed -OH group
(OOH), respectively (Figure 3B).[24] Olatt and Oads were the predom-
inant O species on Mn5O8. Compared with Mn5O8, the peaks of
Olatt and Oads of NaÀ Mn5O8 and Na1.1Mn4O8 shifted to higher
binding energies, which might be contributed to the changes
in the chemical environment of oxygen atoms after rearrange-
ment of crystal. The formation of atom vacancy would reduce
DE � 7:88 À 0:85n
(1)
Where ΔE is the Mn 3 s binding energy difference and n
(2�n�4) is the average oxidation state (AOS) of the Mn atoms.
The value of ΔE decreased from 5.50 eV for Mn5O8 to 4.98 and
4.74 eV for NaÀ Mn5O8 and Na1.1Mn4O8, respectively (Figure 3A).
Correspondingly, the AOS of Mn increased from 2.8 for Mn5O8
to 3.4 for NaÀ Mn5O8 and further to 3.7 for Na1.1Mn4O8 (Table 1),
in line with the Mn2p results. It is reasonable that the surface
electron density on the treated samples was decreased by the
intercalation of Na+, leading to the increase of valence state of
Mn. It has been reported that the Mn mixed oxide with higher
Mn (IV) content possess higher reactivity in the oxidation.[23]
Therefore, it suggests that alkali treated samples would more
active in the HCHO oxidation. In contrast, the valence state of
MnO2 kept unchanged at Mn (IV) before and after the alkali
the electron density around Olatt and therefore increase the
[7c]
binding energies of Olatt
.
Owing to the lower valence state of
Na+ than Mn (II) / Mn (IV), furthermore, the O atoms around the
Mn defects are unsaturated, which are easy to be
hydroxylated.[24] Therefore, the content of OOH increased from
7.7% for Mn5O8 to 11.8-13.2% for the alkali treated samples
(Figure 3B). In addition, the O vacancies and /or unsaturated O
atoms are migratable as active oxygen species due to the weak
interaction between O and Na+. The active oxygen species play
a key role in the oxidation of HCHO. On one hand, the Oads
tends to participate into the oxidation reaction due to its high
activity and the interaction between catalyst surface and
ChemCatChem 2020, 12, 1–12
www.chemcatchem.org
4
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202000449
ChemCatChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 3. (A) Mn 3 s, (B) O 1s and (C) Na 1s XPS spectra and (D) FT-IR spectra of Mn5O8, NaÀ Mn5O8 and Na1.1Mn4O8.
reactant. On the other hand, Olatt with high mobility enhances
the formation of surface active oxygen species through the
complex migration between surface lattice oxygen and oxygen
vacancies.[7c] In this case, NaÀ Mn5O8 owned the highest Oads
content, i.e. 32.9% and the largest Oads/Olatt ratio of 0.61 among
these samples (Table 1 and Figure 3B), suggesting that
NaÀ Mn5O8 contains abundant atom (oxygen) vacancies and
thus possesses high reactivity for HCHO oxidation.
The Na content on sample surface increased from 0 for
Mn5O8 to 5.1 and 10.7 at.% for NaÀ Mn5O8 and Na1.1Mn4O8,
respectively (Figure 3C and Table 1), in line with the observation
of XRD. Compared with the Mn5O8 based samples, the Na+
content in NaÀ MnO2 is much lower (Figure S5 and Table 1),
confirming the relative inertness of MnO2 in the alkali
conditions. Both experimental and simulation have suggested
that alkali metal ions can facilitate the activation of oxygen
molecules and lattice oxygen.[17] The surface groups of catalysts
were measured by FTÀ IR (Figure 3D and Figure S6). The bands
at 3250–3,500 cmÀ 1 and 1650 cmÀ 1 can be ascribed to the
stretching vibration of À OÀ H and the bending vibration of À OH
groups in water, respectively.[25] Among these samples,
NaÀ Mn5O8 possesses more À OH groups, which is consistent
with the O1s XPS analysis.
The local structure state and the lattice oxygen activity of
Mn5O8, NaÀ Mn5O8 and Na1.1Mn4O8 were analyzed by Raman
spectroscopy (Figure 4A). The Raman band located around
270 cmÀ 1 could be attributed to the stretching vibration of
Mn5O8,[22] which is weakened for NaÀ Mn5O8 and then disap-
ChemCatChem 2020, 12, 1–12
www.chemcatchem.org
5
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202000449
ChemCatChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 4. (A) Raman spectra and (B) MnÀ O bond force constant of Mn5O8, NaÀ Mn5O8 and Na1.1Mn4O8.
peared for Na1.1Mn4O8. In contrast, there is no signal around
270 cmÀ 1 for MnO2 and its alkali treated sample (Figure S7),
suggesting that it is a characteristic band for Mn5O8. It is clear
that NaÀ Mn5O8 is an intermediate mixed-crystal containing
unconverted Mn5O8 while Na1.1Mn4O8 is single phase without
Mn5O8 residues. The Raman band around 640 cmÀ 1 can be
attributed to the [MnO6] symmetric stretching vibrations.[26] The
band intensity gradually weakened with the alkali treatment. In
addition, the Raman shift moved from 643 cmÀ 1 of Mn5O8 and
Na1.1Mn4O8 to 638 cmÀ 1 of NaÀ Mn5O8, which may be contrib-
uted by the oxygen vacancies in NaÀ Mn5O8.[27] The Raman band
around 570 cmÀ 1 is assigned to the stretching vibration caused
mainly by Mn4+ in the basal plane of the [MnO6] sheet.[28] This
band was found as a tiny shoulder signal for NaÀ Mn5O8 but a
strong peak for Na1.1Mn4O8. It confirms that more Mn (IV)
formed in sequence for NaÀ Mn5O8 and Na1.1Mn4O8, in agree-
ment with the XPS results. In addition, no band ascribed to Na+
/O interaction was found in the Raman results. It reveals that no
strong bond of NaÀ O was formed and the unrestricted Na+ and
lattice oxygen would help to adsorb and activate oxygen and
HCHO molecules.[13]
NaÀ Mn5O8 is mostly caused by the intermediate state of
NaÀ Mn5O8. In other words, the existence of oxygen vacancies in
NaÀ Mn5O8 contributes to the decline of MnÀ O force constant.
Furthermore, all Mn5O8 based samples displayed lower k than
that of MnO2 and NaÀ MnO2 (Figure S9). It means that MnO2
structure is more stable than Mn5O8 as already proved by XRD
and XPS results, and correspondingly, Mn5O8 is more active
than MnO2, which will be confirmed by the following HCHO
oxidation. According to the above analysis, the uncomplete
phase transition from Mn5O8 to Na1.1Mn4O8 left abundant
oxygen vacancies in NaÀ Mn5O8, which impacts the interactions
of MnÀ O and then the oxidation activity for HCHO.
Figure 5A shows the hydrogen temperature programmed
reduction (H2À TPR) profiles of the Mn5O8 based samples. The
°
profile of Mn5O8 reveals two peaks between 230 and 390 C,
representing the initial reduction of adsorbed active oxygen
species and the reduction of Mn5O8 to MnOy (y<1.6),
respectively.[30] For NaÀ Mn5O8 and Na1.1Mn4O8, the reduction of
°
active oxygen species slightly shifted from 235 C for Mn5O8 to
°
268 and 278 C, respectively. This reveals that the intercalation
of Na+ contributed to the activation and adsorption of oxygen
and then reinforce the interaction force of active oxygen on the
sample.[31] With the decrease of Mn5O8 content in the sample,
on the other hand, the peak intensity of Mn5O8 reduction
gradually declined for NaÀ Mn5O8 and Na1.1Mn4O8. As the new
generated Mn (IV) is easier reduced, furthermore, the reduction
Based on the Hooke’s law, on the other hand, the MnÀ O
bond force constant (k) can be calculated from the Raman
results by using the following equation (2).[29]
sffiffiffi
1
2pc
k
m
°
temperature slightly decreased from 384 C for Mn5O8 to 377
w ¼
(2)
°
and 368 C for NaÀ Mn5O8 and Na1.1Mn4O8, respectively, due to
the increase of Mn (IV) content in these samples.
where ω is the Raman shift (cmÀ 1) around 640 cmÀ 1, c is the
velocity of light (msÀ 1), and μ is the effective mass (g). As shown
in Figure 4B, NaÀ Mn5O8 owned the lowest MnÀ O force constant
(k). However, the k value of pure Na1.1Mn4O8 even a little bit
higher than Mn5O8. This reveals that the decreased k of
The oxygen activation of the samples was measured by
oxygen temperature programmed desorption (O2-TPD) analysis
(Figure 5B). The signals in the O2-TPD profiles at low temper-
°
°
ature (<300 C), moderate temperature (300–600 C), and high
°
temperature (>600 C) correspond to the release of surface
ChemCatChem 2020, 12, 1–12
www.chemcatchem.org
6
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202000449
ChemCatChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 5. (A) H2-TPR and (B) O2-TPD of Mn5O8, NaÀ Mn5O8 and Na1.1Mn4O8.
chemisorbed oxygen molecules / active surface oxygen, sub-
surface lattice oxygen and bulk lattice oxygen, respectively.[32]
Oxygen that released at low temperature was considered as
labile oxygen species, which was highly active for HCHO
oxidation. Mn5O8 shows only one peak at the moderate
temperature, reflecting to a high content of sub-surface lattice
oxygen on Mn5O8 and a relatively low capacity for oxygen
molecule activation. In contrast, the alkali treated samples offer
significantly enhanced activation capacity for oxygen molecules
because that both of them displayed down-shifted signals at
by the special 2D layered structure, the low MnÀ O interaction
force and the mixed valency of Mn5O8 that have been proved
by the measurements above. For the intermediate phase of
NaÀ Mn5O8, it exhibited the highest HCHO conversion in the
whole range of temperature and the T100 value was further
°
declined to 90 C, demonstrating the highest activity among
these samples. In comparison, the pure Na1.1Mn4O8 exhibited a
similar activity to Mn5O8. The HCHO oxidation performance of
NaÀ Mn5O8 is also compared with the literature results, as shown
in Table 2.[7c,15,24,28,30a,33] Under the similar feeding conditions,
NaÀ Mn5O8 offers the lowest T100, which is even better than the
supported metal (e.g. Ag and Ir) catalysts. It is noted that the
powder paraformaldehyde was used as HCHO resource in this
work, and to keep the HCHO concentration of feedstock stable
and reliable, the dehydrated cylinder air was introduced as
sweep gas for HCHO feedings. Therefore, the humidity of the
feedstock is much lower than the real environment and that
used in literature. It has been reported that water/humidity
plays a very important role in the promotion of HCHO catalytic
oxidation.[3,34] Therefore, NaÀ Mn5O8 exhibited highly efficient
HCHO removal under low humidity conditions. Combination
with the structure and chemistry analysis above, the intermedi-
ate phase state of NaÀ Mn5O8, which led to a heteropical 3D
structure of Na1.1Mn4O8-nanosheet@Mn5O8-nanorod and espe-
cially the abundant structure defects (e.g. oxygen vacancies),
contributed to the high activity of this material in HCHO
oxidation because that both mass transfer of reagents and
oxygen/HCHO activation were intensified.
°
~130 C. Importantly, the low temperature signal of NaÀ Mn5O8
is sharper and more intensive than that of Na1.1Mn4O8,
confirming the high capacity for oxygen activation and
adsorption. For the moderate temperature signal, in addition, it
°
°
decreased from ~540 C for Mn5O8 to ~380 C for NaÀ Mn5O8,
indicating that the sub-surface lattice oxygen of NaÀ Mn5O8
were well activated. As a reference, only signals at the temper-
°
ature higher than 500 C were observed from the O2-TPD
measurements of MnO2 and NaÀ MnO2 (Figure S9).
2.3. Catalytic oxidation of HCHO over MnOx
The catalytic oxidation of HCHO on the catalysts was tested via
the continuous flow method in a fixed-bed micro-reactor, as
illustrated in Figure S10. Figure 6A shows the temperature
dependence of HCHO conversion over the MnOx catalysts at
°
50–150 C. The conversion of HCHO were increased with the
rise of reaction temperature. The activity of different samples
The apparent activation energy (Ea) of HCHO oxidation over
the catalysts was also calculated by using the Arrhenius
equation (Figure 6B). Ea for NaÀ Mn5O8, Na1.1Mn4O8, Mn5O8,
NaÀ MnO2 and MnO2 were 44.7, 45.3, 45.7, 53.5 and
58.9 kJmolÀ 1, respectively, which means HCHO was most easily
followed
the
order:
NaÀ Mn5O8 >Na1.1Mn4O8 �Mn5O8 >
NaÀ MnO2 �MnO2. T100 (corresponding to the complete con-
°
version temperature of HCHO) of Mn5O8 is 110 C, which is
°
much lower than that of MnO2 (140 C). It may be contributed
ChemCatChem 2020, 12, 1–12
www.chemcatchem.org
7
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202000449
ChemCatChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 6. (A) HCHO conversion over various MnOx at different temperature (δHCHO =150 ppm, GHSV=60000 mLgÀ 1 hÀ 1); (B) Temperature dependence of HCHO
conversion on various MnOx; (C) Effects of GHSV on the HCHO conversion over NaÀ Mn5O8 at different temperature (δHCHO =150 ppm); (D) Temperature cycling
of NaÀ Mn5O8 in 50–90 C and the stability of NaÀ Mn5O8 at 90 C (δHCHO =150 ppm, GHSV=60000 mLgÀ 1 hÀ 1).
°
°
Table 2. Catalytic performance of reported catalysts for the oxidation of HCHO.
°
Catalyst
Feeding conditions
T100 [ C]
Ref.
[ppmmLgÀ 1 hÀ 1
]
1 K/MnO2
K1/HMO
VMn-3
9000
8400
4800
8400
12000
9000
18000
3600
6000
7200
9000
90
[28]
[30a]
[24]
120
100
130
125
130
90
100
170
100
90
Ag1/HMO
Ir/TiO2À R
[31]
[7c]
[15]
[33b]
[33c]
[37]
α-MnO2-110
3DÀ MnO2
m-NiO-s
MnÀ CuÀ Ce
MnOxÀ CÀ Co3O4À CeO2
NaÀ Mn5O8
This work
activated and oxidized by NaÀ Mn5O8. It is well consistent with
the reactivities of these catalysts. Furthermore, the effect of
GHSV on the HCHO oxidation was evaluated over NaÀ Mn5O8 in
performance during the four-time temperature cycling. During
50 h operation, furthermore, the HCHO conversion maintained
at ~100% without deactivation. After reaction, no visual change
of microstructure was observed from NaÀ Mn5O8 (Figure S10).
These indicated the good stability of NaÀ Mn5O8 for the
elimination of HCHO.
À 1
the range of 30,000 to 120,000 mLgcat hÀ 1 (Figure 6C). The
value of T100 dropped to 75 C at GHSV=30000 mLgcatÀ 1 hÀ 1 and
°
then increased to 105 C under 120000 mLgcatÀ 1 hÀ 1, suggesting
°
that higher GHSV would inhibit the oxidation process. The
stability of NaÀ Mn5O8 for HCHO oxidation was tested by
°
temperature cycling in the range of 50–90 C and long-term
°
operation at 90 C (Figure 6D). The catalyst displayed stable
ChemCatChem 2020, 12, 1–12
www.chemcatchem.org
8
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202000449
ChemCatChem
2.4. Catalytic mechanism of HCHO over NaÀ Mn5O8
and υ(CH) from the intermediates of formate species
(HCOOÀ ).[24] Carbonyl carbon in HCHO is electrophilic, which
tends to adsorb on nucleophilic surface oxygen atom to form
DOM, which will be further oxidized to formate. With the
exposure time, the peaks of DOM decreased while the signals
of formate increased. It reveals that the conversion of DOM
species is a relatively fast step but the further conversion of
formate to carbonate species (CO32À ) is a limited step. The
carbonate species were observed at the band of 1668 cmÀ 1,
which would finally convert into CO2 (2347 cmÀ 1). In addition,
the negative peak of hydroxyLgroups (~3480 cmÀ 1) was
ascribed to the loss of surface À OH, suggesting that À OH
groups were involved and consumed in the HCHO oxidation.[3]
According to the obtained findings, a possible process of
HCHO removal on NaÀ Mn5O8 is proposed, as shown in Fig-
ure 7B. HCHO is firstly adsorbed on the surface of catalyst and
1
2
3
4
5
6
7
8
9
To gain insights into the reaction, the HCHO oxidation on
NaÀ Mn5O8 was investigated by in situ diffuse reflectance infra-
red Fourier transform spectra (DRIFTS) in a flow of HCHO-O2
mixture at room temperature. As shown in Figure 7A, the
absorption bands at 1120, 1340, 1367, 1448, 1518, 1668, 2347,
2846 and 3200–3650 cmÀ 1 were observed from NaÀ Mn5O8. The
band at around 1120 cmÀ 1 was corresponded to molecularly
adsorbed HCHO.[11] As the conversion of adsorbed HCHO into
its intermediates is a fast step, the intensity of this signal is
relatively weak.[11] The band at 1448 cmÀ 1 was assigned to
δ(CH2) of dioxymethylene species (H2CO2, DOM), which are the
intermediates of HCHO oxidation and can be easily converted
to formate species. [35] Otherwise, the bands at 1340, 1367, 1518
and 2846 cmÀ 1 could be attributed to υs(COO), δ(CH), υas(COO)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 7. (A) In situ DRIFTS spectra of NaÀ Mn5O8 exposed to the flow of HCHO-O2 (δHCHO =30 ppm) at room temperature; (B) The proposed mechanism of
HCHO oxidation over NaÀ Mn5O8.
ChemCatChem 2020, 12, 1–12
www.chemcatchem.org
9
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202000449
ChemCatChem
Na1.1Mn4O8 ·3H2O-Mn5O8 biphasic-crystals (denoted as NaÀ Mn5O8)
and NaÀ MnO2 were synthesized through the reaction between
NaOH and the corresponding manganese oxides. In a typical
procedure, 0.1 g of as-synthesized manganese oxides was added to
the adsorption is enhanced by À OH groups via forming hydro-
gen bonding with HCHO. The adsorbed HCHO would be
oxidized by surface active oxygen species into DOM, formate,
carbonate and CO2, in sequence,[11] as proved by the DRIFTS. On
one hand, the surface active oxygen species are released by the
redox cycle of Mn(IV)/Mn(II). On the other hand, the oxygen
molecules in the feedstock would be adsorbed on the oxygen
vacancies of NaÀ Mn5O8 and then dissociated to active oxygen
species. With the consumption of active oxygen, the oxygen
vacancies would be recovered for the next cycle. In this process,
Na+ would enhance the stability of active oxygen species
through Na+/oxygen species interactions. The consumed À OH
groups would be replenished by the reaction between water
and active oxygen species.
1
2
3
4
5
6
7
8
9
100 mL NaOH solution (40 gLÀ 1) and stirred for 20 h at 100 C. The
°
°
resultant solid was then washed with water and dried at 70 C
overnight. For the preparation of pure Na1.1Mn4O8 ·3H2O (denoted
as Na1.1Mn4O8), 0.1 g of NaÀ Mn5O8 biphasic-crystals was further
thermal-treated in NaOH solution (200 gLÀ 1) in autoclave for
°
another 20 h at 160 C.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Formaldehyde oxidation
The HCHO oxidation over the packed catalyst (100 mg, 40–
60 mesh) was carried out in a tubular fixed-bed quartz reactor
under atmospheric pressure (Figure S11). The HCHO feedstock (~
150 ppm) is generated by passing a stream of pure air (Pujiang Gas
Co., Shanghai) through paraformaldehyde powders at room
temperature. The gaseous hourly space velocity (GHSV) was
controlled in the range of 30,000 and 120,000 mLgcatÀ 1 hÀ 1 by
adjusting the flux of air. The HCHO concentration of inlet was
controlled by the amount of paraformaldehyde and the GHSV,
which was confirmed firstly by gas chromatography (GC) for each
test. The relative humidity of the feedstock was measured by a
psychrometer (i.e. ~30% in this work). The HCHO concentration in
the reactant or product gas stream was analyzed by a GC with a
3. Conclusions
The pure phase of Mn5O8 nanorod and Na1.1Mn4O8 nanosheet as
well as their intermediate phase of Mn5O8À Na1.1Mn4O8
(NaÀ Mn5O8) were synthesized by anaerobic calcination of
manganite with and without alkali treatments. During the phase
transformation from Mn5O8 to Na1.1Mn4O8, the morphology
changed from nanorod to ultrathin nanosheet. The intermedi-
ate phase of Na- Mn5O8 kept both morphologies of these pure
phases, resulting in a heteropical 3D structure building by
Na1.1Mn4O8-nanosheet@Mn5O8-nanorod. The resultant Na-
Mn5O8 enables complete conversion of HCHO and reliable
stability at low temperature under demanding HCHO feed
conditions, which is superior to that of Mn5O8, Na1.1Mn4O8 and
the reference of MnO2 and NaÀ MnO2 in this work as well as that
of the reported manganese oxide catalysts. The experimental
results demonstrated that the abundant structure defects
(oxygen vacancies) on NaÀ Mn5O8, which may locate along the
interface of Mn5O8 nanorods and Na1.1Mn4O8 nanosheets,
significantly enhanced the surface active oxygen species and
finally contributed to better activity for HCHO oxidation.
Furthermore, NaÀ Mn5O8 exhibited continuous interconnected
3D structure, which favors mass transfer and fully expose
reaction-active sites for the HCHO adsorption and conversion.
thermal conductivity detector (TCD). The HCHO conversion (ηHCHO
)
was calculated by the Equation 3.
dfeed À doutlet
*
hHCHO
¼
100 %
(3)
dfeed
Where δfeed and δoutlet is the HCHO concentration before and after
reaction. The reliability of GC method was also confirmed by the
phenol spectrophotometric method.[3] To calculate the apparent
activation energy (Ea) of HCHO oxidation on the catalysts, the ηHCHO
was controlled below 15% by adjusting GHSV.
Characterization
The crystal structure of catalysts was determined by XRD (Rigaku
Ultima IV) using Cu Kα radiation (α= 0.15406 nm, 40 kV, 40 mA).
The morphology of the catalysts was characterized by a trans-
mission electron microscopy (TEM, JEM-2100, 200 kV) and
a
scanning electron microscopy (SEM, Zeiss SUPRA 55 SAPPHIRE, 2–
20 kV). The near-surface chemical state and elemental composition
of catalysts were analyzed by an X-ray photoelectron spectroscopy
(XPS, K-Alpha, Al Kα radiation, 1486.6 eV, 12 kV, 3 mA). XPS peak
positions were calibrated with the help of the C 1s peak at
284.8 eV. The specific surface area of the samples was derived from
N2 sorption measurements carried out on an automatic micropore
physisorption analyzer (TriStar II 3020), using the multipoint
Brunauer-Emmet-Teller (BET) analysis method. Raman spectroscopy
measurements were performed using a Renishaw Raman spectrom-
eter using a 12.5 mW laser source at an excitation wavelength of
532 nm. The chemical information of samples was measured by
Fourier transform infrared (FTIR) spectroscopy (Shimadzu IRAffinity-
1). A Micromeritics AutoChem II 2920 apparatus, equipped with a
thermal conductivity detector (TCD), was used for H2 temperature-
programmed reduction (H2-TPR) and O2 temperature programmed
desorption (O2-TPD) analysis. In situ diffuse reflectance infrared
Fourier transform spectra (DRIFTS) of catalysts were recorded in
Thermo Fisher 6700. Samples were pretreated by N2 for 1 min
under room temperature before exposing to HCHO containing gas
Experimental Section
Preparation of MnOx
Potassium permanganate (KMnO4), polyethylene glycol (PEG 400),
and sodium hydroxide (NaOH, �96%) were purchased from
Aladdin Chemicals (China). All the reagents were analyticaLgrade
and used without further purification. The MnOx precursor, MnOOH,
was firstly prepared through a hydrothermal method.[36] In detail,
the well mixed solution, containing PEG 400 (2 mL), KMnO4 (0.1 g)
and deionized water (80 mL), was placed into a Teflon-lined
°
autoclave (100 mL) and then heated at 160 C for 5 h. The resultant
precipitates were washed with deionized water and ethanol for
°
several times and then dried at 70 C overnight. Mn5O8 was
°
obtained by calcining the as-prepared precursor MnOOH at 280 C
in N2 for 2 h. As a reference, MnO2 was prepared under the air
atmosphere at 280 C for 5 h. The Na+ doped MnOx, i.e.
°
ChemCatChem 2020, 12, 1–12
www.chemcatchem.org
10
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202000449
ChemCatChem
(~30 ppm at a flow rate of 20 mLminÀ 1 balanced by pressured
oxygen).
[13] J. Wang, J. Li, P. Zhang, G. Zhang, Appl. Catal. B 2018, 224, 863.
[14] J. Zhang, Y. Li, L. Wang, C. Zhang, H. He, Catal. Sci. Technol. 2015, 5,
2305.
1
2
3
4
5
6
7
8
9
[15] S. Rong, P. Zhang, Y. Yang, L. Zhu, J. Wang, F. Liu, ACS Catal. 2017, 7,
1057.
[16] L. Zhu, J. Wang, S. Rong, H. Wang, P. Zhang, Appl. Catal. B 2017, 211.
[17] F. Xu, Z. Huang, P. Hu, Y. Chen, L. Zheng, J. Gao, X. Tang, Chem.
Commun. 2015, 51, 9888.
[18] a) D. Jeong, K. Jin, S. E. Jerng, H. Seo, D. Kim, S. H. Nahm, S. H. Kim, K. T.
Nam, ACS Catal. 2015, 5, 4624; b) X. Q. Shan, D. S. Charles, Y. K. Lei, R. M.
Qiao, G. F. Wang, W. L. Yang, M. Feygenson, D. Su, X. W. Teng, Nat.
Commun. 2016, 7, 8.
[19] a) R. A. Van Santen, Acc. Chem. Res. 2009, 42, 57; b) A. Vittadini, A.
Selloni, F. P. Rotzinger, M. Gratzel, Phys. Rev. Lett. 1998, 81, 2954; c) A.
Selloni, Nat. Mater. 2008, 7, 613.
[20] M. C. Biesinger, B. P. Payne, A. P. Grosvenor, L. W. M. Lau, A. R. Gerson,
R. S. C. Smart, Appl. Surf. Sci. 2011, 257, 2717.
Acknowledgements
We acknowledge the supports from the National Natural Science
Foundation of China (21506243, 21878322), the Youth Innovation
Promotion Association of Chinese Academy of Sciences, the
Science and Technology Commission of Shanghai Municipality
(19ZR1479200) and the Shanghai Municipal Human Resource and
Social Security Bureau.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
[21] V. R. Galakhov, M. Demeter, S. Bartkowski, M. Neumann, N. A.
Ovechkina, E. Z. Kurmaev, N. I. Logachevskaya, Y. M. Mukovskii, J.
Mitchell, D. L. Ederer, Phys. Rev. B 2002, 65.
[22] T. Gao, P. Norby, F. Krumeich, H. Okamoto, R. Nesper, H. Fjellvag, J. Phys.
Chem. C 2010, 114, 922.
[23] Y. Wu, Y. Lu, C. Song, Z. Ma, S. Xing, Y. Gao, Catal. Today 2013, 201, 32.
[24] J. Wang, J. Li, C. Jiang, P. Zhou, P. Zhang, J. Yu, Appl. Catal. B 2017, 204,
147.
[25] a) G. Zeng, Y. Wang, D. Gong, W. P. Y. Zhang, Y. Sun, ACS Cent. Sci. 2019,
5, 1834; b) J. Yang, D. Gong, G. Li, G. Zeng, Q. Wang, Y. Zhang, G. Liu, P.
Wu, E. Vovk, Z. Peng, X. Zhou, Y. Yang, Z. Liu, Y. Sun, Adv. Mater. 2018,
30, e1705775; c) Z. Zhu, P. Wu, G. Liu, X. He, B. Qi, G. Zeng, W. Wang, Y.
Sun, F. Cui, Chem. Eng. J. 2017, 313, 957.
Conflict of Interest
The authors declare no conflict of interest.
Keywords: Mn5O8
- Na1.1Mn4O8 biphasic-crystal · defects
formation · nanosheet-nanorod three-dimensional structure ·
formaldehyde oxidation · low temperature activity
[26] C. Julien, M. Massot, R. Baddour-Hadjean, S. Franger, S. Bach, J. P.
Pereira-Ramos, Solid State Ionics 2003, 159, 345.
[27] L. Wang, E. Guan, J. Zhang, J. Yang, Y. Zhu, Y. Han, M. Yang, C. Cen, G.
Fu, B. C. Gates, F.-S. Xiao, Nat. Commun. 2018, 9.
[1] a) Y. Huang, W. Fan, B. Long, H. Li, W. Qiu, F. Zhao, Y. Tong, H. Ji, J.
Mater. Chem. A 2016, 4, 3648; b) J. Quiroz, J.-M. Giraudon, A. Gervasini,
C. Dujardin, C. Lancelot, M. Trentesaux, J.-F. Lamonier, ACS Catal. 2015,
5, 2260.
[2] a) J. Peng, S. Wang, Appl. Catal. B 2007, 73, 282; b) C. Zhang, H. He, K.-i.
Tanaka, Catal. Commun. 2005, 6, 211; c) H. Huang, D. Y. C. Leung, ACS
Catal. 2011, 1, 348.
[3] Q. Wang, C. Zhang, L. Shi, G. Zeng, H. Zhang, S. Li, P. Wu, Y. Zhang, Y.
Fan, G. Liu, Z. Jiang, Z. Liu, Y. Sun, iScience 2018, 9, 487.
[4] X. Q. Yang, X. L. Yu, M. Y. Lin, M. F. Ge, Y. Zhao, F. Y. Wang, J. Mater.
Chem. A 2017, 5, 13799.
[28] S. Rong, K. Li, P. Zhang, F. Liu, J. Zhang, Catal. Sci. Technol. 2018, 8,
1799.
[29] Y. J. Wei, L. Y. Yan, C. Z. Wang, X. G. Xu, F. Wu, G. Chen, J. Phys. Chem. B
2004, 108, 18547.
[30] a) J. Gao, Z. Huang, Y. Chen, J. Wan, X. Gu, Z. Ma, J. Chen, X. Tang,
Chem. Eur. J. 2018, 24, 681; b) C. Zhang, P. Wu, G. Liu, Z. Zhu, G. Zeng,
ACS Appl. Nano Mater. 2019, 2, 5224.
[31] Y. Li, X. Chen, C. Wang, C. Zhang, H. He, ACS Catal. 2018, 8, 11377.
[32] Y. Yang, J. Huang, S. Wang, S. Deng, B. Wang, G. Yu, Appl. Catal. B 2013,
142, 568.
[33] a) J. Chen, D. Yan, Z. Xu, X. Chen, W. Xu, H. Jia, J. Chen, Environ. Sci.
Technol. 2018, 52, 4728; b) H. Wang, W. Guo, Z. Jiang, R. Yang, Z. Jiang,
Y. Pan, W. Shangguan, J. Catal. 2018, 361, 370; c) X. Jiang, X. Li, J. Wang,
D. Long, L. Ling, W. Qiao, Catal. Sci. Technol. 2018, 8, 2740.
[34] W. K. Dong, P. W. Seo, G. J. Kim, S. C. Hong, Appl. Catal. B 2015, 163, 436.
[35] C. Zhang, H. He, Catal. Today 2007, 126, 345.
[5] B. B. Chen, X. B. Zhu, M. Crocker, Y. Wang, C. Shi, Appl. Catal. B 2013,
132–133, 245.
[6] Y. B. Li, C. B. Zhang, H. He, J. H. Zhang, M. Chen, Catal. Sci. Technol.
2016, 6, 2289.
[7] a) Z. Dai, X. Yu, C. Huang, M. Li, J. Su, Y. Guo, H. Xu, Q. Ke, RSC Adv.
2016, 6, 97022; b) J. L. Wang, G. K. Zhang, P. Y. Zhang, J. Mater. Chem. A
2017, 5, 5719; c) S. P. Rong, P. Y. Zhang, F. Liu, Y. J. Yang, ACS Catal.
2018, 8, 3435.
[8] B. Bai, H. Arandiyan, J. Li, Appl. Catal. B 2013, 142–143, 677.
[9] X. Tang, Y. Li, X. Huang, Y. Xu, H. Zhu, J. Wang, W. Shen, Appl. Catal. B
2006, 62, 265.
[10] Z. Jin, L. Wang, Q. Hu, L. Zhang, S. Xu, X. Dong, X. Gao, R. Ma, X. Meng,
F.-S. Xiao, ACS Catal. 2018, 8, 5250.
[11] X. Du, C. Li, L. Zhao, J. Zhang, L. Gao, J. Sheng, Y. Yi, J. Chen, G. Zeng,
Appl. Catal. B 2018, 232.
[12] a) Y. Yang, Y. Z. Li, Q. Zhang, M. Zeng, S. W. Wu, L. Lan, X. J. Zhao, J.
Mater. Chem. A 2018, 6, 14195; b) J. K. He, S. Y. Chen, W. X. Tang, Y. L.
Dang, P. Kerns, R. Miao, B. Dutta, P. X. Gao, S. L. Suib, Appl. Catal. B
2019, 255, 11.
[36] Z. C. Bai, B. Sun, N. Fan, Z. C. Ju, M. H. Li, L. Q. Xu, Y. T. Qian, Chem. Eur. J.
2012, 18, 5319.
[37] S. Lu, K. Li, F. Huang, C. Chen, B. Sun, Appl. Surf. Sci. 2017, 400, 277.
Manuscript received: March 16, 2020
Revised manuscript received: April 9, 2020
Accepted manuscript online: April 15, 2020
Version of record online: ■■■, ■■■■
ChemCatChem 2020, 12, 1–12
www.chemcatchem.org
11
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

FULL PAPERS
1
2
3
4
5
6
7
8
9
Formaldehyde oxidation: Complete
C. Zhang, G. Liu, P. Wu, Prof. G. Zeng*,
Dr. Y. Sun
°
HCHO removal was achieved at 90 C
under demanding conditions over
the interconnected 3D structure of
Na1.1Mn4O8-nanosheet@ Mn5O8-
nanorod biphasic-crystals. Structure
defects on the nano-rod/-sheet
shared-edges significantly enhance
the activity.
1 – 12
Complete Formaldehyde Removal
over 3D Structured
Na1.1Mn4O8@Mn5O8 Biphasic-
Crystals
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57