The synergistic catalysis on Co nanoparticles and CoNx sites of aniline-modified ZIF derived Co@NCs for oxidative esterification of HMF

Article abstract of DOI:10.1016/j.cclet.2020.06.027

An efficient, sustainable and scalable strategy for the synthesis of porous cobalt/nitrogen co-doped carbons (Co@NCs) via pyrolysis of aniline-modified ZIFs, has been demonstrated. Aniline can coordinate and absorb on the surface of ZIF (ZIF-CoZn3-PhA), accelerate the precipitation of ZIFs, thus resulting in smaller ZIF particle size. Meanwhile, the aniline on the surface of ZIF-CoZn3-PhA promotes the formation of the protective carbon shell and smaller Co nanoparticles, and increases nitrogen content of the catalyst. Because of these properties of Co@NC-PhA-3, the oxidative esterification of 5-hydroxymethylfurfural can be carried out under ambient conditions. According to our experimental and computational results, a synergistic catalytic effect between CoN_x sites and Co nanoparticles has been established, in which both Co nanoparticles and CoN_x can activate O₂ while Co nanoparticles bind and oxidize HMF. Moreover, the formation and release of active oxygen species in CoN_x sites are reinforced by the electronic interaction between Co nanoparticles and CoN_x.

Full text of DOI:10.1016/j.cclet.2020.06.027

Journal Pre-proof
The synergistic catalysis on Co nanoparticles and CoNx sites of
aniline-modified ZIF derived Co@NCs for oxidative esterification of HMF
Tao Rui, Guo-Ping Lu, Xin Zhao, Xun Cao, Zhong Chen
PII:
S1001-8417(20)30374-0
https://doi.org/10.1016/j.cclet.2020.06.027
CCLET 5710
DOI:
Reference:
To appear in:
Chinese Chemical Letters
Received Date:
Revised Date:
Accepted Date:
7 May 2020
13 June 2020
19 June 2020
Please cite this article as: Rui T, Lu G-Ping, Zhao X, Cao X, Chen Z, The synergistic catalysis
on Co nanoparticles and CoNx sites of aniline-modified ZIF derived Co@NCs for oxidative
esterification of HMF, Chinese Chemical Letters (2020),
doi: https://doi.org/10.1016/j.cclet.2020.06.027
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
©
2020 Published by Elsevier.

Chinese Chemical Letters
journal homepage: www.elsevier.com
Communication
The synergistic catalysis on Co nanoparticles and CoN
derived Co@NCs for oxidative esterification of HMF
x
sites of aniline-modified ZIF
a
a,b
b
b
b
Tao Rui , Guo-Ping Lu , Xin Zhao , Xun Cao , Zhong Chen *
a
School of Chemical Engineering, Nanjing University of Science & Technology, Nanjing 210094, China
School of Materials Science & Engineering,Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore
b
Graphical Abstract
Aniline-modified ZIF derived Co/N codoped carbon: Co@NC-PhA-3 can achieve base-free oxidative esterification of HMF
under ambient conditions due to its higher specific surface area, the protective carbon shell of smaller Co NPs and higher N
content. Furthermore, a synergistic catalytic effect between CoN sites and Co NPs has also been disclosed.
x

Corresponding authors.
E-mail addresses: glu@njust.edu.cn (G. Lu); ASZChen@ntu.edu.sg (Z. Chen)
ARTICLE INFO
Article history:Received 7 May 2020 Received in revised form 15 June Accepted 18 June Available online
ABSTRACT
An efficient, sustainable and scalable strategy for the synthesis of porous cobalt/nitrogen co-doped carbons
(Co@NCs) via pyrolysis of aniline-modified ZIFs, has been demonstrated. Aniline can coordinate and
absorb on the surface of ZIF (ZIF-CoZn3-PhA), accelerate the precipitation of ZIFs, thus resulting in
smaller ZIF particle size. Meanwhile, the aniline on the surface of ZIF-CoZn3-PhA promotes the formation
of the protective carbon shell and smaller Co nanoparticles, and increases nitrogen content of the catalyst.
Because of these properties of Co@NC-PhA-3, the oxidative esterification of 5-hydroxymethylfurfural can
be carried out under ambient conditions. According to our experimental and computational results, a
synergistic catalytic effect between CoN
x
sites and Co nanoparticles has been established, in which both Co
—
——

2
nanoparticles and CoN
formation and release of active oxygen species in CoN
between Co nanoparticles and CoN
x
can activate O
2
while Co nanoparticles bind and oxidize HMF. Moreover, the
x
sites are reinforced by the electronic interaction
x
Keywords:Aniline-modified ZIF Porous cobalt/nitrogen co-doped carbon Synergistic catalysis on CoN
Hydroxymethylfurfural
x
sites and Co nanoparticles Oxidative esterification 5-
5
-Hydroxymethylfurfural (HMF) derived from carbohydrates or cellulose by dehydration, has been used to bridge the gap between
lignocellulosic biomass and some useful chemicals [1-3]. One particularly useful transformation of HMF is its oxidation to 2,5-
furandicarboxylicacid (FDCA), one of the top-10 biobased products from biorefinery carbohydrates [4]. FDCA can be used as a
monomer for biopolymers that can replace the fossil fuel-derived polyethylene terephthalate (PET) [5]. However, there are difficulties
in its direct application in industry due to its low solubility in most solvents and lack of economic methods for its purification [6]. 2,5-
Furan dicarboxylicacid dimethyl ester (FDMC) has good solubility in lots of solvents, which can also be used directly to synthesize
biopolymers through transesterification reaction [7]. Thus, FDMC is an ideal alternative to FDCA.
This transformation is usually catalyzed by noble metals (Au, Pd) with the help of high pressure or excess bases [8-11]. A few
earth-abundant-metal catalysts have also been explored, but high temperature and O
poor in some cases [12-14]. To further optimize the reaction conditions, some multi-metal catalysts including Co
PdCoBi@C, AuPd-Fe have been developed [15-17]. More recently, Co,N-codoped carbon materials were foundto be a kind of
2
pressure are needed, and the selectivity is still
x y
O -N@C/Ru@C,
3
O
4
efficient earth-abundant metal catalysts for the oxidative esterification of HMF under base-free conditions [18-22]. Lin and Zeng
suggested that cobalt nanoparticles (Co NPs) activated by the Mott-Schottky effect may be the main catalytic active sites [21,22]. In
general, there are two potential catalytic active centers: Co NPs and CoN
concept of “active doughnut” in which the Co NPs and CoN activate O while Co NPs bind and oxidize alcohol [23,24]. Meanwhile,
Hu has demonstrated that the electronic interactions between Fe NPs and FeN coordination structure in Fe@NCs favor the adsorption
of O [25]. Along this line, we envisage that there is a synergistic catalytic effect between CoN sites and Co NPs of Co@NCs for
oxidative esterification of HMF, which hasnot been addressed by any reports so far.
x
sites in Co@NCs. Eisenberg et al. have proposed a new
x
2
4
2
x
Compared with other materials, zeolitic imidazolate frameworks (ZIFs) serve as ideal precursors for the formation of Co@NCs
owing to their low-costraw materials, mild synthetic conditions, high nitrogen content, and tunable and uniform pore apertures [26].
Zinc as a self-sacrificialtemplate is also introduced to eliminate the aggregation of Co nanoparticles and improve the specific surface
areas of Co@NCs during the pyrolysis of ZIFs [27,28]. In addition, the synthesis of ZIFs can be achieved in water, which not only
avoids the use of toxic organic solvents, but also helps to save costs of ZIFs [29-33]. There are a few scattered reports mentioning that
the reactant concentration and the use of modulators (amino acids, butylamine, CTAB) can affect the particle size and morphology of
ZIFs [34-36]. Accordingly, we expect that (1) the use of different amines as the modulators may change both morphology and size of
ZIFs in water; (2) amines may adsorb on the surface of ZIFs to form a protective carbon shell after carbonization; (3) some of the
Co@NCs derived from amine-modified ZIFs may possess excellent catalytic performance for oxidative esterification of HMF due to
their controllable size and morphology or protective carbon shell.
With our interestsin exploring the novel ZIF-derived Co@NCs for organic transformations [19,28,37], we demonstrate an efficient,
sustainable and scalable strategy for the synthesis of Co@NCs, in which ZIFs as the precursors, are prepared in water solution using
aniline as the modulators. Aniline can significantly affect the size of the ZIFs by adsorbing on the surface of ZIFs and increasing the
precipitation of ZIFs. Co@NC-PhA-3 exhibits excellent catalytic performance for oxidative esterification of HMF, which is ascribed
to its high BET surface areas, the existence of carbon shell on smaller Co NPs, higher N content and synergistic catalytic effect
x
between its CoN sites and Co NPs.
We started the investigation by employing different amines in the synthesis of ZIF-CoZnX-Y (X denotes the mole ratio of Zn/Co, Y
the type of amine shown in Fig. S1 in Supporting information) in water. These ZIFs were then pyrolyzed at 900 °C with a heating rate
of 5 °C/min for 2 h under pure Ar atmosphere to afford the corresponding Co@NC-Y-X (Fig. S1). Generally, the yields of ZIFs in
water are higher than that in MeOH. For example, the yield of ZIF-CoZn3-PhA is higher than 450 mg (2 mmol scale), but the yield of
ZIF-CoZn3-M (Table 1) is less than 250 mg even with prolonged reaction time to 24 h.
As shown in Fig.1, spherical ZIFs were obtained in the presence of amine or in methanol with different particle sizes (Figs. 1a-d
and f), while irregular flake and rod ZIFs were yielded in the absence of amine (Fig. 1e). The order of the particle sizes of ZIFs is ZIF-
CoZn3-PhA < ZIF-CoZn3-OA < ZIF-CoZn3-nBuA < ZIF-CoZn3-BnA < ZIF-CoZn3-M (Fig. 1, Table S1 in Supporting information).
Thus, it can be concluded that the type of amines plays a crucial role on the size of ZIFs in water. It is well known that all particle ZIFs
are formed by self-assembly of sodalite cages [38]. It is reasonable to assume that amine-modified ZIFs are also derived by the same
way. To verify this assumption, the products ZIF-CoZn3-PhA formed after 5 min, 10 min, 30 min, 1 h, 4 h of reaction were
investigated (Figs. S2a-e, S3a in Supporting information). It is found that (1) the nanoparticles (< 10 nm) as shown in Figs. S2a-e
prove the self-assembly process; (2) the self-assembly is a fast process which can be completedin 5 min because the precipitation is
highly accelerated by the use of external base (amines) [39], which may be also a reason for the smaller ZIF particles; (3) the particle
size and structure have no obvious changes over time.

Fig. 1. SEM images of (a) ZIF-CoZn3-PhA, (b) ZIF-CoZn3-OA, (c) ZIF-CoZn3-nBuA, (d) ZIF-CoZn3-BnA, (e) ZIF-CoZn3, (f) ZIF-CoZn3-
M.
Fig. 2. The effect of preparation process parameters of Co@NCs: (a) amine type and (b) the molar ratio of Zn/Co on catalytic performance for
oxidative esterification of HMF. Conditions: HMF 0.3 mmol, Co@NCs (a) 10 mg or (b) 20 mg, MeOH 3 mL, O 1 atm, 25 °C, (a) 24 h or (b)
2
4
8 h; GC yields with anisole as an internal standard.
According to the powder XRD patterns (Fig. S4a in Supporting information), six peaks belonging to the {100}, {11-1}, {2-10},
200}, {2-11}, {111} planes were detected in the ZIF-CoZn3-M (polyhedron, Fig. 1f) [29]. Additionalpeaks were present in the cases
{
of ZIF-CoZn3 (flat rod, Fig. 1e), ZIF-CoZn3-Y (sphere, Figs. 1a-d). Similarly, six peaks were found in ZIF-CoZn3-Y, but the peak
intensity is different from the ZIF-CoZn3-M. Therefore, we demonstrate that amines may change the ratio of crystal faces in ZIFs by
coordination with metal ions or absorption on the surface of ZIFs, modifying their size and morphology [35,36,39-41].
As shown in IR spectra, signals of amine (aniline and oleylamine) were observed in both ZIF-CoZn3-PhA and ZIF-CoZn3-OA (Fig.
S4b in Supporting information), while benzylamine and n-butylamine were removed from corresponding ZIFs owing to their good

4
solubility in water (Fig. S3a). The poor solubility of aniline and oleylamine results more amines absorbed on the surface of ZIFs,
which may suppress ZIF particle growth and result in smaller ZIF particles than ZIF-CoZn3-BnA and ZIF-CoZn3-nBuA. Furthermore,
the red shift of N-H stretch of aniline (Fig. S4b in Supportinginformation) suggests the metal ions (Co2+ or Zn ) may coordinate with
aniline, which explains why ZIF-CoZn3-PhA has the smallest ZIF particle size.
2+
The topography of amine-modified ZIFs derived Co@NC-Y-3 was also assessed by SEM observation. The morphology of ZIFs can
be preserved with partial shrinkage after calcination (Figs. S5a-d in Supporting information). In the cases of Co@NC-3 and Co@NC-
M-3 (Figs. S5e and f in Supporting information), the structure of ZIFs is destroyed after sintering. The high treatment temperature (900
o
C) is necessary, because it not only promotes the evaporation of Zn, resulting in higher surface areas of Co@NCs, but also facilitates
the formation of carbon with high graphitizationthat is beneficial to the electronic interactions in Co@NCs.
After the successful preparation of these Co@NCs, we investigated the catalytic performance of these materials for oxidative
esterification of HMF (Fig. 2). It can be concluded that the use of amine type and the molar ratio of Zn/Co in the synthetic process
have an obvious influence on the catalytic performance of Co@NCs. Moderate yields of FDMC was achieved at room temperature
under air atmosphere (1 atm) when Co@NC-PhA-3 and Co@NC-OA-3 were employed as thecatalysts. Excellent yields of FDMC
(92%) could be realized by increasing the amount of catalyst to 20 mg and prolonging the reaction time to 48 h.
Compared with previous works (Table S2), Co@NC-PhA-3 exhibited superior catalytic performance. To the best of our knowledge,
this is the first example on the oxidative esterification of HMF under ambient conditions using a non-noble metal catalyst.
Encouraged by the excellent performance of Co@NC-PhA-3, we carried on to investigate the structure-effect relationship between
Co@NCs and the reactions.
As shown in Fig. 3, both amine type and molar ratio of Zn/Co have significant influences on the BET surface areas of Co@NCs,
which affect the yield of FDMC. With smaller particle size of Co@NCs, the corresponding higher BET surface area can enhance the
yield of FDMC owing to the improvement of the interaction between catalyst and reactant (Fig. 3a). Increasing the amount of zinc
results in larger BET surface area due to the evaporation of Zn (Fig. 3b), but it also reduces the amount of Co inCo@NCs (Table S3 in
Supporting information). Among all specimens, Co@NC-PhA-3 provides the best result.
Fig. 3. The relationship between yields (purple points) of FDMC (or BET surfaces areas of Co@NCs (brown square)) with different
preparation process parameters of Co@NCs: (a) amine type; (b) molar ratio of Zn/Co. Conditions: HMF 0.3 mmol, Co@NCs (a) 10 mg or (b)
2
2
0 mg, MeOH 3 mL, O 1 atm, 25 °C, (a) 24 h or (b) 48 h; GC yields with anisole as an internal standard.

Fig. 4. TEM images of (a) Co@NC-PhA-3, (b) Co@NC-OA-3, (c) Co@NC-3, (d) Co@NC-M-3, (e) Co@NC-PhA-Co, (f) Co@NC-PhA-9.
The TEM images of Co@NCs are shown in Fig. 4. The average Co nanoparticle size of Co@NC-PhA-3 is the smallest than all
other Co@NCs (Figs. 4a-e, Table S4 in Supporting information). No obvious Co nanoparticles are found in Co@NC-PhA-9 (Fig. 4f)
owing to its low Co content (Table S3).
A carbon shell on Co nanoparticles was observed in the case of Co@NC-PhA-3 (Fig. 4a). It may be derived from the mixture of
aniline and 2-methylimidazole. Its presence is conducive to the effective protection of active cobalt nanoparticles from leaching, and
inhibitive to the aggregation of cobalt nanoparticles. In the view of BET and TEM results, Co@NC-PhA-Co has low BET surface area
(
Fig. 3b) and large Co NPs (Fig. 4e), thus it shows no reactivity.
The HRTEM images further verify the existence of the carbon shells (Figs. 5a and b). The clear lattice fringes of 0.205 nm, 0.215
nm and 0.356 nm were observed from the HRTEM image of Co@CN-PhA corresponding to Co (111), Co (100) and graphite
respectively (Fig. 5b) [42,43]. The results of TEM-EDS indicate that Co@NC-PhA-3 mainly contains C, O, N, Zn and Co elements,
and most of cobalt element is from Co NPs. (Fig. 5c). The residual zinc in the catalyst can increase Lewis basic sites and decrease the
strong Lewis acidic sites, which can improve the performance of the catalyst [20].

6
Fig. 5. (a), (b) HRTEM images of Co@NC-PhA-3; (c) TEM and its corresponding element mapping images of Co@NC-PhA-3.
In order to further check the elemental composition and chemical state, the XRD (Fig. S6 in Supporting information) and XPS
o
o
o
results of Co@NCs (Fig. 6) were studied. Characteristic diffraction peaks observed at 44.4 , 51.6 , and 75.9 could be attributed to fcc-
structured Co(111), Co(200) and Co(220) planes respectively [43]. The weak diffraction peaks of hcp-structured Co(100) (or Co C)
2
o
o
and Co(101) correspond to the shifts in 41.5 and 48.4 [44,45]. No obvious diffraction peak of metal is found in the cases of Co@NC-
PhA-9 owing to its low Co content. It can also be found that the intensity of Co signal is inversely proportional to the Zn/Co molar
ratio. Meanwhile, the structure of Co NPs tends to disorder by increasing the Zn/Co molar ratio.
The XPS also proves that Co@NC-PhA-3 mainly contains C, O, N, Zn and Co elements (Fig. S7 in Supporting information). CoN
sites exist in the material basedonthe N 1s region of XPS spectra (Fig. 6b) [23]. Because cobalt phthalocyanine(PcCo) hasCoN sites,
the single of CoN can be further confirmed by comparison of XPS N 1s region ofPcCo and Co@NC-PhA-3 (Fig. S8b in Supporting
x
4
x
information) [25]. According to XPS results, the N content of Co@NC-PhA-3 is higher than Co@NC-M-3 (Table S5 in Supporting
information), presumably owing to the presence of aniline in the surface of ZIF-CoZn3-PhA, which is not only an N source but also
promote the formation of carbon shell around Co NPs. The high N content can enhance the Co NPs performance by the Mott-Schottky
effect for the oxidative esterification of 5-hydroxymethylfurfural [21,22].
To clarify the catalytically active sites of Co@NC-PhA-3 for this reaction, several controlled experiments were carried out (Fig. 7).
Firstly, the catalyst was leached with acid (1 mol/L HCl, 120 °C, 12 h) to remove Co NPs, denoted as Co@NC-PhA-3A. The TEM and
XRD results suggest that no obviously Co NPs are found in Co@NC-PhA-3A (Fig. S9 in Supporting information). A remarkable
decrease in yield of FDMC (19%) was observed using Co@NC-PhA-3A as the catalyst, which suggests Co NPs in Co@NC-PhA-3 are
-
necessary for the reaction. It is well known that SCN can poison MN
x
sites in catalysis [25,46]. The yield of FDMC was poor (29%)
sites can boost the reaction.
when both Co@NC-PhA-3 and NaSCN (10 mol%) were used, indicating that CoN
x

Fig. 6. XPS spectra of (a) Co 2p3/2 and (b) N 1s of Co@NC-M-3, Co@NC-PhA-3, Co@NC-PhA-9.
Notably, naked Co NPs and Co@AC as the analogs of Co NPs in Co@NC-PhA-3 failed to yield the final product. AsshowninFig.
S10 (Supporting information), the naked Co NPs have irregular shape and large size (> 150 nm), and the Co NPs mainly on the surface
of AC are easy to aggregate and disappear in the case of Co@AC, which can explain the poor catalytic performance of naked Co NPs
x
and Co@AC.Moreover only trace of FDMC was obtained when PcCo was used as the analog of CoN in Co@NC-PhA-3. These
results demonstrate the crucial roles of N-doped porous carbon with high graphitization for the reaction, which may promote the
absorption and activation of reactants, improve the catalytic performance of Co sites, and enhance the interaction between Co sites and
reactants[18-20].
Based on these results, we can safely conclude that there is a synergistic catalytic effect between CoN
x
sites and Co NPs in
Co@NC-PhA-3 for the reaction. According to the XPS results (Fig. 6a, Fig. S8 and Table S6 in Supporting information), The binding
II
III
energy of Co and Co in Co@NC-PhA-3 shift to a lower value compared with PcCo and Co@NC-M-3, implying Co NPs transfer the
electron to CoN
x
.
Fig. 7. Comparison of catalytic performance over different catalysts for the oxidative esterification of HMF. Co@NC-PhA-3A means
Co@NC-PhA-3 leaching with acid; Co@NC-PhA-3S means the use of Co@NC-PhA-3 with NaSCN (10 mol%); Co@AC meansactive
carbon supported Co nanoparticles; PcCo means cobalt phthalocyanine; Co NPs means naked Co nanoparticles.

8
Three simple models CoN
respectively (Fig. 8). The electron number of Co in CoN
adding of Co to CoN can increase the electron number of Co in CoN
To further verify the synergistic catalytic effects of CoN sites and Co NPs, the adsorption energies (Eads) of 2-hydroxymethylfurane
FM) and O on different Co sites in all three models are calculated (Fig. 9). Cased on these calculations, we are able to conclude that
1) Both the Eads of O and FM on Co are stronger than CoN . (2) Too strong Eads of O on Co may be harmful to the release of
. Thus CoN may be the better sites for the activity of O due to its weak Eads
can enhance the electron transfer from CoN to O and reduce the Eads of O
which are beneficial to the formation and release of active oxygen species [22,47]. (4) The huge difference in Eads of FM between
Co and CoN suggests that the absorption of FM mainly takes place on Co
4
, Co
6
O
2
and CoN
4
-Co
6
O
2
were established which represent CoN
was calculated from baderde composition of charge density. It is found that
, thus supporting that Co NPs transfer the electron to CoN
x
, Co NPs and Co@NC-PhA-3
4
6
O
2
4
4
x
.
x
(
(
2
2
6
O
2
4
2
6 2
O
active oxygen species, resulting in the oxidation of Co
6
O
2
4
2
of O [25]. (3) The electron transfer from C to CoN
2
6
O
2
4
4
2
2
6
O
2
4
6 2
O .
4 6 2 4 6 2 4
Fig. 8. Three calculation models of (a) CoN , (b) Co O and (c) CoN -Co O and electron number of Co in CoN . Brown is C; red is O; blue
is Co; silver is N.
Considering all above results, the tentative synergistic catalysis by CoN
rate determination step in the oxidative esterification of HMF [18-20], was illustrated in Fig. S11 (Supporting information). Both Co
NPs and CoN in Co@NC-PhA-3 can activate O while Co NPs bind and oxidize alcohol owing to that the absorption of alcohol
x
and Co NPs for α-C-H dissociation of alcohol that is the
x
2
mainly takes place on CoNPs. Meanwhile, the electron transfer from Co NPs to CoN
active oxygen species.
x
is beneficial to the formation and release of
Fig. 9. The adsorption energies of 2-hydroxymethylfurane (FM) and O
Co , (d) Co of CoN -Co ; O on (e) CoN , (f) Co , (g) CoN
is Co; silver is N.
2
on different Co sites. FM on (a) CoN
4
, (b) Co
6 2 4 4
O , (c) CoN of CoN -
6
O
2
6
O
2
4
6
O
2
2
4
6
O
2
4
of CoN -Co , (h) Co of CoN -Co O
4
6
O
2
6
O
2
4
6 2
. Brown is C; red is O; blue
The reusability of Co@NC-PhA-3 was also been studied (Fig. S13ain Supporting information). This material could be reused for at
least six times without significant loss of catalytic activity. The ICP, TEM, SEM and XPS results of Co@NC-PhA-3 after six cycles
(
Fig. S12 and Table S3 in Supporting information) suggest that (1) the cobalt leaching was negligible; (2) the shape and size of carbon
and cobalt particles were well reserved; (3) no obvious valence change on the surface of Co inCo@NC-PhA after six cycles
wasobserved. The heterogeneous nature of the catalyst was demonstrated by a filtration experiment (Fig. S13b in Supporting
information). The catalyst was filtered off after 12 h at room temperature, and the isolated solution was allowed to react for a further
3
6 h. No further increase in yield was observed.
In summary, we have developed an efficient, sustainable and scalable strategy for the synthesis of porous Co@NCs using aniline-
modified ZIFs as the precursor. Aniline can coordinate and absorb on the surface of ZIFs, and accelerate the precipitation of ZIFs, by
which smaller particle size of ZIF (ZIF-CoZn3-PhA) are obtained. Notably, several excellent properties of Co@NC-PhA-3 including

higher specific surface area, the protective carbon shell of smaller Co NPs and higher the nitrogen content.Theseare attributed to
aniline on the surface of ZIF-CoZn3-PhA, and can enhance the performance of the catalyst. Co@NC-PhA-3 exhibits promising
performance and recyclability for base-free oxidative esterification of HMF. Based on our characterization, experiments and
calculations, a synergistic catalytic effect between CoN
while Co NPs bind and oxidize HMF. Meanwhile, the electron transfer from Co NPs to CoN
of active oxygen species.
x
sites and Co NPs has been disclosed. Both Co NPs and CoN
x
activate O
x
is beneficial to the formation and release
2
Acknowledgments
We gratefully acknowledge the Fundamental Research Funds for the Central Universities (No. 30920021120) and Key Laboratory
of Biomass Energy and Material, Jiangsu Province (No. JSBEM201912) for financial support. This work was a project funded by the
Priority Academic Program development of Jiangsu Higher Education Institution. We also thank Analysis and Test Center Nanjing
University & Technology for the help in obtaining the IR and XRD date.
References
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
1] Y. Romn-Leshkov, C.J. Barrett, Z.Y. Liu, Nature 447 (2007) 982-986.
2] C. Liu, M. Wei, J. Wang, J. Xu, J. Jiang, K. Wang, ACS Sustain. Chem. Eng. 8 (2020) 5776-5786.
3] X. Kong, Y. Zhu, Z. Fang, et al., Green Chem. 20 (2018) 3657-3682.
4] Z. Zhang, K. Deng, ACS Catal. 5 (2015) 6529-6544.
5] F.A. Kucherov, E.G. Gordeev, A.S. Kashin, V.P. Ananikov, Angew. Chem. Int. Ed. 56 (2017) 15931-15935.
6] I. Delidovich, P.J. Hausoul, L. Deng, et al., Chem. Rev. 116 (2016) 1540–1599.
7] C.F. Araujo, M.M. Nolasco, P.J.A. Ribeiro-Claro, et al., Macromolecules 51 (2018) 3515–3526.
8] E. Taarning, I.S. Nielsen, K. Egeblad, R. Madsen, C.H. Christensen, ChemSusChem 1 (2008) 75-78.
9] O. Casanova, S. Iborra, A. Corma, J. Catal. 265 (2009) 109-116.
10] F. Menegazzo, T. Fantinel, M. Signoretto, F. Pinna, M. Manzoli, J. Catal. 319 (2014) 61-70.
11] F. Menegazzo, M. Signoretto, D. Marchese, F. Pinna, M. Manzoli, J. Catal. 326 (2015) 1-8.
12] J. Deng, H.J. Song, M.S. Cui, Y.P. Du, Y.Fu, ChemSusChem 7 (2014) 3334-3340.
13] Y. Sun, H. Ma, X. Jia, et al., ChemCatChem 8 (2016) 2907-2911.
14] X. Tong, L. Yu, H. Chen, et al., Catal. Commun. 90 (2017) 91-94.
15] A. Cho, S. Byun, J.H. Cho, B.M. Kim, ChemSusChem 12 (2019) 2310-2317.
16] F. Li, X.L. Li, C. Li, J. Shi, Y. Fu, Green Chem. 20 (2018) 3050-3058.
17] A. Salazar, P. Hunemorder, J. Rabeah, et al., ACS Sustain. Chem. Eng. 7 (2019) 12061-12068.
18] H. Zhou, S. Hong, H. Zhang, et al., Appl.Catal. B: Environ. 256 (2019) 117767.
19] K.K. Sun, S.J. Chen, Z.L. Li, G.P. Lu, C. Cai, Green Chem. 21 (2019) 1602-1608.
20] Y. Lin, G.P. Lu, X. Zhao, et al., Mole. Catal. 482 (2020) 110695.
21] H. Su, K.X. Zhang, B. Zhang, et al., J. Am. Chem. Soc. 139 (2017) 811-818.
22] Y. Feng, W. Jia, G. Yan, et al., J. Catal. 381 (2020) 570-578.
23] D. Eisenberg, T.K. Slot, G. Rothenberg, ACS Catal. 8 (2018) 8618-8629.
24] T.K. Slot, D.Eisenberg, G. Rothenberg, ChemCatChem 10 (2018) 2119-2124.
25] W.J. Jiang, L. Gu, L. Li, et al., J. Am. Chem. Soc. 138 (2016) 3570-3578.
26] G. Zhong, D. Liu, J. Zhang, J. Mater. Chem. A 6 (2018) 1887-1899.
27] Y.Z. Chen, C. Wang, Z.Y. Wu, et al., Adv. Mater. 27 (2015) 5010-5016.
28] K.K. Sun, S. Chen, J. Zhang, G.P. Lu, C. Cai, ChemCatChem 11 (2019) 1264-1271.
29] D.W. Lewis, A.R. Ruiz-Salvador, A. Gómez, et al., CrystEngComm 11 (2009) 2272-2276.
30] D. Yu, L. Ge, B. Wu, et al., J. Mater. Chem. A 3 (2015) 16688-16694.
31] Q. Liao, M. He, Y. Zhou, et al., ACS Appl. Mater. Inter. 10 (2018) 29136-29144.
32] G. Fang, J. Zhou, C. Liang, et al., Nano Energy. 26 (2016) 57-65.
33] Y. Pan, Y. Liu, G. Zeng, L. Zhao, Z. Lai, Chem. Commun. 47 (2011) 2071-2073.
34] Y. Deng, Y. Dong, G. Wang, et al., ACS Appl. Mater. Inter. 11 (2017)9699-9709.
35] K. Liang, R. Ricco, C.M. Doherty, M.J. Styles, P. Falcaro, CrystEngComm 18 (2016) 4264-4267.
36] Y. Pan, D. Heryadi, F. Zhou, et al., CrystEngComm 13 (2011) 6937.
37] K.K. Sun, J.L. Sun, G.P. Lu, C. Cai, Green Chem. 21 (2019) 4334-4340.
38] J. Zhang, T. Zhang, D. Yu, K. Xiao, Y. Hong, CrystEngComm 17 (2015) 8212-8215.
39] C. Avci, J. Ariñez-Soriano, A. Carné-Sánchez, et al., Angew. Chem. Int. Ed. 54 (2015) 14417-14421.
40] T. Tsuruoka, S. Furukawa, Y. Takashima, et al., Angew. Chem. 48 (2009)4739-4743.
41] N. Stock, S. Biswas, Chem. Rev. 112 (2012)933-969.
42] B. Liu, L. Jin, H. Zheng, et al., ACS Appl. Mater. Inter. 9 (2017)1746-1758.
43] V.L. Kuznetsov, I.L. Zilberberg, Y.V. Butenko, A.L. Chuvilin, B. Segall, J. Appl. Phys. 86 (1999) 863-870.
44] J.C. Wang, R.F. Nie, L. Xu, X.L. Lyu, X.Y. Lu, Green Chem. 21 (2019) 314-320.

1
0
[45] Z. Zhao, W. Lu, R. Yang, et al., ACS Catal. 8 (2018) 228-241.
[46] Z. Ma, T. Song, Y. Yuan, Y. Yang, Chem. Sci. 10 (2019) 10283-10289.
[47] Q. Gu, P. Sautet, C. Michel. ACS Catal. 8 (2018) 11716-11721.
Table 1
The naming rules and correspondence of ZIFs and Co@NCs.
Entry ZIFs
Co@NCs
Co@NC-Y-X
Co@NC-PhA-Co No use Zn(NO
NC-PhA
Co@NC-M-3
Co@NC-3
Note
/
a
1
2
3
4
5
ZIF-CoZnX-Y
ZIF-Co-PhA
ZIF-Zn-PhA
ZIF-CoZn3-M
ZIF-CoZn3
3
)
2
2
.6H O
No use Co(NO
/
No use of amine
3
)
2
2
.6H O
b
aXdenotes the mole ratio of Zn(NO)
6H O/Co(NO) 6H O (Zn/Co), Y denotes the type of amine(aniline = PhA, benzylamine = BnA,
3
2
3
2
oleylamine = OA, n-butylamine = nBuA).
b
M means the synthesis of ZIFs in MeOH.