Atomic zinc dispersed on graphene synthesized for active CO2 fixation to cyclic carbonates

Article abstract of DOI:10.1039/c8cc09449g

CO₂ fixation to cyclic carbonates is important but depends on the catalyst. Here, atomic zinc (1.62 at%) dispersed on graphene was synthesized as a high-performance heterocatalyst for the cycloaddition reaction of epoxides and CO₂. H

Full text of DOI:10.1039/c8cc09449g

ChemComm
View Article Online
View Journal
COMMUNICATION
Atomic zinc dispersed on graphene synthesized
for active CO2 fixation to cyclic carbonates†
Cite this:DOI: 10.1039/c8cc09449g
Congwei Wang, ‡*a Qingwen Song, ‡*a Kan Zhang,a Ping Liu,a Junying Wang,a
a
Jianmei Wang,b Hengxuan Zhangb and Junzhong Wang
Received 28th November 2018,
Accepted 4th January 2019
DOI: 10.1039/c8cc09449g
rsc.li/chemcomm
CO2 fixation to cyclic carbonates is important but depends on the activity still hinder the exploration of next-generation heterogeneous
catalyst. Here, atomic zinc (1.62 at%) dispersed on graphene was catalysts for efficient CO2 fixation. Moreover, the broad size dis-
synthesized as a high-performance heterocatalyst for the cyclo- tributions and irregular morphology of the aforementioned catalysts
addition reaction of epoxides and CO2. High yield (99%) and high could bring about multiple active sites with different performances,
selectivity (98%) of propylene carbonate with a TOF of 2889 hÀ1 resulting in unwanted by-products and an ambiguous catalytic
were achieved. [ZnN3.76Æ0.2] was the active site, which was proved mechanism.9 Therefore, developing a robust supported catalyst with
by advanced characterization, including synchrotron XANES, EXAFS efficient activity, uniform active sites, economic metal loading and
and XPS and comparative performance tests.
gentle catalytic conditions for chemical transformation is imperative
but challenging.
The rapidly increasing level of carbon dioxide (CO2) in the atmo-
Single atom catalysts (SACs), in which metal atoms are dis-
sphere has raised global environmental and societal concerns. persed on the support at an atomic level, featuring a theoretical
Tremendous efforts have been devoted to CO2 fixation, regardless maximized 100% dispersion and extremely high efficiency of
of the debate about the relationship between CO2 level and global metal utilization, have triggered vast attention and been applied
warming. Indeed, the nontoxic, nonflammable, abundant and in many heterogeneous reactions.10 Significantly, compared with
renewable C1 feedstock features CO2 as a green carbon source for conventional particle catalysts with different activities brought
organic transformations.1 Cyclic carbonates are industrially impor- about by the diverse reactive sites,9 the uniform active sites with
tant as polar aprotic solvents and electrolytes in batteries, and as higher selectivity of SACs offer great promise for potentially
valuable intermediates for dimethyl carbonate, polycarbonates and making the catalytic conditions gentler with increasing energy
other fine chemicals. Currently, the commercial production of cyclic efficiency. Recently, atomically dispersed transition metals have
carbonates is based on the catalyzed cycloaddition of epoxides and exhibited advances for the electrochemical reduction of CO2 to
renewable CO2.2 Among the developed methods, greater attention CO.11 To the best of our knowledge, the use of a SACs-catalyzed
has been paid to heterogeneous catalytic schemes than to homo- strategy in the conversion of CO2 to value-added chemicals has
geneous strategies due to their low cost, easy separation and not been reported. Herein, an unprecedented atomic Zn coordi-
recyclability benefits.3 Traditional well-defined heterogeneous cata- nated@N doped graphene (NG–aZnN) SAC was fabricated and
lysts mainly include metal-containing salts,4 poly(ionic liquid) and exhibited remarkable activity for fixing CO2 to cyclic carbonates.
ionic liquid/metal/amino-supported types (with various supports: The NG–aZnN presents a novel and superior catalytic species for
MOFs,5 polymers,6 silica,7 and other materials8). Although great CO2 conversion via activating the epoxides and further catalyzing
efforts have been made, their complicated preparations and limited the reaction smoothly, with excellent TON and TOF values. The
unsaturated local chemical environment of atomic Zn, coordi-
nated with doped nitrogen atoms, was carefully characterized
through electron microscopy and synchrotron spectroscopy,
a CAS Key Laboratory of Carbon Materials, State Key Laboratory of Coal
Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan,
unveiling the key active sites for CO2 fixation to cyclic carbonates.
Initially, Zn-based SAC was prepared as shown in Fig. 1a.
Electrochemically exfoliated graphene (G) (Fig. S1, ESI†) was mixed
with zinc–oleate complex and melamine. The N atoms would be
doped on planar graphene to provide additional electron lone pairs
to anchor and stabilize the metal ions in a mononuclear organo-
metallic complex during pyrolysis, as previously reported.10b Due to
030001, P. R. China. E-mail: wangcongwei@sxicc.ac.cn, songqingwen@sxicc.ac.cn
b Institute of Mining Technology, Key Laboratory of In situ Property-improving
Mining of Ministry of Education, Taiyuan University of Technology, Taiyuan,
030024, P. R. China
† Electronic supplementary information (ESI) available: Details of general experi-
mental methods, characterization details, supplementary experimental results
and NMR spectra of the products. See DOI: 10.1039/c8cc09449g
‡ These authors contributed equally to this work.
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun.

View Article Online
Communication
ChemComm
Fig. 2 (a) HAADF-STEM images of NG–aZnN, (b) Zn K-edge XANES
spectra of Zn foil, ZnO and NG–aZnN; the inset shows the simulated Zn
configuration, (c) Fourier transform k3-weighted w(k)-function of the
EXAFS spectra, (d) first shell fitting of NG–aZnN and corresponding EXAFS
fitting curves at R space and k space (inset).
Fig. 1 (a) Schematic graph showing the synthesis of NG–aZnN, (b) TEM
and (c) STEM images of NG–aZnN, and EDS colour mapping of (d) C, (e) O,
(f) N and (g) Zn, respectively.
the limited doping level in graphene, the excessive metal ions would
inevitably aggregate and form Zn-based nanoparticles (NG–ZnO).
After acid leaching to remove oxide species and impurities, the k3-weighted EXAFS curves are extracted to probe the local structure.
novel single atom catalyst, namely the atomic Zn coordinated@N
doped graphene (NG–aZnN), was fabricated.
Transmission electron microscopy (TEM) was used to charac-
terize the obtained NG–aZnN catalyst, as shown in Fig. 1b, where
The FT-EXAFS curves show that the Zn atom in NG–aZnN displays
a different local structure with neither metallic Zn nor ZnO.
NG–aZnN did not present any prominent peaks at the positions
of Zn–Zn (2.27 Å) or Zn–O (1.63 Å). Remarkably, a distinctive first
hollow graphitic shells were dispersed on nitrogen doped graphene coordination shell scattering peak at 1.56 Å for potential Zn–N
(NG). The graphitic degree of raw G and NG–aZnN is shown in bonding could be labelled, implying that most Zn atoms in
Fig. S3 (ESI†). The absence of large nanoparticles validates the NG–aZnN form Zn–N moieties, which is consistent with the
successful removal of excess Zn-based nanoparticles. To further previous EDS obervation.12a EXAFS spectra fitting was performed
verify the composition of NG–aZnN, energy-dispersive spectro-
scopic (EDS) mapping was performed, as shown in Fig. 1c–g. The
chemical mapping confirmed that the bright spots correspond to
the presence of C, N, O and Zn, respectively. Significantly, the
to extract the structural parameters and obtain the quantitative
chemical configuration of the Zn atoms (Fig. 2d). The obtained
coordination number of the Zn atom is about 3.76 Æ 0.2 for
NG–aZnN with an R factor of 0.006 (Table S1, ESI†). Moreover, the
overlapping distribution of Zn and N signals suggests that Zn absence of Zn–Zn in NG–aZnN is in good agreement with TEM and
atoms could be surrounded by doped N atoms, which would be HAADF-STEM results that show that all the Zn species are atom-
further verified via synchrotron spectroscopy. The detailed ically coordinated without any particle aggregation. The consis-
chemical composition was further obtained by XPS, ICP-OES and tency of the EXAFS spectra and the corresponding fittings shown in
elemental analysis (Tables S1 and S2, ESI†). The atomic Zn loading
in NG–aZnN was about 1.62 at% (ICP-OES), which is significantly
higher than that reported for transition metal based SACs.10
To give direct evidence of atomic Zn coordinated@N doped
graphene, sub-angstrom-resolution, high-angle annular dark field
Fig. 2d, therefore, confirms our simulated atomic Zn bonding
configurations. This Zn–N bonding could also be verified by the
subtle red-shift observed in the Zn 2p3/2 region for NG–aZnN and
NG–ZnO,12 (Fig. S7d, ESI†) and the C/N element XPS (Fig. S8, ESI†).
As in previous studies, the cycloaddition of epoxides and
(HAADF) imaging in an aberration-corrected STEM (AC-STEM) was CO2 undergoes an electrophilic activation and nucleophilic attack
performed. As shown in Fig. 2a, abundant bright dots, corres- synergistic catalytic mechanism.2a,3,6a Therefore, NG–aZnN could
ponding to heavier Zn atoms, are well-dispersed on the NG surface. potentially activate the epoxides and further initiate the reaction
The measured size of these bright dots ranges from 1.2 to 2.5 Å,
corresponding to the size of individual Zn atoms. Synchrotron
X-ray absorption near-edge structure (XANES) and extended X-ray
absorption fine structure (EXAFS) at the Zn K-edge were employed
efficiently. Firstly, the model cycloaddition reaction of propylene
oxide (PO) and CO2 to propylene carbonate (PC) was investigated.
In order to find the most suitable catalytic system and conditions,
detailed experiments were performed, as listed in Table S4 (ESI†).
to explore the metal coordination environment and unveil Zn A low PC yield was detected when only tetra-n-propylammonium
bonding configurations in NG–aZnN, along with model samples bromide (TPAB) was used (entry 1). Interestingly, the combination
of Zn foil and ZnO, as shown in Fig. 2b. The XANES curve of of graphene or NG–aZnN with TPAB as catalyst rendered remark-
NG–aZnN shows a near-edge absorption energy between those of able yields up to 41.5% and 75.5%, respectively (entries 2 and 3).
ZnO and Zn foil, implying isolated Zn atoms carrying positive
charges. In Fig. 2c, the corresponding Fourier transformed (FT)
The increase in the carbon chain in quaternary ammonium salt is
also beneficial to improving the product yield (entries 3–6), which
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2019

View Article Online
ChemComm
Communication
Table 1 Substrate scopea
Entry
1
Substrate 1
Product 2
Yieldb (%)
Sel.c (%)
98(98)
97(98)
98(97)
93(97)
499(99)
Fig. 3 The yields of PC (2a) using different catalysts; reaction conditions:
1 (20 mmol), catalyst (15 mg), (nC7H15)4NBr (49.1 mg, 0.5 mol%), 120 1C,
1 MPa, 3 h. The inset: cycloaddition of CO2 and PO to produce PC.
2
3
4
99(499)
100(99)
99(499)
could be attributed to the enhanced solubility of the co-catalyst in
the reaction system. The quantitative yield could be obtained for
3 hours under the optimized conditions (entry 7). Moreover, the
increase in temperature and pressure could distinctly accelerate
the reaction rate (Fig. S9 and S10, ESI†). Therefore, based on a
screening of the reactive parameters, 120 1C, 1 MPa and 3 h
reaction time (or 160 1C, 3 MPa, 1 h) with (nC7H15)4NBr as the
co-catalyst would be the optimized conditions and were employed
for the following tests.
Fig. 3 exhibits the PC yields using different graphene and zinc-
based catalysts (Material preparation, ESI†) under the optimized
catalytic conditions. A limited yield of 47.0% was detected when
only (nC7H15)4NBr was used (only co-catalyst). The addition of the
following catalysts, pure G, NG and nitrogen doped carbon–zinc
oxide (NC–ZnO), exhibited surprisingly negative effects on the
cycloaddition reaction. This performance degradation should be
attributed to the lack of active sites in either G/NG or ZnO for the
electrophilic activation process during the reaction; moreover,
the large lateral size and limited surface areas of G could
potentially block the transport of reactive species, leading to
compromised performances. The iron content also shows limited
catalytic activity, exhibiting a similar yield to only co-catalyst.
5
98(94)
499(97)
6
7
98(99)
98(99)
499(99)
99(99)
8
9
41(67)
(39)
99(98)
(99)
10
(52)
(99)
a
Reaction conditions: 1 (20 mmol), NG–aZnN (15 mg, 0.12 mol% Zn
Interestingly, NG–Co could enhance the yield of PC to 65.1%, content on the basis of substrate 1), (nC7H15)4NBr (49.1 mg, 0.5 mol%);
revealing that the cobalt species could facilitate the cycloaddition
protocol A (120 1C, 1 MPa, 3 h); protocol B (160 1C, 3 MPa, 1 h), the data
in the parenthesis are obtained via protocol B. Isolated yield. Deter-
mined by GC using biphenyl as the internal standard.
b
c
of CO2 and propylene oxide.13 Comparatively, the intermediate
NG–ZnO demonstrates a yield of 2a of 94.1%, which could be
further improved to more than 99.0% after acid leaching to
remove inactive ZnO for the zinc-based NG–aZnN SAC. It should
be noted that even when the feeding ratio of zinc precursor was However, the internal epoxides such as cyclohexene oxide
decreased to half of the original, the yield of 2a was still above exhibited limited reactivity, probably owing to the steric effect,
75.1%, indicating the excellent catalytic activity of atomic Zn and meanwhile the rise in temperature is beneficial to improving
coordinated@N doped graphene.
the efficiency (entry 8).
To investigate the scope of this novel atomic Zn catalyst, a series
To further validate the efficaciousness of SAC and the method-
of typical terminal and internal epoxide substrates (1a–h) were ology, a TON/TOF test was performed with a lower catalyst
examined. As shown in Table 1, under two different schemes, a amount (Scheme 1). As a result, a quantitative yield of 2a was
variety of target cyclic carbonates 2a–d were obtained in almost obtained with a high TON of up to 8666 and a TOF of up to
quantitative yield and excellent selectivity through the reaction of 2889 hÀ1. Comparatively, other analogous heterocatalysts such as
the corresponding terminal epoxides bearing an aliphatic or aro- graphitic carbon nitrides and nitrogen doped carbons gave the
matic group and CO2 (entries 1–4). Notably, representative terminal target product with low TOF values (100–600).14 Moreover, with a
epoxide substrates 1e–g embodying different functionalized lower catalytic loading, the higher TON was reached together with a
groups were also verified, and delightfully, corresponding high yield. These results confirmed the highly catalytic efficacious-
products 2e–g were acquired in good yields (entries 5–7). ness of an atomically dispersed zinc catalyst. The recyclability of
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun.

View Article Online
Communication
ChemComm
Financial support from the National Natural Science Foun-
dation of China (51502320, 21602232, U1662102), the Natural
Science Foundation of Shanxi Province (201701D221057, 20160
1D021060), the Shanxi Scholarship Council of China and the DNL
Cooperation Fund (DNL180401) is gratefully acknowledged.
Scheme 1 TON/TOF experiments. (1) Reaction conditions: 1a (11.6 g,
0.2 mol), 3 h; results: 2a, 21 g, 499% yield, TON = 8666, TOF: 2889 hÀ1
.
(2) Reaction conditions: 1a (23.2 g, 0.4 mol), 9 h; results: 2a, 36.2 g, 89%
yield, TON = 15 380, TOF: 1 710 hÀ1
.
Conflicts of interest
There are no conflicts to declare.
Notes and references
¨
¨
¨
1 (a) B. Schaffner, F. Schaffner, S. P. Verevkin and A. Borner, Chem.
´
Rev., 2010, 110, 4554–4581; (b) C. Martın, G. Fiorani and A. W. Kleij,
ACS Catal., 2015, 5, 1353–1370; (c) X.-D. Lang and L.-N. He, Chem.
Rec., 2016, 16, 1337–1352.
2 (a) M. Cokoja, M. E. Wilhelm, M. H. Anthofer, W. A. Herrmann and F. E.
Ku¨hn, ChemSusChem, 2015, 8, 2436–2454; (b) Q. He, J. W. O’Brien,
K. A. Kitselman, L. E. Tompkins, G. C. T. Curtis and F. M. Kerton, Catal.
Sci. Technol., 2014, 4, 1513–1528.
3 V. B. Saptal and B. M. Bhanage, Curr. Opin. Green Sustainable Chem.,
2017, 3, 1–10.
Scheme 2 Plausible atomic [ZnN] based reaction mechanism.
4 H. Yasuda, L.-N. He, T. Sakakura and C. Hu, J. Catal., 2005, 233,
119–122.
5 (a) H. He, Q. Sun, W. Gao, J. A. Perman, F. Sun, G. Zhu, B. Aguila,
K. Forrest, B. Space and S. Ma, Angew. Chem., Int. Ed., 2018, 57,
4657–4662; (b) Z. F. Dai, Q. Sun, X. L. Liu, C. Q. Bian, Q. M. Wu,
S. X. Pan, L. Wang, X. J. Meng, F. Deng and F. S. Xiao, J. Catal., 2016,
338, 202–209; (c) J. Zhou, Y. Dou, A. Zhou, L. Shu, Y. Chen and J.-R.
Li, ACS Energy Lett., 2018, 3, 1655.
6 (a) Z. Z. Yang, Y. N. Zhao, L. N. He, J. Gao and Z. S. Yin, Green Chem.,
2012, 14, 519–527; (b) X. Wang, Y. Zhou, Z. Guo, G. Chen, J. Li,
Y. Shi, Y. Liu and J. Wang, Chem. Sci., 2015, 6, 6916–6924.
7 (a) V. D’Elia, H. Dong, A. J. Rossini, C. M. Widdifield, S. V. C. Vummaleti,
Y. Minenkov, A. Poater, E. Abou-Hamad, J. D. A. Pelletier, L. Cavallo,
L. Emsley and J.-M. Basset, J. Am. Chem. Soc., 2015, 137, 7728–7739;
(b) X.-B. Lu, H. Wang and R. He, J. Mol. Catal. A: Chem., 2002, 186, 33–42;
(c) Q.-N. Zhao, Q.-W. Song, P. Liu, Q. X. Zhang, J.-H. Gao and K. Zhang,
Chin. J. Chem., 2018, 36, 187–193.
8 (a) D. H. Lan, F. M. Yang, S. L. Luo, C. T. Au and S. F. Yin, Carbon,
2014, 73, 351–360; (b) Z. F. Dai, Q. Sun, X. L. Liu, L. P. Guo, J. X. Li,
S. X. Pan, C. Q. Bian, L. Wang, X. Hu, X. J. Meng, L. H. Zhao and F. S.
Xiao, ChemSusChem, 2017, 10, 1186–1192; (c) V. B. Saptal, T. Sasaki,
K. Harada, D. Nishio-Hamane and B. M. Bhanage, ChemSusChem,
2016, 9, 644–650.
9 J. K. Nørskov, F. Studt, F. Abild-Pedersen and T. BligaardJens, Funda-
mental Concepts in Heterogeneous Catalysis, John Wiley & Sons, 2014.
10 (a) H. Wei, X. Liu, A. Wang, L. Zhang, B. Qiao, X. Yang, Y. Huang,
S. Miao, J. Liu and T. Zhang, Nat. Commun., 2014, 5, 5634;
(b) C. Wang, H. Zhang, Y. Zhang, M. Cheng, H. Zhao, J. Wang and
J. Wang, Chem. Mater., 2017, 29, 9915–9922; (c) B. Qiao, A. Wang,
X. Yang, L. Allard, Z. Jiang, Y. Cui, J. Liu, J. Li and T. Zhang, Nat.
Chem., 2011, 3, 634–641; (d) X. Guo, G. Fang, G. Li, H. Ma, H. Fan,
L. Yu, C. Ma, X. Wu, D. Deng, M. Wei, D. Tan, R. Si, S. Zhang, J. Li,
L. Sun, Z. Tang, X. Pan and X. Bao, Science, 2014, 344, 616–619;
(e) A. Li, C. Wang, H. Zhang, Z. Zhao, J. Wang, M. Cheng, H. Zhao,
J. Wang, M. Wu and J. Wang, Electrochim. Acta, 2018, 276, 153–161.
NG–aZnN was further examined (Fig. S11, ESI†); the resultant
decreased PC yield could be attributed to surface partial Zn
aggregation after the reaction, as shown in Fig. S12 (ESI†).
On the basis of previous reports and experimental investigation
in this work, a tentative mechanism for the NG–aZnN-catalyzed
cycloaddition reaction of CO2 and epoxides could be plotted. It
should be noted that for simplicity, a [ZnN4] unit is schematically
shown in Scheme 2. Initially, the functional site, i.e. [ZnN4] in
NG–aZnN, interacts closely with the epoxide to polarize the C–O
bond of the epoxide. Subsequently, the activated epoxide con-
currently undertakes a ring-opening step upon nucleophilic attack
by the bromide anion from the less-sterically-hindered carbon atom
and gives an intermediate A (step 1). Next, the active site-stabilized
alkoxide intermediate A is formed. Then, nucleophilic attack of the
intermediate A at the CO2 molecule generates the alkyl carbonate
anion B (step 2). Finally, cyclic carbonate is produced via ring
closing with regeneration of the catalytic species (step 3). In this
process, the activation of epoxides and stabilization of intermediate
were significantly enhanced via the functional site [ZnN4] and
reactants’ intermolecular interactions, leading to the excellent
catalytic performances.
In summary, a novel single zinc atom catalyst was prepared
through a facile pyrolysis method with an oleic zinc precursor.
The atomic zinc was coordinated with doped nitrogen atoms and
supported on a graphene sheet. This Zn-based SAC has demon- 11 (a) H. B. Yang, S.-F. Hung, S. Liu, K. Yuan, S. Miao, L. Zhang,
X. Huang, H.-Y. Wang, W. Cai, R. Chen, J. Gao, X. Yang, W. Chen,
Y. Huang, H. M. Chen, C. M. Li, T. Zhang and B. Liu, Nat. Energy,
2018, 3, 140–147; (b) F. Yang, P. Song, X. Liu, B. Mei, W. Xing,
strated high efficiency for the synthesis of cyclic carbonates from
epoxides and CO2, with superior catalytic species for CO2 con-
version with a TON up to 8666 (TOF: 2889 hÀ1). The unique and
carefully characterized ZnN3.76Æ0.2 units were revealed as the
dominant catalytic sites in the chemical transformation. We
believe that our strategy can be further developed into a general
approach to fabricate inexpensive heterogeneous catalysts for
upgrading CO2 through its incorporation into valuable chemicals
with synergistic activation and catalysis.
Z. Jiang, L. Gu and W. Xu, Angew. Chem., Int. Ed., 2018, 130, 1–6.
12 (a) H. Zhang, J. Wei, J. Dong, G. Liu, L. Shi, P. An, G. Zhao, J. Kong,
X. Wang, X. Meng, J. Zhang and J. Ye, Angew. Chem., Int. Ed., 2016,
55, 14310–14314; (b) S. Kumar, A. Baruah, S. Tonda, B. Kumar,
V. Shanker and B. Sreedhar, Nanoscale, 2014, 6, 4830–4842.
13 X.-B. Lu and D. J. Darensbourg, Chem. Soc. Rev., 2012, 41, 1462–1484.
14 (a) Z. Huang, F. Li, B. Chen and G. Yuan, Catal. Sci. Technol., 2016, 6,
2942–2948; (b) J. Xu, F. Wu, Q. Jiang, J.-K. Shang and Y.-X. Li, J. Mol.
Catal. A: Chem., 2015, 403, 77–83.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2019