Double-metal cyanide as an acid and hydrogenation catalyst for the highly selective ring-rearrangement of biomass-derived furfuryl alcohol to cyclopentenone compounds

Article abstract of DOI:10.1039/c9gc04432a

Herein, novel green synthetic routes to 4-hydroxy-2-cyclopentenone (HCP) and 2-cyclopentenone (CPE) from biomass-derived furfuryl alcohol via double-metal cyanide catalysis are proposed. For the synthesis of HCP, in comparison to conventional solid acids (i.e., Amberlyst-15), MOFs with coordinatively unsaturated metal ions as pure Lewis acid sites exhibit advantageous catalytic selectivity in the reaction under an N₂ atmosphere in a bi-phasic water/n-hexane solvent system. FeZn and FeZn-P result in an HCP yield of 77.4% and 88.2%, respectively. For the CPE synthesis, the reaction conditions are the same as those for HCP, except a mono-phasic water solvent system and H₂ atmosphere were employed. In addition to the acid-catalyzed rearrangement reaction, FeZn-DMC exhibits catalytic hydrogenation capability via heterolytic cleavage of the H-H bond over Zn-N frustrated Lewis pairs, and a CPE yield of 61.5% is obtained. The DFT simulation indicates that the acid sites and catalytic acid sites are ascribed to the tri-coordinatively unsaturated Zn²⁺ site (Zn(N)₃) on the catalyst surface. Moreover, the DMC catalyst shows excellent stability and recycling performance. This work not only provides an efficient and green catalytic system for CPE and HCP preparation but also demonstrates the interesting bifunctional catalysis of both acid and hydrogenation catalysis over DMC.

Full text of DOI:10.1039/c9gc04432a

Green Chemistry
View Article Online
View Journal
PAPER
Double-metal cyanide as an acid and
hydrogenation catalyst for the highly selective
ring-rearrangement of biomass-derived furfuryl
alcohol to cyclopentenone compounds†
Cite this: DOI: 10.1039/c9gc04432a
Xiang Li,a Qiang Deng, *a Lian Yu,a Rui Gao,a Zhikun Tong,a Chenxi Lu,b
b
Jun Wang, a Zheling Zeng, a Ji-Jun Zou c and Shuguang Deng
*
Herein, novel green synthetic routes to 4-hydroxy-2-cyclopentenone (HCP) and 2-cyclopentenone (CPE)
from biomass-derived furfuryl alcohol via double-metal cyanide catalysis are proposed. For the synthesis
of HCP, in comparison to conventional solid acids (i.e., Amberlyst-15), MOFs with coordinatively unsatu-
rated metal ions as pure Lewis acid sites exhibit advantageous catalytic selectivity in the reaction under an
N2 atmosphere in a bi-phasic water/n-hexane solvent system. FeZn and FeZn-P result in an HCP yield of
77.4% and 88.2%, respectively. For the CPE synthesis, the reaction conditions are the same as those for
HCP, except a mono-phasic water solvent system and H2 atmosphere were employed. In addition to the
acid-catalyzed rearrangement reaction, FeZn-DMC exhibits catalytic hydrogenation capability via hetero-
lytic cleavage of the H–H bond over Zn–N frustrated Lewis pairs, and a CPE yield of 61.5% is obtained.
The DFT simulation indicates that the acid sites and catalytic acid sites are ascribed to the tri-coordina-
tively unsaturated Zn2+ site (Zn(N)3) on the catalyst surface. Moreover, the DMC catalyst shows excellent
stability and recycling performance. This work not only provides an efficient and green catalytic system
for CPE and HCP preparation but also demonstrates the interesting bifunctional catalysis of both acid and
hydrogenation catalysis over DMC.
Received 29th December 2019,
Accepted 25th March 2020
DOI: 10.1039/c9gc04432a
rsc.li/greenchem
cyclopentyl molecule.5 Therefore, the synthesis of 4-hydroxy-2-
cyclopentenone (HCP) and 2-cyclopentenone (CPE) is of great
Introduction
α,β-Unsaturated carbonyl compounds (e.g., 2-cyclopentenone interest in organic chemistry. Traditionally, three pathways are
and 4-hydroxy-2-cyclopentenone) are important building available for CPE preparation (Scheme 1). (1) Palladium-cata-
blocks for natural products, drug molecules, and bulk lyzed dehydrosilylation of 1-trisilyloxy-1-cyclopent-1-ene can be
chemicals,1–4 as well as versatile synthetic reactants in diverse achieved with a CPE yield more than 90% using an oxidant
organic transformations (e.g., Michael addition, Diels–Alder,
and Heck reaction) because these compounds simultaneously
contain active olefin, carbonyl, and hydroxyl groups in a small
aKey Laboratory of Poyang Lake Environment and Resource Utilization
(Nanchang University) of Ministry of Education, School of Resource, Environmental
and Chemical Engineering, Nanchang University, No. 999 Xuefu Avenue,
Nanchang 330031, PR China. E-mail: dengqiang@ncu.edu.cn
bSchool for Engineering of Matter, Transport and Energy, Arizona State University,
51 E. Tyler Mall, Tempe, AZ 85287, USA. E-mail: Shuguang.Deng@asu.edu
cKey Laboratory for Green Chemical Technology of the Ministry of education,
School of Chemical Engineering and Technology, Tianjin University; Collaborative
Innovative Center of Chemical Science and Engineering (Tianjin),
No. 92 Weijin Road, Tianjin 300072, PR China
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c9gc04432a
Scheme 1 Synthetic routes for the production of CPE and HCP.
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
(e.g., p-benzoquinone, tert-butyl hydroperoxide, and oxygen) in widely accepted that the Lewis acidity in the acid support has
an organic solvent.6–9 (2) Photo-oxygenation of cyclopentene a crucially positive effect for the rearrangement step of FA to
with 71% CPE yield has been demonstrated using oxygen as HCP, but the Brønsted acidity more easily promotes the inter-
the oxidant in dichloromethane solvent in the presence of molecular C–C coupling reaction of intermediates, leading to
acetic anhydride, base and an acylation catalyst (i.e., 4-(di- the production of humins.18–20,30,31 Unfortunately, the hydro-
methylamino)pyridine).10 (3) Palladium-catalyzed dehydro- genation of furfural and rearrangement of FA are often rate-
genative oxidations of cyclopentanone have been reported with limiting steps, and active HCP and CPE are difficult to pre-
a CPE yield of 51% using oxygen as the oxidant with the coca- serve due to these subsequent rapid steps.21–23 Based on the
talytic effect of trifluoroacetic acid under dimethyl sulfoxide.11 current development of HCP and CPE syntheses, we were
The development of these catalytic systems is limited, due to inspired to synthesize a pure solid Lewis acid catalyst for the
the various issues including the use of expensive noble metal selective synthesis of HCP and CPE from FA under mild reac-
catalysts, difficulty with the separation and reuse of a homo- tion conditions.
geneous catalyst, corrosivity of acidic agents, large quantities
Metal–organic frameworks (MOFs) with a modular assem-
of organic solvent and expensive raw materials. In addition, bly structure, high surface area, tailorable pores and struc-
these synthesis pathways exhibit low atomic economy, and the tural versatility have the potential for application for gas
reactants are derived from fossil resources, which is not con- storage, separation, catalysis, sensing, drug delivery and
ducive to green and sustainable development. Therefore, the proton conductivity.32–36 The coordinately unsaturated metal
development of an innovative green route from biomass-based centers in a MOF can be used as Lewis acid sites to catalyze
raw materials to CPE is important.
numerous reactions. However, MOFs often exhibit poor
On the other hand, HCP has been synthesized using chemical stability and hydrothermal stability, especially
biomass-derived furfuryl alcohol (FA) via an acid-catalyzed for reactions of aqueous solvents.37,38 As a classical MOF,
rearrangement reaction in water. FA, an important biomass double-metal cyanides (DMCs) which are famous for the use
platform compound, is primarily derived from hemicelluloses of Prussian blue, are constructed by octahedral metal cyanide
via several chemical transformations.12–15 Hronec′s group anion groups bridged with metal ions through a cyanide
reported that FA can be converted to HCP, which is catalyzed group (CuN). Typically, on the edge of a DMC, the coordi-
by the hydrogen proton formed by auto-dissociation of water nately unsaturated vacancy metal ions can be used as Lewis
under high temperature (180–200 °C).16 Unfortunately, the acid site. Generally, triblock copolymers that are used as a
yield of HCP was low (i.e., approximately 50%) under these complexing agent are added to the synthesis process to sup-
severe reaction conditions because some intermediates are press overgeneration of crystals and adjust the surface area.
unstable and easily converted into humins via intermolecular Currently, these DMCs are primarily used in acid-catalyzed
C–C coupling reactions. Zhang′s group used a mild solid base polymerization, esterification, transesterification, polyesterifi-
to dilute the density of the in situ water-produced hydrogen cation, polymerization, hydroamination, and Prins conden-
proton at 240 °C. Although the conversion of FA decreased, the sation reactions.39–43 However, in most cases, these DMC cat-
HCP selectivity increased, and a high HCP yield of approxi- alysts do not exhibit obvious advantages over other catalysts
mately 80% was obtained.17 In addition, under the same reac- based on activity and selectivity.
tion temperature (i.e., 240 °C), Reiser′s group reported an
In our previous work, Pd supported acidic DMCs show an
excellent activity of acetic acid catalysis to convert FA to HCP excellent catalytic activity for the hydrogenative rearrangement
in a high yield of 87% using water–toluene mixed solvent.18–20 of furfural to cyclopentanone. However, the amount of inter-
However, these catalytic systems employ high reaction temp- mediates HCP and CPE is low because of the strong hydrogen-
eratures and homogeneous acids that are difficult to separate ation ability of Pd nanoparticles.24 Herein, for the first time,
and reuse, and experience equipment corrosion, restricting FeZn-DMCs are used as Lewis-acid catalysts for the rearrange-
their development for real-life applications.
ment of furfuryl alcohol (FA) in the water at 150 °C. In com-
Recently, the hydrogenative ring-rearrangement of furfural parison to the Amberlyst-15, Al-MCM-41 as well as a familiar
to cyclopentanone was widely reported over various Group MOF (i.e. Cr-MIL-101, Fe-MIL-100), DMCs exhibit great cata-
VIII or IB metals (such as, Au, Ni, Pd) supported acidic sup- lytic activity and selectivity for HCP preparation under an N2
ports (such as TiO2, Al2O3, metal–organic frameworks) under atmosphere in a bi-phasic solvent system. FeZn and FeZn-P
H2 atmosphere and relatively mild reaction temperature produce 88.8% and 95.2% conversion of FA with HCP yield of
(140–180 °C).21–29 In the hydrogenative rearrangement reac- 77.4% and 88.2%, respectively. To our delight, in addition to
tion, furfural is first hydrogenated to FA, then acid-catalyti- the acid-catalyzed rearrangement reaction, FeZn-DMC exihibits
cally rearranged to HCP, further generated CPE followed by catalytic hydrogenation abilities, and a CPE yield of 61.5% is
subsequent hydrogenation and dehydration steps. Finally, obtained in an H2 atmosphere and mono-phasic water solvent
cyclopentanone is obtained by hydrogenation of CPE. In the system. Moreover, this catalyst exhibits excellent stability and
bifunctional catalysis, the hydrogenation active sites and recycling performance under these reaction conditions. This
acidic sites are attributed to reduced metals and acid sup- work not only provides a green and efficient route for CPE and
ports, respectively. In general, cyclopentanone is synthesized HCP synthesis but also demonstrates the interesting hydrogen-
with a yield higher than 80%. In the reaction, it has been ation performance of DMC.
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
out using inductively coupled plasma optical emission spec-
trometry on a PerkinElmer Optima 8000 spectrometer. The N
element was characterized using a CHNS elemental analyzer
Experimental
Materials
Furfuryl alcohol, hydrofluoric acid (35% HF), terephthalic acid (Vario MICRO). The morphologies were determined using
(H2BDC), N,N-dimethyl formamide, ethanol, n-hexane, buty- a JEM-2100 transmission electron microscope. The acid
lenes oxide, isopropanol, dimethyl sulfoxide, potassium hexa- properties were evaluated by infrared spectroscopy with
cyanoferrate (K4[Fe(CN)6]·3H2O), iron powder, chromium(III) adsorbed pyridine using a Bruker VERTEX 70 spectrometer.
nitrate nonahydrate (Cr(NO3)3·9H2O), nitric acid (65% HNO3), The self-supported wafers (10 mg cm−2) were degassed for
trimesic acid (H3BTC), potassium hexacyanocobaltate K3Co 1 h under vacuum at 200 °C. Then, adsorption of pyridine
(CN)6, zinc chloride (ZnCl2) and tert-butanol were purchased occurred at 60 °C for 30 min followed by desorption at
from Sinopharm Chemical Reagent Company. Amberlyst-15, 150 °C for 20 min, and the transmission spectra were
tert-butyl alcohol, dichloromethane, N-methyl pyrrolidone and recorded. The thermal stabilities were investigated by
acetonitrile were supplied by the Aladdin Reagent Company. thermal gravimetric analysis using a NETZSCH-STA 2500 in
Poly(ethylene glycol)-block-poly(propylene glycol)-block-poly- an N2 atmosphere.
(ethylene glycol) (P123, average molecular weight = 5800) was
obtained from Sigma-Aldrich. All chemicals were used without
Catalytic reactions
further purification.
The synthesis of HCP and CPE from FA was conducted in a
100 mL batch autoclave equipped with a mechanical stirrer
(Anhui Kemi Machinery Technology Co., Ltd). For the HCP
Synthesis of catalysts
DMC was synthesized using a simple precipitation method. synthesis, a mixture consisting of 0.1 g of catalyst, 10.2 mmol
Typically, for FeZn-DMC, a mixture of 0.1 mol of ZnCl2, 20 mL of furfuryl alcohol, and 40 mL of solvent was heated to 150 °C
of water and 20 mL of tert-butanol was slowly added to 40 mL under an N2 atmosphere. For the CPE synthesis, the reaction
of a 0.25 mol L−1 K4[Fe(CN)6]·3H2O aqueous solution over 1 h was carried out at 150 °C with 0.1 g of catalyst, 10.2 mmol of
at 50 °C under stirring. Then, 20 g of P123 (i.e., complexing furfuryl alcohol and 40 mL of water under an H2 atmosphere.
agent) in 2 mL of water and 40 mL of tert-butanol (i.e. co-com- These reaction mixtures were periodically sampled and ana-
plexing agent) were added over 10 min and stirred continu- lyzed using Agilent 6890N GC/5973 MS detector and quantified
ously for 1 h. Then, the crystal (FeZn-P) was obtained after fil- using a Trace 1300 gas chromatograph equipped with an FID
tering and washing with H2O and drying at 60 °C for 8 h. FeZn detector and a TG-WAXMS capillary column (30 m × 0.32 mm).
was synthesized using the same process without P123 and tert-
Computational methodology
All calculations were performed using the DMol3 program in
butanol. Similarly, ZnCo-P was synthesized by replacing K4[Fe
(CN)6]·3H2O with K3[Co(CN)6].
For comparison, Cr-MIL-101 and Fe-MIL-100 were syn- the Materials suite (Accelrys Inc.).45,46 The exchange–corre-
thesized using a hydrothermal method. For the synthesis of lation energy was calculated using the generalized gradient
Cr-MIL-101, a mixture of Cr(NO3)3·9H2O (14 mmol), H2BDC approximation of Perdew and Wang (PW91) with the Grimme
(14 mmol), HF (14 mmol), and water (70 mL) were stirred, method for DFT-D correction, which yields reliable results for
transferred into a Teflon-lined autoclave, sealed and heated at various reactions.47,48 The double-numerical basis set with a
220 °C for 8 h. The resulting light-green crystal was obtained polarization d-function (DNP) was chosen to expand the
after filtering and washing with DMF, ethanol, an NH4F solu- valence electron functions. The orbital cutoff range was 4.6 Å,
tion and H2O, followed by drying for 8 h at 150 °C. For the syn- and the smearing value was set to 0.005 Hartrees. The core
thesis of Fe-MIL-100, iron powder (4.9 mmol), H3BTC electrons for the Zn and Fe atoms were represented by density-
(3.27 mmol), hydrofluoric acid (4 mmol) and nitric acid functional semi-core pseudopotentials (DSSPs). The conver-
(2.78 mmol) were mixed in water (20 mL), and heated for 12 h gence criterion for the optimal geometry was based on the
at 150 °C. The brown Fe-MIL-100 was further purified with energy, force and displacement convergence (i.e. 1 × 10−5
water and ethanol, then dried overnight in a 100 °C vacuum Hartrees, 2 × 10−3 Hartrees Å−1 and 5 × 10−3 Å, respectively).
drying oven. Al-MCM-41 (SiO2/Al2O3 = 26.8) was synthesized The linear and quadratic synchronous transit (LST/QST) com-
according to previous studies.44
plete search method was used to search for the transition state
of the reactions.49 The nature of the minima on the potential
energy surface was verified by inspection of the harmonic
Catalyst characterizations
The crystal structures of the catalysts were determined by vibrational frequencies.
powder X-ray diffraction (XRD) using a Panalytical Empyrean
For the theoretical calculations, a Zn13Fe9(CN)66H58 cluster
diffractometer with Cu-Kα radiation. The porous structures model was employed to represent the catalytic structure of
were characterized by N2 adsorption–desorption at −196 °C FeZn-DMC. Due to the high expense of a DFT calculation on a
on an ASAP 2460 surface analyzer. Fourier transform infrared large system, a small cluster with a di-coordinatively or tri-
(FTIR) spectra were recorded in a range of 400–4000 cm−1 on coordinatively unsaturated Zn site (Zn4Fe3(CN)23H14 and
a
Nicolet 5700 FTIR spectrometer using KBr pellets. Zn4Fe3(CN)24H15) was employed to study the dissociation of H2
Elemental analysis of the Fe and Zn elements was carried on the catalyst surface.
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
Table 2 Element analysis of FeZn-DMC
Results and discussion
Structural characteristics of DMC
N
(wt%)
Fe
(wt%)
Zn
(wt%)
Molar ratio
of N : Fe
Molar ratio
of Zn : Fe
Samples
Fig. 1A shows the XRD patterns of FeZn-DMC. In agreement
FeZn-P
FeZn
18.68
17.89
18.15
11.76
12.25
11.06
24.19
24.19
22.44
6.352
5.842
6.562
1.771
1.57
1.747
ˉ
with previous results, the FeZn-DMC has an R3c space group
structure with tetra-coordinated Zn2+ connected to the four
FeZn-Pa
NuC groups of octahedral [Fe(CN)6]4−, which is consistent
a Represents the recycled sample.
39,40,42
with K2Zn3[Fe(CN)6]2
The anionic zinc-ferrocyanide
framework is characterized by cages of the type depicted in
Fig. 1B, where each cage contains two guest K+ ions. The TEM
image indicates that DMC has a uniform crystal size of
200–300 nm, and the particle size did not change upon
addition of the complexing agents (Fig. S1†). The N2 adsorp-
tion–desorption isotherm of FeZn is consistent with a type II
isotherm, and FeZn-P shows a slight type IV isotherm, indicat-
ing the existence of mesopores after using P123 (Fig. 1C). The
topological pore of FeZn is too small for N2 to enter. The BET
surface area and pore volume of FeZn and FeZn-P are 33.2 and
42.6 m2 g−1, 0.051 and 0.068 cm3 g−1, respectively (Table 1). In
addition, Fig. 1A and S2† show the successful preparation of
Cr-MIL-101 and Fe-MIL-100. In these MIL series MOFs, the
Cr3+/Fe3+ trimer is coordinated to twelve O atoms from H2BDC/
H3BTC, one F− and two water molecules. The BET surface area
and pore volume of Cr-MIL-101 and Fe-MIL-100 are 3225.7 and
841.2 m2 g−1, 1.540 and 0.542 cm3 g−1, respectively.
As shown in Table 2, elemental analysis results indicate
that the N : Fe molar ratios of FeZn and FeZn-P (i.e., 5.842 and
6.352, respectively), are close to the stoichiometrically expected
value (i.e., 6), demonstrating the stability of the octahedral [Fe
(CN)6]4− anion groups. Whereas, the Zn : Fe molar ratio of
FeZn is 1.570, which is higher than the theoretical value (i.e.,
1.5). Due to the excess Zn2+ ions in the synthesis mixture, the
Fe2+ ions are surrounded by the CuN bonds and Zn2+ ions ter-
minate the crystals. The coordinately unsaturated Zn2+ ions
provide the Lewis acidity.43 Meanwhile, the addition of P123
and tert-butanol can suppress overgeneration of the crystal
and increases the Zn : Fe molar ratio to 1.771 for FeZn-P. As
shown in Fig. 1D, the pyridine-absorbed FTIR spectra were
used to determine the acidity of these acid catalysts. The
adsorption bands at 1450 and 1540 cm−1 are assigned to the
chemisorbed pyridine corresponding to the Lewis and
Brønsted acid sites, respectively. The acid density was calcu-
lated using the Emeis equation.50 FeZn-P shows a larger acid
density of both Lewis (1.349 mmol g−1
) and Brønsted
(0.105 mmol g−1) acids than FeZn, because more coordinately
unsaturated Zn2+ ions provide more acid sites that adsorb
more pyridine. Both FeZn and FeZn-P exhibit a high acid
density ratio (i.e. Lewis/Brønsted) with 10.2 and 12.8, respect-
ively. Al-MCM-41 shows a both Lewis (0.034 mmol g−1) and
Brønsted acidity (0.012 mmol g−1) with the ratio of 2.8. For
MIL series MOFs, the Lewis acid density of Cr-MIL-101 and Fe-
MIL-100 are 2.790 and 3.072 mmol g−1 according to its
format.51 This value cannot be obtained from the pyridine-
absorbed FTIR spectra because the strong bands of arising
from carboxylate and benzene ring vibrations disturb the pyri-
dine adsorption signal in the 1400–1600 cm−1 region. As
shown in Fig. S3,† the thermal stability results for FeZn-DMC
indicate a weight loss between 60 °C to 200 °C due to the loss
of crystal and noncrystal water. The structure remained stable
below 300 °C, indicating that their stability under the reaction
conditions (150 °C).
Fig. 1 (A) X-ray diffraction patterns, (B) crystal structure, (C) N2 adsorp-
tion–desorption isotherms and (D) infrared spectra of adsorbed pyridine
for FeZn-DMC. arepresents the recycled sample.
Table 1 Physicochemical properties of catalysts
Lewis acid Brønsted
SBET
Vtotal
density
acid density Lewis/
Samples
(m2 g−1
)
(cm3 g−1
)
(mmol g−1
)
(mmol g−1
)
Brønsted
FeZn
FeZn-P
33.2
42.6
3225.7
841.2
851.2
0.051
0.068
1.540
0.542
0.965
0.31
0.746
1.349
2.790
3.072
0.034
0
0.073
0.105
0
10.2
12.8
—
—
2.8
0
Cr-MIL-101
Fe-MIL-100
Al-MCM-41
0
Synthesis of HCP
0.012
4.7
Amberlyst-15a 45
As mentioned above, the rearrangement of FA to HCP involves
a hydrolysis reaction to 2-pentene-1,4-dione (PED) intermedi-
ate, followed by intramolecular aldol condensation reactions to
a BET surface area and pore volume value of Amberlyst-15 were obtained
from our previous work.52,53
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
dimethyl sulfoxide (7.2) > acetonitrile (6.2) > ethanol (6.0) >
isopropanol (4.3) > tetrahydrofuran (4.2) > hexane (0.06). Using
a strong polar solvent (e.g., dimethyl sulfoxide, acetonitrile,
and ethanol), both the FA conversion rate and HCP selectivity
are low. Interestingly, the conversion and selectivity increase
as the polarity value decreases. The weak polar (e.g., isopro-
panol, tetrahydrofuran) and non-polar (e.g. n-hexane) solvents
exhibit considerable reaction performance. First, the weak
polar solvent is more difficult to dissolve in water, which is the
reactant for hydrolysis of FA. Therefore, these solvents exert a
weak negative influence on the hydrolysis reaction, and the
activity in the water/weak-polar solvent is higher than that in
the water/strong-polar solvent. Second, because the active PED
is generated in the water phase, the weak polar solvent may
easily extract it to maintain a low concentration on the acid
site. An intramolecular aldol reaction is more likely to occur
than some intermolecular C–C bond reactions. Hence, the
HCP selectivity is the highest in the water/non-polar solvent bi-
phasic system.
Fig. 2E further compares the catalytic performance of
several acid catalysts in the aqueous/n-hexane bi-phasic
system. As a common Brønsted acid, Amberlyst-15 exhibits
only 16.5% selectivity to HCP, and a large amount of levuli-
nic acid is generated. Al-MCM-41 (42.3%) exhibits a higher
HCP selectivity than Amberlyst-15 (16.4%). According to pre-
vious work, Brønsted acid (e.g., H2SO4, HCl, p-toluenesulfo-
nic acid, H-ZSM-5, HY and Amberlyst-15) more easily cata-
lyzes the hydrolysis reaction.14,15 In contrast, FeZn, FeZn-P,
Fe-MIL-100 and Cr-MIL-101 with Lewis acidity possess the
high HCP selectivity of more than 70%. However, Cr-MIL-101
Fig. 2 (A) Proposed reaction pathway for HCP preparation, (B) time-
dependent product distribution using FeZn-P as a catalyst in 40 mL of
water or (C) a bi-phasic solvent with 20 mL of water and 20 mL of
n-hexane, (D) effect of different solvents, and (E) catalytic performance
of several acid catalysts. Reaction conditions: FA (10.2 mmol), catalyst
(0.1 g), 150 °C, 4.0 MPa of N2 pressure. (D) Water (20 mL) and organic
solvent (20 mL), FeZn-P catalyst, reaction time 6 h. (E) 20 mL water and
20 mL n-hexane, reaction time 6 h.
HCP (Fig. 2A and B).18–20 PED is very reactive, and can only be and Fe-MIL-100 exhibit relatively low activity below 85% con-
detected in small amounts with GC (Fig. S4A†). The product version, possibly attributing to the weak acidity of the coordi-
was identified by GC-MS (Fig. S5A†). In addition to the main nately unsaturated Cr3+/Fe3+ ions. In addition, FeZn, which
reaction to HCP, some by-products including levulinic acid has a low surface area and acid density, exhibits a FA conver-
(LA) and humins can be generated through ring-opening and sion that is lower than that of FeZn-P. Hence, FeZn-P with
intermolecular C–C reactions, respectively. Different from pre- the highest activity and selectivity is considered as the
vious reports using a high temperature without an acid cata- best catalyst for subsequent optimization of the reaction
lyst, FA cannot be substantially converted at 150 °C due to a conditions.
low density of protons formed by H2O auto-dissociation
As shown in Fig. S6,† when the reaction temperature
(Fig. 2D).16,17 After the addition of FeZn-P, the FA conversion decreases to 130 °C, only 52.3% of FA is converted. However,
rate largely increased. The amount of HCP, LA, and humins when the reaction temperature is higher than 160 °C, the HCP
also synchronously increased with the reaction time, indicat- yield begins to decrease, because some HCP undergoes inter-
ing the presence of parallel reactions (Fig. 2B). After 6 h of molecular aldol condensation at higher temperatures. If the
reaction, FA is completely consumed with a HCP yield of only water/hexane ratio is changed from 40 : 0 to 0 : 40, the conver-
38.2%, and 56.2% carbon loss due to the generation of sion decreases (Fig. S7†). Under the optimal conditions, 95.2%
humins. A stable increase in the HCP yield implies that the of FA is converted with a 92.6% HCP selectivity and 88.2%
humins are primarily produced by the reactive PED.
HCP yield via FeZn-P catalysis in a 1 : 1 mixture containing
Fortunately, as shown in Fig. 2C, when the solvent is 40 mL of water/hexane at 150 °C for 6 h.
changed from 40 mL of water to the bi-phasic solvent consist-
As shown in Fig. S6,† when the reaction temperature
ing of 20 of mL water and 20 mL of n-hexane, the reaction exhi- decreases to 130 °C, only 52.3% of FA is converted. However,
bits a much higher selectivity in the time-dependent product when the reaction temperature is higher than 160 °C, the HCP
curve, although the activity is slightly decreased. Subsequently, yield begins to decrease, because some HCP undergoes inter-
some other stable organic solvents (e.g. ethanol, n-hexane, iso- molecular aldol condensation at higher temperatures. If the
propanol, dimethyl sulfoxide, acetonitrile and tetrahydrofuran) water/hexane ratio is changed from 40 : 0 to 0 : 40, the conver-
were selected to mix with water for use as the reaction medium sion decreases (Fig. S7†). Under the optimal conditions, 95.2%
(Fig. 2D). The polarities of these solvents are as follows: of FA is converted with a 92.6% HCP selectivity and 88.2%
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
HCP yield via FeZn-P catalysis in a 1 : 1 mixture containing tion pattern analysis clearly reveal the hydrogenation step
40 mL of water/hexane at 150 °C for 6 h.
(Fig. S5C†). Meanwhile, the yield of CPE is affected by the H2
pressure (Fig. 3C). The yield of CPE increases from 20.4% to
61.5% with the H2 pressure from 2.0 to 6.0 MPa. Under a low
H2 pressure (2.0 MPa), it shows a similar time-dependent
product curve with that of 4.0 MPa H2 pressure. The synchro-
Synthesis of CPE
In water, when the N2 atmosphere is changed to 4 MPa of H2
pressure, FeZn-P exhibits a similar FA conversion, indicating
that the hydrolysis of FA is not affected. Surprisingly, the pro- nous increase of HCP and CPE indicates their synthesized
reactions are parallel (Fig. S8†). To explore the reaction mecha-
ducts contained a 52.3% yield of CPE is generated in addition
nism, HCP was used as the reactant but it cannot be trans-
to the HCP yield of 42.0% after 6 h (Fig. 3A, B and Fig. S4B†).
The experiment was repeated more than five times. In com- formed without a catalyst under an H2 pressure of 4.0 MPa
(Table S1†). After adding FeZn-P, CPE is generated. It clearly
proves that FeZn-DMC also can be used as a hydrogenation
parison to the N2 atmosphere, the carbon loss decreases from
56.2% to 4.6% under the H2 atmosphere. Because HCP is
unstable under acidic reaction conditions, the transformation catalyst in addition to an acid catalyst. However, compared
with FA reaction with a CPE yield of 52.3%, the yield is rela-
tively lower (38.3%) with HCP as the reactant under the same
of HCP to CPE is presumed to occur via hydrogenation and
then dehydration. Whereas, CPE is stable and cannot be
further hydrogenated. The fact that HCP is active and CPE is reaction conditions. In addition to the HCP pathway, the PED
intermediate can also be converted to CPE via hydrogenation,
inactive for hydrogenation may be ascribed to the conjugation
intramolecular aldol condensation, and dehydration. Because
effect of the hydroxyl group, which increases the electron
density in the CvC bond and facilitates the hydrogenation PED plays a crucial role in the formation of humins, the PED
reaction.54 When H2 is replaced with deuterium hydrogen D2,
hydrogenation route can also be used to explain the low
humin amount observed under H2 conditions.
isotopic labeling investigations with detailed mass fragmenta-
Subsequently, other acidic catalysts were also employed in
the reaction under an H2 atmosphere. Amberlyst-15 and Al-
MCM-41 only exhibit acid catalysis abilities and cannot cata-
lyze the synthesis of CPE (Fig. 3D). Meanwhile, Cr-MIL-101, Fe-
MIL-100, FeZn, and FeZn-P exhibit considerable CPE yield,
indicating the Lewis acidic site is also important for the hydro-
genation reaction. Among of them, FeZn and FeZn-P with the
highest CPE yield exhibit the strongest both acid and hydro-
genation catalytic abilities. In the H2 atmosphere with a mono-
phasic water solvent, the conversion tendency of various cata-
lysts is the same as that in an N2 atmosphere with a water/
n-hexane bi-phasic solvent (Fig. 2E and 3D). Therefore, H2
does not affect the hydrolysis step of FA, but does affect the
subsequent hydrogenation steps. Similar to the results
obtained for acid catalysis to synthesize HCP, FeZn-P exhibits
higher FA conversion of 96.0% than FeZn (90.3%), because of
the presence of more acid sites on the surface. Meanwhile, the
selectivity of CPE over FeZn-P is also higher than that of FeZn,
which indicates that the catalytic hydrogenation active site is
on the surface of the crystal. Table S1† confirm that FeZn-P
has a stronger hydrogenation ability than FeZn when HCP acts
as a reactant. In addition, a rapid, hot-catalyst filtration test
was performed, and no further generation of CPE in the filtrate
was found after the removal of FeZn-P (Fig. S10†), which
proves the true heterogeneously catalytic nature of this reac-
tion. The recyclability of DMC was evaluated. The catalyst was
centrifuged after the reaction, washed with ethanol, and used
for the next run. As shown in Fig. 3E, after 5 runs, the conver-
sion and yield remain at 90.5% and 52.6%, respectively,
without obvious degradation compared to the first run. The N2
adsorption–desorption isotherm of reused FeZn-P shows an
unclosed loop, which can be ascribed to the structural degra-
dation of pores (Fig. 1C).55 The BET surface area and pore
Fig. 3 (A) Proposed reaction pathway for CPE preparation, (B) time-
dependent product distribution using FeZn-P as catalyst, (C) effect of H2
pressure, (D) catalytic performance of several acid catalysts, and (E) re-
cycling performance of FeZn-P. Reaction conditions: FA (10.2 mmol),
catalyst (0.1 g), water (40 mL), 150 °C (B, D and E) 4.0 MPa H2 pressure,
reaction time 6 h. (C) FeZn-P catalyst, reaction time 6 h.
volume are decreased to 28.6 m2 g−1 and 0.048 cm3 g−1
,
respectively (Table S2†). Meanwhile, XRD results also show a
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
slight degradation of the crystalline structure after recycling. ation. Then, olefin (PED, HCP) insertion occurs at the Zn–H
The slight decrease of catalytic activity can be attributed to the bond followed by proton transfer to the olefin. Ultimately, the
degradation of specific surface area and crystalline structure hydrogenation compound is desorbed, and the catalytic cycle
(Table 1 and Fig. S11†). The chemical compositions of the is complete. The cycle does not change the Zn oxidation state.
used catalysts are well preserved as shown by element analysis Based on the structure of FeZn-DMC, the DFT simulation
(Fig. 1C, Table 2 and Fig. S11†), showing their stability under results indicate that the terminal two-coordinative Zn2+ site
the current reaction conditions.
(Zn(N)2) is unstable but tri-coordinative Zn2+ site (Zn(N)3) is
Recently, a single-site Zn2+ was supported on silica, and the stable (Fig. S13†). Compared with the HCP reaction conducted
Zn/SiO2 catalyzed propylene hydrogenation.56–59 The coordina- under 4.0 MPa H2 pressure, a lower CPE yield (38.3% vs.
tively unsaturated Zn2+ activates H2 via a non-redox (metal) 13.5%) is obtained under 2.0 MPa (Fig. S9†). It is shown that
mechanism consisting of heterolytic cleavage of H–H and C–H the H2 pressure have a great effect on the product distribution
bonds in contrast to the typical hemolytic mechanism in the (Table S1†). As the pressure increases, the yield of CPE
metal-catalyzed hydrogenation reaction. In addition, some Zn– increases gradually. Combined with the result of FA reaction,
N–C compounds with an acid–base pair can hydrogenate we can infer that the H2 pressure does strongly affect the
imines via heterolytic cleavage of H2.60,61 Currently, many hydrogenation activity and hydrogen heterolysis is rate-limit-
studies of homogeneous catalysis have been reported for ing. The FTIR spectra of hydrogen-activated catalyst shows a
hydrogen heterolysis by frustrated Lewis pairs, such as metal- sight new peak at 1734 cm−1, which is assigned to the Zn–H
free, non-noble metal, and noble metal-based catalysts.62–65 bond (Fig. S14†).66 The DFT simulation indicates that the
Due to their convenient processing and separation from the energy barrier for H2 heterolysis on Zn(N)3 is 28.4 kcal mol−1
reaction solution, heterogeneous catalysts are favorable for Therefore, heterolysis hydrogenation is possible over Zn(N)3.
process implementation.
The N-coordinatively unsaturated Zn2+ not only coordinates to
.
FeZn and FeZn-P exhibit the catalytic hydrogenation ability. a lone electron pair for acid-catalyzed reactions but also
To further confirm the importance of the coordinatively unsa- attacks the H–H covalent bond for hydrogenation reactions.
turated Zn2+ in the Zn–N–C structure of the catalyst surface,
CoZn-DMC consistent with K2Zn3[Co(CN)6]2 was successfully
synthesized (Fig. S12†). The XRD pattern indicates a Fm3 m Conclusions
space structure, where Zn2+ hexa-coordinates with four NuC
In summary, we propose a novel synthesis route for 4-hydroxy-
bonds and two water molecules. It is important to note that
2-cyclopentenone and 2-cyclopentenone from biomass-derived
ZnCo-DMC cannot catalyze the acid-catalyzed rearrangement
furfuryl alcohol via double-metal cyanide catalysis. Under an
N2 atmosphere in a water-organic bi-phasic solvent medium, a
reaction for HCP, but it can catalytically convert HCP to CPE
with 16.5% yield via hydrogenation reaction (Table S1†). This
DMC with pure Lewis acidity allows for highly selective syn-
result indicates that the catalytic active site of hydrogenation is
thesis of HCP compared with conventional acid catalysts.
attributed to the coordinatively unsaturated Zn2+ in the Zn-N–
FeZn-P with stronger acidity and a larger surface area exhibits
C structure. To best of our knowledge, DMCs catalysts are
the highest yield (88.2%). Under an H2 atmosphere in water,
widely reported acid catalyst, but have not been employed as a
FeZn-P exhibits a CPE yield of more than 61.5% yield via
hydrogenation catalyst, not mentioned to as the bifunctional
bifunctional catalysis involving both acid and hydrogenation
catalyst for both acid and hydrogenation catalysis as reported
catalysis. The DFT simulation indicates that the tri-coordina-
herein.
tively unsaturated Zn2+ site (Zn(N)3) is used as the acid and
For hydrogenation catalysis, the catalytic hydrogenation
hydrogenation active site. Furthermore, this catalyst exhibits
mechanism is proposed in Fig. 4. H2 heterolytically reacts at
great stability and can be reused at least five times. This work
the coordinatively unsaturated Zn–N frustrated Lewis pairs,
producing Zn–H (hydride ion) and N–H (proton). The heteroly-
not only demonstrates a highly efficient catalytic reaction
system to HCP but also provides a green and novel synthesis
tic dissociation of hydrogen can be ascribed to the electron
route to CPE. In addition, it seems opens a new door for DMC
transfer from Zn to N atoms resulting in the charge polariz-
hydrogenation catalysis via heterolytic cleavage of H–H bond
on frustrated Lewis pairs.
Conflicts of interest
The authors declare no conflict of interests.
Acknowledgements
The authors appreciate the supports from the National Natural
Fig. 4 Proposed catalytic hydrogenation mechanism.
Science Foundation of China (21878138, 21666021, 21706112),
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
Postdoctoral Science Foundation of China (2017M622104, 23 Y. Wang, S. Sang, W. Zhu, L. Gao and G. Xia, Chem. Eng. J.,
2018T110660) and the new faculty startup funds from both
Nanchang University and Arizona State University.
2016, 299, 104–111.
24 X. Li, Q. Deng, S. Zhou, J. Zou, J. Wang, R. Wang, Z. Zeng
and S. Deng, J. Catal., 2019, 378, 201–208.
25 Y. Yang, Z. Du, Y. Huang, F. Lu, F. Wang, J. Gao and J. Xu,
Green Chem., 2013, 15, 1932–1940.
26 X.-L. Li, J. Deng, J. Shi, T. Pan, C.-G. Yu, H.-J. Xu and Y. Fu,
Green Chem., 2015, 17, 1038–1046.
27 M. Hronec, K. Fulajtárová, I. Vavra, T. Sotak, E. Dobrocka
and M. Micusik, Appl. Catal., B, 2016, 181, 210–219.
Notes and references
1 W. Xie, C. J. Mirocha and Y. Wen, J. Nat. Prod., 1995, 58,
124–127.
2 A. M. Rodrguez, R. D. Enriz, L. N. Santagata, E. A. Juregui,
M. J. Pestchanker and O. S. Giordano, J. Med. Chem., 1997, 28 R. Fang, H. Liu, R. Luque and Y. Li, Green Chem., 2015, 17,
40, 1827–1834. 4183–4188.
3 H.-L. Qin, J. Leng, C.-P. Zhang, I. Jantan, M. W. Amjad, 29 G. S. Zhang, M. M. Zhu, Q. Zhang, Y.-M. Liu, H.-Y. He and
M. Sher, M. Naeem-ul-Hassan, M. A. Hussain and
S. N. A. Bukhari, J. Med. Chem., 2016, 59, 3549–3561.
4 R. A. Hill and A. Sutherland, Nat. Prod. Rep., 2017, 34,
1180–1184.
5 G. Desimoni, G. Faita and P. Quadrelli, Chem. Rev., 2018,
118, 2080–2248.
6 Y. Ito, T. Hirao and T. Saegusa, J. Org. Chem., 1978, 43,
1011–1013.
7 R. C. Larock, T. R. Hightower, G. A. Kraus, P. Hahn and
D. Zheng, Tetrahedron Lett., 1995, 36, 2423–2426.
8 J.-Q. Yu, H.-C. Wu and E. J. Corey, Org. Lett., 2005, 7, 1415–
1417.
Y. Cao, Green Chem., 2016, 18, 2155–2164.
30 S. K. R. Patil and C. R. F. Lund, Energy Fuels, 2011, 25,
4745–4755.
31 T. D. Swift, H. Nguyen, Z. Erdman, J. S. Kruger, V. Nikolakis
and D. G. Vlachos, J. Catal., 2016, 333, 149–161.
32 D. Alezi, Y. Belmabkhout, M. Suyetin, P. M. Bhatt,
L. J. Weseliński, V. Solovyeva, K. Adil, I. Spanopoulos,
P. N. Trikalitis, A.-H. Emwas and M. Eddaoudi, J. Am.
Chem. Soc., 2015, 137, 13308–13318.
33 X. Li, J. Liu, A. I. Rykov, H. Han, C. Jin, X. Liu and J. Wang,
Appl. Catal., B, 2015, 179, 196–205.
34 G.-W. Xu, Y.-P. Wu, W.-W. Dong, J. Zhao, X.-Q. Wu, D.-S. Li
and Q. Zhang, Small, 2017, 13, 1602996.
9 T. Baba, K. Nakano, S. Nishiyama, S. Tsuruya and
M. Masai, J. Chem. Soc., Chem. Commun., 1989, 22, 1697– 35 M.-X. Wu and Y.-W. Yang, Adv. Mater., 2017, 29, 1606134.
1699.
36 N. T. T. Nguyen, H. Furukawa, F. Gándara, C. A. Trickett,
H. M. Jeong, K. E. Cordova and O. M. Yaghi, J. Am. Chem.
Soc., 2015, 137, 15394–15397.
10 E. D. Mihelich and D. J. Eickhoff, J. Org. Chem., 1983, 48,
4135–4137.
11 G.-F. Pan, X.-Q. Zhu, R.-L. Guo, Y.-R. Gao and Y.-Q. Wang, 37 G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange,
Adv. Synth. Catal., 2018, 360, 4774–4783.
12 M. A. Mellmer, J. M. R. Gallo, D. M. Alonso and
J. A. Dumesic, ACS Catal., 2015, 6, 3354–3359.
13 G. M. G. Maldonado, R. S. Assary, J. Dumesic and
L. A. Curtiss, Energy Environ. Sci., 2012, 5, 6981–6989.
J. Dutour, S. Surblé and I. Margiolaki, Science, 2005, 309,
2040–2042.
38 P. Küsgens, M. Rose, I. Senkovska, H. Fröde, A. Henschel,
S. Siegle and S. Kaskel, Microporous Mesoporous Mater.,
2009, 120, 325–330.
14 F. M. A. Geilen, B. Engendahl, A. Harwardt, W. Marquardt, 39 J. Sebastian and D. Srinivas, Appl. Catal., A, 2014, 482, 300–308.
J. Klankermayer and W. Leitner, Angew. Chem., Int. Ed., 40 R. Srivastava, D. Srinivas and P. Ratnasamy, J. Catal., 2006,
2010, 49, 5510–5514.
15 R. Rinaldi and F. Schüth, Energy Environ. Sci., 2009, 2, 610– 41 J. K. Satyarthi, D. Srinivas and P. Ratnasamy, Energy Fuels,
626. 2010, 24, 2154–2161.
16 M. Hronec, K. Fulajtárová and T. Soták, J. Ind. Eng. Chem., 42 J. Sebastian and D. Srinivas, Chem. Commun., 2011, 47,
2014, 20, 650–655. 10449–10451.
17 G. Li, N. Li, M. Zheng, S. Li, A. Wang, Y. Cong, 43 A. Peeters, P. Valvekens, F. Vermoortele, R. Ameloot,
241, 34–44.
X. Wang and T. Zhang, Green Chem., 2016, 18, 3607–
3613.
C. Kirschhock and D. D. Vos, Chem. Commun., 2011, 47,
4114–4116.
18 K. Ulbrich, P. Kreitmeier and O. Reiser, Synlett, 2010, 2037– 44 J.-J. Zou, Y. Xu, X. Zhang and L. Wang, Appl. Catal., A, 2012,
2040. 421–422, 79–85.
19 K. Ulbrich, P. Kreitmeier, T. Vilaivan and O. Reiser, J. Org. 45 B. Delley, J. Chem. Phys., 1990, 92, 508–517.
Chem., 2013, 78, 4202–4206. 46 B. Delley, J. Chem. Phys., 2000, 113, 7756–7764.
20 J. Li, T. S. Ahmed, C. Xu, B. M. Stoltz and R. H. Grubbs, 47 W. Shi, L. Zhang, Z. Ni, X. Xiao and S. Xia, RSC Adv., 2014,
J. Am. Chem. Soc., 2019, 141, 154–158. 4, 27003–27012.
21 G.-S. Zhang, M.-M. Zhu, Q. Zhang, Y.-M. Liu, H.-Y. He and 48 J. A. White, D. M. Bird, M. C. Payne and I. Stich, Phys. Rev.
Y. Cao, Green Chem., 2016, 18, 2155–2164. Lett., 1994, 73, 1404–1407.
22 Y. Yang, Z. Du, Y. Huang, F. Lu, F. Wang, J. Gao and J. Xu, 49 N. Govind, M. Petersen, G. Fitzgerald, D. King-Smith and
Green Chem., 2013, 15, 1932–1940.
J. Andzelm, Comput. Mater. Sci., 2003, 28, 250–258.
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
50 C. A. Emeis, J. Catal., 1993, 141, 347–354.
51 D. W. Kim, H. G. Kim and D. H. Cho, Catal. Commun.,
2016, 73, 69–73.
59 N. M. Schweitzer, B. Hu, U. Das, H. Kim, J. Greeley,
L. A. Curtiss, P. C. Stair, J. T. Miller and A. S. Hock, ACS
Catal., 2014, 4, 1091–1098.
52 Y. Zhong, Q. Deng, P. Zhang, J. Wang, R. Wang, Z. Zeng 60 S. Werkmeister, S. Fleischer, S. Zhou, K. Junge and
and S. Deng, Fuel, 2019, 240, 270–277. M. Beller, ChemSusChem, 2012, 5, 777–782.
53 H. Li, Q. Deng, H. Chen, X. Cao, J. Zheng, Y. Zhong, 61 P. Jochmann and D. W. Stephan, Angew. Chem., Int. Ed.,
P. Zhang, J. Wang, Z. Zeng and S. Deng, Appl. Catal., A,
2019, 580, 178–185.
54 H. Liu, Y. Li, R. Luque and H. Jiang, Adv. Synth. Catal.,
2011, 353, 3107–3113.
2013, 52, 9831–9835.
62 Á. Gyömöre, M. Bakos, T. Földes, I. Pápai, A. Domján and
T. Soós, ACS Catal., 2015, 5, 5366–5372.
63 Y. Ding, X. Huang, X. Yi, Y. Qiao, X. Sun, A. Zheng and
D. S. Su, Angew. Chem., Int. Ed., 2018, 57, 13800–13804.
55 M. Kruk and M. Jaroniec, Adsorption, 2000, 6, 47–51.
56 A. L. Dent and R. J. Kokes, J. Am. Chem. Soc., 1969, 91, 64 H. Bauer, M. Alonso, C. Fischer, B. Rçsch, H. Elsen and
7207–7208. S. Harder, Angew. Chem., Int. Ed., 2018, 57, 15177–15182.
57 A. L. Dent and R. J. Kokes, J. Am. Chem. Soc., 1970, 92, 65 T. Mahdi and D. W. Stephan, Angew. Chem., Int. Ed., 2015,
1092–1093. 54, 8511–8514.
58 W. C. Conner, R. A. Innes and R. J. Kokes, J. Am. Chem. 66 P. Jochmann and D. W. Stephan, Angew. Chem., Int. Ed.,
Soc., 1968, 90, 6858–6858.
2013, 52, 9831–9835.
This journal is © The Royal Society of Chemistry 2020
Green Chem.