Double metal cyanides as heterogeneous Lewis acid catalysts for nitrile synthesis: Via acid-nitrile exchange reactions

Article abstract of DOI:10.1039/c9cc05382d

A series of transition metal-based double metal cyanides (DMCs) were studied as catalysts for the synthesis of nitriles via acid-nitrile exchange reaction. The best nitrile yields were obtained with Co₃[Co(CN₆)]₂ DMC, which proved to be a versatile, stable, reusable and highly active catalyst, exhibiting an activity comparable to that of homogeneous Lewis acid catalysts.

Full text of DOI:10.1039/c9cc05382d

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Double metal cyanides as heterogeneous Lewis
acid catalysts for nitrile synthesis via acid-nitrile
exchange reactions†
Cite this: Chem. Commun., 2019,
5, 12984
5
Received 12th July 2019,
Accepted 8th October 2019
a
b
a
b
Carlos Marquez,
Matthieu Corbet, Simon Smolders,
Philippe Marion* and
a
Dirk De Vos
*
DOI: 10.1039/c9cc05382d
rsc.li/chemcomm
A series of transition metal-based double metal cyanides (DMCs) the reaction equilibrium by cyclic amide formation. Acid-nitrile
were studied as catalysts for the synthesis of nitriles via acid-nitrile exchange proceeds without catalysts at high temperatures or in
exchange reaction. The best nitrile yields were obtained with the presence of Brønsted or Lewis acids at lower temperatures
3 6 2
Co [Co(CN )] DMC, which proved to be a versatile, stable, reusable (Scheme 1). However, to the best of our knowledge there are
and highly active catalyst, exhibiting an activity comparable to that no reports of a heterogeneously catalyzed acid-nitrile exchange
of homogeneous Lewis acid catalysts.
reaction; which would be desired for industrial applications,
where catalysts recovery and reusability is an important factor.
Having obtained the best catalytic results with Lewis acid
Nitriles are a very important type of organic compound, found
to be useful intermediates for the synthesis of amines, amides, catalysts like AlCl and Al(OTf) , we decided to study the use of
6i
3
3
carbonyl compounds and heterocycles. Furthermore, the –CN double metal cyanides (DMCs) as heterogeneous catalysts for
motif is present in many useful compounds, like methyl the acid-nitrile exchange. DMCs are purely Lewis acid catalysts
cyanoacrylate, nitrile rubber and numerous valuable pharma- industrially used in the ring opening polymerization (ROP) of
1
8
ceutical products. Several routes for the synthesis of nitriles epoxides for obtaining polyols. Recently, several studies have
2
have been reported, such as the Kolbe nitrile synthesis, the focused on expanding their catalytic properties, applying
Sandmeyer reaction and the Wittig reaction. Furthermore, them in diverse organic reactions, such as hydroamination,
more recently, nitriles have also been obtained from alcohols esterification,
under oxidant-free conditions in a flow reactor. However, these they consist of two different transition metal atoms (or 2 atoms
synthetic procedures usually require multiple complicated steps, in different oxidation states of the same transition metal)
generate large amounts of waste, and use hazardous compounds bonded by a cyanide group. They can be defined by the general
and/or ammonia. Another common method of nitriles prepara- formula M
tion consists in the conversion of carboxylic acids to nitriles,
usually through a reaction with ammonia at high temperature
3
4
9
1
0
3
11
12
A coupling, among others. Structurally,
5
6
0
00
0
00
00
u
[M (CN)
] ꢀxH
n v 2
O (M –M DMCs), where typically,
M is a divalent metal (v = 2) and M a trivalent metal (u = 3).
1b
0
7
(
B300 1C) via amide formation (ammoniation–dehydration).
A more straight-forward transformation within this group of acid
6
to nitrile reactions is definitely the acid-nitrile exchange.
Acid-nitrile exchange involves a functional exchange between
the two reactants, a carboxylic acid and a sacrificial nitrile, thus
representing an ammonia-free alternative to the synthesis of
nitriles from carboxylic acids. To displace the acid-nitrile equili-
brium, the sacrificial nitrile is normally a low boiling point nitrile
6g
6h
used in large excess or a short chain dinitrile, which displaces
a
Centre for Membrane Separations, Adsorption, Catalysis and Spectroscopy for
Sustainable Solutions, KU Leuven, Celestijnenlaan 200F, 3001 Leuven, Belgium.
E-mail: dirk.devos@kuleuven.be
b
Solvay Recherches & Innovation Centre de Lyon, 85 rue des fr `e res Perret, BP62,
6
9192 Saint Fons Cedex, France. E-mail: philippe.marion@solvay.com
Electronic supplementary information (ESI) available: Characterization of the
DMCs and additional catalytic data. See DOI: 10.1039/c9cc05382d
†
Scheme 1 Acid-nitrile exchange reactions.
1
2984 | Chem. Commun., 2019, 55, 12984--12987
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Communication
ChemComm
0
This u–v combination usually leads to the DMC adopting a face the M (Table S2 and Fig. S5, ESI†). Even though these materials
1
3
center cubic structure, nevertheless, the occurrence of other give rise to a typical type I isotherm, corresponding to their
1
4
phases has also been observed. In the present work, the first microporous nature, the N2 isotherms of the Cu–Co DMC,
heterogeneously catalyzed synthesis of nitriles via the acid- Co–Co DMC and Fe–Co DMC exhibit a hysteresis loop. This
nitrile exchange reaction is reported.
A series of M –Co DMCs based on earth-abundant transition crystal voids or by aggregation of zones with high concentration
metals were synthesized by simple mixing of aqueous solutions of [Co(CN)
hints at the presence of mesopores, most likely due to inter-
0
3
ꢁ
18
6
]
voids. The acid properties of selected samples
0
of a M salt (FeCl
2
ꢀ4H
2
O, CoCl
2
, NiCl
[Co(CN)
PTMEG (poly(tetramethylene ether)glycol) were added as com- 1450, 1490 and 1610 cm are attributed to the presence of
2
ꢀ6H
2
O, CuCl
2
ꢀ2H
2
O or were studied with pyridine adsorption followed by FTIR
00
ZnCl
2
) and a M cyanide salt (K
3
6
]). tert-Butanol and spectroscopy (Table S2 and Fig. S6, ESI†). While the bands at
ꢁ
1
ꢁ
1
plexing and co-complexing agents, respectively. These organic Lewis acid sites, the absence of bands at 1639 and 1545 cm
additives have been found to increase the activity of DMCs in indicates that the DMCs contain no Brønsted acid sites and are
9
,15
19
ROP and hydroamination reactions.
ICP analyses show a purely Lewis acid catalysts.
0
M : Co ratio of B1.6 for all the DMC samples (Table S1, ESI†),
As a test reaction, the acid-nitrile exchange between 2-phenyl-
which is higher than the stoichiometric 1.5, indicating a slight butyric acid (1a) and glutaronitrile was selected. Considering that
0
excess of M in the structure. The crystallinity of the DMC the benzylic proton of the former is easily discerned in its
1
materials was studied by powder X-ray diffraction (PXRD,
H NMR spectrum and that the latter aids in the shift of the
Fig. S1, ESI†). The reflections around 151, 171, 24.51 and 351 reaction equilibrium by cyclic imide formation, the product
y proved the presence of a cubic lattice structure with space analysis of this test reaction is relatively straightforward. Batch
2
0
13
group Fm3%m, typical of M [Co(CN )] -type DMCs. Likewise, reactions were carried out with the DMC catalysts at 215 1C and
3
6 2
selected scanning electron microscopy (SEM, Fig. S2, ESI†) 165 1C. The results are presented in Table 1 and Tables S3, S4
images of Co–Co DMC reveal small crystalline nanoparticles (ESI†). Blank experiments already give a considerable yield of
(
o1 mm) exhibiting a cubic morphology, consistent with its 2-phenylbutyronitrile (2a) after 16 h reaction time (entry 1);
crystal system. Thermogravimetric analyses (TGA, Fig. S3, ESI†) nonetheless, the use of a catalyst is necessary to form the product
revealed that the materials are stable even above 250 1C, with in higher yields. The use of purely Lewis DMC catalysts consi-
the cyanide ligands being burnt off as (CN)
2
between 300 and derably increased the yield of nitrile 2a, as yields of 465% after
1
6
350 1C. This was also confirmed by in situ high temperature 16 h reaction time were obtained irrespective of the metal
PXRD measurements. As can be seen in Fig. 1, the Co–Co DMC combination (entries 2–6). The tentative reaction mechanism
6
maintains its crystalline structure (cubic phase) until B270 1C. has been previously studied and probably involves the activa-
After this temperature, the pattern shows the characteristic tion of the nitrogen atom of the CN moiety by the Lewis acid
reflections of the cubic phase of Co
3 4
O (JCPDS card no. 00-042- site, forming a nitrilium ion, which in turn, is attacked by
1467). Both PXRD and TGA analyses thus indicate that the the carboxylic acid to form an O-acyl imidate (Scheme 2).
samples can endure the temperatures (215 1C) necessary for the This intermediate undergoes Mumm rearrangement to give an
acid-nitrile exchange, making DMCs potential heterogeneous imide, which in this particular case, transforms into an acyl
catalysts for this process. The Fourier-transformed infrared amidine. In this step, the second nitrile moiety is activated by the
(
FTIR) spectra of the samples exhibited a blue shift in the
0
0
n(CN) compared to the initial M cyanide salt (K [Co(CN) ])
3
6
(
Fig. S4, ESI†). This shift is dependent of the M’ and is an
Table 1 Yield (%) of 2-phenylbutyronitrile (2a) obtained with different
0
indication of the formation of M –CN–Co bridges, since the
electron donation from the CN group to both metals is known acid (1a) and glutaronitrile
to shift the n(CN) to higher wavenumbers. The textural
catalysts in the acid-nitrile exchange reaction between 2-phenylbutyric
a
1
7
properties of the DMCs were also found to be dependent on
b
Nitrile 2a yield (%)
Entry
Catalyst
2.5 h
5 h
16 h
1
2
3
4
5
6
7
8
—
12
41
45
35
39
40
50
42
31
51
58
45
50
49
65
53
52
74
86
65
70
67
90
81
Fe–Co DMC
Co–Co DMC
Ni–Co DMC
Cu–Co DMC
Zn–Co DMC
AlCl
Zr-BEA
3
a
Reaction conditions: 0.375 mmol of 2-phenylbutyric acid (1a), 6 mmol
b
Fig. 1 PXRD patterns of Co–Co DMC during heating from 200 1C of glutaronitrile, 10 mol% of catalyst at 215 1C. Determined by
1
to 350 1C.
H NMR spectroscopy.
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 12984--12987 | 12985

View Article Online
ChemComm
Communication
be obtained with Co–Co DMC as catalyst, a notable increase in
comparison to the yield achieved with conventional heating
under identical temperature conditions after 5 h reaction time.
As can be observed in Table 1, the sample Co–Co DMC
exhibits the highest activity among all the studied DMC
materials. In the light of this, we decided to employ this catalyst
in further experiments to expand the substrate scope of the
DMCs. The acid-nitrile exchange reaction was carried out in the
presence of 10 mol% of Co–Co DMC with various carboxylic
acids and the results are presented in Table S5 (ESI†).
A comparison between entries 1–3 shows that when the phenyl
substituent is further away from the –COOH group, the nitrile
yields are lowered. This reflects that the overall acidity of the
acid is increased with the presence of a phenyl group on
the a-carbon; which in turn, facilitates the activation of the
nitrogen of the nitrile moiety, increasing the reaction rate.
A similar effect is observed when comparing entries 4 and 5.
Scheme 2 Tentative reaction mechanism for the acid-nitrile exchange
with glutaronitrile as a sacrificial nitrile.
Lewis acid catalyst. Finally, this four-membered-ring inter-
mediate spontaneously opens to yield the desired nitrile and
the cyclic imide. The nitrile 2a yield was found to be highly
0
dependent on the M . In order to derive structure–activity rela-
tions, the physicochemical properties of the DMC catalysts,
namely Lewis acidity and external surface area (Sext), were corre-
lated with their catalytic activity. For the Lewis acidity, the
amount of pyridine adsorbed on the Lewis acid sites (LAS) per
g of DMC was plotted versus the nitrile 2a yield (Fig. S7, ESI†).
No clear trend was observed, so we can infer that the acidity of
the cation does not have a direct effect on its activity. Another
determining factor of the differences in activity between the
samples may be the Sext of the DMC materials, since it is known
that the pores in the DMC framework are usually too narrow
2
In this case, the presence of a NO substituent in the para
position of the phenyl group results in a higher nitrile yield.
6f
This is in line with the findings of Toland and Ferstandig, who
suggested that electron withdrawing substituents accelerated
the acid-nitrile exchange reaction. Remarkably, the catalyst
Co–Co DMC is even able to form in respectable yields nitriles
bearing quaternary centers, such as pivalonitrile (entry 6). This
moiety is of particular interest as it is present in several
1e
pharmaceutical products. This substrate scope showcases
Co–Co DMC as a versatile catalyst for the synthesis of nitriles
via acid-nitrile exchange.
9
,12
(B5 Å)
to allow the diffusion of large organic molecules like
the acid 1a. In this sense, a fairly linear correlation between the
yield of the nitrile 2a after 2.5 h reaction time and the Sext of the
DMC catalysts was obtained (Fig. 2). Furthermore, the best yields
among all the DMCs were obtained with the samples Co–Co DMC
and Fe–Co DMC (Table 1, entries 2 and 3), which are the two
In all these examples, one equivalent of a cyclic imide
glutarimide) is formed concomitantly with the production of
(
the desired nitrile. While the imide can be eliminated relatively
easily by flash chromatographic purification, it is interesting to
investigate alternative methods to displace the equilibrium
of the acid-nitrile exchange reaction. In this sense, (phenyl-
sulfonyl)acetonitrile was studied as sacrificial nitrile. It has
been reported that the use of such nitrogen donor can allow
the decarboxylation of the formed carboxylic acid, constituting
an alternative irreversible step towards the production of the
2
ꢁ1
materials with the highest S (327 m g for Co–Co DMC and
ext
2
ꢁ1
3
11 m
sample Co–Co DMC (86% after 16 h reaction time, entry 3) is
comparable to that of homogeneous Lewis acids, such as AlCl
entry 7), under the same reaction conditions. Additionally,
g
for Fe–Co DMC). Remarkably the activity of the
3
(
the reaction rate can be markedly accelerated by microwave
irradiation. After only 0.5 h reaction time, 78% of nitrile 2a can
6i
desired nitrile. As can be observed in Fig. S8 (ESI†), Co–Co
DMC was able to catalyze the acid-nitrile exchange between the
acid 1a and (phenylsulfonyl)acetonitrile, forming the nitrile 2a
in substantial amounts and methyl phenyl sulfone – a lower
boiling point side product, which would simplify the purification.
This represents a promising result which could lead to additional
studies on heterogeneously catalyzed decarboxylative acid-nitrile
exchange reactions.
The heterogeneous nature of Co–Co DMC was demonstrated
by a hot filtration test (Fig. S9, ESI†). To this end, a sample of
the product mixture was taken after 2.5 h reaction time and the
catalyst was separated by filtration without cooling. This hot
filtrate was left stirring for an additional 2.5 and 13.5 h.
A considerably lower nitrile 2a yield was obtained, with the
small increase attributed to the catalyst-free conversion of
the acid 1a. This suggests that no active species leach from
the heterogeneous catalyst, which was further proven by ele-
mental analyses (ICP) performed after reaction (Table S1, ESI†).
Fig. 2 Linear correlation between Sext and yield of nitrile (2a) after 2.5 h
reaction time.
1
2986 | Chem. Commun., 2019, 55, 12984--12987
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Communication
ChemComm
The sample Co–Co DMC kept its activity after three consecutive
reaction runs (Fig. S10, ESI†), indicating that it is a stable and
recyclable catalyst. Finally, Fig. S11 (ESI†) shows the PXRD
pattern of Co–Co DMC before and after reaction. No notable
changes were observed, which demonstrates that the structure
of the catalyst is preserved at the reaction temperature and after
a lengthy contact time with the reactants and products.
4 (a) M. Matsui and G. Yabuta, Agric. Biol. Chem., 1968, 32, 1044–1045;
(
b) M. Ezawa and H. Togo, Eur. J. Org. Chem., 2017, 2379–2384.
(a) Y. Wang, S. Furukawa and N. Yan, ACS Catal., 2019, 9,
681–6691; (b) Y. Wang, S. Furukawa, Z. Zhang, L. Torrente-
5
6
Murciano, S. A. Khan and N. Yan, Catal. Sci. Technol., 2019, 9, 86–96.
(a) A. Gautier, Compt. Rend., 1868, 67, 1255–1259; (b) E. H. Miller,
J. Am. Chem. Soc., 1894, 16, 433–462; (c) J. A. Mathews, J. Am. Chem.
Soc., 1898, 20, 648–668; (d) R. H. Wiley and W. B. Guerrant, J. Am.
Chem. Soc., 1949, 71, 981–982; (e) F. Becke and T. F. Burger, Liebigs
Ann. Chem., 1968, 716, 78–82; ( f ) W. G. Toland and L. L. Ferstandig,
J. Org. Chem., 1958, 23, 1350–1351; (g) D. Cantillo and C. O. Kappe,
J. Org. Chem., 2013, 78, 10567–10571; (h) D. A. Klein, J. Org. Chem.,
1971, 36, 3050–3051; (i) D. Cartigny, A. Dos Santos, L. El Kaim,
L. Grimaud, R. Jacquot and P. Marion, Synthesis, 2014, 1802–1806.
A. Mekki-Berrada, S. Bennici, J.-P. Gillet, J.-L. Couturier, J.-L. Dubois
and A. Auroux, ChemSusChem, 2013, 6, 1478–1489 and references
therein.
6
0
In conclusion, a series of M –Co DMCs were studied as
catalysts for the synthesis of nitriles via acid-nitrile exchange
reaction. These materials were easily synthesized by mixing
0
aqueous solutions of a M Cl ꢀxH O salt with K [Co(CN) ].
2
2
3
6
7
The sample Co–Co DMC was found to be the most active one,
showing an excellent catalytic activity in a wide range of acid-
nitrile reactions with different carboxylic acids, under thermal
and microwave conditions. Furthermore, the use of (phenyl-
sulfonyl)acetonitrile as sacrificial nitrile showed potential as
alternative for equilibrium displacement (decarboxylative acid-
nitrile exchange). The application of DMCs as purely Lewis
acid, stable and reusable catalysts represents the first report of
a heterogeneously catalyzed acid-nitrile exchange reaction and
can induce further studies on nitrile synthesis via acid-nitrile
exchange and the catalytic properties of these still under-
explored materials.
8
9
(a) J. Milgrom, US Pat., 3278457, 1966; (b) R. J. Herold, US Pat.,
3
278457, 1966.
(a) A. Peeters, P. Valvekens, F. Vermoortele, R. Ameloot,
C. Kirschhock and D. De Vos, Chem. Commun., 2011, 47,
4
114–4116; (b) A. Peeters, P. Valvekens, R. Ameloot, G. Sankar,
C. E. A. Kirschhock and D. De Vos, ACS Catal., 2013, 3, 597–607;
c) C. Marquez, M. Rivera-Torrente, P. P. Paalanen, B. M. Weckhuysen,
F. G. Cirujano, D. De Vos and T. De Baerdemaeker, J. Catal., 2017, 354,
2–99.
0 (a) J. K. Satyarthi, S. Radhakrishnan and D. Srinivas, Energy Fuels,
011, 25, 4106–4112; (b) M. Kotwal, S. S. Deshpande and D. Srinivas,
Catal. Commun., 2011, 12, 1302–1306.
1 (a) C. Marquez, F. G. Cirujano, C. Van Goethem, I. Vankelecom,
D. De Vos and T. De Baerdemaeker, Catal. Sci. Technol., 2018, 8,
(
9
1
1
2
The authors thank Solvay for financial support. F. W. O.
Vlaanderen (Research Foundation Flanders) is acknowledged
for project funding (D. D. V.: research projects) and aspirant
funding (S. S.). D. D. V. thanks KU Leuven for the Metusalem grant
CASAS. The authors thank Thomas Cuypers and Michiel Janssen
2
061–2065; (b) C. Marquez, F. G. Cirujano, S. Smolders, C. Van
Goethem, I. Vankelecom, D. De Vos and T. De Baerdemaeker, Dalton
Trans., 2019, 48, 3946–3954.
2 P. Valvekens and D. De Vos, in New Materials for Catalytic Applications,
ed. V. I. Parvulescu and E. Kemnitz, Elsevier, Amsterdam, 2016, ch. 1,
pp. 1–12 and references therein.
3 (a) M. Verdaguer and G. Girolami, in Magnetism: Molecules to
Materials V, ed. J. S. Miller and M. Drillon, Wiley-VCH, Weinheim,
1
1
(KU Leuven) for NMR support and Miguel Rivera-Torrente (Utrecht
University) for collecting additional XRD patterns.
2005, ch. 9, pp. 283–346; (b) D. F. Mullica, W. O. Milligan, G. W.
Beall and W. L. Reeves, Acta Crystallogr., Sect. B: Struct. Crystallogr.
Cryst. Chem., 1978, 34, 3558–3561; (c) A. Ludi, H. U. Guedel and
M. Ruegg, Inorg. Chem., 1970, 9, 2224–2227; (d) G. W. Beall, W. O.
Milligan, J. Korp and I. Bernal, Inorg. Chem., 1997, 16, 2715–2718.
4 (a) J. C. Wojdeł, S. T. Bromley, F. Illas and J. C. Jansen, J. Mol. Model.,
Conflicts of interest
1
There are no conflicts to declare.
2007, 13, 751–756; (b) J. Roque, E. Reguera, J. Balmaseda,
J. Rodr ´ı guez-Hern ´a ndez, L. Reguera and L. F. del Castillo, Micro-
porous Mesoporous Mater., 2007, 103, 57–71; (c) C. Marquez,
A. Simonov, M. T. Wharmby, C. Van Goethem, I. Vankelecom,
B. Bueken, A. Krajnc, G. Mali, D. De Vos and T. De Baerdemaeker,
Chem. Sci., 2019, 10, 4868–4875.
References
1
(a) A. J. Fatiadi, in Preparation and Synthetic Applications of Cyano
Compounds, ed. S. Patai and S. Z. Rappaport, Wiley, New York, 1983;
(
b) K. Friedrich and K. Wallensfels, in The Chemistry of the Cyano 15 I. K. Lee, J. Y. Ha, C. Cao, D. W. Park, C. S. Ha and I. Kim, Catal.
Group, ed. Z. Rappoport, Wiley-Interscience, New York, 1970; Today, 2009, 148, 389–397.
c) F. F. Fleming, L. Yao, P. C. Ravikumar, L. Funk and B. C. 16 R. Zboril, L. Machala, M. Mashlan and V. Sharma, Cryst. Growth
(
Shook, J. Med. Chem., 2010, 53, 7902–7917; (d) T. Oishi,
Des., 2004, 4, 1317–1325.
K. Yamaguchi and N. Mizuno, Angew. Chem., Int. Ed., 2009, 48, 17 (a) I. Kim, J. T. Ahn, C.-S. Ha, C. S. Yang and I. Park, Polymer, 2003,
6
2
286–6288; (e) D. Nath and F. F. Fleming, Angew. Chem., Int. Ed.,
011, 50, 11790–11793; ( f ) F. Le Vaillant, M. D. Wodrich and
44, 3417–3428; (b) X.-H. Zhang, Z.-J. Huab, S. Chenc, F. Liua,
X.-K. Suna and G.-R. Qi, Appl. Catal., A, 2007, 325, 91–98;
(c) J. Fernandez Bertran, J. Blanco Pascual and E. Reguera Ruiz,
Acta Crystallogr., Sect. A: Found. Crystallogr., 1990, 46, 685–689.
J. Waser, Chem. Sci., 2017, 8, 1790–1800.
L. Friedman and H. Shechter, J. Org. Chem., 1960, 25(6), 877–879.
2
3
(a) J. Kim, H. J. Kim and S. Chang, Angew. Chem., Int. Ed., 2012, 51, 18 L. Boudjema, J. Long, F. Salles, J. Larionova, Y. Guari and P. Trens,
11948–11959; (b) T. Sandmeyer, Ber. Dtsch. Chem. Ges., 1884, 17,
633–1635; (c) C. Galli, Chem. Rev., 1988, 88, 765–792.
Chem. – Eur. J., 2019, 25, 479–484.
19 M. Orazov and M. E. Davis, Chem. Sci., 2016, 7, 2264–2274.
1
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 12984--12987 | 12987