Magnetic Fe3O4 nanoparticles bearing CuI-NHC complexes by an “auto-click” strategy

Article abstract of DOI:10.1016/j.ica.2021.120312

Two copper(I)–NHC complexes bearing an azide group were reacted with alkyne-decorated magnetic Fe₃O₄ nanoparticles without the use of an external copper source. In the case of the least sterically congested complex, the resulting nan

Full text of DOI:10.1016/j.ica.2021.120312

Inorganica Chimica Acta 520 (2021) 120312
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Research paper
Magnetic Fe3O4 nanoparticles bearing CuI-NHC complexes by an
“auto-click” strategy
*
´
´
Kevin Fauche, Federico Cisnetti
´
Universite Clermont Auvergne, CNRS, SIGMA Clermont, ICCF, F-63000 Clermont-Ferrand, France
A R T I C L E I N F O
A B S T R A C T
Keywords:
Two copper(I)–NHC complexes bearing an azide group were reacted with alkyne-decorated magnetic Fe3O4
nanoparticles without the use of an external copper source. In the case of the least sterically congested complex,
the resulting nanoparticles displayed a catalytic activity in copper-catalyzed azide alkyne cycloaddition and
some reusability highlighting the covalent grafting of the molecular catalyst by this mild and simple strategy.
Magnetic nanoparticles
Nanocatalysts
Click chemistry
Azide-alkyne cycloaddition
Copper(I)–NHC complexes
1. Introduction
decided to preliminarily explore its potential for the functionalization of
magnetic nanoparticles bearing alkyne groups [23] and the use of the
N-heterocyclic carbene complexes are nowadays a tool for the
chemist of great significance, among others in the field of catalysis by
metal complexes [1,2]. In the recent years, there has been a strong effort
towards the heterogenization of molecular catalysts based on N-het-
erocyclic carbene complexes, although this endeavor is met with high
challenges. Among the heterogeneous systems actively considered,
magnetic nanoparticles (MNPs) attract a considerable interest due to
their facile separation from a reaction mixture by the use of an external
magnet [3,4]. Several examples of covalent grafting of metal NHC
complexes on magnetic nanoparticles have been reported in the recent
years [5–14].
resulting nanoparticles in catalysis of the azide-alkyne cycloaddition
reaction. In the design of this study, we reasoned that the fact that the
functionalization reaction is conducted with dissolved azide-tagged
NHC complexes and no other reactant besides the nanoparticles en-
sures any catalytic activity is to be ascribed to immobilized complexes
and not to undissolved impurities. This might constitute an asset in
comparison to more conventional strategies in which complexes or li-
gands react in the presence of other reagents with a functionalized
surface bearing the complementary function. To this end, we selected
two NHC complexes with a different shape and steric crowding around
the metal center: C1 and C2 and examined first their reactivity towards
alkyne-functionalized MNPs, then the reactivity of the resulting nano-
catalysts towards the azide-alkyne cycloaddition as a proof of concept of
our novel immobilization strategy (Scheme 1).
Copper-catalyzed azide-alkyne cycloaddition (CuAAC) is one of the
most ubiquitous reactions for novel developments in diverse areas
ranging from polymers to medicinal chemistry [15]. Copper(I) NHC
complexes allow this reaction (see references [13;14] for the use of
copper NHC grafted on nanoparticles as CuAAC catalysts), along with
other important catalytic transformations [16]. Moreover, although not
as widespread as the same reaction from their silver(I) counterparts, the
transmetalation from CuI-NHC complexes allows the access to other
metal-NHC compounds (Au, Pd… ) [17]. In the past years, it has been
shown that copper(I) NHC complexes bearing one or two azide groups
were able to catalyze their own functionalization in solution with an
alkyne reagent without any other reactant or catalyst, albeit in an
intermolecular manner [18–20]. Due to the simplicity of this ‘click’ [21]
reaction and its suitability to heterogeneous conditions [15,22], we
2. Experimental section
2.1. Synthesis of the copper complexes
The ligand precursors L1⋅HCl and L2⋅HCl were prepared by adapting
a previously reported procedure [19] (see ESI for details).
Procedure for the synthesis of copper complexes:
The imidazolinium salt L⋅HCl (2 mmol, 1 eq.) was suspended in
water (20 mL). CuCl (238 mg, 2.4 mmol, 1.2 eq) was added and the
resulting suspension was degassed with argon. Aqueous ammonia (14.2
* Corresponding author.
E-mail address: federico.cisnetti@uca.fr (F. Cisnetti).
https://doi.org/10.1016/j.ica.2021.120312
Received 8 December 2020; Received in revised form 15 February 2021; Accepted 19 February 2021
Available online 25 February 2021
0020-1693/© 2021 Elsevier B.V. All rights reserved.

´
K. Fauche and F. Cisnetti
Inorganica Chimica Acta 520 (2021) 120312
Scheme 1. Azide-tagged NHC complexes and synthesis of heterogenized nanocatalysts.
mol/L as determined by acidimetric titration, 0.85 mL, 12 mmol, 6 eq.)
was added, the reaction vessel was degassed for one more minute and
the mixture was stirred at room temperature for 1 h. After completion,
an extraction with dichloromethane (3 × 20 mL) was performed. The
product was dried over anhydrous Na2SO4 and Na2CO3 and condensed
under vacuum.
was added and the suspension was allowed to settle over a 1.2 T NdFeB
magnet. The supernatant was removed and the nanoparticles were
washed twice with acetone with 10 min shaking for each washing. The
joint organic phase was condensed under vacuum. In the case of recy-
cling experiments, the recovered nanoparticles were dried in vacuo and
reused immediately by adding again the reactants and water.
Work-up for C1: the crude product was dissolved in ~5 mL
dichloromethane (previously treated with Na2CO3) and recrystallized by
the dropwise addition of n-pentane (~30 mL). The resulting solid was
filtered and washed with n-pentane. The recrystallization was repeated a
second time giving pure C1 (418 mg, 48% yield). 1H NMR (CDCl3) δ =
6.98 (s, 2H, HAr), 6.93 (s, 2H, HAr), 3.7 (broad, 4H, CH2), 2.36 (s, 6H, o-
CH3), 2.32 (s, 9H, o-CH3, p-CH3). 13C NMR (DMSO–d6) δ = 202.7
(Ccarbene), 140.0 (CAr), 138.7 (CAr), 137.8 (CAr), 135.2 (CAr), 134.7 (CAr),
134.3 (CAr), 129.7 (CHAr), 119.3 (CHAr), 51.0 (CH2), 50.8 (CH2), 21.0
(CH3), 18.2 (CH3), 18.0 (CH3). ESI-HRMS: calcd. for (C20H23ClCuN5 - Cl
3. Results and discussion
The corresponding imidazolinium salts L1⋅HCl and L2⋅HCl were
obtained by a straightforward synthetic sequence at the gram scale
previously reported for other similar compounds [19]. Dissymmetric
oxalamides were obtained by the reaction of easily prepared iodinated
anilines with a monoamide derivative of oxalyl chloride. After reduction
of the amide groups, an Ullmann-type reaction was performed to
introduce an azide group. The imidazolinium salts were obtained by a
cyclization reaction with triethyl orthoformate. Synthetic details are
given in the Supporting Information.
+ MeCN)+: 437.1510, found: 437.1513; calcd. for (C23H29ClCuN5
–
CuCl + H)+: 334.2026, found: 334.2023. CHN Microanalysis calcd. C:
55.55, H: 5.36, N:16.20; found C: 55.39, H: 5.46, N:16.01.
Finally, the copper(I)–NHC complexes were obtained by the metal-
ation of the imidazolinium salts with aqueous ammonia as basic and
complexing medium and copper(I) chloride [24]. The compounds C1
and C2 were characterized spectroscopically to ensure their identity and
purity. The neutral heteroleptic nature of the complexes NHC-Cu-Cl was
verified by measuring their conductivity in DMSO solution (see ESI for
data) and in the case of C1 by its 13C NMR chemical shift of the carbenic
carbon which is very similar to the reported value for CuCl(SIMes)
(202.7 vs. 202.8 ppm) and thus is diagnostic of the lepticity of the
complex [25]. However, as described in a previous paper by Roland et al.
for silver complexes [26] there may be an equilibration in solution in
coordinating solvents for complex C1, which results in a concentration-
dependent behavior and in the observation of a minor species in the 1H
NMR spectrum of C1, likely a homoleptic [Cu(L1)2](CuCl2) species,
which shares the elemental composition of C1.
Work-up for C2: the crude product was suspended in EtOH and
stirred for 30 min. The resulting solid was filtered and washed with
EtOH giving pure C2 (650 mg, 69% yield). 1H NMR (acetone–d6) δ =
7.28–7.19 (m, 3H, HAr), 7.02 (s, 2H, HAr), 4.29–4.15 (m, 2H, CH2),
4.04–3.98 (m, 2H, CH2), 3.26 (hept, 2H, J = 6.9 Hz, CH), 2.42 (s, 3H,
CH3,Me), 1.37 (d, 6H, J = 6.9 Hz, CH3,iPr), 1.35 (d, 6H, J = 6.9 Hz,
CH3,iPr). 13C NMR (DMSO–d6), δ = 200.8 (Ccarbene), 148.9 (CAr), 140.4
(CAr), 137.7 (CAr), 135.8 (CAr), 131.7 (CAr), 128.7 (CHar), 128.6 (CHAr),
115.0 (CHAr), 53.5 (CH2), 50.7 (CH2), 28.2 (CH), 24.7 (CH3), 23.2 (CH3),
17.4 (CH3). IR (neat):
ν
(cmꢀ 1) = 2960, 2099, 1595, 1481, 1472, 1341,
1323, 1265, 1252, 880. ESI-HRMS: calcd. for (C23H29ClCuN5 - Cl +
MeCN)+: 479.1979, found: 479.1981. CHN Microanalysis calcd. C:
58.09, H: 6.36, N:14.73; found C: 58.08, H: 6.21, N:14.62.
Magnetic nanoparticles functionalized with alkyne groups were
prepared using a 3-step sequence described by Gun’ko without any
modification: preparation of Fe3O4 cores, coating with amino-
propyltriethoxysilane and coupling with propiolic acid promoted by
EDC.[23] The magnetic nanoparticles were subjected to an auto-click
reaction with either of copper(I) complexes (C1 or C2) [19]. The reac-
tion was conducted overnight at 50 ◦C under ultrasonic irradiation. After
the functionalization, a mass similar to the one of the starting nano-
particles was recovered by magnetic decantation followed by several
washes.
2.2. Auto-Click reaction with alkyne-bearing nanoparticles
In a 5 mL round bottom flask, 50 mg of complex C1 or C2 were
dissolved in 2 mL of acetonitrile. 100 mg of alkyne-bearing nano-
particles prepared following a protocol by Gun’ko et al. [23] (see details
in ESI) were added. The resulting suspension was sonicated for 6 h. After
magnetic decanting of the product, the supernatant was removed and
the product was washed with acetonitrile (2 × 2 mL) with 10 min in the
ultrasonic bath for each washing. The nanoparticles were dried in vacuo,
giving 105 mg of MNP1 or 95 mg of MNP2.
TEM images of a batch of functionalized nanoparticles with complex
C1 (MNP1) show that the mean radius of the nanoparticles is about 20
nm as reported by Gun’ko and that the functionalization reaction did not
alter the nanoparticle appearance (ESI). ICP-AES analysis was per-
formed on the nanoparticles after an acidic treatment to uncomplex the
copper(I) ions. This analysis shows that the copper loading on MNP1 is
2.3. Catalysis experiments
1 mmol of benzyl azide, 1 mmol of alkyne and 20 mg of MNP1 or
MNP2 were suspended in 1 mL of water in a Falcon® tube. The mixture
was stirred with a mechanical shaker for 20 h. Acetone (approx. 10 mL)
2

´
K. Fauche and F. Cisnetti
Inorganica Chimica Acta 520 (2021) 120312
and washing (see NMR spectra in the supporting information). Non
activated oct-1-yne (entry 13) resulted in only trace amount of the
product. Hydrophilic alkynes (entries 14,15) resulted in limited
conversions.
N
N
N
N3
MNP1 or MNP2
R
Interestingly we have shown that MNP1 catalysts may be recycled a
limited number of times, which corroborates a covalent grafting of the
complex. Moreover, as the magnetic nanoparticles are compatible with
the metalation reactions in aqueous ammonia used for the complex
synthesis,[24] we performed a “reloading” experiment with MNP1
previously used for 4 catalytic runs (benzyl azide + phenylacetylene).
We were able to show that a significant catalytic activity was restored
(although not quite as good as pristine MNP1) which again correlates
with a covalently linked complex, or - after decomplexation during
multiple catalytic runs - NHC precursor. Concurrently, if alkyne tagged
NHC nanoparticles were subjected to “ammonia metalation conditions”
and isolated no catalytic activity was observed suggesting that the cat-
alytic activity belongs to intact heterogenized NHC complexes or from
Cu+ ions leached from the same grafted complexes [28].
+
R
Scheme 2. Test of the catalytic potential of MNP1 and MNP2 in
a
CuAAC reaction.
about 2.2 mass-%, while MNP2 displays an 8-times lesser loading (while
a blank experiment allows to verify that alkyne-bearing magnetic
nanoparticles prior to the auto-click reaction do not contain any
detectable copper amounts). The metal loading of MNP1 is very similar
(only 8% difference) to the figure reported by Diez-Gonzalez et al. for
copper-NHC MNPs for which the copper was introduced on NHC ligand
precursors previously grafted onto the nanoparticles [13], hinting to an
efficient auto-click reaction in the case of MNP1.
Then, we decided to test the catalytic potential of MNP1 and MNP2.
To this end, we selected the CuAAC reaction with benzyl azide and
phenylacetylene (Scheme 2, R = Ph).
To verify the assumption of lesser activity of C2 which results in
lower copper loading and likely in reduced catalytic activity of MNP2
we performed an auto-click reaction in homogenous conditions as
described in a previous paper [19]. Even though it is not feasible to
reproduce conditions identical to the MNP functionalization experi-
ments (large excess of copper-NHC complex in comparison to the alkyne
counterpart), our results indicate C1 displays a higher reactivity than C2
towards an alkyne (propargyl alcohol in our model experiments) albeit
the reaction results in the formation of multiple minor compounds (most
likely products of ligand rearrangement around the metal center, details
are given in the supporting information). Indeed, this observation is
compatible with either (or both) hypotheses explaining a reduced cat-
alytic efficiency of MNP2 as compared to MNP1.
The first tests (Table 1, entries 1 and 2) show that MNP1 is able to
catalyze the reaction in heterogenous aqueous conditions (dispersion of
an immiscible organic phase in an aqueous medium) with 20 mg cata-
lytic material for 1 mmol of each substrate, while MNP2 resulted in very
limited conversions. On the other hand, a classical water/tert-butanol
solvent mixture able to dissolve the reactants did not afford detectable
product. This highlights that our nanocatalyst are usable efficiently in
on-water conditions as was pointed out for other copper-NHC complexes
[27]. We tentatively ascribe this reduced reactivity of MNP2 to reduced
copper loading and to reduced reactivity of the copper(I) center
With MNP1 first substrate scope experiment shows that the azide-
alkyne cycloaddition reaction may be conducted with aromatic or
other activated alkynes in aqueous dispersion (entries 10–12) and that
the products may be obtained in high purity after magnetic decantation
4. Conclusion
In conclusion, although the catalytic potential of MNP1 for the
CuAAC reaction remains modest in comparison to other copper-loaded
MNPs, this communication provides a first proof of concept of an orig-
inal and simple strategy to heterogenize organometallic complexes.
These results could inspire further research on the optimization of the
auto-click reaction for nanoparticle functionalization. Moreover, ave-
nues might be opened with other metals, provided that transmetalation
from copper is feasible on heterogenized systems similar to MNP1.
Table 1
Outcome of catalytic reactions with MNP1 or MNP2.
Entry
Catalyst
Alkyne
Remarks/
Yield
deviation from
standard
conditionsa
1
2
3
4
5
MNP1b
MNP2c
MNP1
MNP2
MNP1
Phenylacetylene
Phenylacetylene
Phenylacetylene
Phenylacetylene
Phenylacetylene
93%
22%
81%
10%
–
0.5 mL H2O
0.5 mL H2O
H2O/tBuOH v/v
1:1
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
6
7
8
9
MNP1
MNP1
MNP1
MNP1
Phenylacetylene
Phenylacetylene
Phenylacetylene
Phenylacetylene
Catalyst
81%
37%
16%
75%
recycling #1
Catalyst
recycling #2
Catalyst
Acknowledgments
recycling #3
Catalyst
The participation to this study of Zohra Kouidri, Cassandra Perez and
Victoria Garcia (undergraduate trainees) is acknowledged. We thank
Mhammed Benbakkar (Laboratoire Magma et Volcans, UCA) for ICP
measurements and Christelle Blavignac (Centre Imagerie Cellulaire
recycling #3 +
copper
reloadingd
10
MNP1
2-
97%
Hydroxymethylphenylacetylene
4-Methoxyphenylacetylene
Methyl propiolate
Oct-1-yne
´
Sante, UCA) for transmission electronic microscopy.
11
12
13
14
15
MNP1
MNP1
MNP1
MNP1
MNP1
98%
quant.
trace
30%
Appendix A. Supplementary data
4-Ethynylaniline
N-Acetylpropargylamine
15%
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.ica.2021.120312.
a
Standard conditions: 20 mg nanoparticles, 1 mL H2O, 1 mmol of each sub-
strate. Azide: benzyl azide. Reaction duration: 20 h, RT.
b
Copper loading: 0.705 mol-%.
References
c
Copper loading: 0.088 mol-%.
d
see ESI.
[1] M.N. Hopkinson, C. Richter, M. Schedler, F. Glorius, Nature 510 (2014) 485–496.
3

´
K. Fauche and F. Cisnetti
Inorganica Chimica Acta 520 (2021) 120312
[2] S. Diez-Gonzalez, N-heterocyclic Carbenes: from laboratory curiosities to Efficient
Synthetic Tools, Royal Society of Chemistry, 2011.
(c) K. Bozorov, J.Y. Zhao, H.A. Aisa, Bioorg. Med. Chem. 27 (2019) 3511–3531;
(d) G. Yi, J. Son, J. Yoo, C. Park, H. Koo, Biomater. Res. 22 (2018) 13;
(e) L. Li, Z.Y. Zhang, Molecules 21 (2016) 1393. A. Mandoli, Molecules, 21
(2016), 1174;
[3] W.L. Wang, L.F. Cui, P. Sun, L.J. Shi, C.T. Yue, F.W. Li, Chem. Rev. 118 (2018)
9843–9929.
[4] R. Zhong, A.C. Lindhorst, F.J. Groche, F.E. Kuhn, Chem. Rev. 117 (2017)
1970–2058.
(f) V.K. Tiwari, B.B. Mishra, K.B. Mishra, N. Mishra, A.S. Singh, X. Chen, Chem.
Rev. 116 (2016) 3086–3240.
[5] V. Kandathil, B.D. Fahlman, B.S. Sasidhar, S.A. Patil, New J. Chem. 41 (2017)
9531–9545.
[16] F. Lazreg, F. Nahra, C.S.J. Cazin, Coord. Chem. Rev. 293 (2015) 48–79.
´
[17] F. Nahra, A. Gomez-Herrera, C.S.J. Cazin, Dalton Trans. 46 (2017) 628–631.
[6] A. Salamatmanesh, M.K. Miraki, E. Yazdani, A. Heydari, Catal. Lett. 148 (2018)
3257–3268.
[18] C. Gibard, D. Avignant, F. Cisnetti, A. Gautier, Organometallics 31 (2012)
7902–7908.
[7] A. Taber, J.B. Kirn, J.Y. Jung, W.S. Ahn, M.J. Jin, Synlett (2009) 2477–2482.
[8] A.R. Hajipour, N.S. Tadayoni, Z. Khorsandi, Appl. Organomet. Chem. 30 (2016)
590–595.
[19] H. Ibrahim, C. Gibard, C. Hesling, R. Guillot, L. Morel, A. Gautier, F. Cisnetti,
Dalton Trans. 43 (2014) 6981–6989.
[20] W. Wang, L. Zhao, H. Lv, G. Zhang, C. Xia, F.E. Hahn, F. Li, Angew. Chem., Int. Ed.
55 (2016) 7665–7670.
[9] K. Vishal, B.D. Fahlman, B.S. Sasidhar, S.A. Patil, Catal. Lett. 147 (2017) 900–918.
[10] R. Fareghi-Alamdari, M.S. Saeedi, F. Panahi, Appl. Organomet. Chem. 31 (2017).
[11] F. Rafiee, N. Mehdizadeh, Catal. Lett. 148 (2018) 1345–1354.
[12] H.J. Yoon, J.W. Choi, H. Kang, T. Kang, S.M. Lee, B.H. Jun, Y.S. Lee, Synlett (2010)
2518–2522.
[21] V.V. Rostovtsev, L.G. Green, V.V. Fokin, K.B. Sharpless, Angew. Chem., Int. Ed. 41
(2002) 2596–2599.
[22] S. Chassaing, V. Beneteau, P. Pale, Catal. Sci. Technol. 6 (2016) 923–957.
[23] S.A. McCarthy, G.-L. Davies, Y.K. Gun’ko, Nat. Protoc. 7 (2012) 1677–1693.
[24] C. Gibard, H. Ibrahim, A. Gautier, F. Cisnetti, Organometallics 32 (2013)
4279–4283.
[13] J.M. Collinson, J. Wilton-Ely, S. Diez-Gonzalez, Chem. Commun. 49 (2013)
11358–11360.
[14] I. Misztalewska-Turkowicz, K.H. Markiewicz, M. Michalak, A.Z. Wilczewska,
Journal of Catalysis 362 (2018) 46–54.
[25] D. Tapu, D.A. Dixon, C. Roe, Chem. Rev. 109 (2009) 3385–3407.
[26] E. Caytan, S. Roland, Organometallics 33 (2014) 2115–2118.
[27] S. Diez-Gonzalez, S.P. Nolan, Angew. Chem., Int. Ed. 47 (2008) 8881–8884.
[28] This issue may be solved by monitoring the copper leaching after each catalytic
run.
[15] (a) See recent reviews E. Haldon, M.C. Nicasio, P.J. Perez, Org. Biomol. Chem. 13
(2015) 9528–9550;
(b) S. Neumann, M. Biewend, S. Rana, W.H. Binder, Macromol. Rapid Commun.
41 (2020), https://doi.org/10.1002/marc.201900359;
4