Copper-Catalysed Allylic Substitution Using 2,8,14,20-Tetrapentylresorcinarenyl-Substituted Imidazolium Salts

Article abstract of DOI:10.1002/ejoc.201501070

Unsymmetrical imidazolium salts, each having one nitrogen atom (N1) substituted by a cavity-shaped TPR group (TPR = 2,8,14,20-tetrapentylresorcinaren-5-yl), were tested in situ as proligands for the copper-catalysed allylic arylation of cinnamyl bromide w

Full text of DOI:10.1002/ejoc.201501070

FULL PAPER
DOI: 10.1002/ejoc.201501070
Copper-Catalysed Allylic Substitution Using 2,8,14,20-
Tetrapentylresorcinarenyl-Substituted Imidazolium Salts
[a,b]
Neslihan Sahin,[a,c] David Sémeril,*[a] Eric Brenner,[a]
¸
˘
Murat Kaloglu,
[a]
[b]
[c]
[d]
˙
Dominique Matt,* Ismail Özdemir, Cemal Kaya, and Loïc Toupet
Keywords: Cavitands / Calixarenes / Carbenes / Copper / Allylic substitution / Regioselectivity
Unsymmetrical imidazolium salts, each having one nitrogen
atom (N1) substituted by a cavity-shaped TPR group (TPR =
2,8,14,20-tetrapentylresorcinaren-5-yl), were tested in situ as
proligands for the copper-catalysed allylic arylation of cinn-
amyl bromide with arylmagnesium halides. The catalytic
systems produced mixtures of linear (l) and branched (b)
arylated compounds in variable proportions, with the b/l ratio
being the highest (78:22) for the most crowded imidazolium
salt used, namely that in which the second nitrogen atom
(N2) was substituted by a mesityl group. An N-heterocyclic
carbene complex obtained from one of the imidazolium salts
was characterised by an X-ray diffraction study.
often proceeds with SN2Ј-regioselectivity,[6–11] this being
relevant to the synthesis of chiral compounds using an in-
expensive and readily available metal.
Introduction
Allylic substitution, the metal-catalysed reaction between
a nucleophile and a substrate containing a leaving group in
an allylic position (Scheme 1), is a well-established synthetic
procedure of modern organic chemistry.[1–3] Being appli-
cable to the formation of carbon–carbon, carbon–nitrogen,
and carbon–oxygen bonds, it has become very popular in
the field of natural product synthesis as well as in biomolec-
ular and medicinal chemistry. Regioselectivity is a key issue
in this reaction. The nucleophilic attack can lead to two
compounds, the SN2-product (or α-product) or the SN2Ј-
product (or γ-product), with regiocontrol of the reaction
depending on various factors, such as the leaving group, the
nature of the solvent, and the catalytic ligand-metal combi-
nation.[4] Whereas palladium remains the most frequently
employed metal for this reaction, copper-catalysed allylic
substitution,[5] which allows the use of Grignard reagents
as well as organozinc compounds, has recently emerged as
a method that usefully complements palladium catalysis.
Interestingly, allylic substitution with this particular metal
Scheme 1. Allylic substitution.
We have recently described a series of nickel and palla-
dium complexes containing N-heterocyclic carbene ligands
(NHCs) in which one nitrogen atom (termed N1 hereafter)
is substituted by a cavity-shaped 2,8,14,20-tetrapentyl
resorcin[4]aren-5-yl group (abbreviated TPR). Suzuki–
Miyaura (palladium)[12,13] and Kumada–Tamao–Corriu
(nickel)[14] cross-coupling studies with these complexes re-
vealed the beneficial role of the flexible pentyl fragments,
which, in combination with the rigidity of the cavitand core,
can sterically interact with the primary coordination sphere
of the metal and thus provide assistance in the reductive
elimination step. As such, the TPR group should be consid-
ered as a bulky group. Its effective steric bulk in the corre-
sponding carbene complexes is, however, difficult to quan-
tify accurately, particularly because a coordinated metal
centre may be positioned either above the cavity entrance
or pushed towards its exterior, this leading to different
steric effects. As an extension to our studies on complexes
with bulky, cavity-shaped ligands, we have now examined
the arylation properties of NHC-copper based systems de-
rived from the TPR-substituted imidazolium salts 1–4 (Fig-
[a] Université de Strasbourg, Laboratoire de Chimie Inorganique
et Catalyse, UMR 7177 CNRS,
4 rue Blaise Pascal, 67070 Strasbourg Cedex, France
E-mail: dsemeril@unistra.fr
dmatt@unistra.fr
http://lcimc.u-strasbg.fr
[b] Department of Chemistry, Faculty of Science and Art,
Inönü University,
Malatya 44280, Turkey
[c] Department of Chemistry, Faculty of Science,
Cumhuriyet University,
Sivas 58140, Turkey
[d] Institut de Physique, UMR 6251 CNRS, Université de Rennes 1,
Campus de Beaulieu, 35042 Rennes Cedex, France
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201501070.
7310
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 7310–7316

Copper-Catalysed Allylic Substitution
ure 1). These imidazolium salts differ only in the substituent
attached to the N2 nitrogen atom (benzyl, CH2(TPR),
CH2{(2,6-OMe)2C6H3}, mesityl). The azolium salts were all
assessed in situ in the allylic arylation of cinnamyl bromide
with arylmagnesium halides. We presumed that all the
NHCs generated from these salts would have comparable
electronic donor properties and thus only differ in the ex-
tent of steric encumbrance. The use of Cu-NHC catalysts
in allylic substitution is well documented.[4,6–11]
Scheme 2. Synthesis of imidazolium salts 3 and 4.
The ease of forming copper carbene complexes with the
above salts was demonstrated formally only for
1
Figure 1. The four TPR-substituted imidazolium salts 1–4 used in
this study, ranged in order of increasing bulk. TPR stands for
tetrapentylresorcinaren-5-yl. Being connected to a nitrogen atom
via the C-5 carbon atom, the TPR group does not behave as a
substituent with axial symmetry.
(Scheme 3). Thus, when this salt reacted with NaOtBu and
CuBr (in 1:1:1 ratio) in tetrahydrofuran (THF), copper(I)
complex 6 was formed.[15] In the 13C NMR spectrum, the
carbenic carbon atom appears at δ = 135.56 ppm. The
structure of 6 was confirmed by an X-ray diffraction study
(Figure 2). The core of the resorcin[4]arene shows the classi-
Results and Discussion
The catalytic systems tested in the present study (see be- cal bowl-shaped structure found in other resorcinarene-
derived cavitands[15–18] with separations between opposed
–
low) were generated in situ by mixing [Cu(OTf)2] (OTf
=
–
CF3SO3 , triflate) and the corresponding imidazolium salt pairs of C-2 atoms (the C-2 atoms are the OCCCO atoms
in Et2O at room temperature. Imidazolium salts 3 and 4 are of the wider rim) of 7.93 and 7.96 Å. Interestingly, in the
new compounds. The former was obtained quantitatively solid, the CuBr unit is pushed outside the cavity, although
by alkylating N-TPR-imidazole (5) with 2-bromomethyl- positioning of the Cu atom above the cavity could be envis-
1,3-dimethoxybenzene in refluxing chloroform (Scheme 2). aged from a steric point of view. In the observed configura-
Salt 4 was obtained in 48% yield by reacting 5 with tion, the percent buried volume (%Vbur) is 28.8, vs. 36.2
(Mes2I)OTf in N,N-dimethylformamide (DMF) at 100 °C estimated were the carbene to be positioned above the cav-
1
for 16 h. Both compounds were characterised by H and ity. The interplanar angle between the imidazoylidene unit
13C NMR spectroscopy, ESI-TOF MS, and elemental and the aromatic ring to which it is connected is 85°. The
analysis (see Experimental Section). Consistent with Cs molecule crystallises with two molecules of diethyl ether,
symmetry of the molecules, the corresponding 1H NMR one of which is poised at the upper entrance of the cavity.
spectra each display two AB patterns for the four OCH2O
groups and two triplets for the methine protons.
To assess the catalytic systems, cinnamyl bromide was
treated with various arylmagnesium halides (Scheme 4).
Eur. J. Org. Chem. 2015, 7310–7316
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7311

D. Sémeril, D. Matt et al.
FULL PAPER
(this period of time including the time of addition of the
Grignard reagent).
We began the study by determining whether the time
taken for the addition of phenylmagnesium bromide to a
solution containing a mixture of cinnamyl bromide, [Cu-
(OTf)2] and 1, influenced the catalytic outcome. We ob-
served that irrespective of the duration of addition (1 h,
10 min, or less than 1 min), the substitution reaction was
complete 1 h after having started the addition of the Grig-
nard reagent (Table 1). In keeping with previous reports, we
also found that the proportion of linear product increased
slightly when the Grignard reagent was added rapidly, with
its highest value reaching 78% (entries 4).[4] Notably, no
significant differences were seen between runs using pre-
formed complex 6 and tests carried out with the catalyst
obtained from 1 in situ (entries 1 and 2); this observation
justified the use of catalysts generated in situ.
Table 1. Copper-catalysed allylic arylation of cinnamyl bromide
with phenylmagnesium bromide: Influence of the time taken for
Grignard addition.[a]
Entry
Salt or Addition of
complex PhMgBr
Conversion
[%][b]
Linear
[%][b]
Branched
[%][b]
Scheme 3. Synthesis of copper complex 6.
1
2
3
4
1
6
1
1
over 1 h
over 1 h
over 10 min
fast (Ͻ 1 min)
100
100
100
100
68
70
72
78
32
30
28
22
[a] Reaction conditions: [Cu(OTf)2] (1 mol-%), imidazolium salt
(1 mol-%), PhCH=CHCH2Br (0.32 mmol), PhMgBr (0.39 mmol),
Et2O (3 mL), –78 °C, 1 h. [b] The conversion and the linear/
branched product ratios were determined by 1H NMR spectro-
scopic analysis (accuracy of Ϯ5%).
Given that the rate of addition had practically no influ-
ence on the conversion, all further tests were carried out
under fast addition of Grignard reagent, and stopped after
1 h. As described above, each run was performed at –78 °C
(Table 2). The proportion of branched product increased in
the order 1Ͻ 2 Ͻ 3 Ͻ 4, with the b/l ratios varying from
22:78 for 1 to 78:22 for 4 (entries 1–4). Thus, the observed
b/l ratios seemingly correlate with the steric bulk of the li-
gand.[19] Comparison of 2 with two related salts, namely 7
and 8, in which the TPR group was replaced by a smaller
phenyl or calix[4]arenyl substituent, confirmed this trend
(Figure 3). Here, the b/l ratios were 20:80 (7) and 16:84 (8)
Figure 2. Molecular structure of 6. A molecule of Et2O occupies
the cavity entrance, another one (not shown for clarity) is remote
from the cavity. The interplanar angle between the imidazolylidene
unit and the aromatic ring to which it is connected is 85°.
Table 2. Copper-catalysed allylic arylation of cinnamyl bromide
and phenylmagnesium bromide. Influence of imidazolium salts 1–
4, 7 and 8.[a]
Entry
NHC·HCl Conversion [%][b] Branched [%][b] Linear [%][b]
1
2
3
4
5
6
1
2
3
4
7
8
100
100
100
100
100
100
22
31
47
78
20
16
78
69
53
22
80
84
Scheme 4. Copper-catalysed allylic arylation of cinnamyl bromide
with an arylmagnesium halide.
[a] Reaction conditions: [Cu(OTf)2] (1 mol-%), imidazolium salt
(1 mol-%), PhCH=CHCH2Br (0.32 mmol), PhMgBr (0.39 mmol),
Et2O (3 mL), –78 °C, 1 h. [b] The conversion and the linear/
The reactions were performed at –78 °C with a cinnamyl
1
bromide/Cu ratio of 100. The conversions were determined
branched proportions were determined by H NMR spectroscopic
1
by H NMR spectroscopic analysis after 1 h reaction time analysis (accuracy of Ϯ5%).
7312
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 7310–7316

Copper-Catalysed Allylic Substitution
(vs. 31:69 for 2) (entries 2, 5 and 6). Note also that the and the subsequent reductive elimination.[4,6,20–28] In these
proportion of branched product was significantly higher for two possible pathways, a bulky ligand should accelerate the
2 than for 8, which is consistent with a TPR group having reductive elimination step leading to the γ-product. This
a greater steric bulk than a calixarenyl group.
was indeed observed with the ligands 1–4 (in this order) in
the reaction between cinnamyl bromide and PhMgBr. In
the same vein, the reductive elimination step leading to the
γ-product should also be favoured with bulky Grignard rea-
gents. This was notably the case with o-Me-C6H4MgCl.
Thus, the present results conform to the generally accepted
mechanism. The fact that the γ-selectivity of the most bulky
ligand used in this study, namely the carbene derived from
4, does not reach that of Tomioka’s crowded complex 9[9,10]
(Figure 4), reflects the seemingly moderate steric encum-
brance of 4 (electronic factors are not relevant here because
the electronic influence of a mesityl group compares with
that of an alkyl group[29]). We note, however, that the rate
of arylation with the [Cu(OTf)2]/4 system (100% conversion
in 1 h using 1 mol-% of copper catalyst) is comparable to
that of 9. Overall, the unexpectedly small influence of the
bulky TPR group on the γ selectivity of the reaction con-
trasts with the previously reported major activity increase
induced by this group in Suzuki–Miyaura cross coupling.
Possibly, the steric influence on the selectivity of the TPR
group in copper-catalysed allylic substitution could be en-
Figure 3. Imidazolium salts 7 and 8.
To establish whether the selectivity was dependent upon
the nature of substituents on the Grignard reagent, p-
chlorophenylmagnesium bromide, p-fluorophenylmagne-
sium bromide and o-methylphenylmagnesium chloride were
tested with the catalytic system [Cu(OTf)2]/4 (1 mol-% Cu)
(Table 3). The reaction rates observed for the two p-substi-
tuted Grignard reagents were relatively high, and were com-
parable to those obtained for phenylmagnesium bromide
(entries 1–3). Both substrates gave the branched compound
as the major product, but the proportion of the latter was
significantly lower than that obtained with PhMgBr. As an-
ticipated because of its greater encumbrance, the use of o-
methylphenylmagnesium chloride resulted in a very low
conversion (2% conversion after 1 h reaction time) al-
though the branched product formed selectively (entry 4).
Table 3. Copper-catalysed allylic alkylation of cinnamyl bromide
with arylmagnesium halide.[a]
Entry ArMgX
NHC·HX Conversion Branched
Linear
[%][b]
[%][b]
[%][b]
1
2
3
4
PhMgBr
4
4
4
4
100
94
100
2
78
52
62
22
48
38
0
p-Cl-(C6H4)MgBr
p-F-(C6H4)MgBr
o-Me-(C6H4)MgCl
100
[a] Reaction conditions: [Cu(OTf)2] (1 mol-%), salt 4 (1 mol-%),
PhCH=CHCH2Br (0.32 mmol), ArMgX (0.39 mmol), Et2O
(3 mL), –78 °C, 1 h. [b] The conversion and the linear/branched
proportions were determined by 1H NMR spectroscopic analysis
(accuracy of Ϯ5%).
Scheme 5. Possible pathway for the copper-catalysed allylic substi-
tution of cinnamyl bromide with arylmagnesium halides (ArMgX).
In the generally accepted mechanism for copper-cata-
lysed allylic substitutions (Scheme 5), the initially formed
arylcuprate A preferentially reacts with the allylic bromide
to form the σ-allyl copper(III) complex B. Reductive elimi-
nation of B leads to the branched product, whereas its iso-
merisation into the π-allyl complex C gives the sterically less
hindered σ-allyl complex D, precursor of the linear product.
Thus, the regioselectivity depends on the relative rates of
Figure 4. Tomioka’s catalyst used in the arylation of cinnamyl
the reductive elimination of B, the isomerisation of B to D, bromide.
Eur. J. Org. Chem. 2015, 7310–7316
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7313

D. Sémeril, D. Matt et al.
FULL PAPER
2-N-(2,6-Dimethoxybenzyl)-5-N-[4(24),6(10),12(16),18(22)-tetra-
hanced by rigidly fixing the TPR unit (after chemical modi-
fication of the ligand) above the cavity entrance, thereby
increasing its time-averaged steric bulk.
methylenedioxy-2,8,14,20-tetrapentylresorcin[4]arene-5-yl]imid-
1
azolinium Bromide (3): Yield 0.249 g (90%). H NMR (400 MHz,
3
CDCl3): δ = 9.85 (s, 1 H, NCHN), 7.34 [t, J = 8.5 Hz, 1 H, ArH,
(CH3O)2C6H3], 7.33 (s, 1 H, ArH, resorcinarene), 7.29 (br. s, 1 H,
NCHCHN), 7.21 (br. s, 1 H, NCHCHN), 7.12 (s, 3 H, ArH, resor-
cinarene), 6.64 (s, 2 H, ArH, resorcinarene), 6.60 [d, J = 8.5 Hz,
Conclusions
3
We have shown that the mono-TPR-substituted imid-
azolium salts used in this study efficiently catalyse the cop-
per-catalysed arylation of cinnamyl bromide with phenyl-
magnesium bromide. The intrinsic ability of the bulky TPR
group to induce γ-selectivity was found to be moderate, but
formation of the branched product could be significantly
increased by attaching appropriate substituents to the sec-
ond nitrogen atom, with the highest b/l ratio being obtained
for the imidazolium salts combining TPR and mesityl sub-
stituents. A rational way to improve the γ-selectivity with
TPR substituted NHCs could involve restricting the rota-
tional freedom of the carbene ring about the N–C(resorcin-
arenyl) moiety so as to force the metal centre to remain
located permanently above the cavity entrance, and conse-
quently render the steric effects of the TPR group predomi-
nant over those of the N2-substituent. This could then also
promote possible cavity effects associated with the receptor
properties of the resorcinarene unit. Modification of the
electronic properties of such ligands could also be envis-
aged.
2 H, ArH, (CH3O)2C6H3], 6.53 (s, 1 H, ArH, resorcinarene), 5.77
2
(s, 2 H, NCH2Ar), 5.72 and 4.65 (AB spin system, J = 7.2 Hz, 4
H, OCH2O), 5.60 and 4.77 (AB spin system, 2J = 7.5 Hz, 4 H,
3
3
OCH2O), 4.75 (t, J = 7.8 Hz, 2 H, CHCH2), 4.70 (t, J = 8.1 Hz,
2 H, CHCH2), 3.89 (s, 6 H, OCH3), 2.30–2.16 (m, 8 H, CHCH2),
3
1.49–1.29 (m, 24 H, CH2CH2CH2CH3), 0.91 (t, J = 6.9 Hz, 12 H,
CH2CH3) ppm. 13C NMR (100 MHz, CDCl3): δ = 158.92–104.23
(ArC), 137.56 (s, NCN), 122.90 (s, NCHCHN), 117.79 (s,
NCHCHN), 101.17 (s, OCH2O), 99.57 (s, OCH2O), 56.30 (s,
OCH3), 43.12 (s, NCH2Ar), 36.80 (s, CHCH2), 36.49 (s, CHCH2),
32.17 (s, CH2CH2CH3), 32.05 (s, CH2CH2CH3), 30.12 (s, CHCH2),
29.90 (s, CHCH2), 27.70 (s, CHCH2CH2), 27.65 (s, CHCH2CH2),
22.82 (s, CH2CH3), 22.79 (s, CH2CH3), 14.24 (s, CH2CH3), 14.22 (s,
CH2CH3) ppm. MS (ESI-TOF): m/z = 1033.54 [M – Br]+ (expected
isotopic profile). C64H77BrN2O10 (1114.21): calcd. C 68.99, H 6.97,
N 2.51; found C 69.12, H 7.17, N 2.39.
2-N-[4(24),6(10),12(16),18(22)-Tetramethylenedioxy-2,8,14,20-tetra-
pentylresorcin[4]arene-5-methyl]-5-N-phenyl-imidazolinium Bromide
1
(7): Yield 0.249 g (95%). H NMR (400 MHz, CDCl3): δ = 10.86
(s, 1 H, NCHN), 7.69 (d, 3J = 7.2 Hz, 2 H, ArH, Ph), 7.60–7.51
(m, 3 H, ArH, Ph), 7.47 (br. s, 1 H, NCHCHN), 7.36 (br. s, 1 H,
NCHCHN), 7.21 (s, 1 H, ArH, resorcinarene), 7.09 (s, 1 H, ArH,
resorcinarene), 7.08 (s, 2 H, ArH, resorcinarene), 6.55 (s, 2 H, ArH,
resorcinarene), 6.48 (s, 1 H, ArH, resorcinarene), 6.17 and 4.55 (AB
Experimental Section
2
spin system, J = 7.4 Hz, 4 H, OCH2O), 5.80 (s, 2 H, NCH2), 5.64
2
3
and 4.45 (AB spin system, J = 7.2 Hz, 4 H, OCH2O), 4.72 (t, J
General Experimental Methods: All manipulations were performed
in Schlenk-type flasks under dry nitrogen. Solvents were dried by
conventional methods and distilled immediately prior to use.
CDCl3 was passed down a 5 cm thick alumina column and stored
under nitrogen over molecular sieves (4 Å). Routine 1H and
13C{1H} NMR spectra were recorded with Bruker FT instruments
3
= 8.2 Hz, 2 H, CHCH2), 4.70 (t, J = 8.2 Hz, 2 H, CHCH2), 2.33–
2.08 (m, 8 H, CHCH2), 1.43–1.29 (m, 24 H, CH2CH2CH2CH3),
0.90 (t, 3J = 7.2 Hz, 6 H, CH2CH3), 0.90 (t, 3J = 7.2 Hz, 6 H,
CH2CH3) ppm. 13C NMR (100 MHz, CDCl3): δ = 155.37–116.81
(ArC), 134.56 (s, NCN), 122.05 (s, NCHCHN), 119.88 (s,
NCHCHN), 100.53 (s, OCH2O), 99.68 (s, OCH2O), 44.23 (s,
NCH2 ), 36.76 (s, CHCH2 ), 36.45 (s, CHCH2 ), 32.14 (s,
CH2CH2CH3), 32.11 (s, CH2CH2CH3), 30.02 (s, CHCH2), 29.92
(s, CHCH2), 27.67 (s, CHCH2CH2), 22.80 (s, CH2CH3), 22.78 (s,
CH2CH3), 14.21 (s, CH2CH3) ppm. MS (ESI-TOF): m/z = 973.53
[M – Br]+ (expected isotopic profiles). C62H73BrN2O8 (1054.16):
calcd. C 70.64, H 6.98, N 2.55; found C 70.49, H 7.02, N 2.64.
1
(AVANCE 400). H NMR spectra were referenced to residual pro-
tiated solvents (δ = 7.26 ppm for CDCl3) and 13C NMR chemical
shifts are reported relative to deuteriated solvents (δ = 77.16 ppm
for CDCl3). Chemical shifts and coupling constants are reported
in ppm and in Hz, respectively. Elemental analyses were performed
by the Service de Microanalyse, Institut de Chimie, Université de
Strasbourg. 2-N-Benzyl-5-N-[4(24),6(10),12(16),18(22)-tetrameth-
ylenedioxy-2,8,14,20-tetrapentylresorcin[4]arene-5-yl]imidazolin-
ium bromide (1),[13] 2-N-[4(24),6(10),12(16),18(22)-tetramethyl-
enedioxy-2,8,14,20-tetrapentylresorcin[4]arene-5-methyl]-5-N-
[4(24),6(10),12(16),18(22)-tetramethylenedioxy-2,8,14,20-tetrapent-
ylresorcin[4]arene-5-yl]imidazolinium bromide (2),[12] 5-N-imid-
azolyl-4(24),6(10),12(16),18(22)-tetramethylenedioxy-2,8,14,20-
tetrapentylresorcin[4]arene (5),[13] 2-N-[4(24),6(10),12(16),18(22)-
tetramethylenedioxy-2,8,14,20-tetrapentylresorcin[4]arene-5-
methyl]-5-N-[25,26,27,28-tetrabenzyloxycalix[4]arene-5-yl]imid-
azolinium bromide (8),[12] and 2,6-dimethoxybenzyl bromide[30]
were prepared according to reported procedures.
2-N-[2,4,6-Trimethylphenyl]-5-N-[4(24),6(10),12(16),18(22)-tetra-
methylenedioxy-2,8,14,20-tetrapentylresorcin[4]arene-5-yl]imid-
azolinium Triflate (4): A mixture of N-resorcinarenyl-imidazole 5
(0.500 g, 0.57 mmol), dimesitylidonium salt (0.437 g, 0.85 mmol)
and [Cu(OTf)2] (0.010 g, 0.03 mmol, 5 mol-%) in DMF (3 mL) was
stirred at 100 °C for 16 h. The solvent was then removed under
reduced pressure and the solid residue was purified by flash
chromatography (acetone/MeOEt, 10:90 v/v), yield 0.309 g (48%).
1H NMR (500 MHz, CDCl3): δ = 9.1 (s, 1 H, NCHN), 7.55 (br. s,
1 H, NCHCHN), 7.41 (s, 1 H, ArH, resorcinarene), 7.36 (s br, 1
H, NCHCHN), 7.15 (s, 2 H, ArH, resorcinarene), 7.13 (s, 1 H,
ArH, resorcinarene), 7.03 (s, 2 H, ArH, mesityl), 6.53 (s, 2 H, ArH,
resorcinarene), 6.46 (s, 1 H, ArH, resorcinarene), 5.66 and 4.47 (AB
spin system, 2J = 7.5 Hz, 4 H, OCH2O), 5.53 and 4.68 (AB spin
system, 2J = 7.5 Hz, 4 H, OCH2O), 4.73 (t, 3J = 8.0 Hz, 2 H,
General Procedure for the Preparation of the Imidazolium Salts 3
and 7: N-Arylimidazole (0.25 mmol) and alkyl bromide
(0.25 mmol) were dissolved in CHCl3 (10 mL). The reaction mix-
ture was then heated to reflux for 2 d. After cooling to room tem-
perature, the solvent was removed under vacuum. The solid was
washed with pentane and recrystallised from CH2Cl2/isopropyl
ether to afford the corresponding imidazolium salt.
3
CHCH2), 4.73 (t, J = 8.0 Hz, 2 H, CHCH2), 2.34 (s, 3 H, p-CH3-
mesityl), 2.32–2.07 (m, 8 H, CHCH2), 2.07 (s, 6 H, o-CH3-mesityl),
3
1.46–1.32 (m, 24 H, CH2CH2CH2CH3), 0.92 (t, J = 7.5 Hz, 6 H,
7314
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 7310–7316

Copper-Catalysed Allylic Substitution
CH2CH3), 0.91 (t, 3J = 7.5 Hz, 6 H, CH2CH3) ppm. 13C NMR around 200 K and led to decomposition of the sample. The check-
(100 MHz, CDCl3): δ = 155.56–117.19 (ArC), 100.55 (s, OCH2O),
99.50 (s, OCH2O), 36.81 (s, CHCH2), 36.46 (s, CHCH2), 32.16 (s,
cif contains two level A alerts, both due to the disordered diethyl
ether lying outside the cavity. Given that it was not possible to
CH2CH2CH3), 32.04 (s, CH2CH2CH3), 30.09 (s, CHCH2), 29.91 (s, assign the corresponding residuals peaks, the SQUEEZE procedure
CHCH2), 27.69 (s, CHCH2CH2), 27.66 (s, CHCH2CH2), 22.79 (s, was applied.
CH2CH3), 22.78 (s, CH2CH3), 21.24 (s, p-CH3-mesityl), 17.18 (s, o-
Crystallographic data CCDC-844297 for compound 6 contains the
supplementary crystallographic data for this paper. These data can
be obtained free of charge from The Cambridge Crystallographic
CH3-mesityl), 14.20 (s, CH2CH3) ppm. MS (ESI-TOF): m/z =
1001.58 [M – TfO]+ (expected isotopic profiles). C65H77F3N2O11S
(1151.37): calcd. C 67.81, H 6.74, N 2.43; found C 67.77, H 6.82,
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
N 2.79.
Typical Procedure for the Copper-Catalysed Allylic Arylations at
Bromo-[2-{4(24),6(10),12(16),18(22)-tetramethylenedioxy-2,8,14,20-
–78 °C: A 15 mL Schlenk tube was filled with [Cu(OTf)2] (1.2 mg,
tetrapentylresorcin[4]aren-5-yl}-5-benzylimidazol-2-ylidene]copper(I)
1 mol-%), ligand (1 mol-%) and Et2O (2 mL). The reaction mixture
(6): A mixture of [CuBr] (0.041 g, 0.28 mmol), imidazolium salt
was stirred for 10 min at 0 °C before addition of the Grignard rea-
1 (0.300 g, 0.28 mmol) and NaOtBu (0.027 g, 0.28 mmol) in THF
gent (0.39 mmol). After cooling the reaction mixture to –78 °C, a
(5.5 mL) was stirred at room temp. for 4 h. The reaction mixture
solution of cinnamyl bromide (64 mg, 0.32 mmol) in Et2O (1 mL)
was then filtered through Celite. The filtrate was evaporated under
was added quickly. After 1 h (the temperature being maintained at
vacuum, and the solid residue purified by flash chromatography
–78 °C), 1 m aq HCl (1 mL) was added. After extraction of the
(MeOH/CH2Cl2, 5:95 v/v) to afford the corresponding white cop-
aqueous layer with Et2O (2 ϫ 2 mL), the organic phases were com-
bined. The resulting solution was washed with NaHCO3 (5 mL),
brine (5 mL), dried with MgSO4, filtered and evaporated under re-
1
per complex 6, yield 0.120 g (38%). H NMR (400 MHz, CDCl3):
δ = 7.39–7.23 (m, 3 H, ArH, Ph), 7.33 (br. s, 1 H, NCHNCH), 7.18
(d, J = 6.5 Hz, 2 H, ArH, Ph), 7.15 (s, 2 H, ArH, resorcinarene),
3
duced pressure. The product distribution was analysed by 1H NMR
7.14 (br. s, 1 H, NCHNCH), 6.98 (s, 1 H, ArH, resorcinarene), 6.94
spectroscopy.
(s, 1 H, ArH, resorcinarene), 6.53 (s, 3 H, ArH, resorcinarene), 5.71
and 4.55 (AB spin system, 2J = 7.0 Hz, 4 H, OCH2O), 5.36 and
4.78 (AB spin system, 2J = 7.0 Hz, 4 H, OCH2O), 5.30 (s, 2 H,
Acknowledgments
3
NCH2Ph), 4.73 (t, J = 8.0 Hz, 4 H, CHCH2), 2.31–2.21 (m, 8 H,
CHCH2), 1.46–1.33 (m, 24 H, CH2CH2CH2CH3), 0.93 (t, 3J =
The authors thank the Scientific and Technological Research of
Turkey (TÜBITAK-BIDEB), the International Research Fellow-
ship Programme for grants to M. K. and N. S., and the French
Agence Nationale de la Recherche (ANR) (ANR-12-BS07-0001-
3
7.0 Hz, 6 H, CH2CH3), 0.92 (t, J = 7.0 Hz, 6 H, CH2CH3) ppm.
13C NMR (100 MHz, CDCl3): δ = 155.76–117.00 (ArC), 135.56 (s,
NCN), 126.52 (s, NCHCHN), 124.50 (s, NCHCHN), 100.15 (s,
OCH2O), 100.09 (s, OCH2O), 55.31 (NCH2Ph), 36.80 (s, CHCH2),
36.52 (s, CHCH2), 32.23 (s, CH2CH2CH3), 32.15 (s, CH2CH2CH3),
30.20 (s, CHCH2), 29.89 (s, CHCH2), 27.76 (s, CHCH2CH2), 22.86
01-RESICAT) for financial support.
[1] Copper-Mediated Cross-Coupling Reactions (Eds.: G. Evano, N.
(s, CH2CH3), 14.27 (s, CH2CH3) ppm. MS (ESI-TOF): m/z =
Blanchard), John Wiley & Sons, Hoboken, New Jersey, USA,
1076.48 [M – Br + CH3CN]+ (expected isotopic profile).
2014.
C62H72BrCuN2O8·iPr2O (1116.69 + 60.09): calcd. C 66.34, H 6.85,
[2] S. Díez-González, S. P. Nolan, Synlett 2007, 2158–2167.
N 2.38; found C 66.42, H 6.91, N 2.18. The sample used for ele-
mental analysis was obtained by recrystallisation of 6 from a
CH2Cl2/iPr2O solution.
[3] J.-B. Langlois, A. Alexakis, Top. Organomet. Chem. 2012, 38,
235–268.
[4] J.-E. Bäckvall, M. Sellén, B. Grant, J. Am. Chem. Soc. 1990,
112, 6615–6621.
X-ray Analysis of 6: Single crystals of 6·1.5Et2O suitable for an X-
ray diffraction study were obtained by slow diffusion of diethyl
ether into an acetonitrile solution of the complex. Mr = 1222.81;
[5] G. Fouquet, M. Schlosser, Angew. Chem. Int. Ed. Engl. 1974,
13.
[6] S. Ando, H. Matsunaga, T. Ishizuka, Tetrahedron 2013, 69,
1687–1693.
¯
triclinic; space group P1; a = 14.2710(4), b = 17.1546(5), c =
17.4082(4) Å, α = 115.880(3), β = 94.672(2), γ = 109.233(3)°; V =
3490.76(22) Å3; Z = 2; Dx = 1.163 mgm–3; λ(Mo-Kα) = 0.71073 Å;
μ = 0.937 mm–1; F(000) = 1288; T = 220(2) K. The sample
(0.45 ϫ0.41 ϫ0.27 mm) was studied with an Oxford Diffraction
Xcalibur Sapphire 3 diffractometer (graphite-monochromated Mo-
Kα radiation, λ = 0.71073 Å). The data collection (2θmax = 27°,
omega scan frames via 0.7° omega rotation and 30 s per frame,
range HKL: H –18, 18 K –21, 21 L –22,22) gave 15029 reflections.
The data led to 8705 independent reflections with IϾ2.0σ(I). The
structure was solved with SIR-97,[31] which revealed the non-
hydrogen atoms of the molecule. After anisotropic refinement, all
the hydrogen atoms were found from the Fourier difference map.
The whole structure was refined with SHELX97[32] by full-matrix
least-squares techniques (use of |F2|; x, y, z, βij for C, Br, Cu, N
and O atoms, x, y, z in riding mode for H atoms); 736 variables
and 8705 observations with I Ͼ 2.0σ(I); calcd. w = 1/[σ2(Fo2) +
[7] D. Grassi, H. Li, A. Alexakis, Chem. Commun. 2012, 48,
11404–11406.
[8] M. Magrez, Y. Le Guen, O. Baslé, C. Crévisy, M. Mauduit,
Chem. Eur. J. 2013, 19, 1199–1203.
[9] K. B. Selim, Y. Matsumoto, K. Yamada, K. Tomioka, Angew.
Chem. Int. Ed. 2009, 48, 8733–8735; Angew. Chem. 2009, 121,
8889.
[10] K. B. Selim, H. Nakanishi, Y. Matsumoto, Y. Yamamoto, K.
Yamada, K. Tomioka, J. Org. Chem. 2011, 76, 1398–1408.
[11] S. Tominaga, Y. Oi, T. Kato, D. K. An, S. Okamoto, Tetrahe-
dron Lett. 2004, 45, 5585–5588.
[12] N. S¸ahin, D. Sémeril, E. Brenner, D. Matt, I. Özdemir, C.
Kaya, L. Toupet, Turk. J. Chem., DOI: 10.3906/kim-1503–82.
[13] N. S¸ahin, D. Sémeril, E. Brenner, D. Matt, I. Özdemir, C.
Kaya, L. Toupet, ChemCatChem 2013, 5, 1115–1125.
[14] N. S¸ahin, D. Sémeril, E. Brenner, D. Matt, I. Özdemir, C.
Kaya, L. Toupet, Eur. J. Org. Chem. 2013, 4443–4449.
[15] The relatively low isolated yield arises from the fact that sepa-
ration of this complex from unreacted imidazolium salt was
difficult.
(0.1128P)2] where P = (Fo2 + 2Fc )/3 with the resulting R = 0.0539,
2
RW = 0.1814 and SW = 1.023, Δρ Ͻ 1.19 eÅ–3. The major issues
in the structure determination arose from the tendency of the crys-
tal to de-solvate rapidly and a phase transition, which occurs
[16] H. El Moll, D. Sémeril, D. Matt, L. Toupet, Eur. J. Org. Chem.
2010, 1158–1168.
Eur. J. Org. Chem. 2015, 7310–7316
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7315

D. Sémeril, D. Matt et al.
FULL PAPER
[17] H. El Moll, D. Sémeril, D. Matt, L. Toupet, J.-J. Harrowfield,
Org. Biomol. Chem. 2012, 10, 372–382.
[18] H. El Moll, D. Sémeril, D. Matt, M.-T. Youinou, L. Toupet,
Org. Biomol. Chem. 2009, 7, 495–501.
[25] J. Levisalles, M. Rudler-Chauvin, H. Rudler, J. Organomet.
Chem. 1977, 136, 103–110.
[26] R. M. Magid, Tetrahedron 1980, 36, 1901–1930.
[27] E. S. M. Persson, M. van Klaveren, D. M. Grove, J.-E.
Bäckvall, G. van Koten, Chem. Eur. J. 1995, 1, 351–359.
[28] T. L. Underiner, H. L. Goering, J. Org. Chem. 1991, 56, 2563–
2672.
[29] R. A. Kelly III, H. Clavier, S. Giudice, N. M. Scott, E. D. Ste-
vens, J. Bordner, I. Samardjiev, C. D. Hoff, L. Cavallo, S. P.
Nolan, Organometallics 2008, 27, 202–210.
[30] L. Feng, K. Lv, M. Liu, S. Wang, J. Zhao, X. You, S. Li, J.
Cao, H. Guo, Eur. J. Med. Chem. 2012, 55, 125–136.
[31] A. Altomare, M. C. Burla, M. Camalli, G. Cascarano, C. Gia-
covazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R.
Spagna, J. Appl. Crystallogr. 1998, 31, 74–77.
[19] The degree of steric encumbrance of the carbene ligands de-
rived from 1–4 is difficult to quantify accurately because it de-
pends on the position of the metal with respect to the cavity.
The proposed order of increasing bulk (1 Ͻ 2 Ͻ 3 Ͻ 4) pre-
sumes that the metal is pushed away from the cavity, as shown
in Figure 2, and that the N2 substituent is oriented so as to
come as close as possible to the coordinated metal centre.
[20] J.-E. Bäckvall, E. S. M. Persson, A. Bombrun, J. Org. Chem.
1994, 59, 4126–4130.
[21] E. J. Corey, N. W. Boaz, Tetrahedron Lett. 1984, 25, 3063–3066.
[22] H. L. Goering, S. S. Kantner, J. Org. Chem. 1984, 49, 422–426.
[23] H. L. Goering, V. D. J. Singleton, J. Am. Chem. Soc. 1976, 136,
7854–7855.
[32] G. M. Sheldrick, SHELXL-97, Program for the Refinement of
Crystal Structures, University of Göttingen, Germany, 1997.
Received: August 19, 2015
[24] H. L. Goering, V. D. J. Singleton, J. Org. Chem. 1983, 48, 1531–
1533.
Published Online: October 22, 2015
7316
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 7310–7316