C1-Symmetric Atropisomeric NHC Palladium(II) Complexes: Synthesis, Resolution and Characterization

Article abstract of DOI:10.1002/adsc.202100491

Series of chiral palladium(II) allyl and cinnamyl complexes bearing a C₁-symmetric N-heterocyclic carbenes were synthesized from achiral precursors. The chirality of theses complexes results from the formation of the carbene-palladium bond whic

Full text of DOI:10.1002/adsc.202100491

RESEARCH ARTICLE
doi.org/10.1002/adsc.202100491
C1-Symmetric Atropisomeric NHC Palladium(II) Complexes:
Synthesis, Resolution and Characterization
Lingyu Kong,a Yajie Chou,a Marion Jean,a Muriel Albalat,a Nicolas Vanthuyne,a
Paola Nava,a Stéphane Humbel,a and Hervé Claviera,*
a
Aix Marseille Univ, CNRS, Centrale Marseille, iSm2, Marseille, France
Phone: +33 (0)413945630
E-mail: herve.clavier@univ-amu.fr
Manuscript received: April 22, 2021; Revised manuscript received: June 22, 2021;
Version of record online: July 14, 2021
Supporting information for this article is available on the WWW under https://doi.org/10.1002/adsc.202100491
Abstract: Series of chiral palladium(II) allyl and cinnamyl complexes bearing a C1-symmetric N-heterocyclic
carbenes were synthesized from achiral precursors. The chirality of theses complexes results from the
formation of the carbene-palladium bond which restricts the rotation of dissymmetric N-aryl substituents of the
NHC and thus creates an axis of chirality. Chiral HPLC at preparative scale enabled the resolution of racemic
complexes and provided a straightforward access to complexes with excellent enantiopurities (>99.5% ee).
Enantiopure complexes were studied by crystal X-ray diffraction and electronic circular dichroism (ECD).
Their configuration stabilities were investigated both experimentally and theoretically through the
determination of the rotational barrier values. These complexes were tested for the intramolecular α-arylation
of amides, with a moderate chiral induction (up to 54% ee).
Keywords: axial chirality; NHC ligand; Palladium; Chiral HPLC; ECD
Introduction
Grubbs disclosed that the use of dissymmetric aryl
groups as N-substituents allow for improving the
Nowadays, N-heterocyclic carbenes (NHC) are well- enantioinduction through a relay effect (structure D).[7]
established ancillary ligands in organometallic Despite successful and widespread applications,[8]
chemistry.[1] Their unique stereoelectronic properties conformational insights have been thoroughly studied
allow to obtain highly reactive catalytic species and only in rare occasions.[9] In absence of determinations
contributed to the development of a wide array of of rotational barriers of aryl groups along the CÀ N
metal-catalyzed transformations.[2] Moreover, well-de- bonds, configurational stabilities are difficult to assess
fined complexes bearing an NHC ligand are often (rotamers vs. atropisomers). Axial chirality has also
particularly stable toward air and moisture. Many been implemented for the design of chiral NHC ligands
metal-NHC complexes can even be purified by silica (Figure 1c). Among various examples,[10] structures
gel chromatography. Therefore, the development of E[11] and F[12] are of interest, albeit often prepared
chiral versions has attracted much attention.[3] Several through tedious and time-consuming syntheses. Re-
designs of chiral monodentate NHC ligands have been cently, we disclosed a new concept of chiral metal-
investigated, such as the use of chiral N-substituents NHC complexes by virtue of a rotationally restricted
(A, Figure 1a). In spite of remarkable achievements,[4] CÀ N axes leading to the formation of atropisomers
chiral catalysts A generally suffer from a high (Figure 1d).[13] Of note, since the systematic name for
flexibility of chiral N-substituents that can be circum- this category of NHC ligands is quite cumbersome a
vented with fused skeletons, structure B.[5] The main trivial name “IKong” has been assigned.
other strategy consists in the introduction of chiral
Advantageously, our strategy does not require the
centers on the NHC backbone (Figure 1b). Even if use of chiral precursors for which supplies, prices and
chiral centers are positioned far away from the metal purities might fluctuate. As the metal-NHC bond
center, in some case, excellent chiral induction can be formation generates atropisomers by means of the
reached with catalysts of structure C.[6] In 2001, CÀ N bond rotation restriction, chiral HPLC separation
Adv. Synth. Catal. 2021, 363, 4229–4238
4229
© 2021 Wiley-VCH GmbH

RESEARCH ARTICLE
asc.wiley-vch.de
Results and Discussion
Imidazolium Salts Synthesis
For the synthesis of imidazolium salts 4 containing
methyl groups on the backbone, we first experienced
the synthetic route disclosed by Fürstner in 2006.[14] As
depicted in Scheme 1, formamide 3 was prepared in an
excellent yield from 3-hydroxybutan-2-one 1 and 2-
isopropylaniline giving rise to α-aminoketone 2 and
subsequent formylation using acetic formic anhydride.
Of note, the acid catalyzed substitution could not be
achieved with the bulky 2,6-diisopropylaniline. With
cyclohexylamine, the one-pot reaction including three
steps allowed to isolate imidazolium salt 4b with a
satisfying yield of 49%. With the less nucleophilic 2,6-
diisopropylaniline, the yield of 4e dropped down to
15% and the purification step was found demanding.
At room temperature, the 1H NMR spectra of
imidazolium salt 4b displayed clearly signals of
inequivalent chemical shifts due to the presence of the
diastereotopic isopropyl group which suggests a slow
interconversion of enantiomeric conformers on the
NMR timescale. In order to quantify the rotational
1
barrier around the CÀ N bond, variable temperature H
NMR experiments were carried out and showed the
effect of the temperature on lineshape up to signals
coalescence (Figure 2). Eyring analysis allowed to
extract the kinetic parameters for the CÀ N bond
Figure 1. Designs of chiral monodentate NHC ligands.
at preparative scale gave an access to both enantiom- rotation and these results were corroborated by DFT
ers. Copper and palladium complexes bearing IKong theoretical calculations (Figure 3). In order to simplify
have shown good performances in enantioselective calculations, a methyl group was used as N-substituent
catalysis. Herein, we disclosed the synthesis of C1- instead of cyclohexyl. Of note, the results presented
symmetric atropisomeric NHC palladium complexes later demonstrate that the nature of this N-substituent
and their resolution by chiral HPLC at preparative does not play a significant role on the rotational
scale to afford homochiral complexes with excellent barriers values. The experimental value of the rota-
optical purities. Taking advantage of the C1-symmetry tional barrier of imidazolium salt 4b (82 kJ.molÀ 1; i.e.
°
compared to the C2-symmetry which requires the t1/2 =13 s at 25 C) is in good agreement with the
consideration of meso diastereomers, these complexes lowest calculated value (4a, 87.3 kJ.molÀ 1). The DFT
were used to determine accurately the values of the approach allowed us to confirm that the rotation takes
rotational barriers and thus identify the key parameters place on the hydrogen side as demanding on the
¼
to design configurationally stable NHC-metal com- backbone side (ΔG (BB) =128 kJ.molÀ 1). Calculations
plexes such as the necessity of methyl groups on the performed on the carbene 5a showed that the rotation
NHC backbone.
is by 20 kJ.molÀ 1 easier on the carbene side compared
to the hydrogen side for imidazolium 4a. This
magnitude difference is in line with literature results.[9b]
More surprisingly, the rotation on the backbone side
Scheme 1. Preparation of imidazolium salt 4 according to Fürstner’s method. (Dipp=2,6-diisopropylphenyl).
Adv. Synth. Catal. 2021, 363, 4229–4238
4230
© 2021 Wiley-VCH GmbH

RESEARCH ARTICLE
asc.wiley-vch.de
was found also by 20 kJ.molÀ 1 easier for the carbene
¼
form compared to the imidazolium (ΔG
=
(BB)
108.3 kJ.molÀ 1 vs. 128 kJ.molÀ 1).
As the synthetic procedure developed by Fürstner
was found not convenient for the synthesis of
imidazolium salt with methyl substituents on the
backbone, we then tested the methodology reported by
Glorius (Scheme 2).[15] Dissymmetric formamidines 6
were prepared in low to moderate yields from 2-
isopropylaniline and triethylorthoformate and subse-
quent addition of amine. Then, formamidines 6 reacted
with 3-bromo- butanone in presence of diisopropyle-
°
thylamine in acetonitrile at 110 C for 20 h to give rise
to the cyclized products 7 which were not isolated and
treated with acetic anhydride and HBF4 in toluene at
°
80 C for 16 h. Imidazolium salts 4b–d were obtained
in low to moderate yields but products were easily
purified. Unexpectedly, under the same reaction con-
ditions, formamides containing two aromatic substitu-
ents 6e–g did not lead to the corresponding cyclized
products 7, but alkylated products 8 were formed but
not characterized due to a large number of isomeric
forms (Scheme 3). From 8 to get imidazolium salts 4
we applied the procedure reported by Shi.[16] A treat-
ment with triflic anhydride and trimethylamine gave
rise to the imidazolium salts 4e’–g’ with triflate as
counterion in moderate yields (from 34 to 51%).
Figure 2. Variable temperature 1H NMR spectra of imidazolium
salt 4b in C2D2Cl4 and the corresponding lineshape simulations.
Palladium Complexes Preparation and Resolution
With 6 imidazolium salts 4 in hands, we prepared the
corresponding Pd(allyl)Cl complexes 9 by generation
of the carbene with tBuOK in presence of the dimer
[Pd(allyl)Cl]2 (Scheme 4). These complexes were
found very stable and therefore could be purified by
silica gel chromatography to give the pure complexes
in good yields. The same procedure was applied to the
synthesis of Pd(cinnamyl)Cl complexes 10 which were
isolated in good to quantitative yields and displayed
1
also excellent stabilities (Scheme 5). The H NMR
Figure 3. Values for the rotational barriers of imidazolium salts
4a and 4b and the corresponding free carbene 5a (left:
experimentally determined; right DFT calculated).
spectra of complexes 9 and 10 are somewhat intricate
due to the presence of at least two isomeric forms. For
the same reason, 13C NMR spectra show a multitude of
peaks, but characteristic carbenic carbons were observ-
able in the range 178–182 ppm.[17]
Scheme 2. Preparation of imidazolium salt 4 according to Glorius’ synthetic procedure.
Adv. Synth. Catal. 2021, 363, 4229–4238
4231
© 2021 Wiley-VCH GmbH

RESEARCH ARTICLE
asc.wiley-vch.de
Scheme 3. Preparation of imidazolium salt 4 according to the procedure reported by Zhang and Shi.
Scheme 4. Synthesis of (NHC)Pd(allyl)Cl complexes 9.
Owing the excellent stability of these complexes
towards purification by silica gel chromatography, we
investigated the resolution of the racemates by chiral
HPLC at preparative scale. After a screening of
columns with different chiral stationary phases, Chir-
alpak IG was found suitable for the resolution of all
Scheme 5. Preparation of (NHC)Pd(allyl)Cl complexes 10.
complexes. With a 1 cm diameter column, batches
from 50 to 250 mg could be purified in few hours.
Selected details are gathered in Table 1.17 Both
enantiomers of each palladium complexes were iso- containing complexes. For example, the first eluted
lated in good to excellent yields (34–49%; 71–95% enantiomer of 9b is dextrorotary as well as the first
overall yields) and remarkable enantiopurities, in most eluted complex 10b (entries 1 and 6). The only
of the case >99.5% ee. Moreover, no racemization exception is for complexes containing an adamantyl N-
process was observed during the resolution suggesting substituent, complexes 9d and 10d, for which the first
a good configurational stability of the palladium eluted enantiomer 9d is levorotary whereas dextro-
complexes. The examination of specific rotations rotary for 10d (entries 3 and 8). It seems that the first
revealed that the direction of plane-polarized light eluted enantiomers of complexes bearing an alkyl N-
rotations does not change between allyl- and cinnamyl- substituent rotate plane-polarized light clockwise (en-
Adv. Synth. Catal. 2021, 363, 4229–4238
4232
© 2021 Wiley-VCH GmbH

RESEARCH ARTICLE
asc.wiley-vch.de
Table 1. Resolution by chiral HPLC of heterochiral complexes 9 and 10.
Entry Complex
(Racemate quantity) Complex
1st eluted
2nd eluted
Quantity Yield ee (%)
Overall yield
Quantity Yield ee (%) Complex
1
2
3
4
5
6
7
8
9b (69 mg)
(Sa)-(+)-9b[a] 34 mg
49% >99.5 (Ra)-(À )-9b[a] 32 mg
46% >99.5 95%
9c (132 mg)
9d (213 mg)
9e (51 mg)
9g (170 mg)
10b (106 mg)
10c (110 mg)
10d (105 mg)
10e (55 mg)
10f (135 mg)
10g (250 mg)
(Sa)-(+)-9c[b]
49 mg
37% >99.5 (Ra)-(À )-9c[b] 45 mg
44% >99.5 (Ra)-(+)-9d[b] 95 mg
43% >99.5 (Ra)-(À )-9e[b] 24 mg
41% >99.5 (Ra)-(À )-9g[b] 68 mg
43% >98.5 (Ra)-(À )-10b[a] 43 mg
36% >99.5 (Ra)-(À )-10c[a] 42 mg
37% >99.5 (Sa)-(À )-10d[a] 44 mg
34% >99.5 71%
45% >99.5 89%
47% >99.5 90%
40% >99.5 81%
41% >98.5 84%
38% >99.5 74%
42% >99.5 79%
(Sa)-(À )-9d[a] 94 mg
(Sa)-(À )-9e[b]
22 mg
(Sa)-(À )-9g[b] 70 mg
(Sa)-(+)-10b[a] 45 mg
(Sa)-(+)-10c[a] 40 mg
(Ra)-(+)-10d[a] 39 mg
(Sa)-(À )-10e[a] 25 mg
(Sa)-(À )-10f[a] 53 mg
9
10
11
46% >99
(Ra)-(+)-10e[a] 24 mg
44% >99
90%
39% >99.5 (Ra)-(+)-10f[a] 51 mg
38% >99.5 77%
38% >98.5 80%
(Sa)-(À )-10g[b] 105 mg 42% >99.5 (Ra)-(+)-10g[b] 95 mg
[a] Absolute configuration deduced from ECD spectra comparison.
[b] Absolute configuration determined by XRD spectroscopy.
tries 1,2 6 and 7), whereas the first eluted of complexes
containing aryl N-substituents are levorotary (entries 4,
5 and 9–11).
Absolute Configuration Assignments
In spite of repeated attempts, suitable crystals for
single crystal X-ray diffraction studies could be grown
only for enantiopure 2nd eluted complexes 9c, 9e, 9g
and 10g. Ball-and-stick representations depicted in
Figure 4 allowed to unambiguously establish the
structures, the atoms connectivity, as well as the
absolute configurations thanks to Flack parameters
refined to value of zero. These complexes show the
expected distorted square planar geometry around the
metal center. The carbene-Pd bond distances are
between 2.027(3) and 2.037(5) Å, and the Pd-allyl and
Pd-cinnamyl bond lengths are respectively in the
ranges of 2.101(5)–2.194(8) Å and 2.102(6)–
2.264(6) Å. These values are comparable to those
found in similar complexes.[18]
Figure 4. Ball-and-stick representations of enantiopure palladi-
um complexes (hydrogens have been omitted for clarity).
between allyl- and cinnamyl-containing complexes 9g
With the single crystal X-ray diffraction data in and 10g. The steric map of NHC ligand bearing
hands, the buried volumes (%VBur), a parameter complex 9c (upper left corner) does not allow the
describing the bulkiness of a ligand in the first identification of the clear zones of different steric
coordination sphere of a metal, and topographic steric pressures and suggests a low chiral induction in
maps have been determined (Figure 5).[19] As expected, asymmetric catalysis. In contrast, the steric map of
the NHC bearing a benzyl N-substituent (complex 9c) complex 9e NHC (upper right corner) exhibits a
was found significantly less bulky than other NHC substantial difference of steric pressure between the
containing mesityl or Dipp as N-substituent (%VBur
=
northeast and northwest quadrants. A more pronounced
35.2% vs. 36.8–37.5%). No significant difference difference was observed between these quadrants for
could be noticed either for %VBur values or steric maps,
Adv. Synth. Catal. 2021, 363, 4229–4238
4233
© 2021 Wiley-VCH GmbH

RESEARCH ARTICLE
asc.wiley-vch.de
As displayed in Figure 7, ECD spectra of first
eluted chiral Pd(cinnanyl)Cl(NHC) complexes 10e–g
were perfectly superposable with two negative ECD-
active bands at 195 (Δɛ=À 55; À 78 for 10g) and
265 nm (Δɛ=À 5) and one positive band at 225 nm
(Δɛ= +10). This perfect match might be due to both
N-aryl groups that restrict several rotations in com-
plexes structures and thus provide a more rigid chiral
environment. In the case, the absolute configuration of
complexes 10e and 10f can be confidently assigned as
(Sa)-(À ) for the first eluted enantiomers.
The assignment of the absolute configuration of
complex 10e was confirmed by the ECD spectra
comparison between allyl- and cinnamyl complexes 9e
and 10e, respectively (Figure 8). The perfect concord-
ance between both spectra enabled to attribute un-
ambiguously the same configuration to enantiomers
with the same elution order. A (Ra)-(+) absolute
configuration was attributed to the second eluted
enantiomer 9e through single crystal X-ray diffraction
studies.
Figure 5. Steric contour maps of NHC ligands (complexes are
oriented as shown in the central sketch).
However, ECD spectra of allyl- and cinnamyl
complexes 9d and 10d display almost mirror image
the NHC ligand of complexes 9g and 10g and suggests with one negative ECD-active band at 195 (Δɛ=À 26)
a better ability to induce an enantioselectivity. and two small positive bands at 220 and 235 nm (Δɛ=
UV-vis and ECD spectra for both enantiomers of all +2) for 9d whereas 10d exhibited two positive bands
the palladium complexes have been measured in at 195 (Δɛ= +15) and 270 nm (Δɛ= +8) as well as
acetonitrile solutions. Individual ECD spectra, reported two small negative bands at 220 and 235 nm (Δɛ=À 4
in the supporting information, showed expected nearly and À 2, respectively) (Figure 9). This suggests ob-
mirror-images for both enantiomers. As Electronic viously an inversion of the enantiomeric elution order
Circular Dichroism is highly sensitive towards spatial by cHPLC on the same chiral column and with the
conformations,[20] the comparison of ECD spectra, in same mobile phase[21] as hypothesized during the
particular with those of complexes that have been examination of specific rotations.
studied by X-ray crystallography, represents an effi-
cient way to assign absolute configurations through the
identification of invariant bands or bisignate patterns
in a specific region. ECD spectra of 1st eluted
complexes 9b–e are depicted in Figure 6. A similar
shape was observed for complexes 9c–e with a
negative ECD-active band between 195 and 200 nm,
albeit differential molar extinctions (Δɛ) were slightly
different. Due to the low ECD signal observed for 9b,
its absolute configuration cannot be confidently de-
duced from the comparison with other ECD spectra.
Figure 7. ECD spectra of first eluted complexes 10e–g.
Figure 6. ECD spectra of first eluted complexes 9b–e.
Figure 8. ECD spectra of first eluted complexes 9e and 10e.
Adv. Synth. Catal. 2021, 363, 4229–4238
4234
© 2021 Wiley-VCH GmbH

RESEARCH ARTICLE
asc.wiley-vch.de
decay of enantiomeric excesses over time at a given
temperature and allowed to assess the values of
rotational barriers (Figure 10).[17] Of note, the value of
the rotational barrier for complex 10b could not be
determined due to a rapid decomposition of the
complex under the conditions to monitor the kinetic of
°
enantiomerization (132 C in chlorobenzene). Rota-
tional barrier values for complex 10f containing a tert-
butyl in ortho position and complexes 9g and 10g
containing a benzhydryl moiety could not be measured.
Thus, these complexes exhibit excellent conformation
Figure 9. ECD spectra of first eluted complexes 9d and 10d.
Rotational Barriers Determination
stability and degradation processes occurred before
racemization (>170 kJ.mol , t1/2 >17 days at 180 C).
The half-lifes of racemization for palladium complexes All the complexes containing an iso-propyl in ortho
have been experimentally determined by following the position displayed a rotation barrier value in range
131–139 kJ.molÀ 1. This confirms the good conforma-
À 1
°
tional stability of these complexes: t1/2 >2 years at
°
°
50 C or t1/2 >24 hours at 100 C. As hypothesized, the
nature of the second N-substituent (R) as a limited
impact on the values of rotational barriers as well as
the nature of the allyl ligand (allyl vs. cinnamyl).
Theoretical calculations corroborated these findings,
albeit the calculated values are slightly higher than the
experimental values. For example, the lowest theoret-
ically calculated value for structure 9b is
145.3 kJ.molÀ 1 vs. 138.5 kJ.molÀ 1 for the experimental
value of complex 9b. Interestingly, theoretical calcu-
lations allowed to determine that the more favorable
¼
rotations take place on the palladium side (ΔG (Pd)).
In order to demonstrate that methyl groups on the
NHC backbone were mandatory to restrict the rotation
of N-aryl substituents along the CÀ N bond and thus
obtain configurationnally stable chiral palladium-NHC
complexes, complexes 11b and 12b possessing an
unsubstituted saturated and unsaturated NHC backbone
were prepared (Figure 11).[17] Unfortunately, no enan-
tiomers could be detected by chiral HPLC analyses,
even at low temperature. Rotational barriers for
complexes 11a and 12a with a methyl as second N-
Figure 10. Values for the rotation barriers of Pd(NHC) com-
plexes (left: experimentally determined; right DFT calculated).
Figure 11. Influence of the backbone substitution on the rotation barriers.
Adv. Synth. Catal. 2021, 363, 4229–4238
4235
© 2021 Wiley-VCH GmbH

RESEARCH ARTICLE
asc.wiley-vch.de
substituent have been theoretically studied. The calcu- were obtained. For instance, complex (Ra)-(+)-9e
lations show that rotations are more favorable on the afforded the expected oxindole in good yield (93%)
backbone side with lower values by at least but with only 19% ee. When the same reaction was
¼
10 kJ.molÀ 1. The values of ΔG
were found close carried in DMF, reactivity was found higher and
(BB)
to 75 kJ.molÀ 1 (74.3 and 77.2 kJ.molÀ 1 for complexes allowed to decrease the reaction temperature to 40 C.
11a and 12a, respectively) which indicates half-lifes Moreover, an improved enantioselectivity was ob-
of racemization quicker than one second at 25 C. As served: 46% ee. A similar chiral induction was
°
°
we were wondering if the replacement of the iso- obtained with the cinnamyl analogue (Sa)-(À )-10e,
propyl moiety by a more hindered tert-butyl could 44% ee with 97% yield. Complex (Sa)-(À )-10f
allow to significantly restricted the rotations with an containing a tert-butyl group instead of an iso-propyl
unsubstituted backbone, calculations were performed gave a slightly lower enantioselectivity (38% ee). The
for complex 13a. With ΔG (BB) =96.8 kJ.molÀ 1, it was best enantiomeric excess, 54%, was obtained with
¼
expected to observe enantiomers and therefore com- complex (Sa)-(À )-9g whereas the cinnamyl analogue
plex 13b was synthesized.[17] Unfortunately, enantiom- (Sa)-(À )-10g gave 44% ee. These levels of chiral
ers of complex 13b could not be observed by cHPLC induction were significantly lower than those generally
analyses suggesting that the value of the rotational observed with C2-symmetric chiral NHC ligands. This
barrier was slightly lower than 90 kJ.molÀ 1. Of note, might be the result of an unrestricted rotation of the
with a tert-butyl group and methyl substituents on the NHC along the carbene-metal bond.[22f]
backbone, calculations gave rotational barrier values
higher than 200 kJ.molÀ 1 (structure 14a).
Conclusion
In summary, we reported the synthesis of new class of
Asymmetric Catalysis
chiral palladium complexes bearing C1-symmetric
The palladium-catalyzed intermolecular α-arylation of NHC ligands. The chirality of theses complexes
amide has been used as benchmark reaction to evaluate resulted from the metalation step which restricted the
to ability of chiral palladium-based complexes, in rotation of dissymmetric N-aryl substituents of NHCs
particular containing chiral NHC ligands to induce and thus created an axis of chirality. Due to the
efficiently an enantioselectivity.[4b,c,5d,9a,22] Therefore, excellent stability of these complexes, a resolution by
we tested enantiopure complexes 9 and 10 in the chiral HPLC at preparative scale (up to 250 mg scale)
intermolecular α-arylation of amide 15 (Scheme 6). allowed to obtain enantiopure complexes. Remarkably,
Standard conditions reported in the literature, 1,2- all the palladium complexes could be separated by
°
dimethoxyethane (DME) at 60 C in the presence of t- cHPLC with excellent enantiopurities (typically
BuOK, were used, but only low enantiomeric excess >99.5% ee). Analyses by X-ray crystallography of
some complexes confirmed molecular structures and
determined absolute configurations. ECD spectra were
recorded for all complexes and allowed to deduce
absolute configurations. Experimental and theoretical
investigations of the rotational barriers highlighted that
the metalation step enabled the formation of configura-
tionally stable PdÀ NHC complexes from imidazolium
salts existing in solution as rotamers. Moreover, it has
been demonstrated that the substitution the NHC
backbone with methyl groups was a key parameter to
prevent the rotation of N-aryl substituents on the
backbone side. Finally, palladium complexes exhibited
good catalytic activities in the intramolecular α-
arylation of amides, albeit with moderate enantioselec-
tivities. Nevertheless, enantiomeric excesses were
found higher than those reached with analogous
palladium complexes containing chiral C1-symmetric
NHC ligands.
Scheme 6. Palladium-catalyzed Intermolecular α-arylation of
amide 15.
Adv. Synth. Catal. 2021, 363, 4229–4238
4236
© 2021 Wiley-VCH GmbH

RESEARCH ARTICLE
asc.wiley-vch.de
Experimental Section
Acknowledgements
Preparation of Dissymmetrical Imidazolium Salts 4
– General Procedure A
We are grateful to the CNRS, Aix-Marseille Université and the
ECM. L.K. and Y.C. thank the China Scholarship Council for
Ph.D. grants. We acknowledge the Agence Nationale de la
Recherche (ANR-20-CE07-0030 cResolu). We thank the Spec-
tropole, Fédération des Sciences de Marseille, in particular Dr.
Roseline Rosas for VT-NMR spectroscopy, Dr. Michel Giorgi
for DRX and Dr. Valérie Monnier and Dr. Christophe Chendo
for HRMS as well as Arnaud Treuvey (ECM) for IR spectro-
scopy. Umicore AG & Co is acknowledged for a generous gift
of complexes.
To a suspension of formamidine (1 mmol) in acetonitrile
(2 mL), N,N-Diisopropylethylamine (0.35 mL, 2 mmol,
2.0 equiv.) and 3-bromo-2-butanone (0.21 mL, 2 mmol,
2.0 equiv.) were added and the resulting mixture was stirred at
°
110 C for 20 h. The solvent was removed under vacuum. The
residue was purified by silica gel flash chromatography
(petroleum ether/ethyl acetate=10:1) to give the intermediate.
The intermediate was dissolved in dichloromethane (3 mL).
Triethylamine (142 μL, 1.1 mmol, 1.1 equiv.) was added drop-
°
wise at À 40 C. After 5 min, trifluoromethanesulfonic anhydride
References
(188 μL, 1.1 mmol, 1.1 equiv.) was added dropwise. The
mixture was slowly warmed up to room temperature and stirred
at room temperature for 4 hours. The volatiles were removed
under vacuum and the residue was purified by silica gel column
(dichloromethane/acetone=10:1) to give the imidazolium
salt 4.
[1] a) S. P. Nolan (Ed.), N-Heterocyclic Carbenes Effective
Tools for Organometallic Synthesis; Wiley-VCH: Wein-
heim 2014; b) S. Díez-González (Ed.), N-Heterocyclic
Carbenes: From Laboratory Curiosities to Efficient
Synthetic Tools: Edition 2; The Royal Society of
Chemistry: Cambridge, U. K. 2017; c) H. V. Huynh, The
Organometallic Chemistry of N-heterocyclic Carbenes;
Inorganic Chemistry: A Textbook Series, Wiley, Hoboken
2017.
Preparation of Dissymmetric Formamidines 6 –
General Procedure B
Acetic acid (86 μL, 1.5 mmol, 0.05 equiv.) was added to a
round bottom flask charged with the first amine (30 mmol,
1.0 equiv.) and triethyl orthoformate (5 mL, 30 mmol,
1.0 equiv.). The flask was fitted with a distillation head and was
stirred and heated to 140 C until ethanol (3.5 mL, 60 mmol,
2.0 equiv) was collected by distillation. The second amine
(30 mmol, 1.0 equiv) was then added to the reaction mixture.
Heating at 140 C continued until ethanol (1.75 mL, 30 mmol,
1 equiv.) was collected by distillation. Workup was performed
as indicated below, either by washing with cold (5–15 C) n-
hexane and then either recrystallization or flash chromatogra-
phy.
[2] S. Díez-Gonzalez, N. Marion, S. P. Nolan, Chem. Rev.
2009, 109, 3612–3676.
[3] a) F. Wang, L.-J. Liu, W. Wang, S. Li, M. Shi, Coord.
Chem. Rev. 2012, 256, 804–853; b) D. Janssen-Müller,
C. Schlepphorst, F. Glorius, Chem. Soc. Rev. 2017, 46,
4845–4854; c) P. J. Czerwiński, M. Michalak, Synthesis
2019, 51, 1689–1714.
°
°
[4] For representative examples, see: a) W. A. Herrmann,
L. J. Goossen, G. R. Artus, C. Köcher, Organometallics
1997, 16, 2472–2477; b) S. Lee, J. F. Hartwig, J. Org.
Chem. 2001, 66, 3402–3415; c) E. P. Kündig, T. M.
Seidel, Y.-x. Jia, G. Bernardinelli, Angew. Chem. Int. Ed.
2007, 46, 8484–8487; Angew. Chem. 2007, 119, 8636–
8639; d) S. Urban, N. Ortega, F. Glorius, Angew. Chem.
Int. Ed. 2011, 50, 3803–3806; Angew. Chem. 2011, 123,
3887–3890; e) A. Albright, D. Eddings, R. Black, C. J.
Welch, N. N. Gerasimchuk, R. E. Gawley, J. Org. Chem.
2011, 76, 7341–7351; f) A. Albright, R. E. Gawley, J.
Am. Chem. Soc. 2011, 133, 19680–19683; g) D. Shen, Y.
Xu, S.-L. Shi, J. Am. Chem. Soc. 2019, 141, 14938–
14945; h) Z.-C. Wang, P.-P. Xie, Y. Xu, X. Hong, S.-L.
Shi, Angew. Chem. Int. Ed. 2021, doi.org/10.1002/
anie.202103803; Angew. Chem. 2021, doi.org/10.1002/
ange.202103803.
°
Synthesis of the NHC-Palladium Complexes 9–13 –
General Procedure C
A mixture of imidazolium tetrafluoroborate (or imidazolium
trifluoromethylsufonate) (2.3 mmol, 2.3 equiv.), [Pd(cinnamyl)
Cl]2 or [Pd(allyl)Cl]2 (1.0 mmol, 1.0 equiv.), KOtBu (2.3 mmol,
°
2.3 equiv.) in acetone was stirred at 25 C for 5 h. The reaction
mixture was filtered through a Celite® pad and the solvent was
removed under vacuum. The crude product was purified by
silica gel column (PE/Et2O=1:1) give the expected NHC-
palladium complex.
[5] For selected publications, see: a) S. Würtz, C. Lohre, R.
Fröhlich, K. Bergander, F. Glorius, J. Am. Chem. Soc.
2009, 131, 8344–8345; b) D. Baskakov, W. A. Herr-
mann, E. Herdtweck, S. D. Hoffmann, Organometallics
2007, 26, 626–632; c) C. Metallinos, X. Du, Organo-
metallics 2009, 28, 1233–1242; d) C. Xu, Y. Feng, L.-R.
Wang, W.-P. Ma, Y.-M. He, Q.-H. Fan, Organometallics
2020, 39, 1945–1960.
Crystallographic Data
CCDC numbers 1918272 ((Ra)-(+)-9e), 2049183 ((Ra)-(À )-
9c), 2049184 ((Ra)-(+)-9g), 2049185 ((Ra)-(+)-10g) and
2049186 (11b) contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.ca-
m.ac.uk/data_request/cif.
[6] For selected publications, see: a) D. R. Jensen, M. S.
Sigman, Org. Lett. 2003, 5, 63–65; b) J. Pytkowicz, R.
Adv. Synth. Catal. 2021, 363, 4229–4238
4237
© 2021 Wiley-VCH GmbH

RESEARCH ARTICLE
asc.wiley-vch.de
Roland, P. Mangeney, G. Meyer, A. Jutand, J. Organo-
met. Chem. 2003, 678, 166–179; c) T. Arao, K. Kondo,
T. Aoyama, Tetrahedron Lett. 2006, 47, 1417–1420;
d) K. B. Selim, Y. Matsumoto, K.-i. Yamada, K. Tomio-
ka, Angew. Chem. Int. Ed. 2009, 48, 8733–8735; Angew.
Chem. 2009, 121, 8889–8891; e) H. Wang, G. Lu, G. J.
Sormunen, A. Hasnain, H. A. Malik, P. Liu, J. Mont-
Chem. Soc. 2012, 134, 14322–14325; b) F. Grande-
Carmona, J. Iglesias-Sigüenza, E. Álvarez, E. Díez, R.
Fernández, J. M. Lassaletta, Organometallics 2015, 34,
5073–5080; c) I. Varela, H. Faustino, E. Díez, J. Iglesias-
Sigüenza, J. Grande-Carmona, F. Fernández, R. Lassa-
letta, J. M. Mascareñas, J. L. López, ACS Catal. 2017, 7,
2397–2402.
gomery, J. Am. Chem. Soc. 2017, 139, 9317–9324; f) W.- [13] L. Kong, J. Morvan, D. Pichon, M. Jean, M. Albalat, T.
Y. Wang, J.-Y. Wu, Q.-R. Liu, X.-Y. Liu, C.-H. Ding, X.-
L. Hou, Org. Lett. 2018, 20, 4773–4776; g) Y. Chen, L.
Dang, C.-Y. Ho, Nat. Commun. 2020, 11, 2269.
[7] T. J. Seiders, D. W. Ward, R. H. Grubbs, Org. Lett. 2001,
3, 3225–3228.
Vives, S. Colombel-Rouen, M. Giorgi, V. Dorcet, T.
Roisnel, C. Crévisy, D. Nuel, P. Nava, S. Humbel, N.
Vanthuyne, M. Mauduit, H. Clavier, J. Am. Chem. Soc.
2020, 142, 93–98.
[14] A. Fürstner, M. Alcarazo, V. César, C. W. Lehmann,
Chem. Commun. 2006, 2176–2178.
[8] For representative examples, see: a) M. R. Chaulagain,
G. J. Sormunen, J. Montgomery, J. Am. Chem. Soc. [15] K. Hirano, S. Urban, C. Wang, F. Glorius, Org. Lett.
2007, 129, 9568–9569; b) C.-Y. Ho, C.-W. Chan, L. He, 2009, 11, 1019–1022.
Angew. Chem. Int. Ed. 2015, 54, 4512–4516; Angew. [16] J. Zhang, J. Fu, X. Su, X. Wang, S. Song, M. Shi, Chem.
Chem. 2015, 127, 4595–4599; c) R. Kumar, Y. Hoshimo- Asian J. 2013, 8, 552–555.
to, H. Yabuki, M. Ohashi, S. Ogoshi, J. Am. Chem. Soc. [17] For details, see supporting information.
2015, 137, 11838–11845; d) J. S. E. Ahlin, N. Cramer, [18] N. Marion, O. Navarro, J. Mei, E. D. Stevens, N. M.
Org. Lett. 2016, 18, 3242–3245; e) Z. Li, L. Zhang, M.
Nishiura, G. Luo, Y. Luo, Z. Hou, ACS Catal. 2020, 10,
11685–11692.
Scott, S. P. Nolan, J. Am. Chem. Soc. 2006, 128, 4101–
4111.
[19] L. Falivene, Z. Cao, A. Petta, L. Serra, A. Poater, R.
Oliva, V. Scarano, L. Cavallo, Nat. Chem. 2019, 11, 872–
879.
[9] a) X. Luan, R. Mariz, C. Robert, M. Gatti, S. Blumen-
tritt, A. Linden, R. Dorta, Org. Lett. 2008, 10, 5569–
5572; b) X. Luan, R. Mariz, M. Gatti, C. Costabile, A. [20] G. Pescitelli, L. Di Bari, N. Berova, Chem. Soc. Rev.
Poater, L. Cavallo, A. Linden, R. Dorta, J. Am. Chem.
Soc. 2008, 130, 6848–6858.
[10] a) D. S. Clyne, J. Jin, E. Genest, J. C. Gallucci, T. V.
Rajanbabu, Org. Lett. 2000, 2, 1125–1128; b) W. L.
Duan, M. Shi, G.-B. Rong, Chem. Commun. 2003, 2916–
2917; c) L.-j. Liu, F. Wang, M. Shi, Organometallics
2011, 40, 4603–4625.
[21] N. Vanthuyne, C. Roussel, Chiroptical Detectors for the
Study of Unusual Phenomena in Chiral Chromatogra-
phy; In Differentiation of Enantiomers I. Topics in
Current Chemistry, V. Schurig (Ed.), Springer, Cham.,
2013, vol. 340, pp. 107–151.
2009, 28, 4416–4420; d) C. Bartolomé D García-Cuadra- [22] For selected examples, see: a) Y.-X. Jia, J. M. Hillgren,
do, Z. Ramiro, P. Espinet, Inorg. Chem. 2010, 49, 9758–
9764; e) Y.-M. Wang, C. N. Kuzniewski, V. Rauniyar, C.
Hoong, F. D. Toste, J. Am. Chem. Soc. 2011, 133,
12972–12975.
E. L. Watson, S. P. Marsden, E. P. Kündig, Chem.
Commun. 2008, 4040–4042; b) S. Würtz, C. Lohre, R.
Fröhlich, K. Bergander, F. Glorius, J. Am. Chem. Soc.
2009, 131, 8344–8345; c) X. Luan, L. Wu, E. Drinkel,
R. Mariz, M. Gatti, R. Dorta, Org. Lett. 2010, 12, 1912–
1915; d) L. Liu, N. Ishida, S. Ashida, M. Murakami,
Org. Lett. 2011, 13, 1666–1669; e) L. Wu, L. Falivene,
E. Drinkel, S. Grant, A. Linden, L. Cavallo, R. Dorta,
Angew. Chem. Int. Ed. 2012, 51, 2870–2873; Angew.
Chem. 2012, 124, 2924–2927; f) D. Katayev, Y.-X. Jia,
A. K. Sharma, D. Banerjee, C. Besnard, R. B. Sunoj,
E. P. Kündig, Chem. Eur. J. 2013, 19, 11916–11927.
[11] a) J. J. Van Veldhuizen, S. B. Garber, J. S. Kingsbury,
A. H. Hoveyda, J. Am. Chem. Soc. 2002, 124, 4954–
4955; b) S. Handa, L. M. Slaughter, Angew. Chem. Int.
Ed. 2012, 51, 2912–2915; Angew. Chem. 2012, 124,
2966–2969.
[12] a) J. Francos, F. Grande-Carmona, H. Faustino, J.
Iglesias-Sigüenza, E. Díez, I. Alonso, R. Fernández,
J. M. Lassaletta, F. López, J. L. Mascareñas, J. Am.
Adv. Synth. Catal. 2021, 363, 4229–4238
4238
© 2021 Wiley-VCH GmbH