Novel Pd(PN,S)-complexes: Highly active catalysts designed for asymmetric allylic etherification

Article abstract of DOI:10.1016/j.mcat.2021.111763

Six novel thioether-aminophosphine type ligands with a general formula (Ar¹)₂PN(R¹)CHR²(CH₂)_nCH(R³)SAr² has been synthesized. The modular structure of the ligands and the new methodologies developed for their preparation enabled the systematic variation of their bridge length (n = 0 or 1), the substitution pattern of the backbone (R², R³ = H or Me) as well as the P-, N- and S-substituents (Ar¹ = Ph or 3,5-Me₂C₆H₃, R¹ = Et or iPr and Ar² = Ph, 4-MeC₆H₄, or 4-MeOC₆H₄, respectively). The ligands proved to be effective in Pd-catalyzed asymmetric allylic etherification reactions providing the products in high yields (up to 95%) and with good enantioselectivities (up to 86%) using unprecedentedly low (0.2 mol%) loadings of the chiral Pd-catalyst. Based on these findings, a new scalable protocol has been developed for the preparation of chiral allylic ethers. Furthermore, the Pd(II) coordination chemistry of the ligands was thoroughly investigated by 1D and 2D NMR methods as well as by X-ray crystallography with special attention to the conformation of the chelate ring and the stereoselectivity of the sulfur coordination. Based on these studies, the main factors determining activity and selectivity of the catalytic system have been identified.

Full text of DOI:10.1016/j.mcat.2021.111763

Molecular Catalysis 512 (2021) 111763
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.journals.elsevier.com/molecular-catalysis
Novel Pd(PN,S)-complexes: Highly active catalysts designed for asymmetric
allylic etherification
a
a
a,*
b
c
d
´ ´
´
´
´
´
´
Mate M. Major , Maria Guoth , Szabolcs Balogh , Jozsef Simon , Attila C. Benyei ,
a
Jozsef Bakos , Gergely Farkas
a
´
University of Pannonia, Department of Organic Chemistry, Egyetem u. 10., Veszprem 8200, Hungary
b
c
´
University of Pannonia, NMR Laboratory, Egyetem u. 10., Veszprem 8200, Hungary
´
University of Pannonia, Department of Analytical Chemistry, Egyetem u. 10., Veszprem 8200, Hungary
d
´
University of Debrecen, Department of Pharmaceutical Chemistry, Egyetem ter 1, Debrecen 4032, Hungary
A R T I C L E I N F O
A B S T R A C T
Six novel thioether-aminophosphine type ligands with a general formula (Ar1)2PN(R1)CHR2(CH2)nCH(R3)SAr2
has been synthesized. The modular structure of the ligands and the new methodologies developed for their
preparation enabled the systematic variation of their bridge length (n = 0 or 1), the substitution pattern of the
backbone (R2, R3 = H or Me) as well as the P-, N- and S-substituents (Ar1 = Ph or 3,5-Me2C6H3, R1 = Et or iPr and
Ar2 = Ph, 4-MeC6H4, or 4-MeOC6H4, respectively). The ligands proved to be effective in Pd-catalyzed asymmetric
allylic etherification reactions providing the products in high yields (up to 95%) and with good enantiose-
lectivities (up to 86%) using unprecedentedly low (0.2 mol%) loadings of the chiral Pd-catalyst. Based on these
findings, a new scalable protocol has been developed for the preparation of chiral allylic ethers. Furthermore, the
Pd(II) coordination chemistry of the ligands was thoroughly investigated by 1D and 2D NMR methods as well as
by X-ray crystallography with special attention to the conformation of the chelate ring and the stereoselectivity
of the sulfur coordination. Based on these studies, the main factors determining activity and selectivity of the
catalytic system have been identified.
Keywords:
Thioether-aminophosphine
Allylic etherification
Stereoselective coordination
Palladium catalysis
Introduction
metal catalyst may occur. [1] Their application, however, may open up
new horizons towards the direct synthesis of optically active allylic
The palladium-catalyzed asymmetric allylic substitution represents
one of the most versatile methodologies for the enantioselective con-
struction of C-C and C-heteroatom bonds and thus for the efficient
synthesis of biologically active chiral compounds.[1–5] These types of
catalytic reactions typically operate under mild reaction conditions and
coupling partners with a wide array of functional groups are
well-tolerated. In the past several decades, the development of enan-
tioselective C-C and C-N bond forming reactions using racemic allylic
starting materials played a central role in this field of catalyst research.
[6] More recently, however, the focus has been somewhat shifted to-
wards the Pd-catalyzed enantioselective construction of C-O bond that
facilitates the synthesis of chiral ethers and related derivatives. [7–9]
Despite extensive efforts, the direct utilization of alcohols as nucle-
ophiles still represent a challenging direction of research. Amongst O-
nucleophiles alcohols are usually considered as poor nucleophiles and
when applied in an alkoxide form, elimination and deactivation of the
ethers. [10] Indeed, in the last several years novel catalytic systems
emerged that proved to be highly enantioselective (up to 99% ee) in
asymmetric allylic etherification reactions utilizing alcohols as nucleo-
philes. [11–19] However, the practical use of these catalysts may be
strongly limited by their low activity. A detailed survey of the literature
clearly shows that generally 4-5 mol% (substrate/catalyst molar ratio
(S/C) = 20-25) of the chiral catalyst is applied in the Pd-catalyzed allylic
etherification using 1,3-diphenylallyl acetate as benchmark substrate.
Furthermore, to the best of our knowledge there is no example for the
application of the catalyst in less than 1 mol% (S/C = 100) concentra-
tion relative to the substrate in these types of catalytic transformations.
[20] The high loading of the palladium catalyst required significantly
increases the cost of each catalytic run due to the high world market
price of this transition metal. Additionally, high metal loadings relative
to the substrate increase the risk of poisoning subsequent reactions or
contaminating the final product and also strongly hinder the scalability
* Corresponding author.
E-mail address: gerifarkas@almos.uni-pannon.hu (G. Farkas).
https://doi.org/10.1016/j.mcat.2021.111763
Received 27 May 2021; Received in revised form 9 July 2021; Accepted 13 July 2021
Available online 29 July 2021
2468-8231/© 2021 The Author(s).
Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).

M.M. Major et al.
Molecular Catalysis 512 (2021) 111763
of the catalytic reaction. These facts clearly hamper the application of
the present catalytic procedures for efficient palladium-mediated allylic
etherification. As a result, there is a significant need to develop new
synthetic protocols that employ low catalyst loadings, thus making
Pd-catalyzed allylic etherification a cost-effective, safe and scalable
synthetic methodology.
synthesize a novel family of chiral thioether-aminophosphine ligands
and to exploit their advantageous features provided by (i) their high
modularity and convenient synthesis, (ii) the electronic differentiation
induced by their different donoratoms as well as (iii) the stereolabile
nature of the thioether function. Our additional aim was to systemati-
cally change the stereoelectronic features of the ligands in order to
improve the activity and enantioselectvitiy of their Pd-complexes.
In the last several years, chiral phosphorus-thioether ligands became
a privileged class of stereoselectors in transition metal catalysis. [21,22]
The combination of a P,S heteroatom pair facilitates the stereoelectronic
desymmetrization of the catalyst and the independent modification of
the two functionalities enables subtle changes in the catalyst structure.
Furthermore, by the proper choice of the ligand scaffold the donor
sulfur-atom coordinates stereoselectively to the transition metal, thus
promoting the successful stereochemical communication between the
Results and discussion
Synthesis of the new ligands
Recently, we have developed a class of simple alkane-diyl based
thioether-amine type (S,N) chiral ligands with secondary amine func-
tionality using ethyl (S)-lactate (1) [29] or the corresponding cyclic
sulfates as starting materials (2a-d, Fig. 2). [30]
catalyst
and
the
substrate.
[23,24]
In
this
context,
thioether-aminophosphines and -phosphoramidites (PN,S ligands) offer
themselves as promising candidates for stereoelectronic fine-tuning to
improve catalyst activity and enantioselectivity. In these types of com-
pounds, the nitrogen atom may strategically be functionalized by bulky
substituents to further enhance steric control. Furthermore, the varia-
tion of the N-substituents can increase the modularity of the synthetic
protocol and enables the electronic fine-tuning of the phosphorus donor.
Indeed, Xiao and Chen et al. developed a library of highly modular
thioether-phosphoramidites based on 1,2-diphenylethanediyl backbone
(Fig. 1). [25] These ligands proved to be extremely efficient in
Pd-catalyzed indole allylic alkylation as well as in the asymmetric
construction of quaternary stereocenters using palladium-catalyzed
decarboxylative [4+2] cycloaddition reactions. [26] In both types of
catalytic processes very high yields (up to 99%) and ees (up to 98%)
were achieved. Thioether-aminophosphines and -phosphoramidites
were synthesized by Zeng and coworkers and were successfully used in
Pd-catalyzed asymmetric allylic alkylation and in Cu-catalyzed conju-
gate additions. [27] Although with almost quantitative yields, only
moderate ees could be obtained (up to 76% ee in the Cu-catalyzed pro-
cess). Chan et al. developed a small library of ferrocene-based thio-
ether-aminophosphine ligands (FerroNPS, Fig. 1). [28] These
stereoselectors afforded high enantioselectivities (up to 96% ee) in the
asymmetric allylic etherification of racemic 1,3-diphenyl-2-propenyl
acetate with a wide array of alcohols.
Ligands 1 and 2a-d could be successfully applied in palladium-
catalyzed asymmetric allylic alkylation reactions, but proved to be
catalytically inactive in allylic etherification. A prosperous feature of
this ligand class, however, is the possibility of a subsequent function-
alization step due to the presence of the NH unit. Indeed, in a single step,
the addition of 1.5 equiv. of chlorophosphine in the presence of Et3N and
DMAP (4-dimethylaminopyridine) results in the formation of the cor-
responding thioether-aminophosphine ligands in excellent yield (Fig. 3).
Based on the methods leading to compounds 1 and 2a-e, a highly
modular synthetic procedure was developed for the synthesis of PN,S
ligands that allows the easy modification of the S-, N- and P-substituents
and the ligand backbone as well.
Palladium-catalyzed asymmetric allylic etherification
The novel ligands were tested in asymmetric allylic etherification
reactions. In the first set of experiments, the allylic etherification of 1,3-
diphenylallyl acetate as substrate with 4-methoxybenzyl alcohol was
performed to screen ligands L1 and L2a-e (Table 1). Based on the usual
literature procedures, the catalytic reactions were carried out with the
Pd-catalyst prepared in situ from [Pd(
η
3-allyl)Cl]2 and the chiral ligand
or with preformed [Pd(L2)(
η
3-Ph2-allyl)]BF4 complex in the case of L2b
and L2c (entries 4 and 7, Table 1), at a substrate/Pd molar ratio of 20 (5
Encouraged by these literature achievements we envisaged to
mol% catalyst relative to the substrate), at room temperature in toluene
Fig. 1. Thioether-phosphoramidite and thioether-aminophosphine (PN,S) ligands used in asymmetric catalytic transformations.
2

M.M. Major et al.
Molecular Catalysis 512 (2021) 111763
Fig. 2. Synthesis of alkanediyl based thioether-amines.
Fig. 3. Synthesis of thioether-aminophosphine L1 and L2a-e.
and high isolated yields (up to 98%) and good enantioselectivities (up to
88%) could be obtained. Ligand L1 possessing propane-1,2-diyl back-
bone and L2a having pentane-2,4-diyl bridge and N-ethyl substituent
provided much lower ee (entries 1 and 2) compared to N-iPr containing
pentane-2,4-diyl based systems L2b-e (entries 3-12). These results
emphasize the significant role of the backbone as well as the N-sub-
situent in determining enantioselectivity. Interestingly, the S- and P-
substituents seem to have a less significant effect on ee. Nevertheless, the
best enantioselectivity in toluene (86%) could be obtained with ligand
L2c containing S-tolyl and P-phenyl substituents (entries 6 and 7). It is
noteworthy that the reaction proceeds smoothly without loss of enan-
tioselectivity in alternative reaction media like 2-methyl tetrahydro-
furan or methyl-tert-butyl ether (MTBE), although with some lower yield
in the latter.
Table 1
Initial screening of the chiral ligands in asymmetric allylic etherification of 1,3-
diphenylallyl acetate with 4-methoxybenzyl alcohola
Entry
Ligand
Solvent
Yield
(%)b
97
Ee
(%)c
1
L1
toluene
toluene
toluene
toluene
2-Me-
40 (R)
60 (S)
80 (S)
84 (S)
75 (S)
2
L2a (R1 = H, R2 = Et, Ar = Ph)
L2b (R1 = H, R2 = iPr, Ar = Ph)
L2b (R1 = H, R2 = iPr, Ar = Ph)
L2b (R1 = H, R2 = iPr, Ar = Ph)
96
3
96
4d
5
96
97
THF
6
L2c (R1 = Me, R2 = iPr, Ar = Ph)
L2c (R1 = Me, R2 = iPr, Ar = Ph)
L2c (R1 = Me, R2 = iPr, Ar = Ph)
toluene
toluene
2-Me-
95
96
97
86 (S)
86 (S)
75 (S)
7d
8
THF
9
L2c (R1 = Me, R2 = iPr, Ar = Ph)
L2c (R1 = Me, R2 = iPr, Ar = Ph)
L2d (R1 = OMe, R2 = iPr, Ar = Ph)
L2e (R1 = H, R2 = iPr, Ar = 3,5-Me2-
C6H3)
MeCN
MTBE
toluene
toluene
98
92
96
96
71 (S)
88 (S)
76 (S)
83 (S)
Based on these catalytic results we decided to expand the scope of the
reaction with respect to the nucleophilic alcohols (Table 2). The cata-
lytic tests were conducted under the same condition as in the first set of
experiments. Again, high yields and good enantioselectivities could be
obtained in the etherification process. As it was observed earlier, [28]
benzylic alcohols with electron-donating substituents provided higher
enantioselectivities and the ee slowly diminished as the substituent
became more electron-deficient (entries 1-5, Table 2). Although with
somewhat decreased activity and ee, the catalytic reaction took place
with 2-pyridylmethanol (entry 6). In this case, however, the heteroatom
may competitively coordinate to the metal center of the catalyst thus
affecting both the activity and enantioselectivity. Interestingly, using
2-(hydroxymethyl)tiophene and 2-(hydroxymethyl)furane as nucleo-
philes high yields (97 and 98%, respectively) could be obtained (entries
7 and 8). Nevertheless, it was clearly proved that the Pd-catalyst
modified by ligand L2c can effectively be used in the asymmetric
allylic etherification with both aromatic and aliphatic alcohols to pro-
duce chiral allylic ethers in high yields and with good enantioselectiv-
ities (up to 88% ee).
10
11
12
aReaction conditions: catalyst prepared in situ from 2.5 mol% of [Pd(η3-allyl)Cl]2
and 5.25 mol% chiral ligand; substrate and alcohol: 0.25 mmol of 1,3-dipheny-
lallyl acetate and 0.75 mmol of 4-methoxybenzyl alcohol; base: 0.75 mmol of
Cs2CO3; solvent: 1.25 mL; temperature: RT; reaction time: 24 h.
bYield of isolated product after column chromatography.
cDetermined by HPLC on a chiral stationary phase.
dThe reactions were carried out by using preformed [Pd(L2b)(η3-Ph2-allyl)]BF4
(5 mol%) or of [Pd(L2c)(η3-Ph2-allyl)]BF4 (5 mol%) complex instead of in situ
formed catalysts.
as solvent. Cesium-carbonate (Cs2CO3) was chosen as a base due to its
mild basic nature, relatively high solubility in aprotic solvents and to the
fact that Cs-salts are less prone to ion-pairing, thus providing higher
reactivity and selectivity compared to other alkaline metal bases.
[31–33]
In each case, the reactions completed after 24 h as indicated by TLC
In order to explore the potential of our catalytic system we
3

M.M. Major et al.
Molecular Catalysis 512 (2021) 111763
Table 2
Screening of nucleophilic reagents in the catalytic reactiona
Entry
1
Nucleophilic reagent (R-OH)
Yield (%)b
95
Ee (%)c
86
2
3
97
96
82
70
4
5
99
97
72
40
6
87
72
7
8
9
97
98
98
76
80
86
10
11
97
97
88
81
aReaction conditions: catalyst prepared in situ from 2.5 mol% of [Pd(η3-allyl)Cl]2 and 5.25 mol% of L2c; substrate and alcohol: 0.25
mmol of 1,3-diphenylallyl acetate and 0.75 mmol of nucleophile; base: 0.75 mmol of Cs2CO3; solvent: 1.25 mL; temperature: RT;
reaction time: 24 h.
bYield of isolated product after column chromatography
cDetermined by HPLC on a chiral stationary phase. The configuration of the prevailing enantiomer is (S).
investigated the effect of the substrate/catalyst (S/C) molar ratio on the
yield and selectivity of the asymmetric allylic etherification of 1,3-
successful catalytic application of chiral Pd(II)-complexes in this
benchmark allylic etherification reaction at catalyst loadings less than 1
mol%.
diphenylallyl acetate with allyl alcohol using preformed [Pd(L2c)(η3
-
Ph2-allyl)]BF4 as catalyst. Besides raising the S/C molar ratio by
increasing the amount of the substrate, we decided to change the sub-
strate/nucleophile/base ratio from 1/3/3 to 1/1.2/1.2 to enhance the
effectiveness of the reaction and to reduce its environmental impact.
Furthermore, the amount of toluene has not been proportionally
increased relative to the substrate, in each case the reactions were per-
formed in 2.5 mL of the solvent.
Coordination chemistry
Synthesis of [Pd(PN,S)Cl2] and [Pd(PN,S)(
η
3-Ph2-allyl)]BF4 complexes
In order to evaluate the origin of the powerful catalytic properties of
the novel bidentate thioether-aminophosphines, we decided to investi-
gate their palladium coordination chemistry. Therefore, [Pd(PN,S)Cl2]
To our delight, the reactions proceeded smoothly even at a S/C ratio
of 500 and unprecedentedly high yields could be obtained under these
conditions (Table 3). Furthermore, the increase in the substrate/catalyst
molar ratio does not affect the enantioselectivity, only a slight decrease
in ee could be detected at a S/C ratio of 1000.
complexes of ligands L1, L2a and L2b and [Pd(PN,S)(
η
3-Ph2-allyl)]BF4
type coordination compounds of L2b and L2c were synthesized. They
can easily be prepared by using the suitably labile precursors [Pd(COD)
Cl2] or [Pd(COD)(
η
3-Ph2-allyl)]BF4, respectively, by the addition of one
molar equivalent of chiral ligand (Fig. 5). It is important to note that the
two possible configurations of the coordinated sulfur donor and the
different conformations associated with the C1-symmetry seven-
membered chelate ring as well as the isomerism of the diphenylallyl
Encouraged by these promising results we extended this methodol-
ogy to several other representative nucleophiles (Fig. 4). At S/C = 500
the reactions were performed under the same conditions as was
described above. Although the catalytic reactions depend on the
nucleophilic alcohol, it is indicated that high yields and good enantio-
selectivities can be obtained even by using only 0.2 mol% of the Pd-
catalyst containing chiral thioether-aminophosphine L2c. Addition-
ally, not only aromatic but aliphatic alcohols could be applied as nu-
cleophiles. To the best of our knowledge this is the first report on the
moiety in the [Pd(PN,S)(
η
3-Ph2-allyl)]BF4 complexes enable the for-
mation of stereoisomeric structures in the complexation reaction.
X-ray studies
X-ray structure analysis of single crystals grown by slow evaporation
of the solvent from the solution of [Pd(L2a)Cl2] and [Pd(L2b)Cl2] in a
4

M.M. Major et al.
Molecular Catalysis 512 (2021) 111763
Table 3
The effect of substrate/catalyst molar ratioa
Entry
1d
2d
3d
4
S/C molar ratio
Yield (%)b
Ee (%)c
86
20
98
98
98
93
64
100
250
500
1000
86
86
86
5
83
aReaction conditions: catalyst: 0.0025 mmol of preformed [Pd(η3-PhCHCHCHPh)(L2c)]BF4; substrate: diphenylallyl acetate; nucleophile: 1.2 equiv. allyl alcohol; base:
1.2 eqiv. of Cs2CO3; solvent: 2.5 mL of toluene; temperature: RT; reaction time: 24 h.
bIsolated yield.
cDetermined by chiral HPLC. The configuration of the prevailing enantiomer is (S).
dReaction time: 5 h.
Fig. 4. Isolated yields and enantioselectivities of allylic ethers synthesized by asymmetric allylic etherification at a substrate/catalyst molar ratio of 500 (the
additional reaction conditions are given in the footnote of Table 3).
Fig. 5. Synthesis of Pd-complexes of the novel thioether-aminophosphines.
mixture of acetone and dichloromethane has been performed. In both
cases, the coordination sphere of the seven-membered Pd-PN,S chelate
has the usual slightly distorted square planar geometry (Fig. 6). The Pd-
Cl length trans to P is somewhat longer than that trans to S indicating the
larger trans influence of the phosphorus compared to the sulfur
(Table 4).
In the distorted boat-like seven-membered chelate rings the coordi-
nated sulfur donor has (S)-configuration with axially disposed Ph-
substituent in both cases. The nitrogen atom has a trigonal planar ge-
ometry with bond angles of 119.9(7)◦, 119.4(6)◦ and 118.5(6)◦ for [Pd
(L2a)Cl2] and 118.4(9)◦, 118.9(10)◦ and 121.5(11)◦ for [Pd(L2b)Cl2]
around it. The N-atom is displaced by 0.132 and 0.101 Å from the PCC
5

M.M. Major et al.
Molecular Catalysis 512 (2021) 111763
Fig. 6. X-ray structures of [Pd(L2a)Cl2] and [Pd(L2b)Cl2] (top) and the pentane-2,4-diyl moiety of the chelate in [Pd(L2a)Cl2] from different perspectives (bottom)
(ORTEP views at 50% probability level, hydrogen atoms are omitted for clarity).
the 1D spectra. Upon cooling the samples to 193 K, however, the signals
Table 4
of complexes [Pd(L1)Cl2] and [Pd(L2a)Cl2] split, indicating the pres-
Selected bond lengths (Å), bond angles (◦) and torsion angles (◦) of complexes
ence of two equilibrating isomers in a ratio of ~9:1 in both cases. These
isomers may differ in their chelate conformation, in the configuration of
the sulfur donor or both. In contrast to compound [Pd(L2a)Cl2] with N-
ethyl substituent and pentane-2,4-diyl backbone, complex [Pd(L2b)Cl2]
having N-isopropyl group exists in solution as a single diastereomer. The
detailed NMR analysis of [Pd(L2b)Cl2] and the major isomer of [Pd(L2a)
Cl2] proves that they have the same chelate conformation and sulfur
atom configuration as observed in their solid state X-ray structures. It
was unambiguously proved by the characteristic coupling pattern of the
(S,S)-pentane-2,4-diyl framework, where the same coupling constants
could be observed for both methylene protons Hb and Hc (Fig. 7). The
couplings of Hb with the adjacent CH’s, Ha and Hd are significantly
different (~2 and ~14 Hz, respectively) that strongly suggests their syn-
and anti-positions relative to Hb, respectively. As the same argument is
valid for Hc, the planar arrangement of the pentane-2,4-diyl moiety can
nicely be deduced from the solution phase 1H NMR data. Additionally,
the indicative NOESY crosspeaks between the corresponding 1H signals
provide further support for this backbone conformation (Fig. 7) and for
the axial arrangement of the S-Ph substituent. These observations
together with our earlier findings [35] underline the significance of the
pentane-2,4-diyl backbone in the stabilization of ring conformation and
provide unambiguous evidence for its enhanced conformational rigidity.
Obviously, the increased double bond character of the P-N bond and its
sterically crowded environment may also largely contribute to the
remarkable conformational rigidity of the seven-membered chelate in
[Pd(L2a)Cl2] and [Pd(L2b)Cl2]. Although the exact stereochemical
characterization of the equilibrating [Pd(L1)Cl2] isomers and the minor
isomer of [Pd(L2a)Cl2] could not be carried out, their solution phase
dynamics proves that both the chelate ring size and the N-alkyl sub-
stituent has a substantial role in determining the stereoselectivity of the
coordination, a major factor affecting enantioselectivity in catalysis. In
this context, it is important to recall that the ee values achieved by using
ligands L1 (40% ee) and L2a (60% ee) are considerably lower than those
obtained by the application of pentane-2,4-diyl based ligands having
N-iPr substituents (ees up to 86%) under identical conditions (Table 1).
The promising catalytic features of the novel pentane-2,4-diyl based
ligands with N-iPr substituent prompted us to investigate their coordi-
[Pd(L2a)Cl2] and [Pd(L2b)Cl2]a.
[Pd(L2a)Cl2]
[Pd(L2b)Cl2]
Bond lengths (Å)
Pd-S
2.305(2)
2.263(2)
1.673(7)
2.301(2)
2.360(2)
2.315(3)
2.258(4)
1.665(11)
2.301(4)
2.360(4)
Pd-P
P-N
Pd-Cl (trans to S)
Pd-Cl (trans to P)
Bond angles (◦)
P-Pd-S
97.59(7)
90.73(8)
119.4(6)
118.5(6)
119.9(7)
99.24(12)
90.17(15)
118.4(9)
Cl-Pd-Cl
P-N-C4
P-N-C(alkyl)
C4-N-C(alkyl)
Torsion angles (◦)
C20-C2-C3-C4
C40-C4-C3-C2
118.9(10)
121.5(11)
175.7(7)◦
177.5(8)◦,
175.9(13)
178.1(12)
aComplexes were crystallized via slow evaporation of the solvent from their
solution in acetone/dichloromethane mixture.
plane and the P-N bond length is 1.673(7) and 1.665(11) Å, for the
complexes [Pd(L2a)Cl2] and [Pd(L2b)Cl2], respectively. These are clear
indications that the P-N bond has a considerable double bond character.
[34] It is particularly interesting that the torsion angles C20-C2-C3-C4
and C40-C4-C3-C2 in the pentane-2,4-diyl moiety are very close to
180◦ (Fig. 6, Table 4). In other words, the pentane-2,4-diyl framework
adopts a nearly planar zig-zag conformation so that both methyl groups
are directed equatorially in the chelate. It is important to note that the
same arrangement of the backbone has been observed for uncoordinated
pentane-2,4-diyl based phosphine-phosphoramidite ligands. [35]
NMR analysis
[Pd(PN,S)Cl2] type complexes of ligands L1, L2a and L2b have been
characterized by 1H, 13C{1H}, 31P{1H}, 1H-1H NOESY, 1H-1H COSY and
HSQC NMR techniques using CD2Cl2 as solvent in order to have a deeper
insight into their solution phase behavior. The ligands coordinated in a
bidentate fashion in each case as was shown by the significant coordi-
nation shift of the corresponding signals relative to the free ligand in the
1H, 13C{1H} and 31P{1H} NMR spectra of the complexes. At room tem-
perature only one set of signals could be observed for each complex in
nation properties using the ionic complex [Pd(L2b)(
η
3-Ph2-allyl)]BF4.
6

M.M. Major et al.
Molecular Catalysis 512 (2021) 111763
Fig. 7. NOE interactions and the characteristic coupling pattern of the (S,S)-pentane-2,4-diyl skeleton stabilized in planar zig-zag conformation in the major isomer
of [Pd(L2b)Cl2].
The 1H and 31P NMR spectra of [Pd(L2b)(
η
3-Ph2-allyl)]BF4 revealed the
that the substrate-complex is prone to structural changes despite its rigid
chelate ring. Furthermore, the stereolability of the catalyst may strongly
contribute to the acceleration of certain transitions in the catalytic cycle
through the adaptation of the ligand’s structure thus increasing the rate
of the overall process [37,38].
presence of two isomers in 8.6:1.4 molar ratio at room temperature in
CD2Cl2. These isomers can be assigned as the exo- and endo-complexes
(Fig. 8), respectively, based on their characteristic coupling patterns and
the 1H and 13C chemical shifts of the corresponding allylic NMR signals.
[36] The preference of the exo-isomer over the endo can mainly be
attributed to the more pronounced repulsion between the face posi-
tioned equatorial P–Ph group and the allylic Ph substituent in the latter
(Fig. 8). [36] The structural analysis of the diphenylallyl-complex by 1H
and 1H-1H NOESY techniques evidenced that the chelate ring adopts the
same conformation both in the exo- and endo-isomers as in [Pd(L2b)Cl2].
This can for instance be deduced from the very similar coupling patterns
and chemical shifts of the diastereotopic CH2 protons in the backbone.
To our great surprise, however, upon cooling the solution to 193 K the
31P signal of the exo-isomer splits, indicating a fast equilibrium between
two isomers that are present in a molar ratio of 1.6:7. This process can be
attributed to the inversion of the coordinated sulfur atom as the typical
coupling patterns and chemical shifts of the backbone hydrogens and the
allylic protons are very similar in the two species thus excluding the
possibility of changes in the ring conformation or in the allylic moiety.
Consequently, the NMR data suggest that the exo-form of [Pd(L2b)
In order to strengthen this concept, we have synthesized phosphine-
aminophosphine ligand L3, analogously to compounds L2a-e, using the
corresponding aminophosphine (3) [36] as starting material (Fig. 9). L3
can be considered as a structural analogue of L2b as they differ only in
one of their coordinating functionalities: L3 contains a sterically more
demanding PPh2 function instead of the SPh moiety.
In contrast to their structural similarities, however, L3 exhibits quite
different coordination behavior in its Pd-diphenylallyl complex
compared to L2b. In the 31P NMR spectrum of [Pd(L3)(
η
3-Ph2-allyl)]BF4
two pairs of doublets appear indicating the presence of two di-
astereomers in a molar ratio of 10:2, that are identified as the corre-
sponding exo and endo isomers, respectively. The chelate ring in both the
exo- and endo-isomers of [Pd(L3)(
η
3-Ph2-allyl)]BF4 adopts the same
conformation as was found in [Pd(L2b)(
η
3-Ph2-allyl)]BF4. Conse-
quently, both donor phosphorus atoms must have an axially and an
equatorially disposed Ph group, as depicted in Fig. 8. Unlike in the so-
(
η
3-Ph2-allyl)]+ dominates in the solution phase that exhibits fluxional
lution of [Pd(L2b)(
η
3-Ph2-allyl)]BF4, however, no dynamic behavior
behavior due to its stereolabile sulfur atom. It is, however, interesting to
note that the above dynamic behavior of chiral ligand L2b cannot be
observed in [Pd(L2b)Cl2]. It demonstrates the fact that the co-ligand of
the complex (i.e. Cl vs. diphenylallyl) has a remarkable effect on the
stereoselectivity of sulfur coordination and on its dynamics as well.
Additionally, the observed configurational lability in the exo-isomer
point out that the sulfur has a sterically less hindered environment and
could be observed: upon cooling the sample to 193 K the spectrum still
exhibited the same signal sets. This also confirms the fact that the so-
lution phase dynamics of complex [Pd(L2b)(
η
3-Ph2-allyl)]BF4 can be
attributed to the sulfur inversion. The catalytic features of the two
diphenylallyl-complexes, [Pd(L2b)(
η
3-Ph2-allyl)]BF4 and [Pd(L3)(η3
-
Ph2-allyl)]BF4 are also remarkably different. First, in the allylic ether-
ification using allyl alcohol as nucleophile the thioether-containing
Fig. 8. The exo (major)- and endo (minor)- isomers of cations [Pd(L2b)(η3-Ph2-allyl)]+ (top) and [Pd(L3)(η3-Ph2-allyl)]+ (bottom).
7

M.M. Major et al.
Molecular Catalysis 512 (2021) 111763
Fig. 9. Synthesis of chiral phosphine-aminophosphine ligand L3.
Declaration of Competing Interest
catalyst provided the product with 93% yield after 24 h at a substrate/
catalyst molar ratio of 500. In contrast, the catalytic system containing
L3 as chiral selector resulted only 30% isolated yield under the same
conditions. In other words, the stereochemically rather labile catalyst
with a sterically less hindered sulfur environment outperformed the
sterically more congested and rigid system in terms of activity. Addi-
tionally, the enantioselectivity achieved by the latter (20% ee) is
considerably lower than that provided by L2b (84% ee). The higher
enantioselectivity achieved by using the thioether containing catalyst
can mainly be attributed to the remarkably different electronic nature of
its donoratoms. Assuming that the exo-isomer is more reactive than the
endo, and that the nucleophilic attack on the diphenylallyl-moiety takes
place at the allyl carbon trans to the phosphorus and opposite to the Pd
(Fig. 8), [16,39] allylic ethers with (S) configuration will be produced.
Indeed, the prevailing configuration of the optically active allylic
products is (S) when using L2b-e as chiral ligands.
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.
Acknowledgement
´
´
We thank Mr Bela Edes for skillful assistance in analytical mea-
surements and synthetic experiments. The scientific program was sup-
ported by the project NKFIH K128074. The research was also supported
by the ÚNKP-18-3 National Excellence Program of the Ministry of
Human Capacities.
Supplementary materials
Obviously, the stereochemical lability of the Pd(PN,S) complexes
may also be responsible for the imperfect enantioselectivity observed in
the catalytic reactions. [23] It is, however, reasonable to assume that it
is the balanced flexibility of these Pd(PN,S) catalysts, i.e. the expressed
rigidity of the chelate together with the rather non-rigid nature of the
chiral pocket, that enables the catalytic system to provide very high
activities and relatively high enantioselectivities at the same time.
Supplementary material associated with this article can be found, in
the online version, at doi:10.1016/j.mcat.2021.111763.
References
[1] B.M. Trost, T. Zhang, J.D. Sieber, Catalytic asymmetric allylic alkylation employing
heteroatom nucleophiles: a powerful method for C–X bond formation, Chem. Sci. 1
(2010) 427–440.
[2] B.M. Trost, Pd- and Mo-catalyzed asymmetric allylic alkylation, Org. Process Res.
Dev. 16 (2012) 185–194.
Conclusions
[3] R.L. Grange, E.A. Clizbe, P.A. Evans, Recent developments in asymmetric allylic
amination reactions, Synthesis 48 (2016) 2911–2968.
[4] L. Mohammadkhani, M.M. Heravi, Applications of transition-metal-catalyzed
asymmetric allylic substitution in total synthesis of natural products: an update,
Chem. Rec. 21 (2021) 29–68.
In summary, six novel, optically active alkane-diyl based thioether-
aminophosphines L1 and L2a-e with modular structure have been syn-
thesized in a simple step starting from the corresponding thioether-
amines. Based on the new ligand family, a cost-effective, safe and scal-
able synthetic protocol has been developed for the Pd-catalyzed asym-
metric allylic etherification using catalyst loadings as low as 0.2 mol%.
In addition, the substrate/base/nucleophile ratio was significantly
improved to 1/1.2/1.2 from the ratio of 1/3/3 used generally in liter-
ature procedures, while the amount of the solvent could be kept at a
minimum. The optically active allylic ethers could be isolated in high
yields (up to 95%) with good enantioselectivities (up to 86% ee).
Furthermore, combined spectroscopic and X-ray studies, and compara-
tive experiments were performed in order to shed light on the success of
this small ligand family in asymmetric allylic etherification reactions.
The (i) stereoelectronic heterogeneity of the donor atoms, (ii) the ste-
rically less hindered environment around the thioether moiety and (iii)
the carefully balanced stereochemical lability of the catalyst have been
identified as the major factors exerting a profound effect on the activity
and the enantioselectivity. We strongly believe that our findings will
promote the use of chiral PN,S type ligands in asymmetric allylic
etherification reactions as an efficient tool for the synthesis of optically
active chiral building blocks even on an industrial scale.
`
˜
[5] O. Pamies, J. Margalef, S. Canellas, J. James, E. Judge, P.J. Guiry, C. Moberg, J.-
¨
`
´
E. Backvall, A. Pfaltz, M.A. Pericas, M. Dieguez, Recent advances in
enantioselective Pd-catalyzed allylic substitution: from design to applications,
Chem Rev. 121 (2021) 4373–4505.
[6] L. Milhau, P.J. Guiry, Palladium-catalyzed enantioselective allylic substitution, in:
U. Kazmeier (Ed.), Transition Metal Catalyzed Enantioselective Allylic Substitution
in Organic Synthesis, Springer-Verlag, Berlin Heidelberg, 2012, pp. 95–153.
[7] S. Noreen, A.F. Zahoor, S. Ahmad, I. Shahzadi, A. Irfan, S. Faiz, Novel chiral ligands
for palladium-catalyzed asymmetric allylic alkylation/asymmetric Tsuji-Trost
reaction: a review, Curr. Org. Chem. 23 (2019) 1168–1213.
´
[8] M. Biosca, J. Margalef, X. Caldentey, M. Besora, C. Rodriguez-Escrich, J. Salto, X.
`
´
`
C. Cambeiro, F. Maseras, O. Pamies, M. Dieguez, M.A. Pericas, Computationally
guided design of a readily assembled phosphite-thioether ligand for a broad range
of Pd-catalyzed asymmetric allylic substitutions, ACS Catal. 8 (2018) 3587–3601.
[9] Y. Zang, I. Ojima, Pd-catalyzed asymmetric allylic etherification using chiral
biphenol-based diphosphinite ligands and its application for the formal total
synthesis of (ꢀ )-galanthamine, J. Org. Chem. 78 (2013) 4013–4018.
[10] A. Khan, S. Khan, I. Khan, C. Zhao, Y. Mao, Y. Chen, Y.J. Zhang, Enantioselective
construction of tertiary C–O bond via allylic substitution of vinylethylene
carbonates with water and alcohols, J. Am. Chem. Soc. 139 (2017) 10733–10741.
[11] J.-X. Xu, F. Ye, X.-F. Bai, J. Zhang, Z. Xu, Z.-J. Zheng, L.-W. Xu, Fei-Phos ligand-
controlled asymmetric palladium catalyzed allylic substitutions with structurally
diverse nucleophiles: scope and limitations, RSC Adv. 6 (2016) 45495–45502.
[12] Z. Qiu, R. Sun, K. Yang, D. Teng, Spiro indane-based phosphine-oxazolines as
highly efficient P,N ligands for enantioselective Pd-catalyzed allylic alkylation of
indoles and allylic etherification, Molecules 24 (2019) 1575–1591.
`
`
´
[13] J. Margalef, C. Borras, S. Alegre, O. Pamies, M. Dieguez, A readily accessible and
modular carbohydrate-derived thioether/selenoether-phosphite ligand library for
Pd-catalyzed asymmetric allylic substitutions, Dalton Trans. 48 (2019)
12632–12643.
CRediT authorship contribution statement
[14] Q.-L. Liu, W. Chen, Q.-Y. Jiang, X.-F. Bai, Z. Li, Z. Xu, L.-W. Xu, A d-camphor-based
Schiff base as a highly efficient N,P ligand for enantioselective palladium-catalyzed
allylic substitutions, ChemCatChem 8 (2016) 1495–1499.
´ ´
´
´
Mate M. Major: Methodology, Investigation. Maria Guoth: Inves-
´
tigation. Szabolcs Balogh: Investigation, Validation. Jozsef Simon:
´
[15] S. Li, J. Zhang, H. Li, L. Feng, P. Jiao, Preparation and application of amino
phosphine ligands bearing spiro[indane-1,2′-pyrrolidine] backbone, J. Org. Chem.
84 (2019) 9460–9473.
Investigation, Validation. Attila C. Benyei: Investigation, Validation,
´
Conceptualization. Jozsef Bakos: Writing – review & editing, Supervi-
sion, Conceptualization. Gergely Farkas: Writing – original draft, Su-
pervision, Conceptualization.
8

M.M. Major et al.
Molecular Catalysis 512 (2021) 111763
[16] B. Feng, H.-G. Cheng, J.-R. Chen, Q.-H. Deng, L.-Q. Lu, W.-J. Xiao, Palladium/
sulfoxide–phosphine-catalyzed highly enantioselective allylic etherification and
amination, Chem. Commun. 50 (2014) 9550–9553.
[28] F.L. Lam, T.T.-L. Au-Yeung, F.Y. Kwong, Z. Zhou, K.Y. Wong, A.S.C. Chan,
Palladium-(S,pR)-FerroNPS-catalyzed asymmetric allylic etherification: electronic
effect of nonconjugated substituents on benzylic alcohols on enantioselectivity,
Angew. Chem. Int. Ed. 47 (2008) 1280–1283.
`
´
[17] M. Coll, O. Pamies, M. Dieguez, Highly versatile Pd-thioether-phosphite catalytic
systems for asymmetric allylic alkylation, amination, and etherification reactions,
Org. Lett. 16 (2014) 1892–1895.
´
´
´
[29] M.M. Major, Z. Csaszar, A.C. Benyei, S. Balogh, J. Bakos, G. Farkas, Backbone
effects in the synthesis, coordination chemistry and catalytic properties of new
chiral heterobidentate ligands with P,N and S,N donor sets, J. Organomet. Chem.
921 (2020), 121332.
`
´
[18] J. Margalef, M. Coll, P.-O. Norrby, O. Pamies, M. Dieguez, Asymmetric catalyzed
allylic substitution using a Pd/Pꢀ S catalyst library with exceptional high substrate
¨
´
and nucleophile versatility: DFT and Pd-
π
-allyl key intermediates studies,
[30] M.M. Major, G. Rovid, S. Balogh, A.C. Benyei, G. Lendvay, D. Frigyes, J. Bakos,
G. Farkas, Double stereoselective coordination of chiral N,S ligands: synthesis,
coordination chemistry and utilization in Pd-catalyzed allylic alkylation, Appl.
Organomet. Chem. 33 (2019) e4726.
Organometallics 35 (2016) 3323–3335.
[19] J.-X. Xu, F. Ye, X.-F. Bai, Y.M. Cui, Z. Xu, Z.-J. Zheng, L.-W. Xu, A mechanistic
study on multifunctional Fei-Phos ligand-controlled asymmetric palladium-
catalyzed allylic substitutions, RSC Adv. 6 (2016) 70624–70631.
[20] K. Yamamoto, T. Shimizu, K. Igawa, K. Tomooka, G. Hirai, H. Suemune1, K. Usui,
Rational design and synthesis of [5]helicene-derived phosphine ligands and their
application in Pd-catalyzed asymmetric reactions, Sci. Rep. 6 (2016) 36211. http
://www.platinum.matthey.com/prices/price-charts.
´
[31] J. Xie, W. Guo, A. Cai, E.C. Escudero-Adan, A.W. Kleij, Pd-catalyzed enantio- and
regioselective formation of allylic aryl ethers, Org. Lett. 19 (2017) 6388–6391.
[32] J. Xing, P. Cao, J. Liao, Chiral SO/P hybrid ligands: an enantioselective switch in
palladium-catalyzed asymmetric allylic etherifications, Tetrahedron 23 (2012)
527–535.
[21] F.L. Lam, F.Y. Kwong, A.S.C. Chan, Recent developments on chiral P,S-type ligands
and their applications in asymmetric catalysis, Chem. Commun. 46 (2010)
4649–4667.
[33] P. Fang, C.-H. Ding, X.-L. Hou, L.-X. Dai, Palladium-catalyzed regio- and enantio-
selective allylic substitution reaction of monosubstituted allyl substrates with
benzyl alcohols, Tetrahdron 21 (2010) 1176–1178.
`
`
´
[22] J. Margalef, O. Pamies, M.A. Pericas, M. Dieguez, Evolution of
phosphorus–thioether ligands for asymmetric catalysis, Chem. Commun. 56 (2020)
10795–10808.
[34] E.J. Zijp, J.I. van der Vlugt, D.M. Tooke, A.L. Spek, D. Vogt, Chiral bidentate
aminophosphine ligands: synthesis, coordination chemistry and asymmetric
catalysis, Dalton Trans. (2005) 512–517.
´
[23] D.A. Evans, K.R. Campos, J.S. Tedrow, F.E. Michael, M.R. Gagne, Application of
[35] S. Balogh, G. Farkas, T. Holczbauer, J. Bakos, Origin of the chemical stability of
phosphine-phosphoramidites: structural study of an UPPhos-type crystal and
application of UPPhos in the asymmetric hydrogenation of imines, Acta Cryst. C27
(2017) 632–637.
chiral mixed phosphorus/sulfur ligands to palladium-catalyzed allylic
substitutions, J. Am. Chem. Soc. 122 (2000) 7905–7920.
[24] D.A. Evans, F.E. Michael, J.S. Tedrow, K.R. Campos, Application of chiral mixed
phosphorus/sulfur ligands to enantioselective rhodium-catalyzed dehydroamino
acid hydrogenation and ketone hydrosilylation processes, J. Am. Chem. Soc. 125
(2003) 3534–3543.
´
´
´
´
[36] Z. Csaszar, G. Farkas, A.C. Benyei, G. Lendvay, I. Toth, J. Bakos, Stereoselective
coordination: a six-membered P,N-chelate tailored for asymmetric allylic
alkylation, Dalton Trans. 44 (2015) 16352–16360.
[25] B. Feng, X.-Y. Pu, Z.-C. Liu, W.-J. Xiao, J.-R. Chen, Highly enantioselective Pd-
catalyzed indole allylic alkylation using binaphthyl-based phosphoramidite-
thioether ligands, Org. Chem. Front. 3 (2016) 1246–1249.
[37] J.M. Crawford, M.S. Sigman, Conformational dynamics in asymmetric catalysis: Is
catalyst flexibility a design element? Synthesis 51 (2019) 1021–1036.
[38] J.A. Raskatov, A.L. Thompson, J.M. Brown, Asymmetric catalysis with 7-ring
chelate diphosphines: DIOP, BINAP and conformational mobility, Tetrahedron 21
(2010) 1737–1744.
[26] Y. Wei, L.-Q Lu, T.-R. Li, B. Feng, Q. Wang, W.-J. Xiao, H. Alper, P,S ligands for the
asymmetric construction of quaternary stereocenters in palladium-catalyzed
decarboxylative [4+2] cycloadditions, Angew. Chem. 128 (2016) 2240–2244.
[27] J.-L. Zeng, S.-B. Yu, Z. Cao, D.-W. Yang, Preparation of S-containing
aminophosphine and phosphoramidite ligands and their applications in
enantioselective C–C bond forming reaction, Catal. Lett. 136 (2010) 243–248.
[39] The 13C NMR spctrum of the complex revealed that the terminal allylic carbon
trans to P is much more electrophilic that that trans to S. X-ray structure analysis of
the correpsonding dichloro complex proved that the Pd-Cl bond length trans to P is
considerably longer than the one trans to S. These experimental findings confirm
that allylic carbon trans to P is more susceptible to be attacked by the nucleophile.
9