Insights into functional-group-tolerant polymerization catalysis with phosphine-sulfonamide palladium (II) complexes

Article abstract of DOI:10.1002/chem.201404856

Two series of cationic palladium(II) methyl complexes {[(2-MeOC₆H₄)₂PC₆H₄SO₂NHC₆H₃(2,6-R¹,R²)]PdMe}₂[A]₂ (^X1⁺-A: R¹=R²=H: ^H1⁺-A; R¹ = R² = CH(CH₃)₂: ^DIPP1⁺-A; R¹ = H, R² = CF₃: ^CF31⁺-A; A = BF₄ or SbF₆) and neutral palladium(II) methyl complexes {[(2-MeOC₆H₄)₂PC₆H₄SO₂NC₆H₃(2,6-R¹,R²)]PdMe(L)} (^X1-acetone: L = acetone; ^X1-dmso: L = dimethyl sulfoxide; ^X1-pyr: L=pyridine) chelated by a phosphine-sulfonamide were synthesized and fully characterized. Stoichiometric insertion of methyl acrylate (MA) into all complexes revealed that a 2,1 regiochemistry dominates in the first insertion of MA. Subsequently, for the cationic complexes ^X1⁺-A, β-H elimination from the 2,1-insertion product ^X2⁺-A_MA-2,1 is overwhelmingly favored over a second MA insertion to yield two major products ^X4⁺-A_MA-1,2 and ^X5⁺-A_MA. By contrast, for the weakly coordinated neutral complexes ^X1-acetone and ^X1-dmso, a second MA insertion of the 2,1-insertion product ^X2_MA-2,1 is faster than β-H elimination and gives ^X3_MA as major products. For the strongly coordinated neutral complexes ^X1-pyr, no second MA insertion and no β-H elimination (except for ^DIPP2-pyr_MA-2,1) were observed for the 2,1-insertion product ^X2-pyr_MA-2,1. The cationic complexes ^X1⁺-A exhibited high catalytic activities for ethylene dimerization, affording butenes (C₄) with a high selectivity of up to 97.7% (1-butene: 99.3%). Differences in activities and selectivities suggest that the phosphine-sulfonamide ligands remain coordinated to the metal center in a bidentate fashion in the catalytically active species. By comparison, the neutral complexes ^X1-acetone, ^X1-dmso, and ^X1-pyr showed very low activity towards ethylene to give traces of oligomers. DFT analyses taking into account the two possible coordination modes (O or N) of the sulfonamide ligand for the cationic system ^CF31⁺ suggested that the experimentally observed high activity in ethylene dimerization is the result of a facile first ethylene insertion into the O-coordinated PdMe isomer and a subsequent favored β-H elimination from the N-coordinated isomer formed by isomerization of the insertion product. Steric hindrance by the N-aryl substituent in the neutral systems ^CF31 and ^H1 appears to contribute significantly to a higher barrier of insertion, which accounts for the experimentally observed low activity towards ethylene oligomerization.

Full text of DOI:10.1002/chem.201404856

DOI: 10.1002/chem.201404856
Full Paper
&
Palladium Complexes
Insights into Functional-Group-Tolerant Polymerization Catalysis
with Phosphine–Sulfonamide Palladium(II) Complexes
[
a]
[b]
[a]
[a]
[c]
Zhongbao Jian, Laura Falivene, Philipp Wucher, Philipp Roesle, Lucia Caporaso,*
[b]
[a]
[a]
Luigi Cavallo, Inigo Gçttker-Schnetmann, and Stefan Mecking*
X
+
Abstract: Two series of cationic palladium(II) methyl com-
1 -A exhibited high catalytic activities for ethylene dimeri-
1
2
plexes {[(2-MeOC H ) PC H SO NHC H (2,6-R ,R )]PdMe} [A]
zation, affording butenes (C ) with a high selectivity of up to
6
4 2
+
6
4
1
2
6
3
2
2
4
X
+
1
2
H
2
DIPP
+
1
(
1 -A: R =R =H: 1 -A; R =R =CH(CH ) :
1 -A; R =
97.7% (1-butene: 99.3%). Differences in activities and selec-
tivities suggest that the phosphine–sulfonamide ligands
remain coordinated to the metal center in a bidentate fash-
ion in the catalytically active species. By comparison, the
3
2
2
CF3
+
H, R =CF :
1 -A; A=BF4 or SbF ) and neutral palladi-
6
3
um(II) methyl complexes {[(2-MeOC H ) PC H SO NC H (2,6-
6
4 2
6
4
2
6
3
1
2
X
X
R ,R )]PdMe(L)} ( 1-acetone: L=acetone; 1-dmso: L=di-
X
X
X
X
methyl sulfoxide; 1-pyr: L=pyridine) chelated by a phos-
neutral complexes 1-acetone, 1-dmso, and 1-pyr showed
very low activity towards ethylene to give traces of oligo-
mers. DFT analyses taking into account the two possible co-
ordination modes (O or N) of the sulfonamide ligand for the
phine–sulfonamide were synthesized and fully characterized.
Stoichiometric insertion of methyl acrylate (MA) into all com-
plexes revealed that a 2,1 regiochemistry dominates in the
first insertion of MA. Subsequently, for the cationic com-
CF3
+
cationic system
1 suggested that the experimentally ob-
X
+
plexes 1 -A, b-H elimination from the 2,1-insertion product
served high activity in ethylene dimerization is the result of
a facile first ethylene insertion into the O-coordinated PdMe
isomer and a subsequent favored b-H elimination from the
N-coordinated isomer formed by isomerization of the inser-
tion product. Steric hindrance by the N-aryl substituent in
X
+
2
-AMA-2,1 is overwhelmingly favored over a second MA in-
X
+
X +
sertion to yield two major products 4 -A
and 5 -AMA.
MA-1,2
By contrast, for the weakly coordinated neutral complexes
X
X
1
-acetone and 1-dmso, a second MA insertion of the 2,1-
X
CF3
H
insertion product 2
gives 3 as major products. For the strongly coordinated
neutral complexes 1-pyr, no second MA insertion and no b-
H elimination (except for
is faster than b-H elimination and
the neutral systems 1 and 1 appears to contribute signifi-
cantly to a higher barrier of insertion, which accounts for
the experimentally observed low activity towards ethylene
oligomerization.
MA-2,1
X
MA
X
DIPP
2-pyrMA-2,1) were observed for
X
the 2,1-insertion product 2-pyrMA-2,1. The cationic complexes
Introduction
vinyl monomers. Since the finding by Brookhart and co-work-
II
ers that cationic Pd diimine catalysts can catalyze the insertion
[6]
Late-transition-metal catalysts have been intensively studied
for olefin oligomerization and copolymerization, due to their
different incorporation behavior, branch formation, and gener-
copolymerization of ethylene or 1-olefins and acrylates, many
II
studies on insertion polymerization have focused on Pd cata-
lysts chelated by various ligands. Another significant break-
through was reported by Drent and co-workers, who revealed
that a neutral phosphinesulfonato palladium species could
produce highly linear copolymers of ethylene and methyl acryl-
[
1–5]
ally less oxophilic nature.
Among them, palladium catalysts
are of advantage for the insertion copolymerization of polar
[7]
II
ate (MA). In the past decade, this phosphinesulfonato Pd cat-
alyst has been found to promote the formation of linear co-
polymers of ethylene with a broad scope of polar monomers
including not only acrylates, but also acrylonitrile, vinyl acetate,
vinyl ether, vinyl fluoride, vinyl chloride, allylic monomers,
[
a] Dr. Z. Jian, P. Wucher, P. Roesle, Dr. I. Gçttker-Schnetmann,
Prof. Dr. S. Mecking
Chair of Chemical Materials Science, Department of Chemistry
University of Konstanz, 78464 Konstanz (Germany)
E-mail: Stefan.Mecking@uni-konstanz.de
[b] Dr. L. Falivene, Prof. Dr. L. Cavallo
[8,9]
acrylic acid, and acrylamides. An important structural feature
of catalysts containing phosphine–sulfonates (A), which ren-
ders them capable of insertion chain growth, is the presence
Physical Sciences and Engineering, Kaust Catalysis Center
King Abdullah University of Science and Technology (KAUST)
Thuwal 23955-6900 (Saudi Arabia)
[
c] Dr. L. Caporaso
of one hard sulfonate oxygen and one soft phosphine s-donor
Department of Chemistry and Biology, University of Salerno
Via Giovanni Paolo II, 132-84084, Fisciano (SA) (Italy)
E-mail: lcaporaso@unisa.it
[9h]
ligand.
The phosphinesulfonato catalyst system was long
considered to be unique in ethylene/polar vinyl monomer co-
polymerization to give linear copolymer. Recently, Nozaki and
co-workers reported a series of cationic phosphine/phosphine
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201404856.
Chem. Eur. J. 2015, 21, 2062 – 2075
2062
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
monoxide complexes B that catalyze linear copolymerization
SO NHR (e.g., pK (PhSO NH )=10.1), we did not observe any
2
a
2
2
[
10]
1
of ethylene with a range of polar vinyl monomers. This im-
portant finding shows that the phosphinesulfonato motif is
not unique. Beyond the above insights, an understanding of
the criteria for a ligand environment that promotes insertion
chain growth and copolymerization is lacking to date. To shed
light on this case, we chose the phosphine–sulfonamide motif
similar signals in the H NMR spectra of La–c, but the signals
found around 8.14, 7.35, and 8.35 ppm, respectively, could be
assigned to the NH proton. Moreover, this is also exemplified
by the solid-state structure of the phosphine–sulfonamide Lb,
in which H33 is located at the nitrogen atom with a bond
length of about 0.82 ꢁ (d(P1ꢀH33)ꢂ2.70 ꢁ, Figure 1).
(
C and D) for a combined experimental and theoretical
[
11]
study. This allows for the formation of cationic and neutral
palladium(II) complexes with one ligand system coordinating
through different hard donors (N/O), and at the same time the
introduction of steric bulk at the hard donor by means of N-
aryl moieties.
Results and Discussion
Synthesis of phosphine–sulfonamide ligands and cationic
palladium(II) complexes
Figure 1. ORTEP of ligand Lb drawn with 50% probability ellipsoids. Hydro-
gen atoms (except NH) are omitted for clarity.
Phosphine–sulfonamides La–c were synthesized by modified
[
7]
literature procedures. Treatment of sulfonamides with two
equivalents of nBuLi at room temperature for 2 h, followed by
in situ reaction with (2-MeOC H ) PCl at 508C for a further
6
4 2
12 h, afforded the lithium phosphine–sulfonamide salts, which
Phosphine–sulfonamides La–c react readily with one equiva-
were protonated to give La–c as colorless or pale yellow crys-
tals in 47–63% yield (Scheme 1). Ligands La–c were fully char-
lent of [PdMeCl(cod)] at room temperature to straightforwardly
II
X
generate the desired neutral Pd complexes 1-Cl, which were
1
13
31
acterized by multinuclear NMR spectroscopy ( H, C, and P),
X-ray diffraction, and elemental analysis. Generally, as a conse-
quence of the pK values of ArSO H (e.g., pK (PhSO H)=ꢀ2.8)
isolated in satisfactory yields of 62–69% (Scheme 1). The
1
X
H NMR spectra of 1-Cl suggest that the NH group in the
ligand is not deprotonated during the reaction, but 1,5-cyclo-
octadiene (cod) is replaced by the intact neutral ligand. The
a
3
a
3
+
[12]
and Ar P H (pK ꢁ2.7),
the phosphine–sulfonic acids are
3
a
X
usually zwitterionic compounds in which the acidic proton is
NH protons in 1-Cl are unambiguously observed at d=8.50,
located at the phosphorus atom, as evidenced by both the
8.68, and 8.38 ppm, respectively. In addition, doublets at d=
1
coupling constant ( J ꢂ600 Hz) and the chemical shift (d
0.57, 0.96, and 0.76 ppm arising from coupling to phosphorus
PH
[
7]
3
ꢂ9.0 ppm). As expected from the the large pK value of Ar-
( J ) can be assigned to the PdMe group (Table 1). On the
a
PH
basis of NMR spectroscopic data,
one cannot differentiate unam-
2
biguously between a k -P,O and
2
a k -P,N bidentate coordination
X
mode in 1-Cl, and hence the
solid-state structure was deter-
mined by X-ray diffraction. Suit-
DIPP
able crystals of
1-Cl were
grown by layering a CH Cl solu-
2
2
tion of the complex with pen-
tane. X-ray diffraction analysis of
DIPP
1-Cl revealed that the for-
mally neutral phosphine–sulfon-
amide ligand coordinates to the
2
Pd center in a k -P,O fashion,
and H34 is located at the nitro-
II
X
+
Scheme 1. Synthesis of phosphine–sulfonamide ligands La–c and cationic Pd complexes 1 -A.
gen atom (Figure 2). The Pd1ꢀP1
Chem. Eur. J. 2015, 21, 2062 – 2075 www.chemeurj.org
2063
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
and 0.76 ppm (Table 1). Because the reaction took place in
Table 1. Selected chemical shifts [ppm] of palladium complexes.
X
+
X +
a noncoordinating solvent, we inferred that 1 -BF and 1
4
1
13
31
H (PdMe)
C (PdMe)
P
-SbF have dimeric structures in which the fourth coordination
6
II
H
DIPP
CF3
H
site at the Pd center is occupied by a sulfonamide group of
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
-Cl
-Cl
0.57 (d)
0.96 (d)
0.76 (d)
0.65 (s)
0.67 (s)
0.71 (s)
0.55 (s)
0.76 (s)
0.73 (s)
ꢀ0.20 (s)
ꢀ0.12 (s)
ꢀ0.17 (s)
ꢀ0.03 (s)
ꢀ0.04 (d)
ꢀ0.20 (d)
ꢀ0.18 (d)
–
23.2
20.8
21.9
28.6
26.6
24.0
24.5
25.3
23.7
28.3
29.4
28.3
26.7
23.5
24.1
24.4
_
1.62 (s)
2.44 (s)
8.24 (d)
7.96 (s)
6.62 (s)
6.36 (s)
8.41 (s)
7.54 (s)
1.59 (s)
2.02 (d)
1.61 (s)
2.52 (s)
0.99 (d)
2.25 (d)
0.17 (d)
the second [(PSO NHAr)PdMe] fragment. Analogous dimers
2
II
-Cl
have been reported for different phosphinesulfonato Pd com-
+
-BF
4
[9a,13]
plexes.
H
+
+
+
+
+
-SbF
6
6
6
DIPP
DIPP
CF3
CF3
DIPP
DIPP
CF3
CF3
H
-BF
-SbF
-BF
-SbF
4
4
Decomposition of cationic palladium(II) methyl complexes
II
X
+
-acetone
-dmso
-acetone
-dmso
-pyr
To confirm the structures of cationic Pd methyl complexes 1
A, we attempted to isolate crystals suitable for X-ray diffrac-
tion analysis. For example, by layering a CD Cl solution of
-
2
2
DIPP
+
1 -SbF with pentane (Scheme 2), red crystals suitable for
6
DIPP
CF3
-pyr
-pyr
I
DIPP
+
Scheme 2. Formation of binuclear cationic Pd complex
6 decomp.
(1 -SbF )
X-ray diffraction were obtained within 4 d. X-ray diffraction
analysis revealed that these crystals were not the target com-
DIPP
+
DIPP +
(1 -
plex
1 -SbF , but the decomposition product
6
SbF6)decomp.. As reported previously, the decomposition of
II
a phosphinesulfonato Pd methyl complex usually generates
0
II
[14]
DIPP
+
Pd black and Pd L2 or free ligand.
However,
(1 -
II
SbF6)decomp., formed by the decomposition of cationic Pd
methyl complex
DIPP
+
I
1 -SbF , is a binuclear cationic Pd complex
6
DIPP
I
I
Figure 2. ORTEP of complex
1-Cl drawn with 50% probability ellipsoids.
with a PdꢀPd bond (Figure 3). A protonated NH moiety was
Hydrogen atoms (except NH) are omitted for clarity. Selected bond lengths
ꢁ]: Pd1ꢀO3 2.249(2), Pd1ꢀP1 2.216(6), Pd1ꢀC33 2.005(3), Pd1ꢀCl1 2.362(6).
[
distance (2.216(6) ꢁ) is slightly shorter than in the ana-
II
logous phosphinesulfonato Pd complex [{(o-(o-MeOC H ) P)-
6
4 2
[
9l]
C H SO }PdMe(lutidine)] (2.234(9) ꢁ), whereas the Pd1ꢀO3
6
4
3
distance (2.249(2) ꢁ) is significantly longer than the PdꢀO dis-
tance (2.159(2) ꢁ).
[
9l]
In addition, the Pd1ꢀC33 distance
(
2.005(3) ꢁ) is within the range of typical PdꢀC bonds. To study
X
the strength of the PdꢀO bond in 1-Cl, we added an excess
X
of a coordinating solvent such as acetone or DMSO to 1-Cl.
1
However, H NMR signals remained unaltered, that is, the che-
X
lating PdꢀO coordination in 1-Cl is stronger than the binding
strength between Pd and acetone or DMSO.
X
The palladium complexes 1-Cl further reacted with silver(I)
salts (AgBF or AgSbF ) in CH Cl to generate the cationic palla-
4
6
2
2
X
+
dium(II) methyl complexes 1 -A (A=BF and SbF ) in good
4
6
X
+
X +
yields (Scheme 1). Complexes 1 -BF and 1 -SbF were fully
4
6
DIPP
+
1
13
19
31
Figure 3. ORTEP of complex
ellipsoids. Hydrogen atoms, cocrystallized solvent molecules, and SbF
are omitted for clarity. Selected bond lengths [ꢁ]: Pd1ꢀO1 2.157(5), Pd1ꢀP1
2.198(2), Pd1ꢀP1 2.758(2), Pd1ꢀC14 2.326(7), Pd1ꢀPd1 2.544(1).
(1 -SbF
6
)
decomp. drawn with 50% probability
identified by 1D ( H, C, F, and P) and 2D NMR spectroscopy
and elemental analysis. The methyl protons of the PdMe group
in all cationic complexes were clearly observed between 0.55
6
anion
i
i
Chem. Eur. J. 2015, 21, 2062 – 2075
www.chemeurj.org
2064
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
found in the electron density map by X-ray diffraction. An anal-
II
ogous reaction pathway has been observed in Pd zwitterion
I
I
[
(Ph B(PPh ) )PdMe(thf)], which can also generate red PdꢀPd
2
2 2
[
15]
dimer [{(Ph B(PPh ) )Pd} ] on addition of H .
The neutral
2
2 2
2
2
phosphine–sulfonamide ligand coordinates to one Pd center in
2
a k -P,O bidentate mode, and to another Pd center in a weak
2
i
[15]
h -P,C fashion (Pd1ꢀP1 2.758(2), Pd1ꢀC14=2.326(7) ꢁ). The
I
I
DIPP
+
PdꢀPd distance of 2.544(1) ꢁ in
(1 -SbF )
is close to
decomp.
6
[
16a]
that in [Pd (NPN-NPN) ](BF ) (2.5489(7) ꢁ),
but slightly
2
2
4 2
I
shorter than those in [Pd (m-2-C F PPh ) (L) ] (L=PPh ,
2
6
4
2 2
2
3
[16b]
2
.5740(3) ꢁ; L=AsPh , 2.5511(3) ꢁ; L=tBuNC, 2.5803(4) ꢁ)
3
and much shorter than that in [{(Ph B(PPh ) )Pd} ]
2
2 2
2
[
15]
(2.728(6) ꢁ).
Synthesis of neutral palladium(II) methyl complexes
By addition of NaH to phosphine–sulfonamides La–c, the
sodium salts La–c-Na were readily obtained in high yields.
Lb,c-Na reacted with [PdMeCl(cod)] in acetone and CH Cl to
2
2
II
X
readily afford the neutral Pd methyl complexes 1-acetone in
II
good yields (Scheme 3). In comparison with the cationic Pd
X
+
complexes 1 -A, the signals of the PdMe group in the
1
X
H NMR spectra of 1-acetone are shifted upfield to ꢀ0.20 and
ꢀ
0.17 ppm, respectively (Table 1). In addition, the coordinated
acetone molecule gives rise to a strong singlet at d=1.97 and
.20 ppm, respectively. As reported previously, the same reac-
2
_
tion of sodium salt of phosphine–sulfonate [(PO)Na] with
_
[
PdMeCl(COD)] always yielded the dimer [({(PO)Pd(Me)Cl}m-
Na(acetone) ) ], in which the Cl atom is bound to the Pd
2
2
[
17]
center. Strongly coordinating solvents such as pyridine or
silver salts are necessary to abstract the Cl atom. However, X-
ray studies unambiguously revealed that acetone directly
binds to the palladium center with a reasonable PdꢀO distance
DIPP
CF3
Figure 4. ORTEPs of neutral complexes
1-acetone and 1-dmso drawn
with 50% probability ellipsoids. Hydrogen atoms and cocrystallized solvent
molecules are omitted for clarity.
X
in 1-acetone, and the phosphine–sulfonamide ligand coordi-
2
nates to the palladium center in a k -P,N fashion (Figure 4). The
accessed alternatively by addi-
X
tion of dmso to 1-acetone
(
Scheme 3). Generally, DMSO can
II
coordinate to the Pd center in
1
the palladium complexes in a k -
O or a k -S mode.
1
[9a,k,18]
In this
case, X-ray diffraction revealed
that DMSO binds to the palladi-
1
um center in
a
k -O mode
II
X
X
Scheme 3. Synthesis of neutral Pd methyl complexes 1-acetone and 1-dmso.
(
Figure 4). The Pd1ꢀO5 bond
length (2.136(4) ꢁ) in 1-dmso
is reasonably shorter than the
CF3
DIPP
Pd1ꢀO5(acetone) bond (2.148(2) ꢁ) is longer than the Pdꢀ Pd1ꢀO5 bond length of 2.148(2) ꢁ in
1-acetone (Table 2).
_
[(PO)Pd(Me)(dmso)]
O(dmso)
.146(3) ꢁ),
bonds
9a,18]
in
(2.131(2)–
Nozaki and co-workers reported that treatment of a phos-
phine–sulfonate ligand with 2,6-lutidine followed by addition
[
2
that is, acetone is a more weakly coordinating
II
solvent than DMSO (see below, Table 2). To the best of our
knowledge, this is the first structurally characterized palladium
complex directly coordinated by a weakly binding acetone
ligand. In addition, by replacement of acetone with DMSO as
a solvent in the reaction of Lb,c-Na and [PdMeCl(COD)], com-
of [PdMeCl(cod)] affords the neutral Pd complex [{(o-(o-
[9l]
MeOC H ) P)C H SO }PdMe(lutidine)]. In this case, the reac-
6
4 2
6
4
3
tions of phosphine–sulfonamide ligands La–c and pyridine did
not take place owing to the large pK value of sulfonamide;
a
II
X
however, treatment of Pd complexes 1-Cl with pyridine gen-
X
II
X
plexes 1-dmso were obtained in high yields, which could be
erated the neutral Pd methyl complexes 1-pyr, which could
Chem. Eur. J. 2015, 21, 2062 – 2075
www.chemeurj.org
2065 ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
II
2
.097(2) ꢁ) are in the expected range for Pd methyl complexes
Table 2. Selected bond lengths [ꢁ] for neutral palladium complexes.
[9]
(
Table 2).
DIPP
CF3
H
DIPP
CF3
1-acetone
1-dmso
1-pyr
1-pyr
1-pyr
PdꢀC
2.041(2)
2.137(2)
2.051(6)
2.162(5)
2.136(4)
2.203(2)
2.097(2) 2.051(2)
2.131(2) 2.155(2)
2.132(2) 2.118(2)
2.236(6) 2.248(6)
2.046(2)
2.155(1)
2.120(2)
2.228(4)
Stoichiometric insertion of methyl acrylate
PdꢀN1
PdꢀO5/N2 2.148(2)
The insertion of MA is one of the best-studied reactions of the
PdꢀP
2.208(5)
2
II
[9,18,19]
prototypical k -P,O Pd complexes.
We have reported that
II
the first insertion reaction of dianisyl phosphinesulfonato Pd
PSO3
PSO3
complexes
1’-NaCl and
1’-dmso with MA takes place
primarily in a 2,1 mode, and then the 2,1-insertion product
PSO3
2’MA-2,1 can undergo b-H elimination to give methyl croto-
nate or insert another molecule of MA to form the double-in-
PSO3
[9k,19]
sertion product
3’MA [Eq. (1)].
Note that the second in-
sertion step is predominant over b-H elimination from
PSO3
[9b,19]
2’MA-2,1
.
The electronic and steric effects of the phos-
phine–sulfonate on the insertion reactions have been studied
extensively, but a clear and general picture was not ob-
[9b,18,19a,b]
tained.
II
X
Scheme 4. Synthesis of neutral Pd methyl complexes 1-pyr.
II
also be obtained by adding pyridine to the cationic Pd methyl
X
+
X
complexes 1 -A (Scheme 4). In fact, 1-pyr could be generat-
II
ed by addition of pyridine to the neutral Pd methyl complexes
X
X
X
1
-acetone and 1-dmso as well. The formation of 1-pyr indi-
Insertion into cationic complexes
cated that pyridine is capable of promoting deprotonation of
the sulfonamide group in the palladium complexes. The
For the evaluation of the reactivity of the complexes toward
MA, the insertion reactions of an excess of MA (ca. 20 equiv)
1
X
H NMR spectra of 1-pyr clearly revealed the absence of the
X
+
1
NH hydrogen atom and the appearance of coordinated pyri-
with cationic complexes 1 -A were monitored by H NMR
spectroscopy at 258C over a period of 6–15 h (Supporting In-
formation, Figure S32–S42). Insertion products (Scheme 5)
were identified through assignment of their spin systems by
X
dine. The signals of the PdMe group in 1-pyr are shifted up-
field to ꢀ0.04, ꢀ0.20, and ꢀ0.18, respectively (Table 1). The
X
solid-state structures of 1-pyr were determined by X-ray crys-
1
1
1
1
1
13
tallography. The PdMe group and the P atom are located cis to
each other in the square-planar coordination sphere around
the palladium center (Figure 5). The anionic phosphine–sulfon-
H, H COSY and H, H TOCSY experiments, as well as by H, C
HSQC experiments if required. Under these pseudo-first-order
conditions, the first insertion of MA into the PdꢀMe bond to
2
X +
amide ligands coordinate to the palladium center in a k -P,N
give the first insertion product 2 -A is fast at 258C for all
MA
bidentate mode. The PdꢀC(Me) bond lengths (2.046(2)–
cationic complexes and is complete within about 30 min. The
X
Figure 5. ORTEPs of neutral complexes 1-pyr drawn with 50% probability ellipsoids. Hydrogen atoms and cocrystallized solvent molecules are omitted for
clarity.
Chem. Eur. J. 2015, 21, 2062 – 2075
www.chemeurj.org
2066
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
X
+
tion product 2 -A
is generally inhibited, be-
MA-1,2
cause of the formation of a stable five-membered
chelate, but the second insertion of MA and/or the b-
H elimination occur in the major, 2,1-insertion prod-
X
+
uct 2 -AMA-2,1. Notably, in comparison with the rela-
tively small rate constant for b-H elimination (k
=
bH-1st
ꢀ
5
ꢀ1
PSO3
1
.0ꢂ10 s ) of
2’MA-2,1, which is one order of
magnitude smaller than the rate constant for the
ꢀ
5
ꢀ1
PSO3
second MA insertion (k =9.2ꢂ10 s ) of
2’MA-
2nd
[19a]
2,1, analogous b-H elimination rate constants in all
cationic complexes 1 -A are two orders of magni-
tude higher. The largest kbH-1st is up to 280ꢂ10
Table 3, run 4), and thus b-H elimination is over-
whelmingly favored over the second MA insertion for
X
+
ꢀ
5
ꢀ1
s
(
X
+
1
2
-AMA-2,1. A typical H NMR analysis is depicted in
Figure 6. The prominent doublet of doublets (dd,
3
4
JHH =7.2, JHH =2.0 Hz, trans-Me) at d=1.86 ppm is
assigned to trans-methyl crotonate, which results
CF3
+
II
X
+
from the b-H elimination of (2 -SbF6)MA-2,1. The ob-
Scheme 5. Reaction of MA with cationic Pd complexes 1 -A.
CF3
+
served two major products
CF3
(4 -SbF )
and
6
MA-1,2
+
(
5 -SbF ) (rac:meso=6:5) originate from single
6 MA
net rate constants k of the first insertion are in the range
1,2 insertion or two consecutive 2,1 insertions of MA into the
1st
ꢀ
3
ꢀ1
[20]
[22]
(
2.0–3.5)ꢂ10
rate constant for
The observed ratios of 2,1 to 1,2 regioselectivity in the first in-
s
(Table 3, runs 1–6), which is close to the
intermediate PdH species, respectively. The observed ratio
PSO3
ꢀ3 ꢀ1
[19a]
1’-NaCl (k1st =3.2ꢂ10
s
at 258C).
of 2,1 to 1,2 regioselectivity in the second MA insertion is
about 3:1, which is lower than that of MA insertion into the
PdꢀMe bond (11:1). A different regioselectivity for MA insertion
into PdH and PdMe groups has also been found in diazaphos-
H
+
DIPP
+
sertion product are about 10:1, 6:1, and 11:1 for 1 -A,
A, and
1 -
CF3
+
1 -A respectively. This is in agreement with the
II
[18a]
trend that the 2,1 mode generally dominates in the first inser-
pholidinesulfonato Pd complexes.
In conclusion, the
II
CF3
+
CF3
+
tion step for phosphinesulfonato Pd methyl complexes. In the
ratio of insertion products
(5 -SbF )MA/ (4 -SbF )
/
6
6 MA-1,2
CF3
+
CF3
+
next step, the second insertion of MA into the minor, 1,2-inser-
(2 -SbF )MA-1,2/ (3 -SbF ) is about 65:23:9:3, which dif-
6 6 MA
fers strongly from the product distribution of MA insertion into
PSO3
PSO3
1’, in which
3’MA is the major product. This shows again
that b-H elimination is considerably favored for these cationic
Table 3. Rate constants of first insertion and b-H elimination and regiose-
lectivity.
X
+
[
a]
complexes 1 -A.
3
5
Run
Catalyst
10 k1st
10 kbH-1st
Regioselec-
tivity 2,1:1,2-
ꢀ1
ꢀ1
[s
]
[s
]
Insertion into weakly coordinated neutral complexes
H
H
+
[b]
1
2
3
4
5
6
7
8
9
1 -BF
4
2.0
2.2
3.5
3.3
ꢂ3.1
ꢂ3.0
3.2
63
180
ꢂ220
280
83
180
1.0
2.1
2.2
2.1
ꢂ1.5
3.0
n.o.
n.o.
n.o.
ꢂ10:1
The above findings suggested studies on the analogous MA in-
+
1 -SbF
6
6
10:1
6:1
2
X
X
DIPP
DIPP
CF3
CF3
PSO3
DIPP
CF3
PSO3
DIPP
CF3
H
+
[b]
sertions for the neutral k -P,N complexes 1-acetone and 1-
dmso. The reactions of an excess of MA (ca. 20 equiv) with the
1 -BF
4
+
1 -SbF
6:1
11:1
11:1
>15:1
+
[b]
[b]
X
X
1 -BF
4
neutral complexes 1-acetone and 1-dmso were monitored
+
1 -SbF
6
[c]
1
by H NMR spectroscopy at 258C over a period of 13–15 h
1’-NaCl
[b]
(Supporting Information, Figures S45–S49). Kinetic experiments
1-acetone
1-acetone
1’-dmso
1-dmso
1-dmso
1-pyr
1-pyr
1-pyr
1’-pyr
1-pyr
>100
>80
ꢂ7:1
12:1
20:1
revealed that the first insertion of MA into the PdꢀMe bond is
DIPP
10
0.7
11
3.8
0.003
0.005
0.0001
0.012
0.045
0.048
0.013
fast. Especially for
1-acetone, the insertion is complete
[b]
[b]
1
1
ꢂ7:1
within about 5 min. Note that the observed rate constants k
[b]
1st
1
1
1
1
1
1
1
1
2
3
4
5
6
7
8
9
ꢂ12:1
X
X
[d]
[d]
[d]
[e]
are a combination of the pre-equilibrium 1-solvent+MAÐ 1-
MA +solvent (solvent=acetone or dmso) and the actual rate
2,1
DIPP
CF3
PSO3
H
[e]
2,1
2,1
2,1
2,1
2,1
2,1
[e]
X
of insertion for 1-MA. Under the pseudo-first-order conditions,
[f]
[b]
[e]
ꢂ1.8
X
the weakly coordinated acetone complex 1-acetone under-
[f]
[d]
[e]
n.o.
DIPP
CF3
[f]
[b]
[e]
goes the fastest insertion of MA into the PdꢀMe bond (k
1st
>
1-pyr
1-pyr
ꢂ0.6
ꢀ
3
ꢀ1
[f]
[d]
[e]
n.o.
(80–100)ꢂ10 s ), which is significantly faster than those for
PSO3
ꢀ3 ꢀ1
PSO3
1
’-NaCl (k1st =3.2ꢂ10 s ) and dmso complexes
1’-
ꢀ1
[
a] Conditions: Pd (0.025 molL ), MA (20 equiv), CD
2
Cl
2
, T=258C, unless
ꢀ
3
ꢀ1
X
ꢀ3
dmso (k =0.7ꢂ10 s ) and 1-dmso (k1st =11 ꢂ 10 and
otherwise noted. [b] Exact determination not possible due to overlapping
or broad resonances. [c] By addition of 1.1 equiv AgBF , ref. [19a]. [d] No
b-H elimination observed. [e] No first 1,2 insertion product observed due
to the low intensity. [f] C Cl , T=508C.
1st
ꢀ1
ꢀ
3
4
3.8ꢂ10 s ; Table 3), which suggests that weakly coordinated
acetone is easily dissociated from the palladium center in the
presence of MA. The observed ratios of 2,1 to 1,2 regioselectiv-
2
D
2
4
Chem. Eur. J. 2015, 21, 2062 – 2075
www.chemeurj.org
2067
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
CF3
+
2 2 6
Figure 6. H NMR spectrum (CD Cl , 258C, 0–3.5 ppm) of 1 -SbF and MA (20 equiv) after 15 h at 258C.
ity in the first insertion product are about 7:1 and 12:1 for
DIPP
CF3
1
-acetone/dmso and
1-acetone/dmso, respectively. In
of the 2,1-in-
contrast, the b-H elimination rate constant k
bH-1st
X
sertion product 2
on the same order of magnitude (10 s ), which is much
smaller than those for the cationic complexes. The first MA in-
in all acetone and dmso complexes is
MA-2,1
ꢀ
5
ꢀ1
ꢀ
3
ꢀ1
sertion rate constant (k1 =3.8ꢂ10 s ) for the phosphine–
st
CF3
sulfonamide complex 1-dmso is slightly larger than that of
k1st =3.2ꢂ10
ꢀ
3
ꢀ1
PSO3
s
for phosphinesulfonato complex
1’-NaCl,
ꢀ
5
ꢀ1
but the second MA insertion rate constant (k2 =5.1ꢂ10 s )
nd
CF3
ꢀ5 ꢀ1
for
PSO3
1-dmso is smaller than that of k2nd =9.2ꢂ10
’-NaCl. In addition, a second MA insertion is favored over
s
for
1
X
b-H elimination for all cases, and thus 3 is the major prod-
MA
uct after about 15 h (Scheme 6).
II
X
X
Scheme 6. Reaction of MA with neutral Pd complexes 1-acetone and 1-
dmso.
Insertion into strongly coordinated neutral complexes
X
Compared with weakly coordinated acetone and dmso mole-
cules, strongly coordinated pyridine generally impedes the
pre-equilibrium reaction (Pd-pyridine+monomerÐPd-mono-
mer+pyridine) to disfavor monomer insertion, and thus de-
creases the overall insertion rate of monomer (Supporting In-
formation, Figures S54–S56). Under pseudo-first-order condi-
the first 2,1 insertion of MA into the pyridine complexes 1-pyr
are four to five orders of magnitude smaller than those for the
ꢀ
3
ꢀ1
corresponding acetone complexes (k1st >(80–100)ꢂ10 s ) at
258C (Table 3, runs 13–15).
To explore the insertion of MA into the pyridine complexes
X
1-pyr more clearly, the same kinetic experiments were carried
out at higher temperature (508C; Supporting Information, Fig-
ꢀ
7
ꢀ1
tions, the observed rate constants (k1st =(1.0–50)ꢂ10 s ) of
Chem. Eur. J. 2015, 21, 2062 – 2075
www.chemeurj.org
2068
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
CF3
ures S57–S59). The observed rate constants (k =(1.3–4.8)ꢂ
structure of
2-pyrMA-2,1 (cf. Supporting Information, Fig-
1st
ꢀ
5
ꢀ1
[23]
CF3
10
s ) of the first MA insertion at a 508C are about one
ure S64). For the evaluation of the reactivity of 2-pyrMA-2,1
order of magnitude higher than those at 258C (Table 3,
toward MA, the behavior of an excess of MA (ca. 20 equiv) to-
CF3
1
runs 17–19), and are close to that for the phosphinesulfonato
wards 2-pyrMA-2,1 was monitored by H NMR spectroscopy at
508C over 15 h (Supporting Information, Figure S63). The kinet-
ic experiment effectively confirmed that both b-H elimination
PSO3
ꢀ5 ꢀ1
complex
1’-pyr (k (508C)=1.2ꢂ10 s ; Table 3, run 16;
1st
see Supporting Information, Figures S60 and S61), but still sig-
nificantly smaller than those for corresponding acetone com-
CF3
and MA insertion did not take place in 2-pyrMA-2,1. By con-
X
X
CF3
plexes 1-acetone and dmso complexes 1-dmso. In the first
trast, when the reaction of MA with 2-pyrMA-2,1 was moni-
CF3
insertion step, the 2,1 mode is observed exclusively
tored at 958C, fast b-H elimination of
2-pyrMA-2,1 was ob-
(Scheme 7). In addition, b-H elimination of the 2,1-insertion
served.
Dimerization and
oligomerization of ethylene
Ethylene oligomerization to form
linear short-chain a-olefins has
received much attention from
both academia and industry
[24]
over the past decades.
De-
pending on the chain length of
linear a-olefins, they can be
II
X
Scheme 7. Reaction of MA with neutral Pd complexes 1-pyr at 508C.
used for various purposes. For
example, 1-butene, 1-hexene,
and 1-octene can be employed
DIPP
product is not observed at 508C after 15 h, except for
2-
as comonomers in ethylene copolymerization for the produc-
[25]
pyrMA-2,1 (Scheme 7). The observed b-H elimination rate con-
tion of linear low-density polyethylene. To date, a number of
palladium(II) catalysts with various bidentate ligands such as
ꢀ
5
ꢀ1
DIPP
stant (kbH-1st =0.6ꢂ10 s ) of
2-pyrMA-2,1 is on the same
ꢀ
5
ꢀ1
PSO3
[26]
[27]
order of magnitude as that of kbH-1st =1.8ꢂ10 s ) for
1’-
bis-phosphine,
phosphine–trifluoroborate,
NHC–pyri-
[28a,b]
[28c–e]
pyr. Note that we did not observe the consecutive MA inser-
dine,
phosphine–pyridine,
bis(heterocycle)–metha-
and 1,10-phenanthroli-
[
28f]
[28h]
tion product for all pyridine complexes.
ne,
phenacyldiaryl–phosphine,
[
28i,j]
ne
have been reported to dimerize ethylene, with a moder-
ate content of 1-butene. In addition, it is well-known that bi-
II
PSO3
dentate phosphinesulfonato Pd complexes
1’ can catalyze
Further reactivity of the isolated first, 2,1-insertion product
the polymerization of ethylene to produce linear polyethy-
H
CF3
In the pyridine complexes 1-pyr and
1-pyr, neither b-H
[9]
lene.
elimination nor second MA insertion occur in the initial 2,1-in-
H
CF3
sertion products 2-pyr
and 2-pyrMA-2,1. To further eluci-
MA-2,1
Ethylene dimerization catalyzed by cationic complexes
date this observation, studies were conducted starting from
CF3
the isolated 2,1-insertion product
2-pyrMA-2,1. Treatment of
Under pressure-reactor conditions (20 bar, 608C, 30 min), expo-
CF3
the acetone complex 1-acetone with an excess of MA for
II
X +
sure of the cationic Pd methyl complexes 1 -A to ethylene
resulted in a high catalytic activity, as monitored by the ethyl-
ene mass flow. Butenes, small amounts of hexene, and very
ꢀ3
ꢀ1
5
min at 258C (k1st >80ꢂ10 s ), followed by addition of pyri-
CF3
dine, yielded the targeted 2,1-insertion product
2-pyrMA-2,1
1
CF3
(
Scheme 8). In the H NMR spectrum of
2-pyrMA-2,1, triplets
1
little octene were formed, as revealed by H NMR, GC, and GC-
for a-H and g-H appear at d =1.84 ppm and ꢀ0.10 ppm, re-
spectively, and two broad singlets at d =0.58 and 1.13 ppm
are assigned to b-H (Supporting Information, Figure S62). Suit-
[29]
MS analysis of the reaction mixture. Note that the workup
procedure involved cooling of the pressure reactor to 08C
prior to opening after the reaction, after which samples were
withdrawn and analyzed. Hereby, the larger part of the bu-
tenes content is actually recorded. However, unavoidable loss
of the volatile butenes will result in a somewhat overestimated
value of the ratio of C and C to C in Table 4.
CF3
able crystals of 2-pyrMA-2,1 were isolated and analyzed by X-
ray diffraction, which unambiguously proved the molecular
6
8
4
X
+
4
ꢀ1
Complexes 1 -BF showed a high activity of 6.2ꢂ10 h
4
for ethylene dimerization, giving butenes (C ) as the major
4
product (Table 4, runs 1–3). For different N-aryl substitution
patterns, no apparent differences in activity were observed,
but the content of C4 in the short chain a-olefin product
signifycantly increased in the order CF3 (89.9%)>isopropyl
(71.2%)>H (53.2%). This different behavior shows that the sul-
CF3
Scheme 8. Synthesis of initial 2,1-insertion product 2-pyrMA-2,1
.
Chem. Eur. J. 2015, 21, 2062 – 2075
www.chemeurj.org
2069
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[30]
system.
On the other hand, pressure-reactor experiments
II
Table 4. Ethylene oligomerization by cationic and neutral Pd phos-
phine–sulfonamide catalysts.
[
a]
under similar conditions as in these ethylene dimerization ex-
periments but in the presence of MA also yielded butenes as
the major product (808C, 10 bar, 0.3m methyl acrylate,
[
b]
[c]
Run
Catalyst
T
p
E
TOF
Selectivity [%]
CF3
+
2
0 mmol, 1 -SbF ).
ꢀ
1
[d]
6
precursor
[8C]
[bar] [h
]
C
4
(a)
C
6
C
8
H
DIPP
CF3
H
+
1
2
3
4
5
6
7
8
9
1 -BF
4
4
4
60
60
60
60
60
60
20
20
20
20
20
20
5.5
6.2
6.2
4.3
4.8
5.5
–
3.6
5.7
5.7
3.6
5.2
4.8
4.8
3.8
53.2 (99.4) 30.8 16.0
71.2 (99.7) 22.0 6.8
+
1 -BF
Ethylene oligomerization catalyzed by neutral complexes
+
1 -BF
89.9 (99.0) 7.5
2.6
+
1 -SbF
6
6
6
82.7 (97.6) 14.6 2.7
In contrast, under pressure-reactor conditions (20 bar, 25–808C,
DIPP
CF3
X
+
1 -SbF
89.6 (87.7) 8.4
97.3 (92.3) 2.4
oligomers
2.0
0.3
2
II
3
0 min), very low to no activity of the neutral k -P,N Pd
+
1 -SbF
X
[e]
methyl complexes 1-acetone/dmso/pyr towards ethylene was
monitored by mass flow. The H NMR spectra and GC analysis
1-L
25–80 20
CF3
CF3
CF3
CF3
CF3
CF3
CF3
CF3
+
1
1 -BF
4
4
4
60
60
60
60
60
60
80
40
5
10
40
5
10
40
40
40
92.2 (98.7) 6.5
92.1 (99.1) 6.6
90.1 (99.3) 8.2
97.3 (72.5) 2.1
97.4 (83.6) 2.2
97.7 (99.3) 1.8
96.2 (92.6) 3.6
95.8 (99.2) 3.8
1.3
1.3
1.7
0.6
0.4
0.5
0.2
0.4
+
1 -BF
of the reaction solution revealed that only very little oligomer
was obtained.
+
10
1 -BF
+
1
1
1 -SbF
6
6
6
6
6
+
1
1
1
1
2
3
4
5
1 -SbF
+
1 -SbF
+
Computational studies
1 -SbF
+
1 -SbF
The experimental stoichiometric and pressure-reactor studies
[
a] Conditions: Pd (3 mmol), toluene (100 mL), 30 min, 1000 rpm. [b] Turn-
over frequency (TOF)=(10 ꢂmoles ethylene consumed)/(moles Pdꢂh).
c] Determined from the reaction solution by GC and GC-MS and H NMR.
d] a=amount of 1-butene in the C dimers. [e] X=H, DIPP, CF ; L=ace-
tone, dmso, pyridine.
II
showed that the Pd phosphine–sulfonamide complexes can
4
1
rapidly insert acrylate, but no polyethylene chain growth
occurs for the complexes studied. To further understand this
[
[
4
3
CF3
+
CF3
behavior, reactivities of the cationic
1
and the neutral 1,
phosphine–sulfonamide Pd complexes towards ethylene (E)
were investigated by DFT calculations.
H
II
1
fonamide moiety stays coordinated to the metal center during
catalysis. The amount of 1-butene is more than 99.0% in these
experiments, and only traces of internal isomers are formed.
With increasing ethylene pressure from 5 to 40 bar, the
CF3
+
Cationic complex
1
Concerning the free-energy profile for the first ethylene inser-
CF3
+
tion into PdMe, for the k-O-Pd-Me fragment
1 (A) a reaction
amount of C and 1-butene scarcely changed for the active
4
CF3
+
coordinate comprising coordination of ethylene cis to the
oxygen atom, cis–trans isomerization to place ethylene trans to
the oxygen atom, and ethylene insertion to yield the b-agostic
Pd propyl complex (O)b-ago-Pd-propyl was studied
(Scheme 9). In addition, for each intermediate and the insertion
catalyst 1 -BF , although the activity was affected (Table 4,
4
runs 3 and 8–10).
On changing the anion from BF to SbF , a slight decrease of
4
6
the activity for ethylene dimerization was observed with com-
X
+
plexes 1 -SbF (Table 4, runs 4–
6
6). However, the trend of the
amount of C was in agreement
4
X
+
with that of the 1 -BF system.
4
The order of the C4 content
is
CF3
(97.3%)>isopropyl
(
89.6%)>H (82.7%), that is,
ꢀ1
Scheme 9. Free-energy profile [kcalmol in solvent] of the first ethylene insertion into a PdMe species to form
CF3
+
CF3
+
complex 1 -SbF with its elec- a Pd-propyl species for 1 .
6
tron-withdrawing CF3 group
yields the highest content of butenes of all six cationic com-
plexes under the same conditions. Essentially, these findings
confirm that the counterion is noncoordinating and has little
effect on catalytic activity. Additionally, the studies with the
two different counterions mutually confirmed the trends in se-
lectivity observed for the different phosphine–sulfonamides. In
particular, with increasing ethylene pressure from 5 to 40 bar,
transition state (TS), energies for the respective k-N-Pd species
B were also calculated. For the sake of clarity, Scheme 9 con-
tains relevant species and energies along the lowest-energy
pathway (for all other geometries and energies, see Supporting
Information).
These calculations revealed that, for the cationic complex
CF3
+
1 , k-O-coordinated species are more stable than the re-
the content of C in the product was almost unchanged (97.3–
4
9
7.7%), whereas the amount of 1-butene in the C fraction in-
4
creased significantly from 72.5 to 99.3% (Table 4, runs 6 and
1–13). This phenomenon suggests that a high pressure of eth-
1
ylene can suppress the 2,1 reinsertion of the 1-butene into
PdH species to generate the isomerization product 2-butene,
which has also been observed in the analogous Ni catalyst
Chem. Eur. J. 2015, 21, 2062 – 2075
www.chemeurj.org
2070
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[
33]
spective k-N-coordinated species, the only exception being
generally viewed as a proper model of the growing chain,
ꢀ
1
(O)b-ago-Pd-propyl, which is 1.4 kcalmol less stable than the
an essentially similar behavior of Pd butyl and Pd propyl spe-
cies is expected, considering the competition between b-H
elimination and ethylene insertion. Thus, the selective forma-
tion of 1-butene seems reasonable as long as 1-butene itself is
not consumed.
respective (N)b-ago-Pd-propyl. Furthermore, the transition
state for the isomerization of (O)b-ago-Pd-propyl to (N)b-ago-
ꢀ
1
Pd-propyl is only 8.7 kcalmol
higher than (O)b-ago-Pd-
propyl, which makes this O/N isomerization kinetically viable.
Note that the calculated overall free-energy barrier to ethylene
Interestingly, the calculations suggest that the propensity for
CF3
+
CF3
+
insertion starting from cationic
1
(Pd-Me) is only 13.1 kcal
b-H elimination of the
1 system may not primarily be relat-
ꢀ
1
mol , in agreement with the experimentally observed high
ed to the positive charge of the complex, but rather to the
possibility of switching from O coordination to N coordination
of the sulfonamide ligand to give the N-coordinated intermedi-
ate (N)b-ago-Pd-propyl, which allows for a facile b-H elimina-
CF3
+
[31]
catalytic activity of the
1 system towards ethylene.
Starting from both (O)b-ago-Pd-propyl and (N)b-ago-Pd-
propyl, two competitive reactions were further considered:
1
) b-H elimination in the presence of ethylene leading to pro-
tion reaction rather than
(Schemes 10 and 11).
a further chain-growth step
pene plus (O/N)E-cis-Pd-H (Scheme 10), and 2) a second ethyl-
ene insertion into the Pdꢀpropyl bond (Scheme 11). To better
clarify the issues at hand, the energy of (O)b-ago-Pd-propyl
CF3
H
Neutral complexes 1 and 1
[
32]
was used as a reference point.
With the aim of rationalizing the experimentally observed low
activity of the neutral complexes for ethylene polymerization,
CF3
H
the reactivity of complexes 1 and 1 towards ethylene was
studied and compared with that of
II
dianisyl phosphinesulfonato Pd
PSO3
complex
1’ as a reference
1 and 1 were chosen
[34] CF3
H
system.
ꢀ
1
Scheme 10. Free-energy profile [kcalmol in solvent] of the elimination
pathways from Pd-propyl for 1 .
CF3
+
because the corresponding R sub-
stituents (CF , H) differ in terms of
3
both steric and electronic effects,
but the two complexes show the
same behavior in producing ethyl-
ene oligomers with a low activity.
The lowest free-energy pathway for b-H elimination starting
ꢀ1
from (O)b-ago-Pd-propyl lies 11.7 kcalmol higher than (O)b-
ago-Pd-propyl and comprises isomerization of (O)b-ago-Pd-
propyl to (N)b-ago-Pd-propyl prior to b-H elimination. In con-
trast, a b-H elimination starting from (O)b-ago-Pd-propyl and
This analysis (Scheme 12) suggests that the energy profiles
CF3
of ethylene insertion of the neutral sulfonamide systems
1
1
ꢀ
1
H
CF3
directly evolving into (O)P-trans-Pd-H requires 15.3 kcalmol
Scheme 10).
The lowest free-energy pathway for the insertion of ethylene
into (O)b-ago-Pd-propyl proceeds without N/O isomerization
and 1 are higher than for the reference system, and the
(
system shows the highest energy in each catalytic step (i.e., in-
termediates and transition states). In fact, the calculated
ꢀ
1
CF3
energy of ethylene coordination is 5.3 kcalmol for 1 and
ꢀ
1
ꢀ1
H
ꢀ1
and lies 20.9 kcalmol
higher than (O)b-ago-Pd-propyl
0.9 kcalmol for 1, as opposed to only ꢀ3.9 kcalmol for
the reference system. The energy barriers of the cis/trans isom-
(
Scheme 11). This compares to a calculated barrier for b-H elim-
ꢀ
1
[35]
CF3
H
ꢀ1
ination of 12 kcalmol , in agreement with the experimental
erization
for
1 and 1 are roughly 8 and 5 kcalmol
CF3
+
finding that
1
is an ethylene dimerization catalyst.
higher than for the reference system, respectively, and at the
CF3
+
In fact, with respect to the cationic system 1 , the calcula-
tions indicate that, after the first fast ethylene insertion into
the PdMe catalyst precursor, the system prefers to undergo
elimination instead of inserting another ethylene molecule.
The formed Pd hydride complex inserts and eliminates ethyl-
ene in a fast equilibrium reaction. However, the Pd ethyl com-
plex present in this equilibrium eventually undergoes further
ethylene insertion, which ultimately results in the formation of
a Pd butyl complex. Considering that a Pd propyl species is
same time the corresponding E-trans-Pd-Me intermediates are
ꢀ
1
7.3 and 1.6 kcalmol higher, respectively. In view of the rate-
determining step of ethylene insertion, the ethylene insertion
ꢀ
1
CF3
H
TS lies at 23.5 and 18.6 kcalmol for 1 and 1, respectively,
ꢀ
1
which are about 9 and 4 kcalmol higher than that of the ref-
erence system. This agrees with the experimentally observed
CF3
H
very low activity of 1 and 1 towards ethylene.
To further rationalize this result, a steric and electronic analy-
CF3
H
sis of 1 and 1 (Table 5) was performed. By comparing these
ꢀ
1
CF3
+
Scheme 11. Free-energy profile [kcalmol in solvent] of the ethylene insertion pathways from Pd-propyl for 1 .
Chem. Eur. J. 2015, 21, 2062 – 2075 www.chemeurj.org ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2071

Full Paper
CF3
geometry of 1 (Figure 7), in which an interaction between
the fluorine atom of the ligand and the Pd center is estab-
[36]
lished with a PdꢀF distance of 2.34 ꢁ. This interaction, which
stabilizes the intermediate, can account for the slightly smaller
CF3
positive charge calculated for the Pd center in 1 (despite the
presence of the electron-withdrawing CF group) with respect
3
H
to 1.
ꢀ
1
Scheme 12. Free-energy profile [kcalmol ] of the first ethylene insertion
into PdMe for 1 and 1.
Despite some electronic differences between the sulfon-
amide and the sulfonate systems, the greater steric hindrance
CF3
H
CF3
at the Pd center in the neutral sulfonamide catalysts 1 and
H
1
may be responsible for the experimentally observed low ac-
Table 5. Electronic and steric analysis of Pd complexes.
tivity towards ethylene. This differs from the effect of steric
[37]
bulk in cationic a-diimine catalysts.
To test this hypothesis, ethylene insertion into the less en-
CF3
H
1
1
Reference
Pd charge
+0.22
+0.25
+0.30
48.2
N-Me
cumbered neutral system
1 was calculated, in which
%V
Bur
63.0
56.1
a methyl substituent is located at the nitrogen atom of the sul-
fonamide ligand. In agreement with this interpretation, the
CF3
H
N-Me
PSO3
order of %VBur is
1 (63.0)> 1 (56.1)>
1 (51.6)>
1’
PSO3
results with those reported for the reference system
1’, it
(48.2), and the calculated order of energy of the ethylene inser-
CF3
H
CF3
H
N-Me
PSO3
clearly emerges that the sulfonamide complexes 1 and
1
tion TS is 1 (23.5)> 1 (18.6)>
1 (17.7)>
1’ (14.9).
are sterically more hindered at the Pd center, as indicated by
the buried volume %V . Furthermore, electronic analysis indi-
Bur
cates that they have a smaller positive charge on the Pd atom
CF3
than the reference system. Notably, the
1 system features
CF3
H
the lowest positive charge but the highest %V
(
1, 1, and
Bur
PSO3
1
’: +0.22e versus +0.25e and +0.30e; 63.0 versus 56.1 and
Conclusion
48.2), that is, the sulfonamide ligand affects the charge at the
Pd center through the PdꢀN bond, and the N-aryl substituent
also imparts a higher steric hindrance around the metal center.
The difference in the charge and in %VBur between the two sul-
fonamide systems can be rationalized by the analysis of the
These studies on the reactivities of new cationic and neutral
palladium(II) phosphine–sulfonamide complexes towards ole-
fins suggest that:
1
) The sulfonamide moiety in the cationic palladium(II) phos-
phine–sulfonamide complexes stays coordinated to the
metal center throughout all reaction steps. There is no evi-
dence that it only stabilizes the catalyst precursor, but the
actual catalytically active species features a monodentate
[38]
phosphine coordination mode.
2
3
) N coordination of the sulfonamide moiety promotes chain
transfer versus chain growth. This seems to be promoted
by the steric bulk of the additional organic substituents (N-
aryl or N-alkyl) on the coordinating hard donor, rather than
a different charge density on the central metal atom.
) For O-coordinated species, insertion barriers and barriers of
chain transfer could be potentially favorable for insertion-
polymerization chain growth, providing that chain transfer
can be suppressed.
This in-depth experimental and theoretical study on the
phosphine–sulfonamide motif also identified decisive features
of the unique neutral phosphinesulfonato motif, which is also
reflected in recently reported cationic phosphine/phosphine
monoxide and cationic phosphine/diethyl phosphonate com-
[39]
plexes. In addition to the unsymmetrical nature with a soft
P) and hard (O) s-donor and likely an appropriate charge at
(
the active site imparted by the donor characteristics of the bi-
dentate ligand, the lack of further substituents on an O donor
with their steric requirement appears to be beneficial.
CF3
Figure 7. Structure of 1. O red, P orange, N blue, S yellow, Pd green, F
cyan, C gray, H white.
Chem. Eur. J. 2015, 21, 2062 – 2075
www.chemeurj.org
2072
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
free of charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data request/cif.
Experimental Section
General procedures and materials
Further experimental details are given in the Supporting Informa-
tion.
Ethylene oligomerization
Oligomerization of ethylene was carried out in a 250 mL mechani-
cally stirred (1000 rpm) stainless steel pressure reactor equipped
with a heating/cooling jacket supplied by a thermostat controlled
by a thermocouple dipping into the polymerization mixture. A
valve controlled by a pressure transducer allowed for applying and
maintaining a constant ethylene pressure. The required flow of
ethylene, corresponding to ethylene consumed by polymerization,
was monitored by a mass flow meter and recorded digitally. Prior
to polymerization experiments, the reactor was heated under
vacuum to 908C for at least 60 min and then brought to the de-
sired temperature and backfilled with argon. A stock solution of
the catalyst precursor in methylene chloride was prepared daily
and stored in the glovebox at ꢀ308C. The amount required was
transferred into a syringe and removed from the glovebox. 100 mL
of toluene was transferred into the reactor through a cannula in
a slight argon stream before the catalyst solution was injected into
the reactor. After the desired reaction time, the reactor was first
cooled to 08C and then depressurized. The reaction was quenched
by adding 1 mL of methanol. The polymerization solution was im-
Unless stated otherwise, all manipulations of air- and moisture-sen-
sitive compounds were carried out under an inert gas atmosphere
by using standard glovebox and Schlenk techniques. Toluene was
distilled from sodium, methylene chloride and pentane were dis-
tilled from CaH , and THF and diethyl ether (Et O) were distilled
from sodium/benzophenone ketyl. Acetone p.a. and methyl acry-
late (MA) were degassed by repetitive freeze–thaw cycles and used
without further purification. DMSO and pyridine were purchased
2
2
from Aldrich and used without further purification. AgBF and
4
[40]
AgSbF6 were purchased from Aldrich. [PdMeCl(COD)] was pre-
pared according to a literature procedure. Ethylene (3.5 grade) was
purchased from Praxair. All deuterated solvents were supplied by
Eurisotop.
NMR spectra were recorded on a Varian Unity Inova 400 or
1
13
a Bruker Avance 400 spectrometer. H and C chemical shifts were
referenced to the solvent signals. The identity and purity of palladi-
1
13
um complexes and insertion products were established by H, C,
1
9
31
F, and P NMR spectroscopy, elemental analysis, and single-crystal
1
mediately studied by H NMR, GC, and GC-MS.
1
1
X-ray diffraction. NMR assignments were confirmed by H, H COSY,
1
1
1
13
1
13
19
H, H TOCSY, H{ C} HSQC, and H{ C} HMBC experiments. F and
3
1
P NMR chemical shifts were referenced to CFCl and 85% H PO ,
3
3
4
Reactions with ethylene and MA
respectively. Elemental analyses were performed on an Elementar
vario MICRO cube instrument. Gas chromatography (GC) was car-
ried out on a PerkinElmer Clarus 500 instrument with autosampler
and FID detection on a PerkinElmer Elite-5 (5% diphenyl-/95% di-
methylpolysiloxane) Series Capillary Columns (length: 30 m, inner
diameter: 0.25 mm, film thickness: 0.25 mm) by using helium as
Reactions with ethylene and MA were carried out in a 250 mL me-
chanically stirred (1000 rpm) stainless steel pressure reactor
equipped with a heating/cooling jacket supplied by a thermostat
controlled by a thermocouple dipping into the polymerization mix-
ture. A valve, controlled by a pressure transducer, allowed for ap-
plying and maintaining a constant ethylene pressure. Prior to olig-
omerization experiments, the reactor was heated under vacuum to
908C for at least 60 min, brought to the desired temperature, and
backfilled with argon. A solution of MA was prepared by dissolving
the desired amount of MA in toluene (50 mL total volume).
ꢀ1
carrier gas at a flow rate of 1.5 mLmin . The injector temperature
was 3008C. After injection the oven was kept isothermal at 408C
ꢀ
1
for 7 min, heated at 40 Kmin to 2808C, and kept isothermal at
808C for 3 min.
2
X-ray diffraction data were collected performed at 100 K on a STOE
IPDS-II diffractometer equipped with a graphite-monochromated
radiation source (l=0.71073 ꢁ) and an image-plate detection
system. A crystal mounted on a fine glass fiber with silicon grease
was employed. The selection, integration, and averaging procedure
of the measured reflex intensities, the determination of the unit-
cell dimensions by a least-squares fit of the 2q values, data reduc-
tion, LP correction, and space-group determination were per-
formed with the X-Area software package delivered with the dif-
2
00 mg of butylated hydroxytoluene (BHT) was added to this solu-
tion before it was cannula-transferred in a slight argon stream into
the reactor. A stock solution of the catalyst precursor in methylene
chloride was prepared daily and stored in the glovebox at ꢀ308C.
The amount required was transferred into a syringe, transferred
out of the glovebox, and the catalyst solution was injected into
the reactor. Ethylene pressure was applied, and after the desired
reaction time, the reactor was first cooled to 08C and then depres-
surized. The reaction was quenched by adding 1 mL of methanol,
and then three drops of polymerization solution was taken and im-
[41]
fractometer.
A semiempirical absorption correction was per-
formed. The structure was solved by direct methods (SHELXS-97),
completed by difference Fourier syntheses, and refined with full-
matrix least-squares techniques by using SHELXL-97 with minimiza-
1
mediately studied by H NMR, GC, and GC-MS.
À
Á
2
o
2
c
2 [42]
tion of w F ꢀ F
.
Following anisotropic refinement of all non-
Computational details
H atoms, ideal hydrogen positions were calculated in a isotropic
riding model. The weighted R factor (wR) and the goodness of fit S
All DFT geometry optimizations were performed at the GGA
[44]
[45]
2
BP86 level of theory with the Gaussian 09 package. The elec-
tronic configuration of the systems was described by the 6-31G
basis set for H, C, N, S, F, P and O, while for Pd we adopted the
quasi-relativistic LANL2DZ ECP effective core potential. All geo-
metries were characterized as minimum or transition state through
frequency calculations. The reported free energies were obtained
through single-point energy calculations on the BP86/6-31G geo-
metries by using the BP86 functional and the triple-z TZVP basis
set on main group atoms. Solvent effects were included with the
PCM model with toluene as solvent. In this BP86/TZVP electronic
were based on F ; the conventional R factor (R) was based on F. All
scattering factors and anomalous dispersion factors were provided
by the SHELXL-97 program. Hydrogen atoms were treated in
a riding model. Thermal ellipsoid representations were created
[46]
[43]
with ORTEP-3.
DIPP
DIPP
+
CCDC 973617 (Lb), 973618 ( 1-Cl), 973619 ( (1 -SbF )decomp.),
6
H
DIPP
CF3
[47]
9
77700
(
1-acetone), 977701
(
1-dmso), 973620 ( 1-pyr),
DIPP
CF3
9
73621 ( 1-pyr) and 973622 ( 1-pyr) contain the supplementa-
[48]
ry crystallographic data for this paper. These data can be obtained
Chem. Eur. J. 2015, 21, 2062 – 2075
www.chemeurj.org 2073
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
energy in solvent, the thermal corrections from the gas-phase fre-
quency calculations at the BP86/6-31G level were included.
[12] O. Pestovsky, A. Shuff, A. Bakac, Organometallics 2006, 25, 2894.
[
13] J. Vela, G. R. Lief, Z. Sheng, R. F. Jordan, Organometallics 2007, 26, 6624.
14] T. Rꢃnzi, U. Tritschler, P. Roesle, I. Gçttker-Schnetmann, H. M. Mçller, L.
Caporaso, A. Poater, L. Cavallo, S. Mecking, Organometallics 2012, 31,
[
8388.
Acknowledgements
[
15] C. C. Lu, J. C. Peters, J. Am. Chem. Soc. 2002, 124, 5272.
16] a) S. K. Bhargava, S. H. Privꢄr, A. C. Willis, M. A. Bennett, Organometallics
2012, 31, 5561; b) J. Zhang, R. Pattacini, P. Braunstein, Inorg. Chem.
2009, 48, 11954.
17] T. Rꢃnzi, D. Guironnet, I. Gçttker-Schnetmann, S. Mecking, J. Am. Chem.
Soc. 2010, 132, 16623.
18] a) P. Wucher, P. Roesle, L. Falivene, L. Cavallo, L. Caporaso, I. Gçttker-
Schnetmann, S. Mecking, Organometallics 2012, 31, 8505; b) P. Wucher,
L. Caporaso, P. Roesle, F. Ragone, L. Cavallo, S. Mecking, I. Gçttker-
Schnetmann, Proc. Natl. Acad. Sci. USA 2011, 108, 8955; c) A. Hamada, P.
Braunstein, Organometallics 2009, 28, 1688; d) A. Meduri, T. Montini, F.
Ragaini, P. Fornasiero, E. Zangrando, B. Milani, ChemCatChem 2013, 5,
1170; e) V. Rosar, A. Meduri, T. Montini, F. Fini, C. Carfagna, P. Fornasiero,
G. Balducci, E. Zangrando, B. Milani, ChemCatChem 2014, 6, 2403.
[19] a) B. Neuwald, L. Falivene, L. Caporaso, L. Cavallo, S. Mecking, Chem.
Eur. J. 2013, 19, 17773; b) P. Wucher, V. Goldbach, S. Mecking, Organo-
metallics 2013, 32, 4516; c) D. Guironnet, L. Caporaso, B. Neuwald, I.
Gçttker-Schnetmann, L. Cavallo, S. Mecking, J. Am. Chem. Soc. 2010,
132, 4418.
[
Z.J. is grateful to the Alexander von Humboldt Foundation for
a postdoctoral research fellowship and to the University of
Konstanz for a EU FP7 Marie Curie Zukunftskolleg Incoming
Fellowship Programme (grant no. 291784).
[
[
Keywords: coordination
modes
·
density
functional
calculations · dimerization · insertion · palladium
[
1] a) V. C. Gibson, S. K. Spitzmesser, Chem. Rev. 2003, 103, 283; b) S. Meck-
ing, Angew. Chem. Int. Ed. 2001, 40, 534; Angew. Chem. 2001, 113, 550;
c) S. Mecking, Coord. Chem. Rev. 2000, 203, 325; d) S. D. Ittel, L. K. John-
son, M. Brookhart, Chem. Rev. 2000, 100, 1169.
[2] L. K. Johnson, C. M. Killian, M. Brookhart, J. Am. Chem. Soc. 1995, 117,
6414.
[
3] a) L. Zhang, M. Brookhart, P. S. White, Organometallics 2006, 25, 1868;
b) F. A. Hicks, M. Brookhart, Organometallics 2001, 20, 3217; c) C. Wang,
S. Friedrich, T. R. Younkin, R. T. Li, R. H. Grubbs, D. A. Bansleben, M. W.
Day, Organometallics 1998, 17, 3149.
[20] The cationic complexes do not contain coordinated solvents, and thus
the net rate constant can be assumed to be equal to the observed rate
constant.
[
[
4] W. Keim, Angew. Chem. Int. Ed. Engl. 1990, 29, 235; Angew. Chem. 1990,
[21] Detailed assignments of these signals can be found in Table S1 of the
Supporting Information.
[22] To observe the product 4 -AMA-2,1, a full analysis was carried out after
102, 251.
CF3
+
5] a) S. Mecking, Macromol. Rapid Commun. 1999, 20, 139; b) B. L. Small,
M. Brookhart, J. Am. Chem. Soc. 1998, 120, 7143; c) B. L. Small, M. Broo-
khart, A. M. A. Bennett, J. Am. Chem. Soc. 1998, 120, 4049; d) G. J. P. Bri-
tovsek, V. C. Gibson, B. S. Kimberley, P. J. Maddox, S. J. McTavish, G. A.
Solan, A. J. P. White, D. J. Williams, Chem. Commun. 1998, 849.
5 h (see Figure S35 in the Supporting Information).
CF3
[23] In the crystal structure of 2-pyrMA-2,1, the incorporated MA moiety is
disordered; hence, we cannot discuss the crystal data but provide them
quantitatively as a proof and reference.
[
6] a) S. Mecking, L. K. Johnson, L. Wang, M. Brookhart, J. Am. Chem. Soc.
[24] a) D. S. McGuinness, Chem. Rev. 2011, 111, 2321; b) P. W. N. M. van Leeu-
wen, N. D. Cle m´ ent, M. J. L. Tschan, Coord. Chem. Rev. 2011, 255, 1499;
c) T. Agapie, Coord. Chem. Rev. 2011, 255, 861; d) C. Janiak, F. Blank, Mac-
romol. Symp. 2006, 236, 14; e) F. Speiser, P. Braunstein, L. Saussine, Acc.
Chem. Res. 2005, 38, 784; f) J. Skupinska, Chem. Rev. 1991, 91, 613.
[25] a) P. W. N. N. van Leeuwen, Homogeneous Catalysis, Kluwer Academic,
Dordrecht (The Netherlands), 2004, p. 175; b) “Oligomerization of ethyl-
ene to higher linear a-olefins”: D. Vogt, in Applied Homogeneous Cataly-
sis with Organometallic Compounds, Vol. 1 (Eds.: B. Cornils, W. A. Herr-
mann), 1996, Wiley-VCH, Weinheim, p. 245; c) D. E. James, Encyclopedia
of Polymer Science and Engineering, Vol. 6 (Eds.: H. F. Mark, N. M. Bikales,
C. G. Overberger, G. Menges), Wiley-Interscience, New York, 1985,
p. 429.
1998, 120, 888; b) L. K. Johnson, S. Mecking, M. Brookhart, J. Am. Chem.
Soc. 1996, 118, 267.
[7] E. Drent, R. van Dijk, R. van Ginkel, B. van Oort, R. I. Pugh, Chem.
Commun. 2002, 744.
[
8] For reviews, see: a) A. Nakamura, T. M. J. Anselment, J. Claverie, B. Good-
all, R. F. Jordan, S. Mecking, B. Rieger, A. Sen, P. W. N. M. van Leeuwen, K.
Nozaki, Acc. Chem. Res. 2013, 46, 1438; b) A. Nakamura, S. Ito, K. Nozaki,
Chem. Rev. 2009, 109, 5215; c) E. Y.-X. Chen, Chem. Rev. 2009, 109, 5157;
d) A. Berkefeld, S. Mecking, Angew. Chem. Int. Ed. 2008, 47, 2538;
Angew. Chem. 2008, 120, 2572.
[
9] For selected examples, see: a) H. Leicht, I. Gçttker-Schnetmann, S. Meck-
ing, Angew. Chem. Int. Ed. 2013, 52, 3963; Angew. Chem. 2013, 125,
4055; b) B. Neuwald, L. Caporaso, L. Cavallo, S. Mecking, J. Am. Chem.
[26] E. Drent, European Patent 0,170,311 B1, 1989.
Soc. 2013, 135, 1026; c) T. Friedberger, P. Wucher, S. Mecking, J. Am.
Chem. Soc. 2012, 134, 1010; d) L. Piche, J. C. Daigle, G. Rehse, J. P. Clav-
erie, Chem. Eur. J. 2012, 18, 3277; e) S. Ito, M. Kanazawa, K. Munakata, J.
Kuoda, Y. Okumura, K. Nozaki, J. Am. Chem. Soc. 2011, 133, 1232; f) Z.
Shen, R. F. Jordan, Macromolecules 2010, 43, 8706; g) T. Rꢃnzi, D. Frçh-
lich, S. Mecking, J. Am. Chem. Soc. 2010, 132, 17690; h) K. Nozaki, S. Ku-
sumoto, S. Noda, T. Kochi, L. W. Chung, K. Morokuma, J. Am. Chem. Soc.
[27] a) Y. Kim, R. F. Jordan, Organometallics 2011, 30, 4250; b) A. L. Gott, W. E.
Piers, J. L. Dutton, R. McDonald, M. Parvez, Organometallics 2011, 30,
4236.
[28] a) V. Khlebnikov, A. Meduri, H. Mueller-Bunz, T. Montini, P. Fornasiero, E.
Zangrando, B. Milani, M. Albrecht, Organometallics 2012, 31, 976; b) V.
Khlebnikov, A. Meduri, H. Mueller-Bunz, B. Milani, M. Albrecht, New J.
Chem. 2012, 36, 1552; c) J. Flapper, H. Kooijman, M. Lutz, A. L. Spek,
P. W. N. M. van Leeuwen, C. J. Elsevier, P. C. J. Kamer, Organometallics
2009, 28, 3272; d) J. Flapper, P. W. N. M. van Leeuwen, C. J. Elsevier,
P. C. J. Kamer, Organometallics 2009, 28, 3264; e) J. Flapper, H. Kooijman,
M. Lutz, A. L. Spek, P. W. N. M. van Leeuwen, C. J. Elsevier, P. C. J. Kamer,
Organometallics 2009, 28, 1180; f) C. T. Burns, R. F. Jordan, Organometal-
lics 2007, 26, 6726; g) M. D. Doherty, S. Trudeau, P. S. White, J. P.
Morken, M. Brookhart, Organometallics 2007, 26, 1261; h) J. M. Malinos-
ki, M. Brookhart, Organometallics 2003, 22, 5324; i) F. C. Rix, M. Broo-
khart, P. S. White, J. Am. Chem. Soc. 1996, 118, 4746; j) F. C. Rix, M. Broo-
khart, J. Am. Chem. Soc. 1995, 117, 1137.
2010, 132, 16030; i) S. Ito, K. Munakata, A. Nakamura, K. Nozaki, J. Am.
Chem. Soc. 2009, 131, 14606; j) S. Noda, A. Nakamura, T. Kochi, L. W.
Chung, K. Morokuma, K. Nozaki, J. Am. Chem. Soc. 2009, 131, 14088;
k) D. Guironnet, P. Roesle, T. Rꢃnzi, I. Gçttker-Schnetmann, S. Mecking, J.
Am. Chem. Soc. 2009, 131, 422; l) T. Kochi, S. Noda, K. Yoshimura, K.
Nozaki, J. Am. Chem. Soc. 2007, 129, 8948; m) W. Weng, Z. Shen, R. F.
Jordan, J. Am. Chem. Soc. 2007, 129, 15450; n) K. M. Skupov, P. R. Marel-
la, M. Simard, G. P. A. Yap, N. Allen, D. Conner, B. L. Goodall, J. P. Claverie,
Macromol. Rapid Commun. 2007, 28, 2033; o) D. K. Newsham, S. Borkar,
A. Sen, D. M. Conner, B. L. Goodall, Organometallics 2007, 26, 3636.
1
[
[
10] B. P. Carrow, K. Nozaki, J. Am. Chem. Soc. 2012, 134, 8802.
[29] Representative H NMR spectrum and GC trace can be found in the
2
11] A series of nickel complexes with k -P,N coordinated phosphine-sulfon-
Supporting Information; compare Figures S29 and S30.
[30] D. V. Gutsulyak, A. L. Gott, W. E. Piers, M. Parvez, Organometallics 2013,
32, 3363.
amide has been reported by Brookhart et al. to catalyze ethylene oligo-
merization (7<degree of polymerization<25), see: J. L. Bennett, M.
Brookhart, L. K. Johnson, C. M. Killian, Patent Application
WO1998030610A1, 1998.
[31] The calculated energetic barrier for ethylene insertion into the PdꢀMe
CF3
+
ꢀ1
ꢀ1
bond of
1 of 13.1 kcalmol is comparable to that of 14.9 kcalmol
Chem. Eur. J. 2015, 21, 2062 – 2075
www.chemeurj.org
2074
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
II
calculated for the neutral dianisyl phosphinesulfonato Pd complex
TS more than the elimination TS, which is reflected in the catalyst activ-
ity (see Supporting Information).
PSO3
1’.
[
32] a) For the chain transfer reaction, namely (N)P-trans-Pd-H+E!(N)E-
cis-Pd-H+P and (O)P-trans-Pd-H+E!(O)E-cis-Pd-H+P, different
routes, such as associative, dissociative, and concerted mechanisms,
have been considered and were calculated. See also W. Heyndrickx, G.
Occhipinti, V. R. Jensen, Chem. Eur. J. 2014, 20, 7962. b) Note that a com-
parison of the calculated values of energetic barriers for b-H elimination
and insertion needs to take into account the different molecularities of
the reactions.
[38] Triphenylphosphine-stabilized palladium complexes have been found
to catalyze ethylene dimerization to butene with low activity: a) A. Sen,
T. Lai, J. Am. Chem. Soc. 1982, 104, 3520; b) A. Sen, T. Lai, J. Am. Chem.
Soc. 1981, 103, 4627.
[39] N. D. Contrella, J. R. Sampson, R. F. Jordan, Organometallics 2014, 33,
3546.
[40] R. E. Rꢃlke, J. M. Ernsting, A. L. Spek, C. J. Elsevier, P. W. N. M. van Leeu-
wen, K. Vrieze, Inorg. Chem. 1993, 32, 5769.
[
[
33] P. Corradini, G. Guerra, L. Cavallo, Acc. Chem. Res. 2004, 37, 231.
34] Calculations reported here refer to the same geometries as reported in
reference [19a]. However, as outlined in the Experimental Section and
the Supporting Information, here we reported free energies in toluene
obtained by adding the thermal corrections from gas-phase frequency
analysis of the optimized geometries to the internal energies obtained
from single-point calculation in the solvent.
[41] X-RED version 1.31 Data Reduction Program, Stoe, Darmstadt, Germany,
2005.
[42] a) G. M. Sheldrick, SHELXS-97 Program for Crystal Structure Analysis,
University of Gçttingen, Gçttingen, Germany, 1997; b) G. M. Sheldrick,
SHELXL-97 Program for Crystal Structure Refinement, University of Gçt-
tingen, Gçttingen, Germany, 1997.
[43] L. J. Faruggia, ORTEP-3 V2.02 for Windows XP, University of Glasgow,
Glasgow, Scotland, 2008.
[44] a) A. D. Becke, Phys. Rev. A 1988, 38, 3098; b) J. P. Perdew, Phys. Rev. B
1986, 33, 8822; c) J. P. Perdew, Phys. Rev. B 1986, 34, 7406.
[45] M. J. Frisch et al. Gaussian 09 Revision A.1, Gaussian, Inc., Wallingford,
CT, 2009.
[46] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297.
[47] a) T. Leininger, A. Nicklass, H. Stoll, M. Dolg, P. Schwerdtfeger, J. Chem.
Phys. 1996, 105, 1052; b) W. Kꢃchle, M. Dolg, H. Stoll, H. Preuss, J. Chem.
Phys. 1994, 100, 7535; c) U. Hꢅussermann, M. Dolg, H. Stoll, H. Preuss,
Mol. Phys. 1993, 78, 1211.
[
35] Energies of the most stable cis–trans isomerization TSs are reported for
all systems. Energies for ethylene insertion TSs from the most stable cis
coordination intermediate were also calculated. They are higher than
ꢀ
1
the trans insertion TS by about 8 kcalmol
.
[
[
36] D. P. Gates, S. A. Svejda, E. OÇate, C. M. Killian, L. K. Johnson, P. S. White,
M. Brookhart, Macromolecules 2000, 33, 2320.
37] The different effects of bulky substituents in cationic a-diimine catalysts
and neutral sulfonamide catalysts, respectively, in ethylene polymeri-
zation essentially arise from the termination mechanism, which is bimo-
lecular (b-H transfer) in the former and monomolecular (b-H elimina-
tion) in the latter case. Since the bimolecular TS is more spatially de-
manding, it is affected more by a steric hindrance around the metal
center. As a consequence, bulky substituents on the diimine complexes
are essential for the formation of high molecular weight polyethylene.
Vice versa, for the sulfonamide complexes, steric interactions between
the ligand substituents and the growing chain destabilize the insertion
[48] a) V. Barone, M. Cossi, J. Phys. Chem. A 1998, 102, 1995; b) J. Tomasi, M.
Persico, Chem. Rev. 1994, 94, 2027.
Received: August 14, 2014
Published online on December 8, 2014
Chem. Eur. J. 2015, 21, 2062 – 2075
www.chemeurj.org
2075
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim