Insights into CO/styrene copolymerization by using PdII catalysts containing modular pyridine-imidazoline ligands

Article abstract of DOI:10.1002/chem.200306051

Continuing our studies into the effect that N-N′ ligands have on CO/styrene copolymerization, we prepared new C₁-symmetrical pyridine-imidazoline ligands with 4′,5′-cis stereochemistry in the imidazoline ring (5) and 4′,5′-trans stereochemistry (6-10) and compared them with our previously reported ligands (1-4). Their coordination to neutral methylpalladium(II) (5a-10a) and cationic complexes (5b-10b), investigated in solution by NMR spectroscopy, indicates that both the electronic and steric properties of the imidazolines determine the stereochemistry of the palladium complexes. The crystal structures of two neutral palladium precursors [Pd(Me)_2-nCl_n-(N-N′)] (n = 1 for 8a; n = 2 for 9a′) show that the Pd-N coordination distances and the geometrical distortions in the imidazoline ring depend on the electronic nature of the substituents in the imidazoline fragment. Density functional calculations performed on selected neutral and cationic palladium complexes compare well with NMR and X-ray data. The calculations also account for the formation of only one or two stereoisomers of the cationic complexes. The performance of the cationic complexes as catalyst precursors in CO/4-tert-butylstyrene copolymerization under mild pressures and temperatures was analyzed in terms of the productivity and degree of stereoregularity of the polyketones obtained. Insertion of CO into the Pd-Me bond, which was monitored by multinuclear NMR spectroscopy, shows that the N ligand influences the stereochemistry of the acyl species formed.

Full text of DOI:10.1002/chem.200306051

FULL PAPER
Insights into CO/Styrene Copolymerization by Using PdII Catalysts
Containing Modular Pyridine Imidazoline Ligands
Amaia Bastero,*[a] Carmen Claver,*[a] Aurora Ruiz,[a] Sergio CastillÛn,[b] Elias Daura,[a]
Carles Bo,[a] and Ennio Zangrando[c]
Abstract: Continuing our studies into
The crystal structures of two neutral
palladium precursors [Pd(Me)2ꢀnCln-
and X-ray data. The calculations also
account for the formation of only one
or two stereoisomers of the cationic
complexes. The performance of the cat-
ionic complexes as catalyst precursors
in CO/4-tert-butylstyrene copolymeri-
zation under mild pressures and tem-
peratures was analyzed in terms of the
productivity and degree of stereoregu-
larity of the polyketones obtained. In-
ꢀ
the effect that N N’ ligands have on
ꢀ
CO/styrene copolymerization, we pre-
pared new C1-symmetrical pyridine
imidazoline ligands with 4’,5’-cis stereo-
chemistry in the imidazoline ring (5)
and 4’,5’-trans stereochemistry (6 10)
and compared them with our previous-
ly reported ligands (1 4). Their coordi-
nation to neutral methylpalladium(ii)
(5a 10a) and cationic complexes (5b
10b), investigated in solution by NMR
spectroscopy, indicates that both the
electronic and steric properties of the
imidazolines determine the stereo-
chemistry of the palladium complexes.
(N N’)] (n=1 for 8a; n=2 for 9a’)
ꢀ
show that the Pd N coordination dis-
tances and the geometrical distortions
in the imidazoline ring depend on the
electronic nature of the substituents in
the imidazoline fragment. Density
functional calculations performed on
selected neutral and cationic palladium
complexes compare well with NMR
ꢀ
sertion of CO into the Pd Me bond,
which was monitored by multinuclear
NMR spectroscopy, shows that the N
ligand influences the stereochemistry
of the acyl species formed.
Keywords: copolymerization ¥ den-
sity functional calculations ¥ ligand
design ¥ N ligands ¥ palladium
4]
Introduction
erties of the products also make it an attractive reaction.[2
Several mechanistic studies have been performed with well-
However, the search for
chiral ligands that lead to the formation of regular micro-
10]
Late-transition-metal complexes can be used as polymeriza-
tion catalysts to access polymeric structures containing polar
monomers.[1] In particular, metal-catalyzed copolymerization
of carbon monoxide with alkenes provides greater control
over the polymer properties than radical polymerization,
and makes it possible to synthesise alternating polyketones.
The low cost of the starting materials and the material prop-
defined single-site catalysts.[5
structures is still an interesting topic.[11]
Palladium catalysts bearing nitrogen-donor ligands have
proved to be effective for CO/styrene copolymerization,[2,12]
unlike palladium diphosphane catalysts, which generally
21]
form oligomers.[13] In most cases bidentate N ligands[14
and, to a lesser extent, hemilabile P,N ligands have been
used successfully.[7,21] The common feature in all these li-
gands is the sp2 character of the coordinating N atom. Oxa-
zolines are effective ligands in several metal-catalyzed ho-
mogeneous reactions[22] including CO/styrene copolymeriza-
tion.[15,16,21] We believed that imidazolines, which are struc-
[a] Dr. A. Bastero, Prof. Dr. C. Claver, Dr. A. Ruiz, E. Daura, Dr. C. Bo
Departament de QuÌmica FÌsica i Inorg‡nica
Universitat Rovira i Virgili
Pl. Imperial T‡rraco 1, 43005 Tarragona (Spain)
Fax : (+34)977-559-563
E-mail: claver@quimica.urv.es
turally analogous but which have different electronic proper-
[b] Prof. Dr. S. CastillÛn
27]
ties,[23
could be good alternatives. Therefore, we
Departament de QuÌmica Org‡nica i QuÌmica AnalÌtica
Universitat Rovira i Virgili
Pl. Imperial T‡rraco 1, 43005 Tarragona (Spain)
developed racemic 1-substituted (R,S)-4,5-dihydro-4,5-di-
phenyl-2-(2-pyridyl)imidazoles 1 4 with 4’,5’-cis stereochem-
istry in the imidazoline moiety, which have the advantage
that the substituent in the aminic nitrogen atom N1 can be
[c] Dr. E. Zangrando
Dipartimento di Scienze Chimiche
Universit‡ degli Studi di Trieste
Via Licio Giorgieri 1, 34127 Trieste (Italy)
easily modified to lead to
a series of chiral ligands
(Figure 1).[28] This substitution makes it possible to fine tune
the electronic properties of the imidazoline ring over a wide
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2004, 10, 3747 3760
DOI: 10.1002/chem.200306051
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3747

FULL PAPER
nation of these ligands to neutral and cationic palladium
complexes was investigated in solution, and their structures
were compared by X-ray diffraction and DFT calculations.
The activity of the cationic palladium(ii) complexes in CO/
4-tert-butylstyrene copolymerization was studied and com-
pared with our data on pyridine imidazolines with 4’-5’-cis
stereochemistry.[28] Since site-selective coordination is crucial
for stereocontrol of the reaction, we analyzed the reactivity
of the complexes towards CO by NMR spectroscopy, that is,
monitoring the first step of the catalytic cycle.
Figure 1. Pyridine imidazolines 1 4 in comparison with the general struc-
ture of pyridine-oxazolines.
range without changing the chiral environment around the
donor N atom.
Modifying the electronic properties of palladium catalysts
by using different ligands may lead to variations not only in
the activity, but also in the selectivity.[29] The use of palladi-
um(ii) catalysts containing chiral C1-symmetrical pyridine-
oxazoline ligands in CO/styrene copolymerization leads to
syndiotactic polyketones.[15,21] This is because the site-selec-
tive coordination of the styrene cis to the pyridine moiety
means that the stereocontrol of the reaction is provided by
chain-end control and not by the chiral ligand (enantiosite
control) (Figure 2).[21] So it seems that to increase the co-
Results and Discussion
Synthesis and characterization of ligands: We first prepared
pyridine imidazoline ligands 1 4, whose phenyl rings in the
imidazoline moiety are in mutually cis positions (Scheme 1).
We observed that modifying the R substituent in the imida-
zoline ring influenced the properties of the binding N atom.
This led to palladium complexes with different stereochem-
istries, which catalyzed the synthesis of polyketones with
different degrees of stereoregularity.[28] To try to promote
the coordination of styrene cis to the imidazoline, we in-
creased the steric hindrance near the pyridine N atom and
simultaneously decreased the basicity of the pyridine ring by
synthesising racemic (R,S)-4,5-dihydro-4,5-diphenyl-2-(6-cy-
anopyridyl)imidazole (5; Scheme 1). The product is obtained
in a low yield (20%) since the bis-anellation product is also
formed.[30] Since it was recently reported that in CO/ethene
copolymerization palladium catalysts containing (R,S)- or
(S,R)-meso-diphosphanes behave differently to those bear-
ing the ligands with R,R or S,S configuration (Figure 3),[31,32]
we tested the effect of varying the stereochemistry of our
imidazolines by preparing ligands 6 10.
Figure 2. Model proposed by Consiglio et al. for styrene insertion to give
predominantly syndiotactic copolymer. GPC=growing polymer chain.
polymer content of l diads with our pyridine imidazolines,
styrene must be selectively coordinated cis to the chiral
imidazoline. For this reason we decided to take into account
both steric and electronic factors in ligand design.
We have now synthesized 1-substituted (R,R)-4,5-dihydro-
4,5-diphenyl-2-(2-pyridyl)imidazoles 6 10 with 4’,5’-trans
stereochemistry in the imidazoline (Scheme 1). The coordi-
Enantiomerically pure (R,R)-pyridine imidazoline 6 was
prepared in quantitative yield by reaction of the correspond-
ing diamine and 2-cyanopyridine, similar to 1 (Scheme 1).
Reaction of 6 with different
electrophiles (benzyl bromide,
methyl iodide, p-toluenesulfon-
yl chloride, and trifluorome-
thylsulphonyl anhydride) in
the presence of a base gave the
substituted (R,R)-pyridine imi-
dazolines 7 10 (Scheme 1).
1
The H NMR spectra of these
ligands show the characteristic
two doublets corresponding to
H4’ and H5’ of the imidazoline
moiety (see Scheme 1 for num-
bering). In the case of 6, the
two doublets become a singlet
at 5 ppm due to tautomeric
equilibrium. A comparison of
the coupling constants of these
two protons in ligands 1 10
shows that the nature of the R
Scheme 1. Synthesis of ligands 1 10 showing numbering scheme. i) meso-(1R,2S)-1,2-Diphenylethylenediamine.
i’) (1R,2R)-1,2-diphenylethylenediamine; ii) 4-dimethylaminopyridine for 3, 4, 9, 10; NaH for 2, 7, 8; iii) BnBr
for 2, 7; TsCl for 3, 9; MeI for 8; Tf2O for 10.
substituent has an effect on the
coupling constant 3J(4’,5’),[33]
3748
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 3747 3760

Pd Complexes of Pyridine Imidazoline Ligands
3747 3760
cal relationship between the
methyl group and the imidazo-
line ring). In the spectra of
complexes 6a 9a the methyl
group bonded to palladium ap-
pears as an upfield-shifted sin-
Figure 3. meso-Diphosphanes used as ligands in CO/ethene copolymerization.
glet (average shift: 0.47 ppm;
Table 2). Since a methyl group
s-bonded to palladium normal-
which decreases for the electron-withdrawing, bulky p-tolu-
enesulfonyl (Ts) and trifluoromethylsulfonyl (Tf) groups
(Table 1). It is worth noting that for enantiomerically pure
(R,R)-ligands 7 10 these variations are more pronounced
than for racemic (R,S)-ligands 2 4.[28]
ly appears at around 1 ppm,[36] the presence of this signal at
lower frequencies may be due to the unusual proximity be-
tween the methyl group and the phenyl ring in the 4’-posi-
tion in these complexes.
The cis stereochemistry was also observed in the solid
state for complex 8a. Single crystals suitable for X-ray anal-
ysis were obtained for this neutral complex, which contains
a ligand substituted with an electron-donating group (R=
Me; Figure 4). Efforts to obtain single crystals of complex
9a, with a ligand bearing an electron-withdrawing group
(R=Ts), resulted in isolation of [PdCl2(9)] (9a’; Figure 5).[37]
To complete the study on the coordination of the (R,R)-
pyridine imidazolines to palladium, DFT calculations were
performed for complexes 6a and 10a. Table 3 shows a selec-
tion of measured bond lengths and angles of the molecular
structures for the neutral complexes 8a and 9a’, and calcu-
lated values for 6a and 10a.
Table 1. Selected 1H NMR data for ligands 1 10 in CDCl3 at room tem-
perature.
Ligand
R
H4’, H5’
3J4’-5’ [Hz]
1
H
5.51
2
3
4
5
Bn
Ts
Tf
H
5.44, 4.92
5.93, 5.80
5.98, 5.92
5.58
11.6
10
8.7
6
H
5.0
7
8
9
10
Bn
Me
Ts
Tf
5.0, 4.42
4.86, 4.25
5.36, 5.17
5.45, 5.37
9.2
10.5
4.8
3.8
ꢀ
The Pd N(py) distances are longer than those involving
the iminic N atom of the imidazoline N atom N2 (Table 3).
ꢀ
The Pd N1 distance observed in 8a is particularly long
(2.121(6) ä), because of the trans influence exerted by the
Synthesis and characterization of and DFT calculations of
neutral methylpalladium(ii) complexes: Similar to what was
observed for ligands 1 4,[34] the
reaction of ligands 5 10 with
[PdClMe(cod)] (cod=1,5-cy-
clooctadiene) in toluene at
room temperature led to pre-
cipitation of neutral complexes
5a 10a (Scheme 2).
methyl group coordinated to palladium (Figure 4). On the
Characterization of these
complexes in solution by mul-
tinuclear NMR spectroscopy
showed the presence of only
one isomer. For complexes 6a
10a in the aromatic region of
the 1H NMR spectra, the
signal for H6 (see Scheme 1) is
considerably shifted downfield
with respect to that in the free
ligand (Dd=0.52 ppm). This
indicates that H6 is subject to
the anisotropic effect of the
neighboring chloro ligand.[35]
NOE experiments show the in-
teraction between the PdMe
group and H4’ of the imidazo-
line ring, which confirms that
all neutral complexes 5a 10a
are cis isomers (where cis and
trans indicate the stereochemi-
Scheme 2. Synthesis of 1a 10a and 1b 10b. The numbering Scheme of Ar04Bꢀ is included.
Chem. Eur. J. 2004, 10, 3747 3760
www.chemeurj.org ¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3749

C. Claver, A. Bastero et al.
FULL PAPER
Table 2. Selected 1H NMR data for complexes 5a 10a and 5b 10b in
CDCl3 at room temperature.[a]
Compound
H6
PdMe
PdNCMe
5a[b]
6a
7a
8a
9a
10a
5b
6b
0.72
0.34
0.52
0.48
0.55
0.75
0.90
0.56
0.54
0.48
0.57
0.97
0.74
1.07
8.82 (d)
9.26 (d)
9.28 (d)
9.24 (d)
9.29 (d)
2.33
2.30
2.28
2.30
2.20
1.61
2.26
1.73
8.34 (d)
8.38 (d)
8.38 (d)
7b
8b
9b[c]
M: 8.35 (dd)
m: 8.50 (d)
M: 8.38 (d)
m: 8.52 (d)
10b[c]
[a] The signals are singlets unless otherwise stated. Coupling constants
are omitted for clarity. [b] In (CD3)2(CO). [c] M: major isomer; m: minor
isomer.
Figure 5. ORTEP plot (40% probability thermal ellipsoids) of the molec-
ular structure of 9a’.
lated complexes. However, the dihedral angle formed by the
best-fit planes through the rings are significantly different:
2.6(4)8 in 8a, 3.08 in 6a, 7.68 in 10a, and 17.9(3)8 in 9a’. The
large angle in 9a’ might be ascribed to a distortion in the
imidazoline plane that favors intramolecular p stacking of
the tosyl ring with the adjacent phenyl group (distance be-
tween centroids 3.747 ä; Figure 5). The electron-withdraw-
ꢀ
ꢀ
ing tosyl group in 9a’ causes the N2 C7 and N3 C7 distan-
ces to be different (1.281(8) and 1.427(8) ä, respectively),
while in 8a (R=Me) these bond lengths are similar
(1.293(9) and 1.330(8) ä). Correspondingly, the sum of the
bond angles about the imidazoline N atom N3 is 360.08 in
8a and 348.18 in 9a’. This electronic effect is very well re-
produced by the DFT calculations for complexes 6a and
10a.
In view of the large trans influence of the methyl group, it
is expected to be trans to the less basic ring (pyridine in 5a
Figure 4. ORTEP plot (40% probability thermal ellipsoids) of the molec-
ular structure of 8a.
Table 3. Selected bond lengths [ä] and angles [8] for 8a, 9a’, 6a and 10a.
ꢀ
other hand, the Pd N2 bond
8a
9a’
6a[a]
10a[a]
lengths of 2.021(5) ä in 8a and
2.013(5) ä in 9a’ seem to be
slightly influenced by the dif-
ferent electronic properties of
the R group in the imidazoline
ring. The calculated values in
Table 3 compare fairly well
X=C1
X=Cl2
X=C1
X=C1
ꢀ
Pd N1
ꢀ
Pd N2
2.121(6)
2.049(5)
2.168
2.030
2.301
2.033
1.307
1.474
1.382
1.488
2.149
2.017
2.292
2.040
1.308
1.467
1.431
1.509
2.021(5)
2.303(2)
2.181(4)
1.293(9)
1.462(8)
1.330(8)
1.467(9)
1.477(9)
2.013(5)
2.289(2)
2.277(2)
1.281(8)
1.444(8)
1.427(8)
1.480(8)
ꢀ
Pd Cl1
ꢀ
Pd X
ꢀ
N2 C7
ꢀ
N2 C9
ꢀ
N3 C7
ꢀ
ꢀ
with the X-ray values. The Pd
Cl and Pd C coordination dis-
N3 C8
ꢀ
N3 C22
ꢀ
ꢀ
N3 S1
1.693(5)
79.9(2)
1.686
76.5
97.1
tances fall in a range that has
often been observed in other
PdII complexes.[34,38] The che-
lating ligand, the small N1-C6-
C7-N2 torsion angles of 5.2(8)8
in 8a and 2.1(9)8 in 9a’ indi-
cate negligible tilting between
the rings. Similar torsion
angles of 0.78 in 6a and 8.08 in
10a were found for the calcu-
N1-Pd-N2
78.4(2)
96.00(16)
171.9(2)
173.22(16)
94.5(2)
77.5
96.1
N1-Pd-Cl1
N1-Pd-X
N2-Pd-Cl1
N2-Pd-X
Cl1-Pd-X
N3-S1-C22
N1-C6-C7-N2
C16-C8-C9-C10
dihedral angle py/im
95.41(16)
172.57(15)
175.15(14)
92.79(15)
91.81(6)
108.2(3)
2.1(9)
172.6
173.4
95.3
172.7
173.4
97.2
91.30(12)
91.1
89.3
5.2(8)
125.8(6)
2.6(4)
ꢀ0.7
134.7
3.0
8.0
114.8
7.6
139.7(6)
17.9(3)
[a] QM/MM calculations.
3750
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 3747 3760

Pd Complexes of Pyridine Imidazoline Ligands
3747 3760
8a, and imidazoline in 9a and 10a). Because we were in-
trigued by the cis stereochemistry of all the complexes and
wished to know whether an isomeric equilibrium–not de-
tected by low-temperature NMR spectroscopy–was pres-
ent, we performed DFT calculations of some representative
examples. One reason for the cis stereochemistry observed
for all the neutral complexes 1a 10a was found to be the
relative energy of the cis and trans isomers, calculated for
complexes 1a, 4a, 6a, 10a (Table 4). The cis isomer was the
most stable in all cases. Note that the relative stability of the
cis/trans isomers depends on the electronic effect exerted by
the R substituent. The relative stability decreases when the
withdrawing character of R increases (4a and 10a vs 1a and
6a). However, the stereochemistry of the imidazolines (R,S
for complexes 1a, 4a; R,R for complexes 6a, 10a) was
found to have no steric effect.
were determined by means of DFT calculations. No great
differences were found between the molecular structures of
the cationic complexes and those of the neutral complexes
described above.[39] The most significant feature is found in
ꢀ
ꢀ
the N2 C7 and N3 C7 bond lengths (see Figure 4 for num-
bering). Both bond lengths are sensitive to the R substitu-
ents, as was found for the neutral complexes. The values in-
ꢀ
ꢀ
dicate that when R=H, the N2 C7 and N3 C7 distances
are similar because of electronic delocalization through the
amidine fragment (e.g., for cis-1b: 1.310 vs 1.306 ä, respec-
ꢀ
tively). When R=Tf, the N2 C7 distance is shorter than the
ꢀ
N3 C7 distance (e.g., for cis-4b: 1.306 vs 1.407 ä, respec-
tively). The same effect can be observed in the case of 6b
and 10b. However, these bond lengths are not as different
in the cationic complexes as in the neutral complexes.
Therefore, in the cationic complexes conjugation between
the imine and amine moieties in the imidazoline moiety is
more pronounced than in the neutral complexes. This is in
agreement with the fact that the electrophilicity is expected
to be higher for a cationic complex than for a neutral one.
The relative energy of the cis/trans isomers was also calcu-
lated for complexes 1b, 4b, 6b and 10b (Table 5). First, we
Table 4. Relative stabilities of cis/trans isomers of complexes 1a/6a and
4a/10a [energies in kJmolꢀ1].
Stereoisomer
R=H
R=Tf
1a
0.0
14.6
6a
0.0
12.1
4a
10a
cis
trans
0.0
2.9
0.0
6.3
Table 5. Relative stabilities of complexes 1b/6b and 4b/10b [energies in
kJmolꢀ1].
Synthesis and characterization and DFT calculations of cat-
Stereoisomer
R=H
R=Tf
1b
6b
4b
10b
ionic palladium(ii) complexes: The neutral complexes were
treated with NaBAr04 in the presence of acetonitrile to
cis
trans
0.0
8.8
0.0
5.0
0.4
0.0
0.0
2.5
0
ꢀ
obtain the cationic complexes [PdMe(NCMe)(N N’)]BAr4
(5b 10b; Scheme 2). They were completely characterized in
solution by NMR spectroscopy. The most significant signals
6
ꢀ
are those related to H for the ligand and to the Pd Me and
ꢀ
observed that for the cationic complexes the difference in
the relative energy of the stereoisomers is lower than that
calculated for the neutral complexes (Table 4). The cis
isomer was the most stable isomer for complexes 1b, 6b,
and 10b, while for 4b the two isomers were almost degener-
ate in energy and the trans form was slightly more stable.
Although these small energy differences may not be signifi-
cant, we note that they are in good agreement with what we
observed experimentally (see Scheme 2).
To sum up, for the cationic complexes 1b 5b, which bear
the (R,S)-pyridine imidazoline ligands (4’,5’-cis stereochem-
istry), the coordination of the Me group to palladium is de-
termined by the electronic nature of the R substituents in
the imidazoline. Consequently the methyl group is always
trans to the less basic ring (pyridine for complexes 1b, 2b,
5b, and imidazoline for 3b and 4b). However, complexes
6b 10b, which bear the (R,R)-ligands (4’,5’-trans stereo-
chemistry), are always present as cis stereoisomers (the
single species for complexes 6b 8b, and the major one in a
mixture of isomers for 9b, 10b). Therefore, in the latter
case, both the electronic nature and the stereochemistry of
the ligand must be responsible for determining the stereo-
chemistry.
Pd NCMe fragments (Table 2). For complexes 5b 8b,
which contain electron-donating substituents, one set of sig-
nals is evident at room temperature. All signals are shifted
with respect to the free ligand, which indicates that coordi-
nation has taken place. For complexes 6b 8b, the signals of
the methyl and acetonitrile ligands coordinated to palladium
appear as singlets between 0.48 and 0.56, and 2.28 and
2.30 ppm, respectively (Table 2). These shifts are indicative
of cis stereoisomers.[34] In the case of 5b, the Pd Me signal
ꢀ
is subject to the effect of the cyano-substituted pyridine ring
in the trans position and appears further downfield
(0.90 ppm). By irradiating the signal of the methyl group
bound to palladium, an NOE interaction with H4’ of the imi-
dazoline ring became evident. For complexes 5b 8b, this
confirmed the presence of only one species in solution,
ꢀ
whose Pd Me bond is cis to the imidazoline ring
(Scheme 2).
In contrast, 9b and 10b showed two sets of resonances in
solution at room temperature, in both the aromatic and the
aliphatic parts of the spectra (Table 2). COSY experiments
together with selective irradiation of the aromatic signals
confirmed the presence of cis/trans stereoisomers in ratios
of 3:1 for 9b and 2:1 for 10b.
Although crystals suitable for structure determination
were not obtained for the cationic complexes, the structural
parameters for the cis and trans isomers of the cationic com-
plexes 1b, 4b, 6b, and 10b (as representative examples)
These small energy differences between the calculated
values for cis/trans isomers (Table 5) suggest that the two
isomers are in equilibrium. The sharpness of the NMR sig-
nals seems to exclude the presence of an equilibrium,[40] so
we considered that the stereochemistry of the complexes
Chem. Eur. J. 2004, 10, 3747 3760
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3751

C. Claver, A. Bastero et al.
FULL PAPER
should be determined during their formation. Although a
full mechanistic study is beyond the scope of this paper, we
attempted to locate possible intermediates along the reac-
tion path to obtain some insight into the formation of the
cationic palladium stereoisomers (Scheme 2). The first ques-
tion that arises at this point is whether the mechanism that
forms these cationic complexes is associative or dissociative.
Several attempts to determine the geometry of a five-coor-
dinate intermediate (associative path), either a square-based
pyramid or trigonal bipyramid (see Scheme 3), failed and
led to ligand dissociation. We therefore concluded that an
associative mechanism is not plausible.
acetonitrile provides different possibilities. To test some of
them, we considered complexes B and C (Figure 6), which
should involve ligand exchange (chloride by acetonitrile) in
an axial direction, perpendicular to the former square-
planar neutral complex. In both complexes we observed
square-planar coordination, with the pyridine (C) or imida-
zoline (B) moiety in the axial position. The energy differ-
ence between B and C is large enough to suggest that pyri-
dine dissociation is preferred. From complex C, the chloride
ligand should dissociate, and then coordination of the pyri-
dinic arm and reorganization leading to the square-planar
complex should follow. We propose that such intermediates
account for the structures obtained for the cationic com-
plexes. The combination of the electronic effects exerted by
these ligands with the steric hindrance that the phenyl rings
of the imidazoline moiety may impose on acetonitrile coor-
dination and on the reorganization step seems to be a plau-
sible explanation for the formation of only one isomer or
both isomers.
Synthesis and structure of cationic rhodium complexes 1c
4c and 1d 4d: To obtain more information about the basici-
ties of pyridine imidazoline ligands 1 4 (Scheme 1), we syn-
thesised a series of bis-carbonyl rhodium complexes and
measured their CO frequencies by IR spectroscopy. The re-
action of ligands 1 4 with [Rh(cod)2]BF4 in dichloromethane
resulted in the displacement of one molecule of 1,5-cyclooc-
Scheme 3. Possible square-based pyramidal and trigonal-bipyramidal in-
termediates.
ꢀ
tadiene to afford the cationic complexes [Rh(cod)(N
N’)]BF4 (1c 4c). The 1H NMR spectra showed one set of
signals for the N ligand, but various broad signals for the co-
ordinated 1,5-cyclooctadiene (see Experimental Section).
Considering the C1 symmetry of the rhodium(i) cation, four
signals are expected for the olefinic protons, although flux-
ionality may restrict the number of signals observed. Inter-
estingly, the number of signals observed at room tempera-
ture for the olefinic protons depended on the substituent on
the N ligand. Low-temperature NMR spectroscopy showed
Hence, we considered dissociation of a ligand leading to a
three-coordinate species prior to acetonitrile coordination.
As was suggested in a recent study on similar pyridine pyra-
zole ligands,[41] we considered a T-shaped intermediate cor-
responding to dissociation of the pyridine moiety (A in
Figure 6). The energy required to form A from its neutral
counterpart is calculated to be 83.6 kJmolꢀ1, which is a rea-
sonable value for a barrier. The subsequent coordination of
Figure 6. Selected intermediates for the dissociative pathway. [energies in kJmolꢀ1].
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3752
www.chemeurj.org
Chem. Eur. J. 2004, 10, 3747 3760

Pd Complexes of Pyridine Imidazoline Ligands
3747 3760
Table 6. Selected bond lengths [ä] and angles [8] for 2c and 3c.[a]
2c 3c
the expected four signals for the four complexes. This gives
further proof that modifying the remote substituent R tunes
the electronic nature of the pyridine imidazolines.
Suitable crystals for X-ray diffraction were obtained for
complexes 2c and 3c. The X-ray structure shows the rhodi-
um atom with the expected square-planar coordination by
the N donors of the chelating pyridine imidazoline ligand
and the two double bonds of 1,5-cyclooctadiene. Figures 7
and 8 show perspective views of 2c and 3c, respectively. The
ꢀ
Rh N1
2.099(5)
2.078(5)
2.124(8)
2.139(7)
2.139(6)
2.147(6)
2.008
2.112(6)
2.090(6)
2.125(7)
2.131(7)
2.169(7)
2.152(7)
2.009
ꢀ
Rh N2
ꢀ
Rh C1
ꢀ
Rh C2
ꢀ
Rh C5
ꢀ
Rh C6
ꢀ
Rh C1m
ꢀ
Rh C5m
2.033
2.047
ꢀ
N2 C14
1.324(7)
1.484(7)
1.336(8)
1.483(8)
1.480(8)
1.283(8)
1.502(8)
1.387(9)
1.496(8)
ꢀ
N2 C16
ꢀ
N3 C14
ꢀ
N3 C15
ꢀ
N3 C29
ꢀ
N3 S1
1.692(5)
1.375(10)
1.363(11)
77.9(2)
87.32
ꢀ
C1 C2
1.393(10)
1.378(10)
78.2(2)
87.41
ꢀ
C5 C6
N1-Rh-N2
C1m-Rh-C5m
N3-C29-C30
112.0(5)
N3-S1-C29
105.7(3)
ꢀ16.5(9)
ꢀ26.9(9)
N1-C13-C14-N2
C23-C15-C16-C17
C29-N3-C15-C23
S1-N3-C15-C23
Dihedral angle py/im
ꢀ13.5(8)
ꢀ16.9(7)
62.8(7)
110.4(6)
14.9(4)
16.2(3)
ꢀ
ꢀ
[a] C1m and C5m are the midpoints of the C1 C2 and C5 C6 bonds, re-
spectively.
Figure 7. ORTEP plot (40% probability thermal ellipsoids) of the molec-
ular structure of the cation of 2c.
Moreover, the electronic effects exerted by the R group at
N3 are mainly evident in the imidazoline ring, rather than in
ꢀ
the metal coordination environment. In fact, in 2c the N2
ꢀ
C14 and N3 C14 bond lengths of 1.324(7) and 1.336(8) ä
are consistent with delocalization in the N2-C14-N3 frag-
ment, while the corresponding values in 3c (1.283(8),
1.387(9) ä), induced by the tosyl group, indicate quite a
short double bond. The degree of delocalization across the
amidine is confirmed by the sums of the bond angles about
N3 of 359.58 in 2c versus 349.18 in 3c.
The chelating ligands are not coplanar, and the distortions
are very similar to those found in the Pd complexes with cis-
disposed phenyl groups.[34] The N1-C13-C14-N2 torsion
angles are ꢀ13.5(8)8 and ꢀ16.5(9)8 in 2c and 3c, respective-
ly, and the phenyl groups on the imidazoline ring avoid an
eclipsed conformation through torsion angles C23-C15-C16-
C17 of ꢀ16.9(7)8 (2c) and ꢀ26.9(9)8 (3c).
Figure 8. ORTEP plot (40% probability thermal ellipsoids) of the molec-
ular structure of the cation of 3c.
The X-ray structural data show that 1) significant distor-
tions in the imidazoline ring are induced by cis-disposed
phenyl rings; 2) the N atom in the imidazoline ring bearing
a benzyl substituent is planar and shows delocalization in
the amidine fragment, as was found with the methyl sub-
stituent (see 8a, Figure 7); and 3) the R substituent influen-
ces the coordination of the 1,5-cyclooctadiene trans to the
pyridine imidazoline and not the direct coordination of the
imidazoline to the rhodium.
ꢀ
Rh C bonds (Table 6), which lie in a wide range (2.124(8)
ꢀ
2.169(7) ä), and the alkene C C bonds (av1.38 ä) agree
with those found in other Rh(cod) complexes.[42] Taking
C1m and C5m as the midpoints of the alkene C=C bonds,
ꢀ
the calculated distances trans to N2 and N1 are Rh C1m
ꢀ
2.008 and 2.009 ä, and Rh C5m 2.033 and 2.047 ä, for 2c
and 3c, respectively. The coordination mean plane N1-N2-
C1m-C5m forms an angle close to 888 with the plane calcu-
lated through the alkene C atoms.
Bubbling CO through the reddish solutions of the diolefin
complexes displaced the cyclooctadiene ligand and formed
yellow solutions of the corresponding bis-carbonyl com-
ꢀ
ꢀ
In both complexes the Rh N1(pyridine) bond length is
plexes [Rh(N N’)(CO)2]BF4 1d 4d (Scheme 4).[43] The
ꢀ
slightly longer than the Rh N2(imidazoline) bond length.
1H NMR spectra of the isolated complexes showed the dis-
Chem. Eur. J. 2004, 10, 3747 3760
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3753

C. Claver, A. Bastero et al.
FULL PAPER
Interestingly, complexes 6b
10b show higher stability in
solution during copolymeriza-
tion than the corresponding
1b 5b. However, the most sur-
prising feature of these cata-
lysts is their disparate produc-
tivity, given the similarity of
the N ligands. For catalysts
6b 10b, containing the (R,R)-
Scheme 4. Synthesis of the RhI complexes 1c 4c, 1d 4d.
pyridine imidazoline ligands,
higher productivities are ob-
appearance of the signals of 1,5-cyclooctadiene together
with a shift of the pyridine imidazoline signals.
tained with the less basic systems (9b and 10b are more pro-
ductive than 6b 8b). However, for catalysts 1b 4b, contain-
ing the (R,S)-ligands, the differences in productivity are less
significant. The high productivity observed for complex 9b,
which is an order of magnitude higher than that obtained
with other pyridine imidazoline-derived catalysts (entry 9 vs
1), is noteworthy. The amount of copolymer produced by
this system falls in the range of the productivities obtained
with the most active systems reported for CO/styrene co-
polymerization under mild conditions.[15,17]
The size of the polyketones obtained with the new precur-
sors are related to the productivity of the catalyst systems
(Table 9). Catalysts 9b and 10b yield polyketones with mo-
lecular weights up to Mw =76500 gmolꢀ1, while the more
basic catalysts lead to shorter polyketones (Mw ꢁ
20000 gmolꢀ1). The tacticity of the polyketones was ana-
lyzed by integrating the 13C NMR spectra in the
CH(Ph)CH2 region (Figure 9). The electronic nature of the
substituent in the imidazoline moiety in complexes 1b 4b,
influenced the stereoregularity of the polymers (see 1b vs
4b in Figure 9). Catalysts 1b and 2b, which show cis stereo-
chemistry, provided the higher percentage of l diads (up to
65%), while catalysts 3b and 4b yielded essentially syndio-
tactic copolymers. With catalysts 6b 10b this effect was less
significant, and the content of l diads of the polyketones
ranged between 23 and 37% (Table 9). Therefore, a prevail-
ing syndiotactic microstructure was obtained with precursors
6b 10b, as previously reported for a similar pyridine oxazo-
line ligand.[15,21]
Table 7 lists the CO stretching frequencies of the carbonyl
complexes. All the complexes show two bands, assigned to
cis-bis-carbonyl complexes.[43] The frequencies increase in
the order 1<2<3<4, that is, the basicity of the nitrogen li-
gands coordinated trans to the carbonyls decreases in this
order.
ꢀ
Table 7. Selected IR data for [Rh(CO)2(N N’)]BF4 complexes in di-
chloromethane.
Ligand
n(CO) [cmꢀ1
]
Ligand
n(CO) [cmꢀ1
]
1
3
2093, 2030
2104, 2045
2
4
2093, 2033
2106, 2050
Copolymerization of carbon monoxide and 4-tert-butylstyr-
ene: The copolymerization conditions for isolated com-
plexes 5b 10b were identical to those used for precatalysts
1b 4b,[28] as shown in Table 8. Complexes 6b 10b give rise
Table 8. CO/4-tert-butylstyrene copolymerization with complexes 1b
10b.[a]
Entry
Cat.
Productivity
Mw
(Mw/Mn)
% l
diads
gCP(gPd)ꢀ1 hꢀ1
1[b]
2
3
4
5
6
7
8
9
1b
2b
3b
4b
5b
6b
7b
8b
9b
10b
2
8.9
7
46400 (1.1)[c]
74600 (1.5)[c]
71100 (1.2)[c]
59600 (1.5)[c]
65
52
15
18.4
12.8
According to the results previously reported,[15,21] the syn-
diotacticity observed should be produced by chain-end con-
trol, which overcomes the enantiosite control created by the
chiral ligand. However since a different behavior is observed
for the catalysts containing 4’,5’-cis and -trans imidazolines,
we more closely examined the reactivity of the latter to-
wards CO.
3.4
4
5
27.2
14.6
20600 (1.2)[d]
n.d.[e]
37.3
26.4
34.5
30.2
23
17600 (1.3)[d]
76600 (1.4)[d]
52400 (2.0)[d]
10
[a] Reaction conditions: ncat =12.5 mmol; TBS/cat.=620; room tempera-
ture; 1 atm CO; solvent: 5 mL chlorobenzene; t=24 h. [b] ncat =8.3 mmol.
[b] Determined by SEC-MALLS in THF. [d] Determined by GPC in
THF relative to polystyrene standards. [e] n.d.=not determined.
Insertion of carbon monoxide into the palladium cationic
complexes: Cationic complexes 6b 10b were carbonylated
in CD2Cl2 solution by bubbling CO for five minutes at
273 K. Selected 1H and 13C NMR data (Table 10) indicate
to efficient catalysts, while 5b is inactive. The presence of
the cyano substituent ortho to the coordinating N atom in
ligand 5 (Scheme 1), seems to cause enough sterical hin-
drance to interfere with the chain-growth sequence. A simi-
lar behavior was reported for Pd complexes containing 2,9-
substituted phenanthrolines and 6-substituted bipyridines.[44]
selective
formation
of
acyl
carbonyl
complexes
0
ꢀ
[Pd(COMe)(CO)(N N’)]BAr (11 15), which result from in-
4ꢀ
sertion of CO into the Pd Me bond (Scheme 5). The
1H NMR spectra at 273 K show two new signals (e.g., for
7b+CO: d=1.72, 1.97 ppm) in the aliphatic region corre-
sponding to Pd COMe and acetonitrile, respectively. The
ꢀ
3754
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 3747 3760

Pd Complexes of Pyridine Imidazoline Ligands
3747 3760
1
Table 9. Selected H and 13C NMR resonances for the reaction of complexes 6b 10b with 13CO.[a]
Variable-temperature NMR
experiments were performed
in the range 263 183 K for all
acyl carbonyl complexes. Com-
plexes 11 14 showed sharp sig-
nals over the entire tempera-
ture range, that is, they are
single isomers. In the case of
15, the broad signal at
2.05 ppm corresponding to the
acyl group disappeared at
233 K. Two doublets became
evident at 183 K (at 1.56 and
2.69 ppm; major and minor iso-
mers of 15, respectively). In
the 13C NMR spectrum the ini-
tial two signals at 273 K (at
173.2 and 207.2 ppm) became
four at 183 K (172.3 and 208.2
for 15 major, 170.3 and 212.9
for 15 minor), as expected for
the presence of two acyl car-
bonyl species. This indicates
the presence of cis/trans equi-
librium for acyl carbonyl com-
plex 15 (Scheme 5).
T
1H NMR
6b+CO
7b+CO
8b+CO
9b+CO
273
183
273
183
273
183
273
0.51
0.49
0.44
n.o.
n.o.
0.86
0.90
n.o.
1.72
1.53
1.72
1.55
1.69
1.52
1.62
n.o.
1.52 (d)
1.46 (d)
M: 0.52
m: 1.04
183
273
1.36
2.05 (br)
10b+CO
M: 0.72
m: 1.15
n.o.
n.o.
183
M: 1.56
m: 2.69 (br)
13C NMR
6b+CO
7b+CO
183
273
183
273
183
273
183
273
183
n.o.
n.o.
173.2, 211.6
174.0, 210.5
173.1, 213.3
174.1, 210.4
173.4, 212.7
173.2, 206.6
172.5, 208.2
173.2 (br), 207.2 (br)
M:172.3, 208.2
m: 170.3, 212.9
8b+CO
9b+CO
10b+CO
176.3
175.0
n.o.
219.9
215.4
n.o.
[a] NMR spectra recorded in CD2Cl2; n.o.: not observed. M: major isomer, m: minor isomer.
The intermediates of the mi-
gratory carbon monoxide inser-
tion were detected in two addi-
tional experiments with complexes 8b and 9b (Table 10).
After CO bubbling at 273 K the tube was carefully placed in
the probe without shaking, so that the carbon monoxide
could slowly diffuse into the solution. Three more signals in
1
the H NMR and two more in the 13C NMR spectrum were
observed corresponding to the two intermediate species,
which were unequivocally assigned in both cases: the methyl
carbonyl (e.g., for 8b+CO: d(1H)=0.86 ppm, d(13C)=
176.3 ppm) and acyl acetonitrile species (e.g., for 8b+CO:
d(1H)=1.52 and 2.35 ppm (due to coordinated acetonitrile),
d(13C)=219.9 ppm).[14]
The stereochemistry of the acyl carbonyl complexes could
not be unequivocally assigned by NOE difference experi-
ments for all the complexes, since in most of the cases inter-
actions were too weak at lower temperatures. Only in the
case of complex 12 could cis stereochemistry clearly be con-
firmed, on the basis of NOE interactions between the acyl
protons and H4’ of the imidazoline moiety. As in the case of
ꢀ
ꢀ
the cationic complexes 1b 10b, whose Pd Me and Pd
NCMe chemical shifts can be used to establish their stereo-
ꢀ
chemistry, the similarity of the Pd COMe shifts for com-
plexes 11 14 and 15 (major) suggest that they all have the
same stereochemistry.
Figure 9. Comparative 13C NMR spectra in the region of the methylene
carbon atom of copolymers obtained using the complexes indicated. The
reference is an epimerized copolymer.
Conclusion
13C NMR spectra at 273 K show two signals (e.g., for 7b+
Modular pyridine imidazoline ligands 1 10 make it possible
ꢀ
ꢀ
CO: d=174 and 210.5 ppm) corresponding to Pd CO and
to study how the structure of N N’ ligands influence palladi-
[14]
ꢀ
Pd COMe, respectively.
um-catalyzed CO/styrene copolymerization. The basicity of
Chem. Eur. J. 2004, 10, 3747 3760
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3755

C. Claver, A. Bastero et al.
FULL PAPER
Table 10. Crystal data and details of structure refinement for compounds 2c, 3c, 8a, and 9a’.
ionic complexes [PdMe(NCMe)-
0
ꢀ
(N N’)]BAr4 depends on the
compound
2c
3c
8a
9a’¥CH2Cl2
pyridine imidazoline ligands.
Cationic complexes with more
basic imidazolines (R=H, Bn,
Me) are cis isomers, but com-
plexes bearing less basic li-
gands (R=Ts, Tf) can be pres-
ent as single trans isomers or
as mixtures of isomers, de-
pending on the stereochemistry
of the imidazoline moiety. This
is confirmed by DFT calcula-
tions, which show that the elec-
tronic effect of the R substitu-
ent has an influence on the rel-
ative stability of the cationic
cis and trans isomers; this de-
creases when the electron-
withdrawing character of R in-
creases.
formula
C35H35BF4N3Rh
687.38
C35H35BF4N3O2RhS
751.44
C22H22ClN3Pd
470.28
C28H25Cl4N3O2PdS
715.77
Mr [gmolꢀ1
]
crystal system
space group
a [ä]
b [ä]
c [ä]
monoclinic
P21/n
11.068(3)
25.636(5)
11.581(4)
monoclinic
P21/c
15.071(3)
10.445(5)
20.830(4)
orthorhombic
P212121
11.335(3)
13.280(4)
13.327(4)
triclinic
P1
≈
9.481(3)
10.207(3)
16.059(4)
101.71(2)
95.82(2)
108.38(2)
1421.0(7)
2
1.673
1.135
150(2)
720
a [8]
b [8]
105.25(2)
99.86(2)
g [8]
V [ä3]
Z
3170.3(15)
4
1.440
0.591
293(2)
1408
3230.5(18)
4
1.545
0.654
293(2)
1536
2006.1(10)
4
1.557
1.069
150(2)
952
1calcd [gcmꢀ3
]
m(MoKa) [mmꢀ1
T [K]
]
F(000)
q range [8]
reflns collected
unique reflns
R(int)
2.27 26.02
9764
5285
0.0498
3202
397
1.98 25.02
10603
5604
0.0709
3483
2.36 27.10
4678
4252
0.0363
3681
246
2.24 27.10
9780
5805
0.0707
4239
353
observed reflns [I>2s(I)]
refined parameters
Flack parameter
GOF on F2
453
ꢀ0.04(6)
1.007
0.0594
0.1579
0.942, ꢀ0.449
1.038
0.0554
0.1402
0.614, ꢀ0.540
1.076
1.048
0.0635
Copolymerization data sug-
gest that the structural ligand
modifications mentioned above
influence the productivity of
the catalyst and the stereoregu-
larity of the polymer. Greater
R1 [I>2s(I)][a]
0.0539
0.1478
1.137,[b] ꢀ1.011
wR2 [I>2s(I)][a]
0.1914
residuals [eäꢀ3
]
1.407,[b] ꢀ1.043
[a] R1 =ꢀjjFo jꢀjFc jj/ꢀjFo j, wR2 =[ꢀw(F2oꢀFc2)2/ꢀw(Fo2)2]1/2. [b] Close to Pd ion.
stability
against
reduction
during polymerization is ob-
served for the systems contain-
ing the less distorted (R,R)-
4’,5’-trans imidazolines, al-
though this does not necessari-
ly lead to higher productivity.
Here the basicity of the imida-
zolines also plays an important
role: the less basic ligands lead
to the most active precursors,
probably due to favorable sty-
rene coordination and inser-
tion. The microstructure of the
polyketones obtained with the
complexes bearing (R,S)-4’,5’-
cis imidazolines depends on
the electronic nature of the
ligand. When (R,R)-4’,5’-trans
imidazolines are used, how-
ever, syndiotactic polyketones
Scheme 5. Carbonylation of the cationic palladium compounds 6b 10b. M: major isomer; m: minor isomer.
the ligands can be tailored by changing the nature of the
remote R substituents on the imidazoline moiety. Modifying
the ligand stereochemistry ((R,R)-4’,5’-trans instead of
(R,S)-4’,5’-cis) leads to imidazolines with smaller degrees of
distortion, as shown by X-ray diffraction. Both structural
changes are reflected in the coordination to palladium and
rhodium in solution and in the solid state. Density function-
al calculations on palladium complexes reproduced the ex-
perimental observations well.
are always obtained, which indicates that chain-end control
is more effective than enantiosite control. The size of the
polymers does not seem to depend on the structural features
of the nitrogen ligands.
Experimental Section
General procedures: All reactions were carried out under nitrogen at-
mosphere at room temperature by using standard Schlenk techniques.
Solvents for synthetic purposes were distilled under nitrogen. Solvents
ꢀ
While all the neutral [PdClMe(N N’)] complexes are ob-
tained as single cis isomers, the stereochemistry of the cat-
3756
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 3747 3760

Pd Complexes of Pyridine Imidazoline Ligands
3747 3760
for spectroscopy were used without further purification. Carbon monox-
ide (labeled and unlabeled, CP grade, 99%) was supplied by Aldrich.
[PdClMe(cod)],[35] NaBAr40 (Ar’=3,5-(CF3)2C6H3),[45] and [Rh(cod)2]-
for C21H19N3 (313.4): C 80.48, H 6.11, N 13.41; found: C 80.66, H 6.09, N
13.33.
Compound 9:
A solution of p-toluenesulfonyl chloride (75.7 mg,
[46]
BF4 were prepared according to reported methods. 1H and 13C NMR
0.4 mmol) was added dropwise to a solution of 6 (100 mg, 0.33 mmol)
and 4-(dimethylamino)pyridine (73.1 mg, 0.6 mmol) in dichloromethane
(3 mL) at 273 K. The reaction mixture was allowed to warm to room
temperature and stirred for 5 h. Evaporation of the mixture gave a
yellow solid that was purified by column chromatography to obtain a
white solid in 77% yield. Rf =0.54 (hexane/ethyl acetate 1/1); 1H NMR
(400 MHz, CDCl3, RT): d=8.62 (d, 3J=5 Hz, 1H; H6), 7.96 (d, 3J=
spectra were recorded on a Varian Gemini spectrometer with a 1H reso-
nance frequency of 300 MHz and a 13C frequency of 75.4 MHz and on a
Varian Mercury VX spectrometer with a 1H resonance frequency of
400 MHz and a 13C frequency of 100.5 MHz. The resonances were refer-
1
enced to the solvent peak versus TMS (CDCl3 at d=7.26 ppm for H and
1
d=77.23 ppm for 13C, CD2Cl2 at d=5.32 ppm for H and d=54.0 ppm for
3
3
7.4 Hz, 1H; H3), 7.84 (t, J=7.4 Hz, 1H; H4), 7.43 (dd, 3J=7.4, J=5 Hz,
1H; H5), 7.39 7.09 (m, 14H; Ph), 5.36 (d, 3J=4.8 Hz, 1H; H4’ or H5’),
5.17 (d, 3J=4.8 Hz, 1H; H5’ or H4’), 2.39 ppm (s, 3H; CH3); 13C NMR
(100.5 MHz, CDCl3, RT): d=158.5 (s, C2), 148.6 (s, C6), 136.5 (s, C4),
129.1 126.4 (Ph), 125.3 (s, C5), 124.9 (s, C3), 78.6 (s, C4’ or C5’), 72.1 (s, C5’
or C4’), 21.9 ppm (s, CH3); elemental analysis calcd (%) for C27H23N3O2S
(453.6): C 71.50, H 5.11, N 9.26; found: C 70.88, H 5.24, N 9.24.
1
13C). The NOE experiments were run with a H pulse of 12 ms (300 MHz)
or 13.3 ms (400 MHz). Two-dimensional correlation spectra (gCOSY)
were obtained with the automatic program of the instrument. Elemental
analyses were carried out on a Carlo Erba Microanalyser EA 1108. MS
(FAB+) were obtained on a Fisons V6-Quattro instrument. The molecu-
lar weights and molecular weight distributions of the copolymers were
determined by size-exclusion chromatography on a Waters 515-GPC
device using a lineal Waters Ultrastyragel column with a Waters 2410 re-
fractive index detector versus polystyrene standards.
Compound 10: Similar to the synthesis of 9 but with trifluoromethanesul-
fonic anhydride as the electrophile. Purification was performed by
column chromatography with ethyl acetate as eluent. Rf =0.89; 1H NMR
(400 MHz, CDCl3, RT): d=8.71 (dd, 3J=4.8, 4J=1.7 Hz, 1H; H6), 8.01
(d, 3J=7.7 Hz, 1H; H3), 7.85 (td, 3J=7.7, 4J=1.7 Hz, 1H; H4), 7.48 7.31
Synthesis of ligands
Compound 5: 2,6-Dicyanopyridine (100 mg, 0.77 mmol) was treated with
meso-1,2-diphenylethylenediamine (164 mg, 0.77 mmol) in chlorobenzene
(5 mL) in the presence of Yb(OTf)3 (46 mg, 0.14 mmol). The mixture was
stirred for 24 h under reflux. The desired product was separated from the
bis-imidazoline product[30] by column chromatography in 20% yield. Rf =
3
3
(m, 11H; H5, 2Ph), 5.45 (d, J=3.8 Hz, 1H; H4’ or H5’), 5.37 ppm (d, J=
3.8 Hz, 1H; H5’ or H4’); 13C NMR (100.5 MHz, CDCl3, RT): 156.0 (s, C2),
148.8 (s, C6), 147.9 (s, C2’), 139.9 (s, Ph), 139.4 (s, Ph), 136.7 (s, C4), 129.3
126.2 (Ph), 125.8 (s, C5), 124.5 (s, C3), 78.9 (s, C5’ or C4’), 72.8 ppm (s, C4’
or C5’); elemental analysis calcd (%) for C21H16N3F3O2S (431.4): C 58.46,
H 3.74, N 9.74; found: C 58.52, H 3.70, N 9.69%.
1
0.41 (hexane/ethyl acetate 1:2). H NMR (300 MHz, CDCl3, RT): d=8.58
(d, 3J=7.8 Hz, 1H; H5), 7.99 (t, 3J=7.8 Hz, 1H; H4), 7.83 (d, 3J=7.8 Hz,
1H; H3), 7.04 6.95 (m, 10H; Ph), 6.49 (s, 1H; NH), 5.58 ppm (br, 2H;
H4’, H5’); 13C NMR (75.4 MHz, CDCl3, RT): d=162.5 (s, C2), 150.0 (s,
C2’), 138.4 (s, C6), 130.2 (s, C3 C5), 130.0 (s, C3 C5), 127.9 127.1 (Ph),
126.2 (s, C3ꢀ5), 117.0 (s, CꢂN), 67.0 ppm (br, C4’, C5’); elemental analysis
calcd (%) for C21H16N4 (324.4): C 77.76, H 4.97, N 17.27; found: C 77.52,
H 4.93, N 17.21.
ꢀ
Synthesis of [PdClMe(N N’)] (5a 10a): The ligand (5 10) was added in
stoichiometric amount to a solution of [PdClMe(cod)] (50 mg, 0.3 mmol)
in toluene (5 mL). The solution was stirred at room temperature for 1 h
to give a yellow precipitate in 67% average yield. The precipitate was
then collected by filtration and washed with Et2O.
Compound 5a: 1H NMR (400 MHz, [D6]acetone, RT): 8.55 (d, 3J=
7.7 Hz, 1H; H5), 8.46 (s, 1H; NH), 8.32 (t, 3J=7.7 Hz, 1H; H4), 8.12 (d,
3J=7.7 Hz, 1H; H3), 7.26 6.85 (m, 10H; Ph), 5.72 (d, 3J=11.2 Hz, 1H;
Compound 6: 2-Cyanopyridine (500 mg, 2.36 mmol) was treated with
(1R,2R)-1,2-diphenylethylenediamine (228 mg, 2.21 mmol) in chloroben-
zene (10 mL) in the presence of Yb(OTf)3 (50 mg, 0.16 mmol). The mix-
ture was stirred for 72 h under reflux. The resulting mixture was evapo-
rated to dryness, dissolved in CH2Cl2, and washed with three portions of
H2O (15 mL). The organic layers were extracted, dried over MgSO4, and
evaporated to give a light-colored solid. Recrystallization from CH2Cl2/
3
H5’), 5.53 (d, J=11.2 Hz, 1H; H4’), 0.72 ppm (s, 3H; PdCH3).
Compound 6a: 1H NMR (400 MHz, CDCl3, RT): d=8.82 (d, 3J=4.7 Hz,
3
3
1H; H6), 8.60 (s, 1H; NH), 8.41 (d, J=7.8 Hz, 1H; H3), 7.75 (td, J=7.8,
4J=1.9 Hz, 1H; H4), 7.41 (dd, J=7.8, J=4.7 Hz, 1H; H5), 7.28 7.16 (m,
10H; Ph), 4.90 (d, 3J=6.8 Hz, 1H; H4’), 4.83 ppm (d, 3J=6.8 Hz, 1H;
H5’), 0.34 (s, 1H; PdCH3); 13C NMR (100.5 MHz, CDCl3, RT): d=149 (s,
C6), 138.6 (s, C4), 129.2 126.2 (Ph), 128.4 (s, C5), 124.3 (s, C3), 76.6 (s, C4’),
70.3 (s, C5’), ꢀ8.6 ppm (s, PdCH3); elemental analysis calcd (%) for
C21H20N3ClPd (456.3): C 55.28, H 4.42, N 9.21; found: C 55.59, H 4.47, N
9.30.
3
3
1
hexane afforded white crystals in 84% yield. H NMR (300 MHz, CDCl3,
3
3
RT): d=8.64 (d, J=5.4 Hz, 1H; H6), 8.35 (d, J=8 Hz, 1H; H3), 7.85 (t,
3J=8 Hz, 1H; H4), 7.44 (dd, 3J=8, 3J=5.4 Hz, 1H; H5), 7.36 7.31 (m,
10H; Ph), 5 (s, 2H; H4’, H5’); 13C NMR (75.4 MHz, CDCl3, RT): d=162.7
(s, C2), 149 (s, C6), 148.5 (s, C2’), 143.3 (s, Ph), 136.9 (s, C4), 128.9 126.8
(Ph), 125.6 (s, C5), 123 (s, C3), 75.6 (s, C4’, C5’); elemental analysis calcd
(%) for C20H17N3 (299.4): C 80.24, H 5.72, N 14.04; found: C 80.09, H
5.66, N 13.97.
Compound 7a: 1H NMR (400 MHz, CDCl3, RT): d=9.26 (d, 3J=5 Hz,
1H; H6), 7.87 (t, 3J=7.8 Hz, 1H; H4), 7.76 (d, 3J=7.8 Hz, 1H; H3), 7.63
(dd, 3J=7.8, 3J=5 Hz, 1H; H5), 5.15 (d, 3J=6.4 Hz, 1H; H4’), 5.02 (d,
2J=17.2 Hz, 1H; CH2), 4.64 (d, 3J=6.4 Hz, 1H; H5’), 4.42 (d, 2J=
17.2 Hz, 1H; CH2), 0.52 ppm (s, 3H; PdCH3); 13C NMR (100.5 MHz,
CDCl3, RT): 150.8 (s, C6), 138.3 (s, C4), 129.7 129.1 (Ph), 128.4 (s, C5),
127.2 126 (Ph), 123.5 (s, C3), 76.8 (s, C5’), 74.3 (s, C4’), 50.4 (s, CH2),
ꢀ3.4 ppm (s, PdCH3); elemental analysis calcd (%) for C28H26N3ClPd
(546.4): C 61.55, H 4.80, N 7.70; found: C 60.50, H 4.73, N 7.48.
Compound 8a: 1H NMR (400 MHz, CDCl3, RT): d=9.28 (d, 3J=4.8 Hz,
1H; H6), 8.03 (m, 2H; H3, H4), 7.67 (q, 3J=4.8 Hz, 1H; H5), 7.46- 7.25
(m, 10H; Ph), 5.04 (d, 3J=7 Hz, 1H; H4’), 4.58 (d, 3J=7 Hz, 1H; H5’),
3.26 (s, 3H; NCH3), 0.48 (s, 3H; PdCH3); 13C NMR (100.5 MHz, CDCl3,
RT): d=150.8 (s, C6), 138.1 (s, C4), 129.7 (s, Ph), 129.3 (s, C5), 129.0 (s,
Ph), 128.1 (s, Ph), 126.8 (s, Ph), 126.2 (s, Ph), 123.5 (s, C3), 79.5 (s, C4’ or
C5’), 73.9 (s, C5’ or C4’), 35.5 (s, CH3), ꢀ7.1 ppm (s, PdCH3); elemental
analysis calcd (%) for C22H22N3ClPd (470.3): C 56.18, H 4.72, N 8.93;
found: C 56.30, H 5.01, N 8.87.
Compound 7: Compound 6 (100 mg, 0.33 mmol) was dissolved in THF
(3 mL) and treated with NaH (9.5 mg, 0.4 mmol) for 1 h. Benzyl bromide
(42.5 mL, 0.36 mmol) was added dropwise to the reaction mixture at
room temperature. After 5 h, evaporation gave a brown paste which was
purified by column chromatography to yield a white solid in 71% yield.
1
Rf =0.10 (hexane/ethyl acetate 1:1). H NMR (400 MHz, CDCl3, RT): d=
8.72 (ddd, 3J=5.5, 4J=1.5, 5J=1.3 Hz, 1H; H6), 8.17 (dt, 3J=8, 4J=
1.3 Hz, 1H; H3), 7.83 (ddd, 3J=8, 3J=5.5, 4J=1.5 Hz, 1H; H4), 7.4 (m,
3
1H; H5), 7.35- 6.97 (m, 15H; Ph), 5.63 (d, J=15.6 Hz, 1H; CH2), 5.0 (d,
3
3J=9.6 Hz, 1H; H4’ or H5’), 4.42 (d, J=9.6 Hz, 1H; H5’ or H4’), 3.95 ppm
(d, 3J=15.6 Hz, 1H; CH2); 13C NMR (100.5 MHz, CDCl3, RT): 148.9 (s,
C6), 137.1 (s, C4), 129.1 127.2 (Ph), 125.4 (s, C3 or C5), 124.9 (s, C5 or C3),
77.9 (s, C4’ or C5’), 73.6 (s, C5’ or C4’), 49.1 ppm (s, CH2); elemental analy-
sis calcd (%) for C27H23N3 (389.5): C 83.26, H 5.95, N 10.79; found: C
83.02, H 5.92, N 10.74.
Compound 9a: 1H NMR (400 MHz, CDCl3, RT): d=9.24 (d, 3J=4 Hz,
1H; H6), 8.72 (d, 3J=8 Hz, 1H; H3), 8.09 (td, 3J=8, 4J=1.6 Hz 1H; H4),
7.80 (m, 1H; H5), 7.49 6.91 (m, 14H; Ph), 5.31 (d, 3J=5.6 Hz, 1H; H4’),
5.19 (d, 3J=5.6 Hz, 1H; H5’), 2.44 (s, 3H; CH3), 0.55 ppm (s, 3H;
PdCH3); 13C NMR (100.5 MHz, CDCl3, RT): d=150.4 (s, C6), 138.3 (s,
C4), 130.6 125.5 (C3, C5, Ph), 74.7 (s, C4’ or C5’), 73.7 (s, C5’ or C4’), 22.1 (s,
Compound 8: This compound was prepared in a similar way to 7 but
with MeI as the electrophile, as reported for its enantiomer.[23] 1H NMR
(300 MHz, CDCl3, RT): d=8.59 (d, 3J=3.6 Hz, 1H; H6), 7.99 (d, 3J=
7.9 Hz, 1H; H3), 7.70 (td, 3J=7.9, 4J=1.5 Hz, 1H; H4), 7.28 7.15 (m,
3
3
11H; H5, 2Ph), 4.86 (d, J=10.5 Hz, 1H; H4’ or H5’), 4.25 (d, J=10.5 Hz,
1H; H4’ or H5’), 2.88 ppm (s, 3H; CH3N); elemental analysis calcd (%)
Chem. Eur. J. 2004, 10, 3747 3760
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3757

C. Claver, A. Bastero et al.
FULL PAPER
CH3), ꢀ4.9 ppm (s, PdCH3); elemental analysis calcd (%) for
C28H26N3ClO2PdS (610.5): C 55.09, H 4.29, N 6.88; found: C 55.17, H
4.32, N 6.74.
1H; H3), 8.50 (d, 3J=4.4 Hz, 1H; H3), 8.17 (td, 3J=8.2, 4J=1.6 Hz, 1H;
H4), 7.71 (s, 9H; H5, 8Hb), 7.52 (s, 4H; Hd), 7.50 6.80 (m, 14H; Ph), 5.24
(d, 3J=4.8 Hz, 1H; H5’), 5.08 (d, 3J=4.8 Hz, 1H; H4’), 2.45 (s, 3H;
CH3Ts), 1.61 (s, 3H; PdNCCH3), 0.97 ppm (s, 3H; PdCH3); 13C NMR
(100.5 MHz, CDCl3, RT): d=161.7 (q, 1JC-B =198.1 Hz, Ca), 147.2 (s, C6),
137.8 (s, C4), 134.8 (s, 8C, Cb), 130.8- 123.2 (C3, C4, Ph), 117.6 (s, 4C, Cd),
76.4 (s, C4’), 74.1 (s, C5’), 22.0 (s, CH3), 5.67 (s, PdCH3), 2.44 ppm (s,
PdNCCH3); elemental analysis calcd (%) for C62H41N4BClF24O2PdS
(1514.9): C 50.34, H 2.79, N 3.79; found: C 50.21, H 2.68, N 3.66.
1
3
Compound 10a: H NMR (400 MHz, CDCl3, RT): d=9.29 (d, J=4.7 Hz,
1H; H6), 8.22 (d, 3J=7.7 Hz, 1H; H3), 8.10 (td, 3J=7.7 Hz, 4J=1.7 Hz,
1H; H4), 7.84 (dd, 3J=7.7 Hz, 3J=4.7 Hz, 1H; H5), 7.47 7.35 (m, 10H;
Ph), 5.55 (d, J=1.8 Hz, 1H; H4’), 5.43 (d, J=1.8 Hz, 1H; H5’), 0.75 ppm
(s, 3H; PdCH3); 13C NMR (100.5 MHz, CDCl3, RT): d=150.5 (s, C6),
138.4 (s, C4), 130.1 (s, C5), 130.0 (s, Ph), 129.6 (s, Ph), 129.5 (s, C3), 125.8
(s, Ph), 125.4 (s, Ph), 75.8 (s, C4’ or C5’), 75.5 (s, C5’ or C4’), ꢀ3.7 ppm (s,
PdCH3); elemental analysis calcd (%) for C22H19N3ClF3O2PdS (588.3): C
44.91, H 3.26, N 7.14; found: C 44.80, H 3.42, N 7.20.
3
3
Compound 10b: 1H NMR (400 MHz, CDCl3, RT): Ratio major/minor=
3
3
2:1; major: d=8.38 (d, J=5.2 Hz, 1H; H6), 8.28 (d, J=7.9 Hz, 1H; H3),
8.04 (td, 3J=7.9, 4J=1.6 Hz, 1H; H4), 7.70 (s, 8H; Hb), 7.51 (s, 4H; Hd),
7.48 7.14 (m, 11H; H5, 2Ph), 5.55 (d, 3J=2.2 Hz, 1H; H5’), 5.28 (d, 3J=
2.2 Hz, 1H; H4’), 2.26 (s, 3H; PdNCCH3), 0.74 ppm (s, 3H; PdCH3);
13C NMR (100.5 MHz, CDCl3, RT): d=161.7 (q, 1JC-B =198.1 Hz, Ca),
149.6 (s, C6), 140.1 (s, C4), 134.8 (s, Cb), 130.7 123.2 (C5, C3, Ce, Ph), 117.6
(s, Cd), 77.4 (s, C4’ or C5’), 75.5 (s, C5’ or C4’), 3.3 (s, PdCH3), 0.9 ppm (s,
PdNCCH3); minor: d=8.52 (d, 3J=5.8 Hz, 1H; H6), 8.29 (m, 1H; H3),
8.18 (td, 3J=7.9, 4J=1.3 Hz, 1H; H4), 7.74 (dd, 3J=7.9, 3J=5.8 Hz, 1H;
H5), 7.70 (s, 8H; Hb), 7.51 (s, 4H; Hd), 7.48 7.14 (m, 10H; Ph), 5.53 (d,
3J=3.4 Hz, 1H; H5’), 5.32 (d, 3J=3.4 Hz, 1H; H4’), 1.73 (s, 3H;
PdNCCH3), 1.07 ppm (s, 3H; PdCH3); 13C NMR (100.5 MHz, CDCl3,
RT): 161.7 (q, 1JC-B =198.1 Hz, Ca), 150.4 (s, C6), 140.4 (s, C4), 134.8 (s,
Cb), 130.7 123.2 (C5, C3, Ce, Ph), 117.6 (s, Cd), 75.7 (s, C4’ or C5’), 75.2 (s,
C4’ or C5’), 6.8 (s, PdCH3), 2.6 ppm (s, PdNCCH3); elemental analysis
calcd (%) for C56H34N4BF27O2PdS (1457.3): C 46.16, H 2.35, N 3.84;
found: C 46.36, H 2.60, N 3.10.
0
ꢀ
Synthesis of [PdMe(NCMe)(N N’)]BAr4 (5b 10b):
A stoichiometric
amount of NaBAr0 was added together with MeCN (0.5 mL) to a solu-
tion of [PdClMe(N N’)] (0.3 mmol) in CH2Cl2 (5 mL). The light yellow
solution that formed was stirred for 1 h, filtered through kieselghur, and
evaporated to dryness. The light yellow compounds were crystallized
from CH2Cl2/hexane in 76% average yield.
4ꢀ
Compound 5b: 1H NMR (400 MHz, CDCl3, RT): d=9.05 (s, 1H; NH),
7.91 (m, 2H; H4, H5), 7.79 (s, 3J=7.6 Hz, 1H; H3), 7.69 (s, 8H; Hb), 7.51
(s, 4H; Hd), 7.13 6.79 (m, 10H; Ph), 5.70 (d, 3J=11.2 Hz, 1H; H5’), 5.50
(d, 3J=11.2 Hz, 1H; H4’), 2.33 (s, 3H; PdNCCH3), 0.90 ppm (s, 3H;
PdCH3).
Compound 6b: 1H NMR (400 MHz, CDCl3, RT): d=8.34 (d, 3J=4.8 Hz,
1H; H6), 7.84 (t, 3J=8 Hz, 1H; H4), 7.70 (s, 8H; Hb), 7.64 (d, 3J=8 Hz,
1H; H3), 7.45 (s, 4H; Hd), 7.43 7.20 (m, 11H; H5, 2Ph), 6.31 (s, 1H;
3
3
NH), 5.05 (d, J=7 Hz, 1H; H4’), 4.96 (d, J=7 Hz, 1H; H5’), 2.30 (s, 3H;
PdNCCH3), 0.56 ppm (s, 3H; PdCH3); 13C NMR (100.5 MHz, CDCl3,
RT): d=161.6 (q, 1J (C-B)=197.2 Hz, Ca), 149.2 (s, C6), 139.9 (s, C4),
134.8 (s, Cb), 129.8 (s, Ph), 129.7 (s, Ph), 129.5 (s, Ph), 129.4 (s, Ph), 129.2
(m, Ce), 128.9 (s, C5), 126.1 (s, Ph), 123.3 (s, C3), 117.6 (s, Cd), 76.6 (s, C4’),
70.7 (s, C5’), 3.4 (s, PdNCCH3), ꢀ3.1 ppm (s, PdCH3); MS (FAB): m/z:
703.2 [MꢀMeꢀNCMe+6]2+, 404.1 [MꢀMeꢀNCMe]+, 298.1 [6]+; ele-
mental analysis calcd (%) for C55H35N4BF24Pd (1325.1): C 49.85, H 2.66,
N 4.23; found: C 50.02, H 2.60, N 4.14.
ꢀ
Synthesis of [Rh(cod)(N N’)]BF4: When the pyridine imidazoline li-
gands 1 4 (0.12 mmol) were added to a reddish solution of [Rh(cod)2]BF4
(0.12 mmol) in CH2Cl2 (2 mL), the color changed instantaneously. After
5 min, diethyl ether (5 mL) was added to precipitate complexes 1c 4c.
Average yield: 74%.
Compound 1c: 1H NMR (400 MHz, CDCl3, RT): d=8.57 (d, 3J=7.8 Hz,
1H; H3), 8.39 (s, 1H; NH), 8.22 (dd, 3J=7.8, 3J=6.9 Hz, 1H; H4), 7.78
(d, 3J=5.6 Hz, 1H; H6), 7.67 (dd, 3J=6.9, 3J=5.6 Hz, H5), 7.08 6.83 (m,
3
3
10H; Ph), 5.74 (d, J=11.7 Hz, 1H; H4’ or H5’), 5.25 (d, J=11.7 Hz, 1H;
H5’ or H4’), 4.44 (m, 1H; CH= cod), 4.17 (m, 2H; CH= cod), 3.51 (m,
1H; CH= cod), 2.51 2.37 (m, 4H; CH2 cod), 1.90 1.68 ppm (m, 4H; CH2
cod); elemental analysis calcd (%) for C28H29N3BF4Rh (597.4): C 56.29,
H 4.89, N 7.03; found: C 56.14, H 4.96, N 6.70.
Compound 7b: 1H NMR (400 MHz, CDCl3, RT): d=8.38 (d, 3J=4 Hz,
1H; H6), 7.84 (m, 2H; H4, H5), 7.70 (s, 8H; Hb), 7.51 (s, 4H; Hd), 7.45
7.02 (m, 16H; H3, 15Ph), 5.05 (d, 3J=5.6 Hz, 1H; H4’), 5.02 (d, 3J=
17.2 Hz, 1H; CH2), 4.72 (d, 3J=5.6 Hz, 1H; H5’), 4.46 (d, 3J=17.2 Hz,
1H; CH2), 2.28 (s, 3H; PdNCCH3), 0.54 ppm (s, 3H; PdCH3); 13C NMR
(100.5 MHz, CDCl3, RT): d=161.7 (q, 1JC-B =197.2 Hz, Ca), 149.6 (s, C6),
139.9 (s, C4), 134.8 (s, Cb), 130.0 (s, Ph), 129.6 (s, Ph), 129.5 (s, Ph), 129.0
(s, C5), 128.8 (m, Ce), 127.0 (s, Ph), 126.1 (s, Ph), 125.6 (Ph), 124.9 (s, C3),
117.6 (s, Cd), 76.8 (s, C4’ or C5’), 73.6 (s, C5’ or C4’), 50.1 (s, CH2), 3.4 (s,
PdNCCH3), ꢀ2.2 ppm (s, PdCH3); MS (FAB): m/z: 883.3
[MꢀMeꢀNCMe+7]2+, 494.1 [MꢀMeꢀNCMe]+, 390.2 [7]+; elemental
analysis calcd (%) for C62H41N4BF24Pd (1415.2): C 52.62, H 2.92, N 3.96;
found: C 52.35, H 3.00, N 3.83.
Compound 2c: 1H NMR (400 MHz, CDCl3, RT): d=8.37 (d, 3J=8 Hz,
1H; H3), 8.27(t, 3J=8 Hz, 1H; H4), 7.91 (d, 3J=5.5 Hz, 1H; H6), 7.79
(dd, 3J=8, 3J=5.5 Hz, 1H; H5), 5.47 (d, 3J=12 Hz, 1H; H4’ or H5’), 5.32
3
3
3
(d, J=17.2 Hz, 1H; CH2), 5.2 (d, J=12 Hz, 1H; H5’ or H4’), 4.55 (d, J=
17.2 Hz, 1H; CH2), 4.41 (m, 1H; CH= cod), 4.31 (m, 1H; CH= cod), 4.22
(m, 1H; CH= cod), 3.56 (m, 1H; CH= cod), 2.5 (m, 2H; CH2 cod), 2.3
(m, 2H; CH2 cod), 1.97 (m, 2H; CH2 cod), 1.8 ppm (m, 2H; CH2 cod);
elemental analysis calcd (%) for C35H35N3BF4Rh (687.6): C 61.14, H 5.13,
N 6.11; found: C 61.03, H 5.24, N 5.56.
Compound 8b: 1H NMR (400 MHz, CDCl3, RT): d=8.38 (d, 3J=4 Hz,
1H; H6), 8.03 (d, 3J=8 Hz, 1H; H3), 7.86 (t, 3J=8 Hz, 1H; H4), 7.69 (s,
8H; Hb), 7.5 (s, 4H; Hd), 7.46- 7.19 (m, 11H; H5, 2Ph), 4.92 (d, 3J=
7.2 Hz, 1H; H5’), 4.64 (d, 3J=7.2 Hz, 1H; H4’), 3.25 (s, 3H; NCH3), 2.30
(s, 3H; PdNCCH3), 0.48 ppm (s, 3H; PdCH3); 13C NMR (100.5 MHz,
Compound 3c: 1H NMR (300 MHz, CDCl3, RT): d=8.56 (d, 3J=8.1 Hz,
1H; H3), 8.34 (m, 1H; H4), 8.04 (m, 2H; H5, H6), 7.76 (d, 2J=8.3 Hz,
2
2H; ArH Ts), 7.45 (d, J=8.3 Hz, 2H; ArH Ts), 7.06 6.67 (m, 10H; Ph),
5.96 (d, 3J=9.5 Hz, 1H; H4’ or H5’), 5.37 (d, 3J=9.5 Hz, 1H; H5’ or H4’),
4.34 (m, 4H; CH= cod), 2.3 (m, 4H; CH2 cod), 1.81 ppm (m, 4H; CH2
cod); elemental analysis calcd (%) for C35H35N3BF4SO2Rh (751.6): C
55.9, H 3.19, N 5.59; found: C 55.75, H 3.78, N 5.58.
1
CDCl3, RT): d=161.6 (q, JC-B =197.2 Hz, Ca), 149.6 (s, C6), 139.7 (s, C4),
134.8 (s, Cb), 130.0 (s, Ph), 129.8 (s, Ph), 129.4 (s, Ph), 129.2 (m, Ce), 128.9
(s, Ph), 128.7 (s, C5), 126.6 (s, Ph), 125.9 (s, Ph), 124.7 (s, C3), 117.6 (s,
Cd), 79.3 (s, C4’), 73.5 (s, C5’), 35.1 (s, CH3N), 3.5 (s, 1C, PdNCCH3),
Compound 4c: 1H NMR (400 MHz, CDCl3, RT): d=8.30 (m, 2H; H6,
H3), 8.14 (m, 1H; H4 or H5), 8.10 (m, 1H; H5 or H4), 7.15 (m, 6H; Ph),
7.00 (br, 1H; Ph), 6.72 (m, 2H; Ph), 6.52 (br, 1H; Ph), 6.12 (d, 3J=
ꢀ2.5 ppm (s, 1C, PdCH3); MS (FAB): m/z: 731.2 [MꢀMeꢀNCMe+8]2+
,
418.1 [MꢀMeꢀNCMe]+, 314.2 [8]+; elemental analysis calcd (%) for
C56H37N4BF24Pd (1339.3): C 50.23, H 2.78, N 4.18; found: C 49.35, H 3.03,
N 3.99.
3
8.8 Hz, 1H; H4’ or H5’), 5.95 (d, J=8.8 Hz, 1H; H5’ or H4’), 4.56 (m, 2H;
CH= cod), 3.96 (m, 2H; CH= cod), 2.43 (m, 2H; CH2 cod), 2.31 (m, 2H;
CH2 cod), 1.85 ppm (m, 4H; CH2 cod).
Compound 9b: 1H NMR (400 MHz, CDCl3, RT): Ratio major/minor=
3:1; major: d=8.62 (d, 3J=8 Hz, 1H; H3), 8.35 (dd, 3J=5, 4J=1.3 Hz,
1H; H6), 8.05 (td, 3J=8, 4J=1.3 Hz, 1H; H4), 7.71 (s, 8H; Hb), 7.52 (s,
ꢀ
Synthesis of [Rh(CO)2(N N’)]BF4 (1d 4d): Bubbling carbon monoxide
3
5H; H5, 4Hd), 7.50 6.80 (m, 14H; Ph), 5.36 (d, J=3.2 Hz, 1H; H5’), 5.06
through solutions of 1c 4c (0.3 mmol) in CH2Cl2 (5 mL) led to the for-
mation of yellow solutions of the dicarbonyl complexes, which were pre-
cipitated by adding Et2O (5 mL).
(d, 3J=3.2 Hz, 1H; H4’), 2.41 (s, 3H; CH3Ts), 2.20 (s, 3H; PdNCCH3),
0.57 ppm (s, 3H; PdCH3); 13C NMR (100.5 MHz, CDCl3, RT): d=161.7
(q, 1JC-B =198.1 Hz, Ca), 149.3 (s, C6), 140.1 (s, C4), 134.8 (s, Cb), 130.8
123.2 (C3, C4, Ph), 117.6 (s, Cd), 74.3 (s, C5’), 74.1 (s, C4’), 22.0 (s, CH3), 3.2
(s, PdNCCH3), ꢀ0.03 ppm (s, PdCH3); minor: d=8.63 (d, 3J=8.2 Hz,
Compound 1d: 1H NMR (300 MHz, CDCl3, RT): d=9.08 (s, 1H; NH),
8.84 (d, 3J=8.4 Hz, 1H; H3), 8.65 (d, 3J=5.3 Hz, 1H; H6), 8.43 (dd, 3J=
8.4 Hz, 3J=7.1 Hz, 1H; H4), 7.82 (dd, 3J=7.1 Hz, 3J=5.3 Hz, 1H; H5),
3758
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 3747 3760

Pd Complexes of Pyridine Imidazoline Ligands
3747 3760
7.13- 6.92 (m, 10H; Ph), 5.83 (d, 3J=12.2 Hz, 1H; H4’ or H5’), 5.67 ppm
3
(d, J=12.2 Hz, 1H; H5’ or H4’).
[1] For recent reviews on late-transition metal catalyzed polymeriza-
tions, see a) S. D. Ittel, L. K. Johnson, M. Brookhart, Chem. Rev.
2000, 100, 1169 1203; b) S. Mecking, Coord. Chem. Rev. 2000, 203,
325 351.
[2] E. Drent, P. H. M. Budzelaar, Chem. Rev. 1996, 96, 663 681.
[3] K. Nozaki, T. Hiyama, J. Organomet. Chem. 1999, 576, 248 253.
[4] C. Bianchini, A. Meli, Coord. Chem. Rev. 2002, 225, 35.
[5] a) F. C. Rix, M. Brookhart, P. S. White, J. Am. Chem. Soc. 1996, 118,
4746 4764; b) C. S. Shultz, J. Ledford, J. M. DeSimone, M. Broo-
khart, J. Am. Chem. Soc. 2000, 122, 6351 6356; c) J. Ledford, C. S.
Shultz, D. P. Gates, P. S. White, J. M. DeSimone, M. Brookhart, Or-
ganometallics 2001, 20, 5266 5276.
Compound 2d: 1H NMR (400 MHz, CDCl3, RT): d=8.72 (d, 3J=4.8 Hz,
3
3
1H; H6), 8.44 (d, J=8 Hz, 1H; H3), 8.38 (t, J=8 Hz, 1H; H4), 7.82 (dd,
3J=8 Hz, 3J=4.8 Hz, 1H; H5), 7.37 6.89 (m, 15H; Ph), 5.72 (d, 3J=
12.4 Hz, 1H; H4’ or H5’), 5.58 (m, 2H; H5’ or H4’, CH2), 4.64 ppm (d, J=
3
17.2 Hz, 1H; CH2).
Compound 3d: 1H NMR (400 MHz, CDCl3, RT): d=8.86 (m, 2H; H6,
3
3
H3), 8.49 (t, J=8 Hz, 1H; H4), 8.04 (t, J=6.6 Hz, 1H; H5), 7.78 (d, 2J=
8.2 Hz, 2H; ArH Ts), 7.41 (d, 2J=8.2 Hz, 2H; ArH Ts), 7.16 6.74 (m,
10H; Ph), 5.98 (d, 3J=9.6 Hz, 1H; H4’ or H5’), 5.71 ppm (d, 3J=9.6 Hz,
1H; H5’ or H4’).
Compound 4d: 1H NMR (400 MHz, CDCl3, RT): d=8.92 (d, 3J=5.3 Hz,
1H; H6), 8.50 (d, 3J=7.8 Hz, 1H; H3), 8.42 (td, 3J=7.8, 4J=1.2 Hz, 1H;
H4), 8.09 (ddd, 3J=7.8, 3J=5.3 Hz, 4J=1.2 Hz, 1H; H5), 7.18 (m, 8H;
Ph), 6.84 (m, 2H; Ph), 5.58 ppm (m, 2H; H4’ and H5’).
[6] K. Nozaki, H. Komaki, Y. Kawashima, T. Hiyama, T. Matsubara, J.
Am. Chem. Soc. 2001, 123, 534 544.
[7] K. R. Reddy, K. Surekha, G. H. Lee, S. M. Peng, J. T. Chen, S. T.
Liu, Organometallics 2001, 20, 1292 1299.
X-ray crystallography: Crystal data and data-collection and refinement
parameters are summarized in Table 10. All data sets were colllected out
on a Nonius DIP-1030H system with graphite-monochromatized MoKa
radiation (l=0.71073 ä). For each compound a total of 30 frames were
collected with an exposure time of 12 20 min, rotation of 68 about f,
with the detector 90 mm from the crystal. Cell refinement, indexing, and
scaling of the data sets were carried out with the programs Mosflm and
Scala.[47]
All structures were solved by Patterson and Fourier analyses[48] and re-
fined by the full-matrix least-squares method based on F2 for all ob-
served reflections. The final cycles include the contribution of hydrogen
atoms at calculated positions. In 3c the BF4ꢀ anion was found to be disor-
[8] a) C. Carfagna, G. Gatti, D. Martini, C. Pettinari, Organometallics
2001, 20, 2175 2182; b) B. Binotti, C. Carfagna, G. Gatti, D. Marti-
ni, L. Mosca, C. Pettinari, Organometallics 2003, 22, 1115 1123.
[9] M. A. Zuideveld, P. C. J. Kamer, P. W. N. M. van Leeuwen, P. A. A.
Klusener, H. A. Stil, C. F. Roobeek, J. Am. Chem. Soc. 1998, 120,
7977 7978.
[10] G. K. Barlow, J. D. Boyle, N. A. Cooley, T. Ghaffar, D. F. Wass, Or-
ganometallics 2000, 19, 1470 1476.
[11] G. Consiglio, B. Milani in Catalytic Synthesis of Alkene Carbon
Monoxide Copolymers and Cooligomers (Ed.: A. Sen), Kluwer Aca-
demic, Dordrecht, 2003, pp. 189 215.
[12] P. Corradini, C. De Rosa, A. Panunzi, G. Petrucci, P. Pino, Chimia
1990, 44, 52 54.
[13] M. Barsacchi, G. Consiglio, L. Medici, G. Petrucci, U. W. Suter,
Angew. Chem. 1991, 103, 992; Angew. Chem. Int. Ed. Engl. 1991, 30,
989; an exception to this behavior can be found in J.-C. Yuan, S.-J.
Lu, Organometallics 2001, 20, 2697 2703.
[14] M. Brookhart, F. C. Rix, J. M. DeSimone, J. C. Barborak, J. Am.
Chem. Soc. 1992, 114, 5894 5895.
[15] M. Brookhart, M. I. Wagner, G. G. A. Balavoine, H. A. Haddou, J.
Am. Chem. Soc. 1994, 116, 3641 3642.
ꢀ
dered over two positions corresponding to rotation about a B F bond
with refined occupancies of 0.59(2)/0.41(2). A molecule of CH2Cl2 was
detected in the DF map of 9a’. All calculations were performed with the
WinGX System, Ver 1.64.02.[49]
CCDC 234657 234660 contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge via
www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cambridge Crys-
tallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK;
fax: (+44)1223-336-033; or deposit@ccdc.cam.uk).
[16] S. Bartolini, C. Carfagna, A. Musco, Macromol. Rapid Commun.
1995, 16, 9 14.
[17] M. T. Reetz, G. Haderlein, K. Angermund, J. Am. Chem. Soc. 2000,
122, 996 997.
[18] B. Milani, G. Corso, G. Mestroni, C. Carfagna, M. Formica, R. Sera-
glia, Organometallics 2000, 19, 3435 3441.
[19] B. Milani, A. Scarel, G. Mestroni, S. Gladiali, R. Taras, C. Carfagna,
L. Mosca, Organometallics 2002, 21, 1323 1325.
[20] A. Bastero, A. Ruiz, J. A. Reina, C. Claver, A. M. Guerrero, F. A.
JalÛn, B. R. Manzano, J. Organomet. Chem. 2001, 619, 287 292.
[21] A. Aeby, G. Consiglio, Inorg. Chim. Acta 1999, 296, 45 51, and ref-
erences therein.
[22] G. Helmchen, A. Pfaltz, Acc. Chem. Res. 2000, 33, 336, and referen-
ces therein.
[23] A. J. Davenport, D. L. Davies, J. Fawcett, D. R. Russell, J. Chem.
Soc. Perkin Trans. 1 2001, 1, 1500 1503.
[24] a) B. CÀ etinkaya, E. CÀ etinkaya, P. B. Hitchcock, M. F. Lappert, I. ÷z-
demir, J. Chem. Soc. Dalton Trans. 1997, 1359 1362; b) B. CÀ etin-
kaya, B. Alici, I. ÷zdemir, C. Bruneau, P. H. Dixneuf, J. Organomet.
Chem. 1999, 575, 187.
Computational details: DFT calculations were carried with the Amster-
dam density functional program (ADFv2000), developed by Baerends
et al.[50,51] The numerical integration scheme used in the calculations was
developed by te Velde et al.,[52, 53] and the geometry-optimization algo-
rithms were those implemented by Versluis and Ziegler.[54] A triple-zeta
plus polarization Slater-type basis set, as included in the ADF library, de-
scribed the electronic configurations of the molecular systems. The 1s 3d
electrons for Pd, the 1s electrons for C, O, N, and the 2p electrons for Cl
were treated as frozen cores. Geometry was fully optimized using nonlo-
cal corrections self-consistently. The DFT functional was the local VWN
exchange-correlation potential with Becke×s nonlocal exchange-correla-
tion corrections[55] and Perdew×s correlation corrections[56] (BP86). Rela-
tivistic effects were considered by using the scalar zeroth-order regular
approximation (ZORA).[57] No symmetry constraints were used. We
treated most atoms at this QM level, but used QM/MM calculations to
include only the phenyl substituents of the imidazoline group, which are
described in the MM partition. QM/MM calculations were performed by
applying the IMOMM method[58] as implemented in the ADF package.[59]
SYBIL[60] force field was used as implemented in ADF to describe the
atoms included in the MM part. For the palladium atom, we used UFF
parameters from the literature.[61] The ratio between the P C(aromatic)
ꢀ
[25] C. Botteghi, A. Schionato, G. Chelucci, H. Brunner, A. K¸rzinger,
U. Obermann, J. Organomet. Chem. 1989, 370, 17.
ꢀ
and the P H bond length was set to 1.234 ä.
[26] N. A. Boland, M. Casey, S. J. Hynes, J. W. Matthews, M. P. Smyth, J.
Org. Chem. 2002, 67, 3919 3922, and references therein.
[27] J. Elguero, E. Gonzalez, J. L. Imbach, R. Jacquier, Bull. Soc. Chim.
Fr. 1969, 11, 4075 4077.
[28] A. Bastero, S. CastillÛn, C. Claver, A. Ruiz, Eur. J. Inorg. Chem.
2001, 12, 3009 3011.
[29] a) A. Batistini, G. Consiglio, U. W. Suter, Angew. Chem. 1992, 104,
306; Angew. Chem. Int. Ed. Engl. 1992, 31, 303; b) B. Sesto, G. Con-
siglio, Chem. Commun. 2000, 1011 1012.
Acknowledgement
The Ministerio de Ciencia y TecnologÌa (BQU 2001-0656) is acknowl-
edged for financial support and for a grant (to A.B.). The ™Palladium∫
Network, HPRN-CT-2002-00196 and the COST D-17 Action are also ac-
knowledged.
[30] G. Chelucci, S. Deriu, G. A. Pinna, A. Saba, R. Valenti, Tetrahedron:
Asymmetry 1999, 10, 3803.
Chem. Eur. J. 2004, 10, 3747 3760
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3759

C. Claver, A. Bastero et al.
FULL PAPER
[31] C. Bianchini, H. M. Lee, A. Meli, W. Oberhauser, M. Peruzzini, F.
Vizza, Organometallics 2002, 21, 16 33.
[32] B. Sesto, G. Consiglio, J. Am. Chem. Soc. 2001, 123, 4097 4098.
[33] H. G¸nther, NMR Spectroscopy. Basic Principles, Concepts and Ap-
plications in Chemistry, Wiley, Chichester, 1995.
[45] S. R. Bahr, P. J. Boudjouk, J. Org. Chem. 1992, 57, 5545 5547.
[46] M. Green, T. A. Kuc, S. H. Taylor, J. Chem. Soc. A 1971, 2334.
[47] Collaborative Computational Project, Number 4. Acta Crystallogr.
Sect. D 1994, 50, 760 763.
[48] G. M. Sheldrick, SHELX97, Program for crystal structure refine-
ment, University of Gˆttingen, Germany, 1998.
[49] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837 838.
[50] E. J. Baerends, D. E. Ellis, P. Ros, Chem. Phys. 1973, 2, 41 51.
[51] C. Fonseca Guerra, J. G. Snijders, G. te Velde, E. Baerends, Theor.
Chem. Acc. 1998, 99, 391 403.
[52] G. T. Velde, E. J. Baerends, J. Comput. Phys. 1992, 99, 84 98.
[53] P. M. Boerrigter, G. T. Velde, E. J. Baerends, Int. J. Quantum Chem.
1988, 33, 87 113.
[34] A. Bastero, A. Ruiz, C. Claver, B. Milani, E. Zangrando, Organome-
tallics 2002, 21, 5820 5829.
[35] R. E. R¸lke, J. M. Ernsting, A. L. Spek, C. J. Elsevier, P. W. N. M.
van Leeuwen, K. Vrieze, Inorg. Chem. 1993, 32, 5769 5778.
[36] P. Pelagatti, M. Carcelli, F. Franchi, C. Pelizzi, A. Bacchi, H.-W. Fr¸-
hauf, K. Goubitz, K. Vrieze, Eur. J. Inorg. Chem. 2000, 463 475.
[37] A similar reaction was observed in the crystallization of complex
2a. See ref. [34].
[38] V. De Felice, V. G. Albano, C. Castellari, M. E. Cucciolito, A. De R-
enzi, J. Organomet. Chem. 1991, 403, 269.
[54] L. Versluis, T. Ziegler, J. Chem. Phys. 1988, 88, 322 328.
[55] A. Becke, Phys. Rev. A 1988, 38, 3098 3100.
[39] See Supporting Information.
[40] Low-temperature NMR studies confirmed the presence of a single
isomer for complexes 1b 8b.
[56] a) J. P. Perdew, Phys. Rev. B 1986, 34, 7406; b) J. P. Perdew, Phys.
Rev. B 1986, 34, 8822 8824.
[57] E. van Lenthe, A. E. Ehlers, E. J. Baerends, J. Chem. Phys. 1999,
110, 8943 8953, and references therein.
[58] F. Maseras, K. Morokuma, J. Comput. Chem. 1995, 16, 1170 1179.
[59] T. K. Woo, L. Cavallo, T. Ziegler, Theor. Chem. Acc. 1998, 100, 307
313.
[60] a) M. Clark, R. D. Cramer III, N. van Opdenbosch, J. Comput.
Chem. 1989, 10, 982 1012; b) U. C. Singh, P. A. Kollman, J.
Comput. Chem. 1986, 7, 718 730.
[41] J. Elguero, A. Guerrero, F. GÛmez de la Torre, A. de la Hoz, F. A.
JalÛn, B. R. Manzano, A. RodrÌguez, New J. Chem. 2001, 25, 1050
1060.
[42] a) S. W. Kaiser, R. B. Saillant, W. M. Butler, P. G. Rasmussen, Inorg.
Chem. 1976, 15, 2681 2688; b) H. Brunner, R. Storiko, F. Rominger,
Eur. J. Inorg. Chem. 1998, 6, 771 781; c) B. de Bruin, R. J. N. A. M.
Kicken, N. F. A. Suos, M. P. J. Donners, C. J. den Reijer, A. J.
Sandee, R. de Gelder, J. M. M. Smits, A. W. Gal, A. L. Spek, Eur. J.
Inorg. Chem. 1999, 9, 1581 1592.
[61] A. K. Rappÿ, C. J. Casewit, K. S. Colwell, W. A. Goddard III, W. M.
Shiff, J. Am. Chem. Soc. 1992, 114, 10024 10035.
[43] C. Claver, E. Marco, L. A. Oro, M. Royo, E. Pastor, Transition Met.
Chem. 1982, 7, 246.
[44] a) A. Sen, Z. Jiang, Macromolecules 1993, 26, 911 915; b) S. Stoc-
coro, G. Alesso, M. A. Cinellu, G. Minghetti, A. Zucca, A. Bastero,
C. Claver, M. Manassero, J. Organomet. Chem. 2002, 664, 77 84.
Received: December 19, 2003
Revised: April 5, 2004
Published online: June 15, 2004
3760
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 3747 3760