Pyridine-derived N-heterocyclic carbenes: An experimental and theoretical evaluation of the bonding in and reactivity of selected normal and abnormal complexes of nickel(II) and palladium(II)

Article abstract of DOI:10.1021/om100431u

We report a thorough investigation of a series of isomeric complexes with the general formula trans-(pyridylidene)M(PPh₃)₂Cl (M = Pd, Ni). For the first time, a systematic comparison of normal, abnormal, and remote bonding modes is presented. X-ray structural and ¹³C NMR data indicate the importance of carbenoid mesomeric contributions in their compound class. The catalytic performance of the palladium complexes trans-(pyridylidene) Pd(PPh₃)₂Cl as precursors in Suzuki-Miyaura-type cross-coupling suggests a correlation of remote bonding with catalyst robustness and effectivity. When metal precursors M(PPh₃)₄ are reacted with 2,4-dichloropyridinium salts, preferential formation of remote carbene complexes occurs and indications are that electronic parameters rather than steric influences are responsible for the observed selectivity. Calculations at the BP86/TZ2P level of theory support interpretation of the results.

Full text of DOI:10.1021/om100431u

Organometallics 2010, 29, 5821–5833 5821
DOI: 10.1021/om100431u
Pyridine-Derived N-Heterocyclic Carbenes: An Experimental and
Theoretical Evaluation of the Bonding in and Reactivity of Selected
Normal and Abnormal Complexes of Nickel(II) and Palladium(II)
Elzet Stander-Grobler,† Oliver Schuster,*,†,^ Greta Heydenrych,‡ Stephanie Cronje,†
Evangeline Tosh,§ Martin Albrecht,^ Gernot Frenking,*,‡ and
Helgard G. Raubenheimer*,†
†Department of Chemistry and Polymer Science, University of Stellenbosch, Private Bag X1,
‡
Matieland, 7602, Stellenbosch, South Africa, Fachbereich Chemie, Philipps-Universitat Marburg,
€
Hans-Meerwein-Strasse, D-35043 Marburg, Germany, §Inorganic Chemistry Department, Technische
Universita€t Mu€nchen, Lichtenbergstrasse 4, D-85747 Garching, Germany, and ^Department of Chemistry,
ꢀ
University of Fribourg, Chemin du Musee 9, CH-1700 Fribourg, Switzerland
Received May 6, 2010
We report a thorough investigation of a series of isomeric complexes with the general formula
trans-(pyridylidene)M(PPh3)2Cl (M = Pd, Ni). For the first time, a systematic comparison of normal,
abnormal, and remote bonding modes is presented. X-ray structural and 13C NMR data indicate the
importance of carbenoid mesomeric contributions in their compound class. The catalytic perfor-
mance of the palladium complexes trans-(pyridylidene)Pd(PPh3)2Cl as precursors in Suzuki-
Miyaura-type cross-coupling suggests a correlation of remote bonding with catalyst robustness
and effectivity. When metal precursors M(PPh3)4 are reacted with 2,4-dichloropyridinium salts,
preferential formation of remote carbene complexes occurs and indications are that electronic
parameters rather than steric influences are responsible for the observed selectivity. Calculations
at the BP86/TZ2P level of theory support interpretation of the results.
Introduction
extraordinary electron-rich metal centers.6-17 These discov-
eries led to a considerably increased interest in “nonclassical”
NHCs, i.e., carbenes that are not stabilized by two adjacent
heteroatoms.18 We have been active in this field for a decade
and have investigated in particular pyridinium-derived
carbenes.19-25 With now well-established protocols for the
functionalization of pyridine, pyridylidenes offer great po-
tential for variation of steric demand and electronic proper-
ties. Moreover, they can bind to the metal in a normal (I, III)
or abnormal (II) fashion (Scheme 1) with the possibility of
The chemistry of N-heterocyclic carbenes (NHCs) has
seen an explosive development during the last number of
years.1-5 However, despite the vast number of possible
N-heterocycles that could be classified as carbenes, the acti-
vity in NHC chemistry has concentrated by and large on
2-imidazolylidenes, often called “Arduengo” carbenes.1-5
Only recently has it been shown that an abnormal binding
mode via the C4 (or C5) position of imidazolylidenes yields
*To whom correspondence should be addressed. E-mail: oliver.
schuster@unifr.ch.
(14) Heckenroth, M.; Albrecht, M. Chimia 2008, 62, 253.
(15) Heckenroth, M.; Kluser, E.; Neels, A.; Albrecht, M. Dalton
Trans. 2008, 6242.
(1) Bertrand G., Ed. J. Organomet. Chem. 2005, 690 (24-25).
(2) Crabtree R. H., Ed. Coord. Chem. Rev. 2007, 251 (5-6).
(3) Glorius F. Ed. Top. Organomet. Chem. 2007, 21.
(4) Nolan, S. P. N-Heterocyclic Carbenes in Synthesis; Wiley-VCH:
Weinheim, 2006.
(5) Hahn, F. E.; Jahnke, M. C. Angew. Chem., Int. Ed. 2008, 47, 3122.
(6) Gruendemann, S.; Kovacevic, A.; Albrecht, M.; Faller, J. W.;
Crabtree, R. H. Chem. Commun. 2001, 2274.
(7) Gruendemann, S.; Kovacevic, A.; Albrecht, M.; Faller, J. W.;
Crabtree, R. H. J. Am. Chem. Soc. 2002, 124, 10473.
(8) Kluser, E.; Neels, A.; Albrecht, M. Chem. Commun. 2006, 4495.
(9) Heckenroth, M.; Kluser, E.; Neels, A.; Albrecht, M. Angew.
Chem., Int. Ed. 2007, 46, 6293.
(10) Tonner, R.; Heydenrych, G.; Frenking, G. Chem.-Asian J.
2007, 2, 1555.
(11) Tonner, R.; Heydenrych, G.; Frenking, G. ChemPhysChem
2008, 9, 1474.
(12) Yang, L.; Kruger, A.; Neels, A.; Albrecht, M. Organometallics
2008, 27, 3161.
(13) Albrecht, M. Chem. Commun. 2008, 3601.
(16) Heckenroth, M.; Neels, A.; Garnier, M. G.; Aebi, P.; Ehlers,
A. W.; Albrecht, M. Chem.-Eur. J. 2009, 15, 9375.
(17) Aldeco-Perez, E.;Rosenthal, A. J.;Donnadieu, B.;Parameswaran,
P.; Frenking, G.; Bertrand, G. Science 2009, 326, 556.
(18) Schuster, O.; Yang, L.; Raubenheimer, H. G.; Albrecht, M.
Chem. Rev. 2009, 109, 3445.
(19) Raubenheimer, H. G.; Toerien, J. G.; Kruger, G. J.; Otte, R.; van
Zyl, W.; Olivier, P. J. Organomet. Chem. 1994, 466, 291.
(20) Schneider, S. K.; Roembke, P.; Julius, G. R.; Raubenheimer,
H. G.; Herrmann, W. A. Adv. Synth. Catal. 2006, 348, 1862.
(21) Schneider, S. K.; Julius, G. R.; Loschen, C.; Raubenheimer,
H. G.; Frenking, G.; Herrmann, W. A. Dalton Trans. 2006, 1226.
(22) Schneider, S. K.; Rentzsch, C. F.; Krueger, A.; Raubenheimer,
H. G.; Herrmann, W. A. J. Mol. Catal. A: Chem. 2007, 265, 50.
(23) Raubenheimer, H. G.; Cronje, S. Dalton Trans. 2008, 1265.
(24) Heydenrych, G.; von Hopffgarten, M.; Stander, E.; Schuster, O.;
Raubenheimer, H. G.; Frenking, G. Eur. J. Inorg. Chem. 2009, 1892.
(25) Strasser, C. E.; Stander-Grobler, E.; Schuster, O.; Cronje, S.;
Raubenheimer, H. G. Eur. J. Inorg. Chem. 2009, 1905.
r
2010 American Chemical Society
Published on Web 10/22/2010
pubs.acs.org/Organometallics

5822 Organometallics, Vol. 29, No. 22, 2010
Stander-Grobler et al.
Scheme 1. Isomeric Pyridine and Pyridylidenes
C4-coordination exclusively leads to metalation at the
4-position.35-39 This observation has been attributed to the
steric bulk of the N-alkyl group.
We first carried out a thorough investigation of a series
of isomeric complexes with the general formula trans-
(pyridylidene)M(PPh3)2Cl (M = Pd, Ni) using various ex-
perimental and computational parameters. We were parti-
cularly interested in evaluating whether certain mesomeric
contributions would be particularly significant and if the
concept of abnormal carbene bonding, which is defined on
the basis of simple valence bond considerations, transfers to
experimental observations. Second, the catalytic perfor-
mance of the palladium complexes trans-(pyridylidene)Pd-
(PPh3)2Cl as precursors in Suzuki-Miyaura-type cross-
coupling reactions was studied in order to elucidate the
influence of normal, abnormal, or remote carbene bonding.
Finally, we investigated preferential carbene complex for-
mation by oxidative addition using low-valent metal pre-
cursors and 2,4-dichloropyridinium salts. DFT calculations
were used to support our interpretation of the results. As an
extension of our studies on quinolylidene complexes,20,21,23,40-42
we also determined the reactivity of M(PPh3)4 (M = Pd, Ni)
toward 4,7-dichloroquinolinium salts, wherein both chloro sub-
stituents are located three bonds away from the nitrogen atom.
Throughout this article, carbene complexes are drawn
with MdC double bonds where applicable. Our choice of
representation should not be regarded as a statement con-
cerning the actual bond order; it merely helps the reader to
identify formal (normal) carbene metal bonds.
both normal and abnormal carbenes arising in a remote position
(II, III). We have extended Crabtree’s terminology6,7 by
using the term abnormal whenever it is not possible to
illustrate the carbene structure without charge separation
(in contrast to Ia and IIIa).18 Remote denotes a carbene that
contains no stabilizing heteroatom directly adjacent to the
carbene carbon. Accordingly, isomer II is abnormal and
remote, whereas isomer III is normal and remote.
Previous theoretical investigations of the free ligands
indicate that pyridine is by far more stable (0.0 kcal/mol)
than the most stable carbenoid isomer I (ca. 40 kcal mol-1),
while the remote species II and III are even higher in energy
(both ca. 55 kcal mol-1).24,26,27 Interestingly, calculations of
isodesmic reactions by Kassaee et al. suggest a similar stabili-
zation of the methylene fragment in 2- and 4-pyridylidenes.28
Significant barriers for intramolecular interconversion
allowed for detecting the coexistence of isomers I and II
under high-vacuum conditions.26,27 In the condensed phase,
intermolecular interconversion is more facile, and pyridyli-
denes have thus far been observed only in complexes, i.e.,
stabilized by a metal center.29 Such metal-stabilized pyridy-
lidenes have been accessible in a few cases by a ligand
isomerization process involving a formal proton shift from
carbon to nitrogen. Depending on the ring substitution
pattern, both 2- and 4-pyridylidene complexes have been
prepared by such a rearrangement from N-protonated pyri-
dinium salts.30-34 In contrast, C-H activation of N-alkyl-
ated pyridinium precursors that could allow for C2- or
Results and Discussion
a. Synthesis of 2-, 3-, and 4-Pyridylidene Metal Complexes.
The synthetic routes used to prepare two complete series of
ionic 2-, 3-, and 4-pyridylidene metal complexes 3 and 4
are summarized in Scheme 2 and follow our established
strategy that has been reported earlier.23 Pyridinium triflates 2
(triflate = OTf = trifluoromethanesulfonate) were prepared
from the corresponding chloropyridines 1a-c with one equi-
valent of methyltriflate in dry dichloromethane. Subsequent
oxidative addition of the ligand precursor salts to Ni(PPh3)4 in
THF at room temperature afforded the desired yellow cationic
complexes 4 in good yields. Similar procedures in toluene at
60 ꢀC were followed to obtain the corresponding colorless
palladium complexes 3. The oxidative addition leading to the
abnormal complex 3b occurred significantly slower than the
corresponding reaction for normal carbenes.
b. Single-Crystal Structure Determinations. The ORTEP
drawing of the molecular structure of the nickel complex 4b
in Figure 1 is representative for the square-planar geometry
generally observed for complexes 3 and 4, as previously estab-
lished for BF4 analogues of the 2- and 4-pyridylidene metal
complexes.20,21 The phosphine ligands adopt the usual trans
configuration. Unlike the corresponding complexes 3a and 4a
or 3c and 4c (in the BF4 salt), which crystallize in distinct
packing arrangements,20,21 crystals of 3b and 4b (in the OTf
salt) are isomorphous. Relevant parameters for the entire
(26) Lavorato, D. J.; Terlouw, J. K.; Dargel, T. K.; Koch, W.;
McGibbon, G. A.; Schwarz, H. J. Am. Chem. Soc. 1996, 118, 11898.
(27) Lavorato, D. J.; Terlouw, J. K.; McGibbon, G. A.; Dargel,
T. K.; Koch, W.; Schwarz, H. Int. J. Mass Spectrom. 1998, 180, 7.
(28) Kassaee, M. Z.; Shakib, F. A.; Momeni, M. R.; Ghambarian,
M.; Musavi, S. M. Tetrahedron 2009, 65, 10093.
(29) Holloczki, O.; Nyulaszi, L. J. Org. Chem. 2008, 73, 4794.
(30) Weisman, A.; Gozin, M.; Kraatz, H. B.; Milstein, D. Inorg.
Chem. 1996, 35, 1792.
(31) Cave, G. W. V.; Hallett, A. J.; Errington, W.; Rourke, J. P.
Angew. Chem., Int. Ed. 1998, 37, 3270.
(32) Newman, C. P.; Clarkson, G. J.; Alcock, N. W.; Rourke, J. P.
Dalton Trans. 2006, 3321.
(33) Alvarez, E.; Conejero, S.; Paneque, M.; Petronilho, A.; Poveda,
M. L.; Serrano, O.; Carmona, E. J. Am. Chem. Soc. 2006, 128, 13060.
(34) Alvarez, E.; Conejero, S.; Lara, P.; Lopez, J. A.; Paneque, M.;
Petronilho, A.; Poveda, M. L.; del Rio, D.; Serrano, O.; Carmona, E.
J. Am. Chem. Soc. 2007, 129, 14130.
(35) Cordone, R.; Harman, W. D.; Taube, H. J. Am. Chem. Soc.
1989, 111, 2896.
(40) Meyer, W. H.; Deetlefs, M.; Pohlmann, M.; Scholz, R.;
Esterhuysen, M. W.; Julius, G. R.; Raubenheimer, H. G. Dalton Trans.
2004, 413.
(41) Schneider, S. K.; Roembke, P.; Julius, G. R.; Loschen, C.;
Raubenheimer, H. G.; Frenking, G.; Herrmann, W. A. Eur. J. Inorg.
Chem. 2005, 2973.
(36) Koizumi, T.; Tomon, T.; Tanaka, K. Organometallics 2003, 22, 970.
(37) Koizumi, T.; Tomon, T.; Tanaka, K. J. Organomet. Chem. 2005,
690, 1258.
(38) Albrecht, M.; Stoeckli-Evans, H. Chem. Commun. 2005, 4705.
(39) Koizumi, T.; Tomon, T.; Tanaka, K. J. Organomet. Chem. 2005,
690, 4272.
(42) Schuster, O.; Raubenheimer, H. G. Inorg. Chem. 2006, 7997.

Article
Organometallics, Vol. 29, No. 22, 2010 5823
Scheme 2. Preparation of Two Complete Series of Carbene Complexes
Bercaw and co-workers have discussed the bonding situa-
tion for normal pyridylidene complexes in terms of the bond
alternation within the heterocyclic carbene ligand.45 We
followed a similar approach in our efforts to identify relevant
contributing structures for complexes 4 (Scheme 3).46
The 2-pyridylidene moiety in 4a shows a significant in-
crease in the N1-C6 and C2-C3 distances when compared
to free pyridine (Figure 2). However, the N1-C2 bond
remains short. These observations could indicate the relative
importance of contributing structures such as IVa and IVe
in Scheme 3. A weak diene-type character of 2-pyridy-
lidenes has been suggested earlier by others.47 Remarkably,
examination of available structural data on 2-pyridone
reveals a rather similar bond distribution pattern therein
(Figure 2).
Evidence for a diene-like contribution such as VIe
(Scheme 3) is also found in the 4-pyridylidene ligand of 4c.
The lateral bonds C2-C3 and C5-C6 are significantly (4.3σ)
shorter than those of C3-C4 and C4-C5 (Figure 2). Related
bond variations are also found for 4-pyridylidene tungsten25
and platinum complexes.31 The latter comprise a short Pt-C
bond and a 13C carbene resonance as high as 324 ppm.
Again, a similar geometry is featured by the corresponding
4-pyridone (Figure 2).
In contrast to the heterocyclic rings of the two normal
pyridylidene ligands in 4a and 4c, where internal distances
qualitatively support carbenoid mesomers (IVe and VIe,
respectively), the abnormal bonding in complex 4b entails
bond length variations that cannot be readily described in
terms of simple valence bond resonance theory. Neverthe-
less, consistent with the examples above, bonds that involve
the nitrogen atom are relatively short (N1-C2, N1-C6),
while bonds from the carbene carbon atom are rather long
(C2-C3, C3-C4).
Figure 1. Molecular structure of the cationic complex 4b
(50% probability ellipsoids, hydrogen atoms and counterion
omitted).
series of isomeric cationic structures are summarized in
Table 1.
As stated earlier,18 examination of M-C distances is not a
versatile tool to prove or rule out carbene-type bonding
because differences are usually not significant. Similarly,
hindered rotation about the M-C bond does not seem to
reflect increased M-C bond order.43
On the other hand, the observed static trans influences
seem to show a consistent trend: the further the heteroatom
occurs from the carbene carbon atom, the longer the trans
M-Cl bond becomes. Bond lengthening is accompanied by a
decrease in the internal bond angle at the carbene carbon
(CCX; X = N, C). However, the differences are small and
might be influenced by intermolecular dipole-dipole inter-
actions in the solid state.44
Such findings are particularly relevant when considering
that the bond length variations found in complexes 4 are not
observed in the corresponding chloropyridinium precursor
salts (see Supporting Information for details). In contrast to
the situation in the metal complexes, the atom separations
are not significantly dependent on the substitution pattern
and are closely related to those in unsubstituted pyridinium
salts. We therefore conclude that metalation, rather than
alkylation, has the greatest influence on the bonding within
the heterocyclic ligands during ylidene formation from
chloropyridines. Furthermore, this influence of the metal
fragment seems to be related to the effect that a doubly
bound oxygen exerts in pyridones. These arguments all point
toward a non-negligible double-bond character of the metal
carbon bonds in the normal pyridylidene complexes.
(43) As a possible indicator for a hindered rotation around the metal
carbon bond and hence an increased metal-carbon bond order, we
investigated the angles between the mean square planes of the pyridy-
lidene ring and the coordination spheres. Given the fact that an in-plane
arrangement is prevented by the bulky phosphine ligands, efficient π
back-bonding should result in an orthogonal alignment. Notably, both
4-pyridylidene ligands in the two complexes 3c and 4c deviate consider-
ably from this ideal geometry. We are however hesitant to ascribe this
observation to an electronic effect because steric considerations seem to
prevail (cf. Supporting Information).
(44) Jonas, V.; Frenking, G.; Reetz, M. T. J. Am. Chem. Soc. 1994,
116, 8741.
(45) Owen, J. S.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc.
2004, 126, 8247.
(46) We mention that reinvestigation of the crystal structure of 3a (in
the tetrafluoroborate salt) revealed disorder of the pyridylidene ligand
(sof: 0.2), which was not found when the result was published initially.20
The updated structure and selected data can be found in the Supporting
Information. However, the disorder compromises a closer examination
of the bond distances and angles in the carbene ligand, and we shall
therefore concentrate on the nickel complexes, which also, in general,
offered better data.
(47) Conejero, S.; Lara, P.; Paneque, M.; Petronilho, A.; Poveda,
M. L.; Serrano, O.; Vattier, F.; Alvarez, E.; Maya, C.; Salazar, V.;
Carmona, E. Angew. Chem., Int. Ed. 2008, 47, 4380.

5824 Organometallics, Vol. 29, No. 22, 2010
Table 1. Selected Structural Parameters of the Series of Pyridylidene Complexes
Stander-Grobler et al.
b
distortion [A]
torsiona [deg]
˚
˚
˚
d(M-C) [A]
d(M-Cl) [A]
CCX angle [deg]
3a
2.002(3)c
1.996(7)
1.979(7)
1.861(5)
1.874(3)
1.863(5)
2.3638(8)
2.371(2)
2.3938(17)
2.203(1)
2.2055(8)
2.2172(13)
116.22c
117.82
114.64
116.84
116.23
115.04
88.60c
88.21
77.84
90.00
88.81
83.46
0.628c
0.572
0.155
1.161
0.698
0.228
3b
3cd
4ae
4b
4ce
a Angle between the mean square planes defined by (i) the six atoms of the heterocycle (plane1) and (ii) the metal center and the four atoms bound
to the metal (plane2). b Distance of the metal-bound carbon atom from plane2 as a probe for the distortion of the square-planar geometry. c Inaccurate,
due to disorder of the ligand (cf. Supporting Information). d Ref 20. e Ref 21.
Scheme 3. Various Mesomeric Structures of Normal (IV, VI) and Abnormal (V) Pyridylidene Complexes, Some of Them Remote (V, VI)
Interestingly, the nonmethylated 4-pyridyl ligand in palla-
dium complex A (Figure 3) displays similar features to those
of the pyridylidene ligand in 4c, i.e., short lateral bonds.49 It
could be reasoned that coordination to the metal center
already induces localization of double bonds in a carbenoid
fashion (A0). Indeed, according to electron decomposition
analysis (EDA), σ and π orbital contributions in pyridyl-
metal and pyridylidene-metal bonds are remarkably
similar.24 This result also provides an explanation for the in-
creased basicity of the nitrogen in pyridyl metal complexes,
which can even lead to the activation of dichloromethane.50-55
c. NMR Spectroscopic Characterization. Similar to the
difficulties experienced when attempting to determine “car-
bene character” from metal-carbene bond lengths, a com-
parison of absolute carbene carbon NMR shift values also
does not give clear answers.18 We have previously noted that
correlation of NMR data with data of other analytical tech-
niques, e.g., IR ν(CO) stretching frequencies, is often poor,
thus requiring cautious interpretation.18 In the series of com-
plexes 3 and 4, significantly lower chemical shifts are ob-
served for the abnormal complexes 3b and 4b (δC 165 and
174, respectively) than for the normal congeners (δC>190).
When compared to the corresponding carbon resonances in
the precursor pyridinium salts, deshielding between 30 and
50 ppm occurs upon formal carbene formation. For both
metals, abnormal C3 bonding induces the smallest and normal
remote C4 bonding the largest shift difference. Nickel bonding
shows consistently a stronger deshielding effect than palladium
coordination.
In order to correlate the structural investigations with
solution properties, we also elucidated the electronic influ-
ence of alkylation and metalation on the heterocyclic ring by
NMR spectroscopy. Comprehensive interpretation of the 1H
NMR chemical shifts is hampered by the phosphine signals
in the aromatic region that overlap with the protons of the
heterocyle. Hence, we concentrated on 13C NMR data,
which are known to be less affected by long-range magnetic
and solvent effects56 and which correlate well with 1H data in
aromatic systems.56,57 Figure 4 summarizes the most rele-
vant chemical shifts.
(48) Bruno, I. J.; Cole, J. C.; Edgington, P. R.; Kessler, M.; Macrae,
C. F.; McCabe, P.; Pearson, J.; Taylor, R. Acta Crystallogr., Sect. B:
Struct. Sci. 2002, 58, 389.
(49) Meguro, H.; Koizumi, T.; Yamamoto, T.; Kanbara, T.
J. Organomet. Chem. 2008, 693, 1109.
Linear correlations between the chemical shift and the
charge density on a given atom in a homologous series of
compounds are often used for the interpretation of NMR
results.58 Using calculated charge densities, such a depen-
dence has been shown to exist for azine derivatives,59 as well
(50) Isobe, K.; Nakamura, Y.; Miwa, T.; Kawaguchi, S. Bull. Chem.
Soc. Jpn. 1987, 60, 149.
(51) Crociani, B.; Di Bianca, F.; Giovenco, A.; Berton, A.; Bertani,
R. J. Organomet. Chem. 1989, 361, 255.
(52) Crociani, B.; Di Bianca, F.; Fontana, A.; Bertani, R. J. Orga-
nomet. Chem. 1992, 425, 155.
(53) Canovese, L.; Uguagliati, P.; Bianca, F. D.; Crociani, B.
J. Organomet. Chem. 1992, 438, 253.
(56) Lauterbur, P. C. Tetrahedron Lett. 1961, 2, 274.
(57) Retcofsky, H. L.; McDonald, F. R. Tetrahedron Lett. 1968,
2575.
(58) Hesse, M.; Meier, H.; Zeeh, B. Spectroscopic Methods in Organic
Chemistry; Thieme: Stuttgart, 2008.
(54) Crociani, B.; Di Bianca, F.; Fontana, A.; Forsellini, E.;
Bombieri, G. J. Chem. Soc., Dalton Trans. 1994, 407.
(55) Canovese, L.; Visentin, F.; Uguagliati, P.; Di Bianca, F.;
Fontana, A.; Crociani, B. J. Organomet. Chem. 1996, 525, 43.

Article
Organometallics, Vol. 29, No. 22, 2010 5825
On the basis of these considerations, nitrogen methylation
of the chloropyridines 1a-c affects mostly the region remote
to the nitrogen atom. Evidently, the heteroatom ; now with
increased effective electronegativity ; compensates for the
loss of its lone pair by draining electron density from the ring
and thus decreases the electron density especially at the
4-position significantly in the pyridinium salts 2 regardless
of the chloro-substitution pattern. Even though the upfield
shift for C2 and C6 is marginal, it might be attributed to a
decrease in the N1-C2 and N1-C6 bond orders.66
Subsequent metal insertion at the 2-position of 2a to form
3a or 4a induces a drift of the positive charge in the ring
toward the vicinity of the metal, thus indicating contribu-
tions from resonance structures IVa, IVd, and IVe (cf.
Scheme 3). Simultaneously, a secondary process apparently
increases the electron density at the carbon atoms that are
remote to the coordination site. Owing to this effect, the 13C
NMR chemical shifts of the neutral starting compound 1 are
somewhat restored in the ligand backbone, which is in
agreement with predominant charge stabilization within
the N-Ccarbene-M moiety.
The importance of contributing structures similar to VIc
and VIe (Scheme 3) for the remote 4-pyridylidene complexes
3c and 4c is reflected by a significant charge redistribution
within the heterocycle upon carbene formation. Again, metal
insertion effects deshielding of the carbon atoms adjacent to
the metal-bound carbon and concurrent shielding of the
remote nuclei. Even the nitrogen atom possibly becomes less
positively charged, as is shown by a small upfield shift for the
attached methyl group. Due to the mutual trans arrangement
of nitrogen and carbene carbon atoms, the polarization in
the complex is sufficiently large to render the carbons in
ortho and meta position almost magnetically equivalent. This
result lends further support to the carbene character of C4-
pyridylidene complexes (VIe, cf. X-ray discussion), and it is
also in agreement with NBO analysis, which revealed that a
Lewis structure resembling IIIa (Scheme 1) is most appro-
priately describing a free 4-pyridylidene.67
For abnormal C3-pyridylidene complexes, interpretation
of the NMR results in terms of valence bond theory is not
straightforward, thus paralleling the solid-state conclusions.
However, the 13C NMR data indicate consistently for all
pyridylidenes that metalation induces a deshielding effect in
the vicinity of the metal combined with shielding of the remote
heterocyclic nuclei. Similar observations have also been
made by Isobe and co-workers50 for neutral non-methylated
bromopyridines upon insertion of Pd(PEt3)2 into the car-
bon-halide bond. This effect may thus be independent of
the coordination mode of the ligand (normal vs abnormal) or
its partial charge (pyridylidene vs pyridyl). Computational
studies support such a conclusion, since the σ and π orbital
contributions in pyridyl-metal and pyridylidene-metal
bonds have been calculated to be remarkably similar.24
Consistent with our conclusions from crystallographic
analyses, the observed influence of the metal fragment on
the heterocycle might be rationalized by non-negligible
MfL π-back-donation. In general, an essentially similar
but stronger shielding/deshielding behavior is observed
for the nickel(II) fragment compared to palladium(II).
˚
Figure 2. Bond distances (A) in 4a, 4b, and 4c (top) in compar-
ison to those in free pyridine and pyridones. (Bottom) Data
derived from CCSD.48
Figure 3. Metal-induced carbenoid electron distribution prior
to formal carbene formation (via methylation or protonation).
as for palladium pyridyl complexes.60 The latter complexes,
however, show a specific behavior for the metal-bound
carbons, which can be attributed to anisotropic effects.61-63
In our series of compounds, linear regression of the calculated
charges in 1a-c64 and experimental NMR measurements
(Figure 4) also yields an appreciable fit (R2 = 0.91; see
Supporting Information). We therefore assume that the 13C
NMR data of the heterocyclic carbons that are not bound to
the metal offer a qualitative probe for the partial charge
(re)distributions during complex formation. Information
about relative charge densities at the nitrogen atom can be
derived indirectly through the N-methyl carbon resonance.65
Counterion effects may be insignificant since the triflate
salts show negligible alteration of shifts when compared to
the corresponding BF4- analogues, some of which have been
reported earlier.20,21
(59) Lauterbur, P. C. J. Chem. Phys. 1965, 43, 360.
(60) Isobe, K.; Kai, E.; Nakamura, Y.; Nishimoto, K.; Miwa, T.;
Kawaguchi, S.; Kinoshita, K.; Nakatsu, K. J. Am. Chem. Soc. 1980, 102,
2475.
(61) Spiesecke, H.; Schneider, W. G. J. Chem. Phys. 1961, 35, 722.
(62) Kutzelnigg, W. Isr. J. Chem. 1980, 19, 193.
(63) Kutzelnigg, W.; Fleischer, U.; Schindler, M. Basic Princ. Prog.
1990, 23, 165.
(64) Feshin, V. P.; Feshina, E. V.; Zhizhina, L. I. J. Heterocycl. Chem.
(66) Pugmire, R. J.; Grant, D. M. J. Am. Chem. Soc. 1968, 90,
2006, 42, 1435.
697.
(65) Wehrli, F. W.; Giger, W.; Simon, W. Helv. Chim. Acta 1971, 54,
229.
(67) Tukov, A. A.; Normand, A. T.; Nechaev, M. S. Dalton Trans.
2009, 7015.

5826 Organometallics, Vol. 29, No. 22, 2010
Stander-Grobler et al.
Figure 4. Summary of 13C NMR chemical shifts in ppm.
Table 2. Computational Results at the BP86/TZ2P Level and Comparison with Experimental Values, Which Are Given in Italicsa
Erelb [kcal mol-1
]
Eintc [kcal mol-1
]
˚
˚
d(M-C) [A]
d(M-Cl) [A]
CCX angle [deg]
M3a
M3b
M3c
M4a
M4b
M4c
2.021
2.002(3)
2.029
1.996(7)
2.019
1.979(7)
1.873
1.861(5)
1.881
2.342
2.3638(8)
2.347
2.371(2)
2.339
2.3938(17)
2.189
2.203(1)
2.192
116.23
116.22
115.67
117.82
115.34
114.64
116.10
116.84
115.47
116.23
115.13
115.04
0
-95.9
3.25
3.47
0
-105.9
-109.3
-101.6
-111.3
-115.5
3.61
3.72
1.874(3)
1.877
1.863(5)
2.2055(8)
2.179
2.2172(13)
a Compared to experimental values (cf. Table 1). b Relative minimized energies. c Metal-carbene interaction energies.
Scheme 4. Suzuki-Miyaura-Type Test Reactions
This effect may tentatively be assigned to a larger MdC
double-bond character in 4 than in the corresponding
palladium complexes 3.
(Table 2). The calculated M-C bonds are slightly shorter
than the experimental values, while the theoretical M-Cl
distances are a bit longer than the X-ray data, but the
differences are not very large. The very small changes
observed in the M-C and M-Cl bond lengths are within
the error range of the theoretical and experimental data, and
thus, they do not exhibit a significant trend. The theoretical
and experimental CCX bond angles agree well. Also bond
length variations within the rings, which were discussed in
section b, are qualitatively reproduced. The calculated data
of the model compounds may thus be used for the inter-
pretation of the experimental results.
The relative energies of the model systems M3 and M4
follow the trend 2-pyridylidene , 3-pyridylidene < 4-pyr-
idylidene, which is the same order that has been reported for
the corresponding free carbene isomers of pyridine, in which
one proton is located on the nitrogen atom.24,27,28 However,
the energy differences are much smaller in the complexes and
Singlets in the 31P NMR spectra of all complexes indicate
the equivalence of both phosphine ligands, hence confirming
a mutual trans arrangement also in solution (Scheme 2).
Even though the resonances for complexes 3 and 4 fall in a
narrow range (δP 22.7-24.9 and 21.2-23.0, respectively),
both series show a common trend, i.e., 2- < 4- < 3-pyridy-
lidene, suggesting an increasing donor strength of the car-
bene ligand in this order.68
c. Computational Studies. Calculations for the series of
isomeric palladium and nickel model complexes M3 and M4
comprising PH3 ligands instead of PPh3 (as in Scheme 2)
were conducted at the BP86/TZ2P level of theory to
verify and extend our interpretation of experimental results
(68) Iglesias, M.; Albrecht, M. Dalton Trans. 2010, 39, 5213.

Article
Organometallics, Vol. 29, No. 22, 2010 5827
Figure 5. Time/conversion curves under inert conditions (3a
squares; 3b dots; 3c triangles; Pd(PPh3)4 inverted triangles).
Figure 6. Molecular structure of 7 (50% probability ellipsoids,
˚
hydrogen atoms and counterion omitted). Selected bond lengths [A]
and angles [deg]: Ni1-C4 1.855(2), N1-C2 1.356(3), C2-C3
1.372(3), C3-C4 1.399(3), C4-C5 1.410(3), C5-C6 1.1359(3), C6-
N1 1.355(3); C3-C4-C5 116.6(2), C4-Ni1-Cl1 170.24(7).
Scheme 5. Oxidative Addition in a Competitive Environment
Yields Exclusively the 4-Pyridylidene Complexes 6 and 7
unveiled further subtle differences between the pyridylidene-
based catalytic systems (Figure 5). The initial catalytic
activity during the first two hours is very similar for all three
complexes, perhaps slightly higher for the normal carbene
complex 3a. However, this complex seems to be the least
robust and deactivates after extended periods of reaction
time, thus not reaching high conversions. In contrast, the
remote pyridylidene complexes 3b and 3c remain active even
after 20 h of activity. Hence the general trends stated above
need to be refined. While the normal pyridylidene complexes
3a and 3c are slightly more active than abnormal 3b, the
remote complexes 3c and 3b are more robust than 3a. Since
the differences observed here are only small, further tests will
be needed in order to substantiate the deduced trends.
f. Regioselectivity of the Oxidative Addition of Pyridylium
Salts. In addition to comparing the possible bonding modes
of pyridylidenes, we were also interested in determining
how carbene formation would preferentially occur when
the metal complex reactant was presented with two active
Cl-containing sites on the pyridinium ring. Reactions
of M(PPh3)4 (M=Pd, Ni) with the dichlorinated methyl-
pyridinium salt 5 in toluene (Scheme 5) yielded, according to
13C and 1H NMR, exclusively the 4-pyridylidene complexes
6 and 7. In comparison with the unsubstituted analogue 3c,
complex 6 shows a ca. 5 ppm downfield shift of the carbene
resonance, which can be attributed to the meta influence of
the chloride substituent. The 31P signals differ only insignifi-
cantly.
can be rationalized by the calculated stronger interactions in
3- and 4-pyridylidene complexes, partly compensating the
energy difference in the free pyridylidenes. The energy
decomposition analysis of the metal-carbon bonds reveals
increasing interaction energies (Eint) in the order 2-pyridyli-
dene<3-pyridylidene<4-pyridylidene. In an extensive EDA
study of pyridylidene and quinolylidene M-C bonds we
could show that Eint correlates very well with the energy of
the HOMO, i.e., the σ lone-pair orbital of the carbene and
thus the σ-donor properties of a given ligand.24
e. Catalysis. In order to determine the influence of abnor-
mal and remote pyridylidene bonding modes on catalytic
performance, the palladium complexes 3 were used in Suzuki-
Miyaura cross-coupling reactions (Scheme 4).69 Results
of preliminary experiments under standard conditions in-
dicated small but distinct differences in the catalytic activity.
As a general trend, increasing remoteness enhanced the
conversions, and yields of the biaryl product followed
the trend 3c (86%)>3b (82%)>3a (78%). All complexes
displayed better performance than the reference complex
Pd(PPh3)4 (71%). The trend was slightly more pronounced
when using milder reaction temperatures and working
under an inert atmosphere (Figure 5; also see Supporting
Information).
X-ray crystal structure determinations of 6 and 7 unam-
biguously confirmed the coordination sites deduced by
NMR spectroscopy (Figure 6). The observed features are
generally similar to those discussed above. Notably, the
No palladium black was observed during catalysis, thus
suggesting a homogeneous mode of action. This assumption
is further supported by the pertinent time-conversion pro-
files, which do not show any induction time or sigmoidal
substrate conversion. Time-dependent monitoring also
˚
M-C distances of 1.972(4) and 1.855(2) A for 6 and 7,
respectively, both represent the shortest bonds in the series
of pyridylidene complexes discussed in this article. The
M-Cl bonds are within the range described.
On the basis of the steric shielding effect of the N-methyl
group in 5 and with reference to the seminal work by
Carmona and co-workers,33,34,47 the observed regioselectiv-
ity of metalation might be ascribed to steric discrimina-
tion. However, the same selectivity was observed when the
(69) Test reactions under Buchwald-Hartwig conditions yielded
palladium black even at very low temperatures, and no discrimination
between the precatalysts 3 with Pd(OAc)2 or PdCl2 was possible,
probably due to a predominantly heterogeneous mode of action.

5828 Organometallics, Vol. 29, No. 22, 2010
Stander-Grobler et al.
does not depend on the steric requirements of the ligand. We
also calculated the computationally less demanding 4-sub-
stituted model compounds M8/M9 and their 2-substituted
congeners M80/M90 and found a similar energy difference in
favor of the latter isomers, as for the bulkier compounds
(Table 3). Energy decomposition analysis on these simpler
models further shows that the stronger interaction occurs in
the para position (Eint in Table 3). These findings are con-
sistent with the results shown in Table 2 and with our
previous EDA studies on pyridylidenes.24
The theoretically predicted higher stability of the 2-pyr-
idylidene complexes M60 and M70 and the experimental
finding that only the 4-congeners 6 and 7 are formed indicate
that these less stable isomers are the products of a kineti-
cally controlled reaction. This is a very important result,
which we shall investigate in a future study. We do not
think that the calculations predict a wrong stability order,
because we obtain the same trend with similar energy
differences using different models. Also, since the cal-
culations involve different isomers of the same molecule,
it can be expected that the results are quite reliable.
A theoretical study of the reaction course is quite expen-
sive because different pathways are possible. The theore-
tical work shall therefore be complemented by kinetic
studies.
Figure 7. Models of the possible products formed upon oxida-
tive addition in Scheme 5.
Table 3. Computational Results
Erela [kcal/mol]
Eintb [kcal/mol]
M6c
7.9
0
6.9
0
5.6
0
4.9
0
n/a
n/a
n/a
n/a
-103.1
-91.0
-109.3
-96.6
c
M60
M7c
c
M70
M8d
d
M80
M9d
d
M90
a Relative minimized energies. b Bond interaction energies. c BP86/
TZVPP. d BP86/TZ2P.
As an extension of our studies on quinolylidenes,20,21,23,40-42
we also compared the reactivity of M(PPh3)4 toward dichloro-
quinolinium salts 8 where both chloro substituents (Scheme 6)
are three bonds removed from the methylated (R = Me) or
protonated (R = H) nitrogen atom. Again, only one of the
possible products was formed in each reaction. Consistent
with the expected electrophilicity derived from the 13C NMR
spectra of the precursors (C4: δC 155; C7: δC 144), metalation
is preferred within the N-containing ring to yield complexes 9
and 10. In comparison with the monocyclic congeners in 3c
and 4c, the quinolylidene carbene signals are shifted about
10 ppm downfield, while the 31P NMR shifts remain essen-
tially identical.
protonated analogue of 5 was used as ligand precursor.
While the palladium homologue of 6 could not be obtained
in pure form, NMR spectra clearly indicated the formation
of the C4-bound complex as the main product and the only
carbene species (for details see Experimental Section). A
good agreement with the signals in the spectra of 6 and
especially 2JCP couplings for two carbon atoms are indicative
of the selective formation of the C4-bound isomer. Obviously,
metalation of the remote 4-position is not sterically triggered
but is electronically preferred.
The fact that metal attack occurs preferentially at the
remote position can be understood against the background
of established pyridine cross-coupling chemistry.70 Although
the aryl addition mechanism in such reactions is still a matter
of debate, all proposed pathways (i.e., concerted, SN2, free
halide)71-73 feature the electron-rich d10 metal as a nucleo-
phile. Accordingly, preferential attack at the most electron-
deficient 2-position can be assumed and is indeed observed.
For example 2,4-dibromopyridine is metalated by Pd(PPh3)4
almost exclusively in the ortho position.74 This corresponds to
the most deshielded carbon atom according to NMR results
(C2: δC 142.5; C4: δC 133.9). In our case, where the precursor 5
features 13C carbon shifts of 148.9 and 156.0 ppm for C2 and
C4, respectively, attack also occurs at the most deshielded (i.e.,
the presumably more electrophilic) site.
No traces of any dinuclear complexes were found even
when using an excess of metal precursor. Figure 8 depicts
the crystal structure of 9a, indicating that the presence of
bulky phosphine ligands excludes the formation of dinuclear
products.
Conclusions
While X-ray structural data and 13C NMR measurements
of the normal (both adjacent and remote to N) complexes
within a complete set of pyridine-derived carbene complexes
consistently support a carbenoid-type bonding (i.e., partial
MdC double-bond character), results for the abnormal
examples cannot be described in terms of simple valence
bond resonance theory. However, close examination reveals
a coherent and consistent influence of the metal fragments on
the heterocyclic ligand in all bonding modes. Due to an
apparent influence of remote carbene bonding on catalytic
robustness, this concept seems slightly more relevant for
catalytic application than the variation between normal and
abnormal bonding situations. In an investigation of compe-
titive carbene complex formation by oxidative insertion,
remote 4-pyridylidene formation is favored over that of
2-pyridylidene formation even in the absence of steric shield-
ing at nitrogen. Overall, we conclude that the concepts of
remote and (ab)normal carbene bonding are useful and their
Calculation of the relative energies for realistic models of 6
and 7 (Figure 7) at the BP86/TZVPP level of theory revealed
that the 2-pyridylidene complexes M60 and M70 are more
stable than the 4-congeners M6 and M7 (Erel in Table 3). The
same trend was calculated for the 2- and 4-substituted
isomers M3a/M3c and M4a/M4c (Table 2), which suggests
that the relative stability of the 2- and 4-substituted systems
(70) Schroter, S.; Stock, C.; Bach, T. Tetrahedron 2005, 61, 2245.
(71) Fitton, P.; Rick, E. A. J. Organomet. Chem. 1971, 28, 287.
(72) Portnoy, M.; Milstein, D. Organometallics 1993, 12, 1665.
(73) Senn, H. M.; Ziegler, T. Organometallics 2004, 23, 2980.
(74) Sicre, C.; Alonso-Gomez, J.; Cid, M. M. Tetrahedron 2006, 62,
11063.

Article
Organometallics, Vol. 29, No. 22, 2010 5829
Scheme 6. Competitive Oxidative Addition at Dichloroquinolinium Salts
phase was washed with brine, dried with MgSO4, and concen-
trated in a brown Schlenk tube. The oil (2.18 g, 19.2 mmol) was
dissolved in CH2Cl2 and cooled to -78 ꢀC, and methyl trifluoro-
methanesulfonate (3.46 g, 2.40 mL, 21.1 mmol) was added
dropwise. The mixture was stirred overnight while it was
allowed to warm to room temperature. The solvent was re-
moved in vacuo, and the resulting product was washed once with
10 mL of THF and twice with 10 mL of diethyl ether.
Procedure B: Preparation of Protonated Triflate Salts. The
hydropyridinium triflate salts were obtained by reacting the
ligand precursor with a small excess (1.1 molar equiv) of
CF3SO3H in 20 mL of CH2Cl2 at room temperature for 18 h.
Diethyl ether (20 mL) was added and a white precipitate formed.
The solvent was removed via a cannula and the resulting solid
washed with 4 ꢀ 20 mL of diethyl ether and dried in vacuo.
Procedure C: Preparation of trans-Chloro(pyridylidene)bis-
(triphenylphosphine)palladium(II) Triflate Complexes. The re-
spective triflate salts and a small excess (1.01 molar equiv) of
Pd(PPh3)4 were suspended in 30 mL of toluene and stirred for
17 h at 60 ꢀC (in the case of 3b, the reaction mixture was stirred
for 64 h). The white suspension in a light yellow solution was
allowed to cool to room temperature and filtered through Celite.
The solid on the filter was washed with 4 ꢀ 5 mL of toluene, and
the product was dissolved in CH2Cl2 and filtered, to yield after
solvent evaporation in vacuo the microcrystalline palladium
complexes.
Figure 8. Occlusion of the quinolylidene ligand by PPh3 ligands
in 9a.
differences are well reflected in our experimental observa-
tions in various pyridylidene complexes of Pd and Ni.
Experimental Section
Procedure D: Preparation of trans-Chloro(pyridylidene)bis-
(triphenylphosphine)nickel(II) Triflate Complexes. A small ex-
cess (1.1 molar equiv) of Ni(PPh3)4 and the respective triflate salt
were suspended in THF (20 mL), and the mixture was stirred at
room temperature for 17 h. The resulting yellow precipitate in a
brown solution was filtered through Celite and washed with 3 ꢀ
5 mL of toluene. The product was dissolved in CH2Cl2, filtered,
and dried under high vacuum.
General Procedures. All manipulations were performed in a
dry argon atmosphere using standard Schlenk techniques. Pd-
75
(PPh3)4 was prepared using a previously published method.
All other chemicals were commercially available. Diethyl ether,
hexane, and pentane were freshly distilled under N2 from sodium
wire and CH2Cl2 from CaH2. Melting points (uncorrected) were
determined on a Stuart SMP3 apparatus. NMR spectra were
determinedonaVarian300FTorINOVA600MHzspectrometer
(1H NMR at 300/600 MHz, 13C{1H} NMR at 75/151 MHz, and
2-Chloro-1-methylpyridinium Triflate, 2a. According to pro-
cedure A: 88.0% yield of colorless microcrystalline material; mp
163.3-165.0 ꢀC. 1H NMR (CD2Cl2, 20 ꢀC): δ 4.49 (3H, s, NMe),
8.02 (1H, m, H5), 8.06 (1H, dd, 3J = 8.3 Hz, 4J = 1.3 Hz, H3),
8.45 (1H, td, 3J = 8.3 Hz, 3J = 1.7 Hz, H4), 9.18 (1H, dd, 3J =
6.2 Hz, 3J = 1.7 Hz, H6). 13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 48.5
(s, NMe), 127.1 (s, C5), 130.0 (s, C3), 147.3 (s, C4), 149.4 (bs, C2
and C6), CF3 not observed. Anal. Calcd for C7H7NO3SClF3
1
31P{1H} NMR at 121.5/243 MHz). H and 13C chemical shifts
refer to δTMS = 0.00 according to the residual deuterated solvents
and are reported in ppm. 31P chemical shifts are reported in ppm
relative to an 85% H3PO4 external standard. FAB-MS determi-
nations were performed on a VG 70SEQ mass spectrometer in a
2-nitrobenzyl alcohol matrix at the University of Witwatersrand,
South Africa.
Procedure A: Preparation of Methylated Triflate Salts. A
small excess (1.1 molar equiv) of methyl trifluoromethanesulfo-
nate was added dropwise to a solution of the respective pyridine/
quinoline precursor in 20 mL of CH2Cl2. The solution was
stirred for 18 h at room temperature, after which the solvent was
removed by cannula. The resulting precipitate was washed with
ether (3 ꢀ 10 mL) and dried under high vacuum for a few hours.
For the synthesis of the 4-substituted salt, 4-chloropyridi-
(277.65 g mol-1): C 30.28, H 2.54, N 5.04. Found: C 30.01, H
3
2.41, N 4.93.
3-Chloro-1-methylpyridinium triflate, 2b. According to proce-
dure A: 98.5% yield of colorless microcrystalline material; mp
60.8-62.3 ꢀC. 1H NMR (CD2Cl2, 20 ꢀC): δ 4.63 (3H, s, NMe),
8.29 (1H, dd, 3J = 8.3 Hz, 3J = 6.3 Hz, H5), 8.81 (1H, dm, 3J =
8.3 Hz, 4J = 0.5 Hz, H6), 9.15 (1H, d, 3J = 6.3 Hz, H4), 9.38 (1H,
d, 4J = 0.5 Hz, H2). 13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 49.7 (s,
NMe), 130.1 (s, C5), 136.1 (s, C3), 146.1 (s, C6), 146.4 (s, C4),
146.7 (s, C2), CF3 not observed. MS (FAB): m/z (%) 130.1 (35)
[M - CF3SO3]þ, (37Cl); 128.1 (100) [M - CF3SO3]þ, (35Cl).
ne HCl (5.00 g, 33.3 mmol) was dissolved in 50 mL of distilled
water, and 40 mL of 1 M NaOH was added. The aqueous
3
Anal. Calcd for C7H7NO3SClF3 (277.65 g mol-1): C 30.28, H
3
solution was extracted with ether (3 ꢀ 40 mL). The organic
2.54, N 5.04. Found: C 30.80, H 2.30, N 5.09.
4-Chloro-1-methylpyridinium triflate, 2c. According to proce-
dure A: 55.5% yield of colorless microcrystalline material; mp
135.1-136.6 ꢀC. 1H NMR (CD2Cl2, 20 ꢀC): δ 4.45 (3H, s, NMe),
(75) Ozawa, F. In Synthesis of Organometallic Compounds; Komiya,
S., Ed.; John Wiley & Sons: Chichester, 1997.

5830 Organometallics, Vol. 29, No. 22, 2010
Stander-Grobler et al.
8.01 (2H, d, 3J = 6.8 Hz, H3 and H5), 8.82 (2H, d, 3J = 6.8 Hz,
H2 and H6). 13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 49.0 (s, NMe),
129.5 (s, C3 and C5), 147.0 (bs, C2 and C6), 155.3 (s, C4), CF3 not
observed. 31P{1H} NMR (CD2Cl2, 20 ꢀC): δ 23.0 (s, PPh3). MS
(FAB): m/z (%) 711.8 (8) [M - CF3SO3]þ, (37Cl, 58Ni); 449.6 (70)
[M - PPh3 - CF3SO3]þ; 413.0 (24) [M - PPh3 - Cl - CF3SO3]þ.
observed. Anal. Calcd for C7H7NO3SClF3 (277.65 g mol-1): C
Anal. Calcd for C43H37NO3P2SClF3Ni (860.91 g mol-1): C 59.99,
3
3
30.28, H 2.54, N 5.04. Found: C 30.01, H 2.61, N 5.11.
trans-Chloro(2-hydro-1-methyl-2-pyridylidene)bis(triphenylpho-
sphine)palladium(II) Triflate, 3a. According to procedure C:
96.3% yield of colorless microcrystalline material; mp (dec):
H 4.33, N 1.63. Found: C 60.21; H 4.20; N 1.73.
trans-Chloro(4-hydro-1-methyl-4-pyridylidene)bis(triphenylphos-
phine)nickel(II) Triflate, 4c. According to procedure D: 45.7%
1
yield of yellow microcrystalline material; mp (dec) 153 ꢀC. H
1
248 ꢀC. H NMR (CD2Cl2, 20 ꢀC): δ 3.96 (3H,s, NMe), 7.00
NMR (CD2Cl2, 20 ꢀC): δ 3.70 (3H, s, NMe), 6.91 (2H, d, 3J =5.9
Hz, H2 and H6), 7.40, 7.49, 7.68 (32H, m, PPh with obscured H3
and H5). 13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 46.1 (s, NMe), 128.8
(m, PPhmeta), 129.7 (m, PPhipso), 131.2 (s, PPhpara), 134.0 (s, C2 and
C6), 134.7 (m, PPhortho), 136.0 (bs, C3 and C5), 205.0 (t, 3J = 31.2
Hz, C4), CF3 not observed. 31P{1H} NMR (CD2Cl2, 20 ꢀC): δ 22.4
(s, PPh3). Anal. Calcd for C43H37NO3P2SClF3Ni (860.91
(2H, m, H4 and H5), 7.40, 7.52 (31H, m, PPh with obscured H6),
7.89 (1H, d, 3J = 6.0 Hz, H3). 13C{1H} NMR (CD2Cl2, 20 ꢀC): δ
52.8 (s, NMe), 122.3 (s, C5), 128.9(m,PPhipso), 129.3(m, PPhmeta),
131.8 (s, PPhpara), 134.3 (m, PPhortho), 137.3 (s, C4), 138.1 (t,
3JC-P=3.9 Hz, C3), 145.4 (s, C6), 189.3 (t, 2JC-P = 6.5 Hz, C2),
CF3 not observed. 31P{1H} NMR (CD2Cl2, 20 ꢀC): δ 22.7 (s,
PPh3). Anal. Calcd for C43H37NO3P2SClF3Pd (908.64 g mol-1):
g mol-1): C 59.99, H 4.33, N 1.63. Found: C 60.21, H 4.23, N 1.54.
3
3
2,4-Dichloro-1-methylpyridinium Triflate, 5. According to
procedure A: 99.8% yield of colorless microcrystalline material;
C 56.84, H 4.10, N 1.54. Found: C 57.03, H 4.19, N 1.60.
trans-Chloro(3-hydro-1-methyl-3-pyridylidene)bis(triphenylphos-
phine)palladium(II) Triflate, 3b. According to procedure C: 90.8%
yield of colorless crystals; mp (dec): 195 ꢀC. Single crystals suitable
for structure determination were obtained by dissolving the crude
mixture in CH2Cl2 and allowing pentane to diffuse slowly into the
1
mp 95.8-97.7 ꢀC. H NMR (CD2Cl2, 20 ꢀC): δ 4.42 (3H, s,
NMe), 7.99 (1H, dd, 3J = 6.8 Hz, 4J = 2.3 Hz, H5), 8.05 (1H, d,
4J = 2.3 Hz, H3), 9.14 (1H, dd, 3J = 6.8 Hz, 5J = 0.4 Hz, H6).
13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 48.1 (s, NMe), 127.7 (s, C5),
130.1 (s, C3), 148.9 (s, C4), 150.0 (s, C6), 156.0 (s, C2), CF3 not
observed. MS (FAB): m/z (%) 163.0 (77) [M CF3SO3]þ, (37Cl);
161.9 (100) [M - CF3SO3]þ, (35Cl). Anal. Calcd for C7H6-
1
mixture at -20 ꢀC. H NMR (CD2Cl2, 20 ꢀC): δ 3.54 (3H, s,
NMe), 6.73 (1H, dd, 3J = 7.7 Hz, 3J = 6.0 Hz, H5), 7.26 (1H, bs,
H2), 7.40, 7.49, 7.58 (30H, m, PPh), 7.78 (1H, d, 3J = 7.7 Hz, H4),
7.82 (1H, d, 3J = 6.0 Hz, H6). 13C{1H} NMR (CD2Cl2, 20 ꢀC): δ
47.9 (s, NMe), 124.8 (s, C5), 129.0 (m, PPhipso), 129.4 (m, PPhmeta),
131.4 (s, PPhpara), 134.8 (m, PPhortho), 137.9 (s, C6), 148.0 (t, 3JCP
=4.9Hz, C2), 151.7 (t, 3JCP =3.5 Hz, C4), 165.0 (t, 2JCP =6.9Hz,
C3), CF3 not observed. 31P{1H} NMR (CD2Cl2, 20 ꢀC): δ 24.9 (s,
PPh3). MS (FAB): m/z (%) 759.6 (52) [M - CF3SO3]þ, (35Cl,
106Pd); 498.0 (66) [M - PPh3 - CF3SO3]þ; 460.1 (21) [M - PPh3 -
Cl - CF3SO3]þ; 384.7 (10) [M - 2PPh3 - Cl - CF3SO3]þ. Anal.
NO3SCl2F3 (312.09 g mol-1): C 26.94, H 1.94, N 4.49. Found:
3
C 26.88, H 1.99, N 4.57.
2,4-Dichloro-1-hydropyridinium Triflate. According to proce-
dure B: 85.7% yield of colorless hygroscopic microcrystalline
material; mp 115.4-117.0 ꢀC. 1H NMR (CD2Cl2, 20 ꢀC): δ 7.94
(2H, m, H3 and H5), 8.79 (1H, m, H6), NH not observed.
13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 126.8 (s, C5), 128.7 (s, C3),
144.7 (s, C6), 146.6 (s, C2), 157.0 (s, C4), CF3 not observed. MS
(FAB): m/z (%) 147.9 (100) [M - CF3SO3]þ, (35Cl). Anal. Calcd
Calcd for C43H37NO3P2SClF3Pd (908.64 g mol-1): C 56.84, H
3
for C6H4NO3SCl2F3 (298.07 g mol-1): C 24.18, H 1.35, N 4.70.
4.10, N 1.54. Found: C 56.99, H 4.01, N 1.60.
trans-Chloro(4-hydro-1-methyl-4-pyridylidene)bis(triphenylphos-
phine)palladium(II) Triflate, 3c. According to procedure C: 99%
3
Found: C 23.29, H 1.41, N 4.54.
trans-Chloro(2-chloro-4-hydro-1-methyl-4-pyridylidene)bis-
(triphenylphosphine)palladium(II) Triflate, 6. According to
procedure C: 74.3% yield of colorless crystals; mp (dec)
200 ꢀC. 1H NMR (CD2Cl2, 20 ꢀC): δ 3.82 (3H, s, NMe), 7.16
(1H, bs, H3), 7.41 (12H, m, PPhmeta), 7.49 (7H, m, PPhpara with
obscured H5), 7.63 (12H, m, PPhortho), 7.69 (1H, d, 3J=6.4 Hz,
H6). 13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 45.4 (s, NMe), 128.9
(m, PPhmeta), 129.3 (m, PPhipso), 131.5 (s, PPhpara), 134.6 (t,
3JCP=3.7 Hz, C5), 134.8 (m, PPhortho), 136.7 (t, 3JCP = 3.7 Hz,
1
yield of colorless microcrystalline material; mp (dec) 189 ꢀC. H
NMR (CD2Cl2, 20 ꢀC): δ 3.80 (3H, s, NMe), 7.16 (2H, d, 3J = 6.4
Hz, H2 and H6), 7.39 (14H, m, PPh with obscured H3 and H5),
7.50, 7.62 (18H, m, PPh). 13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 46.5
(s, NMe), 129.2 (m, PPhmeta), 130.0 (m, PPhipso), 131.7 (s, PPhpara),
135.2 (m, PPhortho), 136.8 (bs, C3 and C5), 137.4 (s, C2 and C6),
197.7 (t, 2J=6.5 Hz, C4), CF3 not observed. 31P{1H} NMR (CD2-
Cl2, 20 ꢀC): δ 23.6 (s, PPh3). Anal. Calcd for C43H37NO3P2SClF3-
2
C3), 139.2 (s, C2), 140.0 (s, C6), 202.3 (t, JCP = 6.1 Hz, C4),
Pd (908.64 g mol-1): C 56.84, H 4.10, N 1.54. Found: C 56.61, H
3
CF3 not observed. 31P{1H} NMR (CD2Cl2, 20 ꢀC): δ 23.6 (s,
PPh3). MS (FAB): m/z (%) 794.0 (41) [M - -CF3SO3]þ, (35Cl,
106Pd); 758.6 (4) [M - Cl - CF3SO3]þ; 531.9 (46) [M - PPh3 -
CF3SO3]þ; 496.2 (8) [M - Cl - PPh3 - CF3SO3]þ. Anal. Calcd
3.98, N 1.59.
trans-Chloro(2-hydro-1-methyl-2-pyridylidene)bis(triphenylphos-
phine)nickel(II) Triflate, 4a. According to procedure D: 58.9%
1
yield of yellow microcrystalline material; mp (dec) 187 ꢀC. H
for C43H36NO3P2SCl2F3Pd (943.09 g mol-1): C 54.76; H
NMR (CD2Cl2, 20ꢀC): δ4.22 (3H, s, NMe), 6.81 (1H, m, H5), 6.91
(1H, td, 3J = 7.7 Hz, 4J = 1.4 Hz, H4), 7.42, 7.57 (31H, m, PPh
with obscured H6), 7.85 (1H, m, H3). 13C{1H} NMR (CD2Cl2,
20 ꢀC): δ 52.4 (s, NMe), 121.5 (s, C5), 128.9 (m, PPhipso), 129.4 (m,
PPhmeta), 131.8 (s, PPhpara), 134.5 (m, PPhortho), 134.7 (s, C4), 138.2
3
3.85; N 1.49. Found: C 54.57; H 3.78; N 1.55.
trans-Chloro(2-chloro-1,4-dihydro-4-pyridylidene)bis(triphenyl-
phosphine)palladium(II) Triflate. According to procedure C: No
pure material could be obtained. NMR signals of the main
product are overlapping with signals from impurities; thus shifts
were unequivocally assigned by gHSQC and gHMQC measure-
ments. 1H NMR (CD2Cl2, 20 ꢀC): δ 7.15 (s, H3), 7.18 (H6), 7.37
(H5), 7.41 (m, PPhmeta), 7.50 (m, PPhpara), 7.63 (m, PPhortho), NH
3
(bs, C3), 146.0 (s, C6), 193.6 (t, JCP = 33.0 Hz, C2), CF3 not
observed. 31P{1H} NMR (CD2Cl2, 20 ꢀC): δ 21.2 (s, PPh3). Anal.
Calcd for C43H37NO3P2SClF3Ni (860.91 g mol-1): C 59.99, H
4.33, N 1.63. Found: C 60.32, H 4.24, N 1.49.
3
not observed. 13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 128.9 (m,
trans-Chloro(3-hydro-1-methyl-3-pyridylidene)bis(triphenylphos-
phine)nickel(II) Triflate, 4b. According to procedure D: 49.6%
yield of light yellow crystals; mp (dec) 135.0-138.0 ꢀC. Yellow
crystals, suitable for single-crystal X-ray structure determination,
were obtained by slow diffusion of pentane into a concentrated
CH2Cl2 solution at -20 ꢀC. 1H NMR (CD2Cl2, 20 ꢀC): δ 3.36 (3H,
s, NMe), 6.69 (1H, m, H5), 7.11 (1H, bs, H4), 7.40, 7.66 (31H, m,
PPh with obscured H6), 8.09 (1H, d, 4J = 4.4 Hz, H2). 13C{1H}
NMR (CD2Cl2, 20 ꢀC): δ 47.7 (s, NMe), 123.0 (s, C5), 128.9 (m,
PPhmeta), 129.7 (m, PPhipso), 131.2 (s, PPhpara), 134.7 (m, PPhortho),
136.5 (s, C6), 148.2 (s, C4), 150.7 (bs, C2), 174.1 (bs, C3), CF3 not
3
PPhmeta), 129.2 (m, PPhipso), 131.7 (s, PPhpara), 133.2 (t, JCP
=
3.6 Hz, C5), 134.6 (s, C6), 134.7 (m, PPhortho), 135.3 (t, 3JCP = 4.3
Hz, C3), 137.4 (s, C2), 204.6 (t, JCP = 6.1 Hz, C4), CF3 not
2
observed. 31P{1H} NMR (CD2Cl2, 20 ꢀC): δ 24.0 (s, PPh3). MS
(FAB): m/z (%) 780.3 (6) [M - CF3SO3]þ, (35Cl, 106Pd); 744.5 (3)
[M - Cl - CF3SO3]þ; 482.2 (15) [M - Cl - PPh3 - CF3SO3]þ;
446.2 (3) [M - 2Cl - PPh3 - CF3SO3]þ.
trans-Chloro(2-chloro-4-hydro-1-methyl-4-pyridylidene)bis-
(triphenylphosphine)nickel(II) Triflate, 7. According to proce-
dure D: 62.4% yield of yellow crystals; mp (dec) 160 ꢀC.

Article
Organometallics, Vol. 29, No. 22, 2010 5831
1H NMR (CD2Cl2, 20 ꢀC): δ 3.71 (3H, s, NMe), 7.16 (1H, bs,
trans-Chloro(7-chloro-4-hydro-1-methyl-4-quinolylidene)bis-
(triphenylphosphine)nickel(II) Triflate, 10a. According to pro-
cedure D: 80.4% yield of yellow crystals; mp (dec) 176 ꢀC. 1H
NMR (CD2Cl2, 20 ꢀC): δ 3.90 (3H, s, NMe), 7.29 (12H, m,
PPhmeta), 7.42 (7H, m, PPhpara with obscured H8), 7.59 (13H,
m, PPhortho with obscured H6), 7.66 (1H, d, 3J = 6.3 Hz, H3),
7.89 (1H, d, 3J = 6.3 Hz, H5), 9.44 (1H, d, 3J = 8.8 Hz, H2).
13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 43.0 (s, NMe), 117.2 (s, C8),
128.6 (m, PPhmeta), 128.8 (s, C6), 129.3 (m, PPhipso), 130.6 (bs,
C3), 131.2 (s, PPhpara), 133.3 (s, C9), 134.6 (m, PPhortho), 135.3
H3), 7.41 (12H, m, PPhmeta), 7.48 (7H, m, PPhpara with
3
obscured H6), 7.69 (12H, m, PPhortho), 7.76 (1H, d, J=6.4
Hz, H5). 13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 45.0 (s, NMe),
128.9 (m, PPhmeta), 129.4 (m, PPhipso), 131.5 (s, PPhpara), 133.8
(s, C5), 134.7 (m, PPhortho), 136.6 (s, C3), 136.8 (s, C6), 136.9 (s,
C2), 210.8 (s, C4), CF3 not observed. 31P{1H} NMR (CD2Cl2,
20 ꢀC): δ 22.5 (s, PPh3). MS (FAB): m/z (%) 746.3 (40) [M -
CF3SO3]þ, (35Cl, 58Ni); 710.8 (9) [M - Cl - CF3SO3]þ; 484.0
(12) [M - PPh3 - CF3SO3]þ. Anal. Calcd for C43H36NO3P2-
SCl2F3Ni (895.36 g mol-1): C 57.68; H 4.05; N 1.56. Found: C
(s, C10), 136.6 (s, C5), 137.9 (s, C2), 140.9 (s, C7), 216.7 (t, 2JCP
=
3
57.45; H 4.20; N 1.49.
30.5 Hz, C4), CF3 not observed. 31P{1H} NMR (CD2Cl2,
20 ꢀC): δ 21.9 (s, PPh3). MS (FAB): m/z (%) 796.2 (19) [M -
CF3SO3]þ, (35Cl, 58Ni); 533.9 (68) [M - PPh3 - CF3SO3]þ. Anal.
4,7-Dichloro-1-methylquinolinium Triflate, 8a. According to
procedure A: 99.0% yield of colorless microcrystalline material;
mp 108-111 ꢀC. 1H NMR (CD2Cl2, 20 ꢀC): δ 4.65 (3H, s, NMe),
8.03 (1H, dd, 3J = 9.0 Hz, 4J = 1.8 Hz, H6), 8.12 (1H, d, 3J = 6.5
Hz, H3), 8.36 (1H, d, 4J = 1.8 Hz, H8), 8.56 (1H, d, 3J = 9.0 Hz,
Calcd for C47H38NO3P2SCl2F3Ni (945.42 g mol-1): C 59.71; H
3
4.05; N 1.48. Found: C 60.07; H 4.26; N 1.39.
trans-Chloro(7-chloro-1,4-dihydro-4-quinolylidene)bis(triphenyl-
phosphine)nickel(II) Triflate, 10b. According to procedure D:
68.1% yield of yellow crystals; mp (dec) 197 ꢀC. 1H NMR (CD2-
Cl2, 20 ꢀC): δ 7.28 (13H, m, PPh with obscured H3), 7.40 (6H, m,
PPh), 7.48 (1H, dd, 3J = 8.8 Hz, 4J = 1.9 Hz, H6), 7.55 (13H, m,
PPh with obscured H8), 7.77 (1H, d, 3J = 5.8 Hz, H2), 9.19 (1H, d,
3J = 8.8 Hz, H5), not observed NH. 13C{1H} NMR (CD2Cl2,
20 ꢀC): δ 120.3 (s, C8), 128.6 (m, PPhmeta), 128.7 (s, C6), 129.3 (m,
PPhipso), 129.6 (t, 3JCP = 3.7 Hz, C3), 131.3 (s, PPhpara), 132.0 (s,
C2), 132.0 (s, C9), 134.4 (m, PPhortho), 134.7 (bs, C10), 134.9 (s, C5),
3
H5), 9.44 (1H, d, J = 6.5 Hz, H2). 13C{1H} NMR (CD2Cl2,
20 ꢀC): δ 46.4 (s, NMe), 119.1 (s, C8), 123.2 (s, C3), 126.8 (s, C10),
128.7 (s, C5), 132.7 (s, C6), 139.9 (s, C9), 144.4 (s, C7), 151.2 (s,
C2), 154.6 (s, C4), CF3 not observed. MS (FAB): m/z (%) 214.0
(69) [M - CF3SO3]þ, (37Cl); 212.0 (100) [M - CF3SO3]þ, (35Cl).
Anal. Calcd for C11H8NO3SCl2F3 (362.15 g mol-1): C 36.48, H
3
2.23, N 3.87. Found: C 36.26, H 2.29, N 3.78.
4,7-Dichloro-1-hydroquinolinium Triflate, 8b. According to
procedure B: 98.0% yield of colorless microcrystalline material;
mp 146.4-147.2 ꢀC. 1H NMR (CD2Cl2, 20 ꢀC): δ 8.00 (1H, dd,
3J = 9.0 Hz, 4J = 2.0 Hz, H6), 8.08 (1H, d, 3J = 6.0 Hz, H3), 8.47
(2H, m, H5 and H8), 9.10 (1H, d, 3J = 6.0 Hz, H2), not observed
NH. 13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 121.3 (s, C8), 122.8 (s,
C3), 126.4 (s, C10), 127.3 (s, C5), 133.2 (s, C6), 139.1 (s, C9), 143.9
(s, C7), 155.1 (s, C2), 145.1 (s, C4), CF3 not observed. MS (FAB):
m/z (%) 198.0 (100) [M - H - CF3SO3]þ, (35Cl). Anal. Calcd for
2
140.0 (s, C7), 216.6 (t, JCP = 30.0 Hz, C4), CF3 not observed.
31P{1H} NMR (CD2Cl2, 20 ꢀC): δ 22.6 (s, PPh3). MS (FAB): m/z
(%) 782.1 (3) [M - CF3SO3]þ, (35Cl, 58Ni); 746.9 (1) [M - Cl -
CF3SO3]þ; 520.0 (10) [M - Cl - PPh3 - CF3SO3]þ. Anal. Calcd
for C46H36NO3P2SCl2F3Ni (931.39 g mol-1): C 59.32, H 3.90, N
3
1.50. Found: C 59.11, H 4.02, N 1.63.
Typical Procedure for the Suzuki-Miyaura Coupling. The
reactions were performed under inert conditions using dry and
degassed reagents. Phenylboronic acid (3 mmol, 0.3658 g),
bromoacetophenone (2 mmol, 0.3981 g), potassium carbonate/
cesium carbonate (4 mmol, 0.5528/1.303 g), diethyleneglycol-
di-n-butyl ether (2 mmol, 0.4367 g, 0.5 mL), and dimethyl-
acetamide (5 mL) were placed in a 25 mL three-neck round-
bottom flask. The flask was connected to N2 and equipped with
a reflux condenser and a septum. The catalyst solution was
added after heating to 130 ꢀC. Aliquots (0.2 mL) were taken at
regular intervals from the reaction mixture and added to 5 mL of
CH2Cl2. The organic layer was washed with 3 ꢀ 5 mL of water,
dried over MgSO4, and filtered. The solvent was removed
in vacuo, and the residue was analyzed by 1H NMR spectroscopy.
Crystal Structure Determinations. Data associated with the
crystal structures are summarized in Tables 4 and 1S, Support-
ing Information). Intensity data were collected at T=100 K with
a Bruker SMART Apex diffractometer with graphite-mono-
C10H6NO3SCl2F3 (348.13 g mol-1): C 34.50, H 1.74, N 4.02.
3
Found: C 34.28, H 1.81, N 4.12.
trans-Chloro(7-chloro-4-hydro-1-methyl-4-quinolylidene) bis-
(triphenylphosphine)palladium(II) Triflate, 9a. According to
procedure C: 62.9% yield of colorless crystals; mp (dec)
204 ꢀC. 1H NMR (CD2Cl2, 20 ꢀC): δ 4.03 (3H, s, NMe), 7.28
(12H, m, PPhmeta), 7.40 (6H, m, PPhpara), 7.54 (14H, m,
PPhortho with obscured H6 and H8), 7.64 (1H, d, 3J=6.4 Hz,
H3), 7.95 (1H, d, 3J=6.4 Hz, H5), 8.74 (1H, d, 3J = 8.8 Hz, H2).
13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 43.5 (s, NMe), 117.2 (s, C8),
128.7 (m, PPhmeta), 129.3 (m, PPhipso with C6), 130.6 (t, 3JCP
=
3.6 Hz, C3), 131.4 (s, PPhpara), 134.5 (bs, C10), 134.6 (m,
PPhortho), 135.1 (s, C9), 137.0 (s, C5), 141.2 (s, C7), 141.8 (s,
C2), 206.6 (t, 2JCP = 5.5 Hz, C4), CF3 not observed. 31P{1H}
NMR (CD2Cl2, 20 ꢀC): δ 23.2 (s, PPh3). MS (FAB): m/z (%)
844.0 (74) [M - CF3SO3]þ, (35Cl, 106Pd); 581.9 (100) [M -
PPh3 - CF3SO3]þ; 546.0 (13) [M - Cl - PPh3 - CF3SO3]þ;
˚
404.1 (54) [M-L - PPh3-CF3SO3]þ. Anal. Calcd for C47H38
-
chromated Mo KR radiation (λ = 0.71073 A). Intensities were
NO3P2SCl2F3Pd (993.14 g mol-1): C 56.84; H 3.86; N 1.41.
Found: C 56.66; H 3.99; N 1.32.
measured using the ω-scan mode and were corrected for Lorentz
and polarization effects.76-79 The structures were solved by
direct methods (SHELXS-97) and refined by full-matrix least-
squares on F2 (SHELXL-97).80 Non-hydrogen atoms were
refined with anisotropic displacement parameters. Hydrogen
atoms were placed in idealized positions and refined using a
riding model with fixed isotropic contributions if possible.
Reinvestigation of the crystal data of 3b (in the BF4 salt,
previously published)20 yielded a so far unnoticed disorder
(sof: 0.2) of the carbene ligand. Refinement of the less occupied
site was supported by restraining its geometry to fit the major
ligand site using SAME. All crystals were found to contain
3
trans-Chloro(7-chloro-1,4-dihydro-4-quinolylidene)bis(triphenyl-
phosphine)palladium(II) Triflate, 9b. According to procedure C:
85.3% yield of colorless microcrystalline material; mp (dec) 207
ꢀC. 1H NMR (CD2Cl2, 20 ꢀC): δ 7.27 (12H, m, PPh), 7.41 (6H, m,
PPh), 7.43 (1H, dd, 3J = 8.8 Hz, 4J = 2.0 Hz, H6), 7.49 (1H, d,
3J = 5.8 Hz, H2), 7.52, (13H, m, PPh with obscured H3), 7.69 (1H,
d, 4J = 2.0 Hz, H8), 8.55 (1H, d, 3J = 8.8 Hz, H5), not observed
NH. 13C{1H} NMR (CD2Cl2, 20 ꢀC): δ 120.2 (s, C8), 128.7 (m,
PPhmeta), 129.2 (m, PPhipso), 129.3 (s, C6), 129.4 (s, C9), 129.6 (t,
3JCP=4.3 Hz, C3), 131.5 (s, PPhpara), 133.8 (t, 3JCP = 2.5 Hz, C10),
134.5 (m, PPhortho), 135.2 (s, C5), 135.6 (s, C2), 140.2 (s, C7), 207.2
(t, 2JCP = 5.5 Hz, C4), CF3 not observed. 31P{1H} NMR (CD2Cl2,
20 ꢀC): δ 23.8 (s, PPh3). MS (FAB): m/z (%) 830.4 (5) [M -
CF3SO3]þ, (35Cl, 106Pd); 794.5 (1) [M - Cl - CF3SO3]þ; 568.3 (5)
[M - PPh3 - CF3SO3]þ; 532.3 (1) [M - Cl - PPh3 - CF3SO3]þ.
(76) SADABS (version 2.05); Bruker AXS Inc.: Madison, WI, 2002.
(77) SAINT: Data reduction software (version 6.45); Bruker AXS Inc.:
Madison, WI, 2003.
(78) SMART: Data collection software (version 5.629); Bruker AXS
Inc.: Madison, WI, 2003.
(79) Blessing, R. H. Acta Crystallogr. A 1995, 51, 33.
(80) Sheldrick, G. M. Acta Crystallogr. A 2008, 64, 112.
Anal. Calcd for C46H36NO3P2SCl2F3Pd (979.12 g mol-1): C
3
56.43, H 3.71, N 1.43. Found: C 56.22, H 3.87, N 1.46.

5832 Organometallics, Vol. 29, No. 22, 2010
Table 4. Crystal Data, Data Collection, and Structure Refinement Details
Stander-Grobler et al.
3b
4b
empirical formula
Mr
cryst syst
C42H37ClNP2Pd CF3O3S
908.59
C42H37ClNNiP2 CF3O3S
3
3
860.90
monoclinic
P2(1)/n
14.672(5)
18.609(6)
18.404(6)
90
107.724(5)
90
4786(3)
1.261
monoclinic
P2(1)/n
14.6778(18)
18.491(2)
18.317(2)
90
109.068(2)
90
4698.6(10)
1.217
space group
˚
a/A
˚
b/A
˚
c/A
R/deg
β/deg
γ/deg
3
˚
V/A
F
calc/g cm-3
Z
F(000)
4
1848
0.60
100
4
1776
0.63
100
μ(Mo KR/mm-1
T/K
)
refls measd
refls unique
refined params/restraints
24 537
23 577
8709 [Rint = 0.117]
497/1
0.068
8384 [Rint = 0.031]
497/0
0.042
R1 [I g2σ(I )]
wR2
a
0.181
0.113
weighting schemea
a = 0.0896
b = 0.000
1.34/-1.50
a = 0.0617
b = 0.1765
0.66/-0.33
-3
˚
fin(max./min.)/e A
σ
6 2CH2Cl2
3
7 CH2Cl2
3
9a
empirical formula
Mr
cryst syst
C
42H36Cl2NP2Pd CF3O3S 2(CH2Cl2)
C42H36Cl2NNiP2 CF3O3S CH2Cl2
C46H38Cl2NP2Pd CF3O3S
3
3
3
3
3
1112.88
monoclinic
P2(1)/c
17.6792(17)
16.9370(17)
16.3414(16)
90
106.954(2)
90
4680.5(8)
1.579
980.26
monoclinic
P2(1)/n
10.5898(7)
21.8258(14)
19.7422(13)
90
104.7130(10)
90
4413.4(5)
1.475
993.08
monoclinic
C2/c
25.183(6)
11.472(2)
34.952(7)
90
99.278(4)
90
9966(4)
1.324
space group
˚
a/A
˚
b/A
˚
c/A
R/deg
β/deg
γ/deg
3
˚
V/A
F
calc/g cm-3
Z
F(000)
4
2248
0.91
100
4
2008
0.86
100
8
4032
0.63
100
μ(Mo KR) (mm-1
T/K
)
refls measd
refls unique
refined params /restraints
24 156
23 126
26 296
731 [Rint = 0.034]
560/0
0.053
8048 [Rint = 0.028]
533/0
0.038
0.095
a = 0.048
b = 3.1369
9103 [Rint = 0.047]
542/2
0.042
R1 [I g2σ(I )]
wR2
a
0.120
0.097
weighting schemea
a = 0.0457
b = 17.9112
1.67/-1.56
a = 0.0534
b = 0.000
0.91/ -0.54
-3
˚
σ
fin(max./min.)/e A
0.72/-0.59
P
P
a wR2 = { [w(Fo2 - Fc2)2]/ [w(Fo2)2]}1/2; w = 1/[σ2(Fo2) þ (ap)2 þ bp]; p = (Fo þ 2Fc2)/3.
2
solvent. In 3b, 4b, and 9a the dicholormethane molecules
could not be resolved due to disorder and were thus removed
by the SQUEEZE routine in PLATON.81 Detailed information
about the solvent-accessible void and removed electrons can be
found in the CIF along with a complete list of displacement
parameters and tables of interatomic distances and angles.
These have been deposited with the Cambridge Crystallo-
graphic Data Centre and can be obtained free of charge via www.
ccdc.cam.ac.uk/data_request/cif on quoting CCDC 796954 to
796958 and 797119.
using uncontracted Slater-type orbitals (STOs) with TZ2P
quality as basis functions for the SCF calculations.84 An aux-
iliary set of s, p, d, f, and g STOs was used to fit the molecular
densities and to represent the Coulomb and exchange potentials
accurately in each SCF cycle.85 Scalar relativistic effects have
been incorporated by applying the zeroth-order regular approx-
imation (ZORA).86 The nature of the stationary points on the
potential energy surface was determined by calculating the
vibrational frequencies. All structures are minima on the poten-
tial energy surface.
The bonding situation of the metal-carbene bonds was
investigated by an energy decomposition analysis that was
developed by Morokuma87 and by Ziegler and Rauk.88,89 The
Computational Section
With the exception of M6, M7, M60, and M70 (BP86/TZVPP),
all geometries were optimized under Cs-symmetry constraint
using density functional theory at the BP86 level of theory82,83
(84) Snijders, J. G.; Vernooijs, P.; Baerends, E. At. Data Nucl. Data
Tables 1981, 483.
(85) Krijn, J.; Baerends, E. J., Fit Functions in the HFS-Method;
Internal Report (in Dutch); Vrije Universiteit Amsterdam, The Netherlands,
1984.
(81) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7.
(82) Becke, A. D. Phys. Rev. A 1988, 38, 3098.
(86) vanLenthe, E.; vanLeeuwen, R.; Baerends, E. J.; Snijders, J. G.
(83) Perdew, J. Phys. Rev. B: Condens. Matter 1986, 34, 7406.
Int. J. Quantum Chem. 1996, 57, 281.

Article
Organometallics, Vol. 29, No. 22, 2010 5833
bonding analysis focuses on the instantaneous interaction en-
ergy ΔEint of a bond A-B between two fragments A and B in the
particular electronic reference state and in the frozen geometry
of AB. This interaction energy is divided into three main
components [eq 1].
electronic and geometrical ground states, must be added to
ΔEint [eq 2].
ΔEð¼ - DeÞ ¼ ΔEint þ ΔEprep
ð2Þ
To calculate the dissociation energies, we calculated each
fragment in its optimized geometry and derived ΔE by eq 2.
Further details on the EDA can be found in the literature.90,92
The EDA has been used by us for a comprehensive study of
metal-ligand interactions in transition metal complexes.93,94
ΔEint ¼ ΔEelstat þ ΔEPauli þ ΔEorb
ð1Þ
The term ΔEelstat corresponds to the classical electrostatic
interaction between the unperturbed charge distributions of
the prepared atoms and is usually attractive. The Pauli
repulsion, ΔEPauli, is the energy change associated with the
transformation from the superposition of the unperturbed
Acknowledgment. We are very grateful to Dr. Ralf
Tonner, whose additional calculations helped to settle
questions that arose during the peer-review process. We
also thank the South African National Research Founda-
tion for financial support. E.S. thanks SASOL for a Ph.D.
scholarship, O.S., S.C., and H.G.R. appreciate support
from the Alexander von Humboldt Foundation, and OS
thanks the German Academy of Sciences Leopoldina
(BMBF LPD 9901/8-179). G.H. further thanks the
DAAD (grant number A/05/52686) for postdoctoral
funding. M.A. is very grateful for an Alfred Werner
Assistant Professorship.
electron densities FA þ FB of the isolated fragments to the
0
^
ˆ
wave function Ψ =N A [ΨAΨB], which properly obeys the
^
Pauli principle through explicit antisymmetrization (A oper-
ator) and renormalization of the product wave function.90 It
comprises the destabilizing interactions between electrons of
the same spin on either fragment. The orbital interaction
ΔEorb accounts for charge transfer and polarization effects.91
The ΔEorb term can be decomposed into contributions from
each irreducible representation of the point group of the
interacting system. This makes it possible to estimate the
intrinsic strength of orbital interactions from orbitals having
a0 (σ) and a00 (π) symmetry quantitatively. To obtain the
bond dissociation energy, De, the preparation energy,
ΔEprep, which gives the relaxation of the fragments into their
Supporting Information Available: Crystallographic details as
CIF; ORTEP figure of 3a (in the BF4 salt) with resolved
disorder; correlation of calculated charges and 13C NMR data
in chloropyridines; average geometries of chloropyridinium
salts; Cartesian coordinates for all calculated structures. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(87) Morokuma, K. J. Chem. Phys. 1971, 55, 1236.
(88) Ziegler, T.; Rauk, A. Inorg. Chem. 1979, 18, 1755.
(89) Ziegler, T.; Rauk, A. Inorg. Chem. 1979, 18, 1558.
(90) Bickelhaupt, F. M.; Baerends, E. J. In Reviews in Computational
Chemistry, Vol. 15; Wiley-VCH: New York, 2000.
(91) Bickelhaupt, F. M.; Nibbering, N. M. M.; van Wezenbeek,
E. M.; Baerends, E. J. Phys. Chem. 1992, 96, 4864.
(92) Velde, G. T.; Bickelhaupt, F. M.; Baerends, E.; Guerra, C. F.;
Van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J. Comput. Chem.
2001, 22, 931.
€
(93) Frenking, G.; Wichmann, K.; Frohlich, N.; Loschen, C.; Lein,
M.; Frunzke, J.; Rayon, V. M. Coord. Chem. Rev. 2003, 238, 55.
(94) Lein, M.; Frenking, G. In Theory and Application of Computa-
tional Chemistry: The First Forty Years; Dykstra, E., Frenking, G., Kim,
K. S., Scuseria, G. E., Eds.; Elsevier: Amsterdam, 2005.