Cyclopalladated acetate dimers: Crystal structures and VT-NMR

Article abstract of DOI:10.1016/j.jorganchem.2005.11.066

The X-ray structure of two cyclopalladated acetate bridged dimers has been solved, and they are shown to exhibit an open-book type core, analogous to those that have been previously reported. Variable temperature NMR studies on these molecules demonstrates that this core persists in chloroform solution, resulting in restrictions to the movement of ancillary parts of the molecule. The barrier to rotation of a pendant phenyl ring against this core has been measured: ΔH^? = 56 kJ mol^-1, and ΔS ^? = 0.

Full text of DOI:10.1016/j.jorganchem.2005.11.066

Journal of Organometallic Chemistry 691 (2006) 1251–1256
www.elsevier.com/locate/jorganchem
Cyclopalladated acetate dimers: Crystal structures and VT-NMR
Peter G. Evans, Neil A. Brown, Guy J. Clarkson,
*
Christopher P. Newman, Jonathan P. Rourke
Department of Chemistry, Warwick University, Coventry CV4 7AL, UK
Received 15 November 2005; received in revised form 24 November 2005; accepted 28 November 2005
Available online 3 January 2006
Abstract
The X-ray structure of two cyclopalladated acetate bridged dimers has been solved, and they are shown to exhibit an open-book type
core, analogous to those that have been previously reported. Variable temperature NMR studies on these molecules demonstrates that
this core persists in chloroform solution, resulting in restrictions to the movement of ancillary parts of the molecule. The barrier to rota-
tion of a pendant phenyl ring against this core has been measured: DHà = 56 kJ molꢀ1, and DSà = 0.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Cyclometallation; Cyclopalladation; Palladium; VT-NMR; X-ray structure
1. Introduction
with palladium acetate which proceeds cleanly at 60 ꢁC in
acetic acid to yield a cyclopalladated species in high yield
(Scheme 1). A cursory inspection of the NMR spectra
reveals nothing more unusual than overlapping peaks
and further purification is not found to be necessary: crys-
tals of a suitable standard for single-crystal X-ray diffrac-
tion could be isolated easily. A crystal of pure diphenyl
pyridine was also the subject of an X-ray diffraction study;
nothing unusual was revealed, but details are recorded here
for completeness (Fig. 1).
The study of the cyclometallation reaction is of consid-
erable interest, both from a mechanistic point of view [1]
and because of its role in the functionalisation of C–H
bonds [2–6]. We have recently published a couple of papers
dealing with the mono- and di-cycloplatination of diphe-
nylpyridines [7,8], and another one that dealt with the unu-
sual situation that can lead to the formation of a carbene
species [9]. Here, we discuss some aspects of the cyclopalla-
dation of diphenylpyridines, and the structure of the result-
ing complexes. The cyclopalladation reaction is of
particular interest due to the role of palladium catalysts
in C–C bond forming reactions [10], with a recent compu-
tational study identifying likely agostic intermediates in the
reaction pathway [11].
3. Solid-state structures
The crystal structures of both (1) and (2) were solved.
Whilst crystals of (1) were twinned and the structure
could not be refined to publication standard, it is clear
that it has the same principal features exhibited by the
structure of (2) (Fig. 2), which we will discuss in more
detail: selected bond lengths and angles are recorded in
Table 1. Both (1) and (2) exhibit dimeric structures with
the two halves related by a C2 rotation. At the heart of
the molecules are the two palladiums bridged by the
two acetates; these bridges fold the two halves of the mol-
ecule back on each other, rather like an open book. In the
two structures reported here we have a close Pd–Pd
2. Synthesis
Two new organopalladium compounds have been syn-
thesized, via the direct reaction of a substituted pyridine
*
Corresponding author. Tel.: +44 24 76523263.
E-mail address: j.rourke@warwick.ac.uk (J.P. Rourke).
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.11.066

1252
P.G. Evans et al. / Journal of Organometallic Chemistry 691 (2006) 1251–1256
Scheme 1.
Fig. 1. The X-ray structure of the diphenyl pyridine ligand.
˚
˚
contact of 2.9029(12) A in (1) and 2.9067(9) A in (2); the
planes of the cyclometallated rings end up being close to
parallel (inclined at 11.6(6) and11.8(3)ꢁ to each other in
(1) and (2), respectively), and the unmetallated phenyl
rings are inclined at angles of 59.4(3) and 43.9(3)ꢁ to
the pyridine ring. In the structure of (2) the Pd–N dis-
compounds with a carboxyl bridged core. The mean Pd–Pd
˚
distance in these structures is 2.952 A, with max and min
˚
values at 2.823 and 3.281 A (one outlier with a Pd–Pd dis-
˚
tance of 3.413 A is reported for the hugely sterically
crowded cyclopalladated di(orthotolyl)(t-butyl)phosphine
compound [15]). The question of when a Pd–Pd interaction
constitutes a bond seems clear: the covalent radius of a
˚
˚
tance is 2.074(5) A, the Pd–C distance is 1.969(6) A. This
Pd–C distance is significantly shorter that the average Pd–
C single bond, in line with the expectation that the five
membered metallocycle has significant aromatic character
[12]; other authors [13] have related short Pd–C bond dis-
tances with the p back bonding of the metallic centre to
the aromatic system. The Pd–O distances are 2.159(5)
(O trans to C) and 2.058(5) (O trans to N), with this dif-
ference being readily attributed to the greater trans influ-
ence shown by carbon, compared with nitrogen.
˚
square planar palladium is reported as being 1.31 A [16],
thus only with complexes having Pd–Pd distances of
˚
2.62 A or less does the question arise. All of the 78 struc-
tures mentioned above have Pd–Pd distances greater than
˚
2.62 A and thus none should be considered as having a
Pd–Pd bond. Interestingly, of these 78 structures, 14 (mean
˚
Pd–Pd distance = 2.872, min = 2.831, max = 2.939 A) are
shown as having a bond, whereas 64 (mean Pd–Pd dis-
˚
tance = 2.969, min = 2.823, max = 3.281 A) are shown as
A survey of the Cambridge Crystallographic Database
[14] reveals 78 similar structures of dimeric cyclopalladated
not having a bond. Our Pd–Pd distance is thus well within
the normal range.

P.G. Evans et al. / Journal of Organometallic Chemistry 691 (2006) 1251–1256
1253
Fig. 2. The X-ray structure of compound (2).
4. Variable temperature NMR
7.14 and 7.07 ppm on cooling. We measured the separation
of these two peaks to be 32 Hz (on a 500 MHz spectrome-
ter) with a coalescence temperature of 268 K. The only
plausible reason for the signal for proton W being different
from that of proton X (and for that of Y being different to
that of Z) is that the unmetallated phenyl ring cannot
rotate freely with respect to the pyridine ring (on the
NMR timescale) and that the two sides of the pyridine ring
are somehow different. These conditions can only be met if
the acetate bridges, together with the open book shape of
the core, persist in solution. Thus, we regard our results
as proof of exactly that: the cyclopalladated carboxyl
bridged dimers that are known to exist in the solid state
maintain their structure in chloroform solution.
1
A more detailed look at the solution H NMR data on
compounds (1) and (2) shows that some of the protons on
the non-palladated ring are so broad as to be lost in the
baseline at ambient temperature. A more in depth study
of these spectra was thus undertaken. Similar results were
obtained for both (1) and (2), but the less crowded nature
of the 1H spectrum of (2) made it more amenable to study.
Fig. 3 shows a stacked plot of some of the variable tem-
perature 1H NMR spectra of compound (2), in chloroform
solution, in the aromatic region of the spectrum; Fig. 4
shows the labelling system we have used. Only the signals
relating to the protons W, X, Y and Z change significantly
with temperature, and it is worth noting that we can unam-
biguously distinguish protons W and X from Y and Z from
the coupling to the 19F on the ring. At 333 K the signals for
protons W and X are a broad hump, but clearly identifiable
at around 7.5 ppm; as the temperature is decreased this sig-
nal gets broader (and disappears into the baseline at about
room temperature) before separating out into two clear sig-
nals at 7.84 and 7.20 ppm. We measured the separation of
these two peaks to be 323 Hz (on a 500 MHz spectrometer)
with a coalescence temperature of 293 K. The signal for
protons Y and Z are already at a high temperature limit
at room temperature, but separate into the two signals at
It is apparent that the two peaks for W and X are signif-
1
icantly different in the H NMR, and an inspection of the
X-ray structure suggests why: one of the hydrogens W
and X points directly at a fluorine in the other half of the
dimer, whereas the other one points into free space. Simi-
larly the two protons Y and Z are clearly pointing in differ-
ent directions, but both are largely pointing into free space,
and this results in the relatively small difference in their
chemical shifts.
We can use the coalesence temperatures observed to
extract values for the barrier to interconversion of the
two sets of signals, using standard techniques. The DGà

1254
P.G. Evans et al. / Journal of Organometallic Chemistry 691 (2006) 1251–1256
Table 1
observed represents the barrier to rotation of the unmetal-
lated phenyl ring against the pyridine ring. We would
expect this barrier to be greater than that that exists in
2,20-bipyridine (and that of the other, alkoxy substituted,
ring which does not show any additional complications in
the NMR at the temperatures we used) as the palladium
acetate bridged core provides additional steric constraints.
We would expect this barrier to be less than that that exists
for a trisubstituted biphenyl. The zero entropy of activa-
tion is also consistent with this intramolecular process.
We do not feel we need to invoke any additional agostic
type interaction of hydrogens W and X with the palladium
centre to account for the size of the barrier.
˚
Bond lengths (A) and angles (ꢁ) for (1)
Pd(1)–C(1)
Pd(1)–O(02)#1
Pd(1)–N(11)
Pd(1)–O(01)
Pd(1)–Pd(1)#1
C(1)–C(6)
C(1)–C(2)
C(2)–C(3)
C(3)–C(4)
C(3)–F(3)
C(4)–C(5)
C(5)–C(6)
C(6)–C(7)
C(7)–N(11)
C(7)–C(8)
C(8)–C(9)
C(9)–C(10)
C(9)–C(18)
1.969(6)
2.058(5)
2.074(5)
2.159(5)
2.9067(9)
1.413(10)
1.414(9)
1.361(11)
1.362(11)
1.379(8)
1.359(10)
1.380(9)
1.488(9)
1.366(8)
1.380(9)
1.406(9)
1.395(9)
1.485(10)
1.379(9)
1.366(7)
1.479(9)
1.384(10)
92.0(2)
5. Conclusions
The structure of cyclopalladated acetate bridged dimers
is known to exhibit an open-book type core in the solid
state. This core has been shown to persist in chloroform
solution, resulting in restrictions to the movement of ancil-
lary parts of the molecule. We believe the size of the barrier
can be accounted for solely on the basis of a steric
interaction.
C(10)–C(11)
N(11)–C(11)
C(11)–C(12)
F(15)–C(15)
C(1)–Pd(1)–O(02)#1
C(1)–Pd(1)–N(11)
O(02)#1–Pd(1)–N(11)
C(1)–Pd(1)–O(01)
O(02)#1–Pd(1)–O(01)
N(11)–Pd(1)–O(01)
C(6)–C(1)–C(2)
C(6)–C(1)–Pd(1)
C(2)–C(1)–Pd(1)
C(3)–C(2)–C(1)
C(2)–C(3)–C(4)
C(5)–C(6)–C(1)
C(5)–C(6)–C(7)
C(1)–C(6)–C(7)
N(11)–C(7)–C(8)
N(11)–C(7)–C(6)
C(8)–C(7)–C(6)
C(7)–N(11)–C(11)
C(7)–N(11)–Pd(1)
C(11)–N(11)–Pd(1)
81.1(2)
172.9(2)
176.9(2)
85.1(2)
6. Experimental
101.8(2)
117.6(6)
115.4(5)
127.0(6)
118.6(7)
123.6(7)
120.5(6)
125.7(7)
113.7(6)
121.7(6)
114.4(6)
123.9(6)
118.6(5)
112.5(4)
127.8(4)
6.1. General
All chemicals were used as supplied, unless noted other-
wise. All NMR spectra were obtained on a Bruker Avance
500 in CDCl3 and are referenced to external TMS, assign-
ments being made with the use of decoupling, nOe and the
DEPT and COSY pulse sequences. All elemental analyses
were performed by Warwick Analytical Service.
7. Pyridine ligands
The 2,4,6-trisubstituted pyridines were synthesized via
the Kro¨hnke route [19] or its solventless adaptation [20].
Crystals of 4-(4-octyloxy-phenyl)-2,6-diphenylpyridine
suitable for X-ray structure determination were grown
from a chloroform solution, and the structural data are
included as supporting information.
at 293 K can be calculated as 55.78 kJ molꢀ1 and at 268 to
be 55.99 kJ molꢀ1. Given likely uncertainty in the measure-
ment of the coalescence temperatures ( 2 K) and peak sep-
aration ( 5 Hz) our estimate of the uncertainty in the DGà
values is 0.9 kJ molꢀ1. This leads us to conclude that the
Gibbs energy of activation is the same at both tempera-
tures, and thus that (to the accuracy of our measurements)
the DHà is 56 kJ molꢀ1, and the entropy of activation (DSà)
is zero. An entropy of activation that is zero is consistent
with an intramolecular process that does not disrupt sol-
vent–solute interactions.
The enthalpy of activation we measure is very similar to
the barrier to rotation about the C–N bond in isonicitina-
mide [17], which has been measured as 59.2 kJ molꢀ1, and
is in between the 91.6 kJ molꢀ1 reported for a trisubstituted
biphenyl [18] and the 31.5 kJ molꢀ1 reported for 2,20-
bipyridine [18]. We propose that the barrier we have
7.1. Cyclopalladated dimer (1)
4-(4-Octyloxy-phenyl)-2,6-diphenyl)-pyridine (0.116 g,
2.67 · 10ꢀ4 mol) and palladium acetate (0.060 g, 2.67 ·
10ꢀ4 mol) were suspended in acetic acid (80 ml) and heated
(70 ꢁC, 48 h). The acetic acid was then removed under vac-
uum, and the remaining yellow solid dissolved in dichloro-
methane filtered to remove traces of palladium black. The
solvent was removed to give the product (0.223 g, 1.86 ·
10ꢀ4 mol, 70%). A single crystal suitable for X-ray diffrac-
tion was grown by diffusing diethyl ether into a chloroform
solution: the structural data are included as supporting
information.

P.G. Evans et al. / Journal of Organometallic Chemistry 691 (2006) 1251–1256
1255
Fig. 3. 1H NMR spectra of complex (2).
121.9, 119.4, 114.2, 112.3, 67.3, 30.8, 28.4, 28.3 28.22, 25.1,
22.1, 21.7, 13.1.
Mass spectrum (LSIMS) m/z: 1200 (M+), 1141 (M+ꢀ
AcO)
Elemental analysis: C, 65.8; H, 5.82; N, 2.3; C66H70N2-
O6Pd2 requires C, 66.1; H, 5.9; N, 2.3%.
7.2. Cyclopalladated dimer (2)
2,6-Bis-(4-fluoro-phenyl)-4-(4-propoxy-phenyl)-pyridine
(0.0268 g, 6.67 · 10ꢀ5 mol) and palladium acetate (0.0155 g,
6.90 · 10ꢀ5 mol) were suspended in acetic acid (25 ml) and
heated (65 ꢁC, 48 h). The acetic acid was then removed
under vacuum, and the remaining yellow solid dissolved
in dichloromethane filtered to remove traces of palladium
black. The solvent was removed to give the product
(0.0340 g, 3.00 · 10ꢀ5 mol, 90%). A single crystal suitable
for X-ray diffraction was grown by diffusing methanol into
a chloroform solution: the structural data are included as
supporting information.
Fig. 4. The labelling system used.
7.1.1. Spectroscopic data
dH (RT, CDCl3): 7.54 (4H, d, J(HH) = 8.5 Hz, phenoxy
ring), 7.25 (12H, v br m, pyridine + non-palladated ring),
7.02 (4H, d, J(HH) = 8.5 Hz, phenoxy ring), 6.95 (4H, m,
palladated ring), 6.81 (2H, d, J(HH) = 7.5 Hz, palladated
ring), 6.75 (2H, t, J(HH) = 7.5 Hz, palladated ring), 6.60
4
(2H, d, J(HH) = 1.2 Hz, pyridine ring), 4.02 (4H, t,
J(HH) = 7.0 Hz, OCH2), 1.75 (4H, tq, J(HH) = 7.0 Hz,
OCH2CH2), 1.34 (26H, s, AcO + chain), 0.81 (6H, t,
J(HH) = 7.0 Hz, Me).
dC (RT, CDCl3): 177.6, 164.0, 160.6, 156.7, 149.3, 147.6,
144.1, 138.2, 131.3, 127.8, 127.6, 127.2, 126.9, 126.3, 122.3,
7.2.1. Spectroscopic data
dH (RT, CDCl3): 7.56 (4H, d, J(HH) = 9.0 Hz, non-fluo-
rinated ring), 7.50 (4H, v br s, non-palladated ring), 7.18
(2H, d, J(HH) = 1.8 Hz, pyridine ring), 7.04 (4H, dd,
J(HH) = 8.8 Hz, non-palladated ring), 7.02 (4H, d,
4

1256
P.G. Evans et al. / Journal of Organometallic Chemistry 691 (2006) 1251–1256
J(HH) = 9.0 Hz, non-fluorinated ring), 6.80 (2H, dd,
Crystallographic Data Centre, CCDC numbers 289395–
289397, respectively.
4
J(HH) = 9.0 Hz J(HF) = 5.5 Hz, palladated ring), 6.74
4
(2H, d, J(HH) = 1.8 Hz, pyridine ring), 6.55 (2H, dd,
Copies of this information may be obtained free of
charge from: The Director, CCDC, 12 Union Road, Cam-
bridge, CB2 1EZ UK. Fax. (int code) +44 1223 336 033 or
e-mail: deposit@ccdc.cam.ac.uk or www: http://www.ccdc.
cam.ac.uk. Supplementary data associated with this article
can be found, in the online version, at doi:10.1016/
j.jorganchem.2005.11.066.
4
J(HF) = 9.0 Hz J(HH) = 2.5 Hz, palladated ring), 6.45
4
(2H, ddd, J(HH) = 9.0 Hz J(HF) = 9.0 Hz J(HH) = 2.5 Hz,
palladated ring), 4.03 (4H, t, J(HH) = 7.0 Hz, OCH2), 1.84
(4H, tq, J(HH) = 7.0 Hz, OCH2CH2), 1.38 (6H, s, AcO),
1.05 (6H, t, J(HH) = 7.0 Hz, Me).
At 233 K HW and HX appeared as separate peaks at 7.84
and 7.20 ppm, a maximum separation of 323 Hz; they coa-
lesced at 293 K.
References
At 233 K HY and HZ appeared as separate peaks at 7.14
and 7.07 ppm, a maximum separation of 32 Hz; they coa-
[1] A.D. Ryabov, Chem. Rev. 90 (1990) 403.
[2] B.A. Arndtsen, R.G. Bergman, T.A. Mobley, T.H. Peterson, Acc.
Chem. Res. 28 (1995) 154.
[3] G. Dyker, Angew. Chem., Int. Ed. 38 (1999) 1699.
[4] Y. Guari, S. Sabo-Etienne, B. Chaudret, Eur. J. Inorg. Chem. (1999)
1047.
[5] V. Ritleng, C. Sirlin, M. Pfeffer, Chem. Rev. 102 (2002) 1731.
[6] A.D. Ryabov, Synthesis (1985) 233.
[7] G.W.V. Cave, N.W. Alcock, J.P. Rourke, Organometallics 18 (1999)
1801.
[8] G.W. Cave, F.P. Fanizzi, R.J. Deeth, W. Errington, J.P. Rourke,
Organometallics 19 (2000) 1355.
[9] G.W.V. Cave, A.J. Hallett, W. Errington, J.P. Rourke, Angew.
Chem., Int. Ed. 37 (1998) 3270.
[10] A. Suzuki, Chem. Commun. (2005) 4759.
[11] D.L. Davies, S.M.A. Donald, S.A. Macgregor, J. Am. Chem. Soc.
127 (2005) 13754.
lesced at 268 K.
1
dC (RT, CDCl3): 179.2, 164.2, 163.5 (d, J(CF)
=
1
248 Hz), 161.0, 160.8, 160.1 (d, J(CF) = 253 Hz), 152.3
3
(d, J(CF) = 5.5 Hz), 149.3, 141.2, 135.2, 128.4, 128.2,
3
2
124.2 (d, J(CF) = 8.5 Hz), 120.0, 118.4 (d, J(CF)
=
2
19 Hz), 115.4, 115.2 (d, J(CF) = 20 Hz), 113.3, 110.6 (d,
2J(CF) = 23.1 Hz), 69.8, 23.3, 22.6, 10.6.
dF (RT, CDCl3): 110.2 (palladated ring), 112.9 (non-pal-
ladated ring).
Mass spectrum (LSIMS) m/z: 1073 (M+ꢀAcO), 1012
(M+ꢀAcOꢀOC3H7)
Elemental analysis: C, 59.0; H, 4.2; N, 2.3; C56H46F4-
N2O6Pd2 requires C, 59.43; H, 4.10; N, 2.48%.
[12] A. Crispini, M. Ghedini, J. Chem. Soc., Dalton Trans. (1997) 75.
[13] P.J. Bednarski, E. Ehrensperger, H. Scho¨nenberger, T. Burgemeister,
Inorg. Chem. 30 (1991) 3015.
Acknowledgements
[14] D.A. Fletcher, R.F. McMeeking, D. Parkin, J. Chem. Inf. Comput.
Sci. 36 (1996) 746.
We thank Johnson-Matthey for loan of precious
metals.
[15] G.J. Gainsford, R. Mason, J. Organomet. Chem. 80 (1974) 395.
[16] M.R. Churchill, Perspect. Struct. Chem. 3 (1970) 91.
[17] G.M. Leskowitz, N. Ghaderi, R.A. Olsen, K. Pederson, M.E.
Hatcher, L.J. Mueller, J. Phys. Chem. A 109 (2005) 1152.
[18] F. Ceccacci, G. Mancini, P. Mencarelli, C. Villani, Tetrahedron:
Asymm. 14 (2003) 3117.
[19] F. Kro¨hnke, Synthesis (1976) 1.
[20] G.W.V. Cave, C.L. Raston, J. Chem. Soc., Perkin Trans. 1 (2001)
3258.
Appendix A. Supporting information
Experimental data on the crystallographic structures of
the unmetallated ligand and compounds (1) and (2) is
included in the supporting information. Full data for their
structural analysis for has been deposited with the Cambridge