Palladium catalysed alkyne hydrogenation and oligomerisation: A parahydrogen based NMR investigation

Article abstract of DOI:10.1039/b804162h

The role phosphine ligands play in the palladium(ii)-bis-phosphine-hydride cation catalysed hydrogenation of diphenylacetylene is explored through a PHIP (parahydrogen induced polarization) NMR study. The precursors Pd(LL′)(OTf)₂ (1a-e) [LL′ = dcpe (PCy₂CH ₂CH₂PCy₂), dppe, dppm, dppp, cppe (PCy ₂CH₂CH₂PPh₂)] are used. Alkyl palladium intermediates of the type [Pd(LL′)(CHPhCH₂Ph)](OTf) (2 and 3) are detected and demonstrated to play an active role in hydrogenation catalysis. Magnetization transfer experiments reveal chemical exchange from the α-H of the alkyl ligand of 2a (LL′ = dcpe) and linkage isomer 2e′ (LL′ = cppe) into trans-stilbene on the NMR timescale. Activation parameters (ΔH^≠ and ΔS^≠) for this transformation have been determined. These experiments, coupled with GC/MS data, indicate that the catalytic activity is greater in methanol, where it follows the order: dcpe > cppe > dppp > dppe > dppm, than in CD ₂Cl₂. All five of the phosphine systems described are less active than those based on bcope [where bcope is (C₈H ₁₄)PCH₂-CH₂P(C₈H₁₄)] and ^tbucope [where ^tbucope is (C₈H ₁₄)PC₆H₄CH₂P(^tBu) ₂]. cis, cis-1,2,3,4-Tetraphenyl-buta-1,3-diene is detected as a minor reaction product with Pd(LL′)(PhCH-CHPh-CPh=CHPh)⁺ (4) also being shown to play a role in the alkyne dimerisation step.

Full text of DOI:10.1039/b804162h

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Palladium catalysed alkyne hydrogenation and oligomerisation: a
parahydrogen based NMR investigation†
Joaqu´ın Lo´pez-Serrano, Simon B. Duckett,* John P. Dunne, Cyril Godard and Adrian C. Whitwood
Received 12th March 2008, Accepted 9th May 2008
First published as an Advance Article on the web 25th June 2008
DOI: 10.1039/b804162h
The role phosphine ligands play in the palladium(II)-bis-phosphine-hydride cation catalysed
hydrogenation of diphenylacetylene is explored through a PHIP (parahydrogen induced polarization)
NMR study. The precursors Pd(LLꢀ)(OTf)2 (1a–e) [LLꢀ = dcpe (PCy2CH2CH2PCy2), dppe, dppm,
dppp, cppe (PCy2CH2CH2PPh2)] are used. Alkyl palladium intermediates of the type
[Pd(LLꢀ)(CHPhCH2Ph)](OTf) (2 and 3) are detected and demonstrated to play an active role in
hydrogenation catalysis. Magnetization transfer experiments reveal chemical exchange from the a-H of
the alkyl ligand of 2a (LLꢀ = dcpe) and linkage isomer 2eꢀ (LLꢀ = cppe) into trans-stilbene on the NMR
timescale. Activation parameters (DH= and DS=) for this transformation have been determined. These
experiments, coupled with GC/MS data, indicate that the catalytic activity is greater in methanol,
where it follows the order: dcpe > cppe > dppp > dppe > dppm, than in CD2Cl2. All five of the
phosphine systems described are less active than those based on bcope [where bcope is
(C8H14)PCH2–CH2P(C8H14)] and tbucope [where tbucope is (C8H14)PC6H4CH2P(tBu)2]. cis,
cis-1,2,3,4-Tetraphenyl-buta-1,3-diene is detected as a minor reaction product with
ꢀ
+
=
Pd(LL )(PhCH–CHPh–CPh CHPh) (4) also being shown to play a role in the alkyne
dimerisation step.
Introduction
Metal-catalysed carbon–carbon multiple bond hydrogenation
reactions are of fundamental interest in organometallic chemistry.
Although most of the research in this area has focused on alkene
hydrogenation, some information is available on the hydrogena-
tion of triple bonds. These investigations span heterogeneous1,2
and homogeneous palladium systems.3–5 In the latest studies,
a-diimine-based palladium(0) complexes were used.5 At the time
we decided to undergo a comprehensive study on the mechanism
of the palladium(II) bis-phosphine catalysed hydrogenation of
alkynes, little information had been published regarding the role
of phosphines on the mechanism of such a reaction, despite the
Scheme 1
3
key alkyl resting state was confirmed as a cationic g -palladium(II)
benzyl complex, in agreement with the related studies of Brown
et al.13 In these PHIP studies, cationic palladium monohydride
species of the type [Pd(LLꢀ)(Lꢀꢀ)(H)]+ (where Lꢀꢀ = solvent or
pyridine) were proven to provide access to the catalytic cycle with
additional intermediates, including analogous vinyl cations, also
being detected and shown to play an active role in catalysis. These
experimental results proved to be in good agreement with those
presented in a recent theoretical (DFT) study.14 These mechanistic
findings are related to those developed for the methoxycarbonyla-
tion of alkenes15,16 and the palladium catalysed hydroformylation
of alkenes17 because similar reaction intermediates have been
suggested.
fact that palladium complexes containing phosphine complexes
are ubiquitous in homogeneous catalysis.6
Our preliminary PHIP NMR7–9 studies with chelating10 (LLꢀ =
bcope) and monodentate (LLꢀ = PEt3) phosphines11 iden-
tified bis-phosphine alkyl palladium complexes of the type
[Pd(LLꢀ)(CHPhCH2Ph)]+ as key intermediates in alkyne hydro-
genation using these systems (Scheme 1). In addition, an active role
for cationic palladium hydrides in this reaction was established.
Further studies12 described the complete mechanism for alkyne
hydrogenation when the phosphine ligands are bcope and tbucope
t
[where bucope is (C8H14)PC6H4CH2P(tBu)2]. The nature of the
In many studies, the detection and characterization of reaction
intermediates often implies that their isolation or their manipu-
lation is achieved using conditions or ligand systems that differ
from those necessary for useful catalysis. It is even harder to prove
a direct role for such species in catalysis. PHIP NMR overcomes
the problem of detecting short-lived species in low concentration
Department of Chemistry, University of York, Heslington, York, UK YO10
5DD. E-mail: sbd3@york.ac.uk; Fax: +44 1904 43216
† Electronic supplementary information (ESI) available: Ligand
synthesis and crystal data and structure refinement for
[Pd(Ph2PCH2CH2PCy2)Cl2]·2CH2Cl2 and cis,cis-1,2,3,4-tetraphenyl-1,3-
butadiene. CCDC reference number 681984. For ESI and crystallographic
data in CIF or other electronic format, see DOI: 10.1039/b804162h
4270 | Dalton Trans., 2008, 4270–4281
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
by increasing the signal to noise ratio of hydrogen resonances
arising from parahydrogen (p-H2).8,18 It can be extended to allow
the detection of heteronuclei that are scalar coupled to the
enhanced proton signals, and by coherence selection experiments19
that suppress innocent background signals. It also enables the
determination of the fate of the detected intermediates, in situ,
via modified exchange spectroscopy (EXSY) experiments. Thus,
PHIP NMR has been successfully applied in the successful study
of several catalytic reactions.9,20–24
Herein we extend our previous investigations on the palla-
dium(II) bis-phosphine catalyzed hydrogenation of alkynes to
establish the generality of the reaction mechanism. We also
address the effect of the phosphine ligand on the activity of
the catalyst, as revealed by PHIP enhanced 1D EXSY studies
in conjunction with GC/MS data. We explicitly demonstrate
the importance of the phosphine moiety trans to the alkyne
ligand in the key intermediates Pd(LLꢀ)(CHPhCH2Ph)+. Finally,
a competitive multiple alkyne insertion process is also considered
that accounts for the formation of cis,cis-1,2,3,4-tetraphenyl-buta-
1,3-diene as a by-product in this reaction.
Results and discussion
Preparation of [Pd(LLꢀ)(OTf)2] complexes 1a–1e
The triflate complexes [Pd(LLꢀ)(OTf)2] (1a–1e, Table 1) used in
this study were synthesized from their chloride analogues by the
addition of AgOTf and isolated in good yields as pale yellow/white
solids. Details of these syntheses and NMR data for 1a–1e can be
found in the experimental section.
Crystal structure of [Pd(cppe)(Cl)2]
X-Ray quality crystals of the precursor to 1e, [Pd(cppe)Cl2], were
obtained by slow diffusion of diethyl ether into dichloromethane
solutions of the amorphous solid. The compound crystallized
with two solvent molecules (CH2Cl2) whereas the amorphous solid
contained an average of one solvent molecule per metal centre, as
deduced from elemental analysis. The palladium atom proved to
be located in the centre of a square planar arrangement of two
chlorine and two cis phosphorus atoms as shown in Fig. 1. The
P–Pd–P angle in [Pd(cppe)Cl2] is 86.30(2)◦ and close to the ideal
90◦ geometry. It lies in between those reported for the analogous
complexes [Pd(dppe)Cl2]25 and [Pd(dcpe)Cl2]26 where they are
85.82(7)◦ and 87.08(5)◦ respectively. The Pd–Cl(1) and Pd–Cl(2)
bond lengths are however, similar to the Pd–Cl distances in these
two, and other bis-phosphine, dichloride palladium complexes.
˚
The Pd–P(1) (PPh2) bond length is 2.2297(6) A and very similar to
˚
that reported for [Pd(dppe)Cl2] which is 2.230 A (averaged for the
two Pd–P distances). In contrast, the Pd–P(2) (PCy2) bond length
˚
of 2.2486(6) A is slightly longer than the average Pd–P distance
Table 1 Numbering for complexes [Pd(LLꢀ)(OTf)2]
LLꢀ
[Pd(LLꢀ)(OTf)2]
Fig. 1 Molecular structure of [Pd(dcpe)(Cl)2]·2CH2Cl2. ORTEP view
showing 50% probability ellipsoids, with crystal and structure refinement
◦
˚
dcpe (PCy2CH2CH2PCy2)
dppe (PPh2CH2CH2PPh2)
dppm (PPh2CH2PPh2)
dppp (PPh2CH2CH2CH2PPh2)
cppe (PCy2CH2CH2PPh2)
1a
1b
1c
1d
1e
data including selected bond lengths (A) and angles ( ) as indicated.
Hydrogen atoms and crystallization CH2Cl2 molecules have been omitted
for clarity.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4270–4281 | 4271
©

View Article Online
˚
in [Pd(dcpe)Cl2] where it is 2.2325 A. These data suggest
that the phosphorus donor in the unsymmetrical cppe ligand
system exhibits similar properties to those found in both the
unperturbed dppe and dcpe systems.
Hydrogenation of diphenylacetylene
Reactions of the bis-triflate complexes 1a–1e with a fifty-fold
excess of diphenylacetylene-d10 in methanol-d4 or CD2Cl2 under
3 atm of p-H2 were monitored by PHIP NMR spectroscopy over
the temperature range 265–333 K. In the corresponding 1H NMR
spectra, weakly enhanced signals indicating the formation of the
hydrogenation product cis-stilbene appear at d 6.66 as well as
a non-enhanced resonance for the thermodynamically preferred
reaction product trans-stilbene at d 7.18. The two alkenic protons
of the cis isomer are chemically and magnetically equivalent.
Consequently, the observation of this p-H2 enhanced emission
signal at d 6.66 indicates the involvement of a palladium based
intermediate where two hydrogen atoms of the p-H2 molecule
have become magnetically distinct.27 While the proton signal for
trans-stilbene increases in intensity with reaction time, this signal
does not show such p-H2 enhancement. Utilisation of 13C-enriched
diphenylacetylene-d10 changes this situation, however, since now
both the new d 6.66 and 7.18 signals show limited PHIP
enhancement via their 13C satellites. In the species, cis- and
31
Fig. 2 1H{ P} NMR spectrum obtained during the reaction of 1a
with diphenylacetylene-d10 and p-H2 in methanol-d4 showing three PHIP
enhanced signals for the alkyl ligand of 2a (*) at 295 K. The weak singlet
at d 7.18 corresponds to trans-stilbene. The inset shows the weak emission
signal seen for cis-stilbene (64×).
trans-PhCH 13CHPh-d10, that give rise to these signals, the two
=
alkenic protons are magnetically inequivalent because of the 13C
label. This confirms that hydrogenation of the triple bond of
diphenylacetylene to yield cis- and trans-stilbene proceeds in such
a manner as to enable two protons from a single p-H2 molecule to
be placed in the same reaction product.
evident in accordance with little change (or deactivation) of the
initial palladium species during the catalysis.
The PHIP enhanced d 4.38 signal actually possesses a JHH
coupling of 11.5 Hz and two 31P splittings of 8.0 and 4.3 Hz, which
can be removed by appropriate 31P decoupling. This confirms
that species 2a is indeed a palladium bis-phosphine complex. This
deduction was supported by the fact that the corresponding 1H-31P
heteronuclear multiple quantum coherence (HMQC) spectrum,
which utilises magnetisation transfer from the three enhanced
proton resonances, allows the detection of two 31P doublets at
d 72.3 and d 66.4 where JPP is 35 Hz. When the 31P-1H HMQC
experiment was optimisedfor the larger 8.0Hz1H-31P coupling, the
31P resonance that appeared in the corresponding experiment was
at d 72.3. This information suggests that the phosphine yielding
the 31P signal at d 72.3 is trans to the alkyl group.
These results were supported by independent GC/MS studies,
which confirmed that trans-stilbene was the most abundant prod-
uct of the hydrogenation reactions. These studies, however, also
revealed the formation of cis-stilbene and yet smaller amounts of
the fully hydrogenated product dibenzyl and 1,2,3,4-tetraphenyl-
buta-1,3-diene. These results are discussed more fully later in the
manuscript.
Reactions of Pd(dcpe)(OTf)2 (1a) with p-H2 and diphenylacetylene
As indicated above, when the reaction of 1a, diphenylacetylene-
d
10 and p-H2 is monitored by PHIP NMR spectroscopy, a number
A series of 1H-13C HMQC experiments, using 13C C enriched
≡
of organic products are observed. In addition to these species,
a number of resonances associated with palladium complexes of
the type Pd(LLꢀ)(CHPhCH2Ph)+ are detected. In the case of 1a,
these correspond to strongly enhanced proton signals for 2a that
are visible at d 4.38, d 3.40 and d 3.22 in methanol-d4. These
signals are illustrated in Fig. 2. The three PHIP enhanced NMR
signals contain characteristic anti-phase components due to their
origin from p-H2 derived protons and are coupled according
to correlation spectroscopy (COSY) and therefore in the same
molecule. When the sample was investigated using a 90◦ pulse,
which quenches the PHIP effect, the signals due to 2a were no
longer visible and hence we are able to conclude that 2a is not a
stable species. It is therefore only detectable because of the PHIP
diphenylacetylene-d10 were then completed. These experiments
confirmed that the 1H resonance of 2a, which appeared at d
4.38 was connected to a 13C signal at 64.2 (dd, JCP = 46.0 and
10.0 Hz). In contrast, the two proton signals at d 3.40 and d
3.22 connected to a single 13C signal at d 35.5 (dd, JCP = 22.5
and 13.0 Hz) and therefore arise from the CH2 group. The
large 31P-13C coupling of the former 13C signal confirms that
this centre is attached to palladium and trans to phosphorus.
This information is summarized in Tables 2 and 3. The species
giving rise to these signals, 2a, is therefore minimally described as
Pd(dcpe)(CHPhCH2Ph)+.
These observations are in full agreement with the results
previously reported for bcope and tbucope based systems12
where the corresponding alkyl species were described as cationic
1
effect. In addition, it should be noted that when 31P{ H} NMR
spectra that were obtained before the reaction are compared with
those obtained at the end of the run, no spectral changes are
3
g -benzyl complexes (Scheme 2), with the p system of one of
4272 | Dalton Trans., 2008, 4270–4281
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 2 1H NMR data for the alkylpalladium intermediates 2–4 in
t
methanol-d4. Data for the bcope and bucope derived species (labelled
2f and 2g)12 are included for comparison
Pd(LLꢀ)(“alkyl”)+a
d CH
4.38
4.24
3.70
d CH2
Pd(dcpe)(OTf)2 (1a)b
Pd(dppe)(OTf)2 (1b)c
2a
2b
3b
c4b
3.40
3.22
2.99
2.62
3.45
3.38
—
6.64
5.10
4.18
4.96
Pd(dppm)(OTf)2 (1c)c
Pd(dppp)(OTf)2 (1d)b
2c
2d
4d
3.58
3.14
2.98
2.22
—
Scheme 2 Potential structural isomers for Pd(LLꢀ)(CHPhCH2Ph)+.
Species 2, the major isomer detected in this study, are assigned the structure
I on the basis of literature data.12
3.71
6.71
4.87
3.46
4.49
Pd(cppe)(OTf)2 (1e)b
2e (alkyl trans to PPh2)
3.58
3.44
2.95
2.45
3.95
2.94
2eꢀ (alkyl trans to PCy2)
3.99
4.97
It should be noted that while we only refer to NMR data
obtained in methanol-d4, species 2 has also been observed in
CD2Cl2, the only difference being the appreciably smaller intensity
of the PHIP enhanced resonances seen in CD2Cl2. This is in
agreement with the lower catalytic activity shown by these systems
in dichloromethane as determined by GC/MS. This point is
discussed in more detail later in the text.
3eꢀ
a4e
7.25
5.42
4.45
4.98
—
Pd(bcope)(OTf)2 (1f)b
2f
3.06
2.98
3.03
2.91
Pd(tbucope)(OTf)2 (1g)d
2g
4.93
1
We also note that in these H NMR spectra, two additional
polarized signals appear at d 4.94 and d 2.93. While these
signals are too weak to characterize the product that produces
them fully, they are likely to arise from a second isomer of
Pd(dcpe)(CHPhCH2Ph)+. Species with g -benzyl, g -homobenzyl
or b-agostic13 interactions have been seen previously and would
meet this requirement (Scheme 2).28 Unfortunately we are not able
to assign the alkyl ligand a more specific binding mode because of
the weakness of these signals.
a Alkyl = PhCHCH2Ph except for the species 4b, 4d and 4e where “alkyl” =
PhCH–PhCH–CPh CHPh. b T = 295 K. c T = 323 K. d T = 260 K.
=
2
3
the phenyl substituents of the alkyl ligand occupying the fourth
coordination site of the metal centre.
t
Table 3 31P and 13C NMR chemical shifts for species 1–4 in methanol-d4. Data for the bcope and bucope derived species (labelled 2f and 2g)12 are
included for comparison
d 31
P
d 31P (JPP
)
d 13
C
1a
1b
110.65
74.24
2a
2b
72.3 (d, 35 Hz)
63.4 (d, 35 Hz)
50.5 (d, 42 Hz)
47.6 (d, 42 Hz)
64.2 (dd, 46, 10 Hz)
35.5 (dd, 26, 13 Hz)
—
1c
1d
−50 (broad)
—
2d
—
—
—
15.17
17.3 (d, 75 Hz)
3.8 (d, 75 Hz)
4d
2e
2eꢀ
4e
15.2 (d, 75 Hz)
0.7 (d, 75 Hz)
—
1e
107.10 (PCy2) 75.06 (PPh2)
69.5 (PCy2, d, 40 Hz)
49.5 (PPh2, trans to alkyl, d, 42 Hz)
61.0 (PCy2, trans to alkyl, d, 42 Hz)
53.9 (PPh2, d, 42 Hz)
71.8 (PCy2, d, 42 Hz)
52.0 (PPh2, trans to alkyl, d 42 Hz)
65.5 (dd, 49 Hz)
35.6
69.5 (dd, 40, 20 Hz)
34.1 (d, 40 Hz)
144.4
126.3
70.4 (d, 32 Hz)
54.5
62.3 (dd, 54, 16 Hz)
35.6 (dd, 16, 5 Hz)
64.0 (d, 55 Hz)
34.8
1f
73.47
2f
43.4 (d, 40 Hz)
32.2 (d, 40 Hz)
1g
110.00 (PtBu2)
31.10 (PC8H14)
2g
70.4 (PtBu2, trans to alkyl)
19.8 (PC8H14)
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4270–4281 | 4273
©

View Article Online
Reactions of Pd(dppx)(OTf)2 (1b–d; dppx = dppe, dppm and dppp
respectively) with p-H2 and diphenylacetylene
2d is assigned to Pd(dppp)(PhCHCH2Ph)+ by analogy with the
previous data.
Three additional resonances where also seen in these 1H NMR
spectra at d 6.71, 4.87 and 3.46 which also proved to be coupled
by COSY. The d 3.46 resonance correlated to two 31P signals that
appear at d 15.2 and 0.7 and also possess mutual JPP couplings of
75 Hz. The proton resonances for this species are very similar to
1
The associated H NMR spectra for reactions of 1b, dipheny-
lacetylene and p-H2 in methanol-d4 also contain signals for
polarized cis-stilbene at 295 K. Signals for trans-stilbene, however,
now only become readily visible at 323 K. At this temperature,
three much weaker PHIP enhanced, coupled, proton signals due
to 2b also appear at d 4.24, 2.99 and 2.62. The high field d 4.24
signal posses a large JHH coupling of 11.0 Hz and 31P decoupling
now removes two JPH splittings of 12.0 Hz and 5.0 Hz. The
corresponding 1H-31P HMQC spectra now reveal two 31P doublets
at d 50.5 and d 47.6, where JPP = 42 Hz. This suggests that the
structure of 2b is very similar to that of 2a. Unfortunately, 13C data
could not be obtained because of the weak PHIP enhancements
seen for 2b. Nonetheless, we conclude that the species giving rise
to these signals is Pd(dppe)(CHPhCH2Ph)+ (2b).
those of 4e, which are described later, and therefore assigned to
+
=
Pd(dppp)(PhCH–CHPh–CPh CHPh) (4d).
The comparatively larger JPP coupling measured for species 2d
and 4d reflects a change in the P–Pd–P angle, which is facilitated
by the more flexible propyl bridge of the phosphine.
Reactions of Pd(cppe)(OTf)2 (1e) with p-H2 and diphenylacetylene
(a) Exploring the controlling influence of the phosphine trans to
the alkyl. The situation proved to be far more complex when
the low symmetry complex 1e was examined. The PHIP enhanced
resonances of the inorganic complexes seen in the hydrogenation
reaction were readily distinguished from those of their organic
counterparts by 2D COSY spectroscopy, which confirmed that
there were four sets of connected resonances due to four inorganic
In these 1H NMR spectra, three weaker polarized signals
were also detected for a second species, 3b, which appear at d
3.70, 3.45 and 3.38. These resonances correlated with two 31P
resonances, at d 48.1 (d, JPP = 40 Hz) and 40.7 (d, JPP = 40 Hz)
1
in the corresponding H-31P HMQC spectrum. We believe that
1
1
the chemical shifts of the H and 31P resonances are sufficiently
species. The first of these, species 2e shows H PHIP enhanced
similar to those of 2b to indicate the formation of a second isomer
of Pd(dppe)(CHPhCH2Ph)+ (3b). This is supported by the fact
that while the d 3.70 signal for the PdCHPh proton of 3b still
NMR signals at d 4.49 (multiplet, JHH = 10 Hz), 3.58 (m) and d 3.44
(multiplet, JHH = 10 Hz). The second species, 2eꢀ, has resonances
at d 3.99 (m, JHH = 10 Hz), d 2.95 (m, JHH = 10 Hz) with the
possesses a large JHH coupling of 14 Hz, it shows a very limited 31
P
third resonance, obscured in the H spectra by the phosphine
1
splitting of ca. 1 Hz.
ligand signals, but appearing in the COSY spectrum at d 2.45 (m)
(Fig. 3).
1
A third set of three even weaker PHIP enhanced H NMR
signals were visible at d 6.64, 5.10 and 4.18 due to a third species,
4b. However, their signal intensity was so low that neither 31P nor
13C data could be obtained for this product. These resonances
will, however, be shown to be due to Pd(dppe)(PhCH–CHPh–
+
=
CPH CHPh) by analogy with the NMR data for reaction
products seen with 1d that are described later.
When analogous reactions were carried out with Pd(dppm)-
(OTf)2 (1c), diphenylacetylene-d10 and p-H2 in methanol-d4 at
1
323 K, the corresponding H NMR spectra revealed extremely
weak PHIP enhanced resonances at d 4.96, d 3.58 and d 3.14. The
low intensity of these signals prevented the collection of COSY,
13C or 31P data for this species. However, the similarities in these 1H
NMR signals to those of 2a, 2b or 3b suggests that the analogous
alkylpalladium cation 2c has been detected.
Interestingly, when Pd(dppp)(OTf)2 (1d) was used as the
precursor, comparatively intense PHIP signals appeared in the
1
corresponding H NMR spectra at d 3.71 and 2.98. These two
signals proved to be coupled to each other by COSY and a third
proton resonance was located at d 2.22; this overlapped in the
corresponding 1D 1H NMR experiments with signals for the
alkylidene protons of the dppp ligand of 1d.
The signal at d 3.71 contained a large JHH coupling of 11.4 Hz
and two JHP couplings of 13.4 and 3.2 Hz, which are very similar
to those of 2a and 2b. For samples of similar concentrations, the
intensity of these signals were clearly much bigger than those of
2b and 2c but they were not as intense as those observed for
2a. This increase in signal strength facilitated the location of the
two expected phosphorus resonances at d 17.3 and 3.8, but now
JPP proved to be 75 Hz. The intensity of the proton resonances
was not, however, high enough to obtain 13C data. Nonetheless
Fig. 3 p-H2 modified COSY spectrum of a sample containing 1e, dipheny-
lacetylene-d10 and p-H2 in methanol-d4, showing couplings between the
enhanced proton resonances of 4e (red), 2e (yellow) and 2eꢀ (blue).
The relative intensity of the signals for 2e increases with time
and we therefore conclude that its concentration builds up during
the course of the reaction; it also provides the largest signals at
higher reaction temperatures. In contrast, the signals for 2eꢀ are
always weak, although their proportion is higher at lower reaction
temperatures (Fig. 4).
4274 | Dalton Trans., 2008, 4270–4281
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
a second resonance. What remains is to identify which phosphine
arm is trans or cis to the alkyl.
For species 2b, where the alkyl is trans to the PPh2CH2 group
1
of the dppe ligand, H NMR resonances for the alkyl appear at
d 4.24, 2.99 and 2.62 while for species 2c where the alkyl is trans
to the PCy2CH2 moiety they appear at d 4.38, 3.22 and 3.40. The
signals of isomer 2e, appear at d 4.49, 3.58 and 3.44, and are
therefore closer to those of 2c than 2b. In contrast, those for 2eꢀ
appear at d 3.99, 2.95 and 2.45 and are now closer to those of
2b. This suggests that the cis ligand controls the chemical shifts
of the alkyl protons for the main isomer. Unfortunately, for 3e
the corresponding proton resonances lie at d 4.97, 3.95 and 2.94
and therefore do not relate to those of either 2b or 2c. We cannot
therefore draw any conclusions about the relative orientation of
the alkyl ligand in 3e.
Fig. 5 illustrates how the OSPY method can be used successfully
to separate PHIP derived proton signals from those of the solvent
and the starting materials while monitoring these reactions.19
When Fig. 5 is compared with Fig. 4, the benefits of the OPSY
selection method are clearly illustrated.
Fig. 4 1H NMR spectra in methanol-d4 showing PHIP resonances for 2e
(*) and 2eꢀ (ꢀ) at 295 K 1 min after p-H2 addition (lower trace) and 15 min
later at 300 K (upper trace). The structures of isomers 2e and 2eꢀ, which
are differentiated by virtue of the relative orientation of the cppe ligand,
are indicated above.
(b) Detection of an intermediate that is implicated in alkyne
dimerisation. The fourth complex, 4e, seen in these experiments
exhibits enhanced 1H resonances at d 7.25 (d, JHH = 2.5 Hz), d 5.42
(d, JHH = 10 Hz) and d 4.45 (m, JHH = 10 Hz), which are coupled
by COSY (see Fig. 3).
The characterisation of the four species 2e–4e was secured
by multinuclear NMR methods. Specifically, 1H-31P HMQC
spectroscopy shows that 2e contains two inequivalent phosphine
centres by virtue of the location of two 31P resonances at d 49.5
(d, JPP = 40 Hz) and d 69.5 (d, JPP = 40 Hz). A similar situation
holds for 2eꢀ where the corresponding resonances appear at d
53.9 (d, JPP = 42 Hz) and d 61.0 (d, JPP = 42 Hz). For this
These spectral features are similar to those of 2e in as far as the
1
fact that two 31P signals are located in the corresponding H-31P
HMQC spectrum at d 71.8 (d, JPP = 42 Hz, due to the PCy2CH2
moiety) and 52.0 (d, JPP = 42 Hz, due to the PPh2CH2 moiety).
While complex 4e also yielded three enhanced proton signals, the
fact that one resonates at d 7.25 suggests that there is a substantial
difference in structure between those of species 2 and 3; such a
chemical shift would be closer to that expected for a vinylic signal.
This difference is confirmed in the corresponding short range 1H-
31C HMQC spectrum where the three 1H peaks at d 7.26, 5.42 and
4.44 connect to 13C signals at d 126.3, 70.37 (d, JCP = 32 Hz) and
54.5 respectively. In the corresponding long range 1H-13C HMQC
ligand system, 1H-13C HMQC experiments, using 13C C enriched
≡
1
diphenylacetylene-d10 as the substrate, enabled the H resonance
of 2e, which appeared at d 4.49, to be connected to a 13C signal at d
65.5 (d, JCP = 49 Hz) while the proton at d 3.44 for the CH2 group
connected to a signal at d 35.6. The large 31P-13C coupling of the
former signal implies that it is attached to the metal and trans to
phosphine. The analogous 1H resonance of 2eꢀ at d 3.99 correlated
to a 13C signal at d 69.6 (dd, JCP = 40, 20 Hz) while the signals
at d 2.95 and d 2.45 now linked with a carbon resonance at d
34.1. Species 2e and 2eꢀ have therefore been shown to contain
a Pd(PCy2CH2CH2PPh2)(CHPhCH2Ph) core and are therefore
most likely differentiated by virtue of the relative orientation of
the phosphine and alkyl ligands.
1
experiment, the H signal at d 7.26 connects to the d 54.5 signal
and a fourth 13C resonance at d 144.4. There are therefore four
carbon resonances in the alkyl chain of 4e rather than the two
found in 2 and 3. The resonances at d 70.4 (d, JCP = 32 Hz) and
54.5 are, however, still aliphatic in nature while those at d 144.4
and 126.3 are clearly vinylic. Appropriate 31P decoupling of the
1
corresponding H spectra revealed that the organic arm of this
species is trans to the PCy2CH2 moiety.
When the values of the corresponding 1H-31P couplings are
attributed, the structure for 2e has the alkyl trans to the PPh2CH2
moiety. This is based on the fact that decoupling the 31P d 49.5
signal due to the PPh2CH2 group removes a large 31P splitting in
The exact nature of 4e was confirmed when X-ray quality
crystals precipitated from the NMR tube during the reaction.
The resultant X-ray diffraction studies showed that the unit cell
was virtually identical to that reported previously for cis,cis-
1,2,3,4-tetraphenyl-1,3-butadiene29 (see ESI for X-ray diffraction
structure data at 120 K†). The remaining organic signal at d
6.34 was then assigned to the vinylic proton of cis,cis-1,2,3,4-
tetraphenyl-1,3-butadiene in agreement with the literature.30
1
the d 4.49 H signal. An analogous conclusion was made for the
relationship between groups yielding the d 61.0 31P signal of 2eꢀ,
1
the PCy2CH2 moiety, and the d 3.99 H signal. These structures
are illustrated in Fig. 4.
The third species, 3e, provided a set of weakly enhanced signals
at d 4.97 (m, JHH = 10 Hz), d 3.95 (m, JHH = 10 Hz) and d 2.94
(m, JHH = 10 Hz); these signals disappear rapidly as the substrate
is consumed. Species 3e also exhibits one 31P resonance at d 71.0,
but due to the short life of the complex, we were unable to identify
It can therefore be concluded that 4e is Pd(Ph2PCH2CH2PCy2)-
+
=
(C(Ph)H–CHPh–CPh CHPh) . In our previous study, we de-
ꢀ
+
ꢀ
=
tected vinyl complexes of the type Pd(LL )(CPh CHPh) (LL =
bcope and tbucope), which result from the insertion of an
alkyne molecule into a Pd–H monohydride bond. Although
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4270–4281 | 4275
©

View Article Online
Fig. 5 (a) 1H-OPSY spectrum (32 scan) obtained from a sample of 1e with an excess of diphenylacetylene and para-hydrogen (3 atm) in d4-methanol
after 5 min at 295K. (b) 1H spectrum of the same sample showing the antiphase signals for the alkyl proton resonances of the species 2e, 2eꢀ and 3e.
such a species is not observed here, its involvement via binding
diphenylacetylene and subsequent insertion into the Pd–C bond
of the vinyl is necessary to form 4e after p-H2 addition (Scheme 3).
An alternative mechanism for the formation of 4e involving
alkene coupling assisted by triflic acid elimination is possible.31
However, alkyne insertions have been documented,32 and more
importantly, account for the PHIP enhanced resonances seen in
these experiments. It is worth noting that only one isomer of 4e is
detected and based on the structure of 2e, we propose 4e has the
form shown in Scheme 3.
Reactions of the palladium-bis-phosphine triflates with
diphenylacetylene and p-H2 in the presence of pyridine
It has been demonstrated in an earlier study, how the ad-
dition of small amounts of pyridine enables the trapping of
the corresponding palladium monohydride and vinyl cations in
t
the bcope and bucope palladium systems [i.e. Pd(LLꢀ)(py)(H)+
ꢀ
+
=
and Pd(LL )(py)(PhC CHPh) ] and hence their detection. These
adducts where shown to be active in the semihydrogenation
of diphenylacetylene.12 However, the addition of pyridine also
caused a weakening of the observed PHIP polarised signal
intensities. When the systems described in this paper were studied
in the presence of limited amounts of pyridine-d5, p-H2 and
diphenylacetylene, only extremely weak resonances for species
2a, 2e and 2eꢀ were detected at 335 K, and no resonances
Earlier in this manuscript, for the reactions involving 1b and
ꢀ
+
=
1d, similar products, Pd(LL )(PhCH–CHPh–CPH CHPh) 4b
(LLꢀ= dppe) and 4d (LLꢀ = dppp) were tentatively described. We
1
believe that the similarity in H NMR signals for 4b, 4d and 4e
confirms their analogous identities.
4276 | Dalton Trans., 2008, 4270–4281
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
while almost no magnetization transfer into the same reaction
product was observed from the corresponding resonance of 2e
at d 4.49. It is therefore clear that isomer 2eꢀ is more reactive
than isomer 2e (in 2b–d, the dppe, dppm and dppp analogues,
no exchange was again evident). This implies that the electron
donating trans PCy2CH2 moiety of 2eꢀ has a substantial impact on
the underlying reactivity of the system. The rate constant for the
formation of trans-stilbene from 2eꢀ was determined as 0.29 s−1 at
300 K. Eyring analysis of these data at different temperatures was
then used to determine the corresponding activation parameters
and DG= . These observations are similar to the ones reported in
our earlier studies and indicate that the detected PdCHPhCH2Ph
group primarily undergoes transformation into trans-stilbene. The
results are summarized in Table 4 where the data already reported
t
for systems containing the bcope and bucope ligands have been
included for comparison purposes.
In a similar way, when the d 5.42 signal, assigned to the a-H
Scheme
3
Proposed mechanism for the formation of 1,2,3,4-te-
=
of the CHPh–CHPh–CPh CHPh ligand of 4e, was selectively
traphenyl-1,3-butadiene, with the participation of species 4. See Scheme 4
for an outline of the catalytic hydrogenation mechanism.
irradiated, magnetization transfer into the vinyl resonance of
1,2,3,4-tetraphenyl-buta-1,3-diene was observed at temperatures
above 320 K. A similar result was obtained when the vinyl
for the expected palladium vinyl or hydride complexes were
seen.
Table 4 Activation parameters for the formation of trans-stilbene from
the alkyl-palladium intermediates 2 in methanol-d4 and for the formation
of 1,2,3,4-tetraphenyl-buta-1,3-diene from 4e
Kinetic studies (NMR) and organic speciation (GC/MS)
=
The observation of PHIP enhanced resonances for the alkyl
protons of species 2–4 is indicative that these species take part
in a process where p-H2 exchange is facilitated. Direct evidence
for these species corresponding to intermediates involved in the
catalytic formation of diphenylacetylene has been obtained via
EXSY based NMR experiments. These results are now described.
When the dcpe system was examined by EXSY spectroscopy,
magnetization transfer from the CH alkyl resonance of 2a at d 4.38
into the reaction product trans-stilbene was found (Fig. 6); and the
observed rate constant at 300 K proved to be 0.17 s−1. In the case
of the cppe system, when the peak at d 3.99 corresponding to
2eꢀ was selected, magnetization transfer to trans-stilbene was seen
k
300 K/s−1 aDH=
aDS=
aDG 300 K
78
—
—
2a (dcpe)
0.17
—
81
—
—
88
42
94
7
11 21
9
2b–db (Ph2P(CH2)nPPh2)
2e (cppe)
—
—
∼0
2eꢀ (cppe)
bcope system
tbucope system
4ed
0.29
0.53
1.04c
0.49c
3
9
6
39
9
76.6 0.9
−107 31
75
73
8
1
69 22
85 13
33 39
75 31
a Free energies, enthalpies, and entropies in kJ mol−1 and J mol−1 K−1
respectively. b We were unable to observe any exchange from the resonances
of any of the intermediates generated from 1b–d. c Extrapolated from the
Eyring plot in CD2Cl2. d From relatively weak signals, see text.
Fig. 6 1D EXSY experiment (T = 310 K, mixing time = 900 ms) showing exchange of the CH alkylic proton of 2a into trans-stilbene.
This journal is The Royal Society of Chemistry 2008 Dalton Trans., 2008, 4270–4281 | 4277
©

View Article Online
resonance of 4e at d 7.25 was examined. The remaining enhanced
signal of the organic ligand of 4e overlaps with the a-H resonance
of the alkyl ligand of 2e, as previously indicated. Selective
irradiation of this spectral region results in some magnetization
transfer into 1,2,3,4-tetraphenyl-buta-1,3-diene and the d 7.18
resonance of trans-stilbene due to the contribution from 2e. Rate
constants were obtained for diene elimination from 4e between
320 and 333 K and Eyring analysis facilitated the determination
of appropriate activation parameters. The fact that magnetization
transfer was observed only at or above 320 K suggests that the
elimination of the diene from 4e has a higher activation energy
barrier than b-elimination from 2eꢀ to yield stilbene.
The data summarized in Table 4 were obtained in parallel
with GC/MS analysis of the organic product distribution after
one hour of reaction. These data were obtained from a second
NMR tube, which was placed at 322 K and shaken every
15 minutes in order to facilitate H2 dissolution (see Experimental
section). The corresponding GC/MS data are shown in Table 5.
Clearly, the reaction is solvent dependant, with methanol
leading to a higher activity. This matches the fact that the PHIP
signals for the intermediates 2–4 are much stronger in methanol-
d4 than in CD2Cl2. The reactivity of the systems bearing the
PPh2 based phosphines is clearly less than those of the dcpe
system, and the previously described bcope or tbucope phosphine
containing systems. The activity of the PPh2 based systems
follow the order dppp > dppe > dppm and follow the trend in
PHIP enhancements (see ESI†). Interestingly, in the case of 1e,
bearing the low symmetry phosphine cppe (PPh2CH2CH2PCy2),
the overall catalytic activity is lower than that of the system with
dcpe, despite the fact that the rate constant for elimination of
trans-stilbene from 2eꢀ is higher than that of 2a. This information
is consistent with the fact that based on symmetry, 1a always
generates an active complex while for 1e, only form 2eꢀ is active on
the NMR timescale at 300 K. Consequently, it can be concluded
that the concentration of the active form of the catalyst in the 1e
based system is lower than that formed with 1a.
Scheme 4 Proposed mechanism for the catalytic hydrogenation of
diphenylacetylene.12 See Scheme 3 for the proposed mechanism of for-
mation of 1,2,3,4-tetraphenyl-1,3-butadiene via intermediate 4.
The EXSY experiments established the role of the alkyl-
palladium cations in trans-stilbene formation by the observation
of chemical exchange from the a-H of the alkyl ligand into the
alkenyl proton signals of trans-stilbene. The kinetic data obtained
from the EXSY experiments suggest that the elimination of trans-
stilbene proceeds with a higher activation barrier for the dcpe
=
and cppe systems studied here (DG300 K = 78
9 and 76.6
0.9 kJ mol−1 respectively) when compared with those previously
t
described for the more electron rich bcope and bucope ligand
systems (DG300 K = 75 8 and 73.6 1 kJ mol−1 respectively).
=
The product distribution is also dependent on the phosphine.
In the case of the cppe system, nearly 30% diene formation is
indicated, which is in agreement with the NMR observation of 4e.
Nonetheless, trans-stilbene formation is always preferred relative
to diene formation.
It should be noted that based on these data and those reported
previously,12 substantial solvent effects are seen in this reaction in
agreement with the large variations in activation entropy that are
observed.
When the dcpe and cppe systems are compared, it is 2eꢀ, which
bears the cppe phosphine ligand, that provides the faster reacting
intermediate. However, 2e, which has the PPh2CH2- moiety of the
cppe ligand trans to the alkyl, shows very little activity on the
NMR timescale.
Conclusions
In this PHIP based NMR study of palladium-catalysed hydro-
genation of diphenylacetylene, it is confirmed that the reaction
pathway illustrated in Scheme 4 is followed.12,14
Table 5 Conversion percentage and organic product distribution after one hour hydrogenation at 322 K as shown by GC/MS
% Cconversion
CH2Cl2
% cis-Stilbene
% trans-Stilbene
% Dibenzyl
CH2Cl2
% Diene
CH2Cl2
Ligand
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
MeOH
MeOH
dcpe (1a)
dppe (1b)
dppm (1c)
dppp (1d)
cppe (1e)
1.9
1.6
< 1.0
10.5
48.9
2.7
0.0
23.5
27.5
27.3
28.4
100
39.0
20.6
8.9
9.2
38.1
—
18.5
10.7
4.2
67.1
65.9
—
61.0
72.8
89.8
98.8
88.7
53.2
—
78.2
59.4
68.8
98.6
5.6
6.6
—
1.0
5.2
—
—
—
—
—
1.0
1.1
<0.1
1.1
5.2
—
3.3
29.8
1.4
—
—
—
3.4
5.7
0.2
0.2
0.1
25.0
0.6
bcope (1f)
49.8
99.4
tbucope (1g)
99.9a (23.3)
100a (69.4)
1.0
0.9
a Conversion after 10 min in parenthesis.
4278 | Dalton Trans., 2008, 4270–4281
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
◦
When the reaction was examined by GC/MS, the conversion in
methanol after one hour followed the order: bucope > bcope >
glove box equipped with a freezer (−32 C), vacuum pump and
N2 purge facilities. Solvents were obtained as Analytical Grade
from Fisher. Diethyl ether, n-pentane and n-hexane were dried
by refluxing under nitrogen over sodium wire, dichloromethane
by refluxing over calcium hydride and methanol over calcium
hydride. The deuterated solvents methanol-d4 and CD2Cl2 were
purchased from Aldrich and stored under N2 but used without
further purification. Pd(PhCN)2Cl2 was prepared as described
in the literature33 (PdCl2 was purchased from Johnson Matthey
and PhCN from Lancaster). AgOTf was obtained from Aldrich
t
dcpe > cppe > dppp > dppe > dppm. The systems containing the
electron poor PPh2-based phosphines are clearly less active than
systems with phosphines bearing more electron rich and bulkier
substituents on the phosphorus atoms. A greater flexibility in the
phosphine backbone also seems to contribute to increased catalyst
activity, which in the case of the PPh2-based phosphines follows
the order dppp > dppe > dppm.
The overall activity of the dcpe system as shown by GC/MS
is higher than that of the cppe system despite the lower activa-
tion parameters for alkene elimination from 2eꢀ, which bears a
PCy2CH2- moiety trans to the alkyl ligand. This situation arises
because 2e, the isomer with the PPh2CH2- moiety trans to the
alkyl ligand, is essentially inactive, as deduced by the associated
NMR EXSY data. This observation confirms the importance of
an electron-rich phosphine moiety lying trans to the alkyl ligand in
order to promote catalysis. The relative intensity of the enhanced
resonances of 2e and 2eꢀ changes with temperature as shown in
Fig. 4. This change is consistent with an increasing role for 2e at
higher reaction temperatures.
and used as received. PhC 13CPh-d10 and PhC CPh-d10 were
prepared as described previously.12
NMR spectra were obtained on either Bruker DRX 400 or
Bruker DSX 700 spectrometers.
≡
≡
GC/MS data were collected on a Saturn 2000 gas chromato-
grapher–mass spectrometer combination. A Factor-FormTM VF-
6mg capillary column (30 m × 0.25 mm ID and 0.25 lm film
thickness) was used for the GC separation. The initial oven
tem◦perature was 100 ◦C◦, and two temperature ramps (100–145 ◦C,
◦
2.5 C min−1, 145–250 C, 30 C min−1) were employed with the
final temperature retained for 50 min. Helium was used as the
carrier gas (flow 1.0 mL min−1). Mass spectra were recorded in
the electronic impact (EI) mode (70 eV, scanning the 30–650 m/z
range).
In the GC/MS measurements, a dichloromethane or methanol
solution (1 mL) of the palladium complex (ca. 3 mg) and
diphenylacetylene (ca. 40 fold excess, 22 mg) were degassed and
placed under a H2 atmosphere (4 atm). The mixture was heated
to 37 ◦C and reacted at this temperature for 1 h with the sample
being shaken every 15 min.
In order to obtain the proportions of product formed in the
catalytic studies, a series of relative response factors were produced
for the different species by the examination of a control sample
containing diphenylacetylene (9.3 mg, 98% purity), cis-stilbene
(10 lL, 10.14 mg, 97% purity), trans-stilbene (11.3 mg, 96% purity)
and 1,2 diphenylethane (8.9 mg) in CH2Cl2 (5 mL).
The main product of the reaction is trans-stilbene in all cases
(except for the dppm system), although significant amounts of cis-
stilbene and cis,cis-1,2,3,4-tetraphenyl-buta-1,3-diene were also
seen by GC/MS. The relatively high proportion of cis-stilbene
formed in the PPh2- phosphine based systems (those using the
precatalysts 1b, 1c and 1d) is consistent with the observation
that trans-stilbene is detected by NMR only after increasing the
samples’ temperature. This observation suggests that there is a
higher energy barrier for b-H elimination in the corresponding
complexes 2, which must form the (not detected) alkene-hydrides
ꢀ
+
=
Pd(LL )(PhCH CHPh)(H) if alkene isomerisation is to be
achieved (Scheme 4).14
The relatively high percentage of cis,cis-1,2,3,4-tetraphenyl-
buta-1,3-diene observed in the reaction using cppe in methanol,
correlates with the relatively greater intensity of the signals for 4e
when compared with those of 4b and 4d. It should be noted, how-
X-Ray data were obtained using a Bruker Smart Apex
diffractometer with Mo–Ka radiation (k = 0.71073 A) using a
ever, that since the two alkyl protons of Pd(LLꢀ)(C(Ph)HCH(Ph)–
˚
+
=
C(Ph) CHPh) (4) are polarised, an equilibrium must exist
SMART CCD camera. Diffractometer control, data collection
and initial unit cell determination was performed using “SMART”
(v5.625 Bruker-AXS). Structures were solved by “direct methods”
using SHELXS-97 (Sheldrick, 1997) and refined by full-matrix
least squares using SHELXL-97 (Sheldrick, 1997).34 All non-
hydrogen atoms were refined anisotropically. Hydrogen atoms
were placed using a “riding model” and included in the refinement
at calculated positions. Frame integration and unit-cell refinement
was carried out with the “SAINT+” software (v6.22, Bruker
AXS). Absorption corrections were applied by SADABS (v2.03,
Sheldrick).
whereby two hydrogen atoms from the same para-hydrogen
molecule can be placed onto the different carbon centres of
=
the CHPh–CHPh–CPh CHPh ligand in an analogous way to
that described for 2 (Schemes 3 and 4). Based on our earlier
ꢀ
+
=
observations of Pd(LL )(py)(PhC CHPh) ], the binding of the
alkyne and subsequent C–C bond formation which will lead to
Pd(LL )(C(Ph) C(Ph)–C(Ph) CHPh) provides a viable route to
4 as indicated in Scheme 3.
ꢀ
+
=
=
Experimental
General conditions
Synthesis of the complexes
All manipulations were carried out under inert atmosphere
conditions, using standard Schlenk techniques (with a vacuum
of up to 10−2 mbar, with N2 or Ar as an inert atmosphere) or
high vacuum techniques (10−4 mbar). Dry N2 and Ar for Schlenk
lines were purchased from BOC Gases. Storage and manipulation
of samples was carried out using standard glove box techniques,
under an atmosphere of N2 using an Alvic Scientific Gas Shield
The triflate complexes used in this study were synthesized
by chloride displacement from their bis-chloride analogues
using silver triflate, according to literature procedures.35
The NMR data of [Pd(dppm)Cl2],36 [Pd(dppe)Cl2],36,37
[Pd(dppp)Cl2],36 [Pd(dcpe)(Cl)2]38 (dcpe = Cy2PCH2CH2PCy2),
[Pd(dppm)(OTf)2],36 [Pd(dppe)(OTf)2]39,40 and [Pd(dppp)(OTf)2]36
are in agreement with that reported in the literature.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4270–4281 | 4279
©

View Article Online
Synthesis of [Pd(dcpe)(OTf)2] (1a). A solution of [Pd(dcpe)Cl2]
(145 mg, 0.24 mmol) in dry, degassed CH2Cl2 (10 mL) was reacted
with AgOTf (137 mg, 0.53 mmol). The resulting suspension was
stirred for 5 h with the exclusion of light. Then, it was filtered and
the filtrate concentrated to ca. 1 mL. Addition of pentane (10 mL,
mixture of isomers) precipitated a solid, which was collected by
filtration, washed with pentane (2 × 2 mL) and dried under
vacuum for 1 h to give the title compound with crystallisation
is also acknowledged. We are also grateful for discussions with
Eite Drent and Karina Q. Almeida Len˜ero.
References
1 H. Lindlar, Helv. Chim. Acta, 1952, 35, 446–456.
2 H. Lindlar and R. Dubuis, Org. Synth., 1966, 46, 89.
3 A. Dedieu, S. Humbel, C. Elsevier and C. Grauffel, Theor. Chem. Acc.,
2004, 112, 305–312.
4 A. M. Kluwer, T. S. Koblenz, T. Jonischkeit, K. Woelk and C. J. Elsevier,
J. Am. Chem. Soc., 2005, 127, 15470–15480.
5 M. W. van Laren, M. A. Duin, C. Klerk, M. Naglia, D. Rogolino,
P. Pelagatti, A. Bacchi, C. Pelizzi and C. J. Elsevier, Organometallics,
2002, 21, 1546–1553.
1
pentane (NMR) as a white solid. Yield: 90 mg, 45%. 31P{ H}
NMR (161.9 MHz, CD2Cl2): d 109.1 (s, PCy2). Anal. Calcd for
C28H48F6O6P2PdS2·1.5C5H12 C, 45.58; H, 7.11. Found: C, 45.17;
H, 6.82%.
6 E.-I. Negishy, Handbook of Organopalladium Chemistry for Organic
Synthesis, Wiley, New York, 2002.
7 D. Canet, C. Aroulanda, P. Mutzenhardt, S. Aime, R. Gobetto and F.
Reineri, Concepts Magn. Reson., Part A, 2006, 28A(5), 321–330.
8 S. B. Duckett and N. J. Wood, Coord. Chem. Rev., 2008,
DOI: 10.1016/j.ccr.2008.1001.1028.
Synthesis of [Pd(cppe)Cl2] (cppe = Ph2PCH2CH2PCy2). 0.93 g
(2.4 mmol) of [Pd(PhCN)2Cl2] were dissolved in 100 mL of dry
CH2Cl2. A solution of 1 g of Ph2PCH2CH2PCy2 (2.4 mmol)
in 20 mL of CH2Cl2 was slowly added over 15 min with
vigorous stirring. The reaction mixture was then stirred overnight.
The solvent volume was reduced to 15 mL and 50 mL of
pentane was added to precipitate the product as an off-white
precipitate which was isolated by filtration to yield 1.1 g of pure
[Pd(Ph2PCH2CH2PCy2)Cl2]·0.5 CH2Cl2. Yield: 78%. 1H NMR
(400 MHz, CD2Cl2): d 8.03 (o-CH), 7.81 (p-CH), 7.75 (m-H),
9 D. Blazina, S. B. Duckett, J. P. Dunne and C. Godard, Dalton Trans.,
2004, 2601–2609.
10 J. P. Dunne, S. Aiken, S. B. Duckett, D. Konya, K. Q. A. Lenero and E.
Drent, J. Am. Chem. Soc., 2004, 126, 16708–16709.
11 J. Lo´pez-Serrano, S. B. Duckett and A. Lledo´s, J. Am. Chem. Soc.,
2006, 128, 9596–9597.
12 J. Lo´pez-Serrano, S. B. Duckett, S. Aiken, K. Q. Almeida Len˜ero, E.
Drent, J. P. Dunne, D. Konya and A. C. Whitwood, J. Am. Chem. Soc.,
2007, 129, 6513–6527.
13 K. K. M. Hii, T. D. W. Claridge, R. Giernoth and J. M. Brown, Adv.
Synth. Catal., 2004, 346, 983–988.
14 J. Lo´pez-Serrano, A. Lledo´s and S. B. Duckett, Organometallics, 2008,
27, 43–52.
15 W. Clegg, G. R. Eastham, M. R. J. Elsegood, B. T. Heaton, J. A. Iggo,
R. P. Tooze, R. Whyman and S. Zacchini, J. Chem. Soc., Dalton Trans.,
2002, 3300–3308.
16 W. Clegg, G. R. Eastham, M. R. J. Elsegood, B. T. Heaton, J. A. Iggo,
R. P. Tooze, R. Whyman and S. Zacchini, Organometallics, 2002, 21,
1832–1840.
1
3.28 (CH2–PCy3), 1.83 (CH2–PPh2). 31P{ H} NMR (161.9 MHz,
CDCl3): d 93.3 (d, JPP = 15 Hz, PCy2), 66.3 (d, JPP
=
15 Hz, PPh2). Anal. Calcd for C26H36Cl2P2Pd·0.5CH2Cl2: C,
50.50; H, 5.92. Found: C, 50.25; H, 5.88%. Single crystals of
[Pd(Ph2PCH2CH2PCy2)Cl2]·CH2Cl2 suitable for X-ray diffraction
studies were grown by slow diffusion of diethyl ether in CH2Cl2
solutions of amorphous [Pd(Ph2PCH2CH2PCy2)(Cl)2]·0.5CH2Cl2.
Synthesis of [Pd(cppe)(OTf)2] (1e). 770 mg (1.32 mmol) of
[Pd(cppe)Cl2] were dissolved in 150 mL of degassed CH2Cl2. To
this solution, 1.30 g (5.29 mmol) of AgOTf were added and the
schlenk tube protected from sunlight. This reaction mixture was
stirred for 48 h at room temperature. Then it was filtered and
reduced in volume to ca. 10 mL. The product was precipitated
with 50 mL of dry diethyl ether, to yield 468 mg of a white solid.
17 D. Konya, K. Q. Almeida Len˜ero and E. Drent, Organometallics, 2006,
25, 3166–3174.
18 M. S. Anwar, D. Blazina, H. Carterer, S. B. Duckett, T. K. Halstead,
J. A. Jones, C. M. Kozak and R. J. K. Taylor, Phys. Rev. Lett., 2004,
93, 040501–040504.
19 J. A. Aguilar, P. I. P. Elliot, J. Lo´pez-Serrano, R. Adams and S. B.
Duckett, Chem. Commun., 2007, 1183–1185.
20 C. R. Bowers and D. P. Weitekamp, Phys. Rev. Lett., 1986, 57, 2645–
2648.
1
Yield: 44%. H NMR (400 MHz, CD2Cl2): d 7.80–7.70 (m, 6 H,
21 J. Natterer and J. Bargon, Nucl. Magn. Reson. Spectrosc., 1997, 31,
293–315.
meta and para H’s PPh2), 7.69–7.61 (m, 4 H, ortho H’s PPh2),
2.91–2.73 (several m, 2 H, CH2PPh2), 2.41 (m, 2 H, CH’s Cy), 2.27
(m, 2 H, Cy), 2.18 (m, 1H, CH2PCy2), 2.10 (m, 1H, CH2PCy2),
2.00–1.73 (several m, 8 H, Cy), 1.61–1.24 (several m, 10 H, Cy).
22 S. B. Duckett and C. J. Sleigh, Prog. Nucl. Magn. Reson. Spectrosc.,
1999, 34, 71–93.
23 S. B. Duckett and D. Blazina, Eur. J. Inorg. Chem., 2003, 16, 2901–2912.
24 C. Godard, S. B. Duckett, S. Polas, R. P. Tooze and A. C. Whitwood,
J. Am. Chem. Soc., 2005, 127, 4994–4995.
1
4
13C{ H} NMR (100 MHz, CD2Cl2): d 133.93 (d, JCP = 2.8 Hz,
25 W. L. Steffen and G. J. Palenik, Inorg. Chem., 1976, 15, 2432–2439.
26 S. Ganguly, J. T. Mague and D. M. Roundhill, Acta Crystallogr., Sect.
C, 1994, 50, 217–219.
27 S. Aime, R. Gobetto and D. Canet, J. Am. Chem. Soc., 1998, 120,
6770–6773.
28 Despite our efforts to fully characterize some of the species 2 or
3 described in this paper, by location of the protons and carbon
resonances of the aryl substituents of their alkyl ligands, only the
reaction of 1e with protio diphenylacetylene afforded very weak
thermally stable NMR peaks for 2e. However, even in this case, the
NMR resonances were not long-lived enough to identify the related
aryl proton resonances of the alkyl ligand, let alone to allow us the
indirect detection of the carbon resonances associated.
29 I. L. Klarte and K. S. Dragonette, Acta Crystallogr., 1965, 19, 500.
30 A. Fu¨rstner, A. Hupperts, A. Ptock and E. Jansen, J. Org. Chem., 1994,
59, 5215–5229.
p-CH Ph), 133.04 (d, 3JCP = 11.3 Hz, m-CH Ph), 130.02 (d, 2JCP
=
4
12.0 Hz, o-CH Ph), 123.94 (d, JCP = 58.0 Hz, i-C Ph), 35.75
1
2
(d, JCP = 24.0 Hz, CH Cy), 30.99 (dd, JCP = 40.3 Hz, JCP
=
6.4 Hz, CH2PPh2), 28.92 (s, Cy), 28.24 (d, JCP = 4.2, Cy), 26.32
1
(m, Cy), 25.21 (d, JCP = 1.4 Hz, Cy), 17.89 (dd, JCP = 31.1 Hz,
2JCP = 9.2 Hz, CH2PCy2). 31P{ H} (161.9 MHz, CD2Cl2): d 105.45
1
(s, PCy2), 73.46 (s, PPh2). Anal. Calcd for C28H36F6O6P2PdS2: C,
41.26; H, 4.45; S, 7.87. Found: C, 40.74; H, 4.45; S, 7.82%.
Acknowledgements
We are grateful to EU funding under the HYDROCHEM network
(contract HPRN-CT-2002-00176). Financial support from the
Spanish MEC (Project Consolider Ingenio 2010 CSD2007-00006)
31 E. Szuromi and P. R. Sharp, Organometallics, 2006, 25, 558–559.
32 J. Vicente, I. Saura-Llamas, J. Turp´ın and M.-C. Ram´ırez deArellano,
Organometallics, 1999, 18, 2683–2693.
4280 | Dalton Trans., 2008, 4270–4281
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
33 G. K. Anderson and M. Lin, Inorg. Synth., 28, 60–62.
34 G. M. Sheldrick, SHELXS-97, Program for solution of crystal struc-
tures, University of Go¨ttingen, Germany, 1997.
35 P. J. Stang, D. H. Cao, S. Saito and A. M. Arif, J. Am. Chem. Soc.,
1995, 117, 6273–6283.
37 C. Bianchini, G. Mantovani, A. Meli, W. Oberhauser, P. Bruggeller and
T. Stampfl, J. Chem. Soc., Dalton Trans., 2001, 690–698.
38 G. T. L. Broadwood-Strong, P. A. Chaloner and P. B. Hitchcock,
Polyhedron, 1993, 12, 721–729.
39 S. Murata and Y. Ido, Bull. Chem. Joc. Jpn., 1994, 67, 1746–1748.
40 S. Fallis, G. K. Anderson and P. R. Nigam, Organometallics, 1991, 10,
3180–3184.
36 D. L. Oliver and G. K. Anderson, Polyhedron, 1992, 11, 2415–
2420.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 4270–4281 | 4281
©