Synthesis of new allyl palladium complexes bearing purine-based NHC ligands with antiproliferative and proapoptotic activities on human ovarian cancer cell lines

Article abstract of DOI:10.1039/c8dt01831f

A series of new palladium allyl complexes bearing purine-based carbenes derived from caffeine, theophylline and theobromine have been prepared and characterized by NMR spectroscopy, and elemental analysis and in two cases by single crystal X-ray diffraction. The cytotoxic and proapoptotic activities of compounds have been determined in vitro on human ovarian cancer A2780 and SKOV-3 cell lines. These experiments have shown that the palladium-allyl fragment induces a general cytotoxicity, but the choice of the supporting ligands is of paramount importance for achieving the best results. In particular complexes 4c, 4d and 5d exhibit a higher antiproliferative effect (IC₅₀: 0.09, 0.81 and 0.85 μM respectively) than cisplatin (IC₅₀: 1.5 μM) on A2780 cells, and 4d (IC₅₀: 1.7 μM vs. 5.94 μM) on SKOV-3 cell line. Moreover in many cases it has been proved that the cytotoxicity of our complexes is associated with the induction of apoptosis.

Full text of DOI:10.1039/c8dt01831f

View Article Online
View Journal
Dalton
Transactions
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: T. Scattolin, L.
Canovese, N. Demitri, R. Gambari, I. Lampronti, C. Santo, F. Visentin, F. Rizzolio and I. Caligiuri, Dalton
Trans., 2018, DOI: 10.1039/C8DT01831F.
This is an Accepted Manuscript, which has been through the
Volume 45 Number
1 7 January 2016 Pages 1–398
Royal Society of Chemistry peer review process and has been
accepted for publication.
Dalton
Transactions
Accepted Manuscripts are published online shortly after
An international journal of inorganic chemistry
www.rsc.org/dalton
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-9226
PAPER
Joseph T. Hupp, Omar K. Farha et al.
Efficient extraction of sulfate from water using
framework
a Zr-metal–organic
rsc.li/dalton

Page 1 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
Synthesis of new allyl palladium complexes bearing purine-based NHC
ligands with antiproliferative and proapoptotic activity on human ovarian
cancer cell lines
Thomas Scattolina, Isabella Caligiurib, Luciano Canovesea, Nicola Demitric, Roberto
Gambarid, Ilaria Lamprontid, Flavio Rizzolioa,b, Claudio Santoa and Fabiano Visentin*a
aDipartimento di Scienze Molecolari e Nanosistemi, Università Ca’ Foscari, Campus Scientifico Via Torino
155, 30174 VeneziaꢀMestre, Italy. Eꢀmail: fvise@unive.ita;
b Pathology Unit, Department of Molecular Biology and Translational Research, National Cancer
Institute and Center for Molecular Biomedicine ꢀ CRO, Aviano, Italy.
cElettra – Sincrotrone Trieste, S.S. 14 Km 163.5 in Area Science Park, 34149 Basovizza, Trieste, Italy,
dDipartimento di Scienze della Vita e Biotecnologie, Università degli Studi di Ferrara, Via Fossato di Mortara
74, 44121 Ferrara, Italy.
Introduction
In the last decades many efforts have been devoted to develop new strategies against
neoplastic pathologies. Even inorganic chemists have taken part in this challenge and starting
from the epochal discovery of the antiꢀcancer properties of cisplatin1, many researchers have
employed their best energies for planning and preparing new metalꢀbased drugs,
unfortunately without obtaining the exciting results of the early years.2 The general problem
of the pharmacologic approach to cancer is connected to the difficulty of predicting every
possible interaction of the drug with the huge number of biomolecules occurring in cellular
and extraꢀcellular systems. The specificity of a metalꢀbased anticancer drug mainly depends
on the possibility of the metal centre of directly binding to the biological target and thus
altering its functionality. DNA is the most recurring genomic target and the ability of
platinum to coordinate strongly to nucleic acid, forming intrastrand crosslinks, is the reason
of the effectiveness of cisplatin and its second and third generation analogues.3,4 A lot of
studies have subsequently proved that many potential metalꢀbased anticancer drugs have
nonꢀgenomic targets (usually proteins) regulating the cell cycle or inhibiting overꢀexpressed
enzymes present in the cancer cells.2d,5

Dalton Transactions
Page 2 of 37
View Article Online
DOI: 10.1039/C8DT01831F
Unfortunately, the possible reactivity with biomolecules different from the specific diseased
sites may induce a general toxicity, which often is the companion side effect of
chemotherapeutic drugs.
In principle a better selectivity could be obtained when the metal has a structural role i.e. it is
functional to achieve the appropriate shape of the complex. In these cases binding to the
molecular target occurs through nonꢀcovalent interactions in a manner similar to most
organic drugs. Within this context, the problem is the actual difficulty of precisely defining
the shape and the size of the target site and consequently synthesizing the appropriately
tailored compounds.2g
These preliminary remarks highlight that the rational design of a metalꢀbased anticancer drug
is an elegant but often impracticable strategy. Therefore an extensive screening of antiꢀ
tumour properties of different metal compounds may compensate for our inability to control
their almost unpredictable reactivity in the bloodstream and cellular environment. In our
choice to contribute to this systematic search we have preliminarily set some guidelines
suggested by previous literature contributions. Firstly, we have opted for palladiumꢀbased
compounds. Palladium complexes have structures and reactivity strictly comparable to
Platinum analogues and in several cases have shown better antiꢀcancer activity than
cisplatin6. However the fast dissociation pattern of palladium complexes compared to
platinum7 represents a problem since the speciation, which heavily affects the biological
activity and the pharmacokinetic properties, might be increased. To remedy this
contraindication the most direct option is the introduction of ligands firmly anchored to the
metal. For this reason we planned to employ NꢀHeterocylic carbenes (NHCs), which are
known to give strong σꢀbonds with most of the transition metals8. Moreover, several NHCꢀ
palladium complexes have already exhibited an interesting cytotoxic activity9,6a,e and tumour
growth suppression even in vivo.6e
A potential improvement introduced in our work consists in using some innovative NHC
ligands with a purinic framework10 in hope that the natural imprint of the moiety could make
our palladium complexes more compatible with the biological matrix (Chart 1). This kind of
synergy has already given interesting outcomes and in the most favourable cases the ligand
has become a real targeting vector of the metal compound.11

Page 3 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
Chart 1: Purineꢀbased NHCs ligands used in this work
Furthermore, with the aim at enhancing the versatility of our palladium complexes
sometimes also isocyanides or phosphines have been introduced into the coordination sphere.
Among the latter we tested PTA (1,3,5ꢀtriazaꢀ7ꢀphosphadamantane) and TPPTS (3,3',3"ꢀ
phosphinetribenzenesulfonate), which can increase the waterꢀsolubility of the metal
compounds and thereby make their pharmacologic application easier.12
However the most original contribution of this paper is represented by the first attempt to
utilize the palladiumꢀη3ꢀallyl organometallic function for preparing compounds with
cytotoxic properties. This organometallic fragment is well known for its involvement in
13
numerous catalytic processes (the most famous is the TsuijꢀTrost reaction)
but its
behaviour in biological systems remains an almost unexplored field. In previous works some
of us have prepared numerous compounds based on this functional group and studied their
properties and reactivity.14 This experience suggests that the allyl residue could: a) remain
bound to the metal centre, conferring a specific shape to the complex and thereby allowing it
to be hosted in an active biological site; b) be released in the cellular environment and thus
directly interact with some potential bioꢀtarget. The main goal of this paper is not to resolve
this intricate question but analyse its practical consequences. Therefore, in order to verify the
effects of our new palladium complexes, the potential antiproliferative activity was tested on
two tumour cell lines (A2780 and SKOVꢀ3) displaying a different sensitivity to cisplatin. and
for four representative complexes also on human normal fibroblasts MRC5. To obtain
further information on the possible mechanism(s) of action, apoptosis was analysed as proꢀ
apoptotic activity of palladium compounds.15

Dalton Transactions
Page 4 of 37
View Article Online
DOI: 10.1039/C8DT01831F
Results and Discussion
Synthesis of Imidazolium Salts from Functionalized Xanthines
Compounds 1b,16a 1c16b and 1d17 were reꢀsynthesized adopting a slightly modified literature
procedure (Scheme1). We have chosen not particularly severe conditions and the reaction of
theophylline or theobromine with two equivalents of benzylꢀbromide (for 1b and 1d) or 1ꢀ
(phenyl)propargylꢀ3ꢀchloride (for 1c), carried out in DMF at R.T. overnight, gave the
functionalized bases in good yields (93
95%).
Scheme 1 Synthesis of functionalized xanthines
Despite the low reactivity of the sp2 N9 atom, commercial caffeine (1a), the alkylated
theophyllines 1b, 1c and theobromine 1d react with a small excess of Meerwein's salt in
CH3CN at R.T. in the presence of Na2CO3 under not controlled atmosphere to give the
corresponding N9 methylated imidazolium salts in one hour’s time. (Scheme 2). The
acetonitrile is probably more suitable for promoting the nucleophilic substitution than
chlorinated solvents thanks to its high dielectric constant, whereas the presence of Na2CO3 in
heterogeneous phase is necessary since the strong HBF4 acid that is partially formed from the
hygroscopic Me3OBF4 could protonate the N9 nitrogen of the xanthine so that the ensuing
derivative would be no longer available for methylation.

Page 5 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
Scheme 2 Synthesis of imidazolium salts from functionalized xantines
1
As an example we report in Figure S1 (see ESI) the H NMR spectra related to the
methylation of caffeine (1a) which show that, in the absence of Na2CO3, methyl caffeine 2a
coexists with its protonated counterpart, whereas only pure 2a derivative was detected in
solution when the base was added to the reaction mixture.
In this respect, we think that the low yield obtained by other authors in the Meerwein’s salt
mediated methylation of the modified xanthines is probably due to this side reaction, which
makes the workup of the process difficult.15,18
Synthesis of Silver NHC complexes
We recently reported that the reaction between the imidazolium salts 2a, 2b, 2d and Ag2O
carried out in acetonitrile produces a 1:1 mixture of the corresponding silver bisꢀNHC
complexes 3 and AgBF4.19 This result was confirmed by a combination of spectroscopic and
analytical data such as microanalysis, NMR, ESIꢀMS spectra and in one case (3b) even Xꢀ
ray diffraction analysis. In this paper we have applied this protocol for the synthesis of the
new complex 3c. The disappearance of the imidazolic proton and the typical shift of the

Dalton Transactions
Page 6 of 37
View Article Online
DOI: 10.1039/C8DT01831F
1
silverꢀcoordinated carbon at ca. 187 ppm in the H and 13C NMR spectra respectively,
represent simple tests to certify the predictable outcome of the reaction.
Scheme 3 Synthesis of silver NHC purineꢀbased complexes
Synthesis of Mixed NHC/Phosphine and NHC/Isocyanide Palladium η3-allyl complexes
For the synthesis of the palladium complexes 4-7 we can take advantage of the presence of
coꢀprecipitated AgBF4 as a stoichiometric byꢀproduct in the preparations of silver complexes
3. As a matter of fact this condition allows to obtain the final complexes by oneꢀpot reaction,
by mixing the aforesaid mixture with a stoichiometric amount of palladium allyl dimer and
phosphine (PPh3, PTA, TPPTS) or isocyanide (DIC = 2,6ꢀdimetilphenyl isocyanide)
(Scheme 4). The precipitation of silver chloride is the driving force of the process and an
indication of its progress. It is important to highlight that no formation of complexes bearing
carbene/carbene and phosphineꢀphosphine (or isocyanideꢀisocyanide) ligands has been
observed and therefore a selective separation of the carbene/phosphine (or isocyanide)
complexes is always possible from the reaction mixture.

Page 7 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
Scheme 4 Synthesis of mixed NHC/phosphine and NHC/isocyanide palladium allyl
complexes
The characterization of the thirteen new complexes 4a-d, 5a,b,d and 6a,b,d (compounds
7a,b,d were reported in one of our previous works19) was performed by NMR, IR and
elemental analysis. In particular an accurate examination of NMR spectra allows to elucidate
some specific features of these derivatives that we can summarize on the following points:
a) In every NMR spectrum of the complexes containing a coordinated phosphine (4a-d
and 5a,b,d) the presence of two complete sets of signals is manifest. This is a
consequence of the coexistence of a couple of atropoisomers due to the hindered
rotation of the asymmetric carbene ligands around the PdꢀC bond. The two isomers
are always present in virtually equal amounts.
For the less sterically crowded isocyanideꢀcomplexes 6a,b,d this restriction is not
operative and thus only one set of signals is observable in solution at room
temperature.
1
b) The presence in the H NMR spectra of five distinct signals (for each atropoisomer)
ascribable to allyl protons can be explained by the presence of two different spectator
ligands and by the absence of any rearrangement (η3ꢀη1ꢀη3 or syn-syn/anti-anti)
sometimes observed even at room temperature.20
c) The coordination of PPh3 and PTA (for complexes 4a-d and 5a,b,d respectively) is
proved by the marked downfield shift of the two peaks (one for each atropoisomer)
observed in the 31P{1H } NMR spectra (ꢀδ = 30 ꢀ 45 ppm with respect to the free
phosphines).
d) The coordination of DIC in complexes 6a,b,d is certified by the weak signal of the
13
coordinated isocyanide carbon at ~150 ppm in the C{1H }1H NMR spectra and the
significant highfield shift of the ortho methyl protons with respect to their original
1
position in the H NMR spectra of the free ligand. This conclusion is also supported
by the IR spectra which show an intense band at 2170 cmꢀ1 attributable to the CN
stretching of the coordinated isocyanide.

Dalton Transactions
Page 8 of 37
View Article Online
DOI: 10.1039/C8DT01831F
e) The resonances of carbene carbons are always found in a narrow range between 180ꢀ
190 ppm, confirming the coordination of the purineꢀbased NHC ligand at the
palladium(II) centre. Moreover in the spectra of the compounds 4a-d and 5a,b,d
these signals resonate as doublets due to the J2 coupling with the cisꢀphosphine.
The cationic nature of synthesized derivatives is indirectly confirmed by the presence in all
ꢀ
IR spectra of an intense absorption around 1050 cmꢀ1, ascribable to the BF4 anion.
Synthesis of Bis(NHC) Palladium η3-allyl Complexes
This class of complexes were synthesized by reacting the stoichiometric mixture of silver
precursors 3 and AgBF4 with palladium allyl dimer in the presence of KI (Scheme 5). The
addition of potassium iodide is necessary to remove all the silver from the reaction mixture.
Transmetallation proceeds in good yield and the resulting cationic complexes were all stable
in solution and in air.
Scheme 5 Synthesis of bis(NHC) palladium allyl complexes
The most significant conclusion that can be immediately inferred from the NMR spectra of
species 8 is the presence of only one set of signals. This observation indicates that there is a
free rotation of the two Nꢀheterocyclic carbene ligands around the PdꢀC bond and thus the
three hypothetical atropoisomers deriving from hindered rotation are not distinguishable at
room temperature.
The symmetric structure of these complexes is demonstrated by the presence of only three
signals ascribable to the allylic fragment in the 1H NMR spectra (two doublets with different
coupling constants for the syn and anti terminal protons and a multiplet for the central

Page 9 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
proton) and two in the 13C{1H }NMR spectra.
13
Finally the CNHCꢀPd(II) resonances at about 185 ppm represent a compelling evidence of
the coordination of carbene ligands to the palladium centre.
X-Ray Crystal Structure Determination of 4a e 8d
Crystalline forms of 4a and 8d contain one crystallographically independent palladium
ꢀ
complex (Figure1 and 2). The complexes bear a positive charge that is balanced by a BF4
counterion, located close to the allyl ligand, which represents the area where the metal is
more exposed (shortest F···Pd contacts are 3.50(1) Å in 4a and 3.489(8) Å in 8d). Palladium
adopts square planar coordination spheres with bond lengths and angles (Table 2SI) in
agreement with literature structural data of complexes with similar ligands. A query on CSD
(version 5.38), using the AllylꢀPdꢀNHC fragment produces 29 hits with an average bite angle
of 68.3(6)° and 2.04(1) Å Pd···Ccarbenic bond lengths (2.32(1) Å for Pd···P bonds). The allyl
ligand has poor steric bulk constraints and can adopt alternative conformations, specular with
respect to the palladium coordination plane, as can be seen in more than half of similar
structures already published. Models show that the xanthines minimize steric repulsions in
the solid state, adopting a roughly perpendicular orientation with respect to the palladium
coordination plane (85.68° in 4a and 69.19°ꢀ75.98° in 8d, in agreement with the average
79(9)° extracted from CSD). The molecular model of 4a is well superimposable with the
related
triphenylphosphineꢀ(η3ꢀallyl)ꢀ(tetramethylimidazolinꢀ2ꢀylidene)ꢀpalladium
tetrafluoroborate complex21 (CCDC Number: 714135); the comparison highlights a degree of
phosphine ligand conformational freedom (Figure S27), which can be related to different
crystal packing contacts. Crystal packing of 4a and 8d shows hydrophobic contacts among
neighbour molecules, involving several CH···π and minor π···π interactions. Furthermore,
structure 4a has cavities (258 Å3, estimated with PLATON22 ‘CALC VOID’ routine) filled
with disordered solvent molecules (one CH2Cl2 molecule for each cavity).

Dalton Transactions
Page 10 of 37
View Article Online
DOI: 10.1039/C8DT01831F
Figure 1 Ellipsoid representation of 4a (B) crystal ASU contents (50% probability).
Figure 2 Ellipsoid representation of 8d (B) crystal ASU contents (50% probability).

Page 11 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
Antiproliferative Activity on Human A2780, SKOV-3 and MRC-5 Cell Lines
The antiproliferative activity was assayed on two human cancer cell lines: A2780 and
SKOVꢀ3. Stock solutions (25ꢀ50 mM) of each complex were prepared in DMSO (8), H2O (5
and 7), or DMSO/H2O (1/1) (4 and 6); the working solutions (5 mM, 500 ꢁM, 50 ꢁM and
5ꢁM) were obtained using H2O only. Preliminarily the stability of complexes was checked in
1:1 DMSOꢀd6/D2O solution (4, 6 and 8) or D2O (5 and 7). After 48 hours at room
temperature no degradation and no ligand replacement was observed (see NMR spectra in
Figures S28ꢀ S30 ).
All the complexes 4a-d, 5a,b,d, 6a,b,d, 7a,b,d, 8a-d and the synthetic precursor [Pd(ꢁꢀ
Cl)(η3ꢀC3H5)]2 have been tested on cisplatinꢀsensitive A2780 cell line and on cisplatinꢀ
resistant SKOVꢀ3 cell line in comparison to cisplatin (positive control), in order to evaluate
the relative IC50 values expressed in ꢁM concentration. All data obtained are reported in
Table 1 for both the analysed cell lines. As expected cisplatin was more active on A2780
cells (IC50 = 1.5 ± 0.2 ꢁM) with respect to SKOVꢀ3 cells (IC50 = 5.94 ± 0.08 ꢁM). In addition
the results reported in Table 1 show that all the complexes, except for 7a, displayed good
antiꢀproliferative effects on the cisplatinꢀsensitive A2780 cells. The most active of them are
derivatives 4c, 4d and 5d, which exhibit a greater activity than the positive control cisplatin
(IC50: 1.5 ± 0.2 ꢁM), showing 0.09 ± 0.02, 0.81 ± 0.08 and 0.8 ± 0.2 ꢁM IC50 values,
respectively. On the cisplatinꢀresistant SKOVꢀ3 cells, several compounds (5d, 7a, 7b, 7d,
8b, 8c and 8d) exhibited lower activity (IC50 > 30 ꢁM) with respect to cisplatin whereas most
of them displayed a similar activity (IC50 = 3ꢀ8 ꢁM). Remarkably, compounds 4b, 5a, 5b,
6a, 6d exhibited a higer activity on SKOVꢀ3 cells with respect to A2780 cells. Furthermore
the complex 4d showed a very interesting activity on the SKOVꢀ3, exhibiting an IC50 value
(1.7 ± 0.9 ꢁM) lower than that of cisplatin. The imidazolium salts 2a and 2b, also tested on
the same cell models, were completely inactive (data not shown) at the concentrations used
on both cell lines (IC50>100 ꢁM).

Dalton Transactions
Page 12 of 37
View Article Online
DOI: 10.1039/C8DT01831F
Table 1. Effects of the Pdꢀcomplexes on the proliferation of A2780 and SKOVꢀ3
cells. The inhibition of cell growth is represented as IC50.
IC50 (ꢁM)
Complex
A2780 (IC50)
SKOV-3 (IC50)
Cisplatin
1.5 ± 0.2
5.94 ± 0.08
[Pd(ꢁꢀCl)(η3ꢀC3H5)]2
7.8 ± 0.2
4 ± 2
10 ± 4
5.±1
4a
4b
4c
4d
5a
5b
5d
6a
6b
6d
7a
7b
7d
8a
8b
8c
8d
4 ±2
3 ± 2
0.09 ± 0.02
0.81 ± 0.08
7 ± 2
4.02 ± 0.09
1.7 ± 0.9
5.20 ± 0.08
6.5 ± 0.7
50.5 ± 0.5
4.0 ± 0.7
5.2 ± 0.9
3 ± 1
7.60 ± 0.07
0.8 ± 0.2
5 ± 2
3.72 ± 0.06
5 ± 1
80 ± 10
7 ± 2
70 ±10
40 ± 2
9.2 ± 0.8
7 ± 2
57 ± 8
7.7 ± 0.4
61 ± 7
4 ± 1
3.7 ± 0.5
6.3 ± 0.7
60 ± 6
38 ± 6

Page 13 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
To test the selectivity of Pdꢀcomplexes versus cancer cells, normal human fibroblasts (MRCꢀ
5) were treated with four different compunds (4d, 5d, 6b, 8a) representative of different
subclasses with high activity on cancer cells. As shown in table S1, the compounds are
almost inactive on fibroblasts suggesting a preferential activity on cancer cells. In particular,
the compound 4d is more active than 5d, 6b, 8a on A2780 and SKOV3 cancer cell lines but
with an IC50>100 ꢁM on fibroblasts.
Pro-apoptotic Effects on Human A2780 and SKOV-3 Cell Lines
In order to assess whether the antiproliferative activity of the analysed Pdꢀcomplexes is
associated with induction of apoptosis, both A2780 and SKOVꢀ3 cells were also evaluated
employing the Annexin V release assay and the Muse cytometer (Merck Millipore, Billerica,
MA, USA), after treatment with two concentrations close to the IC50 values previously
determined. The found proꢀapoptotic effects are reported in Table 2, Figures 3 and 4
(representative examples) and in Figures 1AꢀE and 2AꢀE in the supplementary materials (all
results). In Table 2, the percentage of total apoptosis is reported. Each data entry is compared
to the analysis of untreated cells (negative control, Cꢀ) in which the observed total apoptosis
was ≤ 5%.
On the A2780 cell line, the complexes that showed the highest proꢀapoptotic activity were
4c, 4d, 6b, 6d, 8b, 8c, and 8d, with total proꢀapoptotic effects between 21.00% (4d) and
98.35% (4c); in addition, complex 4c showed the greatest total proꢀapoptotic activity even
with the lowest concentration tested (54.04%). The remaining compounds were found to be
moderately or poorly active in inducing apoptosis on this cisplatinꢀsensitive cell line, and the
synthetic precursor [Pd(ꢁꢀCl)(η3ꢀC3H5)]2 determined a proꢀapoptotic effect on 29.40% of
cells.
On the cisplatinꢀresistant SKOVꢀ3 cell line, derivatives 4b, 4c, 4d, 5d, 6a, 6b, 6d, 7a, 7b,
7d, 8a, 8b, 8c, 8d were found to be particularly active, with total proꢀapoptotic effects
between 27.49% (5d) and 84.04% (4c). In particular, complex 4c showed the highest activity
even at the lowest concentration used (1 ꢁM) with 43.78% of apoptotic cells. Table 2 also
reports the proportion of dead cells, suggesting that apoptotic effects and antiproliferative
activity are not associated with higher cytotoxicity.

Dalton Transactions
Page 14 of 37
View Article Online
DOI: 10.1039/C8DT01831F
Table 2. Proꢀapoptotic effects of the Pdꢀcomplexes on A2780 and SKOVꢀ3 cell lines detected
at two different concentrations. (Cꢀ: untreated cells)
Total apoptosis (%)
Dead cells (%)
Total apoptosis (%)
Dead cells (%)
Complex
A2780
A2780
SKOV-3
SKOV-3
Cꢀ
2.61
1.96
5.30
0.50
Cisplatin
38.25 (2.5 ꢁM)
3.30 (2.5 ꢁM)
13.33 (5 ꢁM)
1.17 (5 ꢁM)
[Pd(ꢁꢀCl)(η3ꢀC3H5)]2
8.15 (1 ꢁM)
29.40 (10 ꢁM)
3.80 (1 ꢁM)
12.35 (10 ꢁM)
8.35 (1 ꢁM)
2.00 (1 ꢁM)
1.00 (10 ꢁM)
0.30 (1 ꢁM)
0.25 (10 ꢁM)
0.50 (1 ꢁM)
0.33 (10 ꢁM)
6.55 (1 ꢁM)
7.10 (10 ꢁM)
6.35 (1 ꢁM)
12.71 (10 ꢁM)
7.25 (1 ꢁM)
0.55 (1 ꢁM)
2.20 (10 ꢁM)
5.35 (1 ꢁM)
4.86 (10 ꢁM)
8.05 (1 ꢁM)
1.12 (10 ꢁM)
4a
4b
10.66 (10 ꢁM)
44.31 (10 ꢁM)
4c
4d
5a
54.04 (0.1 ꢁM)
98.35 (1 ꢁM)
2.25 (1 ꢁM)
21.00 (10 ꢁM)
3.00 (1 ꢁM)
0.25 (0.1 ꢁM)
0.24 (1 ꢁM)
7.62 (1 ꢁM)
1.45 (10 ꢁM)
0.30 (1 ꢁM)
0.00 (10 ꢁM)
43.78 (1 ꢁM)
84.04 (10 ꢁM)
6.50 (0.5 ꢁM)
52.35 (5 ꢁM)
2.45 (1 ꢁM)
15.83 (0.1 ꢁM)
3.16 (1 ꢁM)
3.70 (1 ꢁM)
0.57 (10 ꢁM)
0.10 (1 ꢁM)
0.65 (10 ꢁM)
15.25 (10 ꢁM)
3.10 (10 ꢁM)
5b
5d
6a
6b
6d
7a
7b
7d
8a
8b
8c
8d
4.25 (1 ꢁM)
7.05 (10 ꢁM)
0.80 (1 ꢁM)
0.15 (10 ꢁM)
3.00 (1 ꢁM)
3.65 (10 ꢁM)
0.45 (1 ꢁM)
0.65 (10 ꢁM)
5.49 (1 ꢁM)
5.76 (10 ꢁM)
2.32 (1 ꢁM)
1.43 (10 ꢁM)
9.82 (50 ꢁM)
27.49 (100 ꢁM)
3.06 (1 ꢁM)
1.71 (10 ꢁM)
9.50 (1 ꢁM)
9.20 (10 ꢁM)
0.60 (1 ꢁM)
0.15 (10 ꢁM)
15.05 (1 ꢁM)
34.55 (10 ꢁM)
0.60 (1 ꢁM)
0.15 (10 ꢁM)
26.52 (1 ꢁM)
61.18 (10 ꢁM)
0.59 (1 ꢁM)
0.18 (10 ꢁM)
6.26 (1 ꢁM)
69.10 (10 ꢁM)
0.59 (1 ꢁM)
0.18 (10 ꢁM)
1.40 (1 ꢁM)
79.45 (10 ꢁM)
8.92 (1 ꢁM)
0.34 (10 ꢁM)
3.45 (1 ꢁM)
52.95 (10 ꢁM)
0.90 (1 ꢁM)
0.15 (10 ꢁM)
3.20 (50 ꢁM)
2.21 (100 ꢁM)
2.10 (50 ꢁM)
1.31 (100 ꢁM)
12.55 (50 ꢁM)
44.96 (100 ꢁM)
0.36 (50 ꢁM)
0.29 (100 ꢁM)
2.66 (1 ꢁM)
1.60 (10 ꢁM)
2.26 (1 ꢁM)
1.30 (10 ꢁM)
11.25 (25 ꢁM)
53.01 (50 ꢁM)
0.15 (1 ꢁM)
0.25 (10 ꢁM)
2.97 (1 ꢁM)
5.65 (10 ꢁM)
0.86 (1 ꢁM)
0.96 (10 ꢁM)
8.73 (50 ꢁM)
36.37 (100 ꢁM)
0.51 (1 ꢁM)
0.34 (10 ꢁM)
2.65 (1 ꢁM)
3.50 (10 ꢁM)
0.20 (1 ꢁM)
0.25 (10 ꢁM)
3.65 (1 ꢁM)
30.89 (10 ꢁM)
0.35 (1 ꢁM)
0.46 (10 ꢁM)
5.69 (1 ꢁM)
43.03 (10 ꢁM)
1.86 (1 ꢁM)
0.34 (10 ꢁM)
6.32 (50 ꢁM)
56.60 (100 ꢁM)
0.41 (1 ꢁM)
0.44 (10 ꢁM)
1.39 (1 ꢁM)
23.66 (10 ꢁM)
7.28 (1 ꢁM)
3.09 (10 ꢁM)
7.84 (50 ꢁM)
52.01 (100 ꢁM)
0.39 (1 ꢁM)
0.38 (10 ꢁM)
5.90 (1 ꢁM)
61.17 (10 ꢁM)
0.56 (1 ꢁM)
0.00 (10 ꢁM)
12.74 (25 ꢁM)
79.32 (50 ꢁM)
1.29 (1 ꢁM)
3.18 (10 ꢁM)

Page 15 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
Figure 3 Representative apoptosis profile of A2780 cells untreated (C-), treated with
cisplatin (C+) and with complexes 4c, 6d and 8b (1ꢀ10 ꢁM) for 72 h.
Figure 4 Representative apoptosis profile of SKOVꢀ3 cells untreated (C-), treated with
cisplatin (C+) and with complexes 4c, 6d and 8b (1ꢀ10 ꢁM) for 72 h.

Dalton Transactions
Page 16 of 37
View Article Online
DOI: 10.1039/C8DT01831F
Conclusions
In this contribution we have proposed a complete and optimised protocol for the synthesis of
new cationic allyl palladium complexes bearing NHC ligands with purinic framework. The
combination of the palladiumꢀallyl organometallic function with the natural imprint of the
spectator ligand seems to play a crucial role in promoting the cytotoxic properties of these
compounds. Their antiproliferative activity has been tested in vitro on cisplatinꢀsensitive
A2780 and cisplatinꢀresistant SKOVꢀ3 cell lines and compared with cisplatin as etalon drug.
The complexes 4c, 4d, and 5d (for A2780 cells) and 4d (for SKOVꢀ3 cells) displayed higher
antiproliferative effects than cisplatin, showing that the simultaneous coordination of one
triphenylphosphine and one Nꢀheterocyclic carbene on the palladium allyl fragment
represents the most promising configuration. However, the most active compunds on cancer
cells are almost inactive on normal cells, suggesting a marked cancer selectivity.
Moreover it was proved that the antiꢀproliferative activity of many of the synthesized
complexes is associated with induction of apoptosis. No major cytotoxicity was observed
after evaluation of the proportion of dead cells (Table 2).
Despite the fact that our data do not clarify the molecular target(s) of the compounds
presented here, the apoptosis network should be considered as a major cellular target.
A future extension of this first encouraging study will have as its main goals the
identification of the true molecular target of these allyl palladium complexes and the
definition of their mode of action. As a further possible application these compounds might
be considered for combined antitumor therapy after coꢀadministration with proꢀapoptotic
molecules exhibiting a verified mechanism of action (paclitaxel, topotecan, doxorubicin,
etoposide, gemcitabine, or with antineoplastic molecules displaying different mechanisms of
actions).33ꢀ41A recognized advantage of the combined antitumor approach is the expected
lower side effects associated with the use of minor concentrations. In any case, in vivo
activity of the most efficient compounds studied here would clarify whether the described
effects on tumor cell lines is accompanied by the absence of side effects on normal cells and
networks. Only in this case these complexes might be further considered for possible
biomedical application in the field of anticancer treatment.

Page 17 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
Experimental
All syntheses of complexes were carried out using standard Schlenk techniques under an
atmosphere of dry nitrogen. Solvents were dried and distilled according to standard methods:
CH2Cl2 was firstly treated with 3Å molecular sieves and then distilled over P2O5 ; CH3CN
was distilled over CaH2 and stored over 3Å molecular sieves. All the other chemicals were
commercially available grade products and were used as purchased.
Complexes 3a, 3b, 3d and 7a, 7b, 7d were synthesized according to the procedure described
in a previous work.19
1
The IR, H, 13C and 31P NMR spectra were recorded on a PerkinꢀElmer Spectrum One
spectrophotometer and on a Bruker 300 or 400 Avance spectrometer, respectively.
Elemental analysis was carried out using an Elemental CHN "CUBO Micro Vario" analyzer.
ESIꢀMS analyses were performed using a LCQꢀDuo (ThermoꢀFinnigan) operating in positive
ion mode. Instrumental parameters: capillary voltage 10 V, spray voltage 4.5 kV, capillary
temperature 200 °C, mass scan range from 150 to 2000 amu, N2 was used as sheath gas and
the He pressure inside the trap was kept constant.
The pressure directly read by an ion gauge (in the absence of N2 stream) was 1.33 x 10ꢀ5
Torr. Sample solutions were prepared by dissolving the compounds in acetonitrile and
directly infused into the ESI source with a syringe pump at 8 µL min−1 flow rate.
General procedure for the synthesis of alkyl functionalized caffeine, theophylline and
theobromine (1)
5.55 mmol (ca. 1g) of the starting xanthine (theophylline or theobromine) and 1.15 g (8.33
mmol) of K2CO3, were suspended in 20 mL of DMF. Two equivalents (11.1 mmol) of the
appropriate alkyl halide were added to the suspension and the resulting mixture was
vigorously stirred overnight at R.T. Finally the compounds were precipitated by addition of
H2O (ca. 100 mL).
7-benzyl-1,3-dimethylxanthine (1b)
White solid, yield 93 %.
1HꢀNMR (400 MHz, CDCl3, T=298K, ppm) δ: 3.43 (s, 3H, NꢀCH3), 3.60 (s, 3H, NꢀCH3),
5.52 (s, 2H, NꢀCH2), 7.30ꢀ7.40 (m, 5H, Ph), 7.58 (s, 1H, NCHN).13C{1H}ꢀNMR (CDCl3,

Dalton Transactions
Page 18 of 37
View Article Online
DOI: 10.1039/C8DT01831F
T=298K, ppm) δ: 28.0 (NꢀCH3), 29.8 (NꢀCH3), 50.3 (NꢀCH2), 107.0 (C5), 128.0, 128.7,
129.1, 135.4 (CPh), 140.9 (NꢀCHꢀN), 148.9 (C4), 151.7 (C=O), 155.3 (C=O).
ESIꢀMS (CH3CN): m/z Calcd. for C14H15N4O2 [M+H]+ 271.12; found: 271.06.
Anal. Calcd. for C14H14N4O2: C 62.21, H 5.22, N 20.73. Found: C 62.32, H 5.14, N 20.64.
7-phenylpropargyl-1,3-dimethylxanthine (1c)
Pink solid, yield 95 %.
1HꢀNMR (400 MHz, CDCl3, T=298K, ppm) δ: 3.45 (s, 3H, NꢀCH3), 3.63 (s, 3H, NꢀCH3),
5.42 (s, 2H, NꢀCH2), 7.32ꢀ7.50 (m, 5H, Ph), 7.94 (s, 1H, NCHN). ≡
13C{1H}ꢀNMR (CDCl3, T=298K, ppm) δ: 28.0 (NꢀCH3), 29.8 (NꢀCH3), 37.4 (NꢀCH2), 80.5
(≡CꢀCH2), 87.6 (≡CꢀPh), 106.8 (C5), 121.5, 128.5, 129.2, 131.9 (CPh), 140.6 (NꢀCHꢀN),
148.9 (C4), 151.7 (C=O), 155.3 (C=O).
ESIꢀMS (CH3CN): m/z Calcd. for C16H15N4O2 [M+H]+ 295.12; found: 294.98.
Anal. Calcd. for C16H14N4O2: C 65.30, H 4.79, N 19.04. Found: C 65.49, H 4.83, N 18.89.
1-benzyl-3,7-dimethylxanthine (1d)
White solid, yield 95 %.
1HꢀNMR (400 MHz, CDCl3, T=298K, ppm) δ: 3.59 (s, 3H, NꢀCH3), 4.00 (s, 3H, NꢀCH3),
5.22 (s, 2H, NꢀCH2), 7.25ꢀ7.55 (m, 5H, Ph), 7.51 (s, 1H, NCHN). 13C{1H}ꢀNMR (CDCl3,
T=298K, ppm) δ: 29.8 (NꢀCH3), 33.6 (NꢀCH3), 44.5 (NꢀCH2), 107.7 (C5), 127.5, 128.4,
128.8, 137.3 (CPh), 141.5 (NꢀCHꢀN), 148.9 (C4), 151.6 (C=O), 155.3 (C=O).
ESI ESIꢀMS (CH3CN): m/z Calcd. for C14H15N4O2 [M+H]+ 271.12; found: 271.14.
Anal. Calcd. for C14H14N4O2: C 62.21, H 5.22, N 20.73 . Found: C 62.45, H 5.07, N 20.99.
General procedure for the synthesis of the imidazolium salts from functionalized
theophylline and theobromine (2)
Into a 50 mL flask, 1.8 mmol of functionalized xanthine (1a d with 1a = commercial
caffeine) were dissolved in ca. 25 mL of CH3CN. After the addition of one equivalent (1.8
mmol) of Me3OBF4 the resulting solution was vigorously stirred for 5 min and then 100 mg
of Na2CO3 were added to the mixture which was stirred for 45 min. Further addition of 0.6
equivalents (1.1 mmol) of Me3OBF4 and 10 min of additional stirring led to virtual

Page 19 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
completion. The excess of base Na2CO3 and the NaBF4 formed were filtered off and the
solvent completely removed under reduced pressure. The solid was washed with three
aliquots of a 2:1 mixture of Et2O/ CH2Cl2 on a sintered glass filter, dried under vacuum and
characterized.
1,3,7,9-tetramethylxanthinium tetrafluoroborate (2a)
White solid, yield 97 %, m. p. = 132ꢀ133 °C.
1HꢀNMR (400 MHz, CD3CN, T=298K, ppm) δ: 3.35 (s, 3H, NꢀCH3), 3.74 (s, 3H, NꢀCH3),
4.08 (s, 6H, 2NꢀCH3), 8.47 (s, 1H, NCHN). 13C{1H}ꢀNMR (CD3CN, T=298K, ppm) δ: 28.7
(NꢀCH3), 31.8 (NꢀCH3), 36.3 (NꢀCH3), 37.6 (NꢀCH3), 109.1 (C5), 139.3(NꢀCHꢀN), 140.2
(C4), 151.1 (C=O), 154.2 (C=O).
ESIꢀMS (CH3CN): m/z Calcd. for C9H13N4O2 [M]+ 209.10; found: 209.07.
Anal. Calcd. for C9H13BF4N4O2: C 36.52, H 4.43, N 18.93. Found: C 36.74, H 4.61, N 18.88
7-benzyl-1,3,9-trimethylxanthinium tetrafluoroborate (2b)
White solid, yield 91 %, m. p. = 251ꢀ253 °C
1HꢀNMR (400 MHz, CD3CN, T=298K, ppm) δ
4.06 (s, 3H, NꢀCH3), 5.70 (s, 2H, NꢀCH2), 7.45ꢀ7.48 (m, 5H, Ph), 8.56 (s, 1H, NCHN).
13C{1H}ꢀNMR (CD3CN, T=298K, ppm) δ
28.8 (NꢀCH3), 31.8 (NꢀCH3), 37.9 (NꢀCH3),
3.33 (s, 3H, NꢀCH3), 3.73 (s, 3H, NꢀCH3),
52.7 (NꢀCH2), 108.4 (C5), 129.2, 129.7, 129.8, 133.9 (CPh), 138.9(NꢀCHꢀN), 140.6 (C4),
151.0 (C=O), 154.0 (C=O).
ESIꢀMS (CH3CN): m/z Calcd. for C15H17N4O2 [M]+ 285.13; found: 285.04.
Anal. Calcd. for C15H17BF4N4O2: C 48.41, H 4.60, N 15.06. Found: C 48.39, H 4.64, N
15.18.
7-phenylpropargyl-1,3,9-trimethylxanthinium tetrafluoroborate (2c)
White solid, yield 92 % (612 mg), m.p. = 231 dec. °C.
1HꢀNMR (400 MHz, CD3CN, T=298K, ppm) δ: 3.37 (s, 3H, NꢀCH3), 3.76 (s, 3H, NꢀCH3),
4.14 (s, 3H, NꢀCH3), 5.57 (s, 2H, NꢀCH2), 7.42ꢀ7.60 (m, 5H, Ph), 8.85 (s, 1H, NCHN).
13C{1H}ꢀNMR (CD3CN, T=298K, ppm) δ: 28.8 (NꢀCH3), 31.9 (NꢀCH3), 37.9 (NꢀCH3), 40.7

Dalton Transactions
Page 20 of 37
View Article Online
DOI: 10.1039/C8DT01831F
(NꢀCH2), 79.2 (≡CꢀCH2), 89.4 (≡CꢀPh), 108.3 (C5), 121.6, 129.4, 130.4, 132.5 (CPh), 138.8
(NꢀCHꢀN), 140.6 (C4), 151.0 (C=O), 154.0 (C=O).
ESIꢀMS (CH3CN): m/z Calcd. for C17H17N4O2 [M]+ 309.13; found: 309.02.
Anal. Calcd. for C17H17BF4N4O2: C 51.54, H 4.33, N 14.14. Found: C 51.73, H 4.28, N
14.31.
1-benzyl-3,7,9-trimethylxanthinium tetrafluoroborate (2d)
White solid, yield 94 % (621 mg), m. p. = 171ꢀ172 °C.
1HꢀNMR (400 MHz, CD3CN, T=298K, ppm) δ: 3.74 (s, 3H, NꢀCH3), 4.08 (s, 3H, NꢀCH3),
4.09 (s, 3H, NꢀCH3), 5.17 (s, 2H, NꢀCH2), 7.28ꢀ7.42 (m, 5H, Ph), 8.48 (s, 1H, NCHN).
13C{1H}ꢀNMR (CD3CN, T=298K, ppm) δ: 31.9 (NꢀCH3), 36.4 (NꢀCH3), 37.6 (NꢀCH3), 45.6
(NꢀCH2), 109.2 (C5), 128.3, 128.6, 129.1, 137.1 (CPh), 139.4 (NꢀCHꢀN), 140.5 (C4), 151.0
(C=O), 154.1 (C=O).
ESIꢀMS (CH3CN): m/z Calcd. for C15H17N4O2 [M]+ 285.13; found: 285.11.
Anal. Calcd. for C15H17BF4N4O2: C 48.41, H 4.60, N 15.06. Found: C 48.65, H 4.72, N
15.14.
Preparation of a 1:1 mixture of 3c and AgBF4
0.1130 g (0.2852 mmol) of the imidazolium salt 2c was dissolved in 30 mL of anhydrous
CH3CN into a 100 mL two necked flask and 0.09364 g (0.1571 mmol) of Ag2O was added
under inert atmosphere (Ar).
The mixture was stirred for 24 h at R.T. in the dark. The solution was filtered on a millipore
membrane filter in order to remove the Ag2O in excess. The resulting clear solution was
concentrated under vacuum and the title complex precipitated by addition of diethylether.
The white complex was separated by filtration and repeatedly washed with diethylether and
nꢀpentane and finally dried under vacuum.
0.1305 g (yield 91%) of 1:1 mixture of 3c and AgBF4 was obtained.
1HꢀNMR (400 MHz, CD3CN, T=298K, ppm) δ: 3.35 (s, 6H, 2NCH3), 3.70 (s, 6H, 2NCH3),
4.03 (s, 6H, 2NCH3), 5.40 (s, 4H, 2NCH2), 7.27ꢀ7.43 (m, 10H, 2Ph). 13C{1H}ꢀNMR
(CD3CN, T=298K, ppm) δ: 27.9 (CH3, NCH3), 31.4 (CH3, NCH3), 40.1 (CH2, NCH2), 40.2
(CH3, NCH3), 82.5 (C, CH2ꢀC≡), 88.7 (C, PhꢀC≡), 108.7 (C, C5), 121.3ꢀ131.8 (Ph), 140.4 (C,
C4), 150.7 (C, C=O), 153.5 (C, C=O), 187.1 (C, carbene).

Page 21 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
IR (KBr): νC≡C= 2218 cmꢀ1, νCO= 1710, 1668 cmꢀ1, νBF=1054 cmꢀ1.
Anal. Calcd. for C34H32Ag2B2F8N8O4: C 40.59, H 3.21, N 11.14. Found: C 40.82, H 3.04, N
11.35.
Synthesis of complex 4a
0.0195 g (0.053 mmol) of the dimer [Pd(ꢁꢀCl)(η3ꢀC3H5)]2 and 0.0279 g (0.106 mmol) of
PPh3 were dissolved in ca. 20 mL of anhydrous CH3CN into a 50 mL two necked flask under
inert atmosphere (Ar). The resulting mixture was treated with 0.0388 g (0.048 mmol) of
3a/AgBF4 and stirred at RT for ca. 15 min.
The precipitated AgCl was removed by filtration on a millipore membrane filter.
Addition of diethylether to the concentrated solution yielded the precipitation of the complex
4a as a yellow solid which was filtered off on a gooch and washed with nꢀpentane.
0.0576 g of 4a was obtained (yield 85%).
1HꢀNMR (400 MHz, T=298K, CDCl3, ppm) δ: 3.16 (d, 1H, J= 13.4 Hz, anti allylꢀH trans C),
3.26 (d, 1H, J= 13.7 Hz, anti allylꢀH trans C), 3.27 (m, 1H, anti allylꢀH trans P), 3.38 (s, 6H,
2NCH3), 3.59 (s, 6H, 2NCH3), 3.61 (s, 6H, 2NCH3), 3.68 (m, 1H, anti allylꢀH trans P), 3.76
(s, 3H, NCH3), 3.78 (s, 3H, NCH3), 4.11 (d, 1H, J= 8.4 Hz, syn allylꢀH trans C), 4.19 (d, 1H,
J= 7.6 Hz, syn allylꢀH trans C), 4.59 (dd, 1H, JHꢀH=JHꢀP= 5.9 Hz, syn allylꢀH trans P), 4.80
(dt, 1H, JHꢀH=JHꢀP= 6.6 Hz, syn allylꢀH trans P), 5.71 (m, 1H, central allylꢀH), 5.99 (m, 1H,
central allylꢀH), 7.25ꢀ7.53 (m, 30H, 6Ph). 13C{1H}ꢀNMR (T=298K, CDCl3, ppm) δ: 28.6
(CH3, NCH3), 31.7 (CH3, NCH3), 36.9 (CH3, NCH3), 37.1 (CH3, NCH3), 38.7 (CH3, NCH3),
38.9 (CH3, NCH3), 68.5 (d, CH2, JCꢀP= 28.5 Hz, allyl trans P), 69.3 (d, CH2, JCꢀP= 27.6 Hz,
allyl trans P), 69.4 (d, CH2, JCꢀP= 1.8 Hz, allyl trans C), 69.4 (d, CH2, JCꢀP= 1.7 Hz, allyl
trans C), 110.4 (C, C5), 110.6 (C, C5), 121.4 (d, CH, JCꢀP= 5.4 Hz, central allyl), 122.8 (d,
CH, JCꢀP= 5.1 Hz, central allyl), 129.2ꢀ133.2 (Ph), 141.0 (C, C4), 141.1 (C, C4), 150.4 (C,
C=O), 153.0 (C, C=O), 185.8 (d, C, JCꢀP= 19.7 Hz, carbene), 186.2 (d, C, JCꢀP= 19.3 Hz,
carbene). 31P{1H}ꢀNMR (T=298K, CDCl3, ppm) δ: 25.9, 25.8.
IR (KBr): νCO= 1709, 1668 cmꢀ1, νBF=1059 cmꢀ1.
Anal. Calcd. for C30H32BF4N4O2PPd: C 51.12, H 4.58, N 7.95. Found: C 51.42, H 4.32, N
8.10.

Dalton Transactions
Page 22 of 37
View Article Online
DOI: 10.1039/C8DT01831F
Synthesis of complex 4b
Complex 4b was prepared in an analogous manner to that described for 4a starting from
0.0144 g of [Pd(ꢁꢀCl)(η3ꢀallyl)]2, 0.0358 g of 3b/AgBF4 and 0.0196 g of PPh3.
0.0511 g (yield 88%) of 4b was obtained.
1HꢀNMR (400 MHz, T=298K, CDCl3, ppm) δ: 2.38 (m, 1H, anti allylꢀH trans P), 2.93 (d,
1H, J= 13.6 Hz, anti allylꢀH trans C), 3.17 (d, 1H, J= 13.4 Hz, anti allylꢀH trans C), 3.40 (s,
3H, NCH3), 3.42 (s, 3H, NCH3), 3.60 (m, 1H, anti allylꢀH trans P), 3.62 (s, 3H, NCH3), 3.63
(s, 3H, NCH3), 3.86 (s, 3H, NCH3), 3.96 (dd, 1H, JHꢀH = 6.8 Hz, JHꢀP = 6.8 Hz, syn allylꢀH
trans P), 4.09 (d, 1H, J= 6.4 Hz, syn allylꢀH trans C), 4.19 (d, 1H, J= 6.1 Hz, syn allylꢀH
trans C), 4.70 (dd, 1H, JHꢀH = 7.7 Hz, JHꢀP = 7.7 Hz, syn allylꢀH trans P), 4.80 and 5.57 (AB
system, 2H, J=14.8 Hz, NCH2), 5.03 and 5.63 (AB system, 2H, J=15.0 Hz, NCH2), 5.22 (m,
1H, central allylꢀH), 5.97 (m, 1H, central allylꢀH), 7.00ꢀ7.60 (m, 40H, 8Ph).
13C{1H}ꢀNMR (T=298K, CDCl3, ppm) δ: 28.7 (CH3, NCH3), 28.8 (CH3, NCH3), 31.9 (CH3,
NCH3), 39.3 (CH3, NCH3), 39.5 (CH3, NCH3), 52.4 (CH2, NCH2), 52.7 (CH2, NCH2), 68.6
(d, CH2, JCꢀP= 1.9 Hz, allyl transꢀC), 68.7 (d, CH2, JCꢀP= 1.8 Hz, allyl transꢀC), 70.1 (d, CH2,
JCꢀP= 27.7 Hz, allyl transꢀP), 70.8 (d, CH2, JCꢀP= 27.8 Hz, allyl transꢀP), 110.1 (C, C5), 110.3
(C, C5), 121.0 (CH, central allyl), 122.8 (CH, central allyl), 127.2ꢀ135.3 (Ph), 141.1 (C4),
141.3 (C4), 143.7 (Ph), 150.3 (C, C=O), 150.4 (C, C=O), 152.8 (C, C=O), 152.9 (C, C=O),
187.5 (d, C, JCꢀP= 18.5 Hz, carbene), 187.8 (d, C, JCꢀP= 18.4 Hz, carbene). 31P{1H}ꢀNMR
(T=298K, CDCl3, ppm) δ: 25.6, 26.3
IR (KBr): νCO= 1709, 1668 cmꢀ1, νBF=1056 cmꢀ1
Anal. Calcd. for C36H36BF4N4O2PPd: C 55.37, H 4.65, N 7.17. Found: C 55.12, H 4.80, N
7.38.
Synthesis of complex 4c
Complex 4c was prepared in an analogous manner to that described for 4a starting from
0.0160 g of [Pd(ꢁꢀCl)(η3ꢀallyl)]2, 0.0418 g of 3c/AgBF4 and 0.0218 g of PPh3.
0.0559 g (yield 84%) of 4c was obtained.
1HꢀNMR (400 MHz, T=298K, CDCl3, ppm) δ: 2.97 (d, 1H, J= 13.3 Hz, anti allylꢀH trans C),
3.20 (d, 1H, J= 12.8 Hz, anti allylꢀH trans C), 3.36 (m, 1H, anti allylꢀH trans P), 3.42 (s, 6H,
2NCH3), 3.61 (s, 3H, NCH3), 3.62 (s, 3H, NCH3), 3.68 (s, 3H, NCH3), 3.76 (m, 1H, anti

Page 23 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
allylꢀH trans P), 3.82 (s, 3H, NCH3), 4.04 (d, 1H, J= 6.1 Hz, syn allylꢀH trans C), 4.22 (d,
1H, J= 6.4 Hz, syn allylꢀH trans C), 4.71 (dd, 1H, JHꢀH = 6.0 Hz, JHꢀP = 6.8 Hz,, syn allylꢀH
trans P), 4.82 and 5.28 (AB system, 2H, J=17.2 Hz, NCH2), 4.88 and 5.47 (AB system, 2H,
J=17.6 Hz, NCH2), 4.89 (m, 1H, syn allylꢀH trans P), 5.63 (m, 1H, central allylꢀH), 6.03 (m,
2H, central allylꢀH), 7.15ꢀ7.46 (m, 40H, 8Ph). 13C{1H}ꢀNMR (T=298K, CDCl3, ppm) δ:
28.7 (CH3, NCH3), 31.8 (CH3, NCH3), 39.4 (CH3, NCH3), 40.0 (CH2, NCH2), 68.9 (CH2,
allyl transꢀC), 69.3 (CH2, allyl transꢀC), 69.9 (d, CH2, JCꢀP= 27.5 Hz, allyl transꢀP), 70.4 (d,
CH2, JCꢀP= 27.6 Hz, allyl transꢀP), 81.6 (C, ≡CꢀCH2), 81.9 (C, ≡CꢀCH2), 86.5 (C, ≡CꢀPh),
87.4 (C, ≡CꢀPh), 109.6 (C, C5), 109.7 (C, C5), 121.4 (d, CH, JCꢀP= 5.1 Hz, central allyl),
122.7 (d, CH, JCꢀP= 5.2 Hz, central allyl), 128.4ꢀ134.0 (Ph), 141.2 (C, C4), 150.4 (C, C=O),
152.6 (C, C=O), 152.7 (C, C=O), 187.3 (d, C, JCꢀP= 18.8 Hz, carbene), 187.7 (d, C, JCꢀP
18.5 Hz, carbene). 31P{1H}ꢀNMR (T=298K, CDCl3, ppm) δ: 25.4, 25.9.
IR (KBr): νCO= 1709, 1667 cmꢀ1, νBF=1058 cmꢀ1
=
Anal. Calcd. for C38H36BF4N4O2PPd: C 56.70, H 4.51, N 6.96. Found: C 56.52, H 4.78, N
7.12.
Synthesis of the complex 4d
Complex 4d was prepared in an analogous manner to that described for 4a starting from
0.0154 g of [Pd(ꢁꢀCl)(η3ꢀallyl)]2, 0.0383 g of 3d/AgBF4 and 0.0210 g of PPh3.
0.0522 g (yield 84%) of 4d was obtained.
1HꢀNMR (400 MHz, T=298K, CDCl3, ppm) δ: 3.13 (d, 1H, J= 13.4 Hz, anti allylꢀH trans C),
3.28 (d+m, 2H, J= 13.2 Hz, anti allylꢀH trans C), 3.29 (m, 1H, anti allylꢀH trans P), 3.58 (s,
3H, NCH3), 3.59 (s, 3H, NCH3), 3.60 (s, 3H, NCH3), 3.62 (s, 3H, NCH3), 3.68 (m, 1H, anti
allylꢀH trans P), 3.74 (s, 3H, NCH3), 3.81 (s, 3H, NCH3), 4.13 (d, 1H, J= 7.4 Hz, syn allylꢀH
trans C), 4.21 (d, 1H, J= 7.8 Hz, syn allylꢀH trans C), 4.58 (dd, 1H, JHꢀH=JHꢀP= 5.2 Hz, syn
allylꢀH trans P), 4.79 (dd, 1H, JHꢀH = 5.5 Hz, JHꢀP = 5.5 Hz, syn allylꢀH trans P), 5.15 (s, 4H,
2 NCH2), 5.70 (m, 1H, central allylꢀH), 5.99 (m, 1H, central allylꢀH), 7.25ꢀ7.47 (m, 40H,
13
8Ph). C{1H}ꢀNMR (T=298K, CDCl3, ppm) δ: 31.8 (CH3, NCH3), 36.9 (CH3, NCH3), 37.0
(CH3, NCH3), 38.7 (CH3, NCH3), 38.9 (CH3, NCH3), 45.1 (CH2, NCH2), 68.3 (d, CH2, JCꢀP
=
28.4 Hz, allyl transꢀP), 69.1 (d, CH2, JCꢀP= 28.1 Hz, allyl transꢀP), 69.2 (d, CH2, JCꢀP= 1.7
Hz, allyl transꢀC), 69.4 (d, CH2, JCꢀP= 1.9 Hz, allyl transꢀC), 110.7 (C, C5), 121.3 (d, CH, JCꢀ

Dalton Transactions
Page 24 of 37
View Article Online
DOI: 10.1039/C8DT01831F
P= 5.5 Hz, central allyl), 122.8 (d, CH, JCꢀP= 5.3 Hz, central allyl), 127.9ꢀ136.4 (Ph), 141.0
(C, C4), 141.1 (C, C4), 150.2 (C, C=O), 152.7 (C, C=O), 186.1 (d, C, JCꢀP= 19.3 Hz, carbene),
186.5 (d, C, JCꢀP= 19.4 Hz, carbene). 31P{1H}ꢀNMR (T=298K, CDCl3, ppm) δ: 25.9, 26.0.
IR (KBr): νCO= 1707, 1668 cmꢀ1, νBF=1056 cmꢀ1.
Anal. Calcd. for C36H36BF4N4O2PPd: C 55.37, H 4.65, N 7.17. Found: C 55.22, H 4.58, N
7.42.
Synthesis of complex 5a
0.0411 g (0.112 mmol) of the dimer [Pd(ꢁꢀCl)(η3ꢀC3H5)]2 and 0.0353 g (0.224 mmol) of
PTA (1,3,5ꢀtriazaꢀ7ꢀphosphadamantane) were dissolved in ca. 30 mL of anhydrous CH3CN
in a 100 mL two necked flask under inert atmosphere (Ar). The resulting mixture was treated
with 0.0787 g (0.0975 mmol) of 3a/AgBF4 and stirred at RT for ca. 1 hour.
The precipitated AgCl was removed by filtration on a millipore membrane filter.
The solution was dried under vacuum and the residue treated with 2 mL of CH2Cl2.
Addition of diethylether to the concentrated solution yielded the precipitation of the complex
5a as a brownish solid which was filtered off on a gooch, washed with nꢀpentane and dried
under vacuum.
0.0991 g of 5a was obtained (yield 97%).
1HꢀNMR (400 MHz, T=298K, D2O, ppm) δ: 2.88 (d, 2H, J= 13.2 Hz, 2 anti allylꢀH transꢀC),
3.15 (m, 2H, 2 anti allylꢀH transꢀP), 3.32 (s, 3H, NCH3), 3.33 (s, 3H, NCH3), 3.74 (s, 3H,
NCH3), 3.77 (s, 3H, NCH3), 3.80 (s, 3H, NCH3), 3.87 (s, 3H, NCH3), 3.92 (s, 3H, NCH3),
4.05 (s, 3H, NCH3), 4.21 (s, 12H, 6 NCH2PPTA), 4.38 (d, 2H, J= 7.4 Hz, 2 syn allylꢀH transꢀ
C), 4.47 (m, 2H, 2 syn allylꢀH transꢀP), 4.55 (m, 12H, 6 NCH2NPTA), 5.48 (m, 2H, 2 central
allylꢀH). 13C{1H}ꢀNMR (T=298K, D2O, ppm) δ: 28.4 (CH3, NCH3), 31.8 (CH3, NCH3), 36.7
(CH3, NCH3), 36.9 (CH3, NCH3), 38.2 (CH3, NCH3), 38.4 (CH3, NCH3), 50.3 (CH2, NCH2P,
JCꢀP =13.6 Hz), 62.7 (CH2, allyl transꢀC), 62.8 (CH2, allyl transꢀC), 68.9 (d, CH2, JCꢀP= 4.8
Hz, allyl transꢀP), 69.2 (d, CH2, JCꢀP= 5.2 Hz, allyl transꢀP), 70.7 (CH2, NCH2N), 70.8 (CH2,
NCH2N), 111.2 (C, C5), 111.4 (C, C5), 121.9 (CH, central allyl), 122.0 (CH, central allyl),
141.8 (C, C4), 141.9 (C, C4), 151.8 (C, C=O), 154.5 (C, C=O), 183.9 (C, JCꢀP= 21.2 Hz,
carbene). 31P{1H}ꢀNMR (T=298K, D2O, ppm) δ: ꢀ52.4
IR (KBr): νCO= 1704, 1665 cmꢀ1, νBF=1031, 1084 cmꢀ1.

Page 25 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
Anal. Calcd. for C18H29BF4N7O2PPd: C 36.05, H 4.87, N 16.35. Found: C 36.34, H 4.97, N
16.14.
Synthesis of complex 5b
Complex 5b was prepared in an analogous manner to that described for 5a starting from
0.0353 g of [Pd(ꢁꢀCl)(η3ꢀallyl)]2, 0.0804 g of 3b/AgBF4 and 0.0304 g of PTA.
0.0925 g (yield 82%) of 5b was obtained.
1HꢀNMR (400 MHz, T=298K, D2O, ppm) δ: 2.76 (m, 2H, 2 anti allylꢀH transꢀC), 3.06 (m,
2H, 2 anti allylꢀH transꢀP), 3.28 (s, 3H, NCH3), 3.29 (s, 3H, NCH3), 3.72 (s, 6H, 3
NCH2PPTA), 3.75 (s, 6H, 3 NCH2PPTA), 3.77 (s, 3H, NCH3), 3.80 (s, 3H, NCH3), 3.88 (s, 3H,
NCH3), 4.06 (s, 3H, NCH3), 4.27ꢀ4.70 (m, 16 H, 6 NCH2NPTA and 2 syn allylꢀH transꢀC and
2 syn allylꢀH transꢀP), 5.07 and 5.64 (AB system, 2H, J=15.4 Hz, NCH2), 5.29 and 5.76 (AB
system, 2H, J=15.7 Hz, NCH2), 5.36 (m, 2H, 2 central allylꢀH), 6.81ꢀ7.36 (m, 10H, 2Ph).
13C{1H}ꢀNMR (T=298K, D2O, ppm) δ: 28.5 (CH3, NCH3), 31.9 (CH3, NCH3), 38.6 (CH3,
NCH3), 38.9 (CH3, NCH3), 49.9 (d, CH2, J= 13.5 Hz, NCH2P), 50.0 (d, CH2, J= 13.6 Hz,
NCH2P), 52.5 (CH2, NCH2), 52.7 (CH2, NCH2), 62.6 (CH2, allyl transꢀC), 63.0 (CH2, allyl
transꢀC), 69.4 (d, CH2, JCꢀP= 26.7 Hz, allyl transꢀP), 69.9 (d, CH2, JCꢀP= 27.8 Hz, allyl transꢀ
P), 70.4 (CH2, NCH2NPTA) 70.5 (CH2, NCH2N), 111.1 (C, C5), 111.4 (C, C5), 121.9 (CH,
central allyl), 126.9ꢀ136.7 (Ph), 141.8 (C, C4), 141.9 (C, C4), 151.8 (C, C=O), 154.2 (C,
C=O), 186.0 (d, C, JCꢀP= 20.4, carbene), 186.1 (d, C, JCꢀP= 20.5 Hz, carbene). 31P{1H}ꢀNMR
(T=298K, D2O, ppm) δ: ꢀ54.1, ꢀ54.0.
IR (KBr): νCO= 1705, 1664 cmꢀ1, νBF=1035, 1083 cmꢀ1
Anal. Calcd. for C24H33BF4N7O2PPd: C 42.66, H 4.92, N 14.51. Found: C 42.82, H 4.78, N
14.22.
Synthesis of complex 5d
Complex 5d was prepared in an analogous manner to that described for 5a starting from
0.0352 g of [Pd(ꢁꢀCl)(η3ꢀallyl)]2, 0.0800 g of 3d/AgBF4 and 0.0301 g of PTA.
0.1057 g (yield 94%) of 5d was obtained.
1HꢀNMR (400 MHz, T=298K, D2O, ppm) δ: 2.83 (d, J= 13.8 Hz, 2H, 2 anti allylꢀH transꢀC),
3.07 (m, 2H, 2 anti allylꢀH transꢀP), 3.69 (s, 3H, NCH3), 3.70 (s, 3H, NCH3), 3.73 (s, 3H,

Dalton Transactions
Page 26 of 37
View Article Online
DOI: 10.1039/C8DT01831F
NCH3), 3.82 (s, 3H, NCH3), 3.87 (s, 3H, NCH3), 4.01 (s, 3H, NCH3), 4.15 (s, 6H, 3
NCH2PPTA), 4.27 (s, 6H, 3 NCH2PPTA), 4.32 (d, J = 6.6 Hz, 2H, 2 syn allylꢀH transꢀC), 4.45
(m, 2H, 2 syn allylꢀH transꢀP), 4.50ꢀ4.70 (m, 12H, 6 NCH2PPTA), 5.10 (2s, 4H, 2 NꢀCH2),
5.42 (m, 2H, 2 central allylꢀH), 7.25ꢀ7.30 (m, 10H, 2Ph).
13C{1H}ꢀNMR (T=298K, D2O, ppm) δ: 31.9 (CH3, NCH3), 36.8 (CH3, NCH3), 37.1 (CH3,
NCH3), 38.2 (CH3, NCH3), 38.5 (CH3, NCH3), 45.2 (CH2, NCH2), 50.3 (d, CH2, JCꢀP= 13.8
Hz, NCH2P), 51.1 (d, CH2, JCꢀP= 15.4 Hz, NCH2P), 62.8 (CH2, allyl transꢀC), 62.9 (CH2,
allyl transꢀC), 69.2 (d, CH2, JCꢀP= 27.7 Hz, allyl transꢀP), 70.8 (CH2, NCH2N), 71.3 (d, CH2,
JCꢀP= 26.2 Hz, allyl transꢀP), 111.4 (C5), 122.0 (CH, central allyl), 123.0 (CH, central allyl),
127.2ꢀ136.2 (Ph), 142.3 (C, C4), 151.6 (C, C=O), 154.2 (C, C=O), 184.2 (d, C, JCꢀP= 19.0 Hz,
carbene), 184.3 (d, C, JCꢀP= 19.8 Hz, carbene). 31P{1H}ꢀNMR (T=298K, D2O, ppm) δ: ꢀ54.2,
ꢀ53.1.
IR (KBr): νCO= 1708, 1668 cmꢀ1, νBF=1062, 1084 cmꢀ1
Anal. Calcd. for C24H33BF4N7O2PPd: C 42.66, H 4.92, N 14.51. Found: C 42.52, H 4.99, N
14.32.
Synthesis of complex 6a
0.0173 g (0.047 mmol) of the dimer [Pd(ꢁꢀCl)(η3ꢀC3H5)]2 and 0.0124 g (0.094 mmol) of
DIC (2,6ꢀdimetilphenyl isocyanide) were dissolved in ca. 20 mL of anhydrous CH3CN in a
50 mL two necked flask under inert atmosphere (Ar). The resulting mixture was treated with
0.0344 g (0.043 mmol) of 3a/AgBF4 and stirred at RT for ca. 15 min.
The precipitated AgCl was removed by filtration on a millipore membrane filter.
Addition of diethylether to the concentrated solution yielded the precipitation of the complex
6a as a brownish solid which was filtered off on a gooch and washed with nꢀpentane.
0.0431 g of 6a was obtained (yield 88%).
1HꢀNMR (400 MHz, T=298K, CDCl3, ppm) δ: 2.41 (s, 6H, 2CH3DIC), 3.20 (d, 1H, J= 13.3
Hz, anti allylꢀH), 3.40 (s, 3H, NCH3), 3.41 (d, 1H, J= 12.7 Hz, anti allylꢀH), 3.88 (s, 3H,
NCH3), 4.05 (s, 3H, NCH3), 4.20 (s, 3H, NCH3), 4.47 (d, 1H, J= 6.4 Hz, syn allylꢀH), 4.79
(d, 1H, J= 7.5 Hz, syn allylꢀH), 5.72 (m, 1H, central allylꢀH), 7.17ꢀ7.31 (m, 3H, PhDIC).
13C{1H}ꢀNMR (T=298K, CDCl3, ppm) δ: 18.7 (CH3, CH3DIC), 28.6 (CH3, NCH3), 32.0
(CH3, NCH3), 37.5 (CH3, NCH3), 39.3 (CH3, NCH3), 65.0 (CH2, allylꢀC), 69.3 (CH2, allylꢀ

Page 27 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
C), 110.8 (C, C5), 122.2 (central allyl), 128.4ꢀ135.6 (Ph), 141.4 (C, C4), 150.6 (C, C=O),
150.7 (C, CNDIC), 153.4 (C, C=O), 181.5 (C, carbene).
IR (KBr): νCN=2175 cmꢀ1, νCO= 1706, 1665 cmꢀ1, νBF=1056 cmꢀ1.
Anal. Calcd. for C21H26BF4N5O2Pd: C 43.97, H 4.57, N 12.21. Found: C 44.24, H 4.77, N
12.04.
Synthesis of complex 6b
Complex 6b was prepared in an analogous manner to that described for 6a starting from
0.0180 g of [Pd(ꢁꢀCl)(η3ꢀallyl)]2, 0.0428 g of 3b/AgBF4 and 0.128 g of DIC.
0.0491 g (yield 85%) of 6b was obtained.
1HꢀNMR (400 MHz, T=298K, CDCl3, ppm) δ: 2.36 (s, 6H, 2CH3DIC), 3.42 (s, 3H, NCH3),
3.48 (bd, 2H, 2anti allylꢀH), 3.91 (s, 3H, NCH3), 4.24 (bd, 1H, syn allylꢀH), 4.25 (s, 3H,
NCH3), 4.67 (d, 1H, J= 7.5 Hz, syn allylꢀH), 5.65 (s, 2H, NCH2), 5.73 (m, 1H, central allylꢀ
13
H), 7.16ꢀ7.33 (m, 8H, Ph and PhDIC). C{1H}ꢀNMR (T=298K, CDCl3, ppm) δ: 18.8 (CH3,
CH3DIC), 28.7 (CH3, NCH3), 32.1 (CH3, NCH3), 39.9 (CH3, NCH3), 53.0 (CH2, NCH2), 65.5
(CH2, allylꢀC), 110.6 (C, C5), 121.9 (CH, central allyl), 127.5ꢀ135.9 (Ph), 141.5 (C, C4),
150.5 (C, C=O), 150.7 (C, CNDIC), 153.2 (C, C=O), 183.5 (C, carbene).
IR (KBr): νCN=2175 cmꢀ1, νCO= 1709, 1670 cmꢀ1, νBF=1057 cmꢀ1.
Anal. Calcd. for C27H30BF4N5O2Pd: C 49.91, H 4.65, N 10.78. Found: C 49.84, H 4.83, N
10.90.
Synthesis of complex 6d
Complex 6d was prepared in an analogous manner to that described for 6a starting from
0.0184 g of [Pd(ꢁꢀCl)(η3ꢀallyl)]2, 0.0437 g of 3d/AgBF4 and 0.0131 g of DIC.
0.0543 g (yield 92%) of 6d was obtained.
1HꢀNMR (400 MHz, T=298K, CDCl3, ppm) δ: 2.41 (s, 6H, 2CH3DIC), 3.19 (d, 1H, J= 13.5
Hz, anti allylꢀH), 3.41 (d, 1H, J= 12.5 Hz, anti allylꢀH), 3.86 (s, 3H, NCH3), 4.05 (s, 3H,
NCH3), 4.19 (s, 3H, NCH3), 4.47 (d, 1H, J= 7.1 Hz, syn allylꢀH), 4.78 (d, 1H, J= 7.5 Hz, syn
allylꢀH), 5.18 (s, 2H, NCH2), 5.72 (m, 1H, central allylꢀH), 7.17ꢀ7.49 (m, 8H, Ph and PhDIC).
13C{1H}ꢀNMR (T=298K, CDCl3, ppm) δ: 18.8 (CH3, CH3DIC), 32.0 (CH3, NCH3), 37.5
(CH3, NCH3), 39.3 (CH3, NCH3), 45.2 (CH2, NCH2), 64.9 (CH2, allylꢀC), 110.9 (C, C5),

Dalton Transactions
Page 28 of 37
View Article Online
DOI: 10.1039/C8DT01831F
122.2 (CH, central allyl), 127.8ꢀ136.3 (Ph), 141.5 (C, C4), 150.5 (C, C=O), 150.6 (C,
CNDIC), 153.2 (C, C=O), 181.8 (C, carbene).
IR (KBr): νCN=2173 cmꢀ1, νCO= 1707, 1665 cmꢀ1, νBF=1056 cmꢀ1.
Anal. Calcd. for C27H30BF4N5O2Pd: C 49.91, H 4.65, N 10.78. Found: C 50.04, H 4.51, N
10.95.
Synthesis of complex 8a
0.0156 g (0.086 mmol) of the dimer [Pd(ꢁꢀCl)(η3ꢀC3H5)]2 was dissolved in ca. 20 mL of
anhydrous CH3CN in a 50 mL two necked flask under inert atmosphere (Ar).
The resulting mixture was treated with 0.0690 g (0.171 mmol) of 3a/AgBF4, 0.0142 g (0.086
mmol) of KI and stirred at RT for ca. 15 min. The precipitated AgCl and AgI were removed
by filtration on a millipore membrane filter. Addition of diethylether to the concentrated
solution yielded the precipitation of the complex 8a as a white solid which was filtered off
on a gooch and washed with nꢀpentane.
0.0446 g of 8a was obtained (yield 80%).
1HꢀNMR (400 MHz, T=298K, CD3CN, ppm) δ: 2.95 (d, 2H, J= 13.3 Hz, anti allylꢀH), 3.30
(s, 6H, NCH3), 3.73 (s, 6H, NCH3), 3.93 (s, 6H, NCH3), 4.02 (s, 6H, NCH3), 4.12 (d, 2H, J=
13
7.4 Hz, syn allylꢀH), 5.59 (m, 1H, central allylꢀH). C{1H}ꢀNMR (T=298K, CD3CN, ppm)
δ: 28.4 (CH3, NCH3), 32.0 (CH3, NCH3), 37.7 (CH3, NCH3), 39.4 (CH3, NCH3), 61.6 (CH2,
allylꢀC), 110.9 (C, C5), 121.0 (CH, central allyl), 141.9 (C, C4), 151.4 (C, C=O), 153.9 (C,
C=O), 184.9 (C, carbene).
IR (KBr): νCO= 1706, 1668 cmꢀ1, νBF=1053 cmꢀ1.
Anal. Calcd. for C21H29BF4N8O4Pd: C 38.76, H 4.49, N 17.22. Found: C 38.50, H 4.82, N
17.35.
Synthesis of complex 8b
Complex 8b was prepared in an analogous manner to that described for 8a starting from
0.0157 g of [Pd(ꢁꢀCl)(η3ꢀallyl)]2, 0.0822 g of 3b/AgBF4 and 0.0142 g of KI.
0.0599 g (yield 87%) of 8b was obtained.

Page 29 of 37
Dalton Transactions
View Article Online
DOI: 10.1039/C8DT01831F
1HꢀNMR (400 MHz, T=298K, CD3CN, ppm) δ: 2.79 (d, 2H, J= 13.2 Hz, anti allylꢀH), 3.23
(s, 6H, NCH3), 3.68 (s, 6H, NCH3), 3.97 (s, 6H, NCH3), 3.97 (bd, 2H, syn allylꢀH), 5.39ꢀ5.61
(m, 5H, NCH2 and central allylꢀH), 6.94ꢀ7.24 (m, 10H, 2Ph).
13C{1H}ꢀNMR (T=298K, CD3CN, ppm) δ: 28.3 (CH3, NCH3), 31.9 (CH3, NCH3), 39.6
(CH3, NCH3), 53.0 (CH2, NCH2), 65.9 (CH2, allylꢀC), 110.4 (C, C5), 121.2 (CH, central
allyl), 125.7ꢀ137.2 (Ph), 142.0 (C, C4), 151.2 (C, C=O), 153.2 (C, C=O), 185.3 (C, carbene).
IR (KBr): νCO= 1709, 1664 cmꢀ1, νBF=1058 cmꢀ1.
Anal. Calcd. for C33H37BF4N8O4Pd: C 49.36, H 4.64, N 13.96. Found: C 49.57, H 4.12, N
14.22.
Synthesis of complex 8c
Complex 8c was prepared in an analogous manner to that described for 8a starting from
0.0157 g of [Pd(ꢁꢀCl)(η3ꢀallyl)]2, 0.0863 g of 3c/AgBF4 and 0.0142 g of KI.
0.0592g (yield 81%) of 8c was obtained.
1HꢀNMR (400 MHz, T=298K, CD3CN, ppm) δ: 3.02 (s, 6H, NCH3), 3.03 (d, 2H, J= 13.4 Hz,
anti allylꢀH), 3.53 (s, 6H, NCH3), 4.20 (m, 8H, NCH3 and syn allylꢀH), 5.44 (bs, 4H, NCH2),
5.66 (m, 1H, central allylꢀH), 7.16ꢀ7.48 (m, 10H, 2Ph). 13C{1H}ꢀNMR (T=298K, CD3CN,
ppm) δ: 28.4 (CH3, NCH3), 31.9 (CH3, NCH3), 39.8 (CH3, NCH3), 41.1 (CH2, NCH2), 65.9
(CH2, allylꢀC), 83.8 (CH2ꢀC≡), 84.7 (C, PhꢀC≡), 110.4 (C, C5), 121.6 (CH, central allyl),
121.9ꢀ131.0 (Ph), 141.7 (C, C4), 150.5 (C, C=O), 153.4 (C, C=O), 185.5 (C, carbene).
IR (KBr): νCO= 1706, 1667 cmꢀ1, νBF=1056 cmꢀ1.
Anal. Calcd. for C37H37BF4N8O4Pd: C 52.22, H 4.38, N 13.17. Found: C 52.37, H 4.10, N
13.55.
Synthesis of the complex 8d
Complex 8d was prepared in an analogous manner to that described for 8a starting from
0.0154 g of [Pd(ꢁꢀCl)(η3ꢀallyl)]2, 0.0806 g of 3d/AgBF4 and 0.0139 g of KI.
0.0595 g (yield 88%) of 8d was obtained.
1HꢀNMR (400 MHz, T=298K, CD3CN, ppm) δ: 2.96 (d, 2H, J= 13.3 Hz, anti allylꢀH), 3.73
(s, 6H, NCH3), 3.93 (s, 6H, NCH3), 4.02 (s, 6H, NCH3), 4.13 (d, 2H, J= 7.4 Hz, 2 syn allylꢀ
H), 5.13 (s, 4H, NCH2), 5.59 (m, 1H, central allylꢀH), 7.28ꢀ7.39 (m, 10H, 2Ph).