Unexpected isomerism in "[Pd(2,9-dimethylphenanthroline)X 2]" (X = Cl, Br, I) complexes: A neutral and an ionic form exist

Article abstract of DOI:10.1039/c2dt11979j

Complexes of composition "[Pd(2,9-dimethylphenanthroline)X ₂]" (X = Cl, Br, I) have long been known and they are used as precursors for the synthesis of other derivatives or as catalysts. In the previous literature, they have invariably been described as neutral square planar complexes, but we have found that a second ionic isomer also exists, having composition [Pd(Neoc)₂X]₂[Pd₂X ₆], and that the formation of this isomer occurs under a wider range of conditions than that of the neutral one. Retrospectively, the ionic isomer had surely been obtained in most previous reports even if formation of the neutral one was claimed. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2dt11979j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 3648
www.rsc.org/dalton
PAPER
Unexpected isomerism in “[Pd(2,9-dimethylphenanthroline)X2]”
(X = Cl, Br, I) complexes: a neutral and an ionic form exist†
Martino Rimoldi,a Fabio Ragaini,*a Emma Gallo,a Francesco Ferretti,a Piero Macchib and Nicola Casati‡a,c
Received 19th October 2011, Accepted 5th January 2012
DOI: 10.1039/c2dt11979j
Complexes of composition “[Pd(2,9-dimethylphenanthroline)X2]” (X = Cl, Br, I) have long been known
and they are used as precursors for the synthesis of other derivatives or as catalysts. In the previous
literature, they have invariably been described as neutral square planar complexes, but we have found that
a second ionic isomer also exists, having composition [Pd(Neoc)2X]2[Pd2X6], and that the formation of
this isomer occurs under a wider range of conditions than that of the neutral one. Retrospectively, the
ionic isomer had surely been obtained in most previous reports even if formation of the neutral one was
claimed.
generated under catalytic conditions. Different synthetic
approaches have been reported for their preparation, but the pro-
Introduction
Phenanthroline and substituted phenanthroline complexes have
played a key role in the development of coordination1 and orga-
nometallic chemistry and are actually finding increasing appli-
cations in homogeneous catalysis.2–5 Among the members of
this class of ligands, 2,9-dimethyl-1,10-phenanthroline (neo-
cuproine, Neoc), its derivative 2,9-dimethyl-4,7-diphenyl-1,10-
phenanthroline (batocuproine), and the related 6,6′-dimethyl-
2,2′-bipyridine have attracted special attention because the steric
hindrance caused by the two methyl groups stabilises unusual
coordination geometries and stabilises the formation of otherwise
unstable adducts.6–9 As a consequence, they have been shown to
lead to higher catalytic activities with respect to unhindered phe-
nanthrolines when employed as ligands for several palladium
catalysed reactions, including the Heck reaction,10 the oxidative
Heck arylation of olefins by boronic acids,11–14 the production
of hydrogen peroxide,15,16 the oxidative carbonylation of
phenols,17–19 and especially the alcohol oxidation by
dioxygen.20–31
ducts are invariably formulated as [Pd(Neoc)X2] (X = Cl, Br, I).
We have recently become interested in the use of palladium neo-
cuproine complexes in the context of a mechanistic study of the
palladium–phenanthroline catalysed carbonylation reaction of
nitroarenes to carbamates.33,34 While further investigating the
reactivity of [Pd(Neoc)Cl2], we realised that even minor vari-
ations in the synthetic procedure led to products that displayed
completely different solubility behaviour despite having the
same elemental analysis, always in agreement with that calcu-
lated for the expected product.
A closer inspection of the literature shows that some inconsis-
tencies in the reported solubility of [Pd(Neoc)X2] (X = Cl, Br, I)
complexes are indeed present, which had been apparently over-
looked by many scientists, including ourselves. Compounds of
general formula [Pd(Neoc)X2] were reported for the first time in
the literature in 1971,32 obtained by stirring a suspension of [Pd
(Neoc)(NO3)2] in acetone–water (X = Cl, Br) or in ethanol
(X = I) with a “large excess of alkali halide” for several hours
(no more detail was given in the paper). The precipitated com-
pounds were then recrystallised from chloroform, indicating that
they are at least partially soluble in this solvent (Scheme 1).
The compounds thus synthesised were characterised by
elemental analysis and UV spectroscopy. In a following paper,35
Palladium neocuproine halide complexes have been known
for 40 years32 and are often used as the starting material for the
synthesis of more complex compounds or are proposed to be
aUniversità degli Studi di Milano, Dipartimento di Chimica Inorganica,
Metallorganica e Analitica “Lamberto Malatesta” and ISTM-CNR, Via
Venezian 21, 20133 Milano, Italy. E-mail: fabio.ragaini@unimi.it;
Fax: + (39)0250314405; Tel: + (39)0250314405
1
compounds obtained in the same way were characterised by H
NMR (CDCl3) spectroscopy, again implying they are soluble
enough in chloroform.
bDepartment of Chemistry and Biochemistry, University of Bern
Freiestrasse 3, CH3012 Bern, CH
cUniversità degli Studi di Milano, Dipartimento di Chimica Strutturale e
Stereochimica Inorganica, via Venezian 21, 20133 Milano, Italy
†Electronic supplementary information (ESI) available: 1H NMR
spectra of 1a, 2a, and 3a. CCDC reference numbers 843108–843112.
For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c2dt11979j
‡Present address: Diamond Light Source Ltd, Harwell Science and
Innovation Campus, Chilton, OX110DE, UK.
Scheme 1
3648 | Dalton Trans., 2012, 41, 3648–3658
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 2
Scheme 4
solvents. However, elemental analyses of both 1a and 1b per-
fectly matched the expected composition. If the reaction was per-
formed in 1,2-dichloroethane, the outcome depended on the
temperature. If the reaction was run at reflux, only 1a was
obtained, whereas running it at RT resulted in a mixture of the
two. Refluxing the so obtained mixture in C2H4Cl2 for several
hours resulted in the complete conversion of 1b into 1a. The
same conversion also occurred in CH2Cl2 at RT, but it took
several months to reach completion (Scheme 4). As mentioned
in the introduction, Newkome reported the product of the reac-
tion of [Pd(CH3CN)2Cl2] with neocuproine in CH3CN to be
insoluble in standard NMR solvents.36,37 However, since the
reported yield was not quantitative, it may be doubted whether
some “soluble” product was also formed and lost during the
workup. We thus repeated this reaction and only insoluble 1b
was obtained (Scheme 4).
Scheme 3
Following works by Newkome36,37 reported a different syn-
thesis for [Pd(Neoc)Cl2] (Scheme 2).
To a stirred acetonitrile solution of neocuproine, PdCl2 dis-
solved in warm acetonitrile ([Pd(CH3CN)2Cl2] is generated
in situ) was added and the solution stirred for 2 h at room temp-
erature. A crystalline precipitate was formed. In the same paper it
was remarked that NMR characterisation was not possible
because of the low solubility of the product in standard NMR
solvents. This suggests that in fact the compound obtained was
not the same as that in the previously mentioned works.
These data strongly suggest that 1a and 1b are isomers and
that the former is the thermodynamically favoured one, although
formation of 1b may be faster under some conditions. Since no
geometric isomerism is possible for a square planar complex
with a chelating ligand, the two compounds should be structural
isomers.
The same series of reactions was performed starting from [Pd
(CH3CN)2Br2]. At variance from the chloride complex, in all
cases only an insoluble compound, 2b, was obtained, irrespec-
tive of the solvent and reaction temperature. In the case of the
reaction run at RT in CHCl3, the product contained less neocu-
proine than expected and was apparently contaminated by some
PdBr2. A small amount of a different soluble compound (see
later) remained in solution. However, the soluble isomer 2a was
cleanly obtained by exchange of 1a with tetrabutylammonium
bromide in chloroform, in agreement with Fanizzi.38
We also investigated another approach to the synthesis of 2,
that is the direct reaction of palladium acetate, neocuproine and
NaBr in methanol. This represents a slight modification of the
procedure already used by Aresta et al. to prepare [Pd(Phen)
I2].42 The procedure resulted in the formation of 2b only if the
reaction was run at RT, whereas formation of metallic palladium
was observed in refluxing methanol.
Later, Fanizzi obtained [Pd(Neoc)Cl2] by treating a chloro-
form solution of [Pd(PhCN)2Cl2] with neocuproine.38 The corre-
sponding bromide and iodide complexes were obtained from [Pd
(Neoc)Cl2] by an exchange reaction with the corresponding tetra-
butylammonium halides in the same solvent (Scheme 3).
1H NMR characterisation was possible only for X = Cl, Br,
whereas the low solubility of [Pd(Neoc)I2] did not allow the
recording of the spectrum. However, elemental analysis
confirmed the supposed composition.
Being interested in revealing the real identity of those pro-
ducts, we started a more careful investigation of these apparently
trivial compounds and found that two structural isomers exist for
each of the compounds in the series, one of which had escaped
identification for more than 40 years, despite having previously
been obtained in several cases. In recent years, it has emerged
that the way palladium–phosphine complexes are prepared can
change the final product and, especially, its catalytic activity.39–41
In this paper, we demonstrate that a similar problem can also
occur with a nitrogen ligand and we stress that more attention
should be paid in correctly identifying the actual species
employed or generated in situ than done in the past.
As already reported in the literature,43 [Pd(CH3CN)2I2] is not
stable and thus it could not be used as starting material to syn-
thesise 3. Both the reaction of palladium acetate with neocu-
proine and KI in methanol and the exchange reaction from 1a
afforded insoluble 3b. Only by applying the original protocol,
employing [Pd(Neoc)(NO3)2] as starting material, was some
soluble 3a, in mixture with 3b, obtained.
Results and discussion
Syntheses and molecular structures
Our initial observations originated by solubility differences
obtained when [Pd(Neoc)Cl2] was prepared by reaction of [Pd
(CH3CN)2Cl2] with neocuproine in either chloroform or methyl-
ene chloride. Whereas the product of the reaction in chloroform
(1a) showed a good solubility in both CHCl3 and CH2Cl2, that
obtained from methylene chloride (1b), both at room and reflux
temperature, was completely insoluble in both chlorinated
Having a clearer picture of which conditions are suitable to
obtain each isomer, we next started a series of attempts to grow
crystals of the different compounds in order to unequivocally
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 3648–3658 | 3649

View Article Online
square planar and the phenanthroline plane coincides with the
coordination plane of the metal. This also holds true for substi-
tuted phenanthrolines, but not those substituted in ortho, like
neocuproine. In this case a huge distortion occurs, obviously due
to a steric effect, so that the plane of Neoc and the square plane
of the metal are quite inclined. This is evident in Fig. 1 and 2 for
compounds 1a and 2a. The distortion implies a loss of planarity
for Neoc and therefore a re-arrangement of the bonds within its
skeleton.
A solid state form of compound 1a has recently been
reported,44 but the form we obtained (1a·1CH2Cl2) is a pseudo-
2
polymorph, having the crystallisation solvent co-crystallised (in
a ratio 1 : 2). The space group type of 1a and 1a·12CH2Cl2 is inci-
dentally the same (P21/n, keeping the same setting adopted in
ref. 44) but the crystal lattice dimensions are quite different.
Crystals of 1a·12CH2Cl2 contains two molecules of 1a in the
asymmetric unit and the unit cell volume is of course more than
2 times larger (having CH2Cl2 inside also). The unprecedented
structure of 2a is instead isomorphic with 1a, therefore with just
one molecule in the asymmetric unit and no co-crystallised
solvent. No traces of a 2a·1CH2Cl2 polymorph were found. The
molecular structures of 1a 2(either with clathrated solvent or not)
and 2a are all very similar, characterised by the classical distor-
tion of Neoc tetra-coordinated Pd or Pt complexes. The angles
between the two planes (Neoc and N2PdX2) are ca. 40°. The Pd
atom is always slightly shifted out of the average plane of the
four coordinated atoms, making a slight distortion toward square
pyramidal.
Fig. 1 Ortep view of the two independent molecules of 1a in the
crystal [Pd(Neoc)Cl2]·12CH2Cl2. The solvent was omitted for clarity.
Ellipsoids are drawn at 50% probability level.
These molecular distortions make the Neoc coordination to Pd
less efficient than the analogous phenanthroline. In [Pd(Phen)
Cl2],45 Pd–N distances are 2.036(2) Å (averaging the two dis-
tances), whereas in 1a they are 2.06(1) Å (an average between
the two crystalline forms; in parentheses standard deviation from
the mean value are given, not the experimental uncertainties).
For the Br derivative, the Pd–N distances are 2.053(6) for [Pd
(Phen)Br2]46 and 2.08(1) Å for 2a, respectively (averaging the
two distances). Similar trends affect the Pd–X distances, which
are longer for the Neoc derivatives.
These PdNeoc derivatives are quite similar to the correspond-
ing PtNeoc ones,38,47 where the same distortions are observed
compared with the phenanthroline ligands. Interestingly, while Pt
47
(Neoc)Br2 is isomorphous to 2a, Pt(Neoc)Cl2 crystallises in a
different form, orthorombic Pbca, without co-crystallised
solvent.38
For both clathrated and solvent free crystalline forms of 1a,
the packing is characterised by antiparallel stacking of the Neoc
ligands, perpendicular to the monoclinic axis in direction (1, 0,
−1) for the clathrated form, along direction (1,1,1) for the
solvent free one. This motif also characterises Pt(Neoc)Cl2.
The very low solubility of type b compounds required
growing single crystals in a different way, namely performing
the reaction in a double layer: an acetonitrile solution of [Pd
(CH3CN)2Br2] was layered over a dichloromethane solution of
neocuproine allowing the diffusion to take place at −40 °C. As
shown in Fig. 3, 2b is a ionic compound of composition [Pd
(Neoc)2Br]2[Pd2Br6]. The CH2Cl2 solvent co-crystallised.
The structure of the cation is similar to that recently reported
for the Cl derivative,48 crystallised as the (PF6)− salt. Despite
apparent penta-coordination, the structure is in fact a highly
Fig. 2 Ortep view of the molecular structure of [Pd(Neoc)Br2], 2a.
Ellipsoids drawn at 50% probability level.
characterise at least a member of each series by X-ray
diffraction.
Crystals of 1a and 2a could be successfully grown by slow
diffusion of hexane into a CH2Cl2 solution of the corresponding
complex at RT. Under these conditions, the chloride complex
crystallised as a CH2Cl2 solvate (Fig. 1).
As shown in Fig. 1 and 2, compounds of type a are the
expected mononuclear neutral complex.
All the structures of [Pd(Phen)X2] or [Pt(Phen)X2] (Phen =
1,10-phenanthroline) known to date are almost undistorted
3650 | Dalton Trans., 2012, 41, 3648–3658
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 4 The molecular structure of [Pd(Neoc)Br2(DMSO)] (2c) in the
solid state. Ellipsoids are drawn at 50% probability level.
Polar non-protic solvents, the only ones in which compounds b
are at least partly soluble, should be considered with caution.
Indeed, when 2b was dissolved in DMSO and the solution was
layered with THF, a new penta-coordinated palladium complex
crystallised, identified as [Pd(Neoc)Br2(DMSO)] (2c) with
DMSO coordinated to Pd (Fig. 4). An analogous complex with
platinum as metal and iodide as ligand is already known,49 but
the solid state form is different; 2c is monoclinic, C2/c, whereas
Fig. 3 Ortep view of the solid state form of [Pd
(Neoc)2Br]2[Pd2Br6]·CH2Cl2 (2b·CH2Cl2); ellipsoids are drawn at 50%
probability level. Solvent omitted for clarity.
distorted square planar with a fifth, very long Pd1–N21 coordi-
nation, 2.518(06) Å. Fig. 3 presents an orientation that highlights
the distorted square planar. The three Pd–N bonds are not identi-
cal, with Pd–N31 being longer than the other two (N11–Pd1
2.045(6) Å; N31–Pd1 2.113(6) Å; N41–Pd1 2.031(5) Å).
ˉ
[Pt(Neoc)I2(DMSO)] is triclinic P1.
Compound 2c is also a pseudo-pentacoordinate complex,
where one of the two Neoc nitrogens is only weakly bound
(Pd1–N21 2.571(5)). In a forthcoming paper,50 we will discuss
in more details this peculiar stereochemistry of penta-coordinated
[Pd(Neoc)X2L], which is quite a function of the nature of the X
and L ligands. The structure of 2c is somewhat analogous to that
of [Pd(Neoc)Cl2(CO)].33
ˉ
Compound 2b·CH2Cl2 crystallises as triclinic (P1) and the
crystal packing is also dominated by the stacking of Neoc
ligands along (1,−1,0) direction. Anion–cation interactions are
not strong and just limited to few C–H⋯Br. This solid state form
is unambiguously different from both forms of a-type com-
pounds, therefore an X-ray powder diffraction (XRPD) exper-
iment on the bulk would immediately reveal the possible
presence of both isomers.
However, XRPD on a sample obtained from the reaction of Pd
(OAc)2 with neocuproine and NaBr in methanol confirmed the
presence of just a single solid state phase, with a diffractogram
identical to that of 2b·CH2Cl2, implying the presence of an iso-
morphic 2b·CH3OH. Therefore, no more than two compounds
(2a and 2b) are needed to explain the obtained results.
The fact that 2c is quickly obtained from 2b supports the idea
that the monomeric complexes are the thermodynamically
favoured compounds and that the formation of compounds of
type b is driven by their fast formation and their insolubility. If
compounds of type b are given the possibility to dissolve, most
times they will isomerise to type a compounds. Such isomerisa-
tion further explains why the existence of isomers b had not
been appreciated before. We also incurred this misunderstanding.
We have recently reported the synthesis and X-ray crystal struc-
ture determination of the full series of compounds [Pd(Neoc)
X2(CO)], obtained by carbonylating “[Pd(Neoc)X2]” in
CH2Cl2.33,34 Some of the reactions were performed several
times. In some cases the starting materials were insoluble in the
solvent, but dissolved during the reaction and the same, soluble,
carbonylated compound was obtained independent of the solubi-
lity of the starting material. Retrospectively, the crystals
employed for the X-ray diffraction analysis of both the chloride
and iodide members of the series had been grown from solutions
of the corresponding compounds obtained employing samples of
1b and 3b, respectively. Only crystals of [Pd(Neoc)Br2(CO)]
Compound 2b has clearly the same composition as 1b thus
explaining why its existence had not been realised up to now (a
slight difference in the amount of co-crystallised solvent may
depend on the crystallisation conditions and would anyway be
difficult to reliably assess by elemental analysis).
1H NMR spectra of 1a–3a in CDCl3 all show the presence of
a single set of signals attributable to a neocuproine ligand in a
symmetrical environment. Except for this, it is not possible to
draw any other conclusion from this data. On the other hand, no
spectrum in CDCl3 can be recorded for any compound of type b.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 3648–3658 | 3651

View Article Online
were obtained by a solution obtained by carbonylating a type a
isomer.
While writing the present work, we became aware of a very
recent paper44in which complex 1a was obtained by reaction of
[PdCl2(DMF)2] and neocuproine in DMSO, followed by recrys-
tallisation in DMF. Given the ability of DMSO to transform
ionic 2b into a neutral species, it is possible that this solvent also
plays a role in addressing the reaction of [PdCl2(DMF)2] and
neocuproine towards neutral 1a.
Although isomerisation of type a into type b species was
observed under several experimental conditions, it must be
remarked that, in our experience, this only holds if other anions
are not added. On the other hand, when a suspension of species
1b was treated with an excess of tetrabutylammonium chloride,
the compound dissolved within 2 h, but reprecipitation with
methanol again afforded the same insoluble isomer. When the
1
reaction was performed in deuterated chloroform, the H NMR
Fig. 5 Ortep
drawing
of
[Pd(Neoc)2Br]Br·2H2O·CHCl3
spectrum of the solution evidenced the presence of free neocu-
proine and another series of signals attributable to a neocuproine
ligand in a symmetrical environment. The former is probably
due to neocuproine displacement associated with the formation
(4·2H2O·CHCl3).51 Ellipsoids are drawn at 50% probability level.
Compound 4 has already been reported in the literature,
obtained by reaction of 2a with excess neocuproine, but no crys-
tallographic characterisation was carried out.32,35 Under our con-
ditions, formation of 4 is clearly associated to the other reactions
observed: loss of acetonitrile from [PdBr2(CH3CN)2] generates
PdBr2 oligomers that precipitate together with 2b. The excess
ligand that remains in solution results in the formation of 4.
2−
in solution of species [PdCl]4 while the latter may be due to a
compound such as [(η1-Neoc)PdCl3]−, although we have not
investigated this aspect in more detail and the composition of the
latter complex must be regarded as speculative. In analogy, it has
been reported38 that the use of an excess of tetraalkylammonium
iodide when performing the ion exchange on compound 1a
leads to formation of [NR4]2[Pd2I6] as a by-product.
1
It should be noted that the H NMR spectrum of 4 (and also
that of the chloride analogue) has been reported35 to show
signals in accordance with an asymmetrical neocuproine.
However, this is in contrast not only with our observation, but
also with the fluxional nature of all other pentacoordinated neo-
cuproine adducts we have analysed. Thus, the original report is
probably partially erroneous and a mixture of complexes was
likely obtained at the time.
As previously mentioned, treating [Pd(CH3CN)2Br2] with
neocuproine in CHCl3 afforded a precipitate of complex 2b,
apparently contaminated by palladium bromide (as indicated by
elemental analysis), but small amounts of other soluble com-
pounds were present in solution. It should first be noticed that
we observed formation of palladium bromide even when a sol-
ution of [Pd(CH3CN)2Br2] in chloroform or methylene chloride
was left to stand, but the decomposition is faster in CHCl3.
Moreover, when analytically pure [Pd(CH3CN)2Br2] was ana-
Species 4 were obtained in crystal form, co-crystallised with
H2O and CHCl3. As for 2b, the cation can be rationalised as a
distorted tetracoordinated square planar with a fifth loose coordi-
nation (Pd–N41 = 2.579(7) Å). The other three Pd–N bonds are
more similar, instead (ranging from 2.048(7) to 2.081(7) Å). The
crystal structure is monoclinc, P21/c and the packing is again
characterised by the stacking of Neoc ligands and weak
C–H⋯Br− interactions. Co-crystallised H2O and CHCl3 are also
involved in hydrogen bond networks.
1
lysed by H NMR in CDCl3, an immediate decomposition was
observed. Even after less than one minute from the solvent
addition, the signal of free acetonitrile was largely dominant (ca.
95% of the total intensity in the CH3 region) and only two very
weak signals were observable at 2.42 and 2.37 ppm. This
strongly suggests that aggregated units of the type [Pd(CH3CN)
Br]2(μ2-Br)2 are easily formed in these solvents, from which
aggregation can further proceed to eventually produced insoluble
oligomeric species containing little, if any, coordinated aceto-
nitrile. It should be noted that the presence of free acetonitrile
1
As previously mentioned, H NMR characterisation of com-
pounds 1–3 can be performed in CDCl3 only for the soluble
isomers a, whereas the use of coordinating solvents should be
avoided. The observed signals for 1a and 2a are in very good
agreement with those reported by Fanizzi,38 whereas the agree-
ment is less good with the only reported spectrum of the iodide
complex.35 This may be due to the low resolution of the instru-
ment employed in that relatively old work. The spectra show a
symmetrical coordination of the neocuproine ligand, in accord
with the solid state structure.
UV spectra could also be recorded only for type a complexes.
Our data agree quite well with those already reported in the lit-
erature32 for the chlorine and bromine derivatives, although they
are quite different from what was reported in the same reference
for the iodine one.
1
was also detected when recording a H NMR spectrum of [Pd
(CH3CN)2Cl2], but in much lower amounts.
Concerning the soluble fraction of the reaction between [Pd
(CH3CN)2Br2] and neocuproine in CHCl3, 1H NMR spec-
troscopy showed the presence of three compounds, two of them
(including 2a) present in low amounts, each showing a sym-
metrical environment for the neocuproine ligand. By cooling this
solution at −20 °C, crystals of one of those compounds were
obtained and its X-ray structure solved. The compound was
shown to be [Pd(Neoc)2Br]Br (4) (Fig. 5).
The cationic part of 4 is the same as in 2b, but the anionic
part does not contain palladium.
3652 | Dalton Trans., 2012, 41, 3648–3658
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 1 Polarity properties of the employed chlorinated solvents
CHCl3
CH2Cl2
C2H4Cl2
Dielectric constant 52
Dipole moment 52
4.89
1.15
23.1
9.02
1.14
20.4
10.74
1.83
16.7
Acceptor number 53–55
Reaction pathway to different isomers
We can now analyse which is the plausible pathway leading to
the different isomers and the reasons why one is favoured over
the other. We will first focus on the reactions employing [Pd
(CH3CN)2X2] type complexes as starting materials and in halo-
genated solvents, because these are the conditions in which
different isomers can be obtained by slight variations of the
experimental conditions.
Because 1b isomerises to 1a under some conditions, but the
reverse was never observed, 1a must be the thermodynamic
product. Isomer 1b is surely the kinetic product under some con-
ditions, but the simplest general scenario implies that it may be
always the kinetic product. Depending on the solvent and temp-
erature, 1b may or may not have the time to isomerise to 1a
before precipitating.
As far as the solvent influence is concerned, some properties
of CHCl3, C2H4Cl2 and CH2Cl2, from which only 1a, a mixture
of 1a and 1b, and 1b only are obtained at RT, are reported in
Table 1.
A clear correlation between a single solvent property and the
obtained results cannot be obtained, but chloroform has the
highest acceptor number and forms the stronger hydrogen bonds.
This implies that it should labilise coordinated chloride more
efficiently than the other two solvents. Dichloroethane, on the
other hand, is the most polar and should better solubilise the
ionic isomer, retarding its precipitation and giving it the time to
isomerise, a feature in accord with the observation that at reflux
temperature only 1a is obtained.
Scheme 5
The influence of the halide can now also be discussed. From
an analysis of the data it emerges that it is increasingly more
difficult to obtain the soluble isomer when descending the
halogen group. There are two reasons that can justify this trend.
The first is that several data in the literature indicates that in pal-
ladium complexes the tendency of halides to assume a bridging
position increases in the order Cl < Br < I,56 and bridging
halides are present in isomer b complexes, but not in type a
ones. The second, in cases where it is pertinent, is that chloride
forms stronger hydrogen bonds than its heavier homologues, and
this feature will facilitate halide abstraction from [Pd(CH3CN)
X3]−, orienting the reaction towards the formation of the neutral
isomer a, as in Scheme 5.
The reaction pathway shown in Scheme 5 accounts for all
experimental observations.
Conclusions
Given the strong tendency of halides to assume a bridging
position in palladium complexes, we propose that the starting
point of the reaction with neocuproine is a dimeric complex
formed from two [Pd(CH3CN)2X2] units by acetonitrile loss.
Observation of free acetonitrile in the 1H NMR spectrum of both
[Pd(CH3CN)2Cl2] and [Pd(CH3CN)2Br2] supports the presence
of such type of complexes in solution.
The neocuproine attack should generate two intermediates,
none expected to be stable under the reaction conditions. They
will evolve to the two components of type b complexes, from
which precipitation will occur. If the solvent can give strong
enough hydrogen bonds, it may displace a halide from the
anionic [Pd(CH3CN)X3]− giving an unsaturated intermediate
that easily reacts with neocuproine to give a type a complex. The
resulting free halide can coordinate to the cationic [Pd(Neoc)
(CH3CN)X]+, affording additional a complex. When X = Br, the
unsaturated [Pd(CH3CN)Br2] may also aggregate to insoluble
PdBr2 (actually a polymer with bridging bromides). The excess
neocuproine and bromide left in solution can result in the pre-
cipitation of [Pd(Neoc)2Br]Br (4).
A summary of the results obtained by different synthetic
approaches is shown is Scheme 6. Out of the reported synthetic
strategies, reaction of [Pd(CH3CN)2X2] with neocu-
proine33,34,36,37 led to either isomer a or b depending on the con-
ditions in the case of the X = Cl, but only to isomer b in the case
of the brominated compound. The exchange reaction of [Pd
(Neoc)Cl2] (1a)38 led to isomer a in case of X = Br, but to
isomer b for X = I. Reaction of palladium acetate with neocu-
proine and NaX (X = Br, I)42 afforded in both cases isomer b,
whereas the firstly reported strategy, metathesis of [Pd(Neoc)
(NO3)2] with NaX,32 gave isomer a for X = Cl, Br and a mixture
of isomers a and b for X = I. This is the only procedure by
which the neutral iodinated derivative could be obtained, even if
in a mixture with the ionic one.
The ionic isomer can be converted into the neutral one under
some conditions and dissolution of the ionic isomer in DMSO
gives an adduct of the neutral compound, a behaviour that
recalls that observed when reacting either ionic or neutral deriva-
tives with CO.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 3648–3658 | 3653

View Article Online
Scheme 6
Synthesis of PdX2
At this stage, a similar, if not the same, behaviour is to be
expected with other sterically hindered phenanthrolines such as
2,9-dimethyl-4,7-diphenylphenanthroline (batocuproine) and,
likely, related derivatives such as 6,6′-dimethylbipyridine. On the
other hand, it is not experimentally determined if complexes of
the type [Pd(L)2X]2[Pd2X6] can also exist for other phenanthro-
lines lacking substituents in the 2 and 9 positions, or whether
the same kind of isomers a and b can be also obtained for
related platinum complexes. This will be the topic of a future
study.
Palladium bromide was prepared dissolving metallic palladium
in a nitric acid (65%) and bromidic acid (48%) mixture accord-
ing to the reported procedure.43 Metallic palladium was comple-
tely dissolved by boiling the mixture; the volume of the solution
was reduced and nitric acid was added three times, evaporating
the solution almost to dryness every time to eliminate the HBr
excess and decompose the [PdBr4]2− ion. The solution was
finally evaporated to dryness and a powdered brown precipitate
was obtained.
Synthesis of [Pd(CH3CN)2X2] (X = Cl, Br)
Experimental
[Pd(CH3CN)2Cl2]. The procedure is adapted from that already
described for the corresponding benzonitrile complexes.57 Palla-
dium chloride was suspended in acetonitrile (120 ml CH3CN per
gram of PdCl2) and refluxed for 2 h allowing PdCl2 to essen-
tially dissolve completely. The hot solution was filtered through
filtering paper with the aid of a cannula to remove very small
amounts of undissolved impurities. The solution was then cooled
at 0 °C to promote [Pd(CH3CN)2Cl2] precipitation. The yellow
solid was recovered by filtration with a Buchner. The remaining
solution was then evaporated in vacuo to afford additional
complex (quantitative yield). Calcd for C4H6N2Cl2Pd: C, 18.52;
H, 2.33; N, 10.80%. Found: C, 18.78, H, 2.26; N, 10.72%. δH
(CDCl3, RT, 300 MHz): 2,40 (s, CH3). The spectrum also shows
a signal at 2.03 ppm indicating free acetonitrile. Aweak signal at
2.39 ppm is also observable. IR (nujol): 2335, 2304 cm−1 (in
General procedures
Unless otherwise stated, all reactions and manipulations were
conducted under a dinitrogen atmosphere employing standard
Schlenk and cannula techniques. CHCl3, CH2Cl2, C2H4Cl2,
CH3CN (CaH2) and MeOH (Mg(OMe)2) were dried by distilla-
tion over the corresponding drying agent under dinitrogen
immediately before use. All glassware was kept in an oven at
120 °C for at least 2 h and left to cool under vacuum before use.
2,9-Dimethyl-1,10-phenanthroline was purchased as the hydrate.
It was dried by dissolving it in CH2Cl2, drying the resulting sol-
ution with Na2SO4, filtering the suspension under dinitrogen,
and evaporating the filtered solution in vacuo. It was then stored
under dinitrogen. It can be weighed in the air without problems,
but must be stored in an inert atmosphere if water uptake is to be
avoided. All other chemicals were purchased from Aldrich,
Acros or Alfa Aesar and used as received. CDCl3 was purified
by passing it through a short column of basic alumina that had
been previously dried by heating it in vacuo with a heating gun
(to remove all acidic impurities and most of the water). The so
purified solvent was degassed and stored over activated molecu-
lar sieves under dinitrogen and in the dark. DMSO-d6 was dis-
tilled over CaH2, degassed and stored over activated molecular
sieves under dinitrogen. 1H NMR spectra were recorded on a
Bruker AC300 FT or on an Avance Bruker DPX300 spec-
trometer, working at 300 MHz. Signals are referred to SiMe4 and
J values are given in Hz. Elemental analyses were recorded on a
Perkin-Elmer 2400 CHN Elemental Analyser.
accord with the literature58) IR (C2H4Cl2): 2337, 2306 cm−1
.
[Pd(CH3CN)2Br2]. The same procedure described above was
employed. Calcd for C4H6N2Br2Pd: C, 13.79; H, 1.74; N,
8.04%. Found: C, 13.61; H, 1.66; N, 7.78%. δH (CDCl3, RT,
300 MHz): 2.41(s). Also observable is a singlet at δ 2.37,
attributable to a dimeric species, and a much more intense signal
(integrated intensity ca. 70 fold that of the other two signals) at δ
2.03 (s, free CH3CN).
Synthesis of [Pd(Neoc)(NO3)2]
The synthesis was performed by adapting the procedure already
reported for the corresponding phenanthroline complex.59
3654 | Dalton Trans., 2012, 41, 3648–3658
This journal is © The Royal Society of Chemistry 2012

View Article Online
Reaction run in CHCl3 at RT. The reaction was performed
analogously to the previous one, but at RT, starting from [Pd
(CH3CN)2Cl2] (45.0 mg, 0.17 mmol) and neocuproine (48.0 mg,
0.23 mmol), in CHCl3 (7 mL). The solution was stirred for 2 h
and then the product was precipitated with diethyl ether. The
solid was collected by filtration on a Buchner and dried in vacuo
(85% yield). The solid is completely soluble in CHCl3.
Palladium sponge was dissolved in 65% HNO3. Then, solid neo-
cuproine was added in small portions at half an hour intervals,
until the brown solution became almost colourless. The orange
precipitate was collected by filtration through a Buchner and
washed with diluted HNO3. After drying in the air and under
vacuum, it was analytically pure and no recrystallisation was
needed. Calcd for C14H12N4O6Pd: C, 38.33; H, 2.76; N,
12.77%. Found: C, 38.58; H, 2.57; N, 12.71%.
Reaction run in CH3CN. [Pd(CH3CN)2Cl2] (105.2 mg,
0.405 mmol) was dissolved in warm (ca. 40 °C) acetonitrile
(10 mL) and solid neocuproine (106.2 mg, 0.510 mmol) was
added to the hot solution. An orange-brown precipitate was
immediately formed. Stirring was continued for 2 h to complete
the precipitation. The solid was collected by filtration on a
Buchner and dried in vacuo. (152.6 mg, 98% yield). The solid is
completely insoluble in CHCl3. Calcd for C14H12N2Cl2Pd: C,
43.61; H, 3.14; N, 7.27%. Found: C, 43.95; H, 3.52; N, 7.57%.
Synthesis of [Pd(Neoc)Cl2]/[Pd(Neoc)2Cl]2[Pd2Cl6] from
[PdCl2(CH3CN)2]
Reaction run in refluxing C2H4Cl2. To a Schlenk flask
equipped with a reflux condenser and under a dinitrogen atmos-
phere, [Pd(CH3CN)2Cl2] (115.2 mg, 0.444 mmol) and neocu-
proine (113.2 mg, 0.544 mmol) were added. C2H4Cl2 (10 mL)
was then added and the suspension was stirred and heated at
reflux. The completeness of the reaction was checked after 2 h
by assessing the absence of IR stretching bands at 2335 and
2304 cm−1 (in nujol, on a solution sample evaporated in vacuo)
due to the reagent. The orange crystalline precipitate was col-
lected by filtration on a Buchner (67% yield). Calcd for
C14H12N2Cl2Pd: C, 43.61; H, 3.14; N, 7.27%. Found: C, 43.96;
H, 3.54; N, 7.16%. δH (CDCl3, 300 MHz, RT) 8.32 (d, 2H, J =
8.4); 7.86 (s, 2H); 7.57 (d, 2H, J = 8.4), 3.29 (s, 6H). λmax
(CH2Cl2)/nm 359, 395 (lit.,32 360, 390). Evaporation of the sol-
ution afforded an additional crop of a material with the same
composition (33% yield, after washing out the excess neocu-
proine with toluene), although a small impurity must be present
because the product appeared slightly darker. Both crops of
product were completely soluble in CHCl3.
Synthesis of [Pd(Neoc)Br2]/[Pd(Neoc)2Br]2[Pd2Br6] from
[PdBr2(CH3CN)2]
Reaction run in refluxing C2H4Cl2. The reaction was per-
formed analogously to that described for the corresponding reac-
tion of the chlorinated derivative, but for 10.5 h, starting from
[Pd(CH3CN)2Br2] (104.6 mg, 0.30 mmol) and neocuproine
(75.0 mg, 0.36 mmol), in C2H4Cl2 (8 mL). The precipitated
product (90% yield) was collected by filtration on a Buchner.
From the remaining solution, only a trace amount of a product
that may be 2a could be isolated. The precipitate product is com-
pletely insoluble in CHCl3. Calcd for C14H12Br2N2Pd
(“[Pd(Neoc)Br2]”): C, 35.44; H, 2.55; N, 5.90%. Found:
C, 35.20; H, 2.48; N, 5.55%.
Reaction run in C2H4Cl2 at RT. The reaction was performed
analogously to that described above, but at RT for 24 h, starting
from [Pd(CH3CN)2Cl2] (220.0 mg, 0.850 mmol) and neocu-
proine (212.0 mg, 1.02 mmol), in C2H4Cl2 (17 mL) The product
recovered by filtration was formed by a mixture of 1a (which
was separated by dissolving it in CHCl3 (17% yield) and 1b
(60% yield). Additional 1a could be recovered by evaporating
the filtered solution in vacuo.
Reaction run in CH2Cl2 at RT. The reaction was performed
analogously to the previous one, but at RT for 24 h, starting
from [Pd(CH3CN)2Br2] (153.4 mg, 0.44 mmol) and neocuproine
(110.0 mg, 0.53 mmol), in CH2Cl2 (12 mL). The brown precipi-
tate (93% yield) was completely insoluble in CHCl3. Calcd
for
C57H50Br8Cl2N8Pd4
(corresponding
to
[Pd
(Neoc)2Br]2[Pd2Br6]·CH2Cl2, in accord with the composition
determined by X-ray diffraction): C, 34.53; H, 2.54; N, 5.65%.
Found: C, 34.38; H, 2.20; N, 5.51%.
Reaction run in refluxing CH2Cl2. The reaction was per-
formed analogously to that described above, but at reflux for
15 h, starting from [Pd(CH3CN)2Cl2] (50.5 mg, 0.19 mmol) and
neocuproine (48.0 mg, 0.23 mmol), in CH2Cl2 (7 mL). The
brown-orange solid was completely insoluble in CHCl3. Calcd
for C57H50Cl10N8Pd4 (corresponding to [Pd(Neoc)Cl2]·1/
4CH2Cl2): C, 42.07; H, 3.10; N, 6.89%. Found: C, 42.49; H,
3.13; N, 6.83%.
Reaction run in CHCl3 at RT and preparation of single crys-
tals of [Pd(Neoc)2Br]Br (4). The reaction was performed analo-
gously to the previous one, but at RT for 3 h, starting from [Pd
(CH3CN)2Br2] (23.8 mg, 0.068 mmol) and neocuproine
(16.1 mg, 0.077 mmol), in CHCl3 (15 mL). A larger solvent
amount than usual was employed in this case, in the hope of
facilitating the solubilisation of intermediate species. The brown
precipitate (93% yield) was completely insoluble in CHCl3.
Calcd for C57H49Br8Cl3N8Pd4 (corresponding to [Pd
(Neoc)2Br]2[Pd2Br6]·CHCl3): C, 33.94; H, 2.45; N, 5.27%.
Found: C, 32.89; H, 2.27; N, 5.34%. The low values for all
elements are attributed to contamination by PdBr2, in accord
with the easy decomposition of the starting complex in chloro-
form, as discussed before. The reaction solution was dried
Reaction run in CH2Cl2 at RT. The reaction was performed
analogously to the previous one, but at RT for 24 h, starting
from [Pd(CH3CN)2Cl2] (52.0 mg, 0.20 mmol) and neocuproine
(50.2 mg, 0.24 mmol), in CH2Cl2 (7 mL). The brown-orange
solid was completely insoluble in CHCl3. Calcd for
C57H50N8Cl10Pd4 (corresponding to [Pd(Neoc)Cl2]·1/4CH2Cl2):
C, 42.07; H, 3.10; N, 6.89%. Found: C, 42.31; H, 3.08; N,
6.86%.
1
in vacuo and H NMR (CDCl3) analysis revealed that, besides
very small amounts of 2a, two more compounds are present in
Dalton Trans., 2012, 41, 3648–3658 | 3655
This journal is © The Royal Society of Chemistry 2012

View Article Online
solution. The NMR tube was placed in a refrigerator at −20 °C
and a few single crystals of compound 4 formed over two days.
suspension was stirred at room temperature for 1 h. The precipi-
tated solid was collected by filtration on a Buchner, washed with
a 1 : 1 MeOH–H2O solution and dried in vacuo (80% yield).
Reaction run in refluxing CH3CN. [Pd(CH3CN)2Br2]
(60.0 mg, 0.17 mmol) was dissolved in hot (70 °C) acetonitrile
(10 mL) and solid neocuproine (40.4 mg, 0.19 mmol) was added
to the hot solution. A dark brown precipitate was immediately
formed. The solid was collected by filtration on a Buchner and
dried in vacuum. The solid is completely insoluble in CHCl3.
Calcd
for
C56H48Br8N8Pd4
(corresponding
to
[Pd
(Neoc)2Br]2[Pd2Br6]): C, 35.44; H, 2.55; N, 5.90%. Found: C,
36.32; H, 2.60; N, 6.13%. Despite the fact that the analysis for
carbon is slightly out of range (which may be due to some clath-
rated solvent), the identity of the compound was confirmed by
X-ray powder diffraction. The spectrum of the product was in
good agreement with that simulated for 2b based on the single
crystal data. The same reaction was repeated at reflux tempera-
ture, but extensive formation of metallic palladium was
observed.
Calcd
for
C58H51Br8N9Pd4
(corresponding
to
[Pd
(Neoc)2Br]2[Pd2Br6]·CH3CN): C, 35.93; H, 2.65; N, 6.50%.
Found: C, 36.24; H, 2.41; N, 6.56%.
Syntheses by exchange reaction from [Pd(Neoc)Cl2] (1a)
[Pd(Neoc)2I]2[Pd2I6] (3b). The synthesis was performed as the
previous one, starting from Pd(OAc)2 (53.0 mg, 0.236 mmol),
neocuproine (63.2 mg, 0.35 mmol) and NaI (85.1 mg,
0.66 mmol), in methanol (5 mL). A black precipitate was
formed (94% yield), which is completely insoluble in CHCl3.
Calcd for C56H48I8N8Pd4 ([Pd(Neoc)2I]2[Pd2I6]): C, 29.58; H,
2.13; N, 4.93%. Found: C, 29.84; H, 2.20; N, 5.17%.
[Pd(Neoc)Br2] (2a). The exchange reaction between [Pd
(Neoc)Cl2] and twice the required molar amount of Bu4NBr in
chloroform was performed as reported in the literature.38 The
product was completely soluble in CHCl3, in accord with the
proposed neutral composition. However, in our hands, a single
exchange reaction was not sufficient to remove all chloride. The
1H NMR spectrum of the complex in CDCl3 does not evidence
the presence of the residual chloride because the signals of 1a
and 2a are too close to each other. However, the elemental analy-
sis of the product was out of range and when it was employed as
a substrate for a carbonylation reaction to give [Pd(Neoc)
Br2(CO)],33 the presence of [Pd(Neoc)Cl2(CO)] could be also
detected, since the signals of the chloride and bromide com-
plexes are more separated in this case. The exchange procedure
had to be repeated twice more with the same amounts to get a
completely chloride-free and analytically pure compound. Calcd
for C14H12Br2N2Pd: C, 35.44; H, 2.55; N, 5.90%. Found: C,
35.80; H, 2.64; N, 5.60%. δH (CDCl3, 300 MHz, RT): 8.31 (d,
2H, J = 8.4 Hz); 7.85 (s, 2H); 7.58 (d, 2H, J = 8.4 Hz); 3.32 (s,
6H). λmax (CH2Cl2)/nm 362, 412 (lit.,32 355, 410).
Syntheses of [Pd(Neoc)I2] (3a) by metathesis from
[Pd(Neoc)(NO3)2]
[Pd(Neoc)(NO3)2] (19.7 mg, 0.045 mmol) and NaI (35.6 mg,
0.237 mmol) were introduced in a Schlenk under a dinitrogen
atmosphere and then ethanol (5 mL) was added under dinitrogen.
The suspension was stirred at room temperature for 24 h. The
solid was collected by filtration on a Buchner, washed with
ethanol and dried in vacuum (94% yield). The same reaction has
been also run in methanol. Calcd for C14H12I2N2Pd (“[Pd
(Neoc)2I2]”): C, 29.58; H, 2.13; N, 4.93%. Found: C, 30.02; H,
2.48; N, 5.03%. The precipitate was treated with CHCl3 to yield
1
a soluble fraction whose H NMR spectrum is in full agreement
with those observed for 1a and 2a, evidencing the formation of
3a (some very small impurities are observable, however). δH
(CDCl3, 300 MHz, RT): 8.32 (d, 2H, J = 8.4 Hz); 7.85 (s, 2H);
7.58 (d, 2H, J = 8.4 Hz); 3.30 (s, 6H). λmax (CH2Cl2)/nm 371
(lit.,32 445, 500 sh). The part that was not dissolved immediately
in CHCl3 was completely insoluble in this solvent, in accord
with its formulation as 3b.
[Pd(Neoc)2I]2[Pd2I6] (3b). The exchange reaction between [Pd
(Neoc)Cl2] and twice the required molar amount of Bu4NI in
chloroform was performed as reported in the literature.38 As
reported, the black product precipitated out of the solution. After
washing out the ammonium salts with methanol, it was comple-
tely insoluble in CHCl3. This is in accord with the reported
results, but contrasts with the attribution of the neutral compo-
sition (3a) to the obtained product, since 3a should be soluble in
this solvent. Calcd for C56H48I8N8Pd4 ([Pd(Neoc)2I]2[Pd2I6]): C,
29.58; H, 2.13; N, 4.93%. Calcd for C57H49I8Cl3N9Pd4 (corre-
sponding to [Pd(Neoc)2I]2[Pd2I6]·CHCl3): C, 28.61; H, 2.02; N,
4.68%. Found: C, 29.01; H, 2.31; N, 4.54%. Elemental analysis
does not allow us to assess whether a crystallisation solvent mol-
ecule is present, as in the case of the chloride and bromide
derivatives, or not, since the calculated values are very close to
the experimental one in both cases.
Crystal structure determination
Compounds 1a, 2a, 2b, 2c and 4 were characterised by single
crystal X-ray diffraction. All samples were mounted in air on a
goniometer head and measured on a Bruker-APEX II CCD dif-
fractometer, at room temperature, except 2a, which was collected
at T = 150 K (using an Oxford Cryosystem 600-cryostream
device). Generator settings were 50 kVand 30 mA, radiation was
Mo Kα, graphite monochromated. Frames were collected with
0.5° ω-scans. All structures were solved by direct methods and
refined by least square refinement (using SHELX97,60 under the
wingx package61) Non-hydrogen atoms were refined including
anisotropic thermal parameters. All hydrogen atoms were placed
in geometrically calculated positions and refined using a riding
model where each H was assigned a fixed isotropic displacement
parameter. 2b·0.5(CH2Cl2) is very close to a monoclinic lattice,
Syntheses from Pd(OAc)2, neocuproine and NaX
[Pd(Neoc)2Br]2[Pd2Br6]
(2b). Pd(OAc)2
(124.2
mg,
0.553 mmol), neocuproine (149.6 mg, 0.718 mmol) and NaBr
(136.6 mg, 1.328 mmol) were introduced into a Schlenk flask
under dinitrogen and methanol (10 mL) was added. The
3656 | Dalton Trans., 2012, 41, 3648–3658
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 2 Details of X-ray diffraction data collections on 1a, 4, 2a, 2b·CH2Cl2, and 2ca
1a
4
2a
2b·CH2Cl2
2c
Formula
T (K)
λ
Crystal system
Space group
A (Å)
b (Å)
c (Å)
α (°)
β (°)
C
29H26Cl6N4Pd2
C29H29Br2Cl3N4O2Pd2
293(2)
MoKα
Monoclinic
P21/c
17.348(5)
7.517(5)
24.825(5)
90
99.147(5)
90
3196(2)
4
0.23 × 0.15 × 0.13
0–26.5°
32 988
C14H12Br2N2Pd
150(2)
MoKα
Monoclinic
P21/n
12.1525(7)
7.8107(5)
14.4621(8)
90
93.298(1)
90
1370.5(1)
4
0.2 × 0.1 × 0.05
2–29
9451
C57H50Br8Cl2N8Pd4
293(2)
MoKα
C16H18Br2N2OPdS
293(2)
MoKα
Monoclinic
C2/c
20.1737(7)
7.8067(3)
23.4069(8)
90
93.160(2)
90
3680.8(2)
8
0.2 × 0.18 × 0.1
1.7–28.7
12 802
293(2)
MoKα
Monoclinic
P21/n
9.897(4)
21.440(8)
15.012(6)
90
102.442(5)
90
3111(2)
4
0.2 × 0.16 × 0.12
0–31°
34 573
9042
0.0226
0.0215
374
Triclinic
ˉ
P1
12.141(5)
8.103(5)
16.299(5)
90.181(5)
109.819(5)
89.497(5)
1508(1)
2
0.15 × 0.12 × 0.08
1–25
10 597
5098
0.0298
γ (°)
V (Å3)
Z
Crystal size (mm3)
θ range
Reflections
Unique
Rint
6580
0.0772
0.0671
357
2520
0.0405
0.0609
174
4291
0.057
0.075
208
Rσ
0.0469
351
1.07
Parameters
GoF
1.04
1.03
0.834
1.015
R1 (I/2σ(I))
wR2 (I/2σ(I))
0.0258
0.0604
−0.47, +0.76
0.0746
0.1904
−1.8, +2.5
0.0353
0.0488
−0.37, +0.48
0.0426
0.1106
−1.42, +2.0
0.0457
0.1017
−1.0, +0.79
Residuals (eÅ−3
)
a Rint = Σ|Fo − Fo (mean)|/ΣFo ; Rσ = Σσ(Fo )/ΣFo R1 = Σ||Fo| − |Fc||/Σ|Fo|; wR2 = {Σ[w(Fo − Fc ) ]/Σ[w(Fo ) ]}1/2 GoF = {S/(n − p)}1/2
=
2
2
2
2
2
2
2 2
2 2
{Σ[w(Fo2 − Fc ) ]/(n − p)}1/2
2 2
Table 3 Metal–ligand distances in Pd Neoc complexes. Values are
in Å
Notes and references
1 W. W. Brandt, F. P. Dwyer and E. D. Gyarfas, Chem. Rev., 1954, 54, 959–
1017.
2 G. Chelucci and R. P. Thummel, Chem. Rev., 2002, 102, 3129–3170.
3 E. Schoffers, Eur. J. Org. Chem., 2003, 1145–1152.
4 F. Fache, E. Schulz, M. L. Tommasino and M. Lemaire, Chem. Rev.,
2000, 100, 2159–2232.
5 Olefin Upgrading Catalysis by Nitrogen-Based Metal Complexes II:
State-of-the-Art and Perspectives, Catal. Met. Complexes, ed.
G. Giambastiani and J. Campora, Springer, Dordrecht, The Netherlands,
2011, vol. 36.
6 V. G. Albano, G. Natile and A. Panunzi, Coord. Chem. Rev., 1994, 133,
67–114.
1aa
2a
2b
2c
4
Pd–N11 2.055(2)
Pd–N21 2.067(2)
Pd–N31 2.067(2)
Pd–N41 2.042(2)
2.093(3)
2.073(3)
2.045(6) 2.105(5)
2.518(6) 2.571(5)
2.113(6)
2.081(7)
2.061(7)
2.048(7)
2.579(7)
2.031(5)
Pd–X1
Pd–X2
Pd–X3
Pd–X4
Pd–S
2.297(1)
2.2815(9) 2.4119(5)
2.2788(9)
2.4148(4) 2.464(1) 2.4510(8) 2.435(2)
2.4687(8)
2.2931(8)
2.238(1)
7 F. P. Fanizzi, L. Maresca, G. Natile, M. Lanfranchi, A. Tiripicchio and
G. Pacchioni, J. Chem. Soc., Chem. Commun., 1992, 333–335.
8 F. P. Fanizzi, N. Margiotta, M. Lanfranchi, A. Tiripicchio, G. Pacchioni
and G. Natile, Eur. J. Inorg. Chem., 2004, 1705–1713.
9 M. E. Cucciolito, V. De Felice, G. Roviello and F. Ruffo, Eur. J. Inorg.
Chem., 2011, 457–469.
a Two molecules per asymmetric unit (X1–X4 are Cl11, Cl12, Cl13,
Cl14, respectively).
however triclinic is more likely, because: a) cell angles deviate
quite significantly from 90°; b) internal agreement for the 2/m
Laue class is significantly higher; c) the structure is more likely
centrosymmetric (based on reflection statistics) and in monocli-
nic this would imply two disordered orientations for Pd
(Neoc)2Br molecules (on the other hand the Pd2Br6 moieties are
closes to 2/m symmetry, being mmm in isolation and lying about
the −1 site in the triclinic structure).
Details of each data collection and results of each refinement
are given in Table 2. Crystallographic information files are avail-
able as Supporting Information.† Metal–ligand distances are also
reported in Table 3.
10 W. Cabri, I. Candiani, A. Bedeschi and R. Santi, J. Org. Chem., 1993,
58, 7421–7426.
11 P. A. Enquist, J. Lindh, P. Nilsson and M. Larhed, Green Chem., 2006, 8,
338–343.
12 M. M. S. Andappan, P. Nilsson and M. Larhed, Chem. Commun., 2004,
218–219.
13 M. M. S. Andappan, P. Nilsson, H. von Schenck and M. Larhed, J. Org.
Chem., 2004, 69, 5212–5218.
14 P. A. Enquist, P. Nilsson, P. Sjoberg and M. Larhed, J. Org. Chem., 2006,
71, 8779–8786.
15 D. Bianchi, R. Bortolo, R. D’Aloisio and M. Ricci, Angew. Chem., Int.
Ed., 1999, 38, 706–708.
16 C. Querci, R. D’Aloisio, R. Bortolo, M. Ricci and D. Bianchi, J. Mol.
Catal. A: Chem., 2001, 176, 95–100.
17 H. Ishii, M. Goyal, M. Ueda, K. Takeuchi and M. Asai, Appl. Catal., A,
2000, 201, 101–105.
18 A. Vavasori and L. Toniolo, J. Mol. Catal. A: Chem., 2000, 151, 37–45.
19 H. Yasuda, K. Watarai, J. C. Choi and T. Sakakura, J. Mol. Catal. A:
Chem., 2005, 236, 149–155.
20 G.-J. ten Brink, I. W. C. E. Arends, M. Hoogenraad, G. Verspui and
R. A. Sheldon, Adv. Synth. Catal., 2003, 345, 1341–1352.
Acknowledgements
We thank the MiUR for financial support.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 3648–3658 | 3657

View Article Online
21 I. W. C. E. Arends, G.-J. ten Brink and R. A. Sheldon, J. Mol. Catal. A:
Chem., 2006, 251, 246–254.
43 S. Sakaki, H. Kanai and K. Tarama, Can. J. Chem., 1974, 52, 2857–
2860.
22 G.-J. ten Brink, I. W. C. E. Arends and R. A. Sheldon, Adv. Synth. Catal.,
2002, 344, 355–369.
23 G.-J. ten Brink, I. Arends, M. Hoogenraad, G. Verspui and R. Sheldon,
Adv. Synth. Catal., 2003, 345, 497–505.
24 G.-J. ten Brink, I. Arends and R. A. Sheldon, Science, 2000, 287, 1636–
1639.
25 B. V. Popp and S. S. Stahl, J. Am. Chem. Soc., 2006, 128, 2804–2805.
26 S. S. Stahl, Angew. Chem., Int. Ed., 2004, 43, 3400–3420.
27 S. S. Stahl, J. L. Thorman, R. C. Nelson and M. A. Kozee, J. Am. Chem.
Soc., 2001, 123, 7188–7189.
28 W. R. Thiel, Angew. Chem., Int. Ed., 1999, 38, 3157–3158.
29 N. R. Conley, L. A. Labios, D. M. Pearson, C. C. L. McCrory and R.
M. Waymouth, Organometallics, 2007, 26, 5447–5453.
30 D. M. Pearson and R. M. Waymouth, Organometallics, 2009, 28, 3896–
3900.
44 V. N. Demidov, S. A. Simanova, A. I. Savinova, A. V. Zinchenko,
T. B. Pakhomova and E. A. Aleksandrova, J. Struct. Chem., 2010, 51,
296–302.
45 K. Ha, Acta Crystallogr. Sect. E.-Struct Rep., Online, 2011, 66, M38–
U471.
46 K. Ha, Acta Crystallogr. Sect. E.-Struct Rep., Online, 2010, 66, M7–
U190.
47 R. J. H. Clark, F. P. Fanizzi, G. Natile, C. Pacifico, C. G. van Rooyen and
D. A. Tocher, Inorg. Chim. Acta, 1995, 235, 205–213.
48 S. Pratihar, J. Marek and S. Roy, Inorg. Chim. Acta, 2011, 372, 362–
366.
49 F. P. Fanizzi, N. Margiotta, M. Lanfranchi, A. Tiripicchio, G. Pacchioni
and G. Natile, Eur. J. Inorg. Chem., 2004, 1705–1713.
50 P. Macchi, N. Casati, A. Krawczuk-Pantula, F. Ragaini, A. Sironi,
submitted.
31 D. M. Pearson, N. R. Conley and R. M. Waymouth, Organometallics,
2011, 30, 1445–1453.
32 R. A. Plowman and L. F. Power, Aust. J. Chem., 1971, 24, 303–308.
33 F. Ragaini, M. Gasperini, S. Cenini, L. Arnera, A. Caselli, P. Macchi and
N. Casati, Chem.–Eur. J., 2009, 15, 8064–8077.
51 The origin of the cocrystallised water is not obvious, since all reactions
had been performed under dinitrogen and with dry solvents. Given the
low amount of crystals obtained, it is possible that they have embedded
most of the water present in trace amounts. However, it should also be
recalled that the same compound has previously obtained in a “dry” form
by letting it stay over P2O5, but quickly regained clathrated water when
exposed to the air (ref. 32).
34 F. Ragaini, H. Larici, M. Rimoldi, A. Caselli, F. Ferretti, P. Macchi and
N. Casati, Organometallics, 2011, 30, 2385–2393.
35 R. D. Gillard, L. A. P. Kane-Maguire and P. A. Williams, Transition Met.
Chem., 1977, 2, 55–57.
52 C. Laurence, P. Nicolet, M. T. Dalati, J.-L. M. Abboud and R. Notario,
J. Phys. Chem., 1994, 98, 5807–5816.
36 G. R. Newkome, W. E. Puckett, G. E. Kiefer, V. K. Gupta,
F. R. Fronczek, D. C. Pantaleo, G. L. McClure, J. B. Simpson and
W. A. Deutsch, Inorg. Chem., 1985, 24, 811–826.
37 G. R. Newkome, V. K. Gupta, H. C. R. Taylor and F. R. Fronczek, Orga-
nometallics, 1984, 3, 1549–1554.
38 F. P. Fanizzi, F. P. Intini, L. Maresca, G. Natile, M. Lanfranchi and
A. Tiripicchio, J. Chem. Soc., Dalton Trans., 1991, 1007–1015.
39 C. Bianchini, A. Meli and W. Oberhauser, Organometallics, 2003, 22,
4281–4285.
40 T. J. Mooibroek, E. Bouwman, M. Lutz, A. L. Spek and E. Drent,
Eur. J. Inorg. Chem., 2010, 298–310.
53 V. Gutmann, Coord. Chem. Rev., 1976, 18, 225–255.
54 V. Gutmann, Electrochim. Acta, 1976, 21, 661–670.
55 K. M. Kadish and J. E. Anderson, et al., Pure Appl. Chem., 1987, 59,
711–714.
56 R. A. Widenhoefer and S. L. Buchwald, Organometallics, 1996, 15,
2755–2763.
57 J. R. Doyle, P. E. Slade, H. B. Jonassen and R. N. Rhoda, in Inorganic
Synthesis, John Wiley & Sons, Inc., 1960, pp. 216–219.
58 J. R. Durig, B. R. Mitchell, D. W. Sink, J. N. Willis Jr and A. S. Wilson,
Spectrochim. Acta, Part A, 1967, 23, 1121–1135.
59 G. Anderegg and H. Wanner, Inorg. Chim. Acta, 1986, 113, 101–
108.
41 A. Marson, A. B. van Oort and W. P. Mul, Eur. J. Inorg. Chem., 2002,
3028–3031.
42 M. Aresta, P. Giannoccaro, I. Tommasi, A. Dibenedetto, A. M. Manotti
Lanfredi and F. Ugozzoli, Organometallics, 2000, 19, 3879–3889.
60 G. M. Sheldrick, SHELX97-Program for the Refinement of Crystal Struc-
ture, University of Göttingen, Germany, 1997.
61 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837–838.
3658 | Dalton Trans., 2012, 41, 3648–3658
This journal is © The Royal Society of Chemistry 2012