η1-Allylpalladium complexes with a tridentate PNP ligand with different phosphino groups

Article abstract of DOI:10.1039/c2dt30746d

The iminodiphosphine 2-(PPh₂)C₆H₄-1- CHNC₆H₄-2-(PPh₂) (P-N-P′) is used for the preparation of the complexes [Pd(η¹-CHR¹-CHCR ²R³)(P-N-P′)]BF₄ [R¹ = R ² = R³ = H: (1); R¹ = R² = Ph, R³ = H: (2); R¹ = R³ = H, R² = Ph: (3); R¹ = H, R² = R³ = Me: (4)]. The P-N-P′ tridentate coordination and the η¹-allyl bonding mode in the solid are confirmed by the X-ray structural analysis of 1. In solution, the complexes 1 and 2 undergo an η¹-η³- η¹ rearrangement at 298 K interconverting the bonding site of the allyl group. A five-coordinate structure with the phosphine ligands in the axial position is proposed for the η³-allyl intermediate. For the dynamic process, a ΔG^≠ value of 53.8 kJ mol^-1 is obtained from ¹H NMR data of 2. In 3 and 4, the allyl ligand is rigidly bound to the metal through the less substituted terminus, in line with the higher free energy content of the corresponding isomers: [Pd(η¹-CHPh-CHCH₂)(P-N-P′)]⁺ +48.78 kJ mol^-1; [Pd(η¹-CMe₂-CHCH ₂)(P-N-P′)]⁺ +69.35 kJ mol^-1. The complexes react with secondary amines in the presence of fumaronitrile at different rates yielding allylamines and the palladium(0) derivative [Pd(η²-fn)(P-N-P′)] (5). On the basis of charge distribution on the allylic carbon atoms and of steric factors, the difference in rate and the regioselectivity in the amination of 1-3 are better rationalized by a mechanism with nucleophilic attack at the η³-intermediate rather than by an S_N2 mechanism with nucleophilic attack at the Pd-CHR¹ carbon atom. The high regioselectivity in the reaction of 4 with piperidine implies an S_N2′ mechanism with nucleophilic attack at the CMe₂ allyl carbon. A dynamic process occurs also for the 18-electron complex 5 consisting in a dissociation-association equilibrium of the olefin.

Full text of DOI:10.1039/c2dt30746d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 12490
www.rsc.org/dalton
PAPER
η1-Allylpalladium complexes with a tridentate PNP ligand with different
phosphino groups†
Bruno Crociani,*a Simonetta Antonaroli,a Paola Paolib and Patrizia Rossib
Received 4th April 2012, Accepted 15th August 2012
DOI: 10.1039/c2dt30746d
The iminodiphosphine 2-(PPh2)C6H4-1-CHvNC6H4-2-(PPh2) (P–N–P′) is used for the preparation of the
complexes [Pd(η1-CHR1–CHvCR2R3)(P–N–P′)]BF4 [R1 = R2 = R3 = H: (1); R1 = R2 = Ph,
R3 = H: (2); R1 = R3 = H, R2 = Ph: (3); R1 = H, R2 = R3 = Me: (4)]. The P–N–P′ tridentate coordination
and the η1-allyl bonding mode in the solid are confirmed by the X-ray structural analysis of 1. In solution,
the complexes 1 and 2 undergo an η1–η3–η1 rearrangement at 298 K interconverting the bonding site of
the allyl group. A five-coordinate structure with the phosphine ligands in the axial position is proposed
for the η3-allyl intermediate. For the dynamic process, a ΔG≠ value of 53.8 kJ mol−1 is obtained from
1H NMR data of 2. In 3 and 4, the allyl ligand is rigidly bound to the metal through the less substituted
terminus, in line with the higher free energy content of the corresponding isomers: [Pd(η1-CHPh–
CHvCH2)(P–N–P′)]+ +48.78 kJ mol−1; [Pd(η1-CMe2–CHvCH2)(P–N–P′)]+ +69.35 kJ mol−1. The
complexes react with secondary amines in the presence of fumaronitrile at different rates yielding
allylamines and the palladium(0) derivative [Pd(η2-fn)(P–N–P′)] (5). On the basis of charge distribution
on the allylic carbon atoms and of steric factors, the difference in rate and the regioselectivity in the
amination of 1–3 are better rationalized by a mechanism with nucleophilic attack at the η3-intermediate
rather than by an SN2 mechanism with nucleophilic attack at the Pd–CHR1 carbon atom. The high
regioselectivity in the reaction of 4 with piperidine implies an SN2′ mechanism with nucleophilic attack at
the CMe2 allyl carbon. A dynamic process occurs also for the 18-electron complex 5 consisting in a
dissociation–association equilibrium of the olefin.
monodentate ligand or of a tooth of the bidentate and tridentate
Introduction
ligands. The η1-allyl group was found to react with secondary
The η1-allylpalladium complexes so far reported in the literature
have been identified in solution1 by 1H and 13C NMR spec-
troscopy, and some of them have also been isolated and charac-
terized in the solid state mainly by X-ray diffraction analysis.1a,b,i,2
The configuration of the solid compounds depends on the nature
of the other ligands in the molecule. With neutral monodentate
or bidentate ligands the complexes are of the type [Pd(η1-allyl)-
(η3-allyl)L],1h trans-[PdCl(η1-allyl)L2]1a,i or [PdX(η1-allyl)-
(L–L′)]1b,c,2c–g,j (X = aryl group or chloride ion), while with
anionic or neutral tridentate ligands the complexes are of the
type [Pd(η1-allyl)(L–Y–L′)]1j and [Pd(η1-allyl)(L–L′–Y)]2i
(Y being the atom formally bearing a negative charge) or [Pd(η1-
allyl)(L–L′–L′′)]+.1e–g,2a,b,h In most cases, the complexes are
fluxional in solution as they undergo a dynamic η1–η3–η1
process, the η3-allylic species being formed by displacement of a
amines at lower rates with respect to the η3-bound ligand.1i,2j,3
From the regioselectivity of the nucleophilic attack it was
inferred that both the SN2 and SN2′ mechanisms are operative.
On the basis of theoretical calculation, however, it has been
reported that the η1-allyl ligand is unreactive toward nucleo-
philes.4 Actually, the η1-allyl ligand is susceptible to electro-
philic attack by maleic anhydride1c or by aldehydes and imines,
the latter reactions being involved in catalytic processes.1j,4a,5
Due to our interest in the chemistry and catalytic properties of
palladium derivatives with iminophosphine ligands,6 we have
recently published a synthetic and structural work on the η1-allyl
complexes [Pd(η1-CHR1–CHvCHR2)(P–N–O)] (R1 = R2 = H,
Ph; R1 = H, R2 = Ph; P–N–O as the anion derived from deproto-
nation of the hydroxylic group of 2-(PPh2)C6H4-1-CHvNC6H4-
2-OH).2i As a continuation of our studies, we wish to report
herein the preparation and structural characterization in solution
and in the solid state of the complexes [Pd(η1-allyl)(P–N–P′)]-
BF4 and [Pd(η2-fumaronitrile)(P–N–P′)] (allyl = C3H5, 1,3-
Ph2C3H3, 3-PhC3H4, 3,3-Me2C3H3; P–N–P′ as the iminodiphos-
phine ligand 2-(PPh2)C6H4-1-CHvNC6H4-2-(PPh2)), where the
diphenylphosphino groups have different substituents. The reac-
tions of the η1-allyl complexes with diethylamine and piperidine
have also been studied in order to get a better understanding of
aDipartimento di Scienze e Tecnologie Chimiche, Università di Roma
“Tor Vergata”, Via delle Ricerca Scientifica, 00133 Roma, Italy.
E-mail: crociani@stc.uniroma2.it; Fax: +39 06 72594328
bDipartimento di Energetica, Università di Firenze, Via S. Marta 3,
50139 Firenze, Italy
†Electronic supplementary information (ESI) available. CCDC 866113.
For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c2dt30746d
12490 | Dalton Trans., 2012, 41, 12490–12500
This journal is © The Royal Society of Chemistry 2012

View Article Online
the factors governing the mechanism and regioselectivity of
amination.
the low-frequency shift of the ν(CvN) band (17–20 cm−1) rela-
tive to the free ligand [ν(CvN) = 1621 cm−1] in the IR spectra,
and by the appearance of the 31P signals as AB quartets in the
range 26–34 ppm with 2J(P–P) values of 312–362 Hz, which are
typical of the non-equivalent 31P nuclei in the trans position
(Table 1).7 The PNP coordination is also confirmed by the X-ray
structural analysis of complex 1 (see further).
Results and discussion
Preparation and solution behavior of the η1-allyl complexes
The η1-allylpalladium complexes 1–4 can be prepared from the
reaction of the iminodiphosphine ligand P–N–P′ with the η3-
allyl dimers [Pd(μ-Cl)(η3-CHR1–CHvCR2R3)]2 (R1 = R2 = R3
= H; R1 = R2 = Ph, R3 = H; R1 = R3 = H, R2 = Ph; R1 = H, R2 =
R3 = Me) (molar ratio P–N–P/Pd = 1 : 1) in the presence of
sodium tetrafluoroborate as reported in Scheme 1.
The complexes 1–4 have been isolated in high yields and
characterized by conductivity measurements, elemental analysis,
IR, 1H and 31P NMR spectroscopy (see the Experimental
section). The tridentate coordination of the ligand through the
imino nitrogen and the phosphorus atoms is clearly indicated by
Complexes 1 and 2 are fluxional in solution as they undergo a
rapid (on the NMR time scale) η1–η3–η1 exchange process
which interconverts the four terminal allyl protons of 1 and the
1
two terminal allyl protons of 2. As can be seen in the H NMR
spectra of 1 in CD2Cl2 in the range 2–6 ppm (Fig. 1) at 298 K,
the terminal allyl protons appear as a broad band centered at
3.30 ppm and the central proton as a quintet at 5.45 ppm. At
213 K, the exchange rate is considerably reduced and four
signals are observed for the η1-allyl group at chemical shifts that
one would expect by comparison with η1-allyl palladium com-
plexes reported in the literature.1,2 In particular, the Pd–CH2
Scheme 1
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12490–12500 | 12491

View Article Online
Table 1 Proton resonances of the allyl ligand in complexes 1–4,a and 31P NMR data
Complex
T/K Pd–CHR1
vCR2R3
3.30(br)
vCH
31Pb
1 (R1 = R2 = R3 = H)
298
243
5.45
27.1d, 32.2d(br) (332)c
26.2d, 34.2d (362)c
213 2.34ddd (5.5)f, (6.7)f, (6.7)g
3.74dd (16.7)g, 4.22de (9.9)g 5.29m
5.40(br)
2 (R1 = R2 = Ph; R3 = H) 298
mkh
mkh
22.0d(br), 26.0d(br) (312)c
213 4.45ddd (5.2)f, (14.0)f, (14.0)g 5.91d (14.0)g
3 (R1 = R3 = H; R2 = Ph) 298 2.70ddd (6.0)f, (6.0)f, (8.4)g
4 (R1 = H; R2 = R3 = Me) 298 2.49ddd (5.9)f, (5.9)f, (8.5)g
5.23d (15.6)g
0.83s, 1.19s
5.84 dt (15.6)g, (8.4)g 23.1d, 31.5d (350)c
4.85t (8.5)g
24.8d, 28.5d (349)c
a In CD2Cl2, chemical shifts in ppm and coupling constants in Hz. b AB system. c 3J(P–P). d Signal of proton anti to the central allylic proton. e Signal
of proton syn to the central allyl proton. f 3J(P–H). g 3J(H–H). h Masked by the allyl proton resonance.
resonance appears as a rather broad multiplet due to coupling
with the two non-equivalent phosphorus atoms and with the
central allyl proton. Actually, this band consists of two overlap-
ping triplets resulting from one of the 3J(P–H) coupling con-
stants being equal to 3J(H–H) (Table 1).
The dynamic behavior of complex 1 was further examined by
31P NMR spectra at different temperatures (Fig. 2).
At 243 K, when the η1-allyl form predominates, a rather sharp
AB quartet is observed. On raising the temperature in the range
260–300 K, the two singlets at lower field progressively broaden
and then become again sharper at 308 K, and even more at
higher temperatures, while the two singlets at higher field remain
almost unchanged. This trend can be rationalized in terms of a
temperature-dependent equilibrium between the η1-allyl complex
and an η3-allyl intermediate, the latter being characterized by a
31P AB quartet with the two high-field singlets almost coincident
with those of the η1-allyl species and the two low-field singlets
rather close to those of the η1-allyl species. It is interesting to
note that under conditions of fast exchange (at temperatures
higher than 338 K) a 2J(P–P) coupling constant of 338 Hz is still
detected which indicates that no Pd–P bond breaking occurs in
Fig. 2 31P NMR spectra of [Pd(η1-C3H5)(P–N–P′)]BF4 (1) at different
temperatures (range 243–308 K: solvent CD2Cl2; 338 K: solvent
Fig. 1 1H NMR spectra (allyl region) of [Pd(η1-C3H5)(P–N–P′)]BF4
(1) in CD2Cl2 at different temperatures.
DMSO-d6).
12492 | Dalton Trans., 2012, 41, 12490–12500
This journal is © The Royal Society of Chemistry 2012

View Article Online
the dynamic process and that a trans P–Pd–P structure is present
in the η3-allyl intermediate.
On raising the temperature to 290 K, the two signals coalesce
into a single absorption at 5.4 ppm. From these data, a ΔG≠
value of 53.8 kJ mol−1 can be estimated for the dynamic
process.8
The occurrence of AB spin systems (due to the two non-
equivalent phosphorus atoms linked to the same metal center) is
confirmed by a 2D COSY experiment on the 31P NMR spectrum
of 1 in DMSO-d6 at 318 K which shows correlation peaks
between the signals of the AB quartet, namely between the two
higher field signals at 22.72 and 25.53 ppm, between the two
lower field signals at 29.03 and 31.84 ppm, between the signal at
22.72 ppm and that at 31.84 ppm, and between the signal at
25.53 ppm and that at 29.03 ppm.
On the basis of the above 1H NMR data we propose the mech-
anism reported in Scheme 2 for the η1–η3–η1 exchange process
which interconverts the bonding site of the allyl ligand. The
η3-intermediate is formulated as a five-coordinate species with
the phosphine ligands in the axial position and with the imino
nitrogen and the terminal allyl carbons on the equatorial plane.
Such an intermediate, however, is never observed even when the
temperature is lowered suggesting an equilibrium shift toward
the η1 bonding mode at low temperatures, which is the bonding
mode found in the solid for complex 1 according to the X-ray
structural data.
1
In the H NMR spectrum of 2 in CD2Cl2 at 213 K (Table 1),
the terminal allyl proton CHPh resonates as a doublet at
5.91 ppm (with a 3J(H–H) of 14.0 Hz typical of a trans-
CHvCH fragment), while the Pd–CHPh proton resonates as a
multiplet at 4.45 ppm resulting from the overlap of three doub-
In contrast to the dynamic behavior of 1 and 2, complexes 3
3
3
3
1
lets (with J(H–H) = J(P–H) = 14.0 Hz and J(P–H) = 5.2 Hz).
and 4 exhibit sharp H and 31P NMR spectra at 298 K, which
Scheme 2
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12490–12500 | 12493

View Article Online
are typical of “frozen” η1-allyl species with the allyl group
bound to the central metal through the unsubstituted terminal
carbon atom.2b,d,i In both cases, the Pd–CH2 proton resonance
Table 2 Most significant geometrical parameters (Å, °) defining the
coordination geometry of the palladium atom in complexes 1I, 3I and
4I, as derived from DFT calculations. For comparative purposes,
experimental data derived from X-ray single crystal data of 1 are also
reported (for atom labelling refer to Scheme 2)
3
shows a J(H–H) coupling with the central allyl proton and two
equal 3J(P–H) couplings with the cis phosphorus atoms
(Table 1). In addition, the 3J(H–H) value of 15.6 Hz between the
central and the terminal allyl protons is indicative of their trans
position in the CHvCHPh unit of 3. In order to check the occur-
rence of equilibria of Scheme 2 also for these complexes with
asymmetrically substituted allyl ligands, we have recorded the
1H NMR spectra of 4 in DMSO-d6 in the temperature range
298–390 K. On increasing the temperature above 350 K, the two
sharp singlets at 1.10 and 0.80 ppm for the methyl protons of the
allyl ligand progressively broaden. Even though the coalescence
was not complete due to decomposition at higher temperature,
the change in shape of the signals suggests a slow exchange of
the methyl groups which can take place in the isomeric structure
II of Scheme 2, when the allyl ligand is η1-bound to the metal
through the more substituted carbon atom. An analogous inter-
conversion of the methyl groups through η1–η3–η1 rearrange-
ment was earlier reported for the complex [Pd(η1-CH2–
CHvCMe2)(PNP)]BF4 (PNP = 1,2-bis[(di-phenylphosphino)-
methyl]pyridine).2b On the basis of the above data, however, the
equilibria of Scheme 2 for 4 (and presumably for 3) involve
undetectable amounts of the η3-intermediate and of the isomer II.
With the results from the solution studies on complexes 3 and
4 in mind, we set up a DFT modelling study aiming to evaluate
the relative energy contents of the corresponding unsymmetrical
isomers I and II which are involved in the equilibria of
Scheme 2.
Complex
1I
[Pd(η1-C3H5)(P–N–P′)]+ (1) 3I
4I
Pd–P
Pd–N
Pd–P′
2.362 2.268(2)
2.233 2.156(7)
2.363 2.304(2)
2.098 2.065(8)
2.364 2.365
2.245 2.251
2.369 2.372
2.107 2.106
Pd–CHR1
P–Pd–N
P–Pd–P′
N–Pd–P′
P–Pd–CHR1
80.3
83.0(2)
80.2
80.3
166.9 171.47(9)
166.3 166.4
87.1
96.3
90.0(2)
91.4(2)
95.7(2)
86.4
96.4
97.0
86.6
96.9
96.2
P′–Pd–CHR1 96.3
N–Pd–CHR1
176.6 173.9(3)
176.5 176.7
Fig. 3 Ball and stick representations of the DFT optimized complexes:
3I [Pd(η1-CH2–CHvCHPh)(P–N–P′)]+ (left) and 4I [Pd(η1-CH2–
CHvCMe2)(P–N–P′)]+ (right).
First of all, the suitability of the model chemistry adopted (see
the Experimental section) was checked by comparing the experi-
mental X-ray structure of [Pd(η1-C3H5)(P–N–P′)]+ (1) (see later)
with the optimized one 1I. The overall shape of the latter is
consistent with the corresponding experimental one, thus
confirming the reliability of the model chemistry. In particular,
the experimental 3D arrangement of the η1-bound allyl ligand
and the palladium coordination sphere is properly reproduced in
the corresponding optimized structure (1 vs. 1I in Table 2) as
well as the relative orientation of the PPh2 aromatic rings.
Importantly, although the Pd–donor atom bond distances are
invariably slightly longer in comparison with the experimental
values (Table 2), the allyl anion is doubtlessly η1-coordinated. In
addition, the angle between the allyl plane and the mean plane
described by the palladium centre with its four donor atoms well
compares with the experimental one (66.4 vs. 66.1(7)°).
As for the complexes 3 and 4, both isomers I (Fig. 3),
corresponding to a Pd–CH2 η1-coordination mode, are signi-
ficantly more stable with respect to isomers II, the latter being
+48.78 and +69.35 kJ mol−1 higher in energy, respectively,
in agreement with the “frozen” η1-allyl species observed in
the NMR spectra. This regioselectivity could be ascribed to
obvious steric reasons (CH2 vs. CHPh and CMe2) and also
to electronic factors, the terminal CH2 carbon atom being
more negatively charged with respect to the CHPh and CMe2
allyl carbons in the corresponding unbound allyl anions
(Table 3), as estimated by the Mullikan and Merz–Singh–
Kollman charge derivation scheme (implemented in the
Gaussian package).9
Within each couple of isomers (3I and 3II, 4I and 4II), while
the overall architecture of the complex is almost identical (data
provided in the ESI†), in the I species the palladium–donor atom
bond distances are always a little bit shorter than in the corres-
ponding II ones. Finally, as expected, there are not significant
differences in the palladium coordination sphere of the com-
plexes 1I, 3I and 4I given that in all cases the metal is η1-bound
to a CH2 moiety (Table 2).
Solid state molecular structure of complex [Pd(η1-C3H5)-
(P–N–P′)]BF4·0.25Et2O
In the asymmetric unit of [Pd(η1-C3H5)(P–N–P′)]BF4·0.25Et2O
−
there is one complex cation 1, one BF4 counterion and a
quarter of a diethyl ether molecule. The palladium ion in the
[Pd(η1-C3H5)(P–N–P′)]+ complex is four-coordinated by two
trans disposed phosphorus atoms, by the nitrogen atom of the
tridentate ligand and by the η1-allyl group (Fig. 4).
The coordination geometry is close to square-planar with the
maximum deviation from the mean plane defined by the four
donor atoms and the palladium atom of 0.123(8) Å (N(1)).
Table 4 lists selected bond distances and angles in the [Pd(η1-
C3H5)(P–N–P′)]+ cation.
In order to compare the geometrical data observed in 1 with
similar metal complexes, a literature survey was made. 13 com-
plexes containing an η1-allylpalladium group were retrieved and
of these just two2b,h contain a PNP type ligand. More in general,
only 4 metal complexes were found which feature a ligand
12494 | Dalton Trans., 2012, 41, 12490–12500
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 3 Atomic charges (e) calculated by using the Mullikan (M) and Merz–Singh–Kollman (MSK) schemes (HF/6-31G* and B3LYP/6-31G*
model chemistries)
HF/6-31G*
B3LYP/6-31G*
Isolat−ed allyl anion
C3H5
CR2R3 (M/MSK)
−0.623/−1.188
−0.362/−0.760
−0.405/−0.837
−0.121/−0.296
CH (M/MSK)
+0.030/+0.525
−0.094/+0.282
−0.045/+0.279
−0.039/−0.012
CHR1 (M/MSK)
−0.623/−1.188
−0.362/−0.760
−0.541/−0.874
−0.624/−0.994
CR2R3 (M/MSK)
−0.505/−1.041
−0.276/−0.489
−0.307/−0.631
+0.101/−0.255
CH (M/MSK)
+0.037/+0.394
−0.090/+0.067
−0.027/+0.144
−0.068/−0.042
CHR1 (M/MSK)
−0.505/−1.041
−0.276/−0.489
−0.447/−0.740
−0.501/−0.858
PhCHCHCHPh−
−
CHPhCHCH2−
Me2CCHCH2
[Pd(η1-allyl(P–N–P′)]+ (most stable isomer)
CR2R3 (M)
−0.352
CH (M)
−0.016
−0.053
−0.070
−0.126
CHR1 (M)
−0.404
−0.338
−0.406
−0.400
C3H5
—
—
—
—
—
—
—
—
—
—
—
—
PhCHCHCHPh
CHPhCHCH2
Me2CCHCH2
−0.199
−0.187
+0.185
Fig. 4 ORTEP3 view of the cation [Pd(η1-C3H5)(P–N–P′)]+. The ellipsoid probability was set to 30%. The atom labelling scheme used in the discus-
sion of the X-ray structure is analogous to that used in ref. 2i.
having the same atomic sequence of the P–N–P′ ligand reported
in the present work (i.e. P–C–C–N–C–C–C–P, – being any
type of bond), and only one of them has palladium as the
metal centre,10a while all the four complexes contain the same
P–N–P′-ligand used in this work.7,10
Bond distances and angles around the metal ion in 1 well
compare with those already reported,11 including the Pd–C(38)
distance. In particular, as found in related compounds, the
M–P(1) distance (where M is the coordinated metal atom and
P(1) belongs to the five-membered chelate ring) is usually
shorter than the M–P(2) one (where P(2) is part of the six-mem-
bered chelate ring); the P–M–N angle belonging to the six-mem-
bered ring is always larger than that of the five-membered one,
and finally the N–M–X and P–M–P angles are slightly less than
180°.10a The plane containing the allyl ligand forms an angle of
66.1(7)° with respect to the mean plane formed by the four
donor atoms and the palladium centre. As for the five- and six-
membered chelate rings, the non-donor atoms in the six-
membered ring lie on the opposite side with respect to the
non-donor atoms of the five-membered ring. Finally, the
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12490–12500 | 12495

View Article Online
Allyl amination and characterization of [Pd(η2-fumaronitrile)-
(P–N–P′)]
Table 4 Selected bond distances (Å) and angles (°) for the cation [Pd-
(η1-C3H5)(P–N–P′]+ as derived from the X-ray diffraction data
Pd(1)–N(1)
Pd(1)–P(1)
Pd(1)–P(2)
Pd(1)–C(38)
N(1)–Pd(1)–P(1)
N(1)–Pd(1)–P(2)
N(1)–Pd(1)–C(38)
P(1)–Pd(1)–P(2)
P(1)–Pd(1)–C(38)
P(2)–Pd(1)–C(38)
2.156(7)
2.268(2)
2.304(2)
2.065(8)
83.0(2)
The complexes 1–4 react with secondary amines (diethylamine
or piperidine) in the presence of fumaronitrile (fn) yielding the
palladium(0) derivative
(Scheme 3).
5 and allylamines 6a and/or 6b
90.0(2)
We have followed the course of the reaction and analyzed the
products by means of 1H and 31P NMR spectroscopy upon
mixing the reactants in a molar ratio complex/R2NH/fn of
1 : 5 : 1.2 at 298 K. The formation of 5 can be easily detected by
its characteristic δ(31P) signals (see further), while the formation
of 6a and/or 6b is indicated by the typical proton resonances of
the allyl group. The results are summarized in Table 5.
173.9(3)
171.47(9)
91.4(2)
95.7(2)
conformations adopted by the two rings can be described as
envelope (E5 in the series N(1), C(18), C(13), P(1), Pd(1)) and
twist boat (TB2,5 in the series P(2), C(25), C(20), C(19), N(1),
Pd(1)), for the five and six-membered rings, respectively.12 The
C(18)–N(1)–C(19)–C(20) fragment, which connects the two aro-
matic rings of the P–N–P′ ligand, lies on a plane and forms
angles of 25.2(7) and 15.7(6)° with the C(13)–C(18) and
C(20)–C(25) aromatic mean planes, respectively, while the two
aromatic rings form with each other an angle of 40.9(3)°. The
diphenylphosphino C(1)–C(6) and C(7)–C(12) aromatic ring
mean planes form an angle of 66.3(3)°, while the other couple of
phenyl rings (C(26)–C(31)/C(32)–C(37)) forms an angle of
76.0(3)°. Finally, the C(1)–C(6) and C(26)–C(31) rings are
facing each other (they form an angle of 11.6(3)°) and the dis-
tance between their geometrical centroid (calculated using only
the carbon atoms) is 8.41 Å. A C⋯H π interaction exists
between the allyl group and the aromatic ring of one diphenyl-
phosphino moiety: the distance of H(40a) from the centroid of
the C(1)–C(6) ring (Ct, in the following) is 2.946(3) Å, while
the angle C(40)–H(40a)⋯Ct is 140.7(7)°. Finally, in the crystal
packing, due to the presence of π–π interactions between two
symmetry related C(20)–C(25) aromatic rings (the symmetry
operation is –x + 1, −y, −z + 1), dimers of the palladium
complex are formed, the distance between the centroids of the
rings being 3.88(1) Å.13
From the completion times of the reactions with diethylamine
it appears that complexes 1 and 2 react faster than 3 and 4. A
high selectivity occurs in the reactions of 3 with both amines
(yielding the regioisomer 6a) and of 4 with piperidine (yielding
the regioisomer 6b). According to literature reports, the observed
regioselectivity would imply a nucleophilic attack at the
Pd–CHR1 carbon atom (SN2 mechanism) for 3 and at the
terminal vCR2R3 carbon atom (SN2′ mechanism) for 4,
respectively.1i,2j,3 In order to get a better understanding of the
differences in rate and regioselectivity, we have calculated the
charge distribution on the allyl carbon atoms of complexes 1–4
(Table 3).
As one can see, in all cases the Pd–CHR1 allyl carbon bears
the highest negative charge which makes it more prone to an
electrophilic attack rather than to a nucleophilic one. In fact, all
the complexes undergo protonolysis of the Pd–CHR1 bond by
trace amounts of hydrochloric acid present in CDCl3 yielding the
cationic species [PdCl(P–N–P′)]+.7 Thus, within the reliability of
the data in Table 3, amination through an SN2 mechanism can
hardly take place by considering also the large steric require-
ments of the Pd(P–N–P′) moiety. A partial negative charge is
also present on the terminal allyl carbon vCR2R3 of complexes
1–3, which renders unlikely a nucleophilic attack at this carbon
atom through an SN2′ mechanism. A partial positive charge,
Scheme 3
12496 | Dalton Trans., 2012, 41, 12490–12500
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 5 Allyl amination dataa
Complex
R2NH
Completion time
Allylamineb
1
2
3
4
3
Et2NH
Et2NH
Et2NH
Et2NH
ca. 1 h
6a
6a
ca. 4 h c
ca. 9 h
6a
Very slow
ca. 4 h
6a + 6bd
6a
4
ca. 12 h
6b
Fig. 5 Proposed structure for the complex [Pd(η2-fumaronitrile)-
(P–N–P′)] (5).
a Experimental condition: complex/R2NH/fn molar ratio of 1 : 5 : 1.2 in
CDCl3 at 298 K. b Predominant product (>98%) of allylation, within the
limit of integration of the 1H NMR signals. c In CD2Cl2. d Molar ratio
6a/6b of 9 : 1. This ratio remains almost constant during the course of
the slow reaction, which is not complete even after 4 days.
fumaronitrile). The 1H NMR spectrum in CD2Cl2 at 298 K
shows the presence of a broad singlet at 2.85 ppm for the
protons of the η2-bound olefin and of a small singlet at 6.33 ppm
for the protons of the free olefin with an integration ratio of
16 : 1. On cooling to 268 K, the proton signal of the coordinated
olefin appears as two doublets of doublets of doublets centered
however, is present on the terminal allyl carbon vCR2R3 (R2 =
R3 = Me) of 4, which accounts for the regioselectivity observed
in its reaction with piperidine (SN2′ mechanism). This finding
parallels that reported by Åkermark for the reaction of dimethyl-
3
3
at 2.94 ppm [3J(H–H) = J(P–H) = 9.0 Hz, J(P–H) = 3.7 Hz]
and 2.83 ppm [3J(H–H) = 3J(P–H) = 8.9 Hz, 3J(P–H) = 2.9 Hz],
respectively. The observed spectral change and the presence of
non-equivalent olefin protons at low temperature may be
accounted for by a dissociation–association equilibrium of the
olefin which occupies a coordination site at a corner of a dis-
torted tetrahedral structure as depicted in Fig. 5.
amine with
a
mixture [PdCl(η3-CH2–CHvCMe2)]2/PPh3
(1 : 4 molar ratio).3 The higher reaction rates of the reactions of 1
and 2 with diethylamine (as compared to the corresponding reac-
tions of 3) and the regioselectivity in the reactions of 3 are better
interpreted in terms of a nucleophilic attack occurring at the η3-
allyl intermediate of Scheme 2, which is present in higher con-
1
centration in the solutions of 1 and 2 at 298 K according to H
NMR data. In literature data, the amination of cationic η3-allyl-
Conclusion
palladium complexes (allyl
= CH2–CHvCMe2 or CH2–
In the cationic complexes 1–4, the iminodiphosphine 2-(PPh2)-
C6H4-1-CHvNC6H4-2-(PPh2) (P–N–P′) acts as a tridentate
CHvCHPh) was found to involve the less substituted allyl ter-
minus.3,14 The steric requirements of the entering amine play an
important role in the rates and regioselectivity to such an extent
that the reaction of 4 with the more sterically hindered diethyl-
amine is extremely slow and yields both regioisomers 6a and 6b
in a molar ratio of 9 : 1. In this case, the rate of attack at the ter-
minalvCMe2 carbon of the η1-allyl ligand (yielding 6b) is con-
siderably depressed by steric repulsion of the substituents, and
becomes comparable (or even lower) to the rate of attack at trace
amounts of the η3-allyl intermediate (yielding 6a).
The complex 5 has been isolated in high yield and charac-
terized by elemental analysis, IR, 1H and 31P NMR spectroscopy
(see the Experimental section). It appears to be an 18 electron
palladium(0) derivative due to the tridentate coordination of the
P–N–P′ ligand and the η2-bonding mode of the olefin. As a
matter of fact, the 31P resonances in CDCl3 are detected as two
doublets at 27.6 and 12.0 ppm, respectively, with a 2J(P–P)
coupling of 18.7 Hz (which rules out a trans-P–Pd–P structure)
while the imino proton signal is observed at 8.21 ppm (0.78 ppm
upfield relative to the free ligand). Correspondingly, in the IR
spectrum the ν(CvN) vibration at 1581 cm−1 is markedly
shifted to lower frequency as compared to the free ligand. The
spectral change of the CHvN unit suggests extensive d → π*
back donation upon N-coordination to the central metal. The
presence of the coordinated olefin is further confirmed by the
observation of a ν(CuN) band at 2204 cm−1 in the IR spectrum
(cf. the corresponding vibration at 2244 cm−1 of the free
1
ligand as indicated by H and 31P NMR data in solution, and by
the X-ray structural analysis of 1 in the solid. The solution be-
havior of 1 and 2 (η1–η3–η1 process), as well as the amination
data with secondary amines in the presence of fumaronitrile
(difference in rate between 1 and 3, and regioselectivity for
complex 3), are better rationalized on the basis of a penta-
coordinate η3-intermediate with the phosphino groups in the
axial position. In agreement with atomic charge calculation, the
nucleophilic attack of piperidine occurs at the more substituted
allyl carbon of the η1-CH2–CHvCMe2 ligand of complex 4.
In addition to allylamines, the amination reactions yield the
palladium(0) derivative 5 containing the tridentate P–N–P′
ligand and the η2-bound fumaronitrile.
Experimental
1H NMR spectra were recorded on Bruker AM400 or Bruker
Avance 300 spectrometers operating at 400.13 and 300.13 MHz,
respectively. 31P{1H} NMR spectra were recorded on a Bruker
Avance 300 spectrometer operating at 121.49 MHz. Chemical
shifts are reported in ppm downfield from SiMe4 for 1H and
from H3PO4 as an external standard for 31P. The spectra were run
at 25 °C except when noted. IR spectra were recorded on a
Perkin-Elmer Spectrum One FT-IR spectrometer. All the reac-
tions were carried out under N2. The solvents and the
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12490–12500 | 12497

View Article Online
3
3
commercially available chemicals, such as sodium tetrafluoro-
borate and fumaronitrile, were used without further purification.
Diethylamine and piperidine were distilled over anhydrous
K2CO3 under N2. The iminodiphosphine P–N–P′ and the com-
plexes [Pd(μ-Cl)(η3-CHR1–CHvCR2R3)]2 (R1 = R2 = R3 = H;
R1 = R2 = Ph, R3 = H; R1 = R3 = H, R2 = Ph; R1 = H, R2 = R3 =
Me) were prepared by literature methods.1a,15,16
3J(H–H) = 8.5 Hz, J(P–H) = J(P–H) 5.9 Hz, Pd–CH2), 4.85
(1 H, t,vCH, central allylic proton), 7.2–7.9 (27 H, m, aryl
protons), 8.2–8.3 (1 H, m, aryl proton), 9.09 (1H, s, NvCH).
Preparation of [Pd(η2-fn)(P–N–P′)]·CH2Cl2
Diethylamine (0.183 g, 2.5 mmol) was added to a solution of
complex 1 (0.392 g, 0.5 mmol) and fumaronitrile (0.047 g,
0.6 mmol) in CH2Cl2 (20 cm3) under N2. After stirring for 2 h at
room temperature, the mixture was treated with H2O (2 ×
10 cm3). The organic phase was dried over anhydrous Na2SO4,
concentrated to a small volume (ca. 5 cm3) and diluted with
Et2O to precipitate the product as a yellow-brown solid, which
was purified by further precipitation from CH2Cl2–Et2O. As
Preparation of [Pd(η1-CHR1–CHvCR2R3)(P–N–P′)]BF4 (1–4)
A solution of NaBF4 (0.066 g, 0.6 mmol) in MeOH (5 cm3) was
added to a solution of P–N–P′ (0.275 g, 0.5 mmol) and [Pd-
(μ-Cl)(η3-CHR1–CHvCR2R3)]2 (0.25 mmol) in CH2Cl2
(20 cm3). The mixture was stirred for 2 h at room temperature
and the solvent was evaporated to dryness at reduced pressure.
The solid residue was extracted with CH2Cl2 (20 cm3). After
addition of activated charcoal and filtration, the clear solution
was concentrated to a small volume (ca. 3 cm3) and diluted with
Et2O to precipitate the solid products with colours ranging from
yellow to purple. The complexes were purified by a further pre-
cipitation from CH2Cl2–Et2O.
1
shown by the H NMR spectrum in acetone-d6 and elemental
analysis, the complex contains a CH2Cl2 molecule of crystalliza-
tion (0.310 g, 76%) (found: C 61.33, H 4.05, N 5.07;
C42H33Cl2N3P2Pd requires C 61.59, H 4.06, N 5.13%);
νmax (Nujol)/cm−1 2204 (CuN), 1581 (CvN); δH (300 MHz,
CD2Cl2, 298 K) 2.85 (2 H, s, br, olefin protons), 6.7–6.8 (1 H,
m, aryl proton), 6.9–7.0 (2 H, m, aryl protons), 7.0–7.1 (2 H, m,
aryl protons), 7.1–7.6 (24 H, m, aryl protons), 7.7–7.8 (1 H, m,
aryl proton), 8.20 (1 H, s, NvCH); δH (300 MHz, acetone-d6,
298 K) 5.64 (2 H, s, CH2Cl2); δp (300 MHz, CD2Cl2, 298 K)
11.7 d, 27.7 d, 2J(P–P) = 16.7 Hz.
Complex 1 (R1 = R2 = R3 = H): yellow solid (0.370 g, 94%)
(found C 60.90, H 4.25, N 1.72; C40H34BF4NP2Pd requires
C 61.29, H 4.37, N 1.79%); ΛM 51.1 S cm2 mol−1 for a 1 ×
10−3 mol dm−3 solution in CH2Cl2 at 298 K; νmax (Nujol)/cm−1
1603 (CvN), 1058 (B–F); δH (400 MHz, CD2Cl2, 213 K) 2.34
(2 H, ddd, 3J(H–H) = 3J(P–H) = 6.7 Hz, 3J(P–H) = 5.5 Hz,
3
Pd–CH2), 3.74 (1H, d, J(H–H) = 16.7 Hz,vCH2 proton trans
Identification of allylamines 6a and 6b
to the central allylic proton), 4.22 (1H, d, 3J(H–H) = 9.9
Hz,vCH2 proton cis to the central allylic proton), 5.29 (1H,
m,vCH, central allylic proton), 7.2–8.0 (28 H, m, aryl protons),
8.72 (1H, s, NvCH).
The allylamines 6a and 6b formed in the reaction of complexes
1–4 with secondary amines (diethylamine or piperidine) in the
presence of fumaronitrile were detected and characterized by the
proton resonances of the allyl group in the 1H NMR spectra and,
in some cases, by GC-MS analysis of the reaction mixture.
Complex 2 (R1 = R2 = Ph, R3 = H): purple solid (0.412 g,
88%) (found C 66.42, H 4.40, N 1.45; C52H42BF4NP2Pd
requires C 66.72, H 4.52, N 1.50%); ΛM 54.1 S cm2 mol−1 for a
1 × 10−3 mol dm−3 solution in CH2Cl2 at 298 K; νmax
(Nujol)/cm−1 1601 (CvN), 1058 (B–F); δH (400 MHz, CD2Cl2,
Et2N–CH2–CHvCH2
(3-diethylaminopropene)
δH
3
(300 MHz, CDCl3, 298 K): 3.12 (2 H, d, J(H–H) = 6.5 Hz,
3
N–CH2), 5.14 (1 H, d, J(H–H) = 10.9 Hz,vCH2 proton cis to
3
3
3
213 K) 4.45 (1 H, ddd, J(H–H) = J(P–H) = 14.0 Hz, J(P–H)
the central allylic proton), 5.19 (1H, d, 3J(H–H) = 18.1 Hz,
vCH2 proton trans to the central allylic proton), 5.89 (1H,
m,vCH, central allylic proton); MS data: m/z 113 (M+, 9%),
98 (68, M − CH3), 86 (22, M − CHvCH2), 84 (5, M − C2H5),
72 (3, M − CH2CHvCH2).
3
= 5.2 Hz, Pd–CH2), 5.91 (1H, d, J(H–H) = 14.0 Hz,vCHPh,
proton trans to the central allylic proton), 6.4–8.0 (39 H, m,
central allylic proton and aryl protons), 8.45 (1H, s, NvCH).
Complex 3 (R1 = R3 = H, R2 = Ph): orange solid (0.414 g,
96%) (found C 64.21, H 4.28, N 1.57; C46H38BF4NP2Pd
requires C 64.24, H 4.45, N 1.63%); ΛM 51.9 S cm2 mol−1 for a
1 × 10−3 mol dm−3 solution in CH2Cl2 at 298 K; νmax
(Nujol)/cm−1 1604 (CvN), 1064 (B–F); δH (300 MHz, CD2Cl2,
Et2N–CHPh–CHvCHPh (trans-3-diethylamino-1,3-diphenyl-
3
propene) δH (300 MHz, CDCl3, 298 K): 4.31 (1H, d, J(H–H)
3
3
= 8.8 Hz, N–CHPh), 6.37 (1 H, dd, J(H–H) = 8.8, J(H–H) =
15.8,vCH, central allylic proton), 6.56 (1 H, d, 3J(H–H) =
15.8,vCHPh, proton trans to the central allylic proton).
3
3
298 K) 2.70 (2 H, ddd, J(H–H) = 8.4 Hz, J(P–H) = 3J(P–H) =
3
6.0 Hz, Pd–CH2), 5.23 (1 H, d, J(H–H) = 15.6 Hz,vCHPh,
Et2N–CH2–CHvCHPh (trans-3-diethylamino-1-phenylpro-
proton trans to the central allylic proton), 5.84 (1H, dt,vCH,
central allylic proton), 6.7–6.9 (2 H, m, aryl protons), 7.0–7.3
(3 H, m, aryl protons), 7.4–8.3 (28 H, m, aryl protons), 8.86
(1H, s, NvCH).
pene) δH (300 MHz, CDCl3, 298 K): 3.28 (2 H, d, 3J(H–H)
3
3
= 6.7 Hz, N–CH2), 6.32 (1 H, dt, J(H–H) = 6.7 Hz, J(H–H) =
3
15.9 Hz,vCH, central allylic proton), 6.53 (1 H, d, J(H–H) =
15.9 Hz,vCHPh, proton trans to the central allylic proton); MS
data: m/z 189 (M+, 8%), 174 (6, M − CH3), 160 (4, M − C2H5),
117 (66, M − N(C2H5)2), 72 (2, M − CH2CHvCHC6H5).
Et2N–CH2–CHvCMe2 (4-diethylamino-2-methyl-2-butene)
δH (300 MHz, CDCl3, 298 K): 1.68 (s, CH3), 1.76 (s, CH3),
2.86 (d, 3J(H–H) = 7.9 Hz, N–CH2), 5.28 (t, 3J(H–H) =
7.9 Hz,vCH, central allylic proton); MS data: m/z 141
Complex 4 (R1 = H, R2 = R3 = Me): orange solid (0.340 g,
84%) (found: C 62.07, H 4.70, N 1.75; C42H38BF4NP2Pd
requires C 62.13, H 4.72, N 1.73%); ΛM = 55.2 S cm2 mol−1 for
a 1 × 10−3 mol dm−3 solution in CH2Cl2 at 298 K; νmax
(Nujol)/cm−1 1604 (CvN), 1058 (B–F); δH (300 MHz, CD2Cl2,
298 K) 0.83 (3 H, s, CH3), 1.19 (3 H, s, CH3), 2.49 (2 H, ddd,
12498 | Dalton Trans., 2012, 41, 12490–12500
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 6 Crystallographic data and refinement parameters for [Pd(η1-
C3H5)(P–N–P′)]BF4·0.25(Et2O)
(M+, 14%), 126 (25, M − CH3), 86 (28, M − CHvC(CH3)2),
72 (17, M − CH2CHvC(CH3)2).
Et2N–CMe2–CHvCH2 (3-diethylamino-3-methyl-1-butene)
δH (300 MHz, CDCl3, 298 K): 1.16 (s, CH3), 4.9–5.1
Chemical formula
Mr/mol−1
Crystal size/mm
Crystal system
Space group
a/Å
b/Å
c/Å
α (°)
β (°)
[Pd(η1-C3H5)(P–N–P′)](BF4)·0.25(Et2O)
802.36
0.50 × 0.45 × 0.30
Triclinic
ˉ
P1
11.0898(8)
12.060(1)
15.869(1)
73.145(7)
76.574(6)
81.511(7)
1968.3(2)
2
3
3
(m,vCH2), 5.91 (dd, J(H–H) = 17.6 Hz, J(H–H) = 10.8 Hz,
=CH, central allylic proton).
C5H10N–CH2–CHvCHPh
(trans-1-phenyl-3-piperidino-
propene) δH (300 MHz, CDCl3, 298 K): 3.14 (2 H, d, 3J(H–H) =
6.7 Hz, N–CH2), 6.32 (1 H, dt, 3J(H–H) = 6.7 Hz, 3J(H–H)
= 15.9 Hz,vCH, central allylic proton), 6.52 (1 H, d, 3J(H–H) =
15.9 Hz,vCHPh, proton trans to the central allylic proton); MS
data: m/z 201 (M+, 12%), 117 (44, M − NC5H10), 98
(16 M − CHvCHC6H5), 84 (9, M − CH2 CHvCHC6H5).
C5H10N–CMe2–CHvCH2 (3-methyl-3-piperidino-1-butene)
δH (300 MHz, CDCl3, 298 K): 1.18 (6 H, s, CH3), 5.05 (1 H, d,
3J(H–H) = 17.4 Hz,vCH2, proton trans to the central allylic
γ (°)
U/Å3
Z
Dc/Mg m−3
μ/mm−1
λ/Å
1.6354
0.601
0.71073
T/K
150
3
2θ range (°)
Index range (hkl)
8.4–57.8
−14 to 13
−16 to 15
−21 to 19
15 832/8511
0.932
0.0861/0.1660
0.1795/0.1943
0.921/−0.770
proton), 5.08 (1 H, d, J(H–H) = 11.0 Hz,vCH2 proton cis to
the central allylic proton), 5.94 (1H, dd,vCH, central allylic
proton), cf. the corresponding allyl proton resonances of the
related compound Me2N–CMe2–CHvCH2.3
Reflections collected/unique
Goodness-of-fit on F2
R1, wR2 [I > 2σ(I)]
R1, wR2 [all data]
Δρmax/min/e Å−3
X-ray crystallography
Intensity data for compound [Pd(η1-C3H5)(P–N–P′)]BF4·
0.25(Et2O) were collected on an Oxford Diffraction Excalibur
diffractometer using Mo Kα radiation (λ = 0.71073 Å). The dif-
fractometer was equipped with a cryocooling device which has
been used to set the temperature at 150 K. Data collection was
performed with the program CrysAlis CCD.17 Data reduction
was carried out with the program CrysAlis RED.18 The absorp-
tion correction was applied using the ABSPACK program.19
The structure was solved by using the SIR-97 package20 and
subsequently refined on the F2 values by the full-matrix least-
squares program SHELXL-97.21
modelled species (1I, 3I, 3II and 4I, 4II) were based on X-ray
diffraction data about analogous species, where available.
Acknowledgements
CRIST (Centro di Cristallografia Strutturale), University of Flor-
ence, where the X-ray measurements were performed is grate-
fully acknowledged.
All the non-hydrogen atoms were refined anisotropically with
the exception of those of the diethyl ether molecule, all the
hydrogen atoms of the ligands were set in calculated positions
and refined with an isotropic thermal parameter depending on
the atom to which they are bound.
References
1 (a) J. Powell and B. L. Shaw, J. Chem. Soc. A, 1967, 1839–1851;
(b) S. Numata, R. Okawara and H. Kurosawa, Inorg. Chem., 1977, 16,
1737–1741; (c) H. Kurosawa, A. Urabe, K. Miki and N. Kasai, Organo-
metallics, 1986, 5, 2002–2008; (d) P. K. Byers and A. J. Canty, J. Chem.
Soc., Chem. Commun., 1988, 639–641; (e) R. E. Rülke, D. Kliphuis,
C. J. Elsevier, J. Fraanje, K. Goubitz, P. W. N. M. van Leeuwen and
K. Vrieze, J. Chem. Soc., Chem. Commun., 1994, 1817–1819;
(f) P. Wehman, R. E. Rülke, V. E. Kaasjager, P. C. J. Kamer,
H. Kooijman, A. L. Spek, C. J. Elsevier, K. Vrieze and P. W. N. M. van
Leeuwen, J. Chem. Soc., Chem. Commun., 1995, 331–332;
(g) R. E. Rülke, V. E. Kaasjager, P. Wehman, C. J. Elsevier, P. W. N.
M. van Leeuwen, K. Vrieze, J. Fraanje, K. Goubitz and A. L. Spek,
Organometallics, 1996, 15, 3022–3031; (h) J. Krause, R. Goddard,
R. Mynott and K.-R. Pörschke, Organometallics, 2001, 20, 1992–1999;
(i) T. Cantat, E. Génin, C. Giroud, G. Meyer and A. Jutand, J. Organo-
met. Chem., 2003, 687, 365–376; ( j) N. Solin, J. Kjellgren and
K. J. Szabó, J. Am. Chem. Soc., 2004, 126, 7026–7033.
2 (a) S. Ramdeehul, L. Barloy, J. A. Osborn, A. De Cian and J. Fischer,
Organometallics, 1996, 15, 5442–5444; (b) L. Barloy, S. Ramdeehul,
J. A. Osborn, C. Carlotti, F. Taulelle, A. De Cian and J. Fischer,
Eur. J. Inorg. Chem., 2000, 2523–2532; (c) P. Braunstein, F. Naud,
A. Dedieu, M.-M. Rohmer, A. De Cian and S. J. Retting, Organo-
metallics, 2001, 20, 2966–2981; (d) M. Kollmar and G. Helmchen,
Organometallics, 2002, 21, 4771–4775; (e) P. Braunstein, J. Zhang and
R. Welter, Dalton Trans., 2003, 507–509; (f) B. Goldfuss, T. Löschmann
and F. Rominger, Chem.–Eur. J., 2004, 10, 5422–5431; (g) J. Zhang,
P. Braunstein and R. Welter, Inorg. Chem., 2004, 43, 4172–4177;
(h) J. I. van der Vludt, M. A. Siegler, M. Jansen, D. Vogt and A. Spek,
The diethyl ether non-hydrogen atoms were refined isotropi-
cally with an occupancy factor of 0.25, the hydrogen atoms of
this molecule were not introduced in the refinement.
Geometrical calculations were performed by PARST9722 and
molecular plots were produced by the ORTEP3 program.23 Crys-
tallographic data and refinement parameters are reported in
Table 6.
Computational methods
The Gaussian 09 (revision B.01)24 package was used. All the
studied species were fully optimized by using the density func-
tional theory (DFT) method by means of Becke’s three-para-
meter hybrid method using the LYP correlation functional.25 The
effective core potential of Hay and Wadt26 was used for the pal-
ladium atom. The 6-31G* basis set was used for the remaining
atomic species.27 The reliability of the found stationary points
(minima on the potential energy surface) was assessed by evalu-
ating the vibrational frequencies. Starting geometries for the
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12490–12500 | 12499

View Article Online
Organometallics, 2009, 28, 7025–7032; (i) B. Crociani, S. Antonaroli,
M. Burattini, P. Paoli and P. Rossi, Dalton Trans., 2010, 39, 3665–3672;
( j) L. Canovese, F. Visentin, C. Santo, G. Chessa and V. Bertolasi,
Organometallics, 2010, 29, 3027–3038.
3 B. Åkermark, G. B. Åkermark, L. S. Egedus and K. Zetterberg, J. Am.
Chem. Soc., 1981, 103, 3037–3040.
15 (a) R. Hartley and S. R. Jones, J. Organomet. Chem., 1974, 66, 465–473;
(b) M. C. Ashraf, T. G. Burrowes and W. R. Jackson, Aust. J. Chem.,
1976, 29, 2643–2649.
16 E. W. Ainscough, A. M. Brodie, P. D. Buckley, A. K. Burrell,
S. T. F. Kennedy and J. M. Waters, J. Chem. Soc., Dalton Trans., 2000,
2663–2671.
4 (a) M. García-Iglesias, E. Buñuel and D. J. Cárdenas, Organometallics,
2006, 25, 3611–3618; (b) P. Fristrup, M. Ahlquist, D. Tanner and
P.-O. Norrby, J. Phys. Chem., 2008, 112, 12862–12867.
17 CrysAlis CCD, Version 1.171.pre23_10 beta, Oxford Diffraction Ltd
(release 21.06.2004 CrysAlis171.NET) (compiled June 21, 2004, 12:00:08).
18 CrysAlis RED,, Version 1.171.pre23_10 beta, Oxford Diffraction Ltd
(release 21.06.2004 CrysAlis171.NET) (compiled June 21,
2004,12:00:08).
19 ABSPACK in CrysAlis RED, Version 1.171.29.2, Oxford Diffraction Ltd
(release 20.01.2006 CrysAlis171.NET), (compiled January 20, 2006,
12:36:28).
20 M. C. Burla, M. Camalli, B. Carrozzini, G. L. Cascarano, C. Giacovazzo,
G. Polidori and R. Spagna, J. Appl. Crystallogr., 2003, 36, 1103.
21 G. M. Sheldrick, Acta Crystallogr, Sect. A: Fundam. Crystallogr., 2008,
A64, 112–122.
5 (a) K. Szabó, Chem.–Eur. J., 2000, 6, 4413–4421; (b) O. A. Wallner
and K. Szabó, Org. Lett., 2002, 4, 1563–1566; (c) O. A. Wallner and
K. Szabó, Chem.–Eur. J., 2003, 9, 4025–4230; (d) O. A. Wallner and
K. Szabó, J. Org. Chem., 2002, 68, 2934–2943; (e) N. Solin,
O. A. Wallner and K. Szabó, Org. Lett., 2005, 7, 689–691.
6 (a) B. Crociani, S. Antonaroli, L. Canovese, P. Uguagliati and F. Visentin,
Eur. J. Inorg. Chem., 2004, 732–742; (b) A. Scrivanti, V. Beghetto,
U. Matteoli, S. Antonaroli, A. Marini and B. Crociani, Tetrahedron,
2005, 61, 9752–9758; (c) B. Crociani, S. Antonaroli, A. Marini,
U. Matteoli and A. Scrivanti, Dalton Trans., 2006, 2698–2705;
(d) A. Scrivanti, M. Bertoldini, U. Matteoli, S. Antonaroli and
B. Crociani, Tetrahedron, 2009, 65, 7611–7615.
7 E. W. Ainscough, A. M. Brodie, A. K. Burrell, A. Derwall,
G. B. Jameson and S. K. Taylor, Polyhedron, 2004, 23, 1159–1168.
8 M. Hesse, H. Meier and B. Zech, Metodi Spettroscopici nella Chimica
Organica, EdiSES, Napoli, 5th edn, 1996, ch. 3, p. 99.
9 (a) U. C. Singh and P. A. Kollman, J. Comput. Chem., 1984, 5, 129–145;
(b) B. H. Besler, K. M. Merz Jr. and P. A. Kollman, J. Comput. Chem.,
1990, 11, 431–439.
22 M. Nardelli, J. Appl. Crystallogr., 1995, 28, 659.
23 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
24 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson,
H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov,
J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov,
T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell,
J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam,
M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma,
V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich,
A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and
D. J. Fox, Gaussian, Inc., Wallingford CT, 2010.
10 (a) S. Doherty, J. G. Knight, T. H. Scanlan, M. R. J. Elsegood and
W. J. Clegg, J. Organomet. Chem., 2002, 650, 231–248;
(b) E. W. Ainscough, A. M. Brodie, A. K. Burrell, A. Derwahl and
S. K. Taylor, Inorg. Chim. Acta, 2004, 357, 2379–2384.
11 Structures found in the Cambridge Structural Database (CSD, v. 5.32):
F. H. Allen, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2002, B58,
380–388.
12 D. Cremer and J. A. Pople, J. Am. Chem. Soc., 1975, 97, 1354–1358.
13 C. Janiak, J. Chem. Soc., Dalton Trans., 2000, 3885–3896.
14 B. Crociani, S. Antonaroli, L. Canovese, F. Visentin and P. Uguagliati,
Inorg. Chim. Acta, 2001, 315, 172–182.
25 A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
26 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 299–310.
27 G. A. Petersson, A. Bennett, T. G. Tensfeldt, M. A. Al-Laham,
W. A. Shirlay and J. Mantzaris, J. Chem. Phys., 1988, 89, 2193–2218.
12500 | Dalton Trans., 2012, 41, 12490–12500
This journal is © The Royal Society of Chemistry 2012