Palladium(II) allyl complexes of chiral diphosphazane ligands: ambident coordination behaviour and stereodynamic studies in solution

View Article Online / Journal Homepage / Table of Contents for this issue

Palladium(II) allyl complexes of chiral diphosphazane ligands:

ambident coordination behaviour and stereodynamic studies in

solution†

Swadhin K. Mandal,^aG. A. Nagana Gowda,^bSetharampattu S. Krishnamurthy*^aand

Munirathinam Nethaji^a

^aDepartment of Inorganic and Physical Chemistry, Bangalore-560012, India

^bSophisticated Instruments Facility, Indian Institute of Science, Bangalore-560012, India

Received 30th July 2002, Accepted 16th December 2002

First published as an Advance Article on the web 4th February 2003

The chemistry of η³-allyl palladium complexes of pyrazolyl substituted diphosphazane ligands, Ph₂P(E)N(R)-

PPh(N₂C₃HMe₂-3,5) [E = lone pair, R = CHMe₂(1); E = lone pair, R = (S)-*CHMePh (2); E = S, R = CHMe₂(3)]

bearing a stereogenic phosphorus centre has been investigated and the complexes, [Pd(η³-1,3-RЈ₂-C₃H₃){κ²-Ph₂P(E)-

N(R)PPh(N₂C₃HMe₂-3,5)}](PF₆) [E = lone pair or sulfur; R = CHMe₂or (S)-*CHMePh; RЈ = H, Me or Ph; 4–13],

have been isolated. Detailed NMR studies reveal that these complexes exist as a mixture of isomers in solution.

The structures of four complexes have been determined by X-ray crystallography and only one isomer is observed

in the solid state in each case. The allyl complexes formed by ligands 1 and 2 display a P,N-coordination mode

except the 1,3-diphenyl allyl complex, [Pd(η³-1,3-Ph₂-C₃H₃){κ²-Ph₂PN(CHMe₂)PPh(N₂C₃HMe₂-3,5)}](PF₆) (8),

which shows the presence of both P,N- and P,P-coordinated isomers in solution with the former predominating.

On the other hand, the complexes bearing the diphosphazane monosulﬁde ligand 3 display P,S-coordination.

Two-dimensional phase sensitive ¹H–¹H NOESY and ROESY measurements indicate that several of the above

allyl palladium complexes undergo syn–anti and/or cis–trans isomerization in solution through an η¹-intermediate

formed by the opening of the η³-allyl group selectively at the trans position with respect to the phosphorus centre.

Preliminary investigations show that the diphosphazanes (1 and (SR)-2) function as eﬃcient auxiliary ligands for

catalytic allylic alkylation reactions but lead to only low levels of enantiomeric excess.

Introduction

The chemistry of palladium() allyl complexes is a topic of

considerable contemporary interest in view of the various ﬂux-

ional processes observed in these systems and also from the

point of view of the potential applications of these type of

complexes in asymmetric synthesis.¹There has been a grow-

ing interest in elucidating the structures and stereodynamic

Scheme 1

behaviour of η³-allyl palladium complexes in solution. Multi-

nuclear and multidimensional NOE based NMR spectroscopy

zane ligands, Ph₂PN(R)PPh(N₂C₃HMe₂-3,5) [R = CHMe₂(1)

or R = (S)-*CHMePh(2)] bearing a stereogenic phosphorus

is a valuable tool for this purpose.²

The use of C₂symmetry as a chiral auxiliary ligand design

centre and the diphosphazane monosulﬁde ligand Ph₂P(S)-

N(CHMe₂)PPh(N₂C₃HMe₂-3,5) (3).¹³In this paper, we

element is a well-recognised strategy for restricting the number

of possible diastereomeric transition states.³Equally powerful

report the synthesis, structural characterisation, and dynamic

stereochemical restrictions may also be realised with chiral

behaviour of a range of cationic symmetrically substituted allyl

ligands lacking C₂symmetry through the use of electronic

palladium complexes of these ligands. The main purpose of this

investigation is to make a systematic study of allyl palladium

complexes of diphosphazane class of ligands and to investigate

the roles of phosphorus chirality and trans inﬂuence as control

eﬀects such as trans inﬂuence.⁴Chiral bidentate ligands

incorporating strong and weak donor heteroatom pairs such as

P,N-,⁵P,S-⁶or S,N-⁷have the capacity to inﬂuence both the

stability and reactivity of the diastereomeric intermediates in

elements in asymmetric catalysis involving these ligands.

the catalytic cycle. In the palladium catalysed allylic alkylation

of symmetrically substituted allyl moiety, the regio-selectivity

Experimental

of nucleophilic attack determines the ratio of the enantiomeric

products (Scheme 1).⁸For the mixed-donor ligand systems it is

generally accepted that the nucleophilic attack takes place pref-

erentially at the allyl terminus trans to the more π-accepting

donor centre.⁹However, uncertainty still exists concerning the

inherent reasons leading to such a site selectivity.¹⁰

As a part of our ongoing research on the organometallic

chemistry of diphosphazane ligands,^11,12we have recently

reported the synthesis and dynamic behaviour of unsymmetric-

ally substituted allyl palladium complexes of the diphospha-

Materials and instrumentation

All reactions and manipulations were carried out under

an atmosphere of dry nitrogen using standard Schlenk

and vacuum-line techniques. The solvents were puriﬁed by

standard procedures¹⁴and distilled under nitrogen prior to

use. The chlorobridged palladium allyl dimers, [Pd(η³-C₃H₅)-

16

(µ-Cl)]₂,¹⁵[Pd(η³-1,3-Me₂-C₃H₃)(µ-Cl)]₂and [Pd(η³-1,3-Ph₂-

17

C₃H₃)(µ-Cl)]₂were prepared as previously described. The

diphosphazane ligands 1, 2 and 3 were prepared by published

procedures.^11b,11e,18In the case of 2, the S_CR_P-diastereomer,

absolute conﬁguration of which has been determined by X-ray

crystallography,¹⁸is used for the preparation of allyl palladium

complexes unless speciﬁed otherwise. The NMR spectra were

† Organometallic chemistry of diphosphazanes. Part 17.^11g

Electronic supplementary information (ESI) available: NMR and

crystallographic data, stereodynamic behaviour schemes. See http://

www.rsc.org/suppdata/dt/b2/b207447h/

D a l t o n T r a n s . , 2 0 0 3 , 1 0 1 6 – 1 0 2 7

T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3

1016

View Article Online

Table 1 Yield, melting point and elemental analysis data for η³-allyl palladium complexes 4–13

Elemental analyses^a

Compound

Yield (%)

Mp/ЊC

C (%)

H (%)

N (%)

4

5

85

90

82

95

87

74

34

89

84

77

195–97 decomp.

135–38 decomp.

198–200 decomp.

204–206 decomp.

200–202 decomp.

184–87 decomp.

196–98 decomp.

178–81 decomp.

155–59 decomp.

190–93 decomp.

46.9 (47.2)

50.9 (51.1)

48.7 (48.6)

52.1 (52.2)

55.1 (55.3)

58.5 (58.0)

58.7 (58.0)

44.9 (45.3)

46.8 (46.7)

53.5 (53.4)

4.6 (4.6)

4.4 (4.5)

5.0 (4.9)

4.8 (4.8)

4.6 (4.7)

4.7 (4.6)

4.4 (4.4)

4.8 (4.8)

4.4 (4.6)

5.6 (5.7)

5.2 (5.2)

5.6 (5.5)

5.0 (5.1)

4.8 (4.7)

4.5 (4.4)

4.6 (4.4)

5.7 (5.5)

5.3 (5.3)

4.5 (4.6)

6

7

8

9

10

11

12

13

^aCalculated values are in parentheses.

Table 2 Crystallographic data for complexes 5a, 7a, 8a and 12a

Complex

5a

7a

8a

12a

Empirical formula

Fw

C₃₄H₃₆F₆N₃P₃Pdؒ0.5CH₃COCH₃

829.01

C₃₆H₄₀F₆N₃P₃Pd

828.02

C₄₁H₄₂F₆N₃P₃Pd

890.09

C₃₁H₃₈F₆N₃P₃PdS

798.01

T /K

Crystal system

Space group

a/Å

b/Å

c/Å

α/Њ

293(2)

monoclinic

P2₁

9.2253(2)

37.0328(9)

10.9457(2)

90.00

293(2)

monoclinic

P2₁/n

10.826(3)

18.176(4)

41.424(10)

90.00

293(2)

monoclinic

P2₁/n

9.836(5)

10.758(2)

33.860(9)

90.00

orthorhombic

P2₁2₁2₁

9.474(4)

19.627(4)

19.837(5)

90.00

β/Њ

90.5650(10)

90.00

3739.3(1)

4

1.473

0.685

90.00

3688.5(2)

4

1.491

0.694

3638

90.75(2)

90.00

8150.0(3)

8

1.451

0.634

97.03(3)

90.00

3556.0(2)

4

1.491

0.773

γ/Њ

V/Å³

Z

Density (calcd)/Mg m^Ϫ3

Abs. coeﬀ./mm^Ϫ1

No. reﬂ.

15784

15726

8210

Indep. reﬂ.

10038

3638

14315

–

0.0609, 0.1604

0.0985, 0.1846

7757

–

0.0522, 0.1404

0.0950, 0.1660

Flack parameter

Final R indices: R₁, wR₂(I >2σ_I)

R indices (all data)

Ϫ0.03(2)

0.0383, 0.0989

0.0441, 0.1091

Ϫ0.08(5)

0.0427, 0.1081

0.0604, 0.1184

recorded using Bruker DRX-500 MHz, Bruker AMX-400 MHz

and Bruker ACF-200 MHz spectrometers. Chemical shifts

downﬁeld from the reference standard were assigned positive

values. Optical rotations were measured on a Jasco DIP-370

digital polarimeter. Elemental analyses were carried out using a

Perkin-Elmer 2400 CHN analyser. The ³¹P{¹H} NMR data are

presented in Table 5 (see below). The ¹H and ¹³C NMR data for

the allyl moiety are listed in Tables ES1 and ES2, respectively, as

Electronic Supplementary Information (ESI);† the remaining

¹H and ¹³C NMR data are also given in the ESI.

diﬀractometer. The crystallographic data and details of data

collections are summarised in Table 2. The structures were

solved by Patterson’s heavy atom method using the program

SHELXS 86¹⁹and reﬁned on F ²values for all unique data by

full-matrix least-squares using SHELXS 97.²⁰The non-hydro-

gen atoms were reﬁned anisotropically. The hydrogen atoms

were ﬁxed at their calculated geometrical positions and allowed

to ride on the non-hydrogen atoms to which they are attached.

The validity of the absolute structure for 5a and 7a was

established by the method of Flack.²¹

CCDC reference numbers 190877–190880.

Synthesis

See http://ww w .rsc.org/suppdata/dt/b2/b207447h/ for crystal-

lographic data in CIF or other electronic format.

[Pd(ꢀ³-C₃H₅){ꢁ²-Ph₂PN(CHMe₂)PPh(N₂C₃HMe₂-3,5)}]-

(PF₆), 4. A mixture of 0.036 g (0.98 × 10^Ϫ4mol) of [Pd-

(η³-C₃H₅)(µ-Cl)]₂, 0.033 g (2.02 × 10^Ϫ4mol) of NH₄PF₆and

0.095 g (2.13 × 10^Ϫ4mol) of 1 was dissolved in 20 cm³of acet-

one. The solution was stirred for 2 h at 298 K and the white

precipitate formed during the reaction was ﬁltered oﬀ. The

resulting colourless ﬁltrate was concentrated under reduced

pressure to 10 cm³and the solution was layered by adding

10 cm³of hexane (bp 40–60 ЊC) to yield colourless crystals. The

other allyl palladium complexes, 5–13 were synthesised by an

analogous procedure by varying the allyl palladium chloro

dimer and the diphosphazane ligand. The yields, melting points

and elemental analyses for these complexes are given in Table 1.

General procedure for palladium-catalysed allylic alkylation

The ligand (2.5 mol%) and [Pd(η³-C₃H₅)(µ-Cl)]₂(1 mol%) were

dissolved in 4 cm³of degassed (by three freeze–thaw cycles)

CH₂Cl₂. A solution of (rac)-1,3-diphenyl-2-propenyl acetate (I)

or (E)-1-phenyl-2-propenyl acetate (III) (1 equiv.), in 2 cm³

CH₂Cl₂, was added followed by dimethyl malonate (2 equiv.),

N,O-bis(trimethylsilyl)acetamide (2 equiv.), and a catalytic

amount of KOAc. The mixture was stirred at 298 K for 24 h

and the reaction was monitored by TLC. A saturated aqueous

solution of NH₄Cl (10 cm³) was added to the reaction mixture

and the organic part was extracted with diethyl ether (10 ×

3 cm³). The ether extract was ﬁltered over activated celite,

washed with water (3 × 10 cm³) and dried over anhydrous

Na₂SO₄. Solvent was removed from the ether extract under

reduced pressure and the crude product was puriﬁed by column

chromatography over silica gel (eluent hexane/ethyl acetate,

X-Ray crystallography

Crystal data for complex 5a was collected using Siemen’s

SMART-CCD diﬀractometer. The intensity data for complexes

7a, 8a and 12a were collected using an Enraf-Nonius CAD-4

D a l t o n T r a n s . , 2 0 0 3 , 1 0 1 6 – 1 0 2 7

1017

View Article Online

9 : 1). The optical rotation was measured for the product,

typical for a two-bond cis coupling of phosphorus and carbon

nuclei.²³The resonances of the central allyl carbon nuclei of all

the isomers appear as doublets as would be expected for P,N-

coordination. The driving force for the P,N-coordination might

be the formation of a stable six-membered ring around pal-

ladium rather than a four-membered ring which would result

from P,P-coordination. Moreover, electronically the coordin-

ation of the σ-donor nitrogen is more favourable than that of

the π-acceptor phosphorus as it can synergistically enhance the

metal–allyl bonding.²⁴

PhCH᎐CH–CHPh{CH(CO Me) } (II) and the enantiomeric

᎐

2

excess (e.e.) was calculated from the optical rotation value using

the literature data.^6b,22The enantiomeric excesses obtained were

29

2% in favour of the ‘R’ enantiomer (observed [α]_D=

ϩ 6.10, c = 3.2, EtOH) when ligand 1 was used and 28 2% in

favour of the ‘S’ enantiomer (observed [α]_D= Ϫ5.96, c = 3.2,

EtOH) when ligand (SR)-2 was used as the chiral auxiliary.

Detailed 2D NMR studies (¹H–¹H DQF-COSY, NOESY,

ROESY and ¹³C–¹H HSQC) show that the two isomers of 4 and

5 have the endo- (major) and exo- (minor) conﬁgurations

(Fig. 1).^2b,25For the major isomers of 4 and 5, there is a NOE

contact between the 3-Me protons of pyrazole and the central

allyl proton, H_cand the syn proton H_s. This result indicates

that the central allyl hydrogen, H_cpoints toward 3-Me

protons of pyrazole i.e. the major isomers (4a and 5a) have the

endo-conﬁguration (3-Me protons and H_care on the same side

of the N–Pd–P coordination plane). By contrast, the pyrazole

3-Me protons of the minor isomers of 4 and 5 do not show

NOE contact to the central allyl proton H_cbut there is a strong

NOE contact to the anti allyl proton, H_a. Hence, the minor

isomers (4b and 5b) are assigned the exo-conﬁguration (3-Me

and H_care on the opposite sides of the N–Pd–P coordination

plane).

Results and discussion

The cationic allyl palladium complexes 4–13 have been pre-

pared from the reactions of the chlorobridged palladium allyl

dimers, [Pd(η³-1,3-RЈ₂-C₃H₃)(µ-Cl)]₂(RЈ = H or Me or Ph) with

the appropriate ligands in the presence of NH₄PF₆in acetone

(Scheme 2). These complexes have been characterised by ele-

mental analyses and NMR spectral data (see ESI† and Table 5

below). The isomeric composition in each case has been deter-

mined from NMR spectroscopy. The salient features of the

spectroscopic data and dynamic behaviour are discussed below.

Fig. 1 The two isomers of Pd()–allyl diphosphazane complexes,

[Pd(η³-C₃H₅){κ²-Ph₂PN(R)PPh(N₂C₃HMe₂-3,5)}](PF₆) [R = CHMe₂-

(4), (S)-*CHMePh(5)] observed in solution. Selective NOE contacts

reveal that in the endo-isomer, the central allyl proton and pyrazole-3-

Me protons are close to each other whereas for the exo-isomer, such an

NOE contact with the central allyl proton is absent.

The phase sensitive ¹H–¹H 2D NOESY and ROESY spectra

of 4 at 298 K clearly reveal syn–anti exchange of the allyl pro-

tons trans to nitrogen. Pregosin and coworkers²⁶have demon-

strated that such an exchange can be explained by a selective

opening of the η³-allyl moiety trans to the coordinated phos-

phorus centre followed by rotation around the sp³–sp²C–C

bond and reversal to the η³-mode of coordination by the other

face of the allyl moiety. Complex 5 does not show any dynamic

behaviour at 298 K; presumably the interconversion is slow on

the NMR time scale at this temperature. However, at 323 K, the

¹H–¹H 2D NOESY and ROESY spectra of 5 indicate the pres-

ence of two simultaneous exchange processes. Close examin-

ation of the allyl proton exchange cross-peaks reveals selective

syn–anti exchange as well as cis–trans isomerisation. The

observed exchange results can be accounted for by considering

the sequence of steps shown in Scheme 3. Step (i) is the selective

opening of the η³-allyl bond to the metal at the carbon trans to

the coordinated phosphorus centre of 5a to generate the inter-

mediate D. This intermediate (D) can undergo either a rotation

around the sp³–sp²C–C bond to form 5b [step (ii)] or it can

undergo a rotation around the Pd–C bond [step (iii)] thus result-

ing in 5c. The isomers 5b and 5c are essentially the same and are

not distinguishable in the NMR spectrum, as the allyl group is

unsubstituted and symmetrical. Thus the isomerisation appears

to be subject to electronic control as observed for η³-allyl

palladium complexes of other heterodonor P,S-,²⁷P,O-²⁸and

P,N-^23b,29bidentate ligand systems.

Scheme 2

Palladium allyl complexes of Ph₂PN(R)PPh(N₂C₃HMe₂-3,5)

[R

؍

CHMe₂(1) or (S )-*CHMePh (2)]

The ³¹P{¹H} NMR spectra of 4 and 5 in CDCl₃show two AX

spin patterns indicating the presence of two isomers in solution.

The relative ratios of the isomers are 3 : 1 and 3.4 : 1 for 4 and 5,

respectively. An analysis of ¹³C NMR spectra of allyl carbons

clearly shows P,N-coordination via the nitrogen atom of the

pyrazole ring instead of the usually observed P,P-coordin-

ation.^11c–eA doublet is observed for each of the two terminal

allyl carbon nuclei. The resonances of the terminal allyl carbon

nuclei trans to the coordinated phosphorus centre appear in the

range ca. 75–81 ppm with a ²J(P,C) of 28–31 Hz which lies in a

region usually observed for trans ³¹P–¹³C coupling whereas the

signal for the terminal allyl carbon nucleus trans to nitrogen

resonates in a more shielded region (ca. 58–62 ppm) with a

²J(P,C) coupling constant of the order of 3–4 Hz which is

D a l t o n T r a n s . , 2 0 0 3 , 1 0 1 6 – 1 0 2 7

1018

View Article Online

Scheme 3

Table 3 Selected bond distances (Å) and angles (Њ) in 5a, 7a, 8a and 12a

When a solution of complex 5 in acetone/hexane (1 : 1 v/v) is

cooled, crystals of 5a are obtained. The crystallographic data

for 5a are summarised in Table 2. Selected bond lengths and

bond angles are listed in Table 3. The solid state structure shows

the presence of two crystallographically independent molecules

5a(A) and 5a(C) (A and C represent the diastereomeric con-

ﬁguration as explained below) shown in Fig. 2. The ‘PdC₂PN’

core shows a distorted square-planar geometry with bond

distances of Pd–C, Pd–P and Pd–N in the expected range for

Pd() allyl compounds with P,N-chelate ligands. The P–N–P

bond angle is expanded to 121.8(2)Њ which is normally ca. 100Њ

for P,P-coordinated complexes.¹¹The relative orientations of

the chelating ligand and the allyl moiety are similar in both

the molecules. The central allyl carbon and 3-methyl group of

the pyrazole ring in 5a(A) lie above the coordination plane

deﬁned by N–Pd–P whereas in 5a(C) they are located below

this plane. The two molecules are not superimposable; the

molecular asymmetry arises because of the presence of chiral

centres in the diphosphazane ligand and the position of the

η³-coordinated allyl moiety which is obliquely placed with

respect to the metal square plane.^30,31The conﬁguration at the

stereogenic palladium centre in 5a(A) and 5a(C) can be desig-

nated as A and C, respectively, following the method adopted

by Lanza et al.³⁰When the molecule is viewed from the central

allyl carbon towards palladium, the other two atoms coordin-

ated to the metal viz. P and N (with decreasing CIP priority)

follow an anticlockwise direction in 5a(A) whereas these atoms

follow a clockwise direction in 5a(C) (see Fig. 2). The six-

membered ring in both the molecules adopts a distorted boat

conformation. The diﬀerence between the two Pd–C bond dis-

tances indicates the greater trans inﬂuence of the phosphorus

donor compared to nitrogen. The Pd–C bond lengths trans to

the coordinated phosphorus centre are 2.213(7) and 2.189(7) Å

whereas for the Pd–C bond trans to nitrogen are 2.127(8) and

2.107(7) Å for 5a(A) and 5a(C), respectively (see Table

3).^10,23b,32This elongation of Pd–terminal allyl carbon bond

length trans to the coordinated phosphorus centre as com-

pared to that trans to nitrogen is consistent with the proposed

mechanism involving selective opening of one of the allyl

termini.

Bond distances/Å

Bond angles/Њ

Molecule 5a(A)

Pd(1)–C(1)

Pd(1)–C(2)

Pd(1)–C(3)

Pd(1)–P(4)

Pd(1)–N(4)

2.127(8)

2.133(7)

2.213(7)

2.297(1)

2.100(5)

P(4)–Pd(1)–N(4)

C(1)–Pd(1)–C(3)

C(3)–Pd(1)–P(4)

C(1)–Pd(1)–N(4)

P(2)–N(6)–P(4)

86.2(1)

67.9(4)

172.0(3)

169.5(3)

121.8(2)

Molecule 5a(C)

Pd(2)–C(38)

Pd(2)–C(39)

Pd(2)–C(40)

Pd(2)–P(3)

Pd(2)–N(1)

2.107(7)

2.155(8)

2.189(7)

2.296(1)

2.116(5)

P(3)–Pd(2)–N(1)

C(38)–Pd(2)–C(40)

C(40)–Pd(2)–P(3)

C(38)–Pd(2)–N(1)

P(5)–N(5)–P(3)

86.6(1)

67.9(4)

171.0(3)

169.7(3)

121.8(2)

Complex 7a

Pd(1)–C(4)

Pd(1)–C(3)

Pd(1)–C(2)

Pd(1)–P(2)

Pd(1)–N(3)

2.12(1)

2.129(8)

2.27(1)

2.304(2)

2.150(6)

P(2)–Pd(1)–N(3)

C(4)–Pd(1)–C(2)

C(2)–Pd(1)–P(2)

C(4)–Pd(1)–N(3)

P(1)–N(1)–P(2)

84.2(2)

66.3(5)

172.2(4)

168.8(4)

121.6(3)

Molecule 8a(A)

Pd(1)–C(3)

Pd(1)–C(2)

Pd(1)–C(1)

Pd(1)–P(1)

Pd(1)–N(3)

2.173(7)

2.179(7)

2.224(7)

2.328(2)

2.119(6)

P(1)–Pd(1)–N(3)

C(3)–Pd(1)–C(1)

C(1)–Pd(1)–P(1)

C(3)–Pd(1)–N(3)

P(1)–N(1)–P(2)

84.2(2)

66.5(3)

176.4(2)

165.6(3)

122.8(3)

Molecule 8a(C)

Pd(2)–C(44)

Pd(2)–C(43)

Pd(2)–C(42)

Pd(2)–P(3)

Pd(2)–N(6)

2.148(7)

2.178(8)

2.232(8)

2.306(2)

2.128(6)

P(3)–Pd(2)–N(6)

C(44)–Pd(2)–C(42)

C(42)–Pd(2)–P(3)

C(44)–Pd(2)–N(6)

P(3)–N(4)–P(4)

83.9(2)

67.1(3)

174.2(2)

168.2(3)

121.9(4)

Complex 12a

Pd(1)–C(2)

Pd(1)–C(3)

Pd(1)–C(4)

Pd(1)–P(1)

Pd(1)–S(1)

2.171(7)

2.127(5)

2.192(6)

2.262(1)

2.338(1)

P(1)–Pd(1)–S(1)

C(2)–Pd(1)–C(4)

C(4)–Pd(1)–P(1)

C(2)–Pd(1)–S(1)

P(1)–N(1)–P(2)

91.7(4)

67.4(3)

167.7(2)

163.6(2)

118.3(2)

To obtain a better understanding of the ﬂuxional process, we

have studied 1,3-disubstituted allyl complexes of the ligands 1

and 2 since the unsubstituted allyl moiety is a poor model to

understand the stereodynamic process. The various possible

allylic arrangements are shown in Scheme 4. For each arrange-

ment, there is also the possibility of exo- or endo-isomers

(Scheme 5). Furthermore, there is the possibility of P,P-

coordination.

D a l t o n T r a n s . , 2 0 0 3 , 1 0 1 6 – 1 0 2 7

1019

View Article Online

intra-allyl NOESY cross-peaks. (Two strong NOE cross-peaks

between the central allylic proton with two terminal allyl methyl

protons indicate a syn/syn-arrangement while a strong NOE

between the central allylic proton and one of the two terminal

allyl protons indicates syn/anti- or anti/syn-arrangements.)

Appearance of an NOE cross-peak between the central allyl

proton and 3-methyl protons of pyrazole moiety would reveal

the endo-conﬁguration, whereas the absence of such an NOE

involving the central allyl proton would clearly point to the

exo-conﬁguration. Additionally, the chemical shift of the

central allyl proton for the endo-isomer appears in a deshielded

region (ca. 5.5 ppm) compared to that for the exo-isomer

(ca. 4 ppm) as observed for complexes 4 and 5. Based on the

above considerations, we assign the major species as endo, syn/

syn- (6a and 7a), the second minor species as exo-, syn/syn- (6b

and 7b), the third minor species as exo, anti/syn- (6c and 7c) and

the fourth minor species as endo, anti/syn- (6d and 7d) isomers

(see Fig. 3). The absence of the anti/anti-isomer is understand-

able on the basis of steric grounds but the absence of any syn/

anti-isomer is most probably due to electronic reasons. The anti

allyl methyl group is invariably present at the trans position

with respect to the nitrogen donor. This is in accord with the

proposed mechanism for the exchange processes observed for 4

and 5 (Scheme 3), in which the sp³carbon is generated at the

trans position with respect to the nitrogen donor; in other

words, the η³-allyl group opens selectively at the trans position

with respect to the coordinated phosphorus donor to form the

‘η¹-coordinated’ intermediate.

Fig. 2 A view of the molecular structure of 5a, showing the presence

of two independent molecules 5a(A) (top) and 5a(C) (bottom) (the

hexaﬂuorophosphate anion, the hydrogen atoms and the lattice held

Me₂CO are not shown).

Scheme 4

Fig. 3 The four isomers observed in solution for 1,3-dimethyl allyl

palladium complexes, [Pd(η³-1,3-Me₂-C₃H₃){κ²-Ph₂PN(R)PPh(N₂C₃-

HMe₂-3,5)}](PF₆) [R = CHMe₂(6), (S)-*CHMePh (7)].

Although complex 7 exists as a mixture of four diastereomers

in solution, the solid state structure shows that the allyl moiety

adopts an endo, syn/syn-conﬁguration (7a) (endo- refers to the

relative orientation of the central C–H vector pointing towards

the 3-methyl group on the pyrazole ring). The crystallographic

data are listed in Table 2. The molecular structure is shown in

Fig. 4. The bonding parameters (Table 3) fall in the expected

ranges.^23b,31A few structural features are worth noting. If one

considers the inner coordination sphere as formed by the P and

N donor atoms of the diphosphazane ligand and by the ter-

minal allyl carbons C(2) and C(4), the coordination geometry

around the metal centre is distorted square planar, with the

P(2)–Pd(1)–N(3) and C(4)–Pd(1)–C(2) angles being 84.2(2) and

66.3(5)Њ, respectively. The two terminal allyl carbon atoms are

not coplanar with the coordination plane formed by N(3)–

Pd(1)–P(2); the distances of C(2) and C(4) from this plane are

Scheme 5

The complexes, [Pd(η³-1,3-Me₂-C₃H₃)(1)](PF₆) (6) and

[Pd(η³-1,3-Me₂-C₃H₃)((SR)-2)](PF₆) (7) exist as a mixture

of four isomers as shown by their ³¹P{¹H} NMR spectra. A

detailed analysis of the ¹³C, ¹H and ³¹P{¹H} NMR spectra con-

ﬁrms that all the four isomers are P,N-coordinated species. The

isomeric distributions are 43 : 35 : 15 : 7 and 69 : 19 : 9 : 3 for 6

and 7, respectively. Two-dimensional NMR studies have been

carried out for complex 7. The assignment of the allyl con-

1

ﬁguration is based on H spin–spin coupling constants and

D a l t o n T r a n s . , 2 0 0 3 , 1 0 1 6 – 1 0 2 7

1020

View Article Online

ated diastereomers (shown in Fig. 5). The magnitude of ²J(P,P)

for P,N-coordination (ca. 30 Hz) is much less than that for

P,P-coordination (ca. 115 Hz). Such a large diﬀerence in ²J(P,P)

is probably due to the diﬀerence in P–N–P bond angle in the

two cases. The assignment of the resonances arising from the

two minor isomers (8b and 8c) are based on the chemical shift

diﬀerence of two methyl groups at 3- and 5-positions of the

pyrazole ring. The value of the chemical shift diﬀerence

∆δ(CH₃) [δ(5-CH₃) Ϫ δ(3-CH₃)] for the isomer 8b is 1.10 ppm

whereas this diﬀerence for the isomer 8c is 0.16 ppm. Such a

large diﬀerence indicates that in 8b the central allyl carbon and

pyrazole ring are on the same side of the P–Pd–P coordination

plane and the allyl phenyl group comes close to one of the

pyrazole methyl groups. Thus the two methyl groups experience

diﬀerent steric and electronic environments. By contrast, the

phenyl groups and the pyrazole methyl groups in 8c are on

opposite sides of the coordination plane resulting in lesser

extent of interaction of the pyrazole methyl groups with the

allylic moiety. The 1,3-diphenyl allyl complexes of the

more bulky diphosphazane ligands, (SR)-Ph₂PN((S)-*CH-

MePh)PPh(N₂C₃HMe₂-3,5) and (SS)-Ph₂PN((S)-*CHMe-

Ph)PPh(N₂C₃HMe₂-3,5) form only the P,N-coordinated

isomers 9 and 10 with endo, syn/syn-conﬁguration as revealed

Fig. 4 A view of the molecular structure of complex 7a (the hydrogen

atoms and the hexaﬂuorophosphate anion are not shown).

0.16 and 0.07 Å, respectively. The central allyl carbon atom

C(3) is located at a distance of 0.36 Å from the same plane. The

3-Me pyrazole carbon, C(6) lies on the same side of the central

allyl carbon with regard to the coordination plane (distance

from the coordination plane is 2.23 Å). The terminal Pd–C allyl

bond lengths are signiﬁcantly diﬀerent from one another; the

carbon atom trans to the coordinated phosphorus centre dis-

plays the expected longer distance [Pd(1)–C(2) = 2.27(1) Å], as

compared to its partner trans to nitrogen [Pd(1)–C(4) = 2.12(1)

Å] (Table 3). This is a reﬂection of the higher trans inﬂu-

ence^10,32,33of the phosphorus donor centre. The six-membered

chelate ring adopts a distorted boat conformation as shown by

the distance calculation from the mean plane deﬁned by N(1)–

N(2)–N(3); both Pd(1) and P(1) lie above the N(1)–N(2)–N(3)

plane (the deviations from this plane are 1.17 and 0.74 Å

respectively) whereas P(2) atom is almost coplanar with this

plane (the deviation is only 0.10 Å below this plane).

1

by H–¹H NOESY NMR spectra. The formation of only one

isomer is not surprising as the syn/anti- or anti/anti-arrange-

ments are not favourable for allylic moiety containing bulky

phenyl groups.

When crystalline 7a is dissolved in CDCl₃at 298 K, it under-

goes slow isomerisation (not detectable by phase sensitive

¹H–¹H NOESY and ROESY spectra at 298 K) to produce 7b,

7c and 7d. The formation of exo, syn/syn-isomer (7b) from the

endo, syn/syn-isomer (7a) can be rationalised by considering a

rotation around the Pd–C bond (step (i), Scheme 6). Thus com-

plex 7b represents 5c shown in the isomerisation pathway for

unsubstituted allyl complexes (see Scheme 3). The exo, anti/syn-

isomer (7c) is the 1,3-dimethyl substituted analogue of 5b

which is formed by rotation around sp³–sp²C–C bond (step (ii),

Scheme 6). Formation of 7d can be explained by considering a

further syn–anti exchange process starting from 7b (step (iii),

Scheme 6). This isomerisation pathway is supported by the

¹H–¹H NOESY and ROESY spectra at 323 K for complex 7.

The exchange results at 323 K reveal that complex 7a undergoes

exchange with complex 7b by cis–trans isomerisation and at the

same time complex 7b also shows ﬂuxional behaviour with

complex 7d by syn–anti isomerisation pathway (on the basis of

exchange pattern observed for the allyl methyl groups of the

two isomers). There is no observable exchange between 7a and

7c at 323 K.

Fig. 5 The isomeric species observed in solution for the 1,3-diphenyl

allyl palladium() diphosphazane complexes, [Pd(η³-1,3-Ph₂-C₃H₃){κ²-

Ph₂PN(R)PPh(N₂C₃HMe₂-3,5)}](PF₆) [R = CHMe₂(8), (S)-*CHMePh

(9, 10)].

The molecular structure of complex 8 reveals that only the

endo, syn/syn-isomer (8a) is present in the solid state. The

crystal data are given in Table 2 and selected bond angles and

bond distances are listed in Table 3. There are two independent

molecules in the unit cell, which are not superimposable. An

ORTEP plot of the two molecules is shown in Fig. 6. Both the

molecules feature P,N-coordination. Furthermore, the central

allyl carbon points towards the 3-methyl of the pyrazole ring

revealing an endo-conﬁguration with syn/syn-arrangement of

the allyl phenyl groups in both the molecules 8a(A) and 8a(C).

The two molecules diﬀer in their relative direction of P–Pd–N

when the molecule is viewed from the central allyl carbon

towards palladium; in 8a(A) it follows an anticlockwise direc-

tion whereas in 8a(C) it is in the clockwise direction. The

coordination geometry around the palladium metal is distorted

square planar, with the allyl ligand markedly displaced from its

idealised position. In molecule 8a(A), the two terminal carbon

When we move to the 1,3-diphenyl allyl system containing

the more bulky substituents with enhanced electron donating

capacity compared to the 1,3-dimethyl allyl system, the com-

plex formed by the diphosphazane ligand 1, viz. [Pd(η³-1,3-

Ph₂-C₃H₃){κ²-Ph₂PN(CHMe₂)PPh(N₂C₃HMe₂-3,5)}](PF₆) (8)

exists as a mixture of three isomers in solution in the ratio

75 : 15 : 10 as shown by its ³¹P{¹H} and ¹H NMR spectra. The

1

¹³C NMR and H–¹H NOESY spectra show that the major

component is the P,N-coordinated endo, syn/syn-isomer (8a)

and the other two minor isomers (8b and 8c) are P,P-coordin-

D a l t o n T r a n s . , 2 0 0 3 , 1 0 1 6 – 1 0 2 7

1021

View Article Online

Scheme 6

0.65 Å from the coordination plane on the same side of C(42),

as shown below:

The deviation of the terminal carbon atoms from the

coordination plane is less for the carbon atom trans to the

coordinated phosphorus than the one trans to nitrogen. The

plane of the allyl ligand is tilted (20.4Њ in molecule 8a(A) and

20.3Њ in molecule 8a(C)) from an axis perpendicular to the N–

Pd–P coordination plane. The allyl ligands are rotated mark-

edly around the Pd–allyl bond axis away from the –PPh₂group

to minimise steric interaction with the allyl phenyl groups. A

similar rotation of the allyl moiety with respect to its idealised

position has also been observed in the related cationic com-

plexes with P,N-chelating ligands, containing the sterically

demanding η³-1,3-diphenylallyl fragment.^10,23a,32As expected,

the palladium–terminal allyl carbon bond distance, which is

trans to the coordinated phosphorus, is longer than that trans to

nitrogen in both the molecules, revealing the greater trans inﬂu-

ence of the phosphorus donor centre. The chelated six-mem-

bered rings adopt boat conformations.

When a crystalline sample of complex 8a is dissolved in

CDCl₃, it undergoes slow isomerisation to produce a mixture of

1

complexes 8a–c. The phase sensitive H–¹H NOESY spectrum

at 298 K indicates no observable exchange between these iso-

mers which is not surprising as the sterically demanding allyl

moiety raises the activation barrier for exchange. When the

temperature is raised (323 K), one observes an exchange

between the isomers 8b and 8c probably via a cis–trans isomer-

isation. The formation of 8b and 8c from 8a (the isomer present

in the solid state) can be rationalised by considering a pathway

as shown in Scheme 7. Alternatively, 8b may be formed ﬁrst and

then it can undergo cis–trans isomerisation to form 8c. When

the ¹H NMR spectrum is recorded immediately after dissolving

the solid 8a in CDCl₃, the three isomers 8a–8c are present in the

ratio 91 : 4 : 3. This ratio changes to 75 : 14 : 11 upon heat-

ing the sample at 323 K for 1 h. It may be noted that the rel-

ative ratio of 8b and 8c remains the same before and after the

Fig. 6 A view of the molecular structure of 8a showing the presence

of two independent molecules 8a(A) (top) and 8a(C) (bottom) (the

hydrogen atoms and the hexaﬂuorophosphate anion are not shown).

atoms are on the same side of the coordination plane formed by

N(3)–Pd(1)–P(1) (the distances are 0.01 and 0.12 Å for C(1) and

C(3) respectively) and the central allyl carbon atom C(2) resides

on the opposite side (the distance is 0.61 Å). On the other hand,

in molecule 8a(C), the terminal allyl carbons are located on

opposite sides to the plane formed by N(6)–Pd(2)–P(3) (the

distances are 0.06 and 0.115 Å for C(42) and C(44) respect-

ively). The central allyl carbon, C(43) is located at a distance of

D a l t o n T r a n s . , 2 0 0 3 , 1 0 1 6 – 1 0 2 7

1022

View Article Online

Scheme 7

Fig. 7 The two Pd()–allyl diastereomers of diphosphazane mono-

sulﬁde,

[Pd(η³-C₃H₅){κ²-Ph₂P(S)N(CHMe₂)PPh(N₂C₃HMe₂-3,5)}]-

(PF₆) (11), observed in solution. Selective NOE contacts reveal that in

the endo-isomer, the central allyl proton and pyrazole-3-Me protons are

close to each other whereas for the exo-isomer, such an NOE contact

with the central allyl proton is absent.

equilibrium is attained although the relative ratios with respect

to 8a vary. This result indicates that the formation of 8b and 8c

from 8a involves a common intermediate formed by selective

opening of the allyl terminus trans to the coordinated phos-

phorus centre accompanied by a concerted attack of the metal

by the non-coordinated phosphorus. Recombination to the η³-

mode before and after a rotation around the Pd–C bond would

generate 8b and 8c, respectively (Scheme 7).

The absence of such an interaction in the minor isomer (11b)

indicates that this isomer has an exo-conﬁguration.

The ³¹P{¹H} NMR spectrum of the 1,3-dimethyl allyl com-

plex 12 shows three pairs of doublets in the ratio 6 : 4 : 1

indicating the presence of three isomers (12a–c) in solution.

The ¹H NMR spectrum shows three diﬀerent sets of allyl

protons each consisting of three resonances. All the proton

resonances are unambiguously assigned by recourse to stand-

Palladium allyl complexes of a diphosphazane monosulﬁde

ligand, Ph₂P(S)N(CHMe₂)PPh(N₂C₃HMe₂-3,5) (3)

The unsymmetrically substituted diphosphazane monosulﬁde

ligand, Ph₂P(S)N(CHMe₂)PPh(N₂C₃HMe₂-3,5) (3) consists of

three potential donor centres namely, S, P and N at the

2-position of the pyrazole ring. It can form complexes using

either P,N- or P,S- or N,S-combinations for binding to the metal

resulting in a four-, ﬁve- or seven-membered chelate ring,

respectively, around the metal. The ³¹P{¹H} NMR spectrum of

the complex, [Pd(η³-C₃H₅){κ²-Ph₂P(S)N(CHMe₂)PPh(N₂C₃-

HMe₂-3,5)}](PF₆) (11) in CDCl₃shows two pairs of doublets in

1

ard H–¹H 2D homo correlation techniques. The ¹³C NMR

spectrum is assigned with the help of 2D ¹H–¹³C heteronuclear

correlation spectrum. The doublet pattern observed for one

of the terminal allylic carbon resonances clearly point to

P,S-coordination. The signal arising from the third isomer 12c

could not be seen in the ¹³C NMR spectrum owing to its relative

1

low abundance. However, the ³¹P{¹H} and H chemical shifts

a ratio 4 : 1 revealing the presence of two diastereomers. The ¹³

C

and ²J(P,P) coupling constant values for 12c are similar to those

of 12a and 12b. Hence, 12c must be structurally close to 12a

and 12b. The ¹H–¹H COSY spectrum indicates the presence of

two anti allyl protons for isomers 12a and 12b i.e. the methyl

groups are in the syn/syn-arrangements (see Fig. 8). This

arrangement is supported by the expected NOE interactions

between the central allyl proton and two allyl methyl groups for

allyl resonances are assigned unequivocally by carrying out a

¹³C–¹H HSQC 2D experiment with the help of pre-assigned

allyl ¹H resonances by a ¹H–¹H COSY experiment. Analysis of

the ¹³C NMR spectrum shows a doublet pattern with a ²J(P,C)

value of ca. 30 Hz for one of the two terminal allyl carbon

resonances in case of isomers 11a and 11b indicating bonding

of one phosphorus with the palladium centre. From the ³¹P

chemical shifts (Table 5), we can conclude that the pyrazol-

bearing phosphorus atom and the sulfur atom are coordinated

to palladium. The doublet at ca. 91.5 ppm is assigned the

coordinated –PPh(N₂C₃HMe₂-3,5) phosphorus nucleus with

coordination shift of 26.0 ppm as compared with the free ligand

³¹P chemical shift. The second doublet at ca. 77.0 ppm with a

coordination shift of 6.5 ppm is assigned the non-coordinated –

P(S)Ph₂phosphorus nucleus. This conclusion is supported by

the ¹³C chemical shifts of the allyl carbon nuclei when com-

pared with the chemical shifts of allyl carbon nuclei of P,N-

coordinated complexes 4–10 (see ESI). It may be noted that the

chemical shifts of allyl carbon nuclei trans to nitrogen are more

shielded (by ∼10–20 ppm) than the chemical shifts of the allyl

carbon nuclei trans to sulfur when we consider the similar allyl

isomers in both the cases. Thus, in contrast to 1 and 2, the

ligand 3 does not coordinate through the nitrogen atom of the

pyrazole ring. The P,S-coordination is presumably favoured

because of the formation of a stable ﬁve-membered ring around

the palladium centre. The competitive N,S-coordination would

lead to the formation of a seven-membered chelate ring around

the palladium atom.

Two sets of ﬁve non-equivalent η³-allyl proton resonances

are observed in the ¹H NMR spectrum of 11 at 298 K. Some of

1

the H resonances of the minor isomer 11b (e.g. H_cand H_a)

overlap with those arising from the major isomer (11a), but they

can be located by 2D NMR ¹H–¹H NOESY measurements. The

most interesting NOE interaction is that observed between the

5-CH₃group of the pyrazole ring and the central allyl proton,

H_cand syn allyl proton trans to the sulfur atom, H_sЈ for the

major isomer (11a) revealing an endo-conﬁguration (see Fig. 7).

Fig. 8 The three isomers observed in solution for 1,3-disubstituted allyl

palladium complexes, [Pd(η³-1,3-RЈ₂-C₃H₃){κ²-Ph₂P(S)N(CHMe₂)-

PPh(N₂C₃HMe₂-3,5)}](PF₆) [RЈ = Me (12); RЈ = Ph (13)].

D a l t o n T r a n s . , 2 0 0 3 , 1 0 1 6 – 1 0 2 7

1023

View Article Online

the isomers, 12a and 12b. It is diﬃcult to determine the allylic

the selective exchanges observed (see ESI†), the major exchange

pathway is found to be the cis–trans isomerisation. Apart from

this, a second isomerisation is also observed via syn–anti

exchange. The syn–anti isomerisation is observed only at the

allyl terminus trans to the sulfur atom, suggesting that the open-

ing of the η³-allyl ligand to η¹-bonded intermediate is con-

trolled electronically. Despite the occurrence of these two pro-

cesses in solution, only two diastereomers are observed as the

allyl group is unsubstituted. (Such a result is also observed in

the case of P,N-coordinated unsubstituted allyl complex 5.)

The dynamic behaviour of the complexes 12 and 13 has been

1

arrangement for 12c from the H–¹H COSY spectrum because

the resonances of both the allyl protons appear at the same

1

position (4.75 ppm). However, a careful analysis of the H–¹H

NOESY spectrum reveals that 12c is the anti/syn-isomer,

because only the methyl proton resonances at 1.99 ppm show an

NOE interaction to the central allyl proton, H_cand the other

methyl proton resonances at 0.31 ppm do not show any NOE

interaction with H_cas would be expected for the anti/syn-isomer

in which the anti-allyl methyl group is located trans to the sulfur

atom (see Fig. 8). In contrast to the situation for complex 11,

the major isomer (12a) observed in solution for complex 12,

possesses the exo-conﬁguration. This is revealed by the NOE

interaction between the 5-CH₃protons of the pyrazole ring and

the anti allyl proton trans to the sulfur centre, H_aЈ; the absence

of such an interaction with the H_cproton also supports the exo-

conﬁguration for isomer 12a. The isomers 12b and 12c exhibit

NOE interaction between the central allyl proton to the respect-

ive 5-CH₃group of the pyrazole ring which is indicative of their

endo-conﬁguration.

The 1,3-diphenyl allyl complex 13 exists as two isomers 13a

and 13b in solution (in a 1 : 1.3 ratio) as revealed by its ³¹P{¹H}

NMR spectrum. NMR studies on these isomers show that they

have the syn/syn- allylic arrangement. An NOE interaction

between 5-CH₃group of pyrazole ring and central allyl proton

establishes that 13a has the exo-conﬁguration whereas 13b has

the endo-conﬁguration (Fig. 8).

1

studied by the phase sensitive H–¹H NOESY and ROESY

spectra in CDCl₃at 298 K. No exchange is observed at 298 K

between the isomers 13a and 13b. On the other hand, the crys-

tals of 12a on dissolution in CDCl₃transform into the other

two minor isomers 12b and 12c. The isomer 12c does not show

any observable exchange. An analysis of the observed exchange

cross-peaks between 12a and 12b shows that the exchange

occurs by a cis–trans isomerisation process. A possible mechan-

ism is shown in Scheme 8. The presence of the anti/syn-isomer

(12c) indicates that the isomerisation involves initial selective

opening of the η³-allyl group at the allylic carbon trans to the

coordinated phosphorus atom to generate an η¹-bonded inter-

mediate. This type of selective opening can account for the

absence of any syn/anti-isomer (i.e. the anti allyl methyl group is

located trans to the coordinated phosphorus centre) and is well

1

known for P,S-ligands.²⁷The H–¹H NOESY spectrum (see

From a solution of 12 containing three isomers, the major

isomer 12a is isolated as colourless crystals. The structure of

12a has been determined by X-ray diﬀraction. An ORTEP view

of the molecule is shown in Fig. 9. The X-ray structure shows

P,S-coordination with a distorted ‘PdC₂PS’ core. The Pd–C and

Pd–S bond distances (Table 3) are in the expected range

observed for Pd() allyl compounds with P,S-chelate ligands.^6b

The 5-CH₃group of the pyrazole and central allyl carbon lie at

the opposite side of the coordination plane formed by P(1)–

Pd(1)–S(1) revealing the exo, syn/syn-conﬁguration in the solid

state.

ESI†) shows NOE cross-peaks between the isomers 12a and 12b.

These intermolecular NOE interactions arise as a result of a

fast exchange process in solution. The rapid dynamic process is

most probably related to the possibility that the free nitrogen on

the pyrazole ring can interact with the palladium centre in the

η¹-bonded intermediate. This internal coordination of the

pyrazole nitrogen stabilises the η¹-bonded intermediates (B and

C in Scheme 8) and thereby can accelerate the cis–trans isomer-

isation process. It is well known that catalytic amounts of

anions such as chloride or ﬂuoride or polar solvents like DMSO

or acetonitrile, which can coordinatively interact with pal-

ladium centre in the η¹-bonded intermediate, can accelerate the

dynamic process.^5e,35The formation of 12c can also be rational-

ised by considering a rotation around C(sp³)–C(sp²) bond

starting from 12a via the intermediate B (see Scheme 8).

Variable temperature NMR studies and activation barriers (ꢂG^≠)

In order to determine the activation barriers for the exchange

process in some of the above complexes, we have carried out

variable temperature ¹H NMR studies in DMSO-d₆at 298–373

K. The coalescence temperatures and the corresponding ∆G^{≠ 36}

values are listed in Table 4. The data show that the change

of the substituent on the nitrogen from CHMe₂to CHMePh

of the diphosphazane ligand backbone increases the activ-

ation barrier of the isomerisation process as observed for the

unsubstituted allyl complexes 4 and 5 (see entries 1 and 2)

bearing Ph₂PN(CHMe₂)PPh(N₂C₃HMe₂-3,5) and Ph₂PN((S)*-

CHMePh)PPh(N₂C₃HMe₂-3,5) ligands, respectively. The in-

corporation of Ph or Me substituents at the 1,3-allylic positions

increases the free energy of activation for the exchange process

as compared to the unsubstituted allyl analogs (see entries 1

and 4, entries 2 and 3). The activation barrier for the allyl iso-

mers bearing P,S-ligand is less as compared to the allyl isomers

of P,N-ligand (see entries 3, 5 and 6). The activation barrier for

cis–trans isomerisation process is slightly less than that for the

syn–anti isomerisation process when the ligand and the allyl

moieties are the same (see entries 5 and 6).

Fig. 9 A view of the molecular structure of 12a showing the exo, syn/

syn-conﬁguration (the hydrogen atoms and the hexaﬂuorophosphate

anion are not shown).

Stereodynamic behaviour of allyl complexes 11–13 in solution

Trends in ³¹P chemical shifts

The phase sensitive 2D H–¹H NOESY and ROESY spectra

The ³¹P{¹H} resonances are assigned on the basis of the ∆δ

[∆δ = δ(complex) Ϫ δ(free ligand)] and sensitivity of the

chemical shifts upon substitution on the allyl moiety (see Table

1

of complex 11 in CDCl₃at 298 K shed light on the intimate

nature of the dynamic processes occurring in solution. From

D a l t o n T r a n s . , 2 0 0 3 , 1 0 1 6 – 1 0 2 7

1024

View Article Online

Table 4 Variable temperature ¹H NMR data^aand calculated free energy of activation for η³-allyl palladium complexes of diphosphazane ligands

Interchanging

isomers

Coalescence

temperature/K

Entry

∆ν/Hz

∆G^≠/kJ mol^{Ϫ1 b}

The protons for which coalescence observed

(1)

(2)

(3)

(4)

(5)

(6)

4a and 4b

5a and 5b

7a and 7b

8b and 8c

12a and 12b

12a and 12c

66.0

11.6

30.8

17.8

44.4

25.4

313

318

363

368

318

324

63.7

69.4

76.7

79.5

65.9

68.7

–N(CHMe₂) methyl protons

–N(*CHMePh) methyl protons

Allyl methyl protons trans to the coordinated phosphorus centre

–N(CHMe₂) methyl protons

^aDMSO-d₆was used as solvent. ^bCalculated according to the Shanan-Atidi and Bar-Eli method.³⁶The estimated error in the calculated free energies

of activation is 1 kJ mol^Ϫ1. ∆ν = frequency separation in the slow exchange limit.

Scheme 8

5). The chemical shift of one of the phosphorus nuclei for all

the palladium allyl complexes except for the P,P-coordinated

complexes (8b and 8c) lies in the range 70–74 ppm with ∆δ

being only 0.5–6 ppm. The constancy in the magnitude of

chemical shifts on changing the allyl part and the low ∆δ value

suggest that this chemical shift can be assigned to the non-

coordinated phosphorus. On the other hand, the chemical

shift of the other phosphorus nucleus shows a greater range

of variation (75–83 ppm). This phosphorus chemical shift lies

very much downﬁeld as compared to that of the free ligand and

the magnitude of ∆δ is in the range 31–36 ppm. Hence this

chemical shift is assigned to the –PPh₂phosphorus. For the

P,P-coordinated complexes (8b and 8c), the –PPh₂and the

–PPh(N₂C₃HMe₂-3,5) chemical shifts are assigned the values

62–63 and 69–73 ppm, respectively, based on the observed

chemical shifts for 1,3-diphenyl palladium allyl complex of the

ligand Ph₂PN(CHMe₂)PPh₂which features P,P-coordination.³⁷

For the P,S-coordinated complexes (11–13), the –P(S)Ph₂and

the –PPh(N₂C₃HMe₂-3,5) chemical shifts are assigned the

values 73–77 and 91–93 ppm, respectively, based on the the

magnitude of ∆δ and chemical shifts observed for palladium

allyl complexes of the ligands, Ph₂P(S)N(CHMe₂)PPh₂and

Ph₂P(S)N(CHMe₂)P(S)Ph₂.³⁷The ²J(P,P) values for P,N-

coordinated complexes of 1 and 2 lie in the range 28–30 Hz

Catalytic studies

We tested the catalytic activity of the two diphosphazane

ligands (1 and (SR)-2) in allylic alkylation reaction of the

model substrate, (rac)-1,3-diphenyl-2-propenyl acetate (I)

(Scheme 9) using dimethyl malonate as the nucleophile.³⁴

Catalytic precursors are generated in situ from [Pd(η³-C₃H₅)-

(µ-Cl)]₂(1 mol%) and the appropriate ligands (2.5 mol%).

The reaction is found to be complete within 24 h as shown by

1

the H NMR spectrum of the reaction mixture. However, only

low levels of enantiomeric excess (<30%) is observed in these

reactions (see Experimental).

The palladium catalysed allylic alkylation with (E)-3-phenyl-

2-propenyl acetate (III) is investigated to study the regio-

selectivity of these systems (Scheme 10). The use of an unsym-

metrical substrate, such as III in allylic alkylation would aﬀord

a mixture of linear and branched regio-isomers IV and V

depending on the nucleophilic attack at one or the other ter-

minal allyl carbon centres. The ¹H NMR spectrum of the

reaction mixture shows complete conversion of the substrate in

24 h at 298 K and the ratio of linear and branched isomer is

ca. 70 : 30 indicating that nucleophilic addition is more

favoured at the unsubstituted terminal carbon.

2

whereas J(P,P) values for P,P-coordinated allyl complexes (8b

Conclusions

The diphosphazane ligands 1–3 bearing

and 8c) are observed in the range of ∼110 Hz. The ²J(P,P)

values for the P,S-coordinated complexes of 3 are in the range

87–97 Hz.

a stereogenic

phosphorus centre exhibit diverse palladium allyl chemistry

D a l t o n T r a n s . , 2 0 0 3 , 1 0 1 6 – 1 0 2 7

1025

View Article Online

Table 5 The ³¹P{¹H} NMR data^afor the η³-allyl palladium complexes 4–13

Complex

–PPh(N₂C₃HMe₂-3,5) (δ_A)^b

∆δ_A

0.6

3.9

–P(E)Ph₂(δ_x)^b

∆δ_x

4a

4b

72.5 d (30.0)

72.5^c

75.6 d

75.3 d

77.7 d

78.4 d

79.5 d

76.9 d

75.4 d

78.3 d

82.3 d

81.2 d

78.3 d

83.4 d

80.2 d

62.4 d

63.1 d

81.3 d

81.8 d

77.1 d

76.6 d

74.4 d

75.8 d

75.5 d

74.0 d

73.5 d

31.7

31.4

30.4

31.1

35.6

33.0

31.5

34.4

35.0

33.9

31.0

36.1

36.3

18.5

19.2

34.0

32.9

11.8

11.3

9.1

5a

72.7 d (28.4)

73.0 d (29.0)

71.6 d (28.3)

71.3 d (30.2)

71.8 d (29.1)

70.1 d (27.5)

72.7 d (28.9)

73.7 d (27.8)

73.8 d (28.2)

73.0 d (29.4)

70.9 d (30.3)

73.1 d (114.3)

69.0 d (109.1)

70.8 d (28.8)

72.5 d (28.7)

91.1 d (95.3)

92.0 d (92.4)

92.8 d (91.9)

93.0 d (96.8)

93.3 d (91.2)

91.4 d (92.3)

93.1 d (87.2)

5b

4.2

6a

Ϫ0.3

Ϫ0.6

Ϫ0.1

Ϫ1.8

3.9

4.9

5.0

4.2

Ϫ1.0

1.2

Ϫ2.9

2.0

6.3

20.6

21.5

22.3

22.5

22.8

20.9

22.6

6b

6c

6d

7a

7b

7c

7d

8a

8b

8c

9

10

11a

11b

12a

12b

12c

13a

13b

10.5

10.2

8.7

8.2

^aThe ³¹P{¹H} NMR spectra were recorded in CDCl₃at 298 K except for 6 in which case a 1 : 1 mixture of CDCl₃and CH₂Cl₂was used. The ³¹P–³¹P

coupling constants are given in parentheses. ∆δ = δ(complex) Ϫ δ(free ligand). ^bd: doublet. ^cOverlapped by a signal arising from the –PPh-

(N₂C₃HMe₂-3,5) phosphorus of isomer 4a.

Scheme 9

Scheme 10

including various ﬂuxional processes in solution. The ligands 1

and 2 prefer P,N-coordination rather than the usually observed

P,P-coordination as the former mode of coordination elec-

tronically stabilises the palladium allyl bonding besides leading

to the formation of a stable six-membered chelate ring. The

usual P,P-coordination is observed only in the case of 1,3-

diphenyl allyl system (complex 8) probably because of the

higher electron releasing ability of two phenyl groups thus

obviating the necessity of having the σ-donor pyrazole nitrogen

to stabilise the metal allyl bonding. On the other hand, the

diphosphazane monosulﬁde ligand, Ph₂P(S)N(CHMe₂)PPh-

(N₂C₃HMe₂-3,5) (3) displays P,S-coordination. The P,S-

coordination is preferred owing to the formation of a stable

ﬁve-membered ring around the palladium centre. Thus the

choice of a particular coordination mode (P,N- or P,P- or P,S-)

for the ligands 1–3 is the result of combined eﬀect of chelate

ring size around the palladium centre as well as electronic

nature of the allyl moiety.

structure of P,S-coordinated complex (12a) reveals exo-con-

ﬁguration. The isomer observed in the solid state undergoes

slow isomerisation to produce a mixture of several isomers in

solution. The exchange pathway involves two well-known pro-

cesses, namely cis–trans isomerisation and syn–anti exchange

through a common σ-bonded intermediate. The σ-bonded

intermediate is generated by the opening of the η³-allyl moiety

at the trans position with respect to the greater π-acceptor

phosphorus centre and is thus subject to electronic control. The

activation barriers for exchange processes among various allyl

isomers bearing the P,S-ligand is less than that observed for the

allyl isomers formed by the P,N-ligands.

Acknowledgements

We thank the Department of Science and Technology (DST),

New Delhi, India for ﬁnancial support.

The allyl palladium complexes formed by these ligands (1–3)

exist as several isomers in solution. However, the solid state

structures reveal the presence of only one isomer. In the solid

state, the P,N-coordinated complexes (5a, 7a and 8a) show the

endo conﬁguration for the allyl moity whereas the solid state

References

1 (a) B. M. Trost and T. R. Verhoeven, in Comprehensive

Organometallic Chemistry, ed. G. Wilkinson, F. G. A. Stone and

E. W. Abel, Pergamon Press, Oxford, 1982, 8, 799; (b) P. von Matt

D a l t o n T r a n s . , 2 0 0 3 , 1 0 1 6 – 1 0 2 7

1026

View Article Online

and A. Pfaltz, Angew. Chem., Int. Ed. Engl., 1993, 32, 566;

(c) T. Hayashi, in Catalytic Asymmetric Synthesis, ed. I. Ojima, VCH

Publishers, New York, 1993, 325; and references cited therein;

(d ) R. Noyori, Asymmetric Catalysis in Organic Synthesis, Wiley,

New York, 1994, 82; (e) B. M. Trost and D. L. V. Vranken, Chem.

Rev., 1996, 96, 395; ( f ) G. Consiglio and R. M. Waymouth, Chem.

Rev., 1989, 89, 257; (g) C. G. Frost, J. Howarth and J. M. J. Williams,

Tetrahedron: Asymmetry, 1992, 3, 1089; (h) G. Helmchen,

J. Organomet. Chem., 1999, 576, 203.

Chem. Rev., 1994, 94, 1163; (c) P. Bhattacharyya and J. D. Woollins,

Polyhedron, 1995, 14, 3367.

13 S. K. Mandal, G. A. Nagana Gowda, S. S. Krishnamurthy,

C. Zheng, S. Li and N. S. Hosmane, Eur. J. Inorg. Chem., 2002, 2047.

14 D. D. Perrin, W. L. F. Armarego and D. R. Perrin, Puriﬁcation of

Laboratory Chemicals, Pergamon Press, Oxford, 3rd edn., 1988.

15 Y. Tatsuno, T. Yshida and S. Otsuka, Inorg. Synth., 1990, 28, 342.

16 P. R. Auburn, P. B. McKenzie and B. Bosnich, J. Am. Chem. Soc.,

1985, 107, 2033.

17 P. von Matt, G. C. Lloyd-Jones, A. B. E. Minidis, A. Pfaltz,

L. Macko, M. Neuburger, H. Rüegger and P. S. Pregosin, Helv.

Chim. Acta, 1995, 78, 265.

2 (a) P. S. Pregosin and R. Salzmann, Coord. Chem. Rev., 1996, 155,

35; (b) P. S. Pregosin and G. Trabesinger, J. Chem. Soc., Dalton

Trans., 1998, 727; (c) P. S. Pregosin, H. Rüegger, R. Salzmann,

A. Albinati, F. Lianza and R. W. Kunz, Organometallics, 1994,

13, 83; (d ) J. M. Canal, M. Gómez, F. Jiménez, M. Rocamora,

G. Muller, E. Duñach, D. Franco, A. Jiménez and F. H. Cano,

Organometallics, 2000, 19, 966; (e) A. Gogoll, C. Johansson,

A. Axén and H. Grennberg, Chem. Eur. J., 2001, 7, 396.

18 R. P. K. Babu, S. S. Krishnamurthy and M. Nethaji, Tetrahedron:

Asymmetry, 1995, 6, 427.

19 G. M. Sheldrick, SHELXS86, Program for solution of crystal

structures, University of Göttingen, Germany, 1986.

20 G. M. Sheldrick, SHELXS-97, Program for solution of crystal

structures, University of Göttingen, Germany, 1997.

21 H. D. Flack, Acta Crystallogr., 1983, A39, 876.

3 J. K. Whitesell, Chem. Rev., 1989, 89, 1581.

4 S. Murray, S. Hartley and F. Hartley, Chem. Rev., 1981, 81, 365.

5 (a) A. Pfaltz, Acta Chem. Scand. Ser. B, 1996, 50, 189 and references

cited therein; (b) S. Kudis and G. Helmchen, Angew. Chem., Int. Ed.,

1998, 37, 3047; (c) G. J. Dawson, C. G. Frost, J. M. J. Williams and

S. J. Coote, Tetrahedron Lett., 1993, 34, 3149; (d ) G. Helmchen,

S. Kudis, P. Sennhenn and H. Steinhagen, Pure Appl. Chem., 1997,

69, 513 and references cited therein; (e) J. M. Brown, D. I. Hulmes

and P. J. Guiry, Tetrahedron, 1994, 50, 4493.

22 A. Togni, Tetrahedron: Asymmetry, 1991, 2, 683.

23 (a) N. Baltzer, L. Macko, S. Schaﬀner and M. Zehnder, Helv. Chim.

Acta, 1996, 79, 803; (b) R. Fernández-Galán, F. A. Jalón, B. R.

Manzano, J. R. Fuente, M. Vrahami, B. Jedlicka, W. Weissensteiner

and G. Jogl, Organometallics, 1997, 16, 3758; (c) J. Powell and B. L.

Shaw, J. Chem. Soc. A, 1967, 1839; (d ) B. Åkermark, B. Kraken-

berger, S. Hansson and A. Vitagliano, Organometallics, 1987, 6,

620.

6 (a) D. A. Evans, K. R. Campos, J. S. Tedrow, F. E. Michael and

M. R. Gagné, J. Org. Chem., 1999, 64, 2994; (b) D. A. Evans,

K. R. Campos, J. S. Tedrow, F. E. Michael and M. R. Gagné,

J. Am. Chem. Soc., 2000, 122, 7905; (c) A. Albinati, P. S. Pregosin

and K. Wick, Organometallics, 1996, 15, 2419; (d ) A. Albinati,

J. Eckert, P. S. Pregosin, H. Rüegger, R. Salzmann and C. Stoggel,

Organometallics, 1997, 16, 579.

7 (a) T. Morimoto, K. Tachibana and K. Achiwa, Synlett., 1997, 783;

(b) J. C. Anderson, D. S. James and J. P. Mathias, Tetrahedron:

Asymmetry, 1998, 9, 753; (c) J. V. Allen, J. F. Bower and J. M. J.

Williams, Tetrahedron: Asymmetry, 1994, 5, 1895; (d ) K. Boog-

Wick, P. S. Pregosin and G. Trabesinger, Organometallics, 1998, 17,

3254.

8 G. Helmchen and A. Pfaltz, Acc. Chem. Res., 2000, 33, 336.

9 (a) O. Reiser, Angew. Chem., Int. Ed. Engl., 1993, 32, 547; (b) P. B.

Mackenzie, J. Whelan and B. Bosnich, J. Am. Chem. Soc., 1985, 107,

2046.

10 A. Togni, U. Burckhardt, V. Gramlich, P. S. Pregosin and

R. Salzmann, J. Am. Chem. Soc., 1996, 118, 1031.

11 (a) M. Ganesan, S. S. Krishnamurthy and M. Nethaji, J. Organomet.

Chem., 1998, 570, 247; (b) R. P. K. Babu, K. Aparna, S. S.

Krishnamurthy and M. Nethaji, Phosphorus, Sulfur Silicon Relat.

Elem., 1995, 103, 39; (c) R. P. K. Babu, S. S. Krishnamurthy and

M. Nethaji, Organometallics, 1995, 14, 2047; (d ) R. P. K. Babu,

S. S. Krishnamurthy and M. Nethaji, Polyhedron, 1996, 15, 2689;

(e) R. P. K. Babu, S. S. Krishnamurthy and M. Nethaji,

J. Organomet. Chem., 1993, 454, 157; ( f ) K. Raghuraman, S. S.

Krishnamurthy and M. Nethaji, J. Chem. Soc., Dalton Trans., 2002,

4289; (g) Part 16. K. Raghuraman, S. S. Krishnamurthy and

M. Nethaji, J. Organomet. Chem., in press.

24 (a) S. Sakaki, K. Takeuchi and M. Sugimoto, Organometallics, 1997,

16, 2995; (b) C. Carfagna, R. Galarini, K. Linn, J. A. Lòpez,

C. Mealli and A. Musco, Organometallics, 1993, 12, 3019;

(c) H. Fujimoto and T. Suzuki, Int. J. Quantum Chem., 1999, 74,

735.

25 J. W. Faller, M. E. Thomson and M. Mattina, J. Am. Chem. Soc.,

1971, 93, 2642.

26 C. Breutel, P. S. Pregosin, R. Salzmann and A. Togni, J. Am. Chem.

Soc., 1994, 116, 4067.

27 J. Hermann, P. S. Pregosin, R. Salzmann and A. Albinati,

Organometallics, 1995, 14, 3311.

28 T. Hosokawa, Y. Wakabayashi, K. Hosokawa, T. Tsuji and S.-I.

Murahashi, Chem. Commun., 1996, 859.

29 J. Sprinz, M. Keifer, G. Helmchen, M. Reggelin, G. Huttner,

O. Walter and L. Zsolnai, Tetrahedron Lett., 1994, 35, 1523.

30 S. Lanza, G. Bruno, F. Nicolò, A. Rotondo, R. Scopelliti and

E. Rondo, Organometallics, 2000, 19, 2462.

31 B. Crociani, S. Antonaroli, G. Bandoli, L. Canovese, F. Visentin and

P. Uguagliati, Organometallics, 1999, 18, 1137.

32 U. Burckhardt, V. Gramlich, P. Hoﬀmann, R. Nesper, P. S. Pregosin,

R. Salzmann and A. Togni, Organometallics, 1996, 15, 3496.

33 For a review, see e.g.: T. G. Appleton, H. C. Clark and L. Manzer,

Coord. Chem. Rev., 1973, 10, 335.

34 B. M. Trost and D. J. Murphy, Organometallics, 1985, 4, 1143.

35 (a) S. Hansson, P-O. Norrby, M. P. T. Sjögren, B. Åkermark,

M. E. Cucciolito, F. Giordano and A. Vittagliano, Organometallics,

1993, 12, 4940; (b) U. Burckhardt, M. Baumann, G. Trabesinger,

V. Gramlich and A. Togni, Organometallics, 1997, 16, 5252.

36 J. Sandström, Dynamic NMR Spectroscopy, Academic Press,

London, 1982.

12 For reviews on this topic see: (a) M. S. Balakrishna, V. S. Reddy,

S. S. Krishnamurthy, J. F. Nixon and J. C. T. R. Burckett St Laurent,

Coord. Chem. Rev., 1994, 129, 1; (b) M. Witt and H. W. Roesky,

37 S. K. Mandal, G. A. N. Gowda, S. S. Krishnamurthy, C. Zheng, S. Li

and N. S. Hosmane, J. Organomet. Chem., in press.

D a l t o n T r a n s . , 2 0 0 3 , 1 0 1 6 – 1 0 2 7

1027

Article Doi

Palladium(II) allyl complexes of chiral diphosphazane ligands: ambident coordination behaviour and stereodynamic studies in solution

DOI: 10.1039/b207447h

Source and publish data:

Authors:

Article abstract of DOI:10.1039/b207447h

Full text of DOI:10.1039/b207447h

Products guided by the article

R&D Labs maybe for 591231-70-8

Relevant to this article

Hot Product