Ligand ambivalence in pallada(platina)cyclic complexes of a rigid phosphine

Article abstract of DOI:10.1039/b307070k

Palladium(II) and platinum(II) complexes of a chiral pentacyclic phosphine, (1S, 4R, 4aS, 5aR, 6R, 9S, 9aS, 10aR)-4,6,11,11,12,12-hexamethyl-10-phenyl-dodecahydro-1,4:6, 9-dimethano-phenoxaphosphinine (phenop), show diverse structures dependent upon the chosen metal-containing starting material and reaction conditions. With palladium acetate, a P,C-cyclometallated dimeric complex [Pd(μ-κ ²-OAc)(μ-κ¹-OAc)(κP,κC ¹⁴-phenop)]₂, 4, is obtained through metallation at the C(14) methyl to form a six-membered chelate. The acetato bridged dimer is readily converted to the halo-bridged species [Pd(μ-X)(κP,κC ¹⁴-phenop)]₂, where X is chloride (5) or bromide (6). Reaction of one equivalent of phenop with Pd(COD)Cl₂ or Na ₂PdCl₄ gives a different phosphapalladacycle dimer [Pd(μ-Cl)(κP,κC⁸-phenop)]₂, 7, with a five-membered chelate and metallation at the C(8) methylene carbon. The analogous platinum derivative [Pt(μ-Cl)(κP,κC ⁸-phenop)]₂, 8, is obtained from the 1 : 1 reaction of phenop and K₂PtCl₄. An unusual ligand-ligand coupled product, 9, has been isolated in low yield from the reaction of phenop and Pd(COD)Cl₂. The zerovalent Pd(κP-phenop)₂, 10, and a monodentate silver(I) derivative, [Ag(κP-phenop)(CF₃SO ₃)], 11, have also been prepared. These new complexes have been fully characterised by spectroscopic and other techniques including single crystal X-ray structure determinations of phenop, 7-9 and 11.

Full text of DOI:10.1039/b307070k

View Article Online / Journal Homepage / Table of Contents for this issue
Ligand ambivalence in pallada(platina)cyclic complexes of a rigid
phosphine
K. M. Abdul Malik and Paul D. Newman*
Department of Chemistry, Cardiff University, PO Box 912, Cardiff, UK CF10 3TB.
E-mail: newmanp1@cardiff.ac.uk
Received 20th June 2003, Accepted 5th August 2003
First published as an Advance Article on the web 13th August 2003
Palladium() and platinum() complexes of a chiral pentacyclic phosphine, (1S, 4R, 4aS, 5aR, 6R, 9S, 9aS, 10aR)-
4,6,11,11,12,12-hexamethyl-10-phenyl-dodecahydro-1,4:6,9-dimethano-phenoxaphosphinine (phenop), show diverse
structures dependent upon the chosen metal-containing starting material and reaction conditions. With palladium
acetate, a P,C-cyclometallated dimeric complex [Pd(µ-κ2-OAc)(µ-κ1-OAc)(κP,κC14-phenop)]2, 4, is obtained through
metallation at the C(14) methyl to form a six-membered chelate. The acetato bridged dimer is readily converted to the
halo-bridged species [Pd(µ-X)(κP,κC14-phenop)]2, where X is chloride (5) or bromide (6). Reaction of one equivalent
of phenop with Pd(COD)Cl2 or Na2PdCl4 gives a different phosphapalladacycle dimer [Pd(µ-Cl)(κP,κC8-phenop)]2,
7, with a five-membered chelate and metallation at the C(8) methylene carbon. The analogous platinum derivative
[Pt(µ-Cl)(κP,κC8-phenop)]2, 8, is obtained from the 1 : 1 reaction of phenop and K2PtCl4. An unusual ligand–ligand
coupled product, 9, has been isolated in low yield from the reaction of phenop and Pd(COD)Cl2. The zerovalent
Pd(κP-phenop)2, 10, and a monodentate silver() derivative, [Ag(κP-phenop)(CF3SO3)], 11, have also been prepared.
These new complexes have been fully characterised by spectroscopic and other techniques including single crystal
X-ray structure determinations of phenop, 7–9 and 11.
Introduction
Heterocyclic phosphorus chemistry has been a highly active
research field as demonstrated in a recent collection of mono-
graphs.1 Unsaturated systems predominate, although a wide
range of saturated systems from 3-membered phosphiranes
to large-ring multiheteroatom species are known.2 The co-
ordination chemistry of small-ring unsaturates, especially the
Fig. 1 Structure and numbering of phenop.
phospholes, has been exploited by, amongst others, Mathey,3
Nelson4 and Leung.5 These workers and their groups have high-
lighted the rich chemistry that is possible with these hetero-
cycles particularly for making other phosphines by controlled
additions to coordinated phospholes. Mathey and coworkers
have also contributed hugely to the development of the satur-
ated systems, whilst Wild2 (phosphiranes, phosphepanes
and higher members), Marinetti6 (asymmetric phosphetanes
and others), Grützmacher7 (stable, fused-ring phosphiranes)
and Pringle8 (various) have all made significant additions to the
area especially regarding the complexation chemistry of such
donors. The discovery of the highly successful DuPHOS9
family of ligands in metal-based asymmetric hydrogenation
brought the phospholanes to the attention of a wider audience
and the chemistry of this ring size and smaller rings has benefit-
ted. However, the next largest member, the six-membered phos-
phinane class, has received little to no attention. We and others
have recently sought to address this oversight with the synthesis
of chiral phosphinane10 and oxaphosphinane11,12 derivatives.
We are interested in developing chiral monophosphines with
superstructures derived from readily available ‘chiral-pool’
precursors and employing them as ligands for broad-range
catalytic application. To this end, we recently reported the
stereoselective synthesis of the oxaphosphinane derivative
(1S, 4R, 4aS, 5aR, 6R, 9S, 9aS, 10aR)-4,6,11,11,12,12-hexa-
methyl-10-phenyl-dodecahydro-1,4:6,9-dimethano-phenoxa-
phosphinine, phenop (Fig. 1) and a complex with palladium()
acetate.12 The impetus for such work stems from the know-
ledge that many organic transformations catalysed by metal–
phosphine complexes are extremely sensitive to the steric and
electronic nature of the phosphine, and that bulky, electron-rich
phosphines are often more effective ligands than those that are
poor σ-bases and/or are sterically slight. This disparity is no
more evident than in the field of C–C and C–heteroatom coup-
ling chemistry where little to no catalytic activity is observed for
palladium based systems using, for example, trialkylphosphines
where the alkyl groups are small (methyl, ethyl) but efficient sys-
tems result when using tri-tert-butyl-, tricyclohexylphosphine or
mixed 1-adamantyl(tert-butyl)phosphines.13–15 Moreover, in the
newly developing field of alkyl–alkyl Suzuki coupling, extreme
ligand sensitivity is observed, so that, of the twelve or so phos-
phines tested by Fu and co-workers, only tricyclohexylphosphine,
and to a lesser extent tri(2-propyl)phosphine and tricyclopentyl-
phosphine, gave the desired product; tri-tert-butylphosphine was
ineffective.16 This highlights a recurring theme: that for selected
organic transformations catalysed by transition metal–phosphine
complexes, a narrow steric and/or electronic ‘window’ of activity
exists, such that only a very limited number of ligands are effect-
ive. For C–C coupling reactions performed at high temperatues
and/or over extended periods of time, cyclometallated-phosphine
complexes have often proved to be more efficient catalysts then
simple P-bound phosphine complexes.17 The cyclometallated
species appear to act as precatalysts for the active species, hence
their advantage as more robust catalyst reservoirs. The ready
availability of a large number of chiral phosphines meant the
transition from achiral C–C coupling to asymmetric variants
was an inevitable one. However, to date, surprisingly little has
been reported on asymmetric Suzuki, and to a lesser extent,
Heck couplings catalysed by phosphine–metal complexes, cyclo-
metallated or otherwise. This paper extends the initial studies of
the chemistry of phenop to include other palladium derivatives,
platinum and silver complexes and report an unusual ambi-
denticity in the formation of P,C chelates in the palladium()
complexes.
D a l t o n T r a n s . , 2 0 0 3 , 3 5 1 6 – 3 5 2 5
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
3516

View Article Online
mixture was stirred at Ϫ78 ЊC for 1 h, then allowed to slowly
Experimental
warm to room temperature and stirred for a further 0.5 h at
RT. The mixture was cooled back to Ϫ78Њ and a solution of
dichlorophenylphosphine (4.4 ml, 32.4 mmol) in THF (20 ml)
added dropwise with stirring. After stirring at low temperature
for 2 h, the solution was allowed to warm slowly to RT, stirred
overnight, subsequently quenched with water (5 ml) and taken
to dryness at the pump. The residue was dissolved in methanol
(300 ml) and oxidised by the dropwise addition of H2O2 (10 ml
of 30% solution) with ice cooling. After stirring overnight, the
excess hydrogen peroxide was destroyed by the addition of a
saturated aqueous solution of sodium sulfite (30 ml) and the
methanol removed in vacuo. The resultant gummy aqueous
residue was partitioned between dichloromethane (200 ml) and
water (200 ml) and the organic phase isolated and dried
(MgSO4). After filtering off the MgSO4ؒxH2O, the volatiles
were removed in vacuo to give a white solid that could be
recrystallised from petroleum ether if required. Yield = 12 g
General
The complexes were synthesised under nitrogen using standard
Schlenk line techniques. All solvents were freshly distilled
from sodium (toluene, 40/60 petroleum ether), sodium/benzo-
phenone (diethyl ether, tetrahydrofuran) or calcium hydride
(acetonitrile, ethanol) under nitrogen before use. The 31P NMR
spectra were recorded on a Jeol Eclipse 300 spectrometer oper-
ating at 121.7 MHz and referenced to 85% H3PO4 (δ = 0 ppm).
1H (400.13 or 300 MHz) and 13C (100 or 75.6 MHz) NMR
spectra were obtained on Bruker DPX400 and Jeol Eclipse 300
spectrometers and are referenced to tetramethylsilane (δ =
0 ppm). Mass-spectra were obtained on a VG Fisons Platform
II spectrometer. Microanalyses were performed by Warwick
Analytical Service Ltd. Optical rotations were acquired on an
Optical Activity Instruments polarimeter. All other chemicals
were of reagent grade and were used as supplied unless other-
wise stated.
1
(87%). 31P{1H} NMR (CDCl3, 121.7 MHz) δ 30.8 ppm. H
NMR (CDCl3, 300 MHz) δ 7.76 (2H, t br, J 10.1 Hz), 7.5–7.4
(4H, m), 4.58 (1H, ddd, J 26.6, 4.0, 2.4 Hz), 3.95 (1H, ddd,
J 14.1, 4.2, 2.2 Hz), 2.63 (1H, t, J 3.9 Hz), 2.05 (1H, m), 1.88
(1H, t, J 4.0 Hz), 1.77 (1H, m), 1.60 (2H, m), 1.30 (2H, m), 0.93
(3H, s), 0.92 (3H, s), 0.91 (3H, s), 0.90 (3H, s), 0.86 (3H, s), 0.84
(3H, s), 0.54 (1H, m) ppm. 13C DEPT NMR (CDCl3, 100 MHz)
δ 215.0 (s, CO), 213.5 (d, J 4.0 Hz, CO), 132.6 (d, J 34.3 Hz,
ipso-Ph), 132.4 (s, p-Ph), 131.8 (d, J 9.8 Hz, o-Ph), 129.0 (d,
J 10.1 Hz, m-Ph), 60.2 (s, C), 60.0 (d, J 3.2 Hz, C), 53.5 (d,
J 78.8 Hz, CH), 52.8 (d, J 79.3 Hz, CH), 47.6 (d, J 1.5 Hz, C),
47.5 (d, J 1.7 Hz, C), 46.3 (d, J 1.6 Hz, CH), 45.9 (d, J 1.6 Hz,
CH), 29.8 (s, CH2), 29.3 (s, CH2), 23.5 (d, J 10.0 Hz, CH2), 23.3
(d, J 13.3 Hz, CH2), 19.7 (s, CH3), 19.5 (s, CH3), 18.9 (s, CH3),
18.7 (s, CH3), 10.1 (s, CH3), 9.9 (s, CH3) ppm. Anal.: Calc. for
C26H35O3P: C, 73.20; H, 8.29%. Found: C, 73.2; H, 8.3%. [α]D =
ϩ 24.7Њ (c = 0.6, CH2Cl2). Mpt = 135–7 ЊC (dec).
X-Ray crystallography
The X-ray intensity data for the free ligand phenop, and the
metal complexes 7–9 and 11 were recorded on a Bruker Nonius
kappa CCD area detector at 150 K using monochromated Mo-
Kα radiation (λ = 0.71073 Å). The unit cell parameters for each
compound were initially determined from the reflections in a ꢀ
range of 10Њ, and finally refined using all the observed reflec-
tions in the data set. Data collection and processing were
carried out by using the programs COLLECT18 and DENZO.19
Empirical absorption corrections were applied to all data sets
using multiple and symmetry-related data measurements via the
program SORTAV.20 The structures were solved by direct
methods (SHELXS-96)21 and refined on Fo by full-matrix
2
least-squares (SHELXL-97)22 using all the unique data. In each
case, the non-hydrogen atoms were all refined anisotropically.
The crystal lattice of 7 contained voids with several low electron
density peaks suggesting that there were some disordered sol-
vent species (acetone). These did not fit into a readily recognis-
able geometry; however, eight such sites, two oxygens and six
carbons, each with occupancy 0.25, were included in the
SHELXL-97 refinement, with ISOR = 0.005 restraint. Com-
pound 8 also contained a half occupied 0.5 acetone per asym-
metric unit showing expected geometry; four sites for these
atoms (three carbons and one oxygen) were refined with occu-
pancy 0.5, C–C distance fixed at 1.52 Å and an ISOR = 0.003
restraint. Complex 9 contained a total of two toluene solvent
molecules per complex unit at four sites (one fully occupied,
one half occupied and two quarter occupied). These solvent
molecules were idealised (C–C = 1.39, phenyl–CH3 = 1.52 Å)
and refined with ISOR = 0.01 restraint. The hydrogen atoms of
the disordered acetone in 7 were ignored; all other hydrogen
atoms in the four compounds were included in calculated posi-
tions (riding model). The final values of the Flack parameter23
Ϫ0.05(10) (phenop), Ϫ0.08(3) (7), Ϫ0.006(12) (8), Ϫ0.06(4) (9)
and Ϫ0.02(4) (11) were all 0 within the limits of error, and this
indicated that the absolute stereochemistry has been deter-
mined correctly in each case. The crystal data and refinement
details are summarised in Table 1. All structures have been
reproduced using ORTEP.24
It is noteworthy that, although the endo,endo-isomer was
always obtained as the major product, small amounts of the
endo,exo-form were observed occasionally; this had no bearing
on the subsequent chemistry and the ultimate synthesis of
phenop.
(1S, 4R, 4aS, 5aR, 6R, 9S, 9aS, 10aR)-4,6,11,11,12,12-
hexamethyl-10-phenyl-dodecahydro-1,4:6,9-dimethano-phen-
oxaphosphinine oxide (phenop oxide). (1R,1ЈR,3S,3ЈS)-3,3Ј-
(phenylphosphoryl)bis(1,7,7-trimethylbicyclo[2.2.1]heptan-2-
one (10 g) was dissolved in 2% aq. ethanol and reduced by the
portionwise addition of solid sodium borohydride (4 g in total)
at 50Њ. The reduction was monitored by 31P{1H} NMR and was
deemed complete when the signal for the starting material at δP
30.8 ppm had disappeared (1–2 days, occasionally longer).
Excess borohydride was digested by the addition of saturated
ammonium chloride, the mixture reduced to small volume (∼50
ml) and partitioned between dichloromethane (200 ml) and
water (200 ml). The organic phase was isolated, washed with
water (2 × 100 ml), dried over MgSO4 and all volatiles removed
to give a white solid. This solid consisted of all possible isomers
of the diol. This mixture was then refluxed in toluene (200 ml)
containing para-toluenesulfonyl chloride (5 g) for 24 h. On
return the light-brown solution was cooled to RT, concentrated
to small volume (40 ml), diluted with CH2Cl2 (200 ml) and
washed thoroughly with water (100 ml), 10% NaOH solution
(100 ml) and water (100 ml) again. After drying, the solvents
were removed to give the desired product. If oily, the compound
could be encouraged to crystallise by stirring with 40/60 petrol-
eum ether. Recrystallisation was effected from hot toluene.
Yield = 5 g (52%). 31P{1H} NMR (CDCl3, 121.7 MHz) δ 26.2
CCDC reference numbers 213238–213242.
See http://www.rsc.org/suppdata/dt/b3/b307070k/ for crystal-
lographic data in CIF or other electronic format.
Preparations
1
(1R,1ЈR,3S,3ЈS )-3,3Ј-(phenylphosphoryl)bis(1,7,7-trimethyl-
bicyclo[2.2.1]heptan-2-one. To a solution of 1R-camphor (10 g,
65.7 mmol) in THF (250 ml) at Ϫ78 ЊC was added a solution of
n-BuLi in hexane (28 ml of 2.5 M, 70 mmol) via syringe. The
ppm. H NMR (CDCl3, 300 MHz) δ 7.80 (2H, m, o-Ph), 7.50
(3H, m, m-Ph, p-Ph), 3.63 (1H, dd obs, H5a), 3.61 (1H, dd,
J 7.5, 4.0 Hz, H4a), 3.00 (1H, m, endo-H8), 2.33 (1H, m, H9a),
2.27 (1H, dd, J 8.1, 3.8 Hz, H1), 2.04 (1H, m, endo-H7), 2.00
D a l t o n T r a n s . , 2 0 0 3 , 3 5 1 6 – 3 5 2 5
3517

View Article Online
Table 1 Crystal data and structure refinement for phenop, 7–9 and 11
Phenop
7
8
9
11
Empirical formula
C26H37OP
C52H72O2P2Cl2Pd2ؒ
0.5(C3H6O)
1103.78
C52H72O2P2Cl2Pt2ؒ
0.5(C3H6O)
1281.16
C52H68O2P2Cl2Pd2ؒ
2(C7H8)
1254.97
C54H74O8P2S2F6Ag2
Formula weight
396.53
1306.93
Crystal system
Space group
a/Å
b/Å
Orthorhombic
P2(1)2(1)2(1)
7.57200(10)
10.3049(2)
28.3367(6)
Monoclinic
P2(1)
Monoclinic
P2(1)
Orthorhombic
P2(1)2(1)2(1)
15.8838(4)
17.5364(5)
25.2450(8)
Orthorhombic
P2(1)2(1)2(1)
16.6065(2)
17.2601(2)
20.6222(3)
101.0500(11)
5910.94(13)
4
10.3021(2)
22.5573(4)
12.4212(2)
101.0630(8)
2832.89(9)
2
10.3287(2)
22.5278(5)
12.4452(3)
101.0500(11)
2842.10(11)
2
c/Å
β/Њ
U/Å3
2211.08(7)
4
7031.9(3)
4
Z
Reflections collected,
independent
Rint
Final R1, wR2 [I > 2σ(I )]
All data R1, wR2
14316, 4712
24754, 9970
11276, 7768
29630, 13557
43065, 11592
0.0560
0.0446, 0.1011
0.0620, 0.1113
0.0549
0.0438, 0.1065
0.0549, 0.1233
0.0501
0.0494, 0.1330
0.0542, 0.1513
0.0623
0.0593, 0.1501
0.0924, 0.1697
0.0619
0.0598, 0.1463
0.0902, 0.1646
(1H, t, J 4.4 Hz, H9), 2.00 (1H, m obs, H10a), 1.80 (2H, m, exo-
H2), 1.56 (1H, dt, J 15.0, 2.6 Hz, exo-H3), 1.48 (1H, m, exo-
H8), 1.33 (3H, s), 1.24 (1H, m, exo-H7), 1.13 (1H, m, endo-H3),
0.99 (3H, s), 0.92 (3H, s), 0.91 (1H, m obs, endo-H2), 0.89 (3H,
s), 0.86 (3H, s), 0.83 (3H, s) ppm. 13C DEPT NMR (CDCl3,
75.6 MHz) δ 137.7 (d, J 98.1 Hz, ipso-Ph), 130.8 (d, J 8.1 Hz,
o-Ph), 130.8 (obs, p-Ph), 128.5 (d, J 10.4 Hz, m-Ph), 89.8 (d,
J 5.8 Hz, C4a), 85.1 (d, J 3.5 Hz, C5a), 50.2 (s, C), 49.7 (s, C),
48.1–47.2 (obs, 2 × C, 1 × CH), 46.0 (d, J 2.3 Hz, CH), 38.5 (s,
CH), 37.6 (s, CH), 33.9 (s, CH2), 31.2 (d, J 12.7 Hz, CH2), 26.9
(s, CH2), 22.3 (s, CH3), 21.8 (s, CH3), 21.4 (d, J 3.5 Hz, CH2),
19.0 (s, CH3), 18.9 (s, CH3), 14.1 (s, CH3), 11.2 (s, CH3) ppm.
Anal.: Calc. for C26H37O2P: C, 75.68; H, 9.06%. Found: C, 76.1;
H, 9.1%. [α]D = ϩ 29.4Њ (c = 1.64, CHCl3). Mpt > 300 ЊC.
H14), 2.75 (2H, dd, J 8.8, 4.3 Hz, H14), 2.46 (2H, m, H9a), 2.38
(2H, m, endo-H7), 2.03 (8H, br, H1, CH3CO2), 1.75 (2H, br,
H9), 1.61 (2H, dd, J 10.7, 7.2 Hz, H10a), 1.7–1.4 (4H, m obs,
exo-H2, exo-H8), 1.48 (2H, m, exo-H3), 1.26 (2H, m, exo-H7),
1.03 (12H, s, H13, H17), 1.02 (4H, m obs, endo-H2, endo-H3),
0.88 (6H, s, H18), 0.78 (6H, s, H15), 0.66 (6H, s, H16) ppm. 13
C
DEPT NMR (C6D6, 75.6 MHz, 55 ЊC) δ 134.5 (d, J 55 Hz, ipso-
Ph), 132.4 (d, J 13.7 Hz, o-Ph), 129.7 (s, p-Ph), 128.3 (d, J 9.2
Hz, m-Ph), 88.0 (d, J 4.6 Hz, C4a), 83.9 (s, C5a), 56.8 (d, J 13.9
Hz, C1), 50.2 (s, C), 50.0 (d, J 1.5 Hz, C), 49.7 (s, C), 47.2 (obs,
C), 47.2 (s, C9), 40.6 (d, J 32.3 Hz, C9a), 39.0 (d, J 34.6 Hz,
C10a), 35.1 (s, C3), 27.3 (obs, 3 × CH2), 24.5 (obs, CH2,
CH3CO2), 25.5 (d, J 13.2 Hz, C8), 24.5 (t, J 7.5 Hz, C14), 19.9
(s, C13), 19.4 (s, C15), 18.2 (s, C16), 13.7 (s, C18), 11.9 (s, C17)
ppm. All other data were as reported previously.12
(1S, 4R, 4aS, 5aR, 6R, 9S, 9aS, 10aR)-4,6,11,11,12,12-
hexamethyl-10-phenyl-dodecahydro-1,4:6,9-dimethano-phenoxa-
phosphinine (phenop). A solution of phenop oxide (5 g,
12.12 mmol) in toluene (100 ml) containing trichlorosilane
(10 ml) was refluxed under nitrogen for 18 h. After cooling, the
mixture was hydrolysed carefully by the dropwise addition of
25% aq. NaOH with ice-cooling. The organic layer was washed
with degassed water (2 × 50 ml) then dried over MgSO4. After
filtering, the solvent was removed in vacuo and the oily residue
crystallised from ethanol. Yield = 4.56 g (95%). 31P{1H} NMR
cis-[Pd(ꢀ-Cl)(ꢁP,ꢁC14-phenop)]2, 5. To a solution of 4 (100
mg, 8.9 × 10Ϫ5 mol) in dichloromethane (5 ml) was added an
excess of tetraethylammonium chloride (100 mg, 3.60 × 10Ϫ4
mol) and the solution stirred for 12 h. On return the solution
was washed with water (2 × 10 ml), dried over MgSO4 and
filtered. Volatiles were removed under reduced pressure and the
resultant solid residue crystallised from acetone at 4 ЊC as pale
yellow crystals. Yield = 89 mg (96%). 31P{1H} NMR (C6D6,
121.7 MHz) δ 18.2 ppm. 1H NMR (C6D6, 300 MHz) δ 7.76 (4H,
t, J 8.2 Hz, o-Ph), 7.17 (4H, obs, m-Ph), 7.06 (2H, t, J 9.0 Hz,
p-Ph), 4.87 (2H, br, endo-H8), 3.20 (2H, d br, J 8.4 Hz, H5a),
3.11 (2H, dd, J 9.2, 4.1 Hz, H14), 3.06 (2H, t, J 7.1 Hz, H4a),
2.86 (2H, t, J 9.2 Hz, H14), 2.66 (2H, br, H9a), 2.45 (2H, br,
endo-H7), 2.10 (2H, br, H9), 1.99 (2H, br, H1), 1.6–1.2 (10H,
br), 0.99 (12H, s), 0.95 (12H, s), 1.0–0.8 (4H, obs), 0.68 (6H, s)
ppm. 13C DEPT NMR (CDCl3, 100 MHz) δ 135.6 (d, J 42.7 Hz,
ipso-Ph), 132.5 (d, J 9.2 Hz, o-Ph), 129.8 (s, p-Ph), 128.0 (d,
J 9.2 Hz, m-Ph) 87.7 (s, C4a), 83.4 (s, C5a), 57.7 (d, J 11.5 Hz,
C1), 50.8 (s, C), 49.9 (s, C), 49.6 (s, C), 47.7 (d, J 11.0 Hz, C),
47.5 (s, C9), 40.9 (d, J 31.2 Hz, C9a), 37.6 (d, J 35.8 Hz, C10a),
35.0 (s, C3), 29.2 (s, C14), 27.2 (d, J 15.0 Hz, C2), 26.9 (s, C7),
24.2 (d, J 15.0 Hz, C8), 20.2 (s, CH3), 19.7 (s, CH3), 18.9 (s,
CH3), 14.1 (s, CH3), 12.3 (s, CH3) ppm. Anal.: Calc. for
C52H72O2P2Cl2Pd2: C, 58.10; H, 6.77%. Found: C, 57.3; H,
6.7%. [α]D = ϩ 3.6Њ (c = 2.8, CHCl3). Mpt = 185 ЊC (dec).
1
(C6D6, 121.7 MHz) δ Ϫ36 ppm. H NMR (C6D6, 300 MHz)
δ 7.47 (2H, m, o-Ph), 7.2–7.0 (3H, m, p-Ph, m-Ph), 3.44 (1H, dt,
J 9.6, 1.5, 1.5 Hz, H5a), 3.28 (1H, d, J 7.9 Hz, H4a), 2.73 (1H,
m, endo-H8), 2.48 (1H, m, H9a), 2.20 (1H, m, endo-H7), 1.95
(1H, dd, J 7.6, 5.9 Hz, H10a), 1.87 (1H, dd, J 6.3, 4.0 Hz, H1),
1.72 (1H, t, J 4.4 Hz, H9), 1.67 (3H, d, J 1.4 Hz, H14), 1.67 (2H,
m obs, exo-H2, exo-H8), 1.48 (1H, m, exo-H3), 1.28 (1H, m,
exo-H7), 1.06 (3H, s), 1.0 (2H, m obs, endo-H2, endo-H3), 0.97
(3H, s), 0.85 (3H, s), 0.84 (3H, s), 0.83 (3H, s) ppm. 13C NMR
(C6D6, 100 MHz) δ 143.0 (d, J 24.0 Hz, ipso-Ph), 130.0 (d,
J 17.4 Hz, o-Ph), 127.2 (d, J 4.8 Hz, m-Ph), 126.4 (s, p-Ph) 88.3
(s, C4a), 84.1 (d, J 1.5 Hz, C5a), 48.6 (d, J 18.5 Hz, C1), 48.3 (s,
C), 48.0 (s, C), 47.9 (d, J 14.4 Hz, C9), 46.8 (s, C), 45.7 (d, J 4.6
Hz, C), 43.7 (d, J 17.2 Hz, C10a), 39.9 (d, J 12.5 Hz, C9a), 33.6
(s, C3), 29.1 (d, J 8.3 Hz, C2), 26.6 (s, C7), 23.2 (d, J 31.0 Hz,
C8), 21.3 (d, J 24.1 Hz, C14), 20.9 (s, CH3), 18.8 (s, CH3), 17.4
(s, CH3), 12.9 (s, CH3), 10.4 (s, CH3) ppm. Anal.: Calc. for
C26H37OP: C, 78.73; H, 9.42%. Found: C, 78.4; H, 9.3%. [α]D
ϩ 40.6Њ (c = 1, CHCl3). Mpt = 102–3 ЊC.
=
cis-[Pd(ꢀ-Br)(ꢁP,ꢁC14-phenop)]2.0.25n-Bu4NBr, 6. This com-
pound was prepared in a similar manner to 5 except using
n-Bu4NBr. Yield of yellow solid (recrystallised from acetone) =
85%. Anal.: Calc. for C56H81O2P2N0.25Br2.25Pd2: C, 54.03; H,
6.57; N, 0.28%. Found: C, 54.6; H, 6.9; N, 0.29%. The included
n-Bu4NBr could be removed by dissolution of the solid in
CH2Cl2 and washing with water. The following spectroscopic
and other data are for the salt-free complex. 31P{1H} NMR
(CDCl3, 121.7 MHz) δ 20.4 ppm. 1H NMR (CDCl3, 300 MHz)
cis-[Pd2(ꢀ-ꢁ1-OAc)(ꢀ-ꢁ2-OAc)(ꢁP,ꢁC14-phenop)2], 4. The
compound was prepared as before.12 31P{1H} NMR (C6D6,
1
121.7 MHz, 55 ЊC) δ 16.7 ppm. H NMR (C6D6, 300 MHz,
55 ЊC) δ 7.84 (4H, t, J 8.5 Hz, o-Ph), 7.2–7.0 (6H, m, p-Ph,
m-Ph), 4.47 (2H, m, endo-H8), 3.27 (2H, dd, J 8.9, 1.5 Hz,
H5a), 3.20 (2H, t, J 7.2 Hz, H4a), 2.93 (2H, dd, J 8.8, 5.2 Hz,
D a l t o n T r a n s . , 2 0 0 3 , 3 5 1 6 – 3 5 2 5
3518

View Article Online
δ 7.73 (4H, br, o-Ph), 7.33 (6H, br, m-Ph, p-Ph), 4.18 (2H, br,
endo-H8), 3.44 (2H, t, J 6.9 Hz, H4a), 3.44 (2H, obs, H5a), 2.92
(2H, br, H9a), 2.81 (2H, dd, J 8.9, 5.2 Hz, H14), 2.54 (2H, t,
J 8.9 Hz, H14), 2.12 (2H, dd, J 7.5, 4.2 Hz, H1), 2.00 (2H, m,
endo-H7), 1.78 (2H, dd, J 10.7, 7.1 Hz, H10a), 1.9–0.9 (14H,
m), 0.95 (6H, s), 0.85 (12H, s), 0.82 (12H, s) ppm. 13C DEPT
NMR (CDCl3, 100 MHz) δ 137.0 (d, J 42.1 Hz, ipso-Ph), 132.0
(d, J 10.4 Hz, o-Ph), 129.6 (s, p-Ph), 128.1 (d, J 10.4 Hz, m-Ph)
87.7 (s, C4a), 83.2 (s, C5a), 58.6 (d, J 15.0 Hz, C1), 51.0 (s, C),
50.1 (s, C), 49.9 (s, C), 47.8 (s, C9), 47.7 (s, C), 40.9 (d, J 31.2
Hz, C9a), 38.1 (d, J 34.7 Hz, C10a), 35.0 (s, C3), 30.3 (s, C14),
27.3 (d, J 15.0 Hz, C2), 24.3 (d, J 11.5 Hz, C8), 20.2 (s, CH3),
19.7 (s, CH3), 19.1 (s, CH3), 13.9 (s, CH3), 12.4 (s, CH3) ppm.
MS (APCI): 1083 (70%), [6 Ϫ Br]ϩ. [α]D = ϩ 11.8Њ (c = 3.3,
CH2Cl2). Mpt = 160–5Њ (dec).
MHz) δ 8.06 (4H, o-Ph), 7.05 (6H, p-Ph, m-Ph), 3.64 (2H, dd,
J 8.8, 7.0 Hz, H5a), 3.28 (2H, dd, J 7.1, 1.8 Hz, H4a), 3.11 (2H,
m, exo-H8), 2.60 (2H, dd, J 8.1, 3.3 Hz, H1), 2.51 (6H, s, H14),
2.50 (2H, m obs, endo-H7), 2.27 (2H, m, H9a), 2.04 (2H, dd,
J 12.7, 7.1 Hz, H10a), 1.78 (2H, t, J 7.0 Hz, H9), 1.5–1.3 (6H,
m), 1.09 (6H, s), 1.01 (6H, s), 0.95 (6H, s), 0.81 (6H, s), 0.77
(6H, s), 1.1–0.7 (4H, m obs) ppm. 13C DEPT NMR (CDCl3,
100 MHz) δ 134.5 (d, J 54.3 Hz, ipso-Ph), 132.9 (d, J 10.9 Hz,
o-Ph), 130.3 (s, p-Ph), 128.4 (d, J 11.0 Hz, m-Ph), 88.2 (s, C4a),
85.0 (s, C5a), 61.1 (d, J 8.7 Hz, C9), 49.6 (s, C), 49.5 (s, C), 48.2
(s, C), 46.7–46.4 (obs, C1, C8, C10a), 44.3 (d, J 19.6 Hz, C),
42.3 (d, J 43.9 Hz, C9a), 36.8 (s, C7), 33.8 (s, C3), 31.6 (d, J 12.1
Hz, C2), 25.5 (s, C14), 22.2 (s, CH3), 21.5 (s, CH3), 19.5 (s, CH3),
13.6 (s, CH3), 11.5 (s, CH3) ppm. Anal.: Calc. for C53.5H75-
O2.5P2Cl2Pt2: C, 50.15; H, 5.91%. Found: C, 50.4; H, 6.0%. [α]D
= ϩ 54.0Њ (c = 0.64, CH2Cl2). Mpt = 250–5 ЊC (dec). MS (APCI):
1218 (100%), [8 Ϫ Cl]ϩ. As for the palladium derivative, a minor
species was observed in solution (31P{1H} NMR, d = 23.9 ppm)
and believed to be the cis-isomer.
trans-[Pd(ꢀ-Cl)(ꢁP,ꢁC8-phenop)]2ؒ0.5(C3H6O), 7. A mixture
of Pd(1,5-cyclooctadiene)Cl2 (200 mg, 0.70 mmol) and phenop
(300 mg, 0.76 mmol) in methanol (12 ml) was refluxed for 14 h
with stirring. During this time a white microcrystalline solid
deposited. After cooling to room temperature and then 4 ЊC
overnight, the solid was filtered off, washed sparingly with cold
methanol and air-dried. The complex could be recrystallised
from acetone as colourless blocks. Yield = 350 mg (93%).
31P{1H} NMR (C6D6, 121.7 MHz) δ 42.8s ppm. 1H NMR
(C6D6, 300 MHz) δ 8.11 (4H, m, o-Ph), 7.10 (6H, m, p-Ph,
m-Ph), 3.85 (2H, m, exo-H8), 3.64 (2H, dd, J 9.2, 7.0 Hz, H5a),
3.22 (2H, d, J 6.9 Hz, H4a), 2.70 (2H, dd, J 7.6, 2.9 Hz, H1),
2.55 (2H, ddd, J 15.0, 8.5, 1.5 Hz, endo-H7), 2.47 (6H, s, H14),
2.30 (2H, m, H9a), 1.90 (2H, dd obs, J 6.9 Hz, H10a), 1.89 (2H,
t, J 5.0 Hz, H9), 1.78 (2H, m, exo-H7), 1.60 (2H, m, exo-H2),
1.43 (2H, dt, J 12.4, 2.5 Hz, endo-H2), 1.09 (6H, s), 0.98 (6H, s),
0.97 (2H, m obs, H3), 0.96 (6H, s), 0.72 (6H, s), 0.70 (2H, m
obs, H3), 0.69 (6H, s) ppm. 13C DEPT NMR (C6D6, 100 MHz)
δ 135.6 (d, J 37.1 Hz, ipso-Ph), 133.1 (d, J 12.6 Hz, o-Ph), 130.1
(s, p-Ph), 128.5 (d, J 10.2 Hz, m-Ph), 87.9 (s, C4a), 85.1 (s, C5a),
61.0 (d, J 12.7 Hz, C9), 51.7 (d, J 2.5 Hz, C8), 50.1 (s, C), 49.6
(s, C), 48.4 (d, J 20.4 Hz, C10a), 48.4 (s, C), 46.8 (s, C1), 45.2 (d,
J 26.0 Hz, C), 43.5 (d, J 35.5 Hz, C9a), 36.7 (s, C7), 33.8 (s, C3),
31.3 (d, J 11.3 Hz, C2), 25.1 (s, C14), 22.1 (s, CH3), 21.2 (s,
CH3), 19.2 (s, CH3), 13.6 (s, CH3), 11.5 (s, CH3) ppm. Anal.:
Calc. for C53.5H75O2.5P2Cl2Pd2: C, 58.20; H, 6.86%. Found: C,
57.7; H, 6.9%. [α]D = Ϫ 2.4Њ (c = 0.15, CH2Cl2). Mpt = 235–7 ЊC
(dec). MS (APCI): 1039 (100%), [7 Ϫ Cl]ϩ. It should be noted
that small amounts of a secondary species (presumably the
cis-isomer) were often observed in solution by 31P{1H} NMR
(δ = 41.8 ppm), but these always constituted less than 10% of
the total. Details for the isolation of 9: When 9 was detected in
the reaction mixture (δP = 2.5 ppm) after isolating 7, the solu-
tion was taken to dryness and 9 was isolated from the residue as
pale yellow blocks by crystallisation from toluene in air. Yield =
4 mg (1%). 31P{1H} NMR (CDCl3, 121.7 MHz) δ 2.5s ppm. 1H
NMR (C6D6, 300 MHz) δ 7.99 (4H, o-Ph), 7.38 (4H, m-Ph),
7.21 (2H, p-Ph), 4.12 (2H, d, J 5.7 Hz, H4a), 4.03 (2H, m, endo-
H7), 3.41 (2H, d, J 7.0 Hz, H5a), 3.34 (2H, m), 2.40 (2H, d, J 3.8
Hz, H9), 2.35 (2H, m), 2.00 (2H, m), 1.75 (2H, m), 1.61 (6H, s),
1.52 (6H, s), 1.33 (6H, s), 0.97 (6H, s), 0.94 (6H, s), 0.85 ppm.
MS(APCI): 1035 (100%), [9 Ϫ Cl]ϩ.
Pd(ꢁP-phenop)2ؒC7H8, 10. A solution of phenop (235 mg,
5.93 × 10Ϫ4 mol) and (η5-Cp)Pd(η3-C3H5) (63 mg, 2.96 × 10Ϫ4
mol) in toluene (20 ml) was heated at 75Њ with stirring for 3 h.
After cooling, the solution was filtered and concentrated to
small volume (∼ 5 ml). After diluting with MeOH (10 ml), the
solution was stored at Ϫ35Њ whereupon the desired compound
crystallised. The pale yellow crystals were isolated by filtration.
Yield = 222 mg (76%). The toluene of crystallisation could be
removed by pumping in vacuo if so desired. 31P{1H} NMR
1
(C6D6, 121.7 MHz) δ Ϫ7.2 ppm. H NMR (C6D6, 400 MHz)
δ 7.84 (2H, m, o-Ph), 7.17 (4H, t, J 7.2 Hz, m-Ph), 7.07 (2H, t,
J 7.2 Hz, p-Ph), 4.54 (2H, m, endo-H8), 3.44 (2H, d, J 9.3 Hz,
H5a), 3.33 (2H, d, J 7.2 Hz, H4a), 2.63 (2H, m, H9a), 2.51 (2H,
dd, J 8.4, 3.0 Hz, H1), 2.25 (2H, m, endo-H7), 2.13 (2H, m,
H10a), 1.95 (6H, s, H14), 1.76 (6H, m br, exo-H2, exo-H8, H9),
1.50 (2H, m, exo-H3), 1.35 (2H, m, exo-H7), 1.08 (6H, s, H17),
1.08 (4H, m obs, endo-H2, endo-H3), 0.97 (6H, s, H18), 0.95
(6H, s, H13), 0.88 (6H, s, H15), 0.78 (6H, s, H16) ppm. 13C
DEPT NMR (C6D6, 100 MHz) δ 146.6 (t, J 10 Hz, ipso-Ph),
131.9 (t, J 8.2 Hz, o-Ph), 128.0 (s, m-Ph), 127.7 (s, p-Ph), 89.8 (s,
C4a), 84.7 (s, C5a), 51.6 (t, J 7.2 Hz, C1), 49.6 (s, C), 49.3 (m
obs, C), 49.2 (s, C9), 48.2 (s, C), 46.9 (d, J 3.4 Hz, C), 45.2 (t,
J 5.5 Hz, C10a), 40.4 (t, J 8.6 Hz, C9a), 34.4 (s, C3), 30.7 (t,
J 5.8 Hz, C2), 27.8 (s, C7), 25.5 (t, J 13.2 Hz, C8), 24.5 (t, J 7.5
Hz, C14), 22.3 (s, C13), 19.8 (s, C15), 18.5 (s, C16), 14.3 (s,
C18), 11.5 (s, C17) ppm. Anal.: Calc. for C59H82O2P2Pd: C,
71.45; H, 8.35%. Found: C, 71.8; H, 8.2%. MS (APCI): 522
(40%), [10 Ϫ phenop ϩ NH4ϩ].
[Ag(ꢁP-phenop)(CF3SO3)]2, 11. A solution of Ag(CF3SO3)
(91 mg, 3.53 × 10Ϫ4 mol) and phenop (140 mg, 3.53 × 10Ϫ4 mol)
were stirred in diethyl ether (15 ml) in the absence of light for
24 h. On return, the solution was diluted with 40/60 petroleum
ether (15 ml) and left at Ϫ30 ЊC overnight. The resultant crys-
tals were filtered off and dried at the pump. Yield = 140 mg
(61%). A second crop was obtained on further cooling of the
solution. Yield = 50 mg (22%). 31P{1H} NMR (CDCl3, 121.7
1
1
MHz) δ Ϫ12.1d (1JAg–P = 763 Hz, JAg–P = 879 Hz) ppm. H
NMR (CDCl3, 300 MHz) δ 7.55 (2H, m), 7.47 (3H, m), 3.56
(2H, dd, J 7.7, 1.5 Hz), 2.83 (1H, br), 2.42 (1H, m), 2.28 (1H,
m), 2.08 (1H, dd, J 11.4, 3.8 Hz), 1.90 (4H, m), 1.62 (1H, m),
1.52 (3H, d, 3.5 Hz), 1.33 (1H, m), 1.15 (2H, m), 0.98 (3H, s),
0.95 (3H, s), 0.93 (9H, s), 0.92 (3H, s) ppm. 13C DEPT NMR
(CDCl3, 100 MHz) δ 135.5 (d, J 33 Hz, C), 131.7 (d, J 16 Hz,
CH), 130.9 (s, CH), 129.6 (d, J 10 Hz, CH), 89.5 (s, CH), 84.2 (s,
CH), 51.2 (d, J 9 Hz, CH), 49.7 (s, C), 48.7 (s, C), 47.6 (s, CH),
47.4 (s, C), 43.7 (d, J 16 Hz, CH), 39.4 (d, J 22 Hz, CH), 33.8 (s,
CH2), 30.8 (d, J 14 Hz, CH2), 27.3 (s, CH2), 23.9 (d, J 23 Hz,
CH2), 23.2 (d, J 14 Hz, CH3), 21.6 (s, CH3), 19.7 (s, CH3), 18.7
(s, CH3), 13.9 (s, CH3), 11.1 (s, CH3) ppm.
trans-[Pt(ꢀ-Cl)(ꢁP,ꢁC8-phenop)]2ؒ0.5(C3H6O), 8. A suspen-
sion of Na2PtCl4 (98 mg, 2.56 × 10Ϫ4 mol) and phenop (105 mg,
2.65 × 10Ϫ4 mol) in n-butanol (15 ml) was refluxed with stirring
for 24 h. The initially insoluble platinum salt dissolved during
this time with the formation of a light-brown solution. After
cooling, the solution was filtered and concentrated to small
volume (3 ml). Cooling at 4 ЊC overnight led to the precipitation
of a colourless, crystalline solid. Filtered off solid and washed
sparingly with n-butanol. Recrystallisation was effected from
acetone. Yield = 150 mg (94%). 31P{1H} NMR (C6D6, 121.7
1
MHz) δ 24.4s (1JPt–P = 5529 Hz), ppm. H NMR (C6D6, 300
D a l t o n T r a n s . , 2 0 0 3 , 3 5 1 6 – 3 5 2 5
3519

View Article Online
Scheme 1 (i) nBuLi, 0.5 PhPCl2, H2O2, 87%; (ii) NaBH4, 93%; (iii) TsCl, 52%; (iv) HSiCl3, 93%.
31P{1H} NMR (C6D6, 121.7 MHz) δ Ϫ16.4d (1JAg–P = 730 Hz,
stereochemistry of the diol is not critical and that a single iso-
mer of 3 can be obtained in >50% yield from the diastereomeric
mixture of diols.
1
1JAg–P = 843 Hz) ppm. H NMR (C6D6, 300 MHz) δ 7.43 (4H,
dd, J 8.0, 7.1 Hz, o-Ph), 7.17 (4H, m, m-Ph), 7.08 (2H, t, J 7.4
Hz, p-Ph), 3.20 (2H, d, J 9.0 Hz, H5a), 3.05 (2H, d, J 7.4 Hz,
H4a), 2.48 (4H, m, endo-H8, H9a), 2.03 (2H, dd, J 10.5, 2.9 Hz,
H1), 1.94 (2H, m, endo-H7), 1.9–1.5 (6H, m), 1.61 (6H, d, J 3.0
Hz, H14), 1.35 (2H, m, exo-H3), 1.22 (2H, m, exo-H7), 0.9–0.7
(4H, obs, endo-H2, endo-H3), 0.89 (6H, s), 0.86 (6H, s), 0.83
(6H, s), 0.76 (6H, s), 0.67 (6H, s) ppm. 13C DEPT NMR (C6D6,
100 MHz) δ 136.8 (d, J 33.5 Hz, ipso-Ph), 131.7 (d, J 16.2 Hz,
o-Ph), 130.1 (s, p-Ph), 129.1 (d, J 9.8 Hz, m-Ph), 89.2 (s, C4a),
84.1 (s, C5a), 50.2 (d, J 9.2 Hz, C1), 49.4 (s, C), 48.6 (s, C), 47.6
(s, C9), 47.0 (d, J 7.5 Hz, C), 43.9 (d, J 16.2 Hz, C10a), 39.0 (d,
J 21.9 Hz, C9a), 33.8 (s, C3), 30.4 (d, J 12.7 Hz, C2), 27.2 (s,
C7), 23.6 (d, J 23.1 Hz, C8), 23.0 (d, J 12.7 Hz, C14), 21.3 (s,
CH3), 19.4 (s, CH3), 18.2 (s, CH3), 13.7 (s, CH3), 11.0 (s, CH3)
ppm. Anal.: Calc. for C27H37O4PF3SAg: C, 49.62; H, 5.72%.
Found: C, 49.6; H, 5.7%. [α]D = ϩ24.3Њ (c = 1.9, CH2Cl2). Mpt =
135–7Њ. MS (APCI): 505 (50%), [11 Ϫ CF3SO3]ϩ.
The structure of phenop is shown in Fig. 2. The structure is
very similar to that of the previously reported phosphine oxide
3, and it is clear that the chiral integrity of the compound is
retained upon reduction. The phosphorus centre in phenop is
described as pseudochiral because, although two of the substi-
tuents are constitutionally equivalent, they have the opposite
chirality; one α-carbon (C1) is S while the other (C12) is R. The
ligand is readily recrystallised from alcohols and is air-stable in
the solid-state. As emphasised previously, the central six-mem-
bered ring is fixed in the boat conformation as dictated by the
rigid nature of the ligand. This introduces the notion of pos-
sible coordination of the oxygen of the oxaphosphinane ring to
give bidentate or hemilabile P∧O chelates. This is unlikely here
because of steric impedance by a methyl group and a di-
methylene bridge, but may be feasible for a related ligand
derived from norcamphor; this is currently under investigation
in our laboratories. The NMR spectra of phenop accord with
the X-ray structure, and pertinent resonances are presented in
Table 2. Full assignments (with respect to the labelling scheme
Results and discussion
1
1
1
The synthesis of phenop is highlighted in Scheme 1. The proto-
col presented here is more reliable than the previously reported
route12 that has, since the original preparation, proven to be
poorly reproducible with highly variable yields for a number of
the synthetic steps. Unlike the earlier synthesis, here the di-
ketophosphine is not isolated but oxidised immediately to give
the phosphine oxide 1 which is obtained as a mixture of
exo,endo (minor) and endo,endo (major) isomers that can be
separated by fractional cystallisation from 40/60 petroleum
ether. Often, only the endo,endo-isomer is isolated. Sodium
borohydride reduction of either isomer gives essentially the
same diastereomeric mixture of the diol precursor 2; no one
isomer was dominant, although the actual proportion of each
did vary from one batch to another. Indeed, if no effort is made
to isolate the diastereomeric ketones the same mixture (contain-
ing at least four isomers) of diols, 2, is obtained on reduction.
No attempt was made to isolate this diastereomeric mixture of
diols and after reflux with 2 mol equivalents of para-toluene-
sulfonyl chloride in toluene the oxide of the desired ligand (4)
was obtained in 52% yield. Phenop was produced after exhaust-
ive reduction of the oxide 3 using trichlorosilane. The reduction
is slow, presumably as a consequence of the steric bulk about
the phosphorus centre, and the oxide requires refluxing with
excess trichlorosilane in toluene for 24 to 48 h.
in Fig. 1), as deduced from H, H{31P}, H-1H-COSY, 2D-
NOESY, H-13C HMQC and where necessary H-13C HMBC
NMR techniques appear in the Experimental. The H4a and
H5a protons of phenop are distinguished by the presence of a
four-bond coupling for the latter (a ‘W’ relation to exo-H7) that
is absent for the former, whilst the H1 and H9 protons are
distinguished by the presence or absence of a coupling to their
neighbouring methine protons H10a and H9a, respectively.
1
1
1
Unlike in the H NMR spectrum of phenop oxide, there is no
discernible 3JH–P coupling for the methine protons H4a and H5a
in the corresponding spectrum of phenop. Contact deshielding
leads to the endo-H7, endo-H8 and the C14 protons being
observed downfield of the other methylene and methyl reson-
ances. It is also noteworthy that the C14 methyl is coupled to
the phosphorus (J = 1.4 Hz). Although the coupling constant is
small, it is unlikely to occur via a through-bond mechanism
(the two nuclei are separated by five bonds in an orientation
that disfavours long-range coupling) and is considered to
result from through-space coupling. The P–C14 separation is
only 3.41 Å in the crystal. In addition, the endo-H8 hydrogen
that resides in close proximity to the phosphorus shows a
greater JH-P coupling of 6.5 Hz. Neither of these couplings are
Aside from the reproducibility of this procedure, a further
advantage is the fact that no separation of diastereomeric
mixtures is required at any stage as the cyclisation to give 3
produces but a single isomer of phenop oxide, determined pre-
viously to be the (1S, 4R, 4aS, 5aR, 6R, 9S, 9aS, 10aR) form.
The formation of the oxaphosphinane ring is likely to proceed
through an initial dehydration reaction to give a phosphine
oxide species with a single remaining alcohol functionality and
a vinyl group α to the P᎐O function on the other bicyclic unit.
᎐
This compound is thus set up for an internal Michael type addi-
tion of the alcohol oxygen to the vinyl β carbon giving 3. The
fact that the stereochemistry is endo,exo- in 3 suggests that the
cyclisation is stereoselective, with residual exo-orientated alco-
hol groups adding to the alkenic β-carbon from the endo-face
and vice versa. Such a mechanism explains why the initial
Fig. 2 ORTEP24 representation of phenop.
D a l t o n T r a n s . , 2 0 0 3 , 3 5 1 6 – 3 5 2 5
3520

View Article Online
observed for phenop oxide reflecting the through space influ-
ence of the lone pair.
It has been shown that the reaction of phenop with pal-
ladium() acetate gives a cyclometallated dimer, [Pd(µ-κ2-
OAc)(µ-κ1-OAc)(κP,κC14-phenop)]2, 4, where the site of metal-
lation is an exo-methyl, namely the C14 carbon centre of Fig. 1.†
The bridging acetates can be replaced by chloride or bromide
by simple metathesis in dichloromethane to give the corre-
sponding halide bridged dimers [Pd(µ-X)(κP,κC14-phenop)]2, 5
(X = Cl) and 6 (X = Br), with the same structure as the acetate.
Such an assignment is based on spectroscopic data, where the
C14 methyl doublet at δ = 21.3 ppm in the 13C{1H} NMR spec-
trum of phenop is lost in the corresponding spectrum of the
complexes and replaced by an extra CH2 resonance as a result
of metallation of the said methyl (Table 2). The extra methylene
is a singlet in the 13C{1H} NMR showing no through-metal
coupling to the phosphorus. A similar situation is observed in
the 1H NMR where only three singlets are present for the
methyl groups of 5 and 6 (two sets of two being coincident).
1
Most of the resonances in the H NMR spectrum of 4 are
broadened at room temperature, but become well-resolved at
higher temperature. The methylene protons of the coordinated
carbon are diastereotopic and are observed as distinct doublets
of doublets when the 1H NMR spectrum of 4 is run at 55 ЊC in
C6D6 (Table 2). These protons have a geminal coupling constant
of 8.8 Hz and each couples further to the phosphorus nucleus
with 3JP–H of 5.2 and 4.3 Hz, respectively. Although some of the
1
resonances in the H NMR spectra of cis-[Pd(µ-Cl)(κP,κC14-
phenop)]2, 5, and cis-[Pd(µ-Br)(κP,κC14-phenop)]2, 6, were
broadened at room temperature, most of the relevant signals
were resolved and assignments could be made at this temper-
ature (Table 2). The broadening in the NMR spectra of 4 was
interpreted12 as resulting from inversion through the diacetate
bridge as noted in related N,C systems.25 Such a process may
explain the slightly broadened spectra of 5 and 6, but low tem-
perature studies were precluded by the poor solubility of the
compounds. Aside from the loss of a methyl signal and the
presence of the extra CH2 protons in the 1H NMR spectrum of
1
4, 5 and 6, other significant differences between the H NMR
spectra of these complexes and the uncoordinated ligand
include a downfield shift for the endo-H8 to δ values > 4 ppm as
opposed to 2.73 ppm for the free ligand and the presence of a
1
JP–H4a coupling in the H NMR spectra of the complexes. The
downfield shift of the endo-H8 proton may reflect its position
over the metal in the complexes; in the crystal structure of 4 it
resides over the palladium at a distance of 2.466 Å. The coup-
ling of H4a to the 31P nucleus is unique to these P∧C14 chelated
complexes.
When Pd(1,5-COD)Cl2 or Na2PdCl4 is reacted with 1 mol
equivalent of phenop, the compound trans-[Pd(µ-Cl)(κP,κC8-
phenop)]2, 7, is obtained in high yield. The crystal structure of
the complex is shown in Fig. 3 with selected bond lengths and
angles presented in Table 3. Remarkably, the phenop ligands in
this dimeric species have metallated at a different site to that
observed for the acetate complex; a methylene carbon of one of
the flanking six-membered carbocyclic rings, i.e. C8 (see Fig. 1),
as opposed to the exo-methyl C14. The overall structure is a
butterfly with the chorides as a hinge as seen in most palladium
dimers of this type. The smallest chelate is now five-membered
as opposed to six-membered in 4, 5 and 6. The average Pd–P
bond length (2.208 Å) in 7 is relatively short, but longer than
that observed in 4 (2.197 Å) as are the Pd–C bonds (av. 2.052 Å
compared to 2.033 Å);12 they do, however, accord with liter-
ature values for related cyclopalladated complexes.26 The Pd–Cl
bonds trans to the cyclometallated carbons are longer than
those trans to the phosphorus donors and it is clear that the
† The superscript on the κC in the formula indicates the site of
metallation, in this case the C14 carbon, with respect to the numbering
in Fig. 1.
D a l t o n T r a n s . , 2 0 0 3 , 3 5 1 6 – 3 5 2 5
3521

View Article Online
D a l t o n T r a n s . , 2 0 0 3 , 3 5 1 6 – 3 5 2 5
3522

View Article Online
1
platinum-195 in the sample with a JP–Pt coupling constant of
5529 Hz.
The exo-H8 proton resonance shifts downfield when its
carbon is bound to palladium and/or platinum as observed in
the 1H NMR spectra of 7 and 8. The named proton is seen as a
multiplet at δ 3.85 (7) and 3.11 ppm (8) that shows no observ-
able coupling to 31P. The cyclometallated carbon is observed as
2
a doublet at δ = 51.7 ppm with JC–P = 2.5 Hz in the 13C{1H}
NMR spectrum of trans-[Pd(µ-Cl)(κP,κC8-phenop)]2, 7, but is
obscured in the 13C{1H} NMR spectrum of the corresponding
platinum complex. No interconversion of 5 7 or vice versa
was observed on heating the complexes to 120 ЊC and the differ-
ent sites of metallation are likely to be the result of kinetic
factors. We are currently seeking simple P-bound complexes of
phenop with palladium halides and actetates in an effort to
identify the contributing factors.
Fig. 3 Molecular structure of 7.
Although the complex trans-[Pd(µ-Cl)(κP,κC8-phenop)]2, 7,
is obtained in good yield from either Pd(1,5-COD)Cl2 or Na2-
PdCl4, 31P{1H} NMR analysis of the reaction mixtures did, on
occasion, reveal the presence of several minor species, one of
which was isolated pure and its structure determined by X-ray
crystallography as shown in Fig. 5. Like the previous structures
(Fig. 3, ref. 12), the compound is a dimer, however here the
secondary ligands (chlorides) are not bridging but bound as
simple terminal donors at single metal sites. Instead, the two
phenop ligands have coupled through the C8 carbons (refer to
Fig. 1 for assignments) to form a new C–C bond connecting the
two ligands. These carbons (denoted C14 and C40 in Fig. 5)
form the dimeric bridge by binding both palladium centres
while the C15 and C41 carbons each coordinate a single Pd();
this is the only example where C–H activation at the C7 carbons
has occurred (Fig. 1). The bridging carbons are those that are
cyclometallated in the non-coupled product (Fig. 3). This C4
link does not appear to be a butendiyl of the classical type
because each of the C–C bond lengths are 1.50 0.05 Å indi-
cating essentially single carbon–carbon bonds throughout the
link. Unfortunately, this compound was only isolated in very
low yield on one occasion and it was only possible to acquire
some basic spectroscopic data (31P, 1H NMR) and a mass
spectrum in addition to the X-ray structure. It is included here
as an example of a ligand–ligand coupled product that may
have relevance to other P∧C cyclometallates that are used as
catalysts.
trans influence of the alkyl donor is greater than that of the
phosphine. Unlike 4, the phosphorus donors (and the metal-
lated carbons) in 7 are orientated trans- with respect to the
Pd–Pd axis. The P–Pd–C bond angles in 7 average 81.8Њ as
compared to 86.7Њ in 4, consistent with the smaller chelate
ring size of the former. As for cis-[Pd(µ-κ1-OAc)(µ-κ2-OAc)-
(κP,κC14-phenop)]2 (4), two new chiral centres are generated
stereoselectively on formation of trans-[Pd(µ-Cl)(κP,κC8-
phenop)]2 (7). The stereogenic phosphorus centres in 7 have the
absolute configuration R, and the C8 carbons (labelled C1 and
C27 in Fig. 3), which also become chiral on metallation (co-
ordination) have the S stereochemistry. In 4 the stereogenic
phosphorus is S and the second new chiral centre that is gener-
ated on cyclometallation, the tertiary carbon C11, is also S.
Ambivalence over the site of C–H activation in palladacycles
has been observed previously in nitrogen donor systems,27 but
to our knowledge this is the first example of such ambidenticity
in P,C-chelates and the first where such discretion can generate
two different diastereomers. The platinum analogue of 7,
namely 8, is isolated in excellent yield (94%) after refluxing a
suspension of Na2PtCl4 with one mol equivalent of phenop in
n-butanol. The structure (Fig. 4, Table 3) is analogous to the
palladium complex 7.
Fig. 4 Molecular structure of 8.
Fig. 5 Molecular structure of 9.
The different site of metallation produces fused six- and
seven-membered chelates in 4, 5 and 6, and fused five- and
The reaction of 2 mol of phenop with (η5-Cp)Pd(η3-C3H5) in
toluene generates the zerovalent bis(phosphine) complex
Pd(κP-phenop)2, 10, in excellent yield. The compound is readily
recrystallised from toluene/methanol mixtures as pale yellow to
golden brown crystals. However, the crystals were solvent-
dependent aggregates that precluded any single crystal X-ray
analysis. The 31P{1H} NMR spectrum of the complex is the
expected singlet whilst the H NMR spectrum, which is well
resolved at room temperature, shows the endo-H8 and the
H14 protons shifted to low field (see above). The ortho-phenyl
seven-membered chelates in 7. This generates quite distinct 31
P
chemical shifts for the two sets of complexes (Table 2), where
the former have δP values around 20 ppm and the latter a δP of
42 ppm; this downfield shift is consistent with the presence of a
five-membered chelate in 7 as opposed to the six-membered
palladacycle in 4–6. The 31P{1H} NMR spectrum of the Pt()
dimer 8 shows a singlet at 24.4 ppm (this is, as expected, upfield
of the palladium analogue) and two satellites from the 33% of
1
D a l t o n T r a n s . , 2 0 0 3 , 3 5 1 6 – 3 5 2 5
3523

View Article Online
Scheme 2
protons show two couplings to phosphorus and appear as a
0.172 Å. The Ag–P bond lengths lie in the range observed for a
number of related systems.28 The phenop ligands are clearly
bound as monodentates through the phosphorus donors, with
both carbon centres that cyclometallate in the two distinct sets
of palladium() complexes in close proximity to the silver
centre with average distances of Ag ؒ ؒ ؒ C8 = 3.49 and
Ag ؒ ؒ ؒ C14 = 3.21 Å.‡ The latter is just within the sum (3.29 Å)
of the van der Waals radius of a methyl group and the ionic
radius of a silver() ion.29 The Ag ؒ ؒ ؒ H distance to the fixed
endo-H8 hydrogen averages as 2.54 Å. The 31P{1H} NMR spec-
trum of 11 consists of two doublets through coupling to the
two nuclei of the metal with the one-bond 109Ag–31P doublet
having a larger value (843 Hz) compared to the 107Ag–31P coup-
ling constant (730 Hz) by virtue of its greater gyromagnetic
ratio. Unlike a number of related complexes,30 these individual
coupling constants and the resultant two doublet pattern is
clear in the 31P{1H} NMR spectrum of 11 at room temperature;
others typically show an averaged broad singlet at this temper-
ature, although the couplings are clear in the 31P{1H} NMR
1
multiplet in the normal H NMR spectrum that collapses to a
doublet in the 1H{31P} NMR spectrum. Similarly, a number of
the carbon resonances occur as virtual triplets in the 13C{1H}
6
NMR spectrum of 10, with a long range JC–P coupling of 7.5
Hz being observed for the C14 carbon. The presence of these
couplings in the 13C NMR spectrum reflects the trans arrange-
ment of the phosphines in this bis(phenop)Pd complex. The
complex is air-stable in the solid state, with no visible decom-
position being observed over several days.
Addition of 1 mol equivalent of phenop to a diethyl ether
solution of silver() triflate gives the dimeric species [Ag(κP-
phenop)(CF3SO3)]2, 11, as the only isolable complex. The com-
pound was isolated as small well-formed colourless crystals that
showed little tendency to decompose in light in the solid-state,
but proved to be slightly sensitive to decomposition in solution
when left exposed to light. The dimeric structure of the com-
plex is shown in Fig. 6 with selected bond lengths and angles
collected in Table 3. The structure consists of two three-co-
ordinate silver() centres with trigonal planar geometry (sum of
angles about silver are 358.4 and 350.8Њ, respectively) bridged
by two µ-κ2-CF3SO3 ligands. The geometry about the Ag(2)
centre is considerably more distorted then that around the
Ag(1) ion, with a difference in the two Ag(2)–O bond lengths of
1
spectrum of [Ag(tBu3P)2].31 The H NMR of 11 is, however,
slightly broadened at room temperature. The spectrum of the
complex is quite different in CDCl3 compared to C6D6 suggest-
ing that different species are dominant in these two solvents;
spectroscopic data for 11 in both solvents are included in the
Experimental, but complete assignments have only been made
for the complex in C6D6. The coordination chemistry of
phenop is summarised in Scheme 2.
Preliminary efforts to extend the coordination chemistry to
nickel() have proved fruitless. No reaction was observed when
two mol equivalents of ligand were added to [Ni(H2O)6]Cl2 or
anhydrous NiCl2 in ethanol. In all cases, no distinct colour
change was observed and only uncoordinated ligand was seen
in solution by 31P{1H} NMR; indeed, the ligand could be
recovered intact from the solutions on work-up. In addition, no
reaction was observed when two mol equivalents of phenop
were refluxed with [(1,5-COD)RhCl]2 in ethanol or n-butanol
and the starting materials were again recovered intact.
‡ The labelling reported here is with respect to the scheme of Fig. 1 and
not to the actual carbon labelling of the crystal structure. This is for
comparison with the other structures and solution studies (NMR).
Fig. 6 Molecular structure of 11.
D a l t o n T r a n s . , 2 0 0 3 , 3 5 1 6 – 3 5 2 5
3524

View Article Online
11 M. Ostermeier, J. Prieβ and G. Helmchen, Angew. Chem., Int. Ed.,
In conclusion, the rigid, bulky tertiary phosphine phenop has
been prepared by an improved method and its complexation
chemistry with a number of metal ions examined. Simple
monodentate P-bound complexes have been obtained with
Pd(0) and Ag(). For palladium(), the nature of the complexes
depends on the choice of starting compound and reaction con-
ditions with Pd(OAc)2 giving a six-membered P∧C chelate
through metallation at an exo-methyl and Pd(1,5-COD)Cl2
2002, 41, 612.
12 R. Haigh, K. M. A. Malik and P. D. Newman, Chem. Commun.,
2002, 2558.
13 C. Dai and G. C. Fu, J. Am. Chem. Soc., 2001, 123, 2719.
14 R. B. Bedford and C. S. J. Cazin, Chem. Commun., 2001, 1540.
15 J. P. Stambuli, R. Kuwano and J. F. Hartwig, Angew. Chem., Int. Ed.,
2002, 41, 4746.
16 J. H. Kirchhoff, C. Dai and G. C. Fu, Angew. Chem., Int. Ed., 2002,
41, 1945.
17 W. A. Herrmann, V. P. W. Böhm and C.-P. Reisinger, J. Organomet.
Chem., 1999, 576, 23.
18 R. Hooft, COLLECT software for data collection, Nonius B. V.,
Delft, The Netherlands, 1998.
producing
a cyclometallated five-membered P∧C chelate
through C–H activation at a ring methylene. New chiral centres
are generated stereoselectively in both these systems. Work on
assessing the use of the ligand in asymmetric catalysis and
extending the coordination chemistry is currently in progress
and will be presented in due course.
19 Z. Otwinowski and W. Minor, in Macromolecular Crystallography,
ed. C. W. Carter, Jr. and R. M. Sweet, Academic Press, New York,
1997, p. 307.
20 R. H. Blessing, Acta Crystallogr., Sect. A, 1995, 51, 33.
21 G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990, 46, 467.
22 G. M. Sheldrick, SHELXL-97, Program for refinement of crystal
structures, University of Göttingen, Germany, 1997.
23 H. D. Flack, Acta Crystallogr., Sect. A, 1983, 39, 876.
24 L. J. Farrugia, Ortep for Windows, Version 1.0, University of
Glasgow, 1997.
Acknowledgements
The authors would like to thank Cardiff University for
financial support and the EPSRC for support of the X-ray
crystallography facility in Cardiff.
25 Y. Fuchita, M. Nakashima, K. Hiraki and M. Kawatani, J. Chem.
Soc., Dalton Trans., 1988, 785.
References
26 (a) V. V. Dunina, O. N. Gorunova, L. G. Kuz’mina, M. V. Livantsov
and Y. K. Grishin, Tetrahedron: Asymmetry, 1999, 10, 3951;
(b) W. A. Herrmann, C. Brossmer, K. Öfele, C.-P. Reisinger,
T. Priermeier, M. Beller and H. Fischer, Angew. Chem., Int. Ed.,
1995, 34, 1844; (c) G. J. Grainsford and R. Mason, J. Organomet.
Chem., 1974, 80, 395; (d ) A. L. Rheingold and W. C. Fultz,
Organometallics, 1984, 3, 1414.
1 Phosphorus–Carbon Heterocyclic Chemistry: the Rise of a New
Domain, ed. F. Mathey, Elsevier Science Ltd., Oxford, 2001.
2 M. Pabel and S. B. Wild, Macro- and Spiro-heterocycles, in ref. 1,
ch. 6.1, p. 631.
3 See the relevant chapters in ref. 1; F. Mathey, Angew. Chem., Int. Ed.,
2003, 42, 1578–1604.
4 N. Gül and J. H. Nelson, Organometallics, 2000, 19, 91.
5 P. H. Leung, G. S. Ge, H. F. Lang, A. M. Liu, S. K. Loh,
S. Selvaratnam, K. F. Mok, A. J. P. White and D. J. Williams,
Tetrahedron, 2000, 57, 7.
6 A. Marinetti and D. Carmichael, Chem. Rev., 2002, 102, 201.
7 C. Laporte, G. Frison, H. Grützmacher, A. C. Hillier, W. Sommer
and S. P. Nolan, Organometallics, 2003, 22, 2202.
8 (a) J. H. Downing, V. Gee and P. R. Pringle, Chem. Commun., 1997,
1527; (b) V. Gee, A. G. Orpen, H. Phetmung, P. R. Pringle and
R. I. Pugh, Chem. Commun., 1999, 901.
27 V. V. Dunina, O. N. Gorunova, E. B. Averina, Y. K. Grishin,
L. G. Kuz’mina and J. A. K. Howard, J. Organomet. Chem., 2000,
603, 138.
28 (a) M. J. Ingleson, M. F. Mahon, N. J. Patmore, G. D. Ruggiero and
A. S. Weller, Angew. Chem., Int. Ed., 2002, 41, 3694; (b) L. Lettko,
J. S. Wood and M. D. Rausch, Inorg. Chim. Acta, 2000, 308, 37;
(c) R. A. Stein and C. Knobler, Inorg. Chem., 1977, 16, 242;
(d ) B. K. Teo and J. C. Calabrese, Inorg. Chem., 1976, 15, 2474;
(e) A. F. M. J. van der Ploegg, G. van Koten and A. L. Spek,
Inorg. Chem., 1979, 18, 1052.
9 M. J. Burk, J. E. Feaster, W. A. Nugent and R. L. Harlow, J. Am.
Chem. Soc., 1993, 115, 10125.
10 C. Crause, L. Bennie, L. Damonense, C. L. Dwyer, C. Grove,
N. Grimmer, W. J. van Rensburg, M. M. Kirk, K. M. Mokheseng,
S. Otto and P. Steynberg, Dalton Trans., 2003, 2036.
29 J. E. Huheey, E. A. Keiter and R. L. Keiter, Inorganic Chemistry:
Principles of Structure and Reactivity, Harper Collins, New York,
1993.
30 S. M. Socol and J. G. Verkade, Inorg. Chem., 1984, 23, 3487.
31 R. G. Goel and P. Pilon, Inorg. Chem., 1978, 17, 2876.
D a l t o n T r a n s . , 2 0 0 3 , 3 5 1 6 – 3 5 2 5
3525