Phosphorus-nitrogen-phosphorus ligands: Cooperative effects between nitrogen and phosphorus substituents on catalytic activity

Article abstract of DOI:10.1039/b312702h

A new generation of PNP compounds bearing different diarylphosphine groups were prepared and used as ligands in palladium-catalysed Suzuki cross-coupling reactions. Rates of oxidative addition of iodobenzene to (PNP)Pd[o] complexes were measured using UV spectroscopy. Synergistic effects between the N- and P- substituents were identified and correlated in redox and catalytic chemistry.

Full text of DOI:10.1039/b312702h

View Article Online / Journal Homepage / Table of Contents for this issue
Phosphorus–nitrogen–phosphorus ligands: cooperative effects
between nitrogen and phosphorus substituents on catalytic activity
Sébastien L. Parisel,a Neil D. Moorcroft,b Anny Jutand,c David J. Aldousb and
King Kuok (Mimi) Hii*†a
a Department of Chemistry, King’s College London, Strand, London, UK WC2R 2LS
b Aventis Pharmaceuticals, Route 202-206, PO Box 6800, Bridgewater, NJ 08807-0800, USA
c Ecole Normale Superieure, Departement de Chimie, UNR CNRS 8640, 24, Rue Lhomond,
75231 Paris Cedex 05, France
Received 13th October 2003, Accepted 19th November 2003
First published as an Advance Article on the web 15th December 2003
A new generation of PNP compounds bearing different diarylphosphine groups were prepared and used as ligands
in palladium-catalysed Suzuki cross-coupling reactions. Rates of oxidative addition of iodobenzene to (PNP)Pd[]
complexes were measured using UV spectroscopy. Synergistic effects between the N- and P- substituents were
identified and correlated in redox and catalytic chemistry.
Introduction
Previously we reported the preparation of a class of hemilabile
phosphorus–nitrogen–phosphorus (PNP) ligands of the type
R–N(CH2CH2PPh2)2, and their coordination to palladium and
ruthenium complexes.1–3 In palladium-catalysed Heck reac-
tions, the nature of the N-substituent has a profound effect on
the rates of turnover.4 In the same study, electron-rich tertiary
arylphosphines (PR3) were found to promote catalytic activity
and stability. This led us to speculate that the catalytic activity
of the (PNP)Pd complex may be enhanced by increasing the
basicity of the arylphosphine moiety. To test this hypothesis,
a series of PNP ligands with different electron-rich aryl-
phosphine groups was prepared and their performance in the
Suzuki cross-coupling reactions were evaluated.
Scheme 1 Preparation of ligands 1 and 2. Method A: (i) KOH,
DMSO, 50 ЊC; (ii) HSiCl3, NEt3, PhMe. Method B: (iii) a. t-BuLi, b. 3
or 4.
Comparative catalytic activity
There have been many reports of highly active palladium
catalysts for Suzuki cross-coupling reactions in recent years.6 A
quick survey revealed three main classes of phosphorus ligands
that have been employed with good effect in this area: (i)
cyclopalladated complexes;7 (ii) multidentate tetraphosphine-
palladium complexes;8 and (iii) sterically bulky alkylphosphine
palladium complexes.9 With the exception of the last class of
examples, elevated reaction temperatures (≥ 100 ЊC) are often
required in the first two systems, especially for the less activated
aryl bromides and chlorides. It is necessary to exercise caution
with the interpretation of these results, as it has been recently
suggested that certain palladacycles are unstable at high tem-
peratures, generating colloidal/monomeric palladium species
which are the true catalytic precursors.10
With this in mind, the lowest feasible reaction temperature
was adopted in our catalytic studies, in order to preserve the
integrity of the organopalladium species. Employing unactiv-
ated 4-bromotoluene and 4-methoxyphenylboronic acid as
substrates in our initial studies, several reaction parameters
were examined, including: metal-to-ligand ratio (1 : 1 to 1 : 2),
solvent (DME, THF, dioxane, toluene), additive (K3PO4, CsF,
Cs2CO3), temperature (65, 85, 100 ЊC) and palladium precursor
[Pd(OAc)2, Pd2(dba)3ؒCHCl3], which led ultimately to the
identification of the optimal reaction conditions (Scheme 2).
Results and discussion
Synthesis of diarylphosphine ligands 1 and 2
The enhanced acidity of secondary diarylphosphine oxides in
DMSO5 was exploited in the syntheses of the new PNP ligands.
Diarylphosphine oxide anions may be generated using potas-
sium hydroxide in wet DMSO. Following P–C bond formation
with the appropriate nitrogen mustards 3 or 4 (Method A,
Scheme 1), amino diphosphine oxides 5 and 6 may be obtained,
which were reduced in a basic silane solution to give P(III)
ligands 1b–d and 2b–d. The reactions generally proceeded in
good to excellent yields, except for the electron-rich bis(4-di-
methylaminophenyl)phosphine oxide. In this case, a strong base
(t-BuLi) was used to generate the diarylphosphide nucleophile
(Method B, Scheme 1).
† Present address: Department of Chemistry, Imperial College London,
South Kensington, UK SW7 2AZ. E-mail: mimi.hii@imperial.ac.uk;
Fax: ϩ44-20-75945804.
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 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 3 0 1 – 3 0 6
301

View Article Online
Table 1 Catalytic activity of aminodiarylphosphine ligands 1 and 2 in Suzuki-coupling reactions between 4-bromotoluene and 4-methoxy-
phenylboronic acid (Scheme 2)a
Entry
Ligand
% Conversion (2 h)
% Conversion (20 h)
Isolated yieldb (%)
1
2
3
4
5
6
7
8
9
1a
1b
1c
1d
1e
2a
2b
2c
2d
2e
12
44
24
22
56
26
72
42
21
64
36
95
83
47
98
79
93
91
80
100
28
88
77
38
91
67
86
85
78
96
10
a Under optimised reaction conditions, using Pd2(dba)3.CHCl3 (1 mol%), ligand (2.2 mol%) and CsF (2 eq.) in DME at 85 ЊC. Reactions were
duplicated to within 3%. b After 20 h, following purification by column chromatography.
Combining electron-rich phosphine and nitrogen substi-
tuents, the ligand 2e was found to afford the highest and
quickest turnovers. The synthetic utility of this ligand was thus
investigated in the coupling of a number of other substrates
(Scheme 3, Table 2). A number of electron-rich and electron-
poor aryl bromides and arylboronic acids were assessed in this
study.
Scheme 2 Suzuki coupling reaction between 4-bromotoluene and
4-methoxyphenylboronic acid.
The progress of the reactions promoted by ligands 1e, 2c and
2e was monitored by GC analysis of reaction aliquots extracted
during the catalytic reactions (Fig. 1). In all three cases no
induction period was observed, thus supporting our belief
that the (PNP)Pd[] complexes generated in situ under these
Scheme 3 Suzuki coupling using ligand 2e.
reaction conditions are the true catalytic precursors.
Homocoupling between arylboronic acids is known to occur
in certain palladium-catalysed systems, especially when the
cross-coupling is very slow.11 Thus it is important to note that
even in the more difficult couplings, we did not detect any
competitive homocoupling – GC analysis of the reaction
mixture and the isolated material showed only the expected
cross-coupled product.
Unsurprisingly, the electron poor 4-bromoacetophenone
(entries 1–3) gave the highest yields of over 90% with the quick-
est rates of turnover. With the unactivated 4-bromotoluene,
isolated yields of above 80% may also be obtained (entries 4–6).
It is commonly acknowledged that electron-rich aryl halides
are often the most difficult substrate to activate. In this regard,
we are pleased to see that ligand 2e was able to activate
4-bromoanisole towards the coupling reaction under these mild
conditions (entries 7–9). Even the sterically demanding
2-bromoanisole gave acceptable yields (entries 10–12).
Examining the yields from different aryl boronic acids, we
noticed that the electronically ‘neutral’ phenylboronic acid gave
noticeably lower yields, compared to the electron-rich and elec-
tron-poor arylboronic acids in each of these series (entries 3, 6,
Fig. 1 Rate of 4-bromotoluene consumption vs. time (᭜ 1e, ᭹ 2c,
9 and 12). This unusual result is particularly interesting when
compared to an earlier report of a tetraphosphine/[PdCl-
(η3-C3H5)]2 system, where the electronic property of the aryl-
boronic acid component was studied, and was found to be
insignificant.8
ꢀ 2e).
Comparing the catalytic activity promoted by ligands 1 and
2, the second generation of amino diarylphosphine ligands do
indeed outperform the aminodiphenylphosphine ligands 1a and
2a (Table 1, entries 1 and 6). On the whole, the tert-butylamino
ligands 2a–e display higher rates of turnover than their counter-
parts 1a–e (entries 1–5 and entries 6–10).
Coordination/redox chemistry of palladium(0) precursors
generated from ligands 1 and 2
Examining the results within each class of ligand, it is clear
that the introduction of electron-donating groups at the para-
position of the diarylphosphine moiety improves the rate and
yield of the reaction (a < b ≈ c < e). Moving the substituent to
the ortho-position clearly imposes a detrimental effect on the
catalytic activity (entries 3 and 4, and entries 8 and 9).
In order to understand the interplay of steric and electronic
effects between the nitrogen and phosphorus donors and their
influence on the resultant catalytic chemistry, the comparative
reactivity of palladium() complexes generated from the
ligands 1 and 2 towards the oxidative addition reaction with
aryl halides was explored.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 3 0 1 – 3 0 6
302

View Article Online
Table 2 Ligand 2e in the Suzuki cross coupling between several aryl
Table 3 Rate of oxidative addition of iodobenzene to [(PNP)Pd(dba)]
complexa
bromides and arylboronic acids (Scheme 3)a
b
Entry
R1
R2
Product
Time/h
Yield (%)b
Entry
PNP ligand
k (× 10Ϫ4 sϪ1
)
1
2
3
4
5
6
7
8
9
4-COMe
4-COMe
4-COMe
4-Me
4-Me
4-Me
4-MeO
4-MeO
4-MeO
2-MeO
2-MeO
2-MeO
MeO
F
H
MeO
F
H
MeO
F
H
MeO
F
H
7a
7b
7c
7d
7e
7f
7g
7h
7i
10
10
10
20
20
20
20
20
20
20
20
20
97
95
92
96
82
62
63
79
57
59
61
48
1
2
3
4
5
6
1a
1b
1e
2a
2b
2e
190
36
25
26
180
7.8
a Experiments were carried out at 25 ЊC using 1 mM of the palladium
complex with 10 equivalents of PhI, and were duplicated. b Linear
correlation constants were >0.995 in every case.
10
11
12
7j
7k
7l
(entries 1–3) revealed that the presence of an electron-donating
P-substituent retards the rate of the oxidative addition (1a >>
1b > 1e). This is in accordance with earlier observations, where
the presence of an electron-rich N-substituent was found to
have a similar effect.1 However, when the reactivity of the tert-
butylaminodiphosphine complexes was examined (entries 4–6),
the p-tolylphosphine ligand 2b was found to promote the fastest
reaction (2b >> 2a > 2e).
In a previous study by Amatore and Jutand, oxidative addi-
tion of iodobenzene to palladium() precursors generated
in situ from Pd(dba)2 and triarylphosphine P(p-Z-C6H4)3, where
Z = OMe, Me, H, F and Cl, led to bell-shaped curves in sub-
sequent Hammett plots.12 This was attributed to the antagon-
istic relationship between the inherent reactivity of the L2Pd()
complex and its effective concentration in situ.
a Reaction conditions are the same as described in Table 2. b Purified
yields after column chromatography.
As was observed previously, coordination of these ligands to
Pd(dba)2 generates a mixture of rapidly exchanging Pd(0)
species 8 and 9 (Scheme 4), which undergo reaction with iodo-
benzene to furnish a single palladium(II) species, as indicated
by the observation of a sharp single resonance peak in the 31P
NMR spectrum, corresponding to a cationic [(PNP)Pd(Ph)]ϩIϪ
complex, 10.1
The results obtained in the present system may be explained
using a similar model: although complex 9 is inherently most
active towards oxidative addition when it is coordinated by
electron-rich aminodiphosphine ligands such as 1e and 2e, its
effective concentration is also lower, as the coordination of
dba (π-acceptor) to these complexes is also enhanced by the
increased Lewis basicity of the metal centre. However, this
shift in equilibrium may be offset by the N-substituent.
From these kinetic studies, it is clear that a close synergistic
relationship exists between N- and P-donor groups, where the
electronic contributions are delicately balanced with the
reaction conditions under which the reaction intermediates are
generated.
The catalytic cycle of the Suzuki cross-coupling reaction has
been generally proposed to consist of three principal steps
(Scheme 5):13 (i) oxidative addition; (ii) anion-assisted trans-
metalation; and (iii) reductive elimination. The reluctance of
unactivated aryl chlorides and bromides to undergo oxidative
addition contributes to their decreased reactivity in cross-
coupling reactions (I > Br > Cl).14 Conversely, the transfer
of the aryl moiety from boron to palladium is known to
decreases in the order: Cl > Br > I,15 i.e. the first two steps of the
catalytic cycle works counteractively under catalytic turnover
conditions.
Scheme
4
Generation of palladium() precursor and oxidative
addition with iodobenzene.
The rate of oxidative addition of iodobenzene to the (PNP)-
Pd[] complexes was investigated at 25 ЊC. Palladium() pre-
cursors were generated by mixing equimolar quantities (1 mM)
of the Pd(dba)2 with the PNP ligands in THF. The solution
gives rise to a single UV absorption peak at λmax = 395 nm. The
rate of disappearance of this absorption peak was monitored
in the presence of an excess (10 equivalents) of iodobenzene
(Fig. 2), and was found to be first order in [Pd]. Rate data
can thus be extrapolated from the resultant logarithmic plots
(Table 3). Comparison of the relative rates of the oxidative
addition of iodobenzene to anisylaminodiphosphine ligands
Fig.
2
Progress of oxidative addition of iodobenzene to [(1b)-
Pd(PNP)] complex, generated in situ in the UV cell.
Scheme 5 Catalytic cycle of the Suzuki cross-coupling reaction.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 3 0 1 – 3 0 6
303

View Article Online
In this case, we have shown that the ligands (1e and 2e) pro-
mote the slowest oxidative addition (Table 3, entries 3 and 6),
yet induce the fastest catalytic turnovers (Table 1, entries 5 and
10). However, this does not neccessarily imply that the trans-
metalation is rate-determining, for it is also known that the
transmetalation step is promoted by the presence of electron
donating substituents on arylboronic acids,11 yet the electron-
deficient 4-fluorophenyl boronic acid gave comparable, if not
better, yields than the electron-rich 4-methoxyphenylboronic
acid (Table 2). These two observations suggest the existence of
a dynamic relationship between the oxidative addition and
transmetalation steps – either can become dominant, depend-
ing on the nature of the substrate and the reaction conditions
under which catalytic turnover take place.
give a white foamy solid (523 mg), which was purified by flash
column chromatography. Foamy solid products were dried
overnight under vacuum at 45 ЊC.
N,N-bis{2-[di-(4-tolyl)phosphoryl]ethyl}-N-4-anisidine
5b.
Column chromatography, acetone (411 mg, 85%). Mp: 81–
83 ЊC (Found: C, 73.9; H, 6.3; N, 2.0. C39H43NO3P2 requires C,
73.7; H, 6.75; N, 2.2%). νmax/cmϪ1 (KBr disc) 1172 (P᎐O).
᎐
δH (360 MHz, CDCl3) 2.30 (12H, s, Me), 2.47 (4H, td, 3JHH 6.8
2
3
Hz, JPH 11.0 Hz, CH2P), 3.37 (4H, td, JHH 6.8 Hz and
3JPH 14.0 Hz, NCH2CH2P), 3.67 (3H, s, OMe), 6.42 (2H, d,
J 8.9 Hz, Hortho), 6.66 (2H, d, J 8.9 Hz, Hmeta), 7.20 (8H,
dd, 4JPH 2.3 Hz, 3JHH 8.0 Hz, Hmeta), 7.53 (8H, dd, 3JHH 8.0 Hz,
3JPH = 11.4 Hz, Hortho). δC (90.5 MHz, CDCl3) 21.9 (s, Me), 29.7
1
(d, JPC = 71 Hz, CH2P), 38.8 (s, NCH2CH2P), 56.2 (s, OMe),
3
114.9 (s, Cortho), 115.2 (s, Cmeta), 129.8 (d, JPC12.0 Hz, Cmeta),
1
2
Conclusion
130.0 (d, JPC = 101 Hz, Cipso), 131.1 (d, JPC = 10.0 Hz, Cortho),
141.9 (s, Cipso), 142.7 (s, Cpara), 152.6 (s, Cpara). δP (146 MHz,
CDCl3) ϩ33.2. m/z (FAB) 637 (M ϩ 2Hϩ), 406, 229, 176, 154,
91.
A series of PNP ligands was prepared with systematic changes
in the N- and P-donor substituents. Palladium complexes of
these ligands were found to promote Suzuki cross-coupling
under mild reaction conditions. It was found that catalysts
generated from electron rich ligands are the most active and
are able to activate difficult substrates such as 4-bromoanisole.
During the course of the study, we observed an unusual elec-
tronic effect exerted by the arylboronic acids. In subsequent
mechanistic investigations, palladium() precursors were gener-
ated from these ligands, and their rates of oxidative addition
reactions with iodobenzene were measured using UV spectro-
scopy. An inverse relationship between rate of the oxidative
addition and the catalytic activity was identified. The trans-
metalation is thus suggested to be just as important as the oxid-
ative addition in the catalytic reactions. We demonstrated that
the rate of at least one of these steps may be altered by the
nature of the donor groups of the aminodiarylphosphine
ligands. Therefore, the reaction outcome of catalytic reactions
may be controlled by tuning these ligands.
N,N-bis{2-[di-(4-anisyl)phosphoryl]ethyl}-N-4-anisidine 5c.
Column chromatography, CH2Cl2/acetone, 1 : 1 (91%). Mp 68–
70 ЊC. νmax/cmϪ1 (KBr disc) 1180 (P᎐O). δ (360 MHz, CDCl )
᎐
H
3
3
2
2.27 (4H, td, JHH 11.2 Hz, JPH 16.0 Hz, CH2P), 3.36 (4H, td,
3JPH 6.8 Hz, 3JHH 11.2 Hz, NCH2CH2P), 3.67 (3H, s, OMe), 3.75
(12H, s, OMe), 6.48 (2H, d, J 9.0 Hz, Hortho), 6.66 (2H, d, J 9.0
Hz, Hmeta), 6.86 (8H, dd, JPH 2.1 Hz, JHH 8.8 Hz, Hmeta), 7.49
(8H, dd, JHH 8.8 Hz, JPH 11.1 Hz, Hortho). δC (90.5 MHz,
CDCl3) 28.3 (d, JPC 69.3 Hz, CH2P), 47.1 (s, NCH2CH2P),
55.8 (s, OMe), 56.1 (s, OMe), 114.5 (d, JPC 12.5 Hz, Cmeta),
115.2 (s, Cortho), 117.8 (s, Cmeta), 124.5 (d, JPC 105 Hz, Cipso),
132.8 (d, JPC 11.0 Hz, Cortho), 141.1 (s, Cipso), 152.3 (s, Cpara),
4
3
3
3
1
3
1
2
162.7 (s, Cpara). δP (162 MHz, CDCl3) ϩ31.6. m/z (FAB) 701 (M
ϩ 2Hϩ), 439, 261, 176, 107.
N,N-bis{2-[di-(2-anisyl)phosphoryl]ethyl}-N-4-anisidine 5d.
Column chromatography, CH2Cl2/MeOH, 19 : 1 increasing to
9 : 1 (495 mg, 93%). Hygroscopic white foamy solid. νmax/cmϪ1
Experimental
(KBr disc) 1181 (P᎐O). δ (400 MHz, CDCl ) 2.69–2.78 (4H, m,
᎐
H
3
The general experimental and analytical techniques were the
same as in other recent papers reported from this laboratory.
Nitrogen mustards 3 and 4,1 secondary phosphine oxides,16
aminodiphosphine ligands 1a and 2a,1 were prepared by pre-
viously published procedures. All other chemicals were
obtained from commercial sources and used without prior
purification. Due to the lack of homogeneity between different
batches of commerically available aryl boronic acids,8 these
were combined and recrystallised from water prior to use, to
ensure reproducible results. GC analyses were performed on a
Unicam 6100 system with a FID detector. A JW Scientific DB
250 column was used, with He as a carrier gas set at a flow
rate of 1.0 mL minϪ1. Oven temperature was kept isothermally
at 90 ЊC. UV spectroscopy was performed on a DU 7400
Beckman spectrophotometer in a thermostated (25 ЊC) 1 mm
path length cell.
CH2P), 3.39–3.32 (4H, m, NCH2CH2P), 3.67 (12H, s, OMe),
3.72 (3H, s, OMe), 6.54 (2H, d, J 9.2 Hz, Hortho), 6.73 (2H, d,
3
4
J 9.2 Hz, Hmeta), 6.88 (4H, dd, JHH 8.2 Hz, JPH 5.3 Hz, H-3),
3
3
6.94–7.00 (4H, m, H-5), 7.46 (4H, dddd, JHH 8.2 Hz, JHH
7.3 Hz, 4JHH 1.8 Hz, 5JPH 0.9 Hz, H-4), 7.54 (4H, ddd, 3JPH 13.6
Hz, 3JHH 7.6 Hz, 4JHH 1.8 Hz, H-6). δC (125 MHz, CDCl3) 26.7
1
(d, JPC 71.0 Hz, CH2P), 45.0 (s, NCH2CH2P), 55.4 (s, OMe),
55.8 (s, OMe), 110.7 (d, 3JPC 6.4 Hz, C-3), 114.7 (Cortho or C meta),
115.0 (Cortho or C meta), 120.6 (d, 3JPC 11.4 Hz, C-5), 120.6 (d, 1JPC
101 Hz, C-1), 133.4 (d, 4JPC 1.8 Hz, C-4), 133.8 (d, 2JPC 6.8 Hz,
C-6), 141.1 (s, Cipso), 151.5 (s, Cpara), 160.5 (s, C-2). δP (162 MHz,
CDCl3) ϩ31.2. m/z (HR-ESI) 1421.4927 (2MNaϩ, C78H86N2-
NaO14P4 requires 1421.4932, 35%), 722.2424 (MNaϩ, C39H43-
NNaO7P2 requires 722.2415, 100), 700.2596 (MHϩ, C39H44-
NO7P2 requires 700.2595, 13), 699.2563 (Mϩ, C39H43NO7P2
requires 699.2517, 13).
Preparation of ligands
N,N-bis{2-[di-(4-tolyl)phosphoryl]ethyl}-N-tert-butylamine,
6b. Column chromatography, acetone (374 mg, 84%). Mp 68–
70 ЊC (Found: C, 74.2; H, 7.9; N, 2.15. C36H45NO2P2 requires C,
Typical procedure for the preparation of aminodiphosphine
oxides. The corresponding diarylphosphine oxide (1.55 mmol)
was added to a solution of KOH (115 mg, 205 mmol) in DMSO
(1 cm3) and water (0.15 cm3), giving a light yellow suspension.
After 5 minutes, the aminodichloride 3 or 4 (0.761 mmol) was
added and the reaction mixture was stirred for 30 minutes, then
heated at 50 ЊC for 1.5 hours. After cooling to room temper-
ature, the reaction mixture was diluted with brine (8 cm3), fol-
lowed by extraction with CH2Cl2 (5 × 10 cm3). The combined
organic extracts were washed with brine (5 × 8 cm3), back
extracting each wash with CH2Cl2 (5 cm3). The combined
organic layer was dried over Na2SO4 and evaporated in vacuo to
73.85; H, 7.7; N, 2.4%). νmax/cmϪ1 (KBr disc) 1178 (P᎐O).
᎐
δH (360 MHz, CDCl3) 0.89 (9H, s, Me), 2.20–2.31 (4H, m,
CH2P), 2.29 (12H, s, CH3), 2.72 (4H, td, 3JPH 6.0 Hz, 3JHH 11.8
4
3
Hz, NCH2CH2P), 7.16 (8H, dd, JPH 2.2 Hz and JHH 8.0 Hz,
Hmeta), 7.48 (8H, dd, 3JHH 8.0 Hz, 3JPH 11.3 Hz, Hortho). δC (90.5
MHz, CDCl3) 21.9 (s, CH3), 27.8 (s, CH3), 31.8 (d, 1JPC 67.8 Hz,
CH2P), 42.7 (s, NCH2CH2P), 56.0 (s, CMe3), 127.5 (d, 1JPC 103
3
2
Hz, Cipso), 129.7 (d, JPC 11.8 Hz, Cmeta), 131.0 (d, JPC 9.6 Hz,
Cortho), 142.4 (s, Cpara). δP(146 MHz, CDCl3) ϩ32.1. m/z (FAB)
586 (MHϩ), 530, 356, 229, 154, 91, 77.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 3 0 1 – 3 0 6
304

View Article Online
N,N-bis{2-[di-(4-anisyl)phosphoryl]ethyl}-N-tert-butylamine
6c. Column chromatography, CH2Cl2/acetone, 1 : 1 (283 mg,
76%). Mp: 61–63 ЊC. νmax/cmϪ1 1174 (P᎐O). δ (360 MHz,
3.89 (3H, s, OMe), 6.87 (4H, d, J 8.0 Hz, Hmeta), 6.93 (4H,
d, J 8.0 Hz, Hmeta), 6.98 (2H, d, J 8.9 Hz, Hortho), 7.27 (4H, t,
J 8.0 Hz, Hortho), 7.42 (4H, t, J 8.0 Hz, Hortho), 7.56 (d, 2H,
J 8.9 Hz, Hmeta), 11.4 (1H, br s, N–H ). δC (free amine, 90.5
᎐
H
2
3
CDCl3) 0.89 (9H, s, CMe3), 2.22 (4H, td, JPH 11.5 Hz, JHH
=
3
3
1
2
15.9 Hz, CH2P), 2.73 (4H, td, JPH 6.0 Hz, JHH 15.9 Hz,
MHz, CDCl3) 26.7 (d, JPC13.0 Hz, CH2P), 49.0 (d, JPC 25.5
4
3
NCH2CH2P), 3.75 (12H, s, OMe), 6.87 (8H, dd, JPH 2.0 Hz,
Hz, NCH2CH2P), 55.6 (s, OMe), 56.1 (s, OMe), 114.5 (d, JPC
7.5 Hz, Cmeta), 115.1 (s, Cortho), 115.8 (s, Cmeta), 129.5 (d, JPC
9.5 Hz, Cipso), 134.5 (d, JPC = 20.0 Hz, Cortho), 142.1 (s, Cipso),
152.1 (s, Cpara), 160.5 (s, Cpara). δP (free amine, 146 MHz, CDCl3)
Ϫ22.3. m/z (free amine, FAB) 668 (MHϩ), 560, 422, 245, 107.
3
3
1
3JHH 8.7 Hz, Hmeta), 7.52 (8H, dd, JPH 11.0 Hz, JHH 8.7 Hz,
=
1
2
Hortho). δC (90.5 MHz, CDCl3) 27.8 (s, Me), 32.0 (d, JPC
=
69.0 Hz, CH2P), 42.7 (s, NCH2CH2P), 55.7 (s, OMe), 56.0 (s,
2
1
CMe3), 114.5 (d, JPC 12.5 Hz, Cortho), 125.2 (d, JPC 105 Hz,
Cipso), 132.9 (d, 3JPC 10.7 Hz, Cmeta), 162.6 (s, Cpara). δP(146 MHz,
CDCl3) ϩ31.9. m/z (FAB) 650 (MHϩ), 592, 388, 332, 261, 245,
154.
N,N-bis{2-[di-(2-anisyl)phosphino]ethyl}-N-4-anisidine
1d.
Foamy solid. 405 mg, 72%. δH (400 MHz, CDCl3) 2.29–2.35
(4H, m, CH2P), 3.33–3.40 (4H, m, NCH2CH2P), 3.73 (3H, s,
OMe), 3.74 (12H, s, OMe), 6.54 (2H, J 9.2 Hz, Hortho), 6.74 (2H,
N,N-bis{2-[di-(2-anisyl)phosphoryl]ethyl}-N-tert-butylamine
6d. Column chromatography, CH2Cl2 increasing to CH2Cl2/
methanol, 49 : 1 (279 mg, 75%). Hygroscopic white foamy solid.
δH (400 MHz, CDCl3) 0.94 (9H, s, CMe3), 2.83–2.64 (8H, m,
3
4
d, J 9.2 Hz, Hmeta), 6.84 (4H, dd, JHH 8.3, JPH4.3 Hz, 3-H),
6.85–6.91 (4H, m, 5-H), 7.09 (4H, ddd, 3JHH 7.5 Hz, 3JPH 5.5 Hz,
4JHH 1.7 Hz, 6-H), 7.28–7.33 (4H, m, 4-H). δC (125 MHz,
3
1
2
CH2CH2P), 3.69 (12H, s, OMe), 6.87 (4H, dd, JPH 5.1 Hz,
CDCl3) 22.6 (d, JPC14.2 Hz, CH2P), 49.1 (d, JPC 28.3 Hz,
NCH2CH2P), 55.4 (s, OMe), 55.8 (s, OMe), 110.2 (3-C), 114.7
3
4
3JHH 8.0 Hz, H-6), 7.00 (4H, tdd, JHH 7.5 Hz, JHH 2.0 Hz,
4JPH 1.0 Hz, H-4), 7.48–7.42 (4H, m, H-5), 7.60 (4H, ddd,
(s, Cortho), 120.8 (s, Cmeta), 120.8 (s, 5-C), 125.1 (d, JPC15.1 Hz,
1
3JPH 13.4 Hz, JHH 7.5 Hz, JHH 2.0 Hz, H-3). δC (125 MHz,
1-C), 130.0 (s, 4-C), 132.8 (d, 2JPC 6.4 Hz, 6-C), 142.0 (s, Cipso),
3
4
1
2
CDCl3) 27.5 (s, Me), 30.5 (d, JPC = 70.0 Hz, CH2P), 42.3 (s,
151.2 (s, Cpara), 161.3 (d, JPC 13.3 Hz, 2-C). δP (162 MHz,
NCH2CH2P), 55.4 (s, OMe), 55.7 (s, CMe3), 110.8 (d,
CDCl3) Ϫ38.9. m/z (FAB) 668.2715 (MHϩ, C39H44NO5P2
3
3JPC 6.4 Hz, C-6), 120.5 (d, JPC 11.4 Hz, C-4), 121.0 (d,
requires 668.2695), 422, 245.
2
1JPC 99.8 Hz, C-2), 133.3 (s, C-5), 133.8 (d, JPC 6.8 Hz, C-3),
2
160.5 (d, JPC 3.6 Hz, C-1). δP(162 MHz, CDCl3) ϩ31.3. m/z
N,N-bis{2-[di-(4-tolyl)phosphino]ethyl}-N-tert-butylamine
hydrochloride salt 2bؒHCl. 426 mg, 92%. Mp: 82–84 ЊC. δH (360
MHz, CDCl3) 1.31 (9H, s, Me), 2.26 (6H, s, CH3), 2.29 (6H, s,
CH3), 2.70 (2H, m, CH2P), 3.01 (2H, m, NCH2CH2P), 3.25 (2H,
m, NCH2CH2P), 3.43 (m, 2H, NCH2CH2P), 7.10 (4H, d, J 7.1
Hz, Hmeta), 7.18 (4H, d, J 7.1 Hz, Hmeta), 7.37 (4H, m, Hortho),
7.47 (4H, m, Hortho), 11.68 (1H, br s, N–H ). δC (free amine, 90.5
MHz, CDCl3) 21.7 (s, CH3), 27.8 (s, CH3), 30.3 (d, 1JPC 11.6 Hz,
(HR-ESI) 1321.5338 (2MNaϩ, C72H90N2NaO12P4 requires
1321.5347, 100%), 672.2621 (MNaϩ, C36H45NNaO6P2 requires
672.2623, 4), 650.2811 (MHϩ, C36H46NO6P2 requires 650.2803,
4).
Typical procedures for the reduction of aminodiphosphine
oxides (Method A). At Ϫ78 ЊC, trichlorosilane (0.52 cm3, 5.15
mmol) was added dropwise to a mixture of the corresponding
amino phosphine oxide 5 or 6 (0.84 mmol) and triethylamine
(6 cm3) in toluene (25 cm3). The mixture was allowed to warm
to room temperature, then refluxed for 4 hours. After cooling
with an ice bath, the suspension was diluted with degassed di-
ethyl ether (20 cm3) and excess reducing agent destroyed with
saturated aqueous sodium carbonate (0.25 cm3). The suspen-
sion was filtered, washing the precipitate with further portions
of diethyl ether. The filtrate was concentrated in vacuo to give
an oil which was dissolved in degassed ethyl acetate (25 cm3)
and washed with degassed saturated brine (2 × 15 cm3), dried
over Na2SO4 and concentrated in vacuo to give the product
either as a colourless oil (which may be precipated as the HCl
salt by the addition of 2M aq. HCl) or a solid white foam.
2
CH2P), 46.9 (d, JPC 26.4 Hz, NCH2CH2P), 55.7 (s, CMe3),
129.6 (d, 2JPC 7.1 Hz, Cortho), 132.0 (d, 3JPC 18.6 Hz, Cmeta), 135.6
1
(d, JPC11.1 Hz, Cipso), 138.7 (s, Cpara). δP(146 MHz, CDCl3)
Ϫ13.8. m/z (FAB, free amine) 553 (MHϩ), 462, 340, 283, 213,
182, 91.
N,N-bis{2-[di-(4-anisyl)phosphino]ethyl}-N-tert-butylamine
hydrochloride salt, 2cؒHCl. 521 mg, 95%. Mp: 81–83 ЊC (Found:
C, 66.0; H, 7.1; N, 2.05%. C36H46NO4P2Cl requires C, 66.1; H,
7.1; N, 2.15%). δH (free amine, 360 MHz, CDCl3) 0.89 (9H, s,
2
3
Me), 2.00 (4H, td, JPH 4.6 Hz, JHH 8.7 Hz, CH2P), 2.48 (4H,
td, 3JPH 5.2 Hz, 3JHH 8.7 Hz, NCH2CH2P), 3.70 (12H, s, OMe),
6.76 (8H, dd, 4JPH2.4 Hz, 3JHH 8.4 Hz, Hmeta), 7.23 (8H, dd, 3JPH
3
7.0 Hz, JHH 8.4 Hz, Hortho). δC (free amine, 90.5 MHz, CDCl3)
1
2
29.6 (s, CH3), 32.6 (d, JPC 12.8 Hz, CH2P), 48.7 (d, JPC 26.9
Hz, NCH2CH2P), 57.3 (s, OMe), 57.5 (s, CMe3), 116.2 (d, 3JPC
7.7 Hz, Cmeta), 131.9 (d, 1JPC = 9.3 Hz, Cipso), 136.2 (d, 2JPC = 19.9
Hz, Cortho), 162.2 (s, Cpara). δP (free amine, 146 MHz, CDCl3)
Ϫ22.7. m/z (free amine, FAB) 618 (MHϩ), 510, 372, 315, 245,
214.
N,N-bis{2-[di-(4-tolyl)phosphino]ethyl}-N-4-anisidine hydro-
chloride salt 1bؒHCl. 489 mg, 91%. Mp: 101–103 ЊC. δH (360
3
2
MHz, CDCl3, free amine) 2.09 (4H, td, JHH 5.6 Hz, JPH
=
8.0 Hz, CH2P), 2.26 (12H, s, Me), 3.17 (4H, td, 3JHH 5.6 Hz 3JPH
11.1 Hz, NCH2CH2P), 3.66 (3H, s, OMe), 6.32 (2H, d, J 9.1 Hz,
Hortho), 6.62 (2H, d, J 9.1 Hz, Hmeta), 7.04 (8H, dd, 4JPH 2.6 Hz,
3JHH 7.5 Hz, Hmeta), 7.19 (8H, d, 3JPH 6.9 Hz, 3JHH 7.5 Hz, Hortho).
N,N-bis{2-[di-(2-anisyl)phosphino]ethyl}-N-tert-butylamine
2d. Foamy solid. 425 mg, 82%. δH (360 MHz, CDCl3) 1.00 (9H,
s, CMe3), 2.27–2.21 (4H, m, CH2P), 2.69–2.63 (4H, m, NCH2-
CH2P), 3.70 (12H, s, OMe), 6.81 (4H, dd, 3JHH 8.2 Hz, 4JPH 4.2
Hz, 6-C), 6.85 (4H, t, 3JHH 7.4 Hz, 4-C), 7.09 (4H, ddd, 3JHH 7.4
1
δC (90.5 MHz, CDCl3, free amine) 20.3 (s, Me), 24.9 (d, JPC
13.4 Hz, CH2P), 47.5 (d, 2JPC = 25.4 Hz, NCH2CH2P), 54.7 (s,
3
OMe), 113.7 (s, Cortho), 114.3 (s, Cmeta), 128.2 (d, JPC 7.0 Hz,
Cmeta), 131.6 (d, 2JPC 18.7 Hz, Cortho), 133.6 (d, 1JPC10.8 Hz, Cipso),
137.6 (s, Cipso), 140.7 (s, Cpara), 150.6 (s, Cpara). δP (146 MHz,
CDCl3, free amine) Ϫ21.9. m/z (FAB) 604.2910 (Mϩ Ϫ Cl,
C39H44NOP2 requires 604.2898).
3
4
Hz, JPH 5.3 Hz, JHH 1.7 Hz, 3-C), 7.30–7.25 (4H, m, 5-C).
1
δC (90.5 MHz, CDCl3) 26.6 (d, JPC 14.7 Hz, CH2P), 27.4 (s,
Me), 47.3 (d, 2JCP 28.5 Hz, NCH2CH2), 55.5 (s, OMe), 110.1 (s,
6-C), 120.8 (s, 4-C), 125.6 (d, 1JPC 16.7 Hz, 2-C), 129.8 (s, 5-C),
2
N,N-bis{2-[di-(4-anisyl)phosphino]ethyl}-N-4-anisidine
132.7 (s, 3-C), 161.4 (d, JPH 11.7 Hz, 1-C) – CMe3 signal is
hydrochloride salt, 1cؒHCl. 555 mg, 94%. Mp: 86–88 ЊC (Found:
C, 66.5; H, 6.25; N, 2.0. C39H42NO5P2.HCl requires: C, 66.5; H,
6.25; N, 2.0%). δH (400 MHz, CDCl3) 1.89 (2H, br s, CH2P),
2.94 (2H, br s, NCH2CH2P), 3.18 (2H, br s, NCH2CH2P), 3.54
(2H, br s, NCH2CH2P), 3.81 (6H, s, OMe), 3.84 (6H, s, OMe),
obscured. δP (146 MHz, CDCl3) Ϫ39.1. m/z (FAB) 618.2926
(MHϩ. C36H46NO4P requires 618.2902).
Synthesis of ligands 1e and 2e (Method B). tert-Butyllithium
(4.4 cm3, 1.5M solution in pentane, 6.6 mmol) was added
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 3 0 1 – 3 0 6
305

View Article Online
dropwise to a solution of di-(4-N,N-dimethylaminophenyl)-
phosphine (1.63 g, 6.0 mmol) in dry Et2O (20 cm3) at 0 ЊC. After
stirring for 30 minutes, it was warmed gradually to room tem-
perature, where it developed an intense red colour over a period
of 1 h. It was cooled down to 0 ЊC, where a solution of the
aminodichloride 4 (0.59 g, 3.0 mmol) in Et2O (5.0 cm3) was
added. The reaction mixture was then left to stir overnight at
room temperature. After dilution with toluene (10 cm3), the
reaction mixture was carefully quenched by the addition of
distilled water (5.0 cm3) at 0 ЊC. The suspension was filtered
and the combined filtrates were evaporated under vacuum to
furnish a pale yellow residue, which was recrystallised from
methanol to afford the product as a white solid.
with 4-bromotoluene (0.12 cm3, 1.0 mmol) and 0.19 cm3 of
decane (GC internal standard). The solution containing the
catalyst precursor was introduced into the Radley’s tube,
along with an additional volume of DME (1.0 cm3). The reac-
tion temperature was adjusted to 85 ЊC and maintained at
the specific temperature ( 0.1 ЊC) with the aid of a thermo-
stat. The progress of the reaction was monitored by periodically
extracting aliquots of the reaction mixtures, which were
analysed by GC. At the end of the reaction, the products
were purified by column chromatography (EtOAc–hexane) and
isolated as white solids.
Characterisation data for biaryl compounds 7a–l listed in
Table 2 agree with previously reported values.17–28
N,N-bis{2-[di-(4-dimethylaminophenyl)phosphino]ethyl}-N-p-
anisidine, 1e. 1.72 g, 80%. Mp 129–131 ЊC (Found: C, 71.35; H,
7.3; N, 9.3. C43H55N5P2O requires C, 71.75; H, 7.65; N, 9.7%).
δH (400 MHz, CDCl3) 2.20 (4H, m, CH2P), 2.99 (24H, s, NMe),
3.20 (4H, td, 3JPH 5.3 Hz, 3JHH 8.1 Hz, NCH2CH2P), 3.77 (3H, s,
OMe), 6.42 (2H, d, J 9.0 Hz, Hortho), 6.72 (8H, dd, 4JPH 2.9 Hz,
3JHH 8.3 Hz, Hmeta), 7.30–7.37 (10H, m, Ar-H). 13C{1H} NMR
Acknowledgements
The authors thank EPSRC (SLP, grant number: GR/M78229)
and Aventis (NDM) for studentship support, Johnson Matthey
plc for the generous loan of palladium salts and Professor
S. E. Gibson (KCL) for access to GLC equipment. The authors
also thank the Centre National de la Recherche Scientifique
(UMR CNRS-ENS-UPMC 8640).
1
δC (100 MHz, CDCl3) 26.8 (d, JPC12.5 Hz, CH2P), 40.7 (s,
NMe), 48.9 (d, 2JPC 27.0 Hz, NCH2CH2P), 56.2 (s, OMe), 112.7
(d, 3JPC 7.5 Hz, Cmeta), 115.3 (s, Cortho), 115.1 (s, Cmeta), 124.3 (d,
1JPC 7.0 Hz, Cipso), 134.1 (d, 2JPC = 20.0 Hz, Cortho), 142.5 (s, Cipso),
151.0 (s, P–Cpara), 151.6 (s, Cpara). δP (146 MHz, CDCl3) Ϫ24.9.
m/z (FAB) 720.3956 (MHϩ. C43H56N5OP2 requires 720.3960).
References
1 K. K. Hii, M. Thornton-Pett, A. Jutand and R. P. Tooze,
Organometallics, 1999, 18, 1887.
2 M. S. Rahman, J. W. Steed and K. K. Hii, Synthesis, 2000, 1320.
3 M. S. Rahman, P. D. Prince, J. W. Steed and K. K. Hii,
Organometallics, 2002, 21, 4927.
4 M. Qadir, T. Möchel and K. K. Hii, Tetrahedron, 2000, 56, 7975.
5 E. N. Tsvetkov, N. A. Bondarenko, I. G. Malakhova and M. I.
Kabachnik, Synthesis, 1986, 198.
6 A. F. Littke and G. C. Fu, Angew. Chem., Int. Ed., 2002, 41, 4176.
7 R. B. Bedford, S. L. Hazlewood, M. E. Limmert, D. A. Albisson,
S. M. Draper, P. N. Scully, S. J. Coles and M. B. Hursthouse,
Chem. Eur. J., 2003, 9, 3216.
8 M. Feuerstein, F. Berthiol, H. Doucet and M. Santelli, Synlett, 2002,
1807. Poisoning of the catalysts was ruled out in our studies, as the
arylboronic acids were recrystallised from water prior to use.
9 (a) J. P. Wolfe and S. L. Buchwald, Angew. Chem., Int. Ed., 1999, 38,
2413; (b) A. F. Littke, C. Dai and G. C. Fu, J. Am. Chem. Soc., 2000,
122, 4020.
10 A. M. de Vries, J. M. C. A. Mulders, J. H. M. Mommers, H. J. W.
Henderickx and J. G. de Vries, Org. Lett., 2003, 5, 3285.
11 M. Moreno-Mañas, M. Pérez and R. Pleixats, J. Org. Chem., 1996,
61, 2346.
12 (a) C. Amatore, A. Jutand and G. Meyer, Inorg. Chim. Acta, 1998,
273, 76; (b) C. Amatore and A. Jutand, Coord. Chem. Rev., 1998,
180, 511.
13 N. Miyaura and A. Suzuki, Chem. Rev., 1995, 95, 2457.
14 C. Amatore and A. Jutand, in Organopalladium Chemistry for
Organic Synthesis, eds. E. Negishi, A. de Meijere, Wiley, New York,
2002, vol. 1, Chapter III, Section 2.19.
15 N. Miyaura, J. Organomet. Chem., 2002, 653, 54.
16 P. N. Kapoor and L. M. Venanzi, Helv. Chim. Acta, 1977, 60, 2824.
17 V. Farina, B. Krishnan, D. R. Marshall and G. P. Roth, J. Org.
Chem., 1993, 58, 5434.
N,N-bis{2-[di-(4-dimethylaminophenyl)phosphino]ethyl}-N-
tert-butylamine, 2e. 1.31 g, 65%. Mp: 86–88 ЊC. Found: C, 71.9;
H, 8.4; N, 10.0. C40H57N5P2 requires C, 71.75; H, 8.5; N, 10.4 %.
δH (400 MHz, CDCl3) 0.90 (9H, s, CMe3), 2.03 (4H, td, 2JPH 4.5
Hz, 3JHH 8.5 Hz, NCH2CH2P), 2.49 (4H, td, 3JPH 4.01 Hz, 3JHH
8.5 Hz, NCH2CH2P), 2.84 (24H, s, NMe), 6.57 (8H, dd, 4JPH 2.0
Hz, 3JHH 8.6 Hz, Hmeta), 7.19 (8H, dd, 3JPH 7.0 Hz, 3JHH 8.6 Hz,
1
Hortho). δC(100 MHz, CDCl3) 27.9 (s, CH3), 30.7 (d, JPC 10.4
2
Hz, NCH2CH2P), 40.7 (s, NMe), 47.1 (d, JPC 27 Hz, NCH2-
CH2P), 55.7 (s, CMe3), 112.8 (d, 3JPC 7.5 Hz, P–Cmeta), 125.0 (d,
2
1JPC 7 Hz, Cipso), 134.1 (d, JPC 20 Hz, Cortho), 150.9 (s, Cpara).
δP (162 MHz, CDCl3) Ϫ23.8. m/z (FAB) 670.4163 (MHϩ.
C40H56N5P2 requires 670.4167).
Kinetic studies
UV spectroscopy. The solution cell was carefully purged with
argon prior to the experiment. The Pd() complex was gener-
ated by mixing stock solutions of Pd(dba)2 (1.0 mM) and
appropriate PNP ligands (1.0 mM) in dry THF at room
temperature. After stirring for 10 minutes, the colour of the
reaction mixture changes from red–purple to yellow–brown,
indicating the formation of Pd() precursors. These solutions
were then transferred via a cannula into the UV cell, to which
10.0 eq. of phenyl iodide were injected via a microlitre syringe.
The decay of the UV absorption band due to the Pd(dba)(PNP)
precursor at 395 nm was followed.
18 D. J. Byron, G. W. Gray and R. C. Wilson, J. Chem. Soc. C, 1966,
840.
19 J. P. Wolfe, R. A. Singer, B. H. Yang and S. L. Buchwald, J. Am.
Chem. Soc., 1999, 121, 9550.
Suzuki cross-coupling reactions
20 A. H. A. Tinnemans and W. H. Laarhoven, J. Am. Chem. Soc., 1974,
96, 4611.
21 H. C. Bell, J. R. Kalman, G. L. May, J. T. Pinhey and T. Sternhell,
Aust. J. Chem., 1979, 32, 1531.
22 M. E. Mowery and P. DeShong, J. Org. Chem., 1999, 64, 1684.
23 H. V. Taylor, A. L. Allred and B. M. Hoffman, J. Am. Chem. Soc.,
1973, 95, 3215.
24 Y. Hatanaka, K. Goda, Y. Okahara and T. Hiyama, Tetrahedron,
1994, 50, 8301.
Catalytic runs were performed in parallel using a Radley’s
12-placed reaction carousel under an Ar or N2 atmosphere.
[Pd2(dba)3.CHCl3] (1.0 mol%) and the appropriate PNP ligand
(2.2 mol%) were weighed into a small vial which was flushed
with argon. 1.0 cm3 of DME was added into the vial and the
reaction mixture was stirred at 40 ЊC. After 10 minutes, the
initial purple–black colour of the solution was discharged, giv-
ing a yellow–brown solution indicating the formation of the
complex. 4-Methoxyphenyl boronic acid (228 mg, 1.5 mmol)
and CsF (2.2 equivalents) were introduced into a Radley’s reac-
tion tube, which was placed in the carousel and purged with
argon. 2.0 cm3 of DME was introduced via the septum along
25 S. Schneider and W. Bannwarth, Helv. Chim. Acta, 2001, 84, 735.
26 R. A. Abramovitch and O. A. Koleoso, J. Chem. Soc. B, 1969, 779.
27 M. Lourak, R. Venderesse, Y. Fort and P. Caubere, J. Org. Chem.,
1989, 54, 4844.
28 G. E. Robinson and J. M. Vernon, J. Chem. Soc. C, 1971, 3363.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 3 0 1 – 3 0 6
306