A new synthesis of π-electron conjugated phosphonates and phosphonic bis(diethylamides) and their SHG activities

Article abstract of DOI:10.1039/a709082j

A series of vinylic and arylic phosphonates and phosphonic bis(diethylamides) were prepared by copper promoted substitution of the corresponding bromides. These π-electron conjugated phosphonates and phosphonic bis(diethylamides) were investigated to eluc

Full text of DOI:10.1039/a709082j

View Online / Journal Homepage / Table of Contents for this issue
A new synthesis of ð-electron conjugated phosphonates and
phosphonic bis(diethylamides) and their SHG activities
Takuji Ogawa,*,† Naoya Usuki and Noboru Ono
Department of Chemistry, Faculty of Science, Ehime University, Matsuyama 790, Japan
A series of vinylic and arylic phosphonates and phosphonic bis(diethylamides) were prepared by copper
promoted substitution of the corresponding bromides. These ð-electron conjugated phosphonates and
phosphonic bis(diethylamides) were investigated to elucidate their optical and second harmonic
generation (SHG) properties; it was found that diphenyl 2-(4-dimethylaminophenyl)ethenylphosphonate
exhibited strong SHG activity, the efficiency being 13 times greater than that of urea.
Dialkyl vinyl- and aryl-phosphonates and vinyl- and aryl-
phosphonic bis(dialkylamides) have been attracting interest as
versatile intermediate reagents in organic synthesis, and have
been synthesized via multi-step methods or palladium catalysed
reactions from dialkyl phosphonates and vinyl bromides.1–4
Here we report that these compounds can be prepared by the
copper promoted reaction which we have been using for the
synthesis of a series of vinylic substituted compounds;5–9 some
of these prepared π-electron conjugated phosphonates have
relatively strong second harmonic generation (SHG) activities
together with good transparency to light.
bromides were heated with phosphonates in the presence of
a base and copper() iodide. The reaction conditions were
optimized by using reactions of 1-bromo-2-phenylethene with
phosphonates in the presence of copper() iodide. The results
are collected in Table 1.
Hexamethylphosphoric triamide (HMPA) as the solvent and
KH as the base afforded the best yields in the shortest reaction
time (Runs 1, 2, 4, 5, 7, 8, 10 and 11), however because of the
potential carcinogenicity of HMPA, we recommend to use
other solvents like tetramethylurea (TMU) (Run 12), dimethyl-
formamide (DMF) (Run 13) or tetrahydrofuran (THF) (Runs
3, 6, 9, 14 and 15) although the yields were generally diminished
and a longer reaction time was required with these solvents
when compared with HMPA. Less basic triethylamine was also
effective as the base for these reactions (Runs 3, 6, 9 and 15).
The reaction is essentially stereospecific and the E:Z ratio of
the starting bromides was retained in the products although a
little isomerization of Z to E was observed with prolonged
heating.
The reaction for their preparation is simple and straight-
forward as shown in Scheme 1; the corresponding vinylic
O
O
Br
Base, CuI
Solvent
P(OR)2
+
H
P(OR)2
The reaction was successful not only with alkyl phosphonates
(Runs 1–9), but also with phenyl phosphonate, to afford the
corresponding 2-phenylethenylphosphonate in good yields
(Runs 10–15). The substitution reaction proceeded well with
phenyl bromide, 2-arylethenyl bromides, and 4-phenylbuta-1,3-
dienyl bromide to give the corresponding phosphonates in
moderate to good yields (Table 2).
1
Scheme 1
† Present address: Institute for Fundamental Research of Organic
Chemistry, Kyushu University, Fukuoka, 812-81, Japan.
Table 1 Reaction conditions for the preparation of 2-phenylethenylphosphonates 1 by the copper-promoted reaction
Product
(%)e
Bromide
Phosphonate
R group
Run
E:Za
Baseb
Solventc
T/ЊC
t/hd
E:Za
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
84:16
19:81
>99:1
84:16
19:81
>99:1
84:16
19:81
>99:1
84:16
19:81
>99:1
>99:1
>99:1
>99:1
Me
Me
Me
Et
Et
Et
KH
KH
TEA
KH
KH
TEA
KH
KH
TEA
KH
KH
KH
KH
KH
TEA
HMPA
HMPA
THF
HMPA
HMPA
THF
HMPA
HMPA
THF
HMPA
HMPA
TMU
DMF
THF
120
120
reflux
120
120
reflux
120
120
reflux
120
120
2
2
24
2
2
24
2
2
24
2
2
4
1a (78)
1a (65)
1a (51)
1b (82)
1b (66)
1b (65)
1c (99)
1c (80)
1c (65)
1d (92)
1d (72)
1d (76)
1d (46)
1d (86)
1d (76)
86:14
38:64
>99:1
84:16
27:73
>99:1
84:16
20:80
>99:1
98:2
21:79
>99:1
>99:1
>99:1
>99:1
Pri
Pri
Pri
Ph
Ph
Ph
Ph
Ph
Ph
120
120
reflux
reflux
24
24
24
THF
a The E:Z ratios were determined by 1H NMR spectroscopy. b KH, potassium hydride; TEA, triethylamine. c HMPA, hexamethylphosphoric
triamide; TMU, tetramethylurea; DMF, dimethylformamide; THF, tetrahydrofuran. d Reactions were terminated when the starting bromide was
consumed. e Yields refer to the isolated values and not optimized.
J. Chem. Soc., Perkin Trans. 1, 1998
2953

View Online
Table 2 Yields and physical properties of π-electron conjugated
proceeded readily even in THF to give the products in moderate
phosphonates, phosphonic bis(ethylamides) and related compounds
to good yields. Probably the phosphonates or their anions
themselves coordinated to the copper promoter to activate the
reaction.
Compounds
Yield (%)
Mp or bp/ЊC
1a
1b
1c
1d
2
3a
3b
4
78a
82a
99a
92a
59b
27c
40c
40c
37c
15c
65d
38c
83c
62d
39d
83d
27e
49e
ca. 90 (0.01 mmHg)
ca. 100 (0.1 mmHg)
ca. 110 (0.05 mmHg)
110–111
Because the previously reported methods for preparation of
phosphonic bis(ethylamides) requires tedious procedures10–13
we tried a more direct vinyl Arbuzov type reaction of tris-
(diethylamino)phosphine with phenyl and vinyl bromides in
the presence of copper() iodide and succeeded in obtaining the
substituted products in moderate yields (Scheme 2).
60 (0.05 mmHg)
ca. 100 (0.05 mmHg)
95–97
125–127f
5a
5b
5c
5d
5e
5f
6
97–99g
O
88–89f
CuI
Br
P(NEt2)2
(Et2N)3P
+
89–91f
Solvent
125–127f
121–123h
104–106i
Scheme 2
86–88
7
133–134f
Organic nonlinear optical materials have been attracting
much attention as promising second or third harmonic gener-
ation materials because of their large hyperpolarizabilities,
ultra fast response times, and synthetic flexibility which permits
modification of their molecular parameters.14–17 Most of the
compounds are based on charge transfer transition states
with strong electron acceptors such as nitro groups, and strong
electron donors like amino groups, linked by π-electron con-
jugated systems. However, a recent theoretical paper suggests
that the maximum molecular first hyperpolarizability, β,
could be achieved by using less strong donor and acceptor
groups having an electronegativity similar to that of the linking
π-electron conjugated system.18 Phosphonates and phosphonic
bis(ethylamides) are good candidates for being such moderately
strong electron acceptor groups. With this in mind we studied
optical and second harmonic generation (SHG) activities of
π-conjugated phosphonates and phosphonic bis(ethylamides)
prepared by the present method.19
8
ca.90(0.1mmHg)
9
113–115f
a KH, HMPA, 120 ЊC, 2 h. b Et3N, TMU, 120 ЊC, 4 h. c Et3N, THF,
reflux, 24 h. d Et3N, DMF, 120 ЊC, 24 h. e (Et2N)3P, CuI, DMF, 120 ЊC,
40 h. f For the (E)-isomer which crystallized out. g For the
mixture of E:Z = 12:88. h For the mixture of E:Z = 67:33. i For the
mixture of E:Z = 97:3.
Generally the copper promoted arylic or vinylic substitution
reactions show significant solvent effects; the reaction pro-
ceeded smoothly only in solvents with a strong coordinating
ability like HMPA, TMU, DMF, or dimethyl sulfoxide
(DMSO); in less coordinating solvents such as THF or
chlorinated hydrocarbons no reaction occurred at all.5–9
Therefore it is noteworthy that with the present reactions they
O
O
O
P(OR)2
P(OEt)2
P(OR)2
Calculated second order hyperpolarizabilities (βcalc) of π-
electron conjugated phosphonates and phosphonic bis(ethyl-
amides) by the semi-empirical molecular orbital method20 were
in the order of 1.6–24 × 10Ϫ30 esu as shown in Table 3. para-
Amino substituted arylvinylphosphonates 5d and 5e showed
relatively high βcalc values (15 and 24 × 10Ϫ30 esu) compared
with the corresponding nitro compound 11 (6 × 10Ϫ30 esu).
Actually the phosphonate 5d exhibited strong SHG activity,
the efficiency being 13 times that of urea (will be abbreviated
as 13U) when measured by powder method at 532 nm with a
Nd-YAG laser (1064 nm).21 The corresponding nitro com-
pound 11 showed no SHG activity with the powder method
(Table 3).
3a R = Et
3b R = Ph
1a R = Me
2
1b R = Et
1c R = iPr
1d R = Ph
R4
O
O
P(OR1)2
P(OPh)2
R3
R2
5a R1 = Ph, R2 = H, R3 = Cl, R4 = H
4
Other phosphonates having comparable or higher βcalc values
than 5d did not always show strong SHG activities; the
efficiency being 0.13U for 1d (βcalc = 3 × 10Ϫ30 esu), 0.7U for
5b (βcalc = 6 × 10Ϫ30 esu) and 0.7U for 7 (βcalc = 15 × 10Ϫ30 esu)
5b R1 = Ph, R2 = OMe, R3 = H, R4 = OMe
5c R1 = Ph, R2 = H, R3 = SMe, R4 = H
5d R1 = Ph, R2 = H, R3 = Me2N, R4 = H
5e R1 = Ph, R2 = H, R3 = Ph2N, R4 = H
5f R1 = Ph, R2 = H, R3 = Ph2P, R4 = H
and no activity observed for 4 (βcalc = 3 × 10Ϫ30 esu), 5c (βcalc
=
9 × 10Ϫ30 esu), 5e (βcalc = 24 × 10Ϫ30 esu), 5f (βcalc = 8 × 10Ϫ30
esu) and 6 (βcalc = Ϫ2 × 10Ϫ30 esu). These non-SHG active
crystals probably have centrosymmetric space groups since the
second order macroscopic coefficient of optical susceptibility
χ(2) will vanish in these space group crystals.22–24 The phos-
phonate 5d with SHG activity of 13U had the non-centro-
symmetric Pca21 space group.
O
P(NEt2)2
P(OPh)2
O
P(OPh)2
O
S
S
6
7
8
Theoretical and experimental studies revealed that polariz-
abilities α, β and γ can be expressed in terms of the bond length
alternation (BLA) and the π-electron bond order alternating
(BOA) structural parameters.25–29 BLA of the phosphonate 5d
was found to be 0.127 Å as measured by X-ray crystallographic
analysis (Table 4) (see Fig. 1 for X-ray structure), which is
significantly smaller than that of the parent 2-phenyl-
ethenylphosphonic acid (0.157 Å).30 The result is consistent
with the theoretical prediction that with increasing donor and/
or acceptor strength the ground state structure of the bridge
O
NO2
P(NEt2)2
N
9
10
NO2
Me2N
11
2954
J. Chem. Soc., Perkin Trans. 1, 1998

View Online
Table 3 Optical properties of π-electron conjugated phosphonates,
phosphonic bis(ethylamides) and related compounds
Table 4 Selected bond distances (Å) and bond angles (Њ) for phos-
phonate 5d with estimated standard deviations in parentheses
λmax(λcutoff)/nm
Bond distances
b
Compound
βcalc
10Ϫ30 esu
/
SHG
P(1)–C(1)
C(1)–C(2)
C(2)–C(3)
C(3)–C(8)
C(8)–C(7)
C(7)–C(6)
C(3)–C(4)
C(4)–C(5)
C(5)–C(6)
1.760(7)
1.324(9)
1.458(9)
1.395(9)
1.378(10)
1.410(9)
1.378(8)
1.360(9)
1.40(1)
(E:Z)a
Efficiencyc Benzene
Acetonitrile
1d (98:2)
4 (11:89)
3.115
0.13
0
278.5 (310)
304.0 (368)
262.5 (320)
299.5 (364)
3.209 (E)
1.647 (Z)
6.414 (E)
8.702 (E)
5b (100:0)
5c (100:0)
5d (100:0)
5e (67:33)
5f (97:3)
6
7 (100:0)
10
11
0.7
0
345.5 (407)
313.5 (354)
349.0 (400)
369.5 (432)
306.0 (381)
233.0 (285)
294.0 (338)
422.0 (528)
345.5 (402)
318.0 (368)
351.0 (415)
363.0 (452)
305.0 (385)
232.0 (287)
298.0 (359)
436.5 (590)
452.0 (500)
15.448 (E) 13
24.093 (E)
8.483 (E)
Ϫ1.922
15.086 (E)
20.352 (E)
6.409 (E)
0
0
0
0.7
2.5d
0
Bond angles
P(1)–C(1)–C(2)
C(1)–C(2)–C(3)
C(2)–C(3)–C(8)
C(2)–C(3)–C(4)
119.3(5)
129.0(6)
118.9(6)
124.4(6)
a E:Z ratio of the measured sample. b MOPAC ver. 6.01.49 c Measured
by the powder method using a Nd-YAG laser (1064 nm). Values are
relative to urea. d Data from Ref. 5.
Fig.
2
Absorption spectra of 1-nitro-2-(4-dimethylaminophenyl)-
ethene 11, and diphenyl 2-(4-dimethylaminophenyl)ethenylphos-
phonate 5d, diphenyl 2-(4-diphenylaminophenyl)ethenylphosphonate
5e and diphenyl 2-(4-diphenylphosphinophenyl)ethenylphosphonate
5f in acetonitrile
that of the corresponding phosphonate 5d was about 415 nm;
about 85 nm shorter wavelength. When the diphenylamino
group of 5d was substituted with a diphenylphosphino group
as in compound 5f, the cut-off wavelength became even
shorter and light of wavelength longer than ca. 385 nm could be
transmitted (Fig. 2).
Usually the SHG efficiency–transparency trade off has been
the most serious problem for conjugated and polar molecules
possessing CT characteristics; large hyperpolarizability
Fig. 1 Single crystal X-ray structure of compound 5d
becomes both more delocalized and dipolar, i.e. it changes
towards a fully delocalized structure.25
inevitably leads to a loss of transparency as depicted in
Judging from the acidity of nitro and phosphonate groups,
the nitro group is a stronger acceptor than the phosphonate
group. Therefore 1-nitro-2-arylethenes should have more
delocalized structures and small BLA parameters according to
the theoretical prediction. However, in reality the phosphonate
compounds have a more delocalized structure than the corre-
sponding nitro compounds; BLA of 1-nitro-2-phenylethene
was 0.179 Å31 and that of 1-(2-chloro-4-dimethylaminophenyl)-
2-nitroethene was 0.147 Å,32 both are larger than those of the
corresponding phosphonate compounds. Probably the larger
orbitals of the phosphine atom, relative to the nitrogen atom,
enable a more efficient delocalization of the π-electrons. This
indicates that by using heavier elements than second row
elements as donor and/or acceptor groups, the BLA and there-
fore polarizabilities α, β, and γ can be tuned most efficiently.
Worthy to note with these π-electron conjugated phosphon-
ates as optical materials is that they have better transparency at
shorter wavelength than nitro compounds. For example nitro
compound 11 had a cut-off wavelength at about 500 nm, while
_
Eqn. (1);14 where βCT is the CT component of β, hω is the
laser fundamental photon energy, W is the energy gap, f is the
oscillator strength of the CT transition in the molecule, and
∆µ is the change in the dipole moment.33 In order to increase
βCT, the energy gap W should be small; as a consequence the
absorption band will shift to longer wavelength.
_
3e2h2Wf∆µ
2m[W 2 Ϫ (2hω)2][W 2 Ϫ (hω)2]
(1)
βCT
=
_
_
The π-electron conjugated phosphonates surmounted this
trade-off problem; even though the βcalc was larger for
phosphonate 5d than nitro compound 11, the transparency of
5d was better than 11. The reason for this is that the π-electron
conjugated phosphonates have a small degree of CT character
as can be seen in their solvent dependent absorption; the lowest
energy absorption of phosphonates did not shift to longer
wavelength in acetonitrile compared with that in benzene
J. Chem. Soc., Perkin Trans. 1, 1998
2955

View Online
solution (Table 3). It is an important finding that compounds
with little CT character such as 5d can exhibit relatively large
SHG character.
Here we have exhibited that π-electron conjugated phos-
phonates and phosphonic bis-amides having moderate electron
withdrawing ability can be good candidates for non-linear
optical materials.34
(m, 6H, Ph᎐CH᎐C); (Z)-isomer: 1.18 (t, 6H, J 7.0, OCH CH ),
᎐
2 3
3.99 (dq, 4H, J 7.3 and 7.5, POCH2CH3), 5.80 (dd, 1H, J 14.3
and 15.6, C᎐CH᎐P), 7.19 (d, 1H, J 14.3, Ph᎐CH᎐C), 7.34–7.70
᎐
᎐
(m, 5H, Ph); m/z (EI) 240 (Mϩ, 14%) and 131 (100).
Diisopropyl 2-phenylethenylphosphonate 1c. Reaction of 1-
bromo-2-phenylethene (E:Z = 84:16) with diisopropyl phos-
phonate, according to the typical procedure, in HMPA using
KH as the base, led to diisopropyl 2-phenylethenylphosphonate
1c (E:Z = 84:16) (99%) as a colourless oil (lit.,42 bp 122 ЊC/0.06
mmHg) (Found: C, 62.3; H, 8.0. C14H21O3P requires C, 62.7; H,
7.9%); νmax(KBr)/cmϪ1 2960 and 1225; δH(CDCl3) (E)-isomer:
1.35 (dd, 12H, J 6.1 and 13.1, POCHCH3), 4.71 (m, 2H,
POCHCH ), 6.27 (dd, 1H, J 17.4 and 17.4, C᎐CH᎐P) and 7.33–
Experimental
Mps were measured on a Yanaco micro melting point ap-
paratus and are uncorrected. 1H NMR Spectra were measured
on a JEOL JNM-GSX270 or JNM-EX 400 spectrometer and
J values are given in Hz. IR and UV–VIS spectra were recorded
with a Hitachi 270-30 and Shimadzu UV-2200 spectrometer,
respectively. Mass spectra were taken on a Hitachi M-80B
instrument. The cyclic voltammetry was carried out at
room temperature under nitrogen using a Yanagimoto P-1100
machine.
᎐
3
7.74 (m, 6H, Ph᎐CH᎐C); (Z)-isomer: 1.21 (dd, 6H, J 6.1 and
᎐
27.2, OCHCH3), 4.68 (m, 2H, POCHCH3), 5.82 (dd, 1H, J 14.5
and 15.1, C᎐CH᎐P), 7.33–7.74 (m, 6H, Ph᎐CH᎐C); m/z (EI)
᎐
᎐
268 (Mϩ, 18%), 209 (14) and 104 (100).
Diphenyl 2-phenylethenylphosphonate 1d. Reaction of 1-
bromo-2-phenylethene (E:Z = 84:16) with diphenyl phos-
phonate, according to the typical procedure, in HMPA using
KH as the base, led to diphenyl 2-phenylethenylphosphonate
1d (E:Z = 98:2) (92%) as colourless crystals, mp 111 ЊC (lit.,43
109 ЊC) (Found: C, 71.0; H, 5.1. C20H17O3P requires C, 71.4; H,
5.1%); νmax(KBr)/cmϪ1 3000 and 1180; δH(CDCl3) (E)-isomer:
1-Bromo-2-phenylethene was purchased from Tokyo
Chemical Industry Co., Ltd. and distilled prior to use. (1E,3E)/
(1Z,3E)-1-Bromo-4-phenylbutadiene,35 1-bromo-2-(4-dimethyl-
aminophenyl)ethene,36
and
1-bromo-2-(4-chlorophenyl)-
ethene37 were prepared by the literature methods. 1-Bromo-2-
(4-methylthiophenyl)ethene was prepared by modification of
the reported method.35 1-Bromo-2-(4-diphenylaminophenyl)-
ethene was prepared from 4-diphenylaminobenzaldehyde38 by
a similar method to the reported one.35 1-Bromo-2-(4-diphenyl-
phosphophenyl)ethene was prepared from 4-diphenylphos-
phobenzaldehyde39 by a similar method to the reported one.35
6.47 (dd, 1H, J 17.4 and 19.2, C᎐CH᎐P), 7.13–7.52 (m, 15H,
᎐
Ph) and 7.69 (dd, 1H, J 17.4 and 23.2, Ph᎐CH᎐C); m/z (EI) 336
᎐
(Mϩ, 57%) and 149 (100).
Typical procedure for the reactions using triethylamine as the
base in solvents other than HMPA
Diethyl phenylphosphonate 2. To the solution of the anion
generated from diethyl phosphonate (829 mg, 6 mmol) and tri-
ethylamine (ca. 1.0 g, 9.88 mmol) in tetramethylurea (TMU)
(4 cm3), were added copper() iodide (1.14 g, 6 mmol) and
bromobenzene (314 mg, 2 mmol) under a nitrogen atmosphere.
The solution was heated to 120–140 ЊC for 24 h until the
starting bromide was consumed completely. Benzene (30 cm3)
and brine (30 cm3) were added to the reaction mixture and the
organic layer was separated and washed with ca. 10% aqueous
sodium hydrogen sulfate (30 cm3), brine (30 cm3 × 5), and dried
with sodium sulfite. The solvent was evaporated and the residue
was purified by silica gel column chromatography with hexane–
dichloromethane as the eluents to give the phosphonate 2
(253 mg, 59%) as a colourless oil (lit.,44 bp 90 ЊC/0.2 mmHg)
which was further purified by bulb to bulb distillation in vacuo
for further analysis; δH(CDCl3) 1.33 (t, J 6.4, OCH2CH3), 4.11
(m, 4H, OCH2CH3) and 7.47–7.85 (m, 5H, aromatic); m/z (EI)
214 (Mϩ, 28%) and 77 (100).
Diethyl 4-phenylbuta-1,3-dienylphosphonate 3a. Reaction of
1-bromo-4-phenylbuta-1,3-diene (1E:1Z = 25:75; 3E:3Z >
99:1) with diethyl phosphonate, according to the typical
procedure in THF using triethylamine as the base, led to diethyl
4-phenylbuta-1,3-dienylphosphonate 3a (27%) as a pale yellow
oil (lit.,45 bp 125–126 ЊC/0.05 mmHg) (Found: C, 62.8; H, 7.3.
C14H19O3P requires C, 63.15; H, 7.2%); νmax(KBr)/cmϪ1 2950
and 1220; m/z (EI) 266 (Mϩ, 12%) and 128 (100).
Preparation of ð-electron conjugated phosphonates
from the corresponding bromides and di-alkyl and -aryl
phosphonates by the copper promoted reaction
Typical procedure for the reactions in hexamethylphosphoric
triamide (HMPA) using potassium hydride as base
Dimethyl 2-phenylethenylphosphonate 1a. To the solution
of the anion generated from dimethyl phosphonate (660 mg,
6 mmol) and KH (684 mg, 17.1 mmol) in hexamethylphos-
phoric triamide (HMPA) (4 cm3), were added copper() iodide
(1.14 g, 6 mmol) and 1-bromo-2-phenylethene (E:Z = 84:16)
(366 mg, 2 mmol) under a nitrogen atmosphere. The solution
was heated at 120 ЊC for 2 h until the starting bromide was
consumed completely. Benzene (30 cm3) and brine (30 cm3)
were added to the reaction mixture and the organic layer
was separated and washed with ca. 10% aqueous sodium
hydrogen sulfate (30 cm3), brine (30 cm3 × 5), and dried with
sodium sulfite. The solvent was evaporated and the residue was
purified by silica gel column chromatography with hexane–
dichloromethane as the eluents to give the phosphonate 1a
(E:Z = 86:14) (662 mg, 78%) as a colourless oil (lit.,40 mp 29–
30 ЊC) which was further purified by bulb to bulb distillation
in vacuo for further analysis (Found: C, 56.65; H, 6.3. C10H13-
O3P requires C, 56.6; H, 6.2%); νmax(NaCl)/cmϪ1 2980, 2930,
2830 and 1240; δH(CDCl3) (E)-isomer: 3.77 (d, 6H, J 11.0,
POCH ), 6.22 (dd, 1H, J 17.7 and 14.3, C᎐CH᎐P), 7.21 (d, 1H,
Diphenyl 4-phenylbuta-1,3-dienylphosphonate 3b. Reaction of
1-bromo-4-phenylbuta-1,3-diene (1E:1Z = 25:75; 3E:3Z >
99:1) with diphenyl phosphonate, according to the typical
procedure in THF using triethylamine as the base, led to
diphenyl 4-phenylbuta-1,3-dienylphosphonate 3b (40%) as
colourless crystals, mp 95–97 ЊC (Found: C, 73.0; H, 5.3.
C22H19O3P requires C, 72.9; H, 5.3%); νmax(KBr)/cmϪ1 2950
and 1220; m/z (EI) 362 (Mϩ, 23%) and 326 (100).
Diphenyl 2-(1-naphthyl)ethenylphosphonate 4. Reaction of
1-bromo-2-(1-naphthyl)ethene (E:Z = 8:92) with diphenyl
phosphonate, according to the typical procedure in THF using
triethylamine as the base, led to diphenyl 2-(1-naphthyl)-
ethenylphosphonate 4 (E:Z = 11:89) (61%) as colourless
crystals, mp 97 ЊC (Found: C, 74.5; H, 5.0. C24H19O3P requires
᎐
3
J 14.3, Ph᎐CH᎐C) and 7.35–7.66 (m, 5H); (Z)-isomer: 3.62 (d,
᎐
6H, J 11.3, POCH ), 5.77 (dd, 1H, J 14.3 and 15.6, C᎐CH᎐P),
᎐
3
7.35–7.66 (m, 6H, Ph᎐CH᎐C); m/z (EI) 212 (Mϩ, 24%) and 117
᎐
(100).
Diethyl 2-phenylethenylphosphonate 1b. Reaction of 1-bromo-
2-phenylethene (E:Z = 84:16) with diethyl phosphonate,
according to the typical procedure in HMPA using KH as the
base, led to diethyl 2-phenylethenylphosphonate 1b (E:Z =
84:16) (82%) as a colourless oil (lit.,41 bp 156–158 ЊC/3 mm Hg)
(Found: C, 59.8; H, 7.3. C12H17O3P requires C, 60.0; H, 7.1%);
νmax(KBr)/cmϪ1 2950 and 1220; δH(CDCl3) (E)-isomer: 1.35 (t,
6H, J 7.1, OCH2CH3), 4.13 (dq, 4H, J 7.0 and 7.1, POCH2-
CH ), 6.26 (dd, 1H, J 17.5 and 17.5, C᎐CH᎐P) and 7.34–7.70
᎐
3
2956
J. Chem. Soc., Perkin Trans. 1, 1998

View Online
C, 74.6; H, 5.0%); νmax(KBr)/cmϪ1 2980 and 1180; δH(CDCl3)
(E)-isomer: 6.60 (dd, 1H, J 17.1 and 18.9, C᎐CH᎐P) and
48 h, led to colourless crystals of mp 120 ЊC, which are dis-
solved in dichloromethane and washed with 10% aqueous
ammonia. As the aqueous solution turned blue indicating
the presence of copper, the crystals might be the copper
salt complex of diphenyl 2-(4-diphenylphosphinophenyl)-
ethenylphosphonate. The organic layer was washed with water
until the aqueous solution became neutral, dried with sodium
sulfate, evaporated and purified by silica gel column chrom-
atography to give diphenyl 2-(4-diphenylphosphinophenyl)-
ethenylphosphonate 5f (E:Z = 97:3) (62%) as colourless crys-
tals, mp 106 ЊC (Found: C, 73.6; H, 5.1. C32H26O3P2 requires C,
73.8; H, 5.0%); νmax(KBr)/cmϪ1 3000 and 1180; δH(CDCl3) (E)-
᎐
6.87–8.17 (m, 18H, Ph᎐CH᎐C); (Z)-isomer: 6.32 (dd, 1H,
᎐
J 13.7 and 17.1, C᎐CH᎐P) and 6.87–8.17 (m, 18H, aromatic
᎐
and Ar᎐CH᎐C); m/z (EI) 386 (Mϩ, 25%) and 152 (100).
᎐
Diphenyl 2-(4-chlorophenyl)ethenylphosphonate 5a. Reaction
of 1-bromo-2-(4-chlorophenyl)ethene (E:Z = 9:91) with
diphenyl phosphonate, according to the typical procedure
in THF using triethylamine as the base, led to diphenyl 2-(4-
chlorophenyl)ethenylphosphonate 5a (E:Z = 12:88) (37%)
as colourless crystals, mp 99 ЊC (Found: C, 65.2; H, 4.7.
C20H16ClO3P requires C, 64.8; H, 4.4%); νmax(KBr)/cmϪ1 3000
and 1180; δH(CDCl3) (E)-isomer: 6.44 (dd, 1H, J 17.6 and
isomer: 6.46 (dd, 1H, J 17.4 and 19.2, C᎐CH᎐P), 7.16–7.47 (m,
᎐
18.8, C᎐CH᎐P) and 7.06–7.69 (m, 15H, Ph᎐CH᎐C); (Z)-
24H, aromatic) and 7.67 (dd, 1H, J 17.4 and 23.8, Ar᎐CH᎐C);
᎐
᎐
᎐
isomer: 6.02 (dd, 1H, J 14.0 and 17.1, C᎐CH᎐P) and 7.06–7.69
m/z (EI) 520 (Mϩ, 100%) and 427(5).
᎐
(m, 15H, aromatic and Ar᎐CH᎐C); m/z (EI) 372 (Mϩ ϩ 2,
Diphenyl 2-thienylphosphonate 6. Reaction of 2-bromothio-
phene with diphenyl phosphonate, according to the typical pro-
cedure in DMF using triethylamine as the base for 24 h, led to
diphenyl 2-thienylphosphonate 6 (39%) as colourless crystals,
mp 88 ЊC (lit.,46 75 ЊC) (Found: C, 61.0; H, 4.2. C16H13O3PS
requires C, 60.8; H, 4.1%); νmax(KBr)/cmϪ1 3080 and 1266;
δH(CDCl3) 7.13–7.34 (m, 11H) and 7.73–7.80 (m, 2H); m/z (EI)
316 (Mϩ, 100%) and 223 (58).
Diphenyl 2-(2-thienyl)ethenylphosphonate 7. Reaction of 1-
bromo-2-(4-diphenylaminophenyl)ethene (E:Z = 22:78) with
diphenyl phosphonate, according to the typical procedure in
DMF using triethylamine as the base for 48 h, led to diphenyl
2-(2-thienyl)ethenylphosphonate 7 (E:Z = 43:57) (83%) as
colourless crystals, which were recrystallized from methanol to
give pure (E)-isomer, mp 134 ЊC (Found: C, 63.0 H, 4.35.
C18H15O3PS requires C, 63.15; H, 4.4%); νmax(KBr)/cmϪ1 3100
and 1194; δH(CDCl3) (E)-isomer: 6.19 (dd, 1H, J 17.2 and 18.7,
᎐
35%), 370 (Mϩ, 100) and 277 (100).
Diphenyl 2-(2,5-dimethoxyphenyl)ethenylphosphonate 5b.
Reaction of 1-bromo-2-(2,5-dimethoxyphenyl)ethene (E:Z =
15:85) with diphenyl phosphonate, according to the typical
procedure in dimethylformamide using triethylamine as the
base, led to diphenyl 2-(2,5-dimethoxyphenyl)ethenylphos-
phonate 5b (E:Z > 99:1) (15%) as colourless crystals, which
were recrystallized from methanol to give pure (E)-isomer,
mp 89 ЊC (Found: C, 66.2; H, 5.25. C22H21O5P requires C, 66.7;
H, 5.3%); νmax(KBr)/cmϪ1 3000, 2900 and 1180; δH(CDCl3) 3.78
(s, 3H, OMe), 3.82 (s, 3H, OMe), 6.55 (dd, 1H, J 17.7 and 21.4,
C᎐CH᎐P) and 6.86–7.39 (m, 13H, aromatic) and 7.95 (dd, 1H,
᎐
J 17.7 and 25.0, Ar᎐CH᎐C); m/z (EI) 396 (Mϩ, 55%), 303 (80)
᎐
and 162 (100).
Diphenyl 2-(4-methylthiophenyl)ethenylphosphonate 5c. Reac-
tion of 1-bromo-2-(4-methylthiophenyl)ethene (E:Z = 7:93)
with diphenyl phosphonate, according to the typical procedure
in dimethylformamide using triethylamine as the base, led to
diphenyl 2-(4-methylthiophenyl)ethenylphosphonate 5c (E:Z =
65:35) (65%) as colourless crystals, which were recrystallized
to give pure (E)-isomer, mp 91 ЊC (Found: C, 65.5; H,
5.0. C21H19O3PS requires C, 66.0; H, 5.0%); νmax(KBr)/cmϪ1
3050 and 1210; δH(CDCl3) (E)-isomer: 2.48 (s, 3H, SMe),
C᎐CH᎐P) and 6.78–8.83 (m, 14H, aromatic and Ar᎐CH᎐C);
᎐
᎐
(Z)-isomer: 5.74 (dd, 1H, J 14.3 and 16.7, C᎐CH᎐P) and
᎐
6.78–8.83 (m, 14H, aromatic and Ar᎐CH᎐C); m/z (EI) 342
᎐
(Mϩ, 53%), 249 (56), 173 (68) and 97 (100).
The copper-promoted Arbuzov type vinylic substitution reaction
to give ð-electron conjugated phosphonates and phosphinamides
Typical procedure. Preparation of diethyl 2-phenylethenyl-
phosphonate 1b. A mixture of copper() iodide (10.26 g, 54
mmol), triethyl phosphite (8.97 g, 54 mmol) and 1-bromo-2-
phenylethene (E:Z = 84:16) (3.29 g, 18 mmol) in tetramethyl-
urea (15 cm3) was heated at 120 ЊC under a nitrogen atmosphere
with stirring for 24 h. To the cooled reaction mixture was added
benzene (30 cm3) and brine (30 cm3), and the organic layer
was separated, washed with ca. 10% aqueous sodium hydrogen
sulfate (30 cm3) and brine (30 cm3 × 5), and dried with sodium
sulfite. The solvent was evaporated and the residue was purified
by silica gel column chromatography with hexane–dichloro-
methane as the eluents to give the phosphonate 1b (E:Z =
86:14) (3.50 g, 81%) as a colourless oil.
6.39 (dd, 1H, J 17.4 and 19.5, C᎐CH᎐P) and 6.78–7.71 (m,
᎐
15H, Ph᎐CH᎐C and PhO); (Z)-isomer: 2.47 (s, 3H, SMe), 5.90
᎐
(dd, 1H, J 14.2 and 17.6, C᎐CH᎐P) and 6.78–7.71 (m, 15H,
᎐
aromatic and Ar᎐CH᎐C); m/z (EI) 382 (Mϩ, 100%) and 288 (40).
᎐
Diphenyl 2-(4-dimethylaminophenyl)ethenylphosphonate 5d.
Reaction of 1-bromo-2-(4-dimethylaminophenyl)ethene (E:Z =
18:82) with diphenyl phosphonate, according to the typical
procedure in THF using triethylamine as the base for 48 h, led
to diphenyl 2-(4-dimethylaminophenyl)ethenylphosphonate 5d
(71%) as colourless crystals, which were recrystallized to give
pure (E)-isomer, mp 127 ЊC (Found: C, 69.65; H, 5.9; N, 3.9.
C22H22NO3P requires C, 69.65; H, 5.8; N, 3.7%); νmax(KBr)/
cmϪ1 3000, 2900 and 1160; δH(CDCl3) (E)-isomer: 3.00 (s, 6H,
Me N), 6.12 (dd, 1H, J 17.2 and 20.0, C᎐CH᎐P) and 6.62–7.40
Using this method phosphonates 1a, 1c and 1d were prepared
in 35, 53 and 25% isolated yields respectively.
᎐
2
(m, 14H, aromatic); m/z (EI) 379 (Mϩ, 100%).
Diphenyl 2-(4-diphenylaminophenyl)ethenylphosphonate 5e.
Reaction of 1-bromo-2-(4-diphenylaminophenyl)ethene (E:Z =
19:81) with diphenyl phosphonate, according to the typical
procedure in DMF using triethylamine as the base for 48 h, led
to diphenyl 2-(4-diphenylaminophenyl)ethenylphosphonate 5e
(E:Z = 67:33) (83%) as colourless crystals, mp 123 ЊC (Found:
C, 76.4; H, 5.2; N, 3.2. C32H26NO3P requires C, 76.3; H, 5.2; N,
2.8%); νmax(KBr)/cmϪ1 3050 and 1190; δH(CDCl3) (E)-isomer:
N,N,NЈ,NЈ-Tetraethyl-P-phenylphosphonic diamide 8
Reaction of bromobenzene with tris(diethylamino)phosphine,
according to the typical procedure in DMF for 40 h, led to
diamide 8 (27%) as a colourless oil, bp ca. 90 ЊC/0.1 mmHg
(lit.,47 95 ЊC/0.5 mmHg); νmax(KBr)/cmϪ1 2920 and 1180;
δH(CDCl3) 1.04 (t, 12H, J 7.2, NCH2CH3), 3.03–3.14 (m, 8H,
NCH2CH3), 7.4–7.5 (m, 3H) and 7.73–7.82 (m, 2H); m/z (EI)
268 (Mϩ, 20%) and 196 (100).
6.24 (dd, 1H, J 17.4 and 19.8, C᎐CH᎐P) and 6.77᎐7.68 (m, 25H,
᎐
aromatic and Ar᎐CH᎐C); (Z)-isomer: 5.76 (dd, 1H, J 14.0
᎐
and 17.7, C᎐CH᎐P) and 6.77–7.68 (m, 25H, aromatic and
N,N,NЈ,NЈ-Tetraethyl-P-(2-phenylethenyl)phosphonic diamide 9
Reaction of 1-bromo-2-phenylethene (E:Z = 84:16) with tris-
(diethylamino)phosphine, according to the typical procedure
in DMF for 40 h, led to diamide 9 (E:Z = 84:16), which was
recrystallized from methanol to give pure (E)-isomer (49%)
as colourless crystals, mp 113–115 ЊC (lit.,10 103.5 ЊC) (Found:
᎐
Ar᎐CH᎐C); m/z (EI) 379 (Mϩ, 100%).
᎐
Diphenyl 2-(4-diphenylphosphinophenyl)ethenylphosphonate
5f. Reaction of 1-bromo-2-(4-diphenylphosphinophenyl)ethene
(E:Z = 23:77) with diphenyl phosphonate, according to the
typical procedure in DMF using triethylamine as the base for
J. Chem. Soc., Perkin Trans. 1, 1998
2957

View Online
13 O. I. Kolodyazhnyi, D. B. Golokhov and V. P. Kukhar, Zh. Obshch.
C, 65.6; H, 9.6; N, 9.5. C16H27N2OP requires C, 65.3; H, 9.2; N,
9.5%); νmax(KBr)/cmϪ1 2900 and 1160; δH(CDCl3) (E)-isomer:
1.11 (t, 12H, J 7.0, NCH2CH3), 3.04–3.13 (m, 8H, NCH2CH3),
Khim., 1986, 56, 1035.
14 D. S. Chemla and J. Zyss, ‘Nonlinear Optical Properties of Organic
Molecules and Crystals’, Academic Press, Orlando, FL, 1987.
15 ‘Organic Materials for Non-linear Optics’, ed. R. A. Hann and
D. Bloor, Royal Society of Chemistry, Cambridge, UK, 1989.
16 ‘Electrical, Optical, and Magnetic Properties of Organic Solid
State Materials’, ed. A. F. Garito, A. K.-Y. Jen, C. Y.-C. Lee and
L. R. Dalton, MRS, Pittsburgh, PA, 1994.
6.35 (dd, 1H, J 17.4 and 18.3, C᎐CH᎐P) and 7.34–7.53 (m, 6H,
᎐
Ph᎐CH᎐C); m/z (EI) 294 (Mϩ, 16%) and 222 (100).
᎐
X-Ray crystallographic study of 5d
Crystal data. C22H22NO3P, M = 379.39. Orthorhombic, a =
19.611(1), b = 5.745(8), c = 17.667(6) Å, V = 1990(1) Å3 (by
least-squares refinement on diffractometer angles for 25 care-
fully centred reflections in the range 20.41 < 2θ < 29.80Њ,
λ = 0.71069 Å), space group Pca21 (No. 29), Z = 4, Dcalc = 1.266
g cmϪ3, colourless prism 0.62 × 0.22 × 0.12 mm, µ(Mo-Kα) =
17 ‘Electrical, Optical, and Magnetic Properties of Organic Solid
State Materials III ’, ed. A. K.-Y. Jen, C. Y.-C. Lee, L. R. Dalton,
M. F. Rubner, G. E. Wnek and L. Y. Chiang, MRS, Pittsburgh,
PA, 1996.
18 S. M. Risser, D. N. Beratan and S. R. Marder, J. Am. Chem. Soc.,
1993, 115, 7719.
19 A preliminary communication was reported in the Abstract of the
68th National Meeting of the Chemical Society of Japan, Nagoya,
1A406 (1994).
20 H. A. Kurtz, J. J. P. Stewart and K. M. Dieter, J. Comp. Chem., 1990,
11, 82.
21 S. K. Kurz and T. T. Perry, J. Appl. Phys., 1968, 39, 3798.
22 J. F. Nye, ‘Physical Properties of Crystals’, Oxford University Press,
Oxford, UK, 1960.
23 J. Zyss and J. L. Oudar, Phys. Rev. A, 1982, 26, 2028.
24 J. F. Nicoud and R. J. Tweig, ‘Design and Synthesis of Organic
Molecular Compounds for Efficient Second-Harmonic Generation’,
ed. D. S. Chemla and J. Zyss, Academic Press, 1987.
25 G. U. Bublitz, R. Ortiz, S. R. Marder and S. G. Boxer, J. Am. Chem.
Soc., 1997, 119, 3365.
26 G. U. Bublitz, R. Ortiz, C. Runser, A. Fort, M. Barzoukas, S. R.
Marder and S. G. Boxer, J. Am. Chem. Soc., 1997, 119, 2311.
27 S. R. Marder, C. B. Gorman, F. Meyers, J. W. Perry, G. Bourhill,
J.-L. Bredas and B. M. Pierce, Science, 1994, 265, 632.
28 F. Meyers, S. R. Marder, B. M. Pierce and J. L. Bredas, J. Am.
Chem. Soc., 1994, 116, 10 703.
29 C. B. Gorman and S. R. Marder, Proc. Natl. Acad. Sci. USA, 1993,
90, 11 297.
30 V. V. Tkachev, L. O. Atovmyan, B. V. Timokhin, O. A. Bragina,
G. V. Ratovskii and L. M. Sergienko, Zh. Strukt. Khim., 1986, 27,
121.
31 V. R. Pedireddi, J. A. R. P. Sarma and G. R. Desiraju, J. Chem. Soc.,
Perkin Trans. 2, 1992, 311.
32 T. S. Cameron, D. J. Cowley and J. E. Thompson, J. Chem. Soc.,
Perkin Trans. 2, 1974, 774.
33 J. L. Oudar, J. Chem. Phys., 1977, 67, 446.
34 N. Ono, N. Bansho, S. Ito, T. Murashima and T. Ogawa,
J. Heterocyclic Chem., 1997, 34, 1243.
35 M. Matsumoto and K. Kuroda, Tetrahedron Lett., 1980, 21,
4021.
36 S. Akiyama, S. Nakatsuji, K. Yoshida, K. Nakashima, T. Hagiwara,
H. Tsuruta and T. Yoshida, Bull. Chem. Soc. Jpn., 1983, 56, 361.
37 M. K. Loar and J. K. Stille, J. Am. Chem. Soc., 1981, 103, 4174.
38 T. N. Baker, W. P. Dohery, W. S. Kelley, W. Newmeyer, J. E. Rogers,
R. E. Spalding and R. I. Walter, J. Org. Chem., 1965, 30, 3714.
39 G. P. Schiemenz, Chem. Ber., 1966, 99, 504.
40 G. Maas and M. Regitz, Chem. Ber., 1976, 109, 2039.
41 Y. Xu, J. X. G. Huang and Y. Huang, Synthesis, 1983, 556.
42 O. I. Kolodyazhnyi and D. B. Golokhov, Zh. Obshch. Khim., 1987,
57, 2640.
43 A. B. Antonova, A. V. Kovalenko, R. A. Cherkasov, V. V.
Ovchinnikov, A. A. Ioganson, E. D. Korniets and N. A. Deikhiha,
Zh. Obshch. Khim., 1987, 57, 1030.
44 T. Hirao, T. Masunaga, Y. Ohshiro and T. Agawa, Synthesis, 1981,
56.
45 M. Mikolajczyk, S. Grzejszczak, W. Midura and A. Zatorski,
Synthesis, 1976, 396.
46 D. W. Allen, J. Chem. Soc., Perkin Trans. 2, 1977, 789.
47 D. W. Allen, B. G. Hutley and T. J. M. Mellor, J. Chem. Soc., Perkin
Trans. 2, 1972, 63.
48 D. T. Cromer and J. T. Waber, ‘International Tables for X-ray
Crystallography’, The Kynoch Press, Birmingham, England,
1974.
49 J. A. Ibers and W. C. Hamilton, Acta Crystallogr., 1964, 17, 781.
50 TEXSAN–TEXRAY Structure Analysis Package, Molecular
Structure Corporation, 1985, 1992.
1.59 cmϪ1
.
Data collection and processing. A Rigaku AFC5R diffrac-
tometer with graphite-monochromated Mo-Kα radiation was
used: ω Ϫ 2θ scan technique to a maximum 2θ value of 55.0Њ.
Scans of (1.47 ϩ 0.30 tan θ)Њ were made at a speed of 16.0
minϪ1 (in omega). A total of 2669 reflections were collected.
The intensities of three representative reflections which were
measured after every 150 reflections declined by Ϫ1.80%. A
linear correction factor was applied to the data to account
for this phenomenon.
Structure solution and refinement. The structure was solved
by direct methods (SIR92). The non-hydrogen atoms were
refined anisotropically. The final cycle of full-matrix least-
squares refinement was based on 1284 observed reflections
[I > 1.00σ(I)] and 245 variable parameters and converged
(largest parameter shift was 0.01 times its esd). Neutral
atom-scattering factors were taken from Cromer and Waber.48
49
Anomalous dispersion effects were included in Fcalc
;
the
values for ∆f Ј and ∆f Љ were those of Cromer.48 Final R = 0.052,
Rw = 0.055, goodness of fit = 1.10. The final ∆F synthesis
showed no peaks outside the range from Ϫ0.23 to ϩ0.21 e ÅϪ3
.
All the calculations were performed using the TEXSAN50
crystallographic software package of Molecular Structure
Corporation. The ORTEP programmes were used to obtain
Fig. 1.
Full crystallographic details, excluding structure factor
tables, have been deposited at the Cambridge Crystallographic
Data Centre (CCDC). For details of the deposition scheme,
see ‘Instructions for Authors’, J. Chem. Soc., Perkin Trans. 1,
available via the RSC Web page (http://www.rsc.org/authors).
Any request to the CCDC for this material should quote the
full literature citation and the reference number 207/244.
Acknowledgements
We thank the Ministry of Education, Science and Culture
of Japan for financial support of this work by a Grant-in-Aid
for Science and Research (No. 05804038).
References
1 L.-B. Han and M. Tanaka, J. Am. Chem. Soc., 1996, 118, 1571.
2 Y. Xu, M. T. Flavin and Z.-Q. Xu, J. Org. Chem., 1996, 61, 7697.
3 T. Minami and J. Motoyoshiya, Synthesis, 1992, 333.
4 A. Bhattacharya and G. Thyagarajan, Chem. Rev., 1981, 81, 415.
5 T. Ogawa, S. Hotta, K. Hayami and H. Suzuki, Chem. Lett., 1992,
1947.
6 T. Ogawa, T. Kiji, K. Hayami and H. Suzuki, Chem. Lett., 1991,
1443.
7 T. Ogawa, K. Hayami, H. Iyama and H. Suzuki, Chem. Lett., 1990,
937.
8 T. Ogawa, K. Kusumi, M. Tanaka, K. Hayami and H. Suzuki,
Synth. Commun., 1989, 19, 2199.
9 T. Ogawa, K. Hayami and H. Suzuki, Chem. Lett., 1989, 769.
10 O. I. Kolodyazhnyi and D. B. Golokhov, Zh. Obshch. Khim., 1989,
59, 293.
51 J. J. Stewart, QCPE Bull., 1989, 9, 10.
11 M. Fild, R. Fischer and W. Handke, Z. Anorg. Allg. Chem., 1988,
561, 157.
12 M.-P. Teulade and P. Savignac, J. Organomet. Chem., 1986, 304,
283.
Paper 7/09082J
Received 18th December 1997
Accepted 17th July 1998
2958
J. Chem. Soc., Perkin Trans. 1, 1998