Demethylation reactions of phosphate esters catalysed by complexes of polyether ligands with metal iodides

Article abstract of DOI:10.1039/b105877k

Metal ion catalysis has been observed in demethylation reactions of methyl di(para-substituted phenyl)phosphate esters 1-4 promoted by complexes of polyether ligands 5-7 with alkali metal iodides MI (M = Li, Na, K) in low polarity media (chlorobenzene, 1,

Full text of DOI:10.1039/b105877k

View Article Online / Journal Homepage / Table of Contents for this issue
Demethylation reactions of phosphate esters catalysed by
complexes of polyether ligands with metal iodides
Dario Landini, Angelamaria Maia* and Cristina Pinna
2
Centro CNR and Dipartimento di Chimica Organica e Industriale dell’Università,
Via Venezian 21, I-20133 Milano, Italy. E-mail: angelamaria.maia@unimi.it
Received (in Cambridge, UK) 3rd July 2001, Accepted 28th September 2001
First published as an Advance Article on the web 8th November 2001
Metal ion catalysis has been observed in demethylation reactions of methyl di(para-substituted phenyl)phosphate
esters 1–4 promoted by complexes of polyether ligands 5–7 with alkali metal iodides MI (M = Li, Na, K) in low
polarity media (chlorobenzene, 1,2-dichlorobenzene). The catalytic effect is found to depend on both the metal ion
ϩ
ϩ
ϩ
charge density and the ligand topology. Rate constants increase, in the order K < Na < Li , with the complexes
of crown ether 5 and PEG 6, whereas they are much lower and independent of the cation with the cryptates of 7.
Reaction rates are favored by electron-withdrawing substituents, with Hammett ρ-values that increase with
ϩ
ϩ
ϩ
decreasing Lewis acid character of the cation (Li < Na < K ) or when the charge of the latter is shielded
by a good complexing agent like cryptand 7. Kinetic data are explained on the basis of a concerted
“
push–pull” mechanism.
Introduction
Macrocyclic and open chain polyethers (crown ethers, cryp-
tands, polyethylene glycols) are synthetic ligands of particular
interest owing to their unique ability to solubilize metal salts in
organic media, even of low polarity, by formation of stable and
1,2
highly structured inclusion complexes. In these complexes the
anion is remarkably activated due to the limited interaction
with both the bulky complexed cation and the reaction medium.
The anionic reactivity is mainly determined by the ability of the
ligand to bind effectively the cation, and hence to activate the
ion-paired anion, increasing in the order: polyethylene glycol <
1,2
crown ether Ӷ cryptand.
The topology of the ligand also plays a fundamental role in
reactions where metal ions are involved in the activation process
Fig. 1 Second-order plot for the reaction of (4-CH C H O) P(O)-
3
6
4
2
Ϫ2
Ϫ3
ϩ
Ϫ
(
“electrophilic catalysis”). In this case the cation complexation
OCH 2 (1.08 × 10 mol dm ) with (Na ʚ PHDB18C6)I (0.481 ×
3
Ϫ2 Ϫ3
1
0
mol dm ) in chlorobenzene, at 60 ЊC.
by the host ligand is expected to cause rate decrease through
1b,3
inhibition of cation participation.
In early studies we
revealed cation assistance in nucleophilic substitution reactions
of alkylsulfonates catalysed by complexes of metal iodides with
[2.2.2,C10] 7 with alkali metal and ammonium iodides MI
(M = Li, Na, K, NH ) in low polarity solvents (chlorobenzene,
4
4
polyether ligands in chlorobenzene. Our results showed that
1,2-dichlorobenzene). In order to obtain a comparison with
the corresponding uncatalysed reactions we also performed
kinetic experiments with the lipophilic quaternary onium salt
catalysis is inhibited by good complexing agents such as cryp-
tands whereas it is favored by less efficient ligands (crown ethers
4
ϩ Ϫ
or polyethylene glycols).
Hexyl N I .
4
Recently, we reported on metal ion effects in demethylation
reactions of phosphinic esters promoted by complexes of macro-
cyclic and open chain polyethers with alkali and alkaline-earth
Results
5
metal salts in low polarity media. Demethylation reactions are
Kinetic determinations were carried out by reacting compar-
Ϫ2
of particular interest since the final removal of the protecting
group (usually a methyl group) from the phosphate function of
the substrate often represents the final step of the synthesis of
able amounts of phosphate esters 1–4 (0.2–1.3 × 10 mol
Ϫ3
ϩ
Ϫ
Ϫ2
dm ) and preformed (M ʚ Lig) I complex (0.1–0.6 × 10
Ϫ3
mol dm ) in the appropriate organic solvent (chlorobenzene or
1,2-dichlorobenzene) at 60 ЊC [reaction (1)].
6–8
oligonucleotides with the phosphotriester method.
Phos-
phate esters are known to be an integral part of a number of
biologically active molecules ranging from DNA and ATP to
Rates have been measured by potentiometric titration of the
complexed nucleophile I . Under these conditions the reactions
follow regular second-order kinetics [eqn. (2)] up to at least two
half-lives (Fig. 1).
For sake of comparison the phosphate ester 1 was treated
with comparable amounts of a non-complexable quaternary
onium salt such as tetrahexylammonium iodide, in chloro-
benzene and 1,2-dichlorobenzene. As expected, the reaction (1)
is slower in both solvents and the corresponding rate constants
Ϫ
6–9
pesticides up to nerve agents. As a consequence, the reactions
that involve phosphate esters play leading roles in the chemical
6–9
processes of life.
In this paper we present a kinetic study of the effect of the
cation in demethylation reactions of methyl di(para-substituted
phenyl)phosphates 1–4 promoted by complexes of polyether
ligands PHDB18-crown-6 5, PEG400Me2 6 and cryptand
2
314
J. Chem. Soc., Perkin Trans. 2, 2001, 2314–2317
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b105877k

View Article Online
ϩ Ϫ
ϩ
Ϫ
Table 1 Second-order rate constants for the demethylation reaction of 1 by (M ʚ Lig)I complexes and tetrahexylammonium iodide (Q I ) in
chlorobenzene, at 60 ЊC
b
3
Ϫ1 Ϫ1
k /dm mol
s
Ϫ1
ϩ
Q
ϩ
ϩ
ϩ
Cation
(z/r)/Å
(M ʚ PHDB18C6)
(M ʚ PEG400Me2)
(M ʚ [2.2.2,C ])
10
ϩ
ϩ
Hexyl N
0.23
0.23
0.69
0.75
1.02
1.28
0.0075
0.0050
4
c
Hexyl N
4
ϩ
NH4
0.016
0.022
0.11
0.0098
0.013
0.048
0.67
0.0184
0.0179
0.0181
0.0180
ϩ
K
ϩ
Na
Li
ϩ
0.53
a
b
3
Ϫ2
Ϫ3
ϩ
Ϫ
ϩ
Ϫ
Ϫ2
Ϫ3
A chlorobenzene solution (20–40 cm ) of substrate 1 (0.5–2 × 10 mol dm ) and of the (M ʚ Lig)I complex or Q I (0.2–2 × 10 mol dm ).
c
Average of at least two determinations. The error in these values is estimated to be 5%. Rate constant value in 1,2-dichlorobenzene.
Table 2 Second-order rate constants for the demethylation reaction
ϩ
Ϫ
ϩ
ϩ
ϩ
ϩ
of 1–4 by complexes (M ʚ Lig)I (M = Li , Na , K ) (Lig =
PEG400Me 6, [2.2.2,C ] 7) in chlorobenzene, at 60 ЊC
2
10
b
3
Ϫ1 Ϫ1
k /dm mol
s
ϩ
M
ʚ PEG400Me2
ϩ
ϩ
Na
ϩ
K
ϩ
M
X
σp
Li
ʚ [2.2.2,C10]
c
c
4
3
1
2
NO2
Cl
H
0.81
0.24
0.0
10.6
1.02
0.67
0.495
1.89
0.16
0.048
0.024
1.80
3.12
0.114
0.018
0.077
0.013
0.0086
c
c
CH3
Ϫ0.14
0.0089
a
3
Ϫ2
A chlorobenzene solution (20–60 cm ) of substrate 1–4 (0.2–2 × 10
Ϫ3
ϩ
Ϫ
Ϫ2
Ϫ3
mol dm ) and of the (M ʚ Lig)I complex (0.1–2 × 10 mol dm ).
b
Average of at least two determinations. The error in these values is
c
ϩ
ϩ
ϩ
estimated to be 5%. Average rate constant value for M = Li , Na ,
K
ϩ
(see Table 1 for 1).
Ϫ
rate = k [substrate] [complexed I ]
(2)
are lower than the slowest catalysed reactions. Results are
reported in Tables 1 and 2. As shown in Table 1, the second-
3
Ϫ1 Ϫ1
order rate constants k (dm mol s ) of demethylation of 1
catalysed by the complexes of crown ether 5 and PEG400-
Me2 6 remarkably depend on the cation, increasing in the
ϩ
ϩ
ϩ
ϩ
Fig. 2 Hammett correlation for the demethylation reaction of phos-
ϩ Ϫ ϩ ϩ
order NH4 < K < Na < Li , whereas they are lower and
independent of the cation with the complexes of cryptand 7.
The effect of substituents [X = H (1); CH (2); Cl (3); NO (4)]
phate esters 1–4 by complexes (M ʚ PEG400Me , 6)I , M = Li (᭺),
2
ϩ
ϩ
ϩ
Na (᭿), K (ᮀ), and (M ʚ [2.2.2,C ], 7) (᭹), in chlorobenzene, at
1
0
3
2
60 ЊC.
3
Ϫ1 Ϫ1
on the rate constant k (dm mol s ) of the reaction (1) was
studied by reacting the series of methyl di(para-substituted
phenyl)phosphates 1–4 with the complexes of ligands 5–7
metal iodides. The findings can be rationalized by assuming
ϩ
a transition state where the metal ion M complexed by the
ϩ
Ϫ
ϩ
ϩ
ϩ
ϩ
ϩ
Ϫ
Ϫ
with the alkali metal iodides M I (M = Li , Na and K )
in anhydrous chlorobenzene, at 60 ЊC. Reaction rates are
favored by electron-withdrawing substituents with ρ values that
markedly change depending on the metal ion charge and the
topology of the ligand (Table 2 and Fig. 2).
ligand, (M ʚ Lig), can interact with the ion-paired anion I
and, at the same time, with the leaving group (ArO) PO2
2
(Scheme 1).
ϩ
Ϫ
In the activation process the (M ʚ Lig)I ion pair reacts
following a concerted “push–pull” mechanism where the com-
plexed metal ion stabilizes the developing negative charge on
the phosphoric oxygens while favoring the simultaneous nucleo-
philic attack of the iodide at the methyl carbon (“electrophilic
catalysis”). We have recently proposed an analogous mechan-
ism in dealkylation reactions of phosphinic esters by com-
Discussion
Results provide strong evidence of metal ion assistance (“electro-
philic catalysis”) in the demethylation of methyl diarylphos-
phates 1–4 promoted by complexes of ligands 5–7 with alkali
5
plexes of metal salts with polyether ligands. The higher the
J. Chem. Soc., Perkin Trans. 2, 2001, 2314–2317
2315

View Article Online
of catalysis was not compensated in part by the increased
nucleophilicity of the iodide ion-paired with the bulky quater-
ϩ 2d,e
nary cation Hexyl N . Metal ion participation is confirmed
4
by the lower rate constant values obtained with the complexes
of cryptand 7 (Table 1). Since cryptands are known to be excel-
lent cation complexing agents, and hence powerful anion
activators, the lower reactivity found, up to about 40 fold, is
another additional proof of cation electrophilic assistance to
reaction (1).
The study of the effect of the para substituents X on the
demethylation rate of phosphate esters 1–4 has revealed that,
ϩ
Ϫ
the complex (M ʚ Lig) I being the same, the second-order
3
Ϫ1 Ϫ1
rate constants k (dm mol s ) are linearly related with the
12
corresponding Hammett σ -values. Reactions are favored by
p
electron-withdrawing groups, that stabilize the developing
negative charge on the phosphoric oxygens (Scheme 1), with
ρ values ranging from 0.7 up to 1.38, depending on both the
ϩ
ϩ
ϩ
3
Ϫ1
metal cation (Li , Na , K ) and the ligand (5–7) (Table 2,
Fig. 3 Correlation between second-order rate constant k/dm mol
Ϫ1
Ϫ1
s
and cation charge density (z/r)/Å in the demethylation reaction
Fig. 2).
ϩ
Ϫ
ϩ
ϩ
ϩ
ϩ
ϩ
ϩ
of 1 by complexes (M ʚ Lig)I (M = NH4 , K , Na , Li )
Interestingly, as shown in Fig. 2 for the complexes (M
ʚ
(
1
Lig = PHDB18C6 5 (᭹), [2.2.2, C ] 7 (᭺)), in chlorobenzene (—) and
Ϫ
ϩ
ϩ
1
0
PEG400Me )I , the ρ values increase, in the order Li < Na <
2
,2-dichlorobenzene (---), at 60 ЊC.
ϩ
K , revealing a different sensitivity of the reaction (1) to the
ϩ
effects of the substituent X on changing the cation M . These
data can be rationalized in terms of a different involvement of
metal ion and substituent in the activation process (Table 2).
With a cation of high density of charge like lithium the electro-
philic catalysis of the metal ion is most likely the dominant
effect and hence the reaction is scarcely affected by the elec-
tronic effects of the substituent (ρ = 0.71). By contrast, with
ϩ
ϩ
cations of lower charge density (Na > K ) the prevailing
effect becomes that of the substituent and the ρ values increase
ϩ
ϩ
in the same order [ρ = 0.99 (Na ); ρ = 1.25 (K )]. This hypo-
thesis is also confirmed by the similar value (ρ = 1.38) obtained
with the complexes of cryptand [2.2.2,C ] 7 where the cation,
10
completely included into the cavity of the tridimensional
ligand, cannot assist the departure of the leaving group in the
transition state (Scheme 1).
Scheme 1
A very similar result (ρ = 1.1) was reported by Osborne in
reactions of methyl di(para-substituted phenyl)phosphates
with pyridine to give the corresponding N-methylpyridinium
ϩ
interaction of the complexed cation (M ʚ Lig) with the leav-
ing group the higher must be the reactivity of the iodide and
hence the higher the corresponding rate of demethylation.
13
phosphates.
3
Ϫ1 Ϫ1
As shown in Table 1, the rate constant k (dm mol s ) of
the demethylation of (PhO) P(O)OMe 1 catalysed by the com-
plexes of crown ether PHDB18-crown-6 5 and polyethylene
2
Conclusions
glycol PEG400Me 6 with the series of the alkali metal iodides
2
The data as a whole reveal metal ion participation (“electro-
philic catalysis”) in the demethylation reaction of methyl
diarylphosphates 1–4 promoted by complexes of polyethers 5–7
with alkali metal iodides in low polarity media. The catalytic
effect depends on the Lewis acid character of the cation,
increases 24 and 52 times respectively on switching from KI to
the corresponding complex with the lithium salt. The reactivity
ϩ
ϩ
ϩ
sequence so obtained, in the order kK < kNa < kLi , is in line
ϩ
with the increasing interaction of M with the leaving group
ϩ
ϩ
ϩ
ϩ
on increasing the Lewis acid character of the cation (K
<
increasing in the order K < Na < Li , as well as on the ability
of the ligand to effectively shield the metal ion charge.
ϩ
ϩ
10
Na < Li ). Remarkably, this trend parallels that previously
observed in the demethylation of the corresponding methyl
Kinetic results show a different sensitivity (different Hammett
ρ values) of the reaction to the substituent effects on changing
diphenylphosphinate Ph P(O)OMe under identical conditions
2
ϩ
ϩ
ϩ
5
ϩ
(
kK < kNa < kLi ).
the cation M . Whereas metal ion catalysis is the dominant
The electrophilic catalysis is also proved by the excellent3
effect with the high charge lithium, the substituent effect
progressively increases with cations of lower charge density
linear correlation found by plotting the rate constant k (dm
Ϫ1 Ϫ1
mol s ) as a function of the charge density of the cation,
(
sodium and potassium) and prevails whatever the cation may
ϩ
defined as M charge vs. ionic radius ratio, for the series MI
be with a better complexing agent like cryptand 7.
(
M = NH , K, Na, Li) (Fig. 3 and Table 1). In addition, the
4
sequence of reactivity, unchanged on going from chloro-
benzene to 1,2-dichlorobenzene, does not reveal any signifi-
cant modification of the transition state with the polarity of
the medium. The cation being the same, the trend found
kchlorobenzene > k1,2-dichlorobenzene simply reflects the decreasing
nucleophilic reactivity of the anion on increasing the solvent
Experimental
General methods
Potentiometric titrations were carried out with a Metrohm 670
Titroprocessor by using a combined silver electrode isolated
with a potassium nitrate bridge. Karl–Fischer determinations
11
polarity (Fig. 3).
1
31
As reported in Table 1, the enhancement of reactivity is up to
about 90 times if the comparison is extended to the quaternary
were performed with a Metrohm 684KF coulometer. H and P
NMR spectra were recorded on Bruker AC 300 and AMX 300
spectrometers, using as external references TMS and aqueous
85% H PO , respectively.
ϩ
Ϫ
onium salt Hexyl N I for which no electrophilic assistance is
4
expected. Such enhancement could be still higher if the absence
3
4
2
316
J. Chem. Soc., Perkin Trans. 2, 2001, 2314–2317

View Article Online
1
31
Materials and solvents
the crude mixture and identified by H and P NMR. The
spectroscopic analysis showed the exclusive formation of the
Tetrahexylammonium iodide, ligands PHDB18-crown-6 5 and
Ϫ
ϩ
demethylation products (4-XC H O) P(O)O M (X = H, CH ,
6
4
2
3
PEG400Me2 6 and [2.2.2,C ] 7 were commercial products,
10
ϩ
ϩ
ϩ
ϩ
31
Cl, NO ; M = Li , Na , K ) in all cases. Indeed, the P NMR
2
utilized as purchased. Alkali metal and ammonium iodides,
AnalaR grade commercial products, were first ground then
dehydrated in an oven at 110–120 ЊC under vacuum for several
spectra (D O) showed only a singlet centred at δ Ϫ 7.8 to Ϫ8.1
2
1
ppm and the H NMR spectra (D O) confirmed the absence
2
of the methoxy group. In particular, they are as follows:
hours. In all cases the water content is ≤0.05 mol of H O per
2
Ϫ
ϩ
31
1
(
C H O) P(O)O M , P NMR, δ Ϫ8.13 (s); H NMR, δ 7.38
6 5 2
Ϫ ϩ 31
mol of salt (Karl Fischer titration). Dry (Fluka) chlorobenzene
(
m). (4-CH C H O) P(O)O M , P NMR, δ Ϫ7.80 (s);
3 6 4 2
7.12 (dd, 8H,
4-ClC H O) P(O)O M , P NMR, δ Ϫ8.01 (s); H NMR,
δ 7.10 (dd, J 7). (4-NO C H O) P(O)O M , P NMR, δ Ϫ8.10
s); H NMR, δ 7.81 (dd, J 9.17).
and 1,2-dichlorobenzene (H O ≤ 15 ppm) were used.
2
1
H
NMR,
δ
J
8.48),
δ
2.2 1 (s, 6H).
Synthesis of methyl di(para-substituted)phosphates 1–4 will
Ϫ
ϩ
31
(
14
6
4
2
be published elsewhere.
Ϫ
ϩ 31
2
6
4
2
1
(
Kinetic measurements
3
In a typical procedure a standardized solution (5–15 cm )
of substrate 1–4 (0.02–0.05 M) was added to a standardized
Acknowledgements
3
solution (15–45 cm ) of preformed complex or tetrahexyl-
This work was supported by CNR and Ministero dell’-
Università e della Ricerca Scientifica e Tecnologica (MURST).
Ϫ3
3
ammonium iodide (0.0028–0.050 mol dm ) in a 100 cm flask
3
kept at 60 ± 0.1 ЊC. Samples (2–15 cm ), withdrawn period-
3
ically, were quenched in ice-cold MeOH (50 cm ), and the
unreacted nucleophile I was potentiometrically titrated with
Ϫ
References
0
.01 M AgNO . The second-order rate constants were evalu-
3
1
(a) N. S. Poonia and A. V. Bajaj, Chem. Rev., 1979, 79, 389; (b) J. M.
Lehn, Pure Appl. Chem., 1980, 52, 2303; (c) R. M. Izatt, K. Pawlak
and J. S. Bradshaw, Chem. Rev., 1991, 91, 1721; (d ) G. W. Gokel,
Chem. Soc. Rev., 1992, 39; (e) T. J. Marrone and K. M. Merz, Jr.,
J. Am. Chem. Soc., 1995, 117, 779.
(a) D. Landini, A. Maia, F. Montanari and P. Tundo, J. Am. Chem.
Soc., 1979, 101, 2526; (b) D. Landini, A. Maia, F. Montanari and
F. M. Pirisi, J. Chem. Soc., Perkin Trans. 2, 1980, 46; (c) D. Landini,
A. Maia, G. Podda, D. Secci and Y. M. Yan, J. Chem. Soc., Perkin
Trans. 2, 1992, 1721; (d ) A. Maia, Pure Appl. Chem., 1995, 67,
697; (e) D. Landini, A. Maia and M. Penso, in Comprehensive
Supramolecular Chemistry, ed. J. M. Lehn, Pergamon Press, Oxford,
ated using a least-squares computer program (Excel program)
from the equation 1/([B] Ϫ [A] )ln ([B][A] /[A][B] ) = kt, where
0
0
0
0
A = substrate and B = complexed MI (or tetrahexylammonium
iodide) or vice versa. All rates involved at least eight samplings
and gave correlation coefficients of 0.997 or better.
2
The solutions of the preformed complexes of 5 and 7 were
prepared by magnetically stirring a standardised solution (20–
3
5
0 cm ) of ligand in the organic solvent (chlorobenzene, 1,2-
Ϫ3
dichlorobenzene) (0.009–0.010 mol dm ) with the appropriate
quantity of salt MI (0.7–0.9 mol per mol of ligand) as a solid
phase in a flask thermostatted at 60 ± 0.1 ЊC. The system was
stirred for 1–3 h and then kept without stirring for an additional
1
996, vol. 1, ch. 11, pp. 417–464.
3
4
J. M. Lefour and A. Loupy, Tetrahedron, 1978, 34, 2597.
(a) A. Gobbi, D. Landini, A. Maia and D. Secci, J. Org. Chem.,
1
0 min to allow good separation of the two phases. Aliquots
1
995, 60, 5954; (b) A. Gobbi, D. Landini, A. Maia and M. Penso,
J. Chem. Soc., Perkin Trans. 2, 1996, 2505.
5 D. Albanese, D. Landini and A. Maia, J. Org. Chem., 2001, 66,
249.
3
(
(
5–8 cm ) of the organic phase were centrifuged and samples
2–4 cm ) were withdrawn and titrated with 0.01 M AgNO .
3
3
Ϫ
ϩ
Potentiometric titrations of cryptates (M ʚ [2.2.2,C ])I
3
10
6
D. J. H. Smith, K. K. Ogilvie and M. F. Gillen, Tetrahedron Lett.,
1980, 21, 861.
L. H. Koole, H. M. Moody, N. L. H. L. Broeders, P. J. L. M.
Quaedflieg, W. H. A. Kuijpers, M. H. P. van Genderen, A. J. J. M.
Coenen, S. van der Wal and H. M. Buck, J. Org. Chem., 1989, 54,
1657.
were performed in acid medium (HNO ) in order to avoid the
3
2
a
simultaneous titration of the free ligand.
7
The fastest kinetics were performed by using the “quench-
ing” technique. In a typical procedure standardized solutions
3
Ϫ3
(
3 cm ) of substrate (0.01045 mol dm ) were added with stand-
3
ardized solutions (27 cm ) of preformed complex (0.00225 mol
dm ) in a series (7–10) of 50 cm flasks kept at 60 ± 0.1 ЊC. At
8
9
R. C. Lum and J. J. Grabowsky, J. Am. Chem. Soc., 1992, 114,
Ϫ3
3
8619.
R. A. Torres and T. C. Bruice, J. Am. Chem. Soc., 2000, 122, 781.
different times the reactions were quenched in ice-cold MeOH
1
1
0 R. G. Pearson, Coord. Chem. Rev., 1990, 100, 403.
1 C. Reichardt, in Solvents and Solvent Effects in Organic Chemistry,
Verlag-Chemie, Weinheim, New York, 1988.
3
(
50 cm ) and the unreacted iodide potentiometrically titrated
Ϫ3
with 0.01 mol dm AgNO3.
1
2 O. Exner, in Correlation Analysis in Chemistry, eds. N. B. Chapman
and J. Shorter, Plenum Press, New York, 1978, ch. 10.
3 D. W. Osborne, J. Org. Chem., 1964, 29, 3570.
Reaction products
1
The products of reaction (1) were isolated as solid salts from
14 D. Landini, A. Maia and C. Pinna; unpublished results.
J. Chem. Soc., Perkin Trans. 2, 2001, 2314–2317
2317