gem-difluorovinyl derivatives as insecticides and acaricides

Article abstract of DOI:10.2533/000942904777678163

The insecticidal lead 1,1-difluorododec-1-ene was optimised. This compound has attractive insecticidal activity against tobacco budworm (Heliothis virescens), banded cucumber beetle (Diabrotica balteata), pea aphid (Aphis cracciovora), brown planthopper (

Full text of DOI:10.2533/000942904777678163

FLUORINE IN THE LIFE SCIENCE INDUSTRY
108
CHIMIA 2004, 58, No. 3
Chimia 58 (2004) 108–116
© Schweizerische Chemische Gesellschaft
ISSN 0009–4293
gem-Difluorovinyl Derivatives
as Insecticides and Acaricides
Thomas Pitterna*, Manfred Böger, and Peter Maienfisch
Abstract: The insecticidal lead 1,1-difluorododec-1-ene was optimised. This compound has attractive insecticidal
activity against tobacco budworm (Heliothis virescens), banded cucumber beetle (Diabrotica balteata), pea aphid
(Aphis cracciovora), brown planthopper (Nilaparvata lugens), and green rice leafhopper (Nephotettix cincticeps).
Among different pharmacophore analogues, only 1,1-dichlorododec-1-ene and 1,1-difluoro-2-iodododec-1-ene
showed weak insecticidal activity, whereas similar compounds such as 1-chloro-1-fluorododec-1-ene, 1-fluorodo-
dec-1-ene, and 1,1-difluoro-2-bromododec-1-ene were inactive. Only bridge analogues with even-numbered car-
bon chains were active, for example 1,1-difluorodec-1-ene and 1,1-difluorotetradec-1-ene. Odd-numbered ana-
logues such as 1,1-difluoronon-1-ene, 1,1-difluoroundec-1-ene, 1,1-difluorotridec-1-ene, and 1,1-difluoro-pen-
tadec-1-ene showed no activity. Modification of the tail group led to the analogues 12,12-difluorododec-11-enoic
acid and its methyl ester, 12,12-difluorododec-11-en-1-ol, 1,1-difluoro-12-methoxydodec-1-ene, and 12,12-diflu-
orododec-11-enylamine, all of which showed insecticidal activity. 12,12-difluorododec-11-enoic acid methyl ester,
12,12-difluorododec-11-enoic acid, and 12,12-difluorododec-11-en-1-ol were also active against spider mites
(Tetranychus ssp). Thus, in a first optimisation cycle, broad activity against insect pests and mites was discovered.
Two requirements, the gem-difluorovinyl pharmacophore and an even-numbered carbon chain, were found to be
essential for activity. This latter requirement is in line with the proposed mode of action, which involves inhibition
of the β-oxidation of fatty acids in insect mitochondria. In a second optimisation cycle, it was found that 6,6-diflu-
orohex-5-enoic acid and its derivatives, such as acids, amides, and hydrazides, possess even superior properties
as insecticides and acaricides. This led to the discovery of 6,6-difluorohex-5-enoic acid 2-[4-(4-trifluoromethyl-
benzyloxy)-phenoxy]-ethyl ester (CGA 304’111). This compound showed excellent performance in field trials
against a wide range of pests, as well as a more favourable toxicological profile than earlier derivatives. For a large-
scale synthesis of CGA 304’111, five different synthetic routes for 6,6-difluorohex-5-enoic acid were developed.
The best route involved radical addition of CBrClF2 to pent-4-enoic acid. Removal of bromine by hydrogenation,
elimination of chloride and hydrolysis of the ester concluded this most efficient sequence. Thus, a practical syn-
thesis for CGA 304’111 was identified, which allowed the preparation of samples on a several 100 g scale.
Keywords: Acaricide · CGA 304’111 · 6,6-Difluorohex-5-enoic acid · gem-Difluorovinyl · Insecticide
Introduction
liothis virescens), banded cucumber beetle provide larger samples for field trials, a
As part of our activities in patent moni- (Diabrotica balteata), pea aphid (Aphis practical synthesis of CGA 304’111 was
toring [1], 1,1-difluorododec-1-ene (1) has cracciovora), brown planthopper (Nila- developed.
been prepared. In our insecticide screening parvata lugens) and green rice leafhopper
In this account, we describe the synthe-
this compound showed attractive insectici- (Nephotettix cincticeps). At the same time, sis of the compounds that were relevant for
dal activity against tobacco budworm (He- we learned from studies by Ruder [2], that our optimisation program, as well as their
the most important commercially estab- pesticidal activities. We also present differ-
lished insecticidal modes of action could be ent synthetic routes for 6,6-difluorohex-5-
excluded. Because of the broad activity enoic acid (35), which is an important
spectrum against key pests and the poten- building block for CGA 304’111 (37).
tially new mode of action, an optimisation
project was started.
Lead structure 1 can be dissected in 1st Optimisation Cycle
pharmacophore, bridge and tail (Fig. 1). In
a first optimisation cycle each of these
Lead structure 1 was obtained from
structural elements was optimised separate- aldehyde 3 by a fluoroylide reaction, (E)-1-
ly. This approach yielded compounds with fluorododec-1-ene (9) by reduction of 1
insecticidal and, in addition, acaricidal ac- with Redal®. Both reactions have been de-
tivity. Further optimisation led to the dis- scribed by Hayashi et al. [3]. 1,1-Difluo-
covery of CGA 304’111, a compound with rododecane (4) was prepared from alde-
improved properties. Because of the need to hyde 2 by fluorination with DAST, as de-
*Correspondence: Dr. T. Pitterna
Syngenta Crop Protection AG,
CH–4002 Basel
Tel.: +41 61 323 8488
Fax: +41 61 323 8529
E-Mail: thomas.pitterna@syngenta.com

FLUORINE IN THE LIFE SCIENCE INDUSTRY
109
CHIMIA 2004, 58, No. 3
scribed by Middleton [4]. 2-Methyltridec-
2-ene (6) was obtained from aldehyde 3 by
Wittig reaction with instant ylide, as de-
scribed by Schlosser and Schaub [5]. The
dihalovinyl-compounds 7 and 8 were also
prepared from aldehyde 3 by reaction with
triphenylphosphine and perhalomethanes,
as described by Appel and coworkers [6]
and Corey and Fuchs [7] (Scheme 1). The
addition of bromine to the double bond of
1, as described by Suda [8], gave 10, which
upon treatment with DBU eliminated bro-
mide to give 11. Related elimination reac-
tions have been previously described by
Burton et al. and by Brahms and Dailey [9],
but only for substrates activated by a
phenyl- or a carboxyl group, yielding 1,1-
difluorostyrenes or 3,3-difluoroacrylic
acid, respectively.
Fig. 1. First optimisation cycle. Structural elements of lead compound 1.
We have found that difluoroolefin 1
adds iodinemonochloride in a regioselec-
tive manner resulting in 1-chloro-1,1-di-
fluoro-2-iodododecane (12), revealing the
nucleophilic character of C(2) of di-
fluoroolefin 1. Although the fluorine sub-
stituents diminish the reactivity of the dou-
ble bond towards electrophiles, they direct
the addition of electrophiles by stabilisation
of a positive charge at C(1). This is known
from the work of Knunyants and Pol-
ishchuck, and Krespan and Petrov [10].
Treatment of 12 with DBU causes the elim-
ination of chloride (rather than iodide) and
of H-C(2) to form 1,1-difluoro-2-iododo-
dec-1-ene (13). As H-C(2) is the most
acidic proton, this outcome of the reaction
was expected, still it is unprecedented to the
best of our knowledge. The closest analogy
described in the literature is the elimination
of hydrogen chloride from Cl-CF2-CFI-H
reported by Kendall and Lemal [11], where
only one β-elimination process is possible.
In a similar manner, (E)-1-fluorododec-
1-ene (9) added bromine to give 14, and
iodinemonochloride to give 16. The two ad-
dition products displayed different reactivi-
ty upon treatment with DBU, however.
From 1,2-dibromo-1-fluorododecane (14)
Br-C(1) and H-C(2) are eliminated, leading
to (E)-2-bromo-1-fluorododec-1-ene (15).
This is the same direction of elimination as
in the reaction of 10 to 11, and of 12 to 13.
In
contrast,
1-chloro-1-fluoro-2-
(16) eliminated
iodododecane
H-C(1) and I-C(2), resulting in the forma-
tion of (E)-1-chloro-1-fluorododec-1-ene
(17).
Both two-step addition/elimination se-
quences leading from 9 to 15 and from 9 to
17, respectively, appear to be highly selec-
tive (Scheme 2). The observed products are
consistent with trans-addition to the double
bond, followed by an antiperiplanar
arrangement for elimination. To the best of
our knowledge, there is no precedence in
the literature for the elimination reactions
of 14 to 15 and of 16 to 17, which would
Scheme 1. Synthesis of modified pharmacophores

FLUORINE IN THE LIFE SCIENCE INDUSTRY
110
CHIMIA 2004, 58, No. 3
Scheme 2. Stereochemical course of addition/
elimination sequences
Fig. 2. Modified pharmacophores, R = C10H21.
Table 1. Activity of compounds with modified pharmacophore against insect pests
have allowed the prediction of the different
directions of elimination.
compound
H.v.a
D.b.b
A.c.c
N.l.d
N.l. syse
Thus, a representative selection of com-
pounds with different pharmacophores has
been prepared (Fig. 2), and screened
against tobacco budworm (Heliothis
virescens), banded cucumber beetle (Dia-
brotica balteata), pea aphid (Aphis crac-
ciovora) and brown planthopper (Nila-
parvata lugens). Detailed results from the
insecticide screening are listed in Table 1.
In addition to the gem-difluorovinyl frag-
ment of lead 1, only the 1,1-difluoro-2-
iodovinyl group of 13 and the gem-
dichlorovinyl group of 7 caused insecticidal
activity, albeit clearly weaker than 1 (Fig.
2).
1
12.5
12.5
400
400
0.75
7
400
>400
>400
>400
200
>400
>400
>400
>400
>400
400
>12.5
>12.5
>12.5
3
9
>400
>400
100
>400
>400
400
11
13
17
Next, the difluorovinyl pharmacophore
was kept constant, and various bridges were
introduced, in particular linear alkylene
groups, branched and unsaturated groups,
as depicted in Scheme 3. The compounds
were synthesised in one step from the cor-
responding aldehydes, again using the pro-
tocol from Hayashi et al. [3]. Yields were
not optimised.
>400
>400
>400
>12.5
Biological activities in the Table are given as EC80 in ppm. aHeliothis virescens ovolarvicidal activ-
ity, bDiabrotica balteata L2 feeding/contact activity, cAphis cracciovora contact activity, dNilaparva-
ta lugens contact activity, eNilaparvata lugens systemic activity.
In Table 2 the results from insecticide
screening against tobacco budworm (Helio-
this virescens) are shown. It can be ob-

FLUORINE IN THE LIFE SCIENCE INDUSTRY
111
CHIMIA 2004, 58, No. 3
served that only saturated, linear and un-
branched bridges with an even number of
carbon atoms possess insecticidal activity
(i.e. 1, 19 and 22). Compounds with odd
numbers of carbon atoms (18, 20, 21 and
23) and unsaturated or branched bridges
(24 and 25) are devoid of insecticidal activ-
ity. On this basis it can be speculated that
the mode of action should involve the inhi-
bition of oxidative metabolism of fatty
acids to acetyl-CoA. It was confirmed later
by independent work of Ruder [2] and
Baumert et al. [12], that gem-difluorovinyl
carboxylic acids act as inhibitors of β-oxi-
dation.
Subsequently, we turned our attention to
the tail group of lead structure 1. Common
intermediate for all other tail group modifi-
cations was the methyl ester 27 (Scheme 4),
which is accessible from methyl 13-oxo-
undecanoate (26) by a fluoroylide reaction
(Hayashi et al. [3]). Carboxylic acid 28 was
prepared from 27 by saponification. Reduc-
tion of 27 gave alcohol 29, from which
ether 30 was obtained by methylation.
Amine 32 was prepared by reaction of am-
monia with bromide 31, which was ob-
tained from the alcohol 29 by treatment
with triphenylphosphine and bromine.
Table 3 shows results from biological
screening of tail group modified derivatives
against tobacco budworm (Heliothis
virescens), brown planthopper (Nilaparva-
ta lugens) and spider mites (Tetranychus
ssp). At this point we were pleased to ob-
serve, that the ester 27, the acid 28, and the
alcohol 29 not only displayed enhanced in-
secticidal activity as compared to lead com-
pound 1, in addition they showed good po-
tency against spider mites.
Scheme 3. Synthesis of bridge modifications
Table 2. Heliothis virescens ovolarvicidal activity of compounds with modified bridge
compound
H.v.a
>400
18
19
20
1
3
>400
12.5
>400
100
From our first optimisation cycle we
could draw a number of important conclu-
sions. Clearly, the gem-difluoroolefin phar-
macophore is needed for activity. A further
requirement is an even-numbered carbon
chain, which is not branched and contains
no additional unsaturation. This appears
consistent with the assumption that the
compounds interfere with, or are subject to,
β-oxidation (Fig. 3). It is reasonable to as-
sume that all even-numbered gem-difluo-
roolefins could be oxidised in vivo to gem-
difluorovinyl carboxylic acids or deriva-
tives thereof. Perhaps they could be
degraded by β-oxidation to a common ac-
tive metabolite, the structure of which re-
mains to be investigated. This metabolite
could subsequently inhibit one or more of
the enzymes of the β-oxidation cycle. Such
a mode of action could be facilitated if the
active ingredient is offered already with the
required oxidation state at the tail position.
While we have no rigorous proof for this
hypothesis, we have shown that alcohol-,
carboxylic acid- and ester-tails show im-
proved insecticidal activity and, in addition,
activity against spidermites.
21
22
23
24
25
>400
>400
>400
Biological activities in the table are given as EC80 in ppm
aHeliothis virescens ovolarvicidal activity

FLUORINE IN THE LIFE SCIENCE INDUSTRY
112
CHIMIA 2004, 58, No. 3
against brown planthopper (Nilaparvata lu-
gens) and european red mite (Panonychus
ulmi). The acute toxicity in the rat is also
listed in Table 4. The combination of ester-
[13] and amide-tails [14] with a propylene-
bridge gave compounds that showed not
only excellent activity against insects and
spider mites but also moderate toxicity in
the rat. Ester 37, from here on referred to as
CGA 304’111, showed the optimal combi-
nation of properties.
CGA 304’111 (37) as an acaricide is ac-
tive against all important mite species in de-
ciduous fruit, citrus, vegetables, tea and
cotton. It is equally effective against sus-
ceptible and resistant mites. In addition, it is
an insecticide against aphids, scales, thrips
and jassids. CGA 304’111 can be used as
foliar spray, with typical use rates of 10–20
g/hl, or 150–300 g/ha. It shows both feed-
ing and contact activity, as well as ovicidal
effects. CGA 304’111 has a new mode of
action; it interferes with the β-oxidation of
fatty acids. It has a favourable toxicological
profile, and would be classified as moder-
ately hazardous (WHO class II).
Scheme 4. Synthesis of modified tail structures
Table 3. Activity of compounds with modified tail structures against insects and mites
compound
H.v.a
N.l.b
N.l. sysc
T. sspd
>400
The synthesis of CGA 304’111 (37)
(Scheme 5) involved the coupling of 6,6-di-
fluoro-hex-5-enoyl chloride (42) and alco-
hol 43. 4-Hydroxy-phenyl-acetophenone
(38) was alkylated with 1,3-dioxolan-2-one
(170 °C, neat). The resulting 4-(2-hydroxy-
ethoxy)-phenyl-acetophenone (39) was
subjected to Bayer-Villiger oxidation and
subsequently hydrolysed to give 4-(2-hy-
droxy-ethoxy)-phenol (40). Alkylation of
this phenol with 1-bromomethyl-4-trifluo-
romethyl-benzene (41) gave alcohol 43. All
steps in this sequence proceeded with high
to excellent yield. A more challenging en-
deavour was the synthesis of 6,6-difluoro-
hex-5-enoic acid (35), the precursor of 42.
Acid chloride 42 was obtained from acid 35
by reaction with SOCl2. Different synthetic
methods for the preparation of 35 are dis-
cussed in the following section.
1
12.5
400
0.75
27
28
29
30
32
0.75
0.75
0.75
12.5
100
200
3
50
12.5
100
3
12.5
12.5
>100
>100
0.2
0.75
n.t.
>100
100
Synthesis of 6,6-difluorohex-5-
enoic acid (35)
Biological activities in the table are given as EC80 in ppm. aHeliothis virescens ovolarvicidal activ-
ity, bNilaparvata lugens contact activity, cNilaparvata lugens systemic activity, dTetranychus ssp
activity.
Our first access to this compound relied
on difluoroylide chemistry, as shown in
Scheme 6. This route worked quite effi-
ciently on a few gram scale. However, for
the preparation of larger amounts, as need-
ed for field trials, it was not useful. 1,5-Pen-
tanediol (44) was treated with benzoyl chlo-
ride to give the mono-benzoate 45. Oxida-
tion to the aldehyde 46, followed by
treatment with CF2Br2 and P[N(CH3)2]3 ac-
cording to the procedure of Hayashi et al.
[3] gave the gem-difluorovinyl compound
47 in useful yield. However, upon scale-up
the yield of this step decreased drastically.
Hydrolysis of ester 47 gave the alcohol 48,
which upon Jones oxidation gave acid 35.
2nd Optimisation Cycle
28, such as esters, amides, and hydrazides.
We recognised that the optimal carbon chain
Starting point for further optimisation length of the bridge would not necessarily
was the carboxylic acid 28. Beside the en- be the same for all different acid derivatives.
couraging improvement of pesticidal activi- Therefore, we re-optimised the bridge for a
ty, this compound displayed some unwanted number of such derivatives. This optimisa-
properties, such as too high volatility, stench tion strategy is shown in Fig. 4.
and high acute toxicity in the rat. In order to
In Table 4 representative compounds
improve these properties, we investigated from this second optimisation cycle are
various different types of derivatives of acid shown, as well as their biological activity

FLUORINE IN THE LIFE SCIENCE INDUSTRY
113
CHIMIA 2004, 58, No. 3
Fig. 3. Mode of action hypothesis. gem-Difluo-
rovinyl derivatives inhibit the fatty acid metabo-
lism in insect mitochondria. gem-Difluorovinyl
acids may enter the β-oxidation cycle as CoA de-
rivatives. This could lead to a common degrada-
tion product from different even numbered gem-
difluorovinyl analogues. Such a product may in-
hibit one or more of the enzymes that are
involved in β-oxidation.
Fig. 4. Second optimisation cycle. Combined
variations of bridge and tail part.
Scheme 5. Synthesis of CGA 304’111 (37)

FLUORINE IN THE LIFE SCIENCE INDUSTRY
114
CHIMIA 2004, 58, No. 3
Table 4. Biological activity of selected compounds from the second optimisation cycle
We reasoned, that elimination of fluo-
ride from 6,6,6-trifluorohexanoic acid (51)
could potentially be an extremely short
route to our target. Indeed, with potassium
tert-butoxide as a base this reaction pro-
ceeded in good yield (Scheme 7). 6,6,6-Tri-
fluoro-hexanoic acid (51) was obtained
from hexanedioic acid (49) after treatment
with SF4 in HF. Similarly, the ethyl ester 52
was obtained from monoester 50. These
fluorination reactions were developed in
collaboration with the former Ciba Geigy
catalysis group (now part of Solvias AG).
However, despite valiant efforts to opti-
mise, the yields of 6,6,6-trifluoro-hexanoic
acid (51) or of its ester 52 never exceeded
50%.
c
compound
n
9
9
9
3
3
3
N.l. sysa P.u.b
ALD50
high
28
33
34
35
36
37
3
3
50
50
moderate
moderate
high
12.5
0.75
0.75
3
50
Therefore, we turned to the use of fluo-
rinated building blocks. Radical addition of
CF2Cl-CCl3 to but-3-enoic acid (53)
(Scheme 8) proceeded in useful yield. The
product 54 was esterified with methanol.
From the resulting ester 55, all chlorine
atoms could be removed by an elimination
– hydrogenation – elimination sequence.
The first elimination step with triethyl-
amine gave 56 in good yield. In the follow-
ing step, both double bonds in 56 were hy-
drogenated and the chlorine atom Cl-C(5)
removed to give 57. Finally, a second elim-
ination with DBU as a base, followed by
100
12.5 moderate
to high
12.5 moderate
Biological activities in the table are given as EC80 in ppm. aNilaparvata lugens systemic activity,
bPanonychus ulmi activity, cacute oral toxicity in the rat.
Scheme 6. Synthesis of 6,6-difluorohex-5-enoic
acid (35) by a difluoro ylide approach
Scheme 7. Synthesis of 6,6-difluorohex-5-enoic
acid (35) by deoxifluorination

FLUORINE IN THE LIFE SCIENCE INDUSTRY
115
CHIMIA 2004, 58, No. 3
Scheme 8. Synthesis of 6,6-difluorohex-5-enoic
acid (35) from CF2Cl-CCl3 and but-3-enoic acid
(53).
Scheme 9. Synthesis of 6,6-difluorohex-5-enoic acid (35) from CBrClF2 and Scheme 10. Synthesis of 6,6-difluorohex-5-enoic acid (35) from CBrClF2
pent-4-enoic acid ethyl ester (58) and 2-allyl-malonic acid diethyl ester (62)
efficiently. They involved removal of
bromine by catalytic hydrogenation to give
64, elimination of chloride from 64 with
DBU as a base, hydrolysis of both ester
groups of 65 and decarboxylation of diacid
66 at 140 °C to give 35.
saponification of the ester group gave acid synthetic route is most efficient, and we
35. Although the yield of the hydrogenation used it successfully for the synthesis of sev-
step was low (26%), we did not further op- eral hundred grams of 6,6-difluorohex-5-
timise it, because the following route turned enoic acid (35).
out more promising.
In a variation of the above, we also in-
Radical addition of CBrClF2 to pent-4- vestigated the radical addition of CBrClF2
enoic acid ethyl ester (58) proceeded very to 2-allyl-malonic acid diethyl ester (62)
efficiently (Scheme 9). Bromide was re- (Scheme 10), which is a potentially cheap-
moved from the product 59 by catalytic hy- er starting material as compared with pent-
drogenation to give 60, chloride eliminated 4-enoic acid ethyl ester (58). However, the
by treatment with base, and the ester 61 hy- yield of the addition product 63 remained
drolysed to give 35.All four steps of this se- unsatisfactory, 42% in the best case. In con-
quence proceeded with good yield. This trast, the following steps proceeded quite
In summary, we have identified five
synthetic routes for 6,6-difluorohex-5-
enoic acid (35). The best synthesis follows
a fluorinated building block approach, and
it involves the radical addition of CBrClF2
to pent-4-enoic acid ethyl ester. This route
was used for scale-up to a several hundred

FLUORINE IN THE LIFE SCIENCE INDUSTRY
116
CHIMIA 2004, 58, No. 3
[12] D. Baumert, H. Franke, D. Giles, G.
gram scale. It allowed us to provide suffi-
cient material for field trials.
Hoemberger, A. Koehn, N. Skuballa, P.
Wegner, Special Publication – Royal
Society of Chemistry, 1994, 147 (Ad-
vances in the Chemistry of Insect Control
III), 171.
Conclusion
[13] P. Maienfisch, T. Pitterna, M. Böger, Eur.
Pat. Appl., EP 577555, 1994.
The insecticidal lead 1,1-difluoro-1-do-
decene (1) was optimised. In a first optimi-
sation cycle, broad activity against insect
pests and mites was discovered. The specif-
ic role of fluorine in this class of insecti-
cides is its contribution to the essential gem-
difluorovinyl pharmacophor, as gem-diflu-
orovinyl derivatives interfere with the
β-oxidation of fatty acids in insect mito-
chondria.
[14] P. Maienfisch, M. Böger, T. Pitterna, H.
Szczepansky, Eur. Pat. Appl., EP 661289,
1995.
In a second optimisation cycle, CGA
304’111 (37) was discovered. It showed ex-
cellent performance in field trials against a
wide range of pests. In addition, 37 also
showed a more favourable toxicological
profile than earlier derivatives.
A practical synthesis was identified,
which allowed the preparation of larger
samples (several hundred gram scale).
Acknowledgement
We wish to thank Alfred Rindlisbacher
(Syngenta AG) for the assessment of biological
activity and Franz Ruder (former Ciba-Geigy
AG) for biochemical studies. For synthetic col-
laborations we thank Tibor Gögh and Marcela
Göghova from Synkola, Hermann Rempfler,
Henry Szczepansky and Rudolf Waditschatka
from Syngenta AG, and Heinz Steiner from
Solvias AG.
Received: December 19, 2003
[1] K. Naumann, B. Becker, W. Behrenz, B.
Homeyer, W. Stendel (Bayer A.-G.), Ger.
Offen., DE 3618115, 1987.
[2] F. Ruder (Ciba-Geigy AG), personal com-
munication.
[3] S. Hayashi, T. Nakai, N. Ishikawa, D.J.
Burton, D.G. Naae, H.S. Kesling, Chem.
Lett. 1979, 983.
[4] W.J. Middleton, J. Org. Chem. 1975, 40,
(5), 574.
[5] M. Schlosser, B. Schaub, Chimia 1982, 36
(10), 396.
[6] a) R. Appel, Angew. Chem. 1975, 87, 863;
b) R. Appel, P. Knoll, W. Michel, W. Mor-
bach, H.D. Wihler, H. Veltmann, Chem.
Ber. 1976, 109, 58.
[7] E.J. Corey, P.L. Fuchs, Tetrahedron Let-
ters 1972, 13 (36), 3769.
[8] M. Suda, Tetrahedron Letters 1980, 21
(26), 2555.
[9] a) J.C. Brahms, W.P. Dailey, J. Org. Chem.
1991, 56 (2), 900; b) D.J. Burton, A.L. An-
derson, R. Takei, H.F. Koch, T.L. Shih, J.
Fluorine Chem. 1980, 16 (3), 229.
[10] a) I.L. Knunyants, V.R. Polishchuck, Usp.
Khim. 1975, 44 696; b) V.R. Polishchuck,
E.I. Mysov, I.V. Stankevich, A.L. Chis-
takov, K.A. Potechin, Y.T. Struchkov, J.
Fluorine Chem. 1993, 65, 233; c) C.G.
Krespan, V.A. Petrov, Chem. Rev. 1996,
96, 3269.
[11] J.T. Kendall, D.M. Lemal, J. Fluorine
Chem. 1997, 83 (2), 171.