An efficient synthesis of (Z)-γ-fluoroallylphosphonates using a base-promoted deconjugation of (E)-γ-fluorovinylphosphonates, and its utility as fluoroolefin-containing building block

Article abstract of DOI:10.1016/S0022-1139(99)00244-4

A three-step synthesis of γ-fluoroallylphosphonates starting with α,β-unsaturated aldehydes is described. Treatment with diethyl phosphite in the presence of KF gives α-hydroxyallyl phosphonate in excellent yield; DAST deoxofluorination produces the corre

Full text of DOI:10.1016/S0022-1139(99)00244-4

Journal of Fluorine Chemistry 102 (2000) 189±197
An ef®cient synthesis of (Z)-g-¯uoroallylphosphonates using a
base-promoted deconjugation of (E)-g-¯uorovinylphosphonates,
and its utility as ¯uoroole®n-containing building block
Gerald B. Hammond*, Daniel J. deMendonca
Department of Chemistry and Biochemistry, University of Massachusetts, 285 Old Westport Road, Dartmouth, MA 02747-2300, USA
Received 28 June 1999; received in revised form 19 July 1999; accepted 12 August 1999
Dedicated to Professor Paul Tarrant on the occasion of his 85th birthday
Abstract
A three-step synthesis of g-¯uoroallylphosphonates starting with a,b-unsaturated aldehydes is described. Treatment with diethyl
phosphite in the presence of KF gives a-hydroxyallyl phosphonate in excellent yield; DAST deoxo¯uorination produces the corresponding
g-¯uorovinylphosphonate through a SN20 mechanism, and ®nally, a base-promoted double bond migration leads to the desired
g-¯uoroallylphosphonate. g-Fluoroallylphosphonate is a useful building block in the synthesis of ¯uoroole®ns. An exploratory study
yielded excellent yields of (Z)-diethyl 1-benzyl-3-¯uoro-2-butenylphosphonate and (Z)-diethyl 1,3-di¯uoro-2-hexenylphosphonate.
Treatment of (E)-diethyl 3-¯uoro-2-hexenylphosphonate with LiN(TMS)2 and benzaldehyde in THF led to the preferential formation of
syn-(Z)-diethyl 3-¯uoro-1-(hydroxybenzyl)-2-butenylphosphonate. # 2000 Elsevier Science S.A. All rights reserved.
Keywords: Fluoroallylphosphonate; Fluorovinylphosphonate; Hydroxyallylphosphonate; HWE ole®nation
1. Introduction
non¯uorinated counterpart (4). Despite its importance as a
synthetic intermediate, 5 has received scarce attention in the
literature [5±7]. For the synthesis of 5 we adapted Kiddle
and Babbler's methodology [20] because the required g-
¯uorovinylphosphonate (3) can be easily prepared by a two-
step sequence involving Pudovik reaction [24] of a,b-unsa-
turated aldehydes with diethyl phosphite [HP(O)(OEt)2]
followed by DAST deoxo¯uorination of allylic a-hydroxy-
phosphonate (1) (Eq. (1)).
The selective replacement of a vinyl hydrogen by ¯uorine
enhances the biological activity of ole®n-containing mole-
cules such as pheromones [1±4], and retinoids [5±8]. Of
paramount importance in the synthesis of the latter is the
regio- and stereocontrol of the ole®nic bond. In this context,
allylic phosphonate (4) provides a powerful building block
for the formation of conjugated dienes Ð via a modi®ed
Wadsworth±Horner ole®nation of the phosphoryl-stabilized
a-carbanion [9±11]; and ole®ns Ð after alkylation with
alkyl halides followed by LiAlH4 reduction and simulta-
neous elimination of the phosphonate group [12±14]. Lit-
erature preparations of 4 are numerous and include the
rearrangement of the corresponding phosphites [15±17],
metal-coordinated allyl addition to P(OEt)3 [18], phospho-
nium salts [19], and base-catalyzed isomerization of vinyl-
phosphonates [20] (2 ! 4, Eq. (1)).
(1)
2. Results and discussion
During our studies on ¯uorinated phosphonate synthons
[21±23], it occurred to us that g-¯uoroallylphosphonate (5)
might be capable of playing a similar role to that of its
In our laboratory, a-hydroxyallylphosphonate (1) was
obtained in higher yields than previously reported [25]
(Table 1) when the reaction was carried out neat in the
presence of KFÁ2H2O and a slight excess of HP(O)(OEt)2
[26]. a-Hydroxyallyl phosphonates were very stable except
* Corresponding author. Tel.: ‡1-508-999-8865; fax: ‡1-508-999-9167.
E-mail address: ghammond@umassd.edu (G.B. Hammond).
0022-1139/00/$ ± see front matter # 2000 Elsevier Science S.A. All rights reserved.
PII: S 0 0 2 2 - 1 1 3 9 ( 9 9 ) 0 0 2 4 4 - 4

190
G.B. Hammond, D.J. deMendonca / Journal of Fluorine Chemistry 102 (2000) 189±197
for 1f (entry 6) who underwent gradual decomposition at
room temperature. The deoxo¯uorination of a-hydroxyal-
lylphosphonate (1) with DAST was ®rst studied by Black-
burn using only acyclic substrates [25]. We extended this
¯uorination method to substituted, conjugated and cyclic a-
hydroxyallyl phosphonates, and in all cases but one (entry
6)1, good to excellent yields of (E)-g-¯uorovinylphospho-
nates 3 were obtained (Table 1). The stereochemical assign-
ment was based on vicinal 13C±31P coupling constants [28]
and Blackburn's NMR data [25]. DAST ¯uorination of 2,4-
hexadienyl hydroxyphosphonate (1d) occurred exclusively
on the E-carbon, ruling out a SNi0 mechanism in this con-
jugated system (entry 4). The stereochemistry of the vinyl-
phosphonate moiety in the ®ve- and six-membered rings
exocyclic ole®ns (3g,h) showed a remarkable preference for
the E-isomer, as determined by NOE studies (entries 7 and
8). A competing dehydro¯uorination produced ®ve- and six-
membered ring conjugated dienylphosphonate byproducts
(5% and 24%).
With an effective synthesis of 3 in hand we proceeded to
investigate the base-catalyzed double bond migration (i.e.,
3 ! 5). Presumably, the ®rst step in this process is the
deprotonation of the g-carbon, leading to a resonance
stabilized system in which the a-anion of 5 should be the
major contributor and therefore the predominant isomer
after workup. Semiempirical calculations (AM1) predicted
acyclic compounds 5a,b,c to be more stable than 3a,c,e.
Similar calculations also predicted that the difference in
stabilization energy between cyclic ¯uorophosphonates 3g
and 5d, and between 3h and 5e, was negligible. Experi-
mental evidence on the prototropic equilibria of diethyl
alkenylphosphonates had shown that the PO3Et2 group
provided only a very weak stabilizing effect of on the
adjacent carbon±carbon double bond [29]. Preliminary trials
to synthesize 5 from 3 using Kiddle and Babbler's condi-
tions (KOt-Bu in DMSO at room temperature) [20], met
with partial success, yields ranging between 30% and 50%.
We speculated that the presence of DMSO in the workup
might have contributed to the low yields because of the
partial solubility of the phosphonate product in the DMSO-
containing aqueous layer. We succeeded in promoting the
double bond migration from 3 to (Z)-5, in excellent yield
and with great stereospeci®city, using strongly basic but
experimentally more friendly conditions (KNTMS2, 18-
crown-6 and THF rather than DMSO) (entries 1, 3, and
5). Analysis of the vinylic 3JFH coupling constants permitted
an unequivocal assignment of the sterochemistry of the
double bond. In the case of ®ve- and six-membered rings
(entries 7 and 8), the regiochemical outcome of the migra-
tion protocol was less speci®c. Whereas 3h produced pre-
dominantly the vinyl¯uoro isomer (5e) (vinyl:allyl 90:10),
3g produced exclusively the allyl¯uoro isomer (5d) regard-
less of whether kinetic or thermodynamic conditions were
applied.
Our ef®cient synthesis of acyclic g-¯uoroallyl phospho-
nate (5) unmasked a new active methylene site on the
a-carbon capable of producing a phosphoryl-stabilized
carbanion and the possibility of further functionalization
at the a-position. With this purpose in mind, we explored the
alkylation and electrophilic ¯uorination of 5. It is well
documented in the literature that lithio derivatives of
g-substituted allylphosphonate (4) undergo alkylation at
the a-carbon when alkyl halides are utilized [12±14,30]2.
Thus, it came as no surprise that the alkylation of 5a and 5b
(LiNTMS2 in THF at 788C) with benzyl bromide pro-
duced a-substituted-g-¯uoroallylphosphonate (6a) in excel-
lent yield (Eq. (2)). Analogously, treatment of the a-
carbanion of 5b with N-¯uorobenzenesulfonimide (NFSI),
an electrophilic ¯uorinating agent, yielded 1,3-di¯uorobu-
tenylphosphonate (6b) in 66% yield after chromatography.
Kondo and coworkers [12±14] have used a-substituted
allylphosphonates as a template for the synthesis of
trans-ole®ns by means of a LiAlH4-induced reductive clea-
vage of the phosphonate group. To probe whether our a-
substituted-g-¯uoroallylphosphonate (6) would behave in a
similar fashion, 6a was treated with LiAlH4 (THF, 788C to
room temperature) and the reaction was monitored using 31P
and 19F NMR spectroscopy. After 2 h, disappearance of the
phosphonate signal was accompanied by the appearance of
two new allylic ¯uorine signals at d 165 (85%) and d
168 ppm (15%) corresponding to Z/E isomers of 7 generated
by SN20 hydride displacement of phosphorus and concomi-
tant double bond transposition. Unfortunately, attempted
puri®cation of 7 by chromatography or vacuum distillation
led to loss of HF.
(2)
It can be argued that if the a-carbanion of 5 was trapped
with an electrophilic reagent (e.g., benzaldehyde), and if the
resulting pair of diastereomeric b-hydroxyphosphonates
(e.g., 8c and 8d) could be separated and the phosphate
group eliminated, then it would be possible to synthesize (E)
and (Z)-¯uoroalkenes almost at will (Scheme 1). Toward
this goal, we turned our attention to Warren's pioneering
1 Reaction mixture gave 3f [31P NMR: d 17.8 (46%)] and a major
byproduct tentatively identified as 9 [31P NMR: d 25.3 (54%)]. The
structure of 9 was assigned based on 1H NMR and decoupling
experiments. Compound 9 could have been produced by migration of
the methoxy group in 1f, triggered by the g-allylic cationic
intermediate shown below:
2 Modro has reported that diethyl prop-2-enylphosphonate (allylic
phosphonate), when treated with n-BuLi gives a lithiated derivative that
adds to aromatic aldehydes via the a- or g-carbon atom of the allylic
system. See [30].

G.B. Hammond, D.J. deMendonca / Journal of Fluorine Chemistry 102 (2000) 189±197
191
Table 1
Synthesis of g-fluoroallylphosphonate 5
Entry 1(%)a
3(%)a
5(%)a
d
±
b
c
d
±
c
d
±
a Isolated yields.
b See [35] for preparation.
c NMR yields.
d This reaction was not attempted.
work on ole®n stereocontrol using the Horner±Wittig reac-
tion [31]. By performing ®rst the addition of a lithiated
alkyldiphenylphosphine oxide to an aldehyde, separating
the two diastereomeric b-hydroxyphosphine oxide inter-
mediates and then proceeding with the extrusion of diphe-
nylphosphinic acid, Warren was able to synthesize
exclusively (E) or (Z)-alkenes [32]. Hitherto, this level of
stereocontrol has rarely been achieved using dialkylpho-
sphonates (i.e., HWE ole®nation) [27] mainly because the
oxyanion intermediate decomposes via a transient oxapho-
sphetane to yield usually E-enriched ole®ns [33]. To our
knowledge, the only literature isolation of b-hydroxyallyl-
phosphonates was reported by Collignon and coworkers [9],
who obtained a slight predominance of the syn adduct [e.g.,
syn-8a (58%), anti-8b (42%)] after acidic hydrolysis at
708C. When we generated the carbanion of 5b using
LiNTMS2 (two equivalents) in THF, added benzaldehyde
(1.5 equivalents) at 808C, and warmed up the mixture to
room temperature, the reaction not only stopped short of
oxaphosphetane elimination but, more importantly, it

192
G.B. Hammond, D.J. deMendonca / Journal of Fluorine Chemistry 102 (2000) 189±197
Scheme 1. HWE addition of benzaldehyde to 5.
favored the preferential formation of the anti adduct [anti-
8d (62%), syn-8c (13%)]. The only major contaminant was
unreacted starting material (25%). The high degree of
selectivity observed in this reaction is comparable to War-
ren's Horner±Wittig reaction using the diphenylphosphoryl
group [31]. Diastereomers anti-8d and syn-8c were identi-
®ed according to 3JHaHb, 2JHbP, 3JHaHb and 3JHbHc coupling
ing to syn isomer (8c), would have entailed an unfavorable
gauche interaction between the Li±O±C moiety and the
phosphoryl group. To determine whether ¯uorine indeed
plays a determining role in the stereoselective outcome of
this reaction we conducted a similar reaction using a non-
¯uorinated substrate (e.g., 4 R1 ˆ H, R ˆ n-Pr) and found
approximately equal amounts of syn and anti diastereomers
in the crude mixture, very much in agreement with Col-
lignon's results. An investigation of the chiral impact of
using g-¯uoroallyl phosphonate (5) in asymmetric HWE
reactions is in progress.
1
constant measurements in the H NMR spectrum which
were compared with existing data (Table 2) [9,34]. The
diastereomeric ratios were determined by 31P and/or 19F
NMR spectroscopy in the crude mixture (Table 3).
To account for the diastereoselective formation of 8d we
hypothesized that ¯uorine exerts an in¯uence on the direc-
tion of approach of the carbonyl group through a C±F---Li±
O±C coordination that favors the anti isomer on steric
grounds (Scheme 1, X ˆ F). The alternative pathway, lead-
In summary, we have reported a new general synthesis of
g-¯uoroallylic phosphonates (5) from commercially avail-
able aldehydes, using a base-promoted deconjugation of
g-¯uorovinyl phosphonates. We have also demonstrated the
preference of DAST for SN20 deoxo¯uorination using cyclic
Table 2
1H NMR data for anti-8 (CDCl3)a
Ha (d)
Hb (d)
Hc (d)
3JaP (Hz)
2JbP (Hz)
3Jab (Hz)
3Jbc (Hz)
JcF (Hz)
b
anti-8b
anti-8d
5.3, dd
5.28, dd
3.0, ddd
3.31, ddd
5.4, m
4.76, ddd
±
8.9
21.8c
22.0
3.1d
2.2
9.8
10.8
±
35.9
a Values for 8a,b were obtained from [9].
b Not reported.
c
2
In the case of syn-8a, JbP ˆ 17.5 Hz.
d
3
For syn-8a, Jab 8.8 Hz.

G.B. Hammond, D.J. deMendonca / Journal of Fluorine Chemistry 102 (2000) 189±197
193
Table 3
31P and 19F NMR data for 8 (CDCl3)a
30 min at room temperature. To the slurry was added
dropwise trans-crotonaldehyde (0.998 g, 10.4 mmol). The
resulting mixture was allowed to stir overnight before
adding ether (75 ml) and water (25 ml). The aqueous phase
was extracted with ether (2 Â 25 ml), dried (MgSO4) and
concentrated to provide 1a as a viscous oil (1.74 g,
24.9 mmol) which was puri®ed using column chromatogra-
phy (silica, EtOAc:hexane 1:1) yielding 1a (4.76 g, 95.3%).
Although this compound has been reported previously [35],
31P (d)
19F (d)
4JPF (Hz)
syn-8a
syn-8c
anti 8b
anti 8d
27.0
28.8
26.8
28.5
±
±
103.9
±
8.5
±
103.1
13.0
a Values for 8a,b were obtained from [9].
1
only physical constants were supplied. H NMR: d 5.96±
and acyclic hydroxyphosphonates and have provided a
glimpse of the building block potential of 5 in the regios-
peci®c synthesis of acyclic a-substituted-g-¯uoroallylpho-
sphonates.
5.82 (m, 1H), 5.68±5.57 (m, 1H), 4.40 (ddt, 2JPH ˆ 10.6 Hz,
4
3JHH ˆ 6.9 Hz, JHH ˆ 1.2 Hz, 1H), 4.17 (quintet, J ˆ
3
7.2 Hz, 4H), 3.55 (br. s, 1H), 1.76 (dt, JHH ˆ 5.1 Hz,
4;5
3
4
J
HH
ˆ 1.3 Hz, 3H), 1.32 (td, JHH ˆ 7.1 Hz, JHP ˆ
1.2 Hz, 6H); 31P NMR: d 23.2 (s); 13C NMR: d 129.9 (d,
2JCP ˆ 14.0 Hz), 125.5 (d, JCP ˆ 3.47 Hz), 69.25 (d, JCP
3
1
2
3
3. Experimental
ˆ 161 Hz), 62.9 (t, JCP ˆ 7 Hz), 17.9, 16.4 (d, JCP
6 Hz); m/e 179, 151, 138, 111, 99, 82, 53, 29.
ˆ
All moisture sensitive reactions were done using ¯ame-
dried glassware ¯ushed with argon, magnetic stirring, and
dry, freshly distilled solvents. THF was distilled from Na/
benzophenone. Toluene and methylene chloride (CH2Cl2)
were distilled from calcium hydride. Other solvents were
HPLC grade and were used without puri®cation. Diethyl
phosphite [HP(O)(OEt)2], benzyl bromide, benzaldehyde,
and decyl iodide were distilled prior to use. All other
reagents were used as received. All reactions were mon-
itored using one of the following techniques: TLC, GC±MS,
31P and/or 19F NMR. Preparative TLC was performed using
E.Merck Silica Gel 60 F254. Analytical TLC was performed
using Macherey±Nagel Polygram Sil G/UV254 precoated
plates. Flash chromatography was performed using silica
gel 230±400 mesh, 40±63 m (Lagand). Dry-column chro-
matography was performed using Florisil1, 60±100 mesh.
Solid phase extraction was performed using Extract±
3.2. Diethyl 1-hydroxy-2-hexenylphosphonate (1c)
trans-2-Hexenal (2.09 g, 21.3 mmol), diethyl phosphite
(2.94 g, 21.3 mmol) and KFÁ2H2O (5.08 g, 53 mmol) were
stirred overnight to yield 1c (3.8 g, 76%) after column
1
chromatography (silica, hexane:EtOAc 1:1). H NMR: d
5.92±5.81 (m, 1H), 5.64±5.54 (m, 1H), 4.41 (m, 1H), 4.17
(quintet, J ˆ 7.2 Hz, 4H), 3.84 (t, J ˆ 6.3 Hz, 1H), 2.11±
2.02 (m, 2H), 1.42 (sextet, J ˆ 7.4 Hz, 2H), 1.33 (td,
4
3
3JHH ˆ 7.0 Hz, JHP ˆ 1.1 Hz, 6H), 0.91 (t, JHH ˆ 7.4,
3H); 31P NMR: d 23.1(s); C NMR: d 134.9 (d, JCP
ˆ
ˆ
13
2
3
1
13.0 Hz), 124.4 (d, JCP ˆ 3.7 Hz), 69.30 (d, JCP
161 Hz), 62.9 (t, 2JCP ˆ 8 Hz), 34.4, 22.05 (d, 4JCP ˆ 3 Hz),
‡
3
3 Hz), 16.4 (d, JCP ˆ 5 Hz), 13.5; m/e 207 (M
29, 9),
151 (7), 138 (40), 111 (100), 83 (50), 82 (80). Anal.:
Calcd. for C10H21O4P: C, 50.84; H, 8.96. Found: C,
50.82; H, 9.08.
CleanTM columns (silica gel, 60 A). Melting points are
Ê
uncorrected. IR spectra were recorded on neat liquids.
1H, 19F and 31P NMR spectra were recorded in CDCl3 at
300, 282, and 121 MHz respectively. 19F NMR spectra
3.3. (E, E)-Diethyl 1-hydroxy-2,4-hexadienylphosphonate
(1d)
1
are referenced against external CFCl3, H NMR spectra
against internal (CH3)4Si, and 31P NMR spectra against
85% H3PO4. 31P and 19F NMR spectra were broadband
decoupled from hydrogen nuclei. J values are given in
Hertz (Hz). Low resolution EI mass spectra were recorded
with an ionization voltage of 70 eV; peaks are reported
as m/e (% intensity relative to base peak). Elemental
analyses were performed by Atlantic Microlab, Norcross,
GA. High resolution MS was performed at the Midwest
Center for Mass Spectrometry, University of Nebraska,
Lincoln.
A mixture of KFÁ2H2O (3.29 g, 34.9 mmol) and diethyl
phosphite (1.44 g, 10.4 mmol) was stirred for 30 min. To the
slurry was added dropwise trans,trans-2,4-hexadienal
(0.99 g, 10.4 mmol). The resulting mixture was allowed
to stir overnight and worked up as above to yield 1d
‡
(2.37 g, 97%) 31P NMR: d 22.7 (s); m/e 234 (M , 2),
205 (10), 177 (5), 138 (12), 111 (75), 82 (90), 41 (100);
1
IR (®lm): 3280, 3000±2860, 1639 cm
.
3.4. Diethyl 1-hydroxy-3-phenyl-2-propenylphosphonate
(1e)
3.1. Diethyl 1-hydroxy-2-butenylphosphonate (1a).
General method
trans-Cinnamaldehyde (3.96 g, 30.0 mmol), KFÁ2H2O
(14.14 g, 150.2 mmol) and diethyl phosphite (4.17 g,
30.2 mmol) were stirred overnight under standard reaction
and workup conditions to produce a solid which was
A mixture of KFÁ2H2O (4.92 g, 52.26 mmol) and diethyl
phosphite (3.32 g, 24.04 mmol) was magnetically stirred for

194
G.B. Hammond, D.J. deMendonca / Journal of Fluorine Chemistry 102 (2000) 189±197
1
recrystallized from cyclohexane to yield 1e (6.09 g, 75%).
Although this compound has been reported previously 25 no
comprehensive NMR data was supplied. H NMR:d 7.41±
0.29 (hexane:EtOAc 1:3) H NMR: d 5.87 (bs, 1H), 4.72
2
2
(d, JHH ˆ 4.1 Hz, 2H), 4.32 (dd, JHP ˆ 11.2 Hz,
3JHP ˆ 4.5 Hz, 1H), 4.22±4.11 (m, 4H), 3.07 (pseudo quar-
tet, 3JHH ˆ 6.5 Hz, 1H), 2.4±1.8 (m, 5H), 1.74 (s, 3H), 1.55±
1.40 (m, 1H), 1.33 (td, 3JHH ˆ 7.0 Hz, 4JHP ˆ 1.5 Hz, 6H);
31P NMR: d 23.1 (s); 13C NMR: two diastereomers: 149.74
[139,7], 133.62 (d, 3JCP ˆ 4 Hz) [133.24 (d, 3JCP ˆ 4 Hz)],
125.94 (d, 2JCP ˆ 12 Hz) [125.35 (d, 2JCP ˆ 12 Hz)], 108.9,
1
3
4
7.25 (m, 5H), 6.78 (ddd, JHH ˆ 16.0 Hz, JHP ˆ 4.80 Hz,
4JHH ˆ 1.48 Hz, 1H, H±C±Ph), 6.32 (ddd, 3JHH ˆ 16.0 Hz,
3JHH ˆ 6.1 Hz, JHP ˆ 0.6 Hz, 1H, H±C = C), 4.67 (ddd,
3
2JHP ˆ 12.9 Hz, JHH ˆ 6.1 Hz, JHH ˆ 1.6 Hz, 1H), 4.19
3
4
3
(quintet, J ˆ 7.2 Hz, 4H), 1.33 (td, JHH ˆ 7.1 Hz,
4JHP ˆ 2.0 Hz, 6H); 31P NMR: d 22.2 (s); 13C NMR: d
72.6 (d, JCP ˆ 158 Hz) [72.1 (d, JCP ˆ 158 Hz)], 63.0
1
1
4
2
3
136.4 (d, JCP ˆ 4 Hz), 132.2 (d, JCP ˆ 14 Hz), 128.5,
(m), 41.06 [40.96], 30.85 (d, JCP ˆ 2 Hz) [30.76 (d,
5
3
3
127.8, 126.6 (d, JCP ˆ 2 Hz), 123.9 (d, JCP ˆ 5 Hz),
3JCP ˆ 2 Hz)], 27.7 [27.6], 26.31 (d, JCP ˆ 2 Hz) [26.0
1
2
3
69.40 (d, JCP ˆ 162 Hz), 63.2 (d, JCP ˆ 7 Hz), 63.1 (d,
(d, JCP ˆ 2 Hz)], 20.94, 16.7 (d, JCP ˆ 5 Hz); IR (®lm) n
2JCP ˆ 7 Hz), 16.4 (d, JCP ˆ 5 Hz), 13.5.
3280, 3060, 1639 cm 1. Anal.: Calcd. for C14H25O4P: C,
58.32; H, 8.74. Found: C, 58.04; H, 8.67.
3
3.5. Diethyl 1-hydroxy-4,4-dimethoxy-2-
butenylphosphonate (1f)
3.8. (E)-Diethyl 3-fluoro-1-butenylphosphonate (3a).
General method
Under standard reaction and workup conditions, fumar-
aldehyde monodimethylacetal [36] (0.501 g, 3.84 mmol).
KFÁ2H2O (0.964 g, 8.5 mmol), and diethyl phosphite
(0.447 g, 3.21 mmol) produced 1f (0.793 g, 85%) after
chromatographic puri®cation (silica, gradient hexane/
To a solution of 1a (2.99 g, 14.4 mmol) in CH2Cl2 (75 ml)
at 808C was added DAST (2.4 ml, 17.3 mmol, 1.2 eq)
dropwise via syringe. The resulting solution was allowed
to warm slowly overnight before quenching with saturated
NaHCO3 (100 ml). The layers were separated and the
aqueous was extracted with CH2Cl2 (3 Â 25 ml). The
organic layer was washed with H2O (40 ml). The combined
organic layers were washed with brine (25 ml), dried
(MgSO4) and concentrated to give a viscous amber oil
(2.71 g) containing 3a (d 178.2 ppm, 98%) and its a-¯uoro-
phosphonate isomer (d 196.9 ppm, 2%). Kugelrohr distilla-
tion (1088C/0.07 mmHg) afforded pure 3a (2.30 g, 76%).
[Note: Although this compound has been reported pre-
1
EtOAc). H NMR: 6.07±5.83 (m, 2H), 4.83 (br. s, 1H),
2
3
4.55 (dd, JHP ˆ 16 Hz, JHH ˆ 5 Hz, 1H), 4.12 (quintet,
3
J ˆ 7.2 Hz, 4H), 3.3 (s, 6H), 1.32 (td, JHH ˆ 7.0 Hz,
4JHP ˆ 1.7 Hz, 6H); 31P NMR: d 22.1 (s). This compound
decomposed gradually upon standing and could not be sent
out for elemental analysis.
3.6. 1-(Diethoxyphosphono-a-
hydroxymethyl)cyclopentene (1g)
1
viously by Blackburn and coworkers [25] our H NMR
analysis differs from theirs on the chemical shift of the
Cyclopentene-1-carboxaldehyde [37] (0.77 g, 8.0 mmol)
was stirred with diethyl phosphite (1.27 g, 9.2 mmol) and
KFÁ2H2O (1.80 g, 19 mmol) for 2 h under standard reaction
and workup conditions. Flash chromatography (silica,
CH2Cl2:EtOAc 1:1) of the residual oil and collection of
the fractions with Rf 0.32 yielded a clear oil (1.30 g, 70%).
1
vinylic hydrogen on carbon No. 1] H NMR: d 6.75 (m,
2
3JHH ˆ 17.2 Hz, 1H, C±CH=), 5.95 (ddt, JHP ˆ 19.2 Hz,
5
3JHH ˆ 17.2 Hz, JHF ˆ 1.6 Hz, 1H, CHP), 5.21 (dm,
3
2JHF ˆ 46.8 Hz, 1H), 4.09 (quintet, JHP ˆ 7.1 Hz, 4H),
3
3
1.46 (dd, JHF ˆ 23.4 Hz, JHH ˆ 6.6 Hz, 1H), 1.34 (td,
1
H NMR: 5.9 (br. s, 1H), 4.59 (d, JHP ˆ 12.1 Hz, 1H),
3JHH ˆ 7.0 Hz, JHP ˆ 0.7 Hz, 6H); 31P NMR: d 18.3 (s);
2
4
19F NMR: d 178.2 (s); 13C NMR: d 149.8 (dd, JCF
ˆ
2
4.21±4.16 (quintet, J ˆ 7.2 Hz, 4H), 3.0 (br. s, 1H), 2.56±
2
1
2.47 (m, 2H, CH2±CH=), 2.42±2.34 (m, 2H, CH2±C=), 1.91
3
19.4 Hz, JCP ˆ 5 Hz, C2), 115.5 (dd, JCF ˆ 188.7 Hz,
3
1 3
(quintet, JHH ˆ 7.4 Hz, 2H), 1.32 (td, JHH ˆ 7.0 Hz,
4JHP ˆ 1.7 Hz, 6H); 31P NMR: d 22.4 (s); 13C NMR:
3JCP ˆ 9 Hz, C1), 87.7 (dd, JCP ˆ 173 Hz, JCF ˆ 22 Hz,
C3), 61.3 (d, 2JCP ˆ 6 Hz), 19.5 (dd, 2JCF ˆ 23 Hz, 4JCP
ˆ
2
1
139.6, 128.7 (d, JCP ˆ 12 Hz), 68.2 (d, JCP ˆ 160 Hz),
2 Hz, C4); m/z 163 (37), 155 (43), 135 (98), 119 (19), 109
(18), 82 (47), 81 (100), 73 (46), 65 (61).
62.8 (m), 32.6 (d, 3JCP ˆ 3 Hz), 32.3 (d, 4JCP ˆ 3 Hz), 23.3,
3
16.3 (d, JCP ˆ 6 Hz). Anal.: Calcd. for C10H19O4P: C,
51.28; H, 8.18. Found: C, 51.10; H, 8.28.
3.9. (E)-Diethyl 3-fluoro-3-methyl-1-butenylphosphonate
(3b)
3.7. 1-(Diethoxyphosphono-a-hydroxymethyl)-4-
isopropylidenecyclohexene (1h)
Using the general ¯uorination procedure 1b [35] (2.52 g,
11.362 mmol) was ¯uorinated with DAST (1.5 ml,
11.0 mmol). Kugelrohr distillation (b.p. 1048C/0.07 mmHg)
afforded 3b (1.29 g, 51%). 1H NMR (400 MHz): d 6.79 (m,
(S)-( ) Perillaldehyde (2.42 g, 16.1 mmol) was stirred
with diethyl phosphite (2.67 g, 19.3 mmol) and KFÁ2H2O
(3.6 g, 38.3 mmol) overnight under standard reaction and
workup conditions to give 1h (3.6 g, 77%) after column
chromatography (silica, gradient hexane:EtOAc 1:1); Rf
3
3JHF ˆ 22.9 Hz, JHH ˆ 17.2 Hz, 1H, C±CH=), 5.92 (ddd,
3
4
2JHP ˆ 19.4 Hz, JHH ˆ 17.2 Hz, JHF ˆ 0.58 Hz, 1H,
3
CHP), 4.10 (quintet, JHP ˆ 7.1 Hz, 2H), 1.66 (d,

G.B. Hammond, D.J. deMendonca / Journal of Fluorine Chemistry 102 (2000) 189±197
195
3JHF ˆ 20 Hz, 6H) 1.34 (t, 3JHH ˆ 7.1 Hz, 6H); 31P NMR: d
methoxy-2-butenylphosphonate (9) (54%) accompanied by
smaller amounts of an unidenti®ed non-¯uorinated vinyl-
phosphonate [31P NMR: d 18.5 (s)]. 3f: 1H NMR: d 6.9±6.6
(m, 1H), 6.14±6.00 (m, 1H), 5.0 (dm, 3JHF ˆ 47.4 Hz, 1H),
4.35 (t, J ˆ 6 Hz, 1H), 4.25±4.05 (m, 4H), 3.47 (s, 3H), 3.43
(s, 3H), 1.34 (m, 6H); 31P NMR: d 17.8 (s), 19F NMR: d
198.5 (s) ppm.
‡
18.9 (s), 19F NMR: d 143.6 (s); m/e 204 (M
20, 10), 163
(20), 149 (20), 148 (20), 135 (50), 133 (45), 95 (60), 81 (70),
67 (100). Anal.: Calcd. for C9H18O3PF: C, 48.21; H, 8.09.
Found: C, 47.73; H, 8.33.
3.10. (E)-Diethyl 3-fluoro-1-hexenylphosphonate (3c)
Using the standard ¯uorination conditions, 1c (5.49 g,
23.0 mmol) in CH2Cl2 (80 ml) was treated with DAST
(3.3 ml, 25.0 mmol) to yield an amber oil (6.24 g) which
yielded 3c (4.27 g, 78%) after vacuum distillation (908C/
0.35 mmHg). 1H NMR (CDCl3): d 6.88±6.62 (m, 1H,
3.14. (E)-1-(Diethoxyphosphonomethylene)-2-
fluorocyclopentane (3g)
Using the general ¯uorination protocol, 1g (311 mg,
1.33 mmol) in CH2Cl2 (15 ml) was treated with DAST
(0.25 ml, 1.65 mmol) to give an amber oil (294 mg, 95%
mass recovery). 19F and 31P NMR indicated the presence of
the E-isomer (31P NMR: d 17.1 ppm, 19F NMR: d
175.7 ppm) (87%), and Z-isomer (31P NMR: d 15.4 ppm,
19F NMR: d 167.5 ppm) (8%) and dehydro¯uorinated pro-
duct (31P NMR: d 20.5 ppm) (5%). Kugelrohr distillation
(b.p. 1008C/0.07 mmHg) did not separate the desired pro-
duct from mixture. Flash chromatography (silica, hexane:E-
tOAc 1:2) of an aliquot (1.024 g) led to the isolation of 3g
(181 mg, 60%) as E:Z 94:6 mixture. 1H NMR: d 5.85 (dm,
2JHP ˆ 17.7 Hz, 1H), 5.18 (dm, 2JHF ˆ 53.9 Hz, 1H), 4.083
(quintet, 3JHP ˆ 7.1 Hz, 2H), 4.080 (quintet, 3JHP ˆ 7.1 Hz,
2H), 2.89±2.80 (m, 1H), 2.66±2.48 (m, 1H), 2.08±1.83 (m,
2
3
C±CH=), 5.94 (dddd, JHP ˆ 19.7 Hz, JHH ˆ 17.2 Hz,
5
5JHF ˆ 1.8 Hz, JHH ˆ 0.4 Hz, 1H, CHP), 5.04 (dm,
3
2JHF ˆ 49.1 Hz, 1H), 4.088 (quintet, JHP ˆ 7.1 Hz, 2H),
4.085 (quintet, 3JHP ˆ 7.1 Hz, 2H), 1.75±1.65 (m, 2H), 1.46
3
3
(sextet, JHH ˆ 7.4, 2H), 1.32 (t, JHH ˆ 7.1 Hz, 6H), 0.94
(t, JHH ˆ 7.2 Hz, 3H); 31P NMR: d 18.4 (s); 19F NMR: d
3
185.9 (s); Anal.: Calcd for C10H20O3PF: C, 50.42; H, 8.46.
Found: C, 50.14; H, 8.47.
3.11. Diethyl 5-fluoro-1,3-hexadienyl-1-phosphonate (3d)
To 1d (1.02 g, 4.36 mmol) in CH2Cl2 (10 ml) was added a
solution of DAST (0.75 ml, 5.7 mmol) in CH2Cl2 (10 ml)
dropwise. After workup a viscous amber oil resulted (1.03 g,
92%) which was judged homogeneous by TLC. 1H NMR: d
7.10 (m, 1H), 6.36 (td, 3JHH ˆ 13.0 Hz, 1 Hz, 1H), 6.08 (td,
3H), 1.80±1.65 (m, 1H), 1.33 (t, JHH ˆ 7.1 Hz, 6H); 31P
3
NMR: d 17.1 (s); 19F NMR: d 175.5 (s). m/z ˆ 216, 188,
160, 141, 106, 78, 51, 29. Anal.: Calcd. for C10H18O3PF: C,
50.85; H, 7.68. Found: C, 50.95; H, 7.71.
3
3JHH ˆ 15.6 Hz, J ˆ 5.2 Hz, 1H), 5.77 (t, JHH ˆ 17.9 Hz,
3
1H), 5.18 (dm, 3JHF ˆ 47.8 Hz, 1H), 4.09 (quintet, JHH
ˆ
7.3 Hz, 4H), 1.45 (dd, 3JHF ˆ 23.6 Hz, 3JHH ˆ 6.5 Hz, 3H),
3.15. (E)-1-(Diethoxyphosphonomethylene)-2-fluoro-4-
isopropylidene cyclohexane (3h)
1.34 (t, JHH ˆ 7.1 Hz, 6H); 31P NMR: d 19.09 (s); 19F
3
‡
NMR: d 172.3 (s); m/z ˆ 236 (M , 25), 216 (8), 189 (75),
160 (55), 135 (40), 111 (20), 84 (100), 79 (90), 65 (35), 51
(38). HRMS C10H18O3PF requires 236.09773; Found:
236.09772.
Using the general method, 1h (1.40 g, 4.86 mmol),
CH2Cl2 (80 ml) and DAST (0.80 ml, 6.05 mmol) produced
an oily mixture (1.17 g). 19F and 31P NMR indicated the
presence of the E-isomer (31P NMR: d 17.0 ppm, 19F NMR:
d 174.8 ppm) (45%), Z-isomer (31P NMR: d 18.7 ppm,
3.12. (E) Diethyl 3-fluoro-3-phenyl-1-
propenylphosphonate (3e)
19
4
4JPF ˆ 4.8 Hz, F NMR: d 179.4, JPF ˆ 5.1 Hz) (31%)
and dehydro¯uorinated product (31P NMR: d 19.4 ppm)
(24%). Kugelrohr distillation (b.p. 1148C/0.05 mmHg) did
not separate the desired product from mixture. Flash chro-
matography (silica, gradient hexane:EtOAc 7:3±4:6) of an
aliquot (0.97 g) led to the isolation of 3h (570 mg, 59%) as
1e (3.66 g, 13.5 mmol) in CH2Cl2 (35 ml) was treated
with DAST (2.4 ml, 18.16 mmol) to afford 3e (2.69 g, 73%).
1H NMR were consistent with that previously reported
for this compound [25] and con®rmed the absence of
the (Z)-stereoisomer. 31P NMR: d 17.8 (s); 19F NMR: d
174.4 (s).
1
2
E:Z 78:22 mixture. E-3g H NMR: d 5.59 (dd, JHP
ˆ
4
2
17.2 Hz, JHF ˆ 2.9 Hz, 1H), 4.93 (dm, JHF ˆ 49.0 Hz,
2
3
1H), 4.75 (d, JHH ˆ 11.8 Hz, 2H) 4.092 (quintet, JHP
ˆ
3
3.13. Diethyl 3-fluoro-4,4-dimethoxy-2-
butenylphosphonate (3f)
7.3 Hz, 2H), 4.085 (quintet, JHP ˆ 7.3 Hz, 2H), 3.18 (dt,
3JHF ˆ 14.0 Hz, JHF ˆ 3.7 Hz, 1H), 2.59±2.43 (m, 2H),
2.28±2.15 (m, 1H), 1.96±1.90 (m, 1H), 1.72 (s, 3H), 1.8±
1.3 (m, 2H), 1.33 (t, 3JHH ˆ 7.1 Hz, 6H); 31P NMR: d 17.0
Under the general ¯uorination protocol, 1f (202 mg,
0.75 mmol), CH2Cl2 (15 ml) and DAST (0.14 ml,
0.95 mmol) produced an oily residue (173 mg, 85% recov-
ery) which was chromatographed (silica, EtOAc) to give a
mixture of two compounds (3f) (46%) and diethyl 4-oxo-3-
(s); 19F NMR: d 174.8 (s); m/e 290 (M , 20), 270 (65), 244
‡
(20), 202 (50), 174 (25), 146 (50), 132 (100), 117 (48), 105
(25), 91 (70), 65 (25), 53 (28). C14H24O3PF required
290.14467. Found 290.14468.

196
G.B. Hammond, D.J. deMendonca / Journal of Fluorine Chemistry 102 (2000) 189±197
4
3.16. (Z)-Diethyl 3-fluoro-2-butenylphosphonate (5a).
General method
J ˆ 10.9); 19F NMR:
d
117.3 (d, JFP ˆ 13.8 Hz);
‡
m/z ˆ 272 (M , 15), 216 (7), 199 (10), 135 (100), 133
(44), 115 (98), 109 (38), 81 (39), 29 (39). Anal.: Calcd.
for C13H18O3PF: C, 57.35; H, 6.66. Found: C, 57.20;
H, 6.57.
Into a vacuum±dried mixture of KN(TMS)2 (0.637 g,
3.15 mmol) and 18-crown-6 (1.614 g, 5.72 mmol) dissolved
in THF (30 ml) and cooled to 808C was added a cold
( 808C) solution of 3a (0.602 g, 2.86 mmol) in THF
(10 ml) via cannula, and the resulting mixture was stirred
for 4 h at 808C before addition of saturated aqueous
NH4Cl (30 ml). The layers were separated and the aqueous
phase was extracted with diethyl ether (2 Â 25 ml). The
combined organic layers were dried over anhydrous
MgSO4, ®ltered and concentrated to give an oil (0.554 g).
An aliquot (77 mg) was puri®ed using ¯ash chromatography
3.19. 1-Diethoxyphosphonomethyl-5-fluorocyclopentene
(5d)
Applying the general methodology described for 5a,
KN(TMS)2 (181 mg, 0.91 mmol), 18-crown-6 (523 mg,
1.98 mmol), THF (20 ml) and a solution of 3g (170 mg,
0.72 mmol) in THF (4 ml) were stirred at 858C for 0.5 h
and then slowly warmed to 158C over 3 h. yielding 5d
(141 mg, 83%). An aliquot (120 mg) was chromatographed
(silica, gradient hexane:EtOAc) yielding a very low recov-
ery of 5d (40 mg, 34%), probably as a result of HF loss and
1
(silica, hexane:EtOAc 3:7) yielding 5a (56 mg, 67%). H
NMR: d 4.62 (dq, 3JHF ˆ 34.7 Hz, 3J ˆ 7.2 Hz, 1H, HC=),
3
2
4.11 (quintet, JHP ˆ 7.2 Hz, 4H), 2.60 (ddd, JHP
ˆ
1
21.3 Hz, JHH ˆ 7.8 Hz, 4J ˆ 1.1 Hz, 2H), 1.92 (ddd,
concomitant diene formation. H NMR: d 6.03 (br. s, 1H),
3
5
3JHF ˆ 16.7 Hz, JHH ˆ 5.2 Hz, J ˆ 1.1 Hz, 3H), 1.32 (t,
5.58 (dm, 2JFH ˆ 57 Hz, 1H), 4.11 (quintet, 3JHP ˆ 7.2 Hz,
4
31
4
2
3JHH ˆ 7.1 Hz, 6H); P NMR: d 28.1 (d, JFP ˆ 10 Hz);
4H), 2.74 (br. d, JHP ˆ 20.5 Hz, 2H), 2.9±1.75 (m, 4H),
19
‡
4
F NMR: d 99.2 (d, JFP ˆ 10 Hz); m/z ˆ 210 (M , 32),
1.30 (t, 3JHH ˆ 7.2 Hz, 6H); 31P: d 27.2 (d, 4JFP ˆ 8.0); 19
F
4
162 (14), 138 (52), 134 (42), 111 (75), 109 (57), 101 (30), 91
(38), 81 (100), 73 (88). Anal.: Calcd. for C8H16O3PF: C,
45.72; H, 7.67. Found: C, 45.85; H, 7.75.
NMR: d 166.1 (d, JFP ˆ 8.8 Hz). Anal.: Calcd. for
C10H18O3PF: C, 50.85; H, 7.68. Found: C, 50.69; H, 7.71.
3.20. 1-Diethoxyphosphonomethyl-2-fluoro-4-
isopropylidenecyclohexane (5e)
3.17. (Z)-Diethyl 3-fluoro-2-hexenylphosphonate (5b)
KN(TMS)2 (479 mg, 2.40 mmol), 18-crown-6 (1.52 g,
5.75 mmol), THF (60 ml) and a solution of 3c (440 mg,
1.85 mmol) in THF (20 ml)were mixed and stirred for 4.5 h
to yield a pale yellow oil (422 mg). Puri®cation using ¯ash
KN(TMS)2 (96 mg, 0.48 mmol), 18-crown-6 (277 mg,
1.16 mmol), THF (10 ml) and a solution of 3h (105 mg,
0.36 mmol) in THF (4 ml) were stirred at 858C for 0.5 h
and then slowly warmed to room temperature overnight
yielding 5e (81 mg, 77%) as a 91:9 mixture of vinyl¯uoro
and allyl¯uoro isomers. An aliquot (77 mg) underwent
extensive degradation when chromatographed (silica, gra-
dient hexane:EtOAc) yielding a considerably low amount of
5e (11 mg, 14%). 1H NMR: d 4.75 (d, 2JHH ˆ 8 Hz, 2H) 4.1
1
chromatography produced 5b (343 mg, 78%). H NMR d
3
3
4.61 (dq, JHF ˆ 35.2 Hz, J ˆ 7.8 Hz, 1H, HC=), 4.105
(quintet, 3JHP ˆ 7.2 Hz, 2H), 4.102 (quintet, 3JHPˆ 7.2 Hz,
4
2
3
2H), 2.61 (ddd, JHP ˆ 21.3 Hz, JHH ˆ 7.8 Hz, J ˆ 0.9
3
Hz, 2H), 2.23±2.11(m, 2H), 1.54 (sextet, JHH ˆ 7.4 Hz,
3
3
3
3
2H), 1.32 (t, JHH ˆ 7.1 Hz, 6H), 0.94 (t, JHH ˆ 7.4 Hz,
(quintet, JHP ˆ 7.3 Hz, 4H), 2.65 (d, JHF ˆ 22 Hz, 2H),
3H); P NMR: d 28.1 (d, JFP ˆ 10.2 Hz); 19F NMR: d
2.4±2.2 (m, 5H), 1.9±1.70 (m, 1H), 1.75 (s, 3H), 1.5±1.3 (m,
31
4
105.7 (d, JFP ˆ 8.2 Hz); m/z ˆ 238 (M , 12), 162 (11),
152 (31), 138 (31), 111 (57), 81 (100), 59 (48). Anal.:
Calcd. for C10H20O3PF: C, 50.42; H, 8.46. Found: C, 49.90;
H, 8.48.
1H), 1.33 (t, JHH ˆ 7.1 Hz, 6H); 31P NMR: d 27.5 (d,
‡
4
3
4JFP ˆ 11 Hz); 19F NMR: d 105.6 (d, 4JFP ˆ 13 Hz). Anal.:
Calcd. for C14H24O3PF: C, 57.92; H, 8.33. Found: C, 57.82;
H, 8.36.
3.18. (Z)-Diethyl 3-fluoro-3-phenyl-2-
propenylphosphonate (5c)
3.21. (Z)-Diethyl 1-benzyl-3-fluoro-2-butenylphosphonate
(6a)
Compound 5a, KN(TMS)2 (83 mg, 0.42 mmol), 18-
crown-6 (297 mg, 1.12 mmol), THF (10 ml) and a solution
of 3e (99 mg, 0.36 mmol) in THF (5 ml) were allowed to
warm up to room temperature with stirring during 20 h to
yield 5c (82 mg, 83%). An analytical sample was isolated
To a solution of 5a (112 mg, 0.53 mmol) in THF (15 ml)
cooled to 608C was added LiN(TMS)2 (1.1 ml, 1 M in
THF, 2 eq), the resulting mixture was removed from the cold
bath for 15 min and then cooled down to 608C before
addition of benzyl bromide (136 mg, 0.79 mmol). The
resulting mixture was allowed to warm to room temperature
overnight and quenched with saturated NH4Cl (25 ml). The
layers were separated and the aqueous phase was extracted
with diethyl ether (2 Â 25 ml). The combined organic layers
were dried over anhydrous MgSO4, ®ltered, concentrated
following ®ltration using a silica plug (hexane:EtOAc
1:1).1H NMR: d 7.58±7.32 (m, 5H), 5.46 (dq, JHF
3
ˆ
34.0 Hz, 3J ˆ 7.6 Hz, 1H, HC=), 4.14 (quintet, JHP
ˆ
3
2
3
7.2 Hz, 4H), 2.84 (dd, JHP ˆ 22.7 Hz, JHH ˆ 7.6 Hz,
2H), 1.32 (t, JHH ˆ 7.2 Hz, 6H); 31P NMR: d 27.5 (d,
3

G.B. Hammond, D.J. deMendonca / Journal of Fluorine Chemistry 102 (2000) 189±197
197
[3] F. Camps, J. Coll, G. Fabrias, A. Guerrero, Tetrahedron 40 (1984)
2871±2878.
and puri®ed using ¯ash chromatography (silica, gradient
hexane:EtOAc) to give 6a (147 mg, 92%). 1H NMR: d 7.3±
[4] A. Alcazar, F. Camps, J. Coll, G. Fabrias, A. Guerrero, Synth.
Commun. 15 (1985) 819±827.
3
7.1 (m, 5H), 4.46 (dm, JHF ˆ 34.6 Hz, 1H, HC=), 4.11
(quintet, 3JHP ˆ 7.1 Hz, 4H), 3.36±3.18 (m, 2H), 2.76±2.64
[5] A.E. Asato, H. Matsumoto, M. Denny, R.S.H. Liu, J. Am. Chem.
Soc. 100 (1978) 5957±5960.
3
4
(m, 1H), 1.81 (ddd, JHF ˆ 16.7 Hz, JHH ˆ 5.0 Hz,
5
3
4
[6] B.A. Pawson, K.-K. Chan, J. DeNoble, R.-J.L. Han, V. Piermattie,
A.C. Specian, S. Srisethnil, J. Med. Chem. 22 (1979) 1059±1067.
[7] A.E. Asato, A. Peng, M.Z. Hossain, T. Mirzadegan, J.S. Bertram, J.
Med. Chem. 36 (1993) 3137±3147.
J ˆ 0.6 Hz, 3H), 1.32 (td, JHH ˆ 7.2 Hz, JHP ˆ 1.9 Hz,
6H); 31P NMR: d 29.2 (d, 4JFP ˆ 8 Hz); 19F NMR: d 98.0 (d,
4JFP ˆ 8 Hz); Anal.: Calcd. for C15H22O3PF: C, 59.99; H,
7.58. Found: C, 59.42; H, 7.44.
[8] T.M.T. Nguyen-Truong, H. Togo, M. Schlosser, Tetrahedron 50
(1994) 7827±7836.
[9] H. Al-Badri, E. About-Jaudet, N. Collignon, Synthesis (1994) 1072±
1078.
3.22. (Z)-Diethyl 1,3-dfluoro-2-hexenylphosphonate (6b)
[10] M. Kamal, G. Dewynter, J.-L. Montero, Bioorg. Med. Chem. Lett. 5
(1995) 2461±2464.
To a solution of 5b (107 mg, 0.45 mmol) in THF (15 ml
cooled to 788C was added LiN(TMS)2 (0.65 ml, 1 M in
THF, 1.3 eq), the resulting mixture was removed from the
cold bath for 15 min and then cooled down to 788C before
addition of NFSI (187 mg, 0.59 mmol). The resulting mix-
ture was allowed to warm to room temperature overnight
before addition of brine (20 ml), diethyl ether (50 ml) and
NaOH 14% (7 ml). The layers were separated and the
aqueous phase was extracted with diethyl ether
(2 Â 25 ml). The combined organic layers were washed
with water, dried over anhydrous MgSO4, ®ltered, concen-
trated and puri®ed using ¯ash chromatography (silica,
gradient hexane:EtOAc) to give 6b (77 mg, 66%). 1H
[11] N. Kann, T. Rein, B. Akermark, P. Helquist, J. Org. Chem. 55 (1990)
5312±5323.
[12] H.J. Bestmann, J. Sub, O. Vostrowsky, Tetrahedron Lett. (1979) 245±
246.
[13] C. Canevet, T. Roder, O. Vostrowsky, J. Bestmann, Chem. Ber. 113
(1980) 1115±1120.
[14] K. Kondo, A. Negishi, D. Tunemoto, Angew. Chem. Internat. Edit.
13 (1974) 407±408.
[15] X. Lu, J. Zhu, J. Organomet. Chem. 304 (1986) 239±243.
[16] T. Janecki, R. Bodalski, Synthesis (1990) 799±801.
[17] X. Lu, J. Zhu, Synthesis (1986) 563±564.
[18] A. Hafner, W. Philipsborn, A. Salzer, Helv. Chim. Acta 69 (1986)
1757±1767.
[19] Y. Shen, J. Yao, J. Chem. Res. (S) (1996) 428±429.
[20] J. Kiddle, J.H. Babler, J. Org. Chem. 58 (1993) 3572±3574.
[21] F. Benayoud, G.B. Hammond, J. Chem. Soc. Chem. Commun.
(1996) 1447±1448.
2
2
NMR: d 5.57 (ddd, JHF ˆ 41.0 Hz, JHP ˆ 10.3 Hz,
3JHH ˆ 6.0 Hz, 1H), 5.08±4.88 (m, 1H, HC=), 4.27±4.07
(m, 4H), 2.28±2.04 (m, 2H), 1.5±1.4 (m, 2H), 1.39±1.25 (m,
[22] F. Benayoud, D.J. deMendonca, C.A. Digits, G.A. Moniz, T.C.
Sanders, G.B. Hammond, J. Org. Chem. 61 (1996) 5159±5164.
[23] F. Benayoud, L. Chen, G.A. Moniz, A.J. Zapata, G.B. Hammond,
Tetrahedron 54 (1998) 15 541±15 554.
6H), 0.94 (t, JHH ˆ 7.4 Hz, 3H); 31P NMR: d 16.7 (dd,
3
2JFP ˆ 88.1 Hz, JPF ˆ 9.8 Hz); 13C NMR: d 166.2 (ddd,
4
3
3
1JCF ˆ 256.6 Hz, JCF ˆ 12.4 Hz, JCP ˆ 11.0 Hz), 98.3
[24] A.N. Pudovik, I.V. Konovalova, Synthesis (1979) 81±96.
[25] G.M. Blackburn, D.E. Kent, J. Chem. Soc. Perkin Trans. 1 (1986)
913±917.
(dd, 2JCF ˆ 21.4 Hz, 2JCP ˆ 11.5 Hz), 81.4 (ddd,
1
3
1JCF ˆ 179.5 Hz, JCP ˆ 174.8 Hz, JCF ˆ 7.2 Hz), 63.4
2
2
(d, JCP ˆ 6.7 Hz), 63.2 (d, JCP ˆ 6.7 Hz), 33.9 (d,
2JCF ˆ 24.8 Hz), 18.9, 16.3 (d, 3JCP ˆ 5.9 Hz), 13.2; Anal.:
Calcd. for C10H19O3PF2: C, 46.88; H, 7.47. Found: C,
46.58; H, 7.36.
[26] F. Texier-Boullet, A. Foucaud, Synthesis (1982) 165±166.
[27] K. Ando, Tetrahedron Lett. 36 (1995) 4105±4108.
[28] S.I. Spassov, L. Markova, D.M. Mondeshka, C.N. Tancheva, C.M.
Angelov, Magn. Resonance Chem. 23 (1985) 578±578.
[29] J.P.M. Gerber, T.A. Wagener, C.C.P. Zwierzak, Heteroatom Chem. A
2 (1991) 643±649.
[30] E.L. Muller, T.A. Modro, Bull. Soc. Chim. Fr. 130 (1993) 668±
672.
Acknowledgements
[31] J. Clayden, S. Warren, Angew. Chem. Int. Ed. Engl. 35 (1996)
241±270.
The generous ®nancial support of the National Science
Foundation (CHE-9711062), the Petroleum Research Fund
(PRF No. 32595-B1), and the Camille and Henry Dreyfus
Foundation (TH-96-012) is deeply appreciated.
[32] D. Cavalla, C. Gueguen, A. Nelson, P. O'Brien, M.G. Russell, S.
Warren, Tetrahedron Lett. 37 (1996) 7465±7468.
[33] B.E. Maryanoff, A.B. Reitz, Chem. Rev. 89 (1989) 863±927.
[34] A.D. Buss, W.B. Cruse, O. Kennard, S. Warren, J. Chem. Soc. Perkin
Trans. 1 (1984) 243±247.
[35] A.N. Pudovik, Doklady Akad. Nauk S.S.S.R. 73 (1950) 499±502
[Chem. Abstracts 45, 2853a].
References
[36] R. Gree, H. Tourbah, R. Carrie, Tetrahedron Lett. 27 (1986) 4983±
4986.
[1] M.J. Hensel, P.L. Fuchs, Synth. Commun. 16 (1986) 1285±1295.
[2] F. Camps, G. Fabrias, A. Guerrero, Tetrahedron 42 (1986) 3623±
3629.
[37] J. Brown, D. Henbest, G. Jones, J. Chem. Soc. (1950) 3634±3639.