Allylic fluorination via an unusual alkene Z/E isomerisation

Article abstract of DOI:10.1016/j.tetlet.2009.03.068

The isomerisation of readily available (Z)-4-(para-methoxybenzyloxy)-1-chloro-2-butene was achieved under mild conditions to afford the much less accessible E-diastereoisomer. This was an effective substrate in Sharpless asymmetric dihydroxylation (AD) re

Full text of DOI:10.1016/j.tetlet.2009.03.068

Tetrahedron Letters 50 (2009) 3571–3573
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Allylic fluorination via an unusual alkene Z/E isomerisation
b,c
James A. B. Laurenson a, Sebastien Meiries b, Jonathan M. Percy a,b, , Ricard Roig
*
a WestCHEM Department of Pure and Applied Chemistry, University of Strathclyde, Thomas Graham Building, 295 Cathedral Street, Glasgow G1 1XL, UK
b Department of Chemistry, University of Leicester, University Road, Leicester LE1 7RH, UK
c International Flavours and Fragrances, Avda. Felipe Klein 2, 1258 Benicarlo (Castellon), Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
The isomerisation of readily available (Z)-4-(para-methoxybenzyloxy)-1-chloro-2-butene was achieved
under mild conditions to afford the much less accessible E-diastereoisomer. This was an effective sub-
strate in Sharpless asymmetric dihydroxylation (AD) reactions, delivering highly enantiomerically
enriched fluorinated butene triol building blocks.
Received 15 January 2009
Revised 25 February 2009
Accepted 9 March 2009
Available online 16 March 2009
Crown Copyright Ó 2009 Published by Elsevier Ltd. All rights reserved.
Allylic fluorination by nucleophilic displacement of a good leav-
ing group by fluoride has proved to be a surprisingly difficult reac-
tion to achieve generally and reliably.1 SN2 reactions at allylic
centres are accelerated by the ability of the alkenyl group to lower
the energy of the trigonal bipyramidal transition state, but when
basic fluoride ion is the nucleophile, little advantage accrues and
side reactions often compete effectively. Nucleophilic fluorinating
agents such as DAST, or the Ichikawa reagent can be effective but
SN20 fluorination is a well-documented competing pathway.2 Our
work on the synthesis of 6-deoxy-6-fluorosugars3 demanded a
route to allylic fluoride building blocks 1 and 2 from which we in-
tended to prepare the corresponding monoprotected fluorinated
butene triols 3 or 4, ideally in highly enantiomerically enriched
form (Fig. 1).
The de novo approach to hexose synthesis was pioneered by
Masamune and co-workers,4 and it is worthwhile if an array of su-
gar analogues is sought, rather than a single compound. We there-
fore sought to develop direct routes to these species.
Monoprotected Z-but-2-ene diol is readily available from the
diol via monoalkylation5 (acceptable for benzylation to 5), or by
acetal formation6 and reductive cleavage with DIBAl-H7 (better
for the formation of the mono PMB ether 7) (Scheme 1).
We prepared both compounds 5 and 7 and chlorinated them;
thionyl chloride in diethyl ether worked well for 6 but the in situ
mesylation and displacement method8 was superior for 8. Fluori-
nation was explored for 6 initially; treatment with ‘anhydrous’
TBAF (prepared by heating TBAFꢀ3H2O at very low pressure below
40 °C for several hours)9 in THF afforded small amounts of 9 with 5
and 10 dominating the reaction mixture (GC–MS) (Scheme 2). The
TBAFꢀ3H2O reagent was more effective, transforming 6 to 9 in
moderate (26%) yield. Based on a solid–liquid phase transfer proce-
dure (KF/TBAI) reported by Loupy and co-workers,10 we reacted 6
with TBAI (2 equiv in THF at reflux) followed by the addition of
neat TBAFꢀ3H2O; we were surprised to isolate E-fluoride 12 (the
major component of a mixture of fluorides) with alcohol 11
(29%).11 We isolated the products of the reaction with TBAI, iden-
tifying chlorides 6, 13 and 14, and followed the reaction by GC–MS.
These processes occurred for both benzyl and PMB ethers;
Figure 2 shows the evolution of the reaction of 8 (3 equiv TBAI,
THF, reflux) with time followed by GC–MS, affording products 15
and 16.
No further change occurred up to 20 h, suggesting strongly that
the reaction had reached equilibrium. Isomerisation to the more
OH
OPG
OH
OPG
or
F
X
F
F
Y
OH
Cl
OH
4
3
1
, X = H, Y = CH2OPG;
2, X = CH2OPG, Y = H.
or their enantiomers
Figure 1.
HO
HO
SOCl2
NaH, DMF
then BnBr
HO
Et2O, 15 o
C
BnO
BnO
5
6
82%
75%
p-anisaldehyde
p-TsOH, PhMe
Dean-Stark
Cl
HO
DIBAl-H
O
MsCl, LiCl
s-collidine
THF, 0 o
C
PMBO
O
PMBO
DMF, 0 o
C
PMP
8
96%
7
85%
72%
* Corresponding author.
E-mail address: jonathan.percy@strath.ac.uk (J.M. Percy).
Scheme 1. Preparation of Z-allylic chlorides.
0040-4039/$ - see front matter Crown Copyright Ó 2009 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.03.068

3572
J. A. B. Laurenson et al. / Tetrahedron Letters 50 (2009) 3571–3573
"anhydrous" HO
TBAF, THF
F
+
HO
O
O
PCC
DCM
+
+
6
OR
BnO
10
BnO
BnO
TBAF.3H2O
reflux
18a
NaBH4
18b 82%
RO
RO
5
9
9
HO
5, R = Bn;
THF
5
11
17, R = allyl
20%
26%
via
RO
i) TBAI, THF
reflux
HO
F
+
Scheme 3. Preparation of 11 via PCC oxidation accompanied by an alkene
isomerisation.
OBn
OBn
OBn
+
ii) TBAF.3H2O
11
12
29%
28%
TBAI
Cl
note that 14 is more hindered than 6 and may therefore react quite
slowly once formed, so it would not be expected to be present in
significant amounts at equilibrium.
Our recent asymmetric route to 6-deoxy-6-fluorosugars via
hexadienoate 19 used two fluorination methods,15 one a melt fluo-
rination with KHF2/TBAFꢀ3H2O and the other the KF/TBAHSO4
method of Hou and co-workers.16
+
+
OBn
6
8
THF, reflux
95% overall
Cl
Cl
13
15
14
16
13:14:6 = 6 : 1 : 1
(by GC-MS)
Cl
TBAI
OPMB
+
OPMB
8
THF, reflux
96% overall
15:16:8 = 7 : 1 : 1
(by GC-MS)
F
CO2Me
F
F
Scheme 2. Discovery of the Z/E isomerisation.
OPMB
19
20
21
PMBO
100
90
80
70
60
50
40
30
20
10
0
The E/Z mixture of 15 and 8 underwent melt fluorination with
KHF2/TBAFꢀ3H2O quickly, but the reaction formed a significant
number of side products. In contrast, the KF/TBAHSO4 method gave
a very clean product, but slowly, and required a large excess of re-
agent. The TBAFꢀ(t-BuOH)4 complex described recently by Kim et
al.,17 was the most effective of all the reagents tried (Scheme 4).
The knowledge gained was next applied in the 20 mmol scale
synthesis of 20. Sharpless asymmetric dihydroxylation (AD) reac-
tions were now attempted with the usual commercial AD mixes
Chloride 16
Chloride 8
Chloride 15
[DHQ (for a) and DHDQ (for b) ligands]. As expected, Z-diastereo-
isomer 21 afforded 23a/23b in good yield (90%) but in low (18%)
ee (determined by chiral HPLC of the acetonides, Chiralcel OD-H
column with 1:999 i-PrOH/hexane). The presence of Z-diastereo-
isomer in the isomerised E-material leads to a low dr (13:2 syn:an-
ti) from the AD, though the ees (92% from AD-a, 84% from AD-b)
are quite acceptable and consistent with the level of enantioen-
0
100
200
300
Time (min)
400
500
600
700
richment obtained for related substrates.
We found that cyclohexylidene acetal formation (Scheme 5)
Figure 2. Following the course of Z-chloride isomerisation.
afforded a mixture from which 24a (from AD-mix-a) and 24b
(from AD-mix-b) could be separated effectively (FCC, Buchi Sepa-
core with Biotage SNAP cartridge, 0–30% EtOAc in hexane). AD with
the (DHQD)2AQN ligand18 afforded no improvement in ee for 24a
or 24b.
In conclusion, the Z to E isomerisation of allylic chlorides 6 and
8 affords a very convenient way of making mixtures enriched in
the diastereoisomeric E-species. PMB ether 20 was an effective
substrate in Sharpless AD reactions, delivering highly enantiomeri-
cally enriched fluorinated butene triol building blocks.
Procedure for isomerisation of 8: Chloride 8 (26.5 mmol, 6.00 g)
stable E-chloride 13 or 15 presumably involves sequential SN20 dis-
placements with iodide (the softest nucleophile present) and oc-
curs via 14 or 16. We have not found examples of alkene
isomerisation by this type of mechanism; Danishefsky and Regan12
prepared 11 via E-butenal 18a by PCC oxidation of 5 (the recent
PCC oxidation of ether 17 affords 18b in excellent yield13); the
isomerisation can proceed through acid-catalysed enolisation and
reprotonation (Scheme 3). This would seem to be a very different
process to the one we describe.
was added to
a
solution of tetrabutylammonium iodide
The persistence of 14 and 16 in these mixtures is interesting. To
gain some insight, we determined the lowest energy conforma-
tions of 6, 13 and 14 using the equilibrium conformer search algo-
rithm within SPARTAN’06 v1.1.0 (MMFF94).14 Geometries were
optimised, with full frequency calculations (B3LYP/6-31G*), allow-
ing free energies (gas phase, 298 K) to be determined relative to
13; 6 (+6.75 kJ molꢁ1) and 14 (+8.44 kJ molꢁ1) are indeed less sta-
ble, but the free energy differences would suggest that 13 should
be favoured more decisively at equilibrium. Of course, these are
gas phase energies and it is possible that they may converge once
solvent polarity is taken into account in the calculations. We also
(52.9 mmol, 19.6 g) in dry tetrahydrofuran (70 mL). After 24 h at
reflux, the mixture was diluted with diethyl ether (20 mL), filtered
TBAF.(t-BuOH)4
Cl
OPMB
F
OPMB
MeCN, 70 o
C
8
21,
Z 52%
, Z
15
8
20 21,
: 7:1, E:Z 47%
:
, 7:1, E:Z
Scheme 4. Fluorination of allyl chlorides with a new fluoride source.

J. A. B. Laurenson et al. / Tetrahedron Letters 50 (2009) 3571–3573
3573
at 70 °C for 15 h. An immediate colour change from pale yellow to
pale brown was observed. After cooling to room temperature, the
mixture was diluted with diethyl ether (15 mL) and washed with
water (2 ꢂ 15 mL) and brine (15 mL). The organic layer was then
dried (MgSO4), concentrated under vacuum and purified by flash
chromatography (Buchi Sepacore, 25% diethyl ether in hexane) to
afford 20 (major component of a 7:1:1 mixture) as a colourless
oil (346 mg, 47%); Rf (20% ethyl acetate in hexane) 0.60; mmax
(neat)/cmꢁ1 2937w (CH2), 2838w (CH3), 1612w (C@C), 1512s,
1245s, 1033s (C–O); dH (300 MHz, CDCl3) 7.25 (2H, d, J 8.8, ArH),
6.85 (2H, d, J 8.8, ArH), 5.96–5.84 (2H, m, H-2, H-3), 4.82 (2H, m,
OH
OH
F
F
OPMB
22a
OPMB
OH
OH
OH
OH
or
23a
AD-α for 22b (79%)
and
20
22b
22a
AD-β for
(84%)
F
F
OPMB
OPMB
OH
OH
23b
Cyclohexanone
p-TsOH, CuSO4
THF, reflux
Cyclohexanone
p-TsOH, CuSO4
THF, reflux
2
inc. app. d, JF–H ꢃ 46.8, H-1), 4.43 (2H, s, OCH2Ar), 4.04–3.97
(2H, m, H-4), 3.74 (3H, s, OCH3); dC (75 MHz, CDCl3) 159.3, 131.5
and
3
2
(d, JC–F 11.6, C-3), 130.2, 129.4, 126.9 (d, JC–F 16.9, C-2), 113.8,
or
O
1
4
O
O
O
82.8 (d, JF–C 162.9, C-1), 72.1 (CH2Ar), 69.3 (d, JC–F 1.5, C-4),
55.3 (OCH3); dF (282 MHz, CDCl3) (ꢁ212.7)–(ꢁ213.1) (m); {1H}dF
(282 MHz, CDCl3)–212.85 (s); [HRMS (EI, M+) Found: 210.10558.
Calcd For C12H15O2F 210.10561]; m/z (EI) 210 (8%, M+), 179 (6),
136 (30, [MꢁOPMB]+), 135 (27), 122 (22), 121 (100, PMB+), 77 (29).
O
O
O
O
F
F
F
F
24a
24b
25a
25b
OPMB
OPMB
OPMB
OPMB
Scheme 5. Sharpless AD and subsequent cyclohexylidene protection. Reagents and
conditions: AD-mix-
a
: K2OsO4ꢀ2H2O (1 mol %), (DHQ)2PHAL (5 mol %), CH3SO2NH2
Acknowledgements
(1 equiv), K3Fe(CN)6 (3 equiv), K2CO3 (3 equiv), t-BuOH/H2O (1:1), 24 h, 0 °C; AD-
mix-b: K2OsO4ꢀ2H2O (1 mol %), (DHQ)2PHAL (5 mol %), CH3SO2NH2 (1 equiv),
K3Fe(CN)6 (3 equiv), K2CO3 (3 equiv), t-BuOH/H2O (1:1), 24 h, 0 °C; AQN AD-mix-
b: K2OsO4ꢀ2H2O (0.4 mol %), (DHQD)2AQN (1 mol %), K3Fe(CN)6 (3 equiv), K2CO3
(3 equiv), t-BuOH/H2O (1:1), 48 h, 0°C.
This work was supported by the University of Leicester (stu-
dentship to R.R.), the Engineering and Physical Sciences Research
Council (EPSRC, GR/S82053/02, consumable support to R.R.,
J.A.B.L.) and WestCHEM (studentship to J.A.B.L.). We also thank
the EPSRC National Mass Spectrometry Service Centre, University
of Wales Swansea for mass spectrometric measurements.
and concentrated in vacuo to obtain a crude mixture of inseparable
allylic chlorides (E:Z:terminal = 7:1:1) in which E-allylic chloride
15 (5.73 g, 96%, 83% purity by GC–MS) was the major component
as a brown and viscous oil; Rf (20% ethyl acetate in hexane) 0.62;
mmax (neat)/cmꢁ1 2954w (CH2), 2837w (CH2), 1612w (C@C),
1512s, 1245s, 1033s; dH (400 MHz, CDCl3) 7.26 (2H, d, J 8.7, ArH),
6.88 (2H, d, J 8.7, ArH), 5.91–5.86 (2H, m, H-2, H-3), 4.45 (2H, s,
CH2Ar), 4.08–4.05 (2H, m, H-1), 4.02–4.00 (2H, m, H-4), 3.80 (3H,
s, OCH3); dC (75 MHz, CDCl3) 159.3, 131.4 (C-3), 130.1, 129.4,
128.3 (C-2), 113.8, 72.1 (CH2Ar), 69.2 (C-4), 55.3 (OCH3), 44.4 (C-
1); [HRMS (EI, M+) Found: 226.07611 Calcd For C12H15O2Cl
226.07606]; m/z (EI) 226 (4%, M+), 191 (12), 161 (5), 136 (16,
[MꢁOPMB]+), 135 (14), 121 (100, PMB+), 109 (10). Data for 8: Rf
(20% ethyl acetate in hexane) 0.62; mmax (neat)/cmꢁ1 2935w
(CH2), 2837w (CH2), 1611w, 1511s, 1245s, 1033s; dH (400 MHz,
CDCl3) 7.26 (2H, d, J 8.8, ArH), 6.88 (2H, d, J 8.8, ArH), 5.82–5.72
(2H, m, H-2, H-3), 4.44 (2H, s, CH2Ar), 4.14–4.06 (4H, m, H-1, H-
4), 3.80 (3H, s, OCH3); dC (75 MHz, CDCl3) 159.3, 130.9 (C-3),
130.3, 129.5, 128.4 (C-2), 114.2, 72.1 (CH2Ar), 64.8 (C-4), 55.6
(OCH3), 39.6 (C-1); [HRMS (EI, M+) Found: 226.07613 Calcd For
C12H15O2Cl 226.07606]; m/z (EI) 226 (3%, M+), 191 (10), 161 (5),
136 (13, [MꢁOPMB]+), 121 (100, PMB+), 109 (9).
References and notes
1. Pacheco, M. C.; Purser, S.; Gouverneur, V. Chem. Rev. 2008, 108, 1943–1981; See
also: Roig, R.; Percy, J. M. Sci. Synth. 2006, 34, 319–338.
2. For a review, see Dax, K. Sci. Synth 2006, 34, 71–142.
3. For recent novel routes to 6-deoxy-6-fluorosugars, see: Audouard, C.; Bettaney
(née Middleton), K.; Doan, C. T.; Rinaudo, G.; Jervis, P. J.; Percy, Org. Biomol.
Chem., in press. doi:10.1039/b817672H.; Percy, J. M.; Roig, R.; Singh, K. Eur. J.
Org. Chem. 2009, 1058–1071.
4. Ko, S. Y.; Lee, A. W. M.; Masamune, S.; Reed, L. A.; Sharpless, K. B.; Walker, F. J.
Science 1983, 220, 949–951.
5. Yadav, J. S.; Reddy, P. S. Synth. Commun. 1986, 16, 1119–1131.
6. Williams, R. M.; Rollins, S. B.; Judd, T. C. Tetrahedron 2000, 56, 521–532.
7. Takano, S.; Akiyama, M.; Sato, S.; Ogasawara, K. Chem. Lett. 1983, 1593–1596.
8. Collington, E. W.; Meyers, A. I. J. Org. Chem. 1971, 36, 2.
9. Cox, D. P.; Terpinski, J.; Lawrynowicz, W. J. Org. Chem. 1984, 49, 3216–3219.
10. Bram, G.; Loupy, A.; Pigeon, P. Synth. Commun. 1988, 18, 1661–1667.
11. For
a recent example, see: Ooi, T.; Sugimoto, H.; Doda, K.; Maruoka, K.
Tetrahedron Lett. 2001, 42, 9245–9248.
12. Danishefsky, S.; Regan, J. Tetrahedron Lett. 1981, 22, 3919–3922.
13. Hansen, E. C.; Lee, D. Org. Lett. 2004, 6, 2035–2038.
14. SPARTAN’06, v1.1.0, Wavefunction, I.: Irvine, CA, 2006.
15. Caravano, A.; Field, R. A.; Percy, J. M.; Rinaudo, G.; Roig, R.; Singh, K. Org. Biomol.
Chem. 2009, 7, 996–1008.
16. Fan, R. H.; Zhou, Y. G.; Zhang, W. X.; Hou, X. L.; Dai, L. X. J. Org. Chem. 2004, 69,
335–338.
17. Kim, D. W.; Jeong, H. J.; Lim, S. T.; Sohn, M. H. Angew. Chem., Int. Ed. 2008, 47,
8404–8406.
Procedure for fluorination of 15 with TBAFꢀ(t-BuOH)4: TBAFꢀ(t-
BuOH)4 (4.02 g, 2.1 equiv) was added to a solution (15 mL) of E-al-
lyl chloride 15 (3.5 mmol, 798 mg, major component of a 7:1:1
mixture with 8 and 16) in acetonitrile and the mixture was heated
18. Becker, H.; Sharpless, K. B. Angew. Chem., Int. Ed. 1996, 35, 448–451.