Synthesis of 3-fluoropyrrolidines and 4-fluoropyrrolidin-2-ones from allylic fluorides

Article abstract of DOI:10.1002/chem.201201576

Various 3-fluoropyrrolidines and 4-fluoropyrrolidin-2-ones were prepared by 5-exo-trig iodocyclisation from allylic fluorides bearing a pending nitrogen nucleophile. These bench-stable precursors were made accessible upon electrophilic fluorination of the

Full text of DOI:10.1002/chem.201201576

FULL PAPER
DOI: 10.1002/chem.201201576
Synthesis of 3-Fluoropyrrolidines and 4-Fluoropyrrolidin-2-ones from Allylic
Fluorides
Lorraine E. Combettes, Marie Schuler, Rakesh Patel, Baltasar Bonillo, Barbara Odell,
Amber L. Thompson, Tim D. W. Claridge,* and Vꢀronique Gouverneur*[a]
Abstract: Various 3-fluoropyrrolidines
and 4-fluoropyrrolidin-2-ones were pre-
pared by 5-exo-trig iodocyclisation
from allylic fluorides bearing a pending
nitrogen nucleophile. These bench-
stable precursors were made accessible
upon electrophilic fluorination of the
corresponding allylsilanes. The pres-
ence of the allylic fluorine substituent
induces syn-stereocontrol upon iodo-
cyclisation with diastereomeric ratios
ranging from 10:1 to > 20:1 for all N-
tosyl-3-fluoropent-4-en-1-amines and
amides. The sense and level of stereo-
control is strikingly similar to the cor-
responding iodocyclisation of structur-
ally related allylic fluorides bearing
pending oxygen nucleophiles. These re-
sults suggest that the syn selectivity ob-
served upon ring closure involves I2–p
complexes with the fluorine positioned
inside.
Keywords: diastereoselectivity
·
fluorine · iodocyclization · lactams ·
nitrogen heterocycles
Introduction
has been extended to organocatalysis based on the observa-
tion that stereoelectronic and electrostatic effects resulting
from fluorine substitution can dramatically improve catalyst
performance for a range of important transformations.[8] In
this general context, our aim was to prepare variously sub-
stituted fluorinated pyrrolidines for conformational analysis
in solution using a combination of 19F–1H scalar couplings
and a 19F–1H 1D HOESY sequence optimised for small mol-
The majority of the drugs produced in the pharmaceutical
industry have structural features inspired by natural prod-
ucts. Among them, pyrrolidines hold a prominent position
as key motifs or sub-motifs of many alkaloids such as pyrro-
lizidines and indolizidines.[1] Since fluorine substitution is a
well-established strategy to improve the efficacy of lead
compounds,[2] fluorinated pyrrolidines have become privi-
leged targets for drug-discovery and in medicinal chemistry
research.[3] For example, a recent study of OꢀHagan and co-
1
ecules, from which H–19F internuclear separation estimates
could be extracted.[9] Known syntheses of 3-fluoropyrroli-
dines include the nucleophilic fluorination of hydroxylated
precursors,[10] the ring opening of suitably functionalised 3,4-
epoxypyrrolidines with fluoride sources,[11] 1,3-dipolar cyclo-
additions of 3-fluoroacrylates,[12] and the rearrangement of
2-hydroxyalkylazetidines.[13] We opted for the development
of a new synthetic route using allylic fluorides as starting
materials. In 2008, our group reported that fluorinated tetra-
hydrofurans and g-lactones are accessible upon iodocyclisa-
tion of allylic fluorides bearing a pending oxygen nucleo-
phile.[14] In these reactions, the fluorine substituent is an ef-
fective syn-directing group, a stereochemical preference dic-
tated by an “inside fluoro effect”. Building on these results,
we questioned whether this effect would operate with simi-
lar efficiency for amino-functionalised allylic fluorides un-
dergoing iodoaminations. An electrophilic fluorination of al-
lylsilanes was envisaged to prepare the necessary allylic flu-
orides (Figure 1).[15]
This study therefore started with the synthesis of allylsi-
lanes bearing a nitrogen nucleophile suitably protected to
undergo both electrophilic fluorination and the programmed
cyclisation. The feasibility and the stereochemical aspects of
the halocyclisation were examined on a range of precursors
leading to fluorinated pyrrolidines and lactams; further
functionalisation of a selected ring-closed product is also
presented leading to a fluoroindolizidine.
À
workers revealed that C F bond incorporation into a pe-
ripheral pyrrolidine of a G-quadruplex DNA ligand imposes
a distinct pyrrolidine ring conformation with respect to the
non-fluorinated analogue and changes the mode of bind-
ing.[4] Perhaps, one of the most striking illustrations of the
impact of fluorine substitution on N-heterocycles is in the
use of diastereomeric fluoroprolines towards the develop-
ment of hyperstable collagen materials.[5] The preferential
Cg-exo puckering and Cg-endo puckering of (4R-F)Pro and
(4S-F)Pro respectively, was documented 40 years ago[6] and
the increased stability of synthetic collagen triple helix bear-
ing (4R-F)Pro was reported in 1998.[7] More recently, the
use of fluorinated pyrrolidines (and structural derivatives)
[a] L. E. Combettes, Dr. M. Schuler, R. Patel, B. Bonillo, Dr. B. Odell,
Dr. A. L. Thompson, Dr. T. D. W. Claridge, Prof. V. Gouverneur
Chemistry Research Laboratory
University of Oxford
12 Mansfield Road, Oxford OX1 3TA (UK)
Fax : (+44)1865-285002
E-mail: tim.claridge@chem.ox.ac.uk
veronique.gouverneur@chem.ox.ac.uk
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201201576.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Scheme 1. Synthesis of 1a and 1b: a) allyltrimethylsilane, Grubbs second-
generation catalyst (5 mol%), 1,4-benzoquinone, CH2Cl2, reflux, 3 days;
b) Selectfluor, NaHCO3, MeCN, 48 h.
Figure 1. Synthesis of fluoropyrrolidines from allylsilanes.
Results and Discussion
protected allylic fluorides. The stereochemistry of the major
diastereomer was assigned based on NOE/HOESY experi-
ments indicating a syn relationship between the fluorine and
the iodomethyl substituents.[9] X-ray analysis provided un-
equivocal evidence of the syn pyrrolidine being formed pref-
erentially.[9,16]
Preliminary validation: The model N-tosylated allylic fluo-
ride 1a was synthesised to study the feasibility of the halo-
cyclisation.[16] Cross metathesis of N-3-butenyl-4-methylben-
zenesulfonamide with allyltrimethylsilane under Ru catalysis
(second-generation Grubbs catalyst) afforded 2a as a mix-
ture of E/Z isomers (2:1). It was necessary to conduct this
reaction in the presence of benzoquinone to minimise olefin
migration, a process leading to the truncated side product
3a.[17] Compound 2a was subjected to fluorination in aceto-
nitrile upon treatment with 1-chloromethyl-4-fluoro-1,2-
Scope and limitation: Having established that the iodocycli-
sations of allylic fluorides 1a and 1b are feasible and superi-
or for 1a, we examined next the scope of this reaction with
a series of racemic compounds 1c–i selected based on diver-
sity in terms of substitution patterns. Compounds 1h and 1i
serve to probe the reactivity of substrates featuring the N-
nucleophile one methylene group closer or further apart
from the double bond. Compound 1g is a potential precur-
sor of 4-fluoropyrrolidin-2-one (Figure 2).
Several synthetic routes were applied to synthesise 1c–i.
A metathesis reaction was the key assembling step to access
most silylated precursors. The ring-closing metathesis of
readily accessible alkenyloxysilanes followed by ring open-
ing with MeLi at À788C in THF led to the formation of the
homoallylic alcohols 6 and 7. Mesylation followed by nucle-
ophilic substitution with tosylamine afforded 2c and 2d. The
electrophilic fluorination was performed in acetonitrile at
room temperature in the presence of NaHCO3. This reac-
diazoniabicycloACHTUNGTRENNUNG[2.2.2]-octane bis(tetrafluoroborate) (Select-
fluor) in the presence of NaHCO3 leading to compound 1a
in 63% yield. A similar protocol delivered the N-Boc-pro-
tected allylic fluoride 1b with comparable yields
(Scheme 1).
The iodoamination of 1a was conducted at room tempera-
ture with I2 in various solvents (Table 1, entries 1–4). The
use of CH2Cl2 and aqueous NaHCO3 was optimal leading to
the 3-fluoro-2-iodomethyl pyrrolidine 4a in 88%. The dia-
stereomeric ratio (d.r.) was 14:1 in favour of the syn isomer
(Table 1, entry 1). This reaction was superior to the ones
performed in CH3CN, Et2O, or acetone both in terms of
yield and diastereocontrol (Table 1, entries 2–4).
As an alternative to I2, N-iodosuccinimide (NIS)
was a suitable I+ source affording 4a in 82% yield
with a d.r. of 12:1 when the reaction was performed
Table 1. Iodo- and bromoamination of allylic fluorides 1a and 1b.
in CH2Cl2/NaHCO3 (Table 1, entry 5). N-Bromosuc-
cinimide (NBS) was slightly less efficient. Under
the best reaction conditions, the brominated fluoro-
pyrrolidine 5a was formed in 71% conversion as a
mixture of diastereomers (8:1; Table 1, entries 7
and 8). The nature of the N-protection had a clear
influence on the reaction efficiency, the more acidic
NHTos nucleophile being more reactive than
NHBoc (NHTos, pKa (DMSO)ꢀ16; NHBoc, pKa
(DMSO)ꢀ25).[18,19] Under the best reaction condi-
tions, the cyclisation of the N-Boc-protected allylic
fluoride 1b did occur but was significantly less effi-
cient as reflected by the yield (42%) and d.r. (4:1;
Table 1, entry 9). The lower level of stereocontrol
for 1a versus 1b is likely to be due to the different
rotational barrier of the carbamate versus the aryl-
sufonamide group.[20] This result prompted us to
Entry
1
Reagent
Solvent
Product
Yield
[%][a]
d.r.
ACHTUNGTRENNUNG
(syn/anti)[b]
1
2
3
4
5
6
7
8
9
1a
1a
1a
1a
1a
1a
1a
1a
1b
I2
I2
I2
I2
NIS
NIS
NBS
NBS
I2
CH2Cl2/NaHCO3 (1:1)
CH3CN
Et2O
acetone
CH2Cl2/NaHCO3 (1:1)
CH3CN
CH2Cl2/NaHCO3 (1:1)
CH3CN
CH2Cl2/NaHCO3 (1:1)
4a
4a
4a
4a
4a
4a
5a
5a
4b
88
50
51
39
82
56
71[c]
61
42
14:1
10:1
10:1
8:1
12:1
12:1
8:1
11:1
4:1
[a] Isolated yields. [b] Determined by 19F NMR analysis of the crude product. [c] Con-
focus our scope and limitation study on N-tosyl- version.
&
2
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Synthesis of 3-Fluoropyrrolidines and 4-Fluoropyrrolidin-2-ones
FULL PAPER
Scheme 4. Synthesis of 1e by cross metathesis and electrophilic fluorina-
tion: a) lithium diisopropylamide (LDA), THF, À788C, 1 h; b) allyltri-
Figure 2. Selection of allylic fluorides for iodo-N-cyclisation.
AHCTUNGERTGmNNUN ethylsilane, Hoveyda–Grubbs second-generation catalyst (3 mol%),
CH2Cl2, reflux, 24 h; c) LDA, MeI, THF, À788C, 1 h; d) 1,1’-carbonyldi-
AHCTUNGTRENNUNG
16 h; f) TsCl, NEt3, 4-dimethylaminopyridine (DMAP), CH2Cl2, RT, 3 h;
g) Selectfluor, NaHCO3, RT, 48 h.
Scheme 5. Synthesis of 1g by cross metathesis and electrophilic fluorina-
tion: a) p-tosyl isocyanate, NEt3, THF, 1 h; b) Selectfluor, NaHCO3,
CH3CN, 48 h.
Scheme 2. Synthesis of 1c by ring-closing metathesis (RCM) and electro-
philic fluorination: a) NEt3, CH2Cl2, 08C, 15 h; b) Hoveyda–Grubbs
second-generation catalyst (2 mol%), CH2Cl2, reflux, 1.5 h; c) MeLi,
THF, À788C, 4 h; d) MsCl, NEt3, CH2Cl2, 08C, 2.5 h; e) NH2Ts, KOH,
DMF, 1208C, 3 h; f) Selectfluor, CH3CN, NaHCO3, RT, 24 h.
an efficient transformation delivering 1g in 87% yield
(Scheme 5).
The synthesis of syn-1 f and anti-1 f began with the allyla-
tion of N-(2-benzyloxy)ethylidene)tosylamine. Cross meta-
thesis of the resulting homoallylic N-tosylamine with allyltri-
methylsilane gave access to 2 f as a mixture of E/Z geomet-
rical isomers (ꢀ1:1). Fluorodesilylation led to the desired
allylic fluoride 1 f in 85% yield (Scheme 6).
Scheme 3. Synthesis of 1d by RCM and electrophilic fluorination:
a) NEt3, CH2Cl2, 08C, 15 h; b) Hoveyda–Grubbs second-generation cata-
lyst (2 mol%), CH2Cl2, reflux, 1.5 h; c) MeLi, THF, À788C, 4 h; d) MsCl,
NEt3, CH2Cl2, 08C, 2 h 30 min; e) NH2Ts, KOH, DMF, 1208C, 3 h; f) Se-
lectfluor, CH3CN, NaHCO3, RT, 24 h.
Scheme 6. Synthesis of syn-1 f and anti-1 f by cross metathesis and elec-
trophilic fluorination: a) TosNH2, sodium toluene sulfimide, HCO2H,
H2O, 24 h; b) allylmagnesium bromide (1m in Et2O), CH2Cl2, À788C,
5 h; c) allyltrimethylsilane, Grubbs second-generation catalyst (5 mol%),
1,4-benzoquinone, CH2Cl2, reflux, 3 days; d) Selectfluor, NaHCO3,
CH3CN, 48 h.
tion was quantitative for 2c giving 1c in 99% yield; 1d was
isolated in 57% yield (Scheme 2 and Scheme 3).
Compounds 1e-i were prepared by using a cross metathe-
sis and an electrophilic fluorination as key steps. The synthe-
sis of the gem-dimethylated trimethylallylsilane 8 com-
menced with the deconjugative isomerisation of tiglic acid.
Cross metathesis of the resulting b,g-unsaturated carboxylic
acid with allyltrimethylsilane followed by methylation led to
the desired 2,3-dimethyl-5-trimethylsilylpent-3-enoic acid 8.
The conversion of 8 into 2e required amide formation, re-
duction and tosylation. The electrophilic fluorination of 2e,
carried out under standard conditions, was high yielding
(84%, Scheme 4).
As expected,[21] the presence of the remote stereogenic
centre in 2 f did not induce stereocontrol upon fluorination
and two diastereomers (1:1) were formed; these could be
separated by silica gel chromatography.
The synthesis of allylic fluorides 1h–i is straightforward
applying the standard cross metathesis/fluorination sequence
(Scheme 7).
Notably, all allylic fluorides 1a–i were found to be bench
stable and could be kept without decomposition at room
temperature for an extended period of time.
The fluorination of amide 2g (prepared in one step upon
treatment of carboxylic acid 8 with tosylisocyanate) is also
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

T. D. W. Claridge, V. Gouverneur et al.
758C). The yield of 4h remained low, just under 30%, and
the selectivity in favour of the syn-fluoropyrrolidine was
moderate (10:1; Table 2, entry 9).[24] Finally, compound 1i
could not be cyclised. Under mild conditions (A or B), the
starting material remained intact with no evidence of 6-exo-
trig ring closure. The diene 11 resulting from elimination
was observed under conditions B (Table 2, entry 10). This
result is consistent with the observation that the structurally
Scheme 7. Synthesis of 1h and 1i via cross-metathesis and electrophilic
fluorination: a) allyltrimethylsilane, Hoveyda–Grubbs second-generation
catalyst (5 mol%), 1,4-benzoquinone, CH2Cl2, reflux, 3 days; b) Select-
fluor, NaHCO3, CH3CN, 48 h.
Allylic fluorides 1c–i were subjected to iodocycli-
sation at room temperature with I2 either in aceto-
nitrile in the absence of base (conditions A) or in
Table 2. Iodocyclisation of allylic fluorides 1a–i.
Entry
1
Product
Conditions[a] Yield d.r.
[%][b]
(syn/anti)[c]
ACHTUNGTRENNUNG
CH2Cl2/aqueous
NaHCO3
(conditions B)
(Table 2).[16] The reaction tolerates gem-disubstitut-
ed alkenes as well as quaternary fluorinated allylic
carbon atoms. The iodocyclisation of allylic fluo-
rides 1c and 1d led to the fluoropyrrolidines 4c and
4d in 85% and 65% yield respectively (Table 2, en-
tries 3 and 4). The diastereomeric ratio was found
to be >8:1 consistently in favour of the syn isomer.
For 1c, the level of diastereocontrol was dependent
on the solvent, better results being obtained in
CH3CN (d.r.=12:1 versus 8:1), but at the expense
of reduced chemical yield (from 85 to 42%)
(Table 2, entry 3). Using acetonitrile, compound 4d
was formed in 65% yield as a 10:1 mixture of dia-
stereomers (Table 2, entry 4). The cyclisation of
compound 1e featuring a gem-dimethyl group was
quantitative and proceeded with complete stereo-
control (d.r.>20:1; Table 1, entry 5).[22] The struc-
turally related gem-disubstituted amide 1g cyclised
successfully under conditions A or B (Table 2,
entry 8). In CH2Cl2/NaHCO3, two separable prod-
ucts 4g and 9 (overall yield 86%) resulting from
both N- and O-cyclisation were formed.[23] Structur-
al assignment was consistent with IR data (N-cycli-
sation: 1738 cmÀ1; O-cyclisation: 1650 cmÀ1) and un-
ambiguously confirmed based on single-crystal X-
ray analysis secured for both structural isomers 4g[9]
and 9.[16] In the 13C NMR spectra, the chemical shift
for the carbon bearing the iodomethyl group was
very distinctive (N-cyclisation: d=61.9 ppm; O-cyc-
lisation: d=85.2 ppm). Both cyclisations products
led to a single syn isomer. In acetonitrile, the lac-
tone syn-10 (d.r.>20:1) was isolated in 84% yield
with no evidence for the lactam product in the
crude mixture. The cyclisation of syn-1 f and anti-1 f
led to the fluoropyrrolidines syn,anti-4 f and
syn,syn-4 f, respectively in good yields and with
comparable control over selectivity (>13:1)
(Table 2, entries 6 and 7). The observed sense of di-
astereocontrol is therefore imposed by the allylic
fluorine substituent, not the stereogenic centre
proximal to the N-nucleophile. The homoallylic
amine 1h underwent 5-endo-trig cyclisation under
more forceful conditions (conditions C: CH3CN,
A
B
50
88
10:1
14:1
1
1a
2
1b
B
42
4:1
A
B
42
85
12:1
8:1
3
4
5
1c
1d
1e
A
B
65
97
10:1 (>20:1)[d]
>20:1
6
7
8
syn-1 f
anti-1 f
1g
B
B
78
95
17:1 (>20:1)[d]
13:1
–
[e]
A
B
–
62[f]
>20:1
9
1h
1i
C
B
29
0
10:1 (>20:1)[d]
10
–
[a] Conditions A: I2, CH3CN, 16 h, RT; Conditions B: I2, CH2Cl2/NaHCO3 (aq) (1:1),
16 h, RT; Conditions C: CH3CN, 758C. [b] Isolated yields. [c] Determined by 19F NMR
analysis of the crude product. [d] d.r. after purification. [e] The reaction yielded the
lactone 10 in 84% yield (d.r.>20:1); no trace of 4g in the reaction mixture. [f] This
reaction also gave 24% yield of 9 (d.r.>20:1).
&
4
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Synthesis of 3-Fluoropyrrolidines and 4-Fluoropyrrolidin-2-ones
FULL PAPER
related N-tosyl-4-hydroxy-5-hexenylamine did not undergo
cyclisation with various halogen electrophiles.[18] The relative
stereochemistry of compound 4a–h was unambiguously as-
signed by a combination of NOE/HOESY experiments and
X-ray diffraction analysis when possible.[9]
To demonstrate that further functionalisation of 3-fluoro-
2-iodomethyl-N-tosyl-pyrrolidines is possible, we trans-
formed the representative fluoropyrrolidine syn,syn-4 f
(d.r.=13:1) into the corresponding indolizidine 13
(Scheme 8). The union of methyl acrylate with 4 f upon
The preferential syn selectivity observed for all 5-exo-trig
cyclisations is in accordance with the fluorine-directed dia-
stereoselective iodocyclisation of structurally related allylic
fluorides bearing a pending oxygen nucleophile (alcohol or
carboxylic acid) (Figure 3).[14] The diastereoselectivities ob-
Scheme 8. Synthesis of indolizidine 13: a) CuI, Zn, methyl acrylate, H2O/
MeOH, sonicated, 45 min; b) Na/Hg (5%), Na2HPO4, MeOH, 3 h.
treatment with CuI and zinc in methanol/water led to com-
pound 12 with no erosion of d.r.[25] The reaction of 12 with
Na/Hg (5%) and Na2HPO4 in MeOH allowed for reductive
deprotection of the amine and its subsequent cyclisation.
The desired indolizidine 13 was isolated in 52% (d.r.>
12:1).
Conclusion
Figure 3. Iodocyclisation of allylic fluoride with O- versus N-nucleophile.
This study presents the first examples of iodocyclisation
with functionalised allylic fluorides bearing nitrogen nucleo-
philes. The 5-exo-trig cyclisation of variously substituted N-
tosylated 3-fluoro-4-pentenylamines is a suitable transforma-
tion to prepare 3-fluoropyrrolidines in high yields (up to
97%) and with good to excellent level of diastereocontrol
(up to d.r.>20:1). The method can be extended to the prep-
aration of 4-fluoropyrrolidinones. The sense and level of dia-
stereocontrol observed upon iodoamination is imposed by
the fluorine substituent and influenced by the N-protecting
group. The more acidic N-tosylamine nucleophiles were
found to be superior nucleophiles (yield and d.r.) in compar-
ison with the corresponding N-Boc carbamates. The striking
similarities observed for allylic fluorides bearing O- and N-
tosyl nucleophiles with respect to stereocontrol, suggest that
the cyclisations studied herein are likely to involve I2–p
complexes with the fluorine adopting the inside position.
The 5-endo-trig cyclisation of N-(p-tolylsulfonyl)-3-fluoro-4-
butenylamine was found to be successful but low yielding.
In contrast, 3-fluoropiperidines could not be accessed using
this methodology. The necessary allylic fluorides were pre-
pared reliably by electrophilic fluorination of the corre-
sponding allylsilanes. As presented, the synthesis of the
starting allylsilanes requires multiple steps and this will need
optimisation for large-scale operation. However, with the
appearance of new transition-metal catalysed (asymmetric)
allylic fluorinations, one can now envisage accessing (enan-
tiopure) aminated allylic fluorides from the corresponding
allylic alcohols or chlorides with fluoride.[26] These new
routes could potentially speed up the access to this impor-
tant class of compounds that are 3-fluoropyrrolidines.
served are strikingly similar between these two series of
compounds, a safe indicator that the same effects are re-
sponsible for the observed sense of stereocontrol.
The reaction appears to be under kinetic control, since
additional experiments indicate that prolonging the reaction
time or heating the reaction mixture did not alter the d.r.
When 1a (d.r. 14:1) was subjected to the reaction condi-
tions, this material was recovered intact after 16 h. It is
therefore reasonable to advance that, by analogy with the
fluorine-directed diastereoselective iodoetherification and
iodolactonisation of allylic fluorides, the preferential syn se-
lectivity of the cyclisations shown in Table 2 arises from the
formation of a loosely associated I2–p complex adopting the
preferential inside fluoro conformation. The proximally ori-
ented tosylated amino group is ideally positioned for back-
side attack on the iodonium ion (Figure 4).
Figure 4. Rationale for syn-preference.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

T. D. W. Claridge, V. Gouverneur et al.
2006, 49, 3614–3627; c) J. Xu, L. Wei, R. J. Mathvink, S. D. Edmond-
son, G. J. Eiermann, H. He, J. F. Leone, B. Leiting, K. A. Lyons, F.
Marsilio, R. A. Patel, S. B. Patel, A. Petrov, G. Scapin, J. K. Wu,
N. A. Thornberry, A. E. Weber, Bioorg. Med. Chem. Lett. 2006, 16,
5373–5377; d) C. G. Caldwell, P. Chen, J. He, E. R. Parmee, B. Leit-
ing, F. Marsilio, R. A. Patel, J. K. Wu, G. J. Eiermann, A. Petrov, H.
He, K. A. Lyons, N. A. Thornberry, A. E. Weber, Bioorg. Med.
Chem. Lett. 2004, 14, 1265–1268; e) B. Hulin, S. Cabral, M. G.
Lopaze, M. A. Van Volkenburg, K. M. Andrews, J. C. Parker,
Bioorg. Med. Chem. Lett. 2005, 15, 4770–4773; f) M. J. Ammirati,
K. M. Andrews, D. D. Boyer, A. M. Brodeur, D. E. Danley, S. D.
Doran, B. Hulin, S. Liu, R. K. McPherson, S. J. Orena, J. C. Parker,
J. Polivkova, X. Qiu, C. B. Soglia, J. L. Treadway, M. A. VanVolken-
burg, D. C. Wilder, D. W. Piotrowski, Bioorg. Med. Chem. Lett.
2009, 19, 1991–1995; g) D. E. Patterson, J. D. Powers, M. LeBlanc,
T. Sharkey, E. Boehler, E. Irdam, M. H. Osterhout, Org. Process
Res. Dev. 2009, 13, 900–906; h) H. Fukushima, A. Hiratate, M. Ta-
kahashi, M. Saito-Hori, E. Munetomo, K. Kitano, H. Saito, Y. Ta-
kaoka, K. Yamamoto, Chem. Pharm. Bull. 2008, 56, 1110–1117.
[4] N. H. Campbell, D. L. Smith, A. P. Reszka, S. Neidle, D. OꢀHagan,
Org. Biomol. Chem. 2011, 9, 1328–1331.
Experimental Section
Representative synthetic procedure—preparation of (Æ)-3-fluoro-2-
(iodomethyl)-1-[(4-methylphenyl)sulfonyl] pyrrolidine (4a): To a solution
of allylic fluoride 1a (100 mg, 0.39 mmol, 1 equiv) in saturated NaHCO3
solution/CH2Cl2 (1:1; 0.13m) in a darkened vessel, was added iodine
(218 mg, 0.86 mmol, 2.2 equiv). The reaction mixture was then stirred
overnight before being quenched by the addition of saturated NaS2O3
solution/saturated NaHCO3 solution (1:1). The aqueous layer was ex-
tracted three times with Et2O, the combined organic phases were washed
with brine, dried over MgSO4, filtered and the solvent was removed in
vacuo. The crude product was purified by flash column chromatography
on silica gel (hexane/Et2O 8:2) to yield 4a (131 mg, 88%, d.r. 14:1) as a
white solid. Data for the major syn diastereoisomer: Rf =0.4 (hexane/
EtOAc 7:3); m.p.=728C; 1H NMR (500 MHz, C6D6) d=0.52 (ddddd,
1H, 3J
1.21 (ddd, 1H, 3J
1.86 (s, 3H), 3.22 (ddd, 1H, 2J
G
G
ACHTUNGTRENNUNG
G
R
ACHTUNGTRENNUNG
G
ACHTUNGTRENNUNG
5.4 Hz), 3.28 (dd, 1H, 3J
G
ACHTUNGTRENNUNG
2J
3J
(H,H)=11.7 Hz, 3J
U
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
3
2
3
[5] a) E. S. Eberhardt, N. Panasik, Jr., R. T. Raines, J. Am. Chem. Soc.
1996, 118, 12261–12266; b) S. K. Holmgren, L. E. Bretscher, K. M.
Taylor, R. T. Raines, Chem. Biol. 1999, 6, 63–70; c) L. E. Bretscher,
C. L. Jenkins, K. M. Taylor, M. L. DeRider, R. T. Raines, J. Am.
Chem. Soc. 2001, 123, 777–778.
3.9 Hz, J
ACHTUNGTRENNUNG(H,H)=3.9 Hz), 4.63 (ddd, 1H, JAHCTUNRTGENNG(UN H,F)=52.2 Hz, JACTHUGNTRENNUGN
3.4 Hz), 6.68 (d, 2H, 3J
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
G
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
[6] J. T. Gerig, R. S. McLeod, J. Am. Chem. Soc. 1973, 95, 5725–5729.
[7] S. K. Holmgren, K. M. Taylor, L. E. Bretscher, R. T. Raines, Nature
1998, 392, 666–667.
[8] a) L. Hunter, Beilstein J. Org. Chem. 2010, 6, 38; b) L. E. Zimmer,
C. Sparr, R. Gilmour, Angew. Chem. 2011, 123, 12062–12074;
Angew. Chem. Int. Ed. 2011, 50, 11860–11871.
[9] L. E. Combettes, P. Clausen-Thue, B. Odell, A. L. Thompson, V.
Gouverneur, T. D. W. Claridge, Chem. Eur. J. 2012, DOI: 10.1002/
chem.201201577.
[10] a) G. Giardina, G. Dondio, M. Grugni, Synlett 1995, 55–57; b) R. P.
Singh, J. M. Shreeve, J. Fluorine Chem. 2002, 116, 23–26; c) L. De-
mange, A. Mꢃnez, C. Dugave, Tetrahedron Lett. 1998, 39, 1169–
1172.
;
383.9931; found: 383.9947. Characteristic data for the minor anti
diastereoisomer: 19F{1H} NMR (377 MHz, CDCl3) d=À176.26 ppm.
X-ray crystallographic analysis: Samples of 4c and 9 were quench cooled
to 150 K using an Oxford Cryosystems Cryostream 600 series open flow
N2 cooling device.[27] Data were collected using a Nonius Kappa-CCD
and intensity data were processed using the DENZO-SMN package.[28]
The structures were solved by direct methods[29] and refined by full-
matrix least squares on F2 using the CRYSTALS suite;[30] hydrogen
atoms were treated in the usual manner.[31] CCDC-896062 (4c) and
CCDC-896063 (9) contain the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[11] a) T. Ayad, Y. Gꢃnisson, S. Broussy, M. Baltas, L. Gorrichon, Eur. J.
Org. Chem. 2003, 2903–2910; b) J. A. Kalow, A. G. Doyle, J. Am.
Chem. Soc. 2010, 132, 3268–3269.
[12] a) B. F. Bonini, F. Boschi, M. C. Franchini, M. Fochi, F. Fini, A. Maz-
zanti, A. Ricci, Synlett 2006, 543–546; b) J. M. Mason, A. S. Murkin,
L. Li, V. L. Schramm, G. J. Gainsford, B. W. Skelton, J. Med. Chem.
2008, 51, 5880–5884.
[13] B. Drouillat, F. Couty, O. David, G. Evano, J. Marrot, Synlett 2008,
1345–1348.
[14] M. Tredwell, J. A. R. Luft, M. Schuler, K. Tenza, K. N. Houk, V.
Gouverneur, Angew. Chem. 2008, 120, 363–366; Angew. Chem. Int.
Ed. 2008, 47, 357–360.
Acknowledgements
We thank the EPSRC (grant no.: EP/F019467/1 to L.E.C.) and the Euro-
pean Community (MEIF-CT-2006–03970) for generous financial support.
[15] a) M. Tredwell, V. Gouverneur, Org. Biomol. Chem. 2006, 4, 26–32;
b) S. Thibaudeau, V. Gouverneur, Org. Lett. 2003, 5, 4891–4893;
c) V. Gouverneur, B. Greedy, Chem. Eur. J. 2002, 8, 766–772; d) B.
Greedy, J.-M. Paris, T. Vidal, V. Gouverneur, Angew. Chem. 2003,
115, 3413–3416; Angew. Chem. Int. Ed. 2003, 42, 3291–3294.
[16] For experimental details, see the Supporting Information.
[17] S. H. Hong, D. P. Sanders, C. W. Lee, R. H. Grubbs, J. Am. Chem.
Soc. 2005, 127, 17160–17161.
[1] a) J. W.-H. Li, J. C. Vederas, Science 2009, 325, 161–165; b) F. E.
Koehn, G. T. Carter, Nat. Rev. Drug Discovery 2005, 4, 206–220;
c) T. J. Donohoe, C. J. R. Bataille, G. H. Churchill, Annu. Rep. Prog.
Chem. Sect. B 2006, 102, 98–122; d) J. R. Liddell, Nat. Prod. Rep.
2002, 19, 773–781; e) D. Enders, T. Thiebes, Pure Appl. Chem. 2001,
73, 573–578; f) D. OꢀHagan, Nat. Prod. Rep. 2000, 17, 435–446;
g) J. P. Michael, Nat. Prod. Rep. 2000, 17, 579–602; h) J. E. Saxton,
Nat. Prod. Rep. 1997, 14, 559–590.
[18] Y. Tamaru, S. Kawamura, T. Bando, K. Tanaka, M. Hojo, Z. Yoshi-
da, J. Org. Chem. 1988, 53, 5491–5501.
[19] For a comprehensive compilation of Bordwell pKa data, see: http://
www.chem.wisc.edu/areas/reich/pkatable/index.htm.
[20] a) C. Cox, T. Lectka, J. Org. Chem. 1998, 63, 2426–2427; b) P. V.
Bharatam, A. A. Gupta, D. Kaur, Tetrahedron 2002, 58, 1759–1764;
c) J. Elguero, P. Goya, I. Rozas, J. Catalꢄn, J. L. G. De Paz, J. Mol.
Struct.: THEOCHEM 1989, 184, 115–129; d) R. D. Bindal, J. T.
Golab, J. A. Katzenellenbogen, J. Am. Chem. Soc. 1990, 112, 7861–
7868.
[2] a) S. Purser, P. R. Moore, S. Swallow, V. Gouverneur, Chem. Soc.
Rev. 2008, 37, 320–330; b) K. Mꢂller, C. Faeh, F. Diederich, Science
2007, 317, 1881–1886.
[3] a) F. Goossens, G. VanHoof, I. De Meester, K. Augustyns, M.
Borloo, D. Tourwe, A. Haemers, S. Scharpꢃ, Eur. J. Biochem. 1997,
250, 177–183; b) S. D. Edmondson, A. Mastracchio, R. J. Mathvink,
J. He, B. Harper, Y.-J. Park, M. Beconi, J. Di Salvo, G. J. Eiermann,
H. He, B. Leiting, J. F. Leone, D. A. Levorse, K. Lyons, R. A. Patel,
S. B. Patel, A. Petrov, G. Scapin, J. Shang, R. S. Roy, A. Smith, J. K.
Wu, S. Xu, B. Zhu, N. A. Thornberry, A. E. Weber, J. Med. Chem.
&
6
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Synthesis of 3-Fluoropyrrolidines and 4-Fluoropyrrolidin-2-ones
FULL PAPER
[21] M. Tredwell, K. Tenza, M. C. Pacheco, V. Gouverneur, Org. Lett.
2005, 7, 4495–4497.
[22] a) R. M. Beesley, C. K. Ingold, J. F. Thorpe, J. Chem. Soc. Trans.
1915, 107, 1080–1106; b) C. K. Ingold, J. Chem. Soc. Trans. 1921,
119, 305–329; c) C. K. Ingold, S. Sako, J. F. Thorpe, J. Chem. Soc.
Trans. 1922, 121, 1177–1198.
[23] M. Fujita, O. Kitagawa, T. Suzuki, T. Taguchi, J. Org. Chem. 1997,
62, 7330–7335.
[24] The preferential syn selectivity is consistent with an envelope reac-
tive conformation featuring a gauche interaction between the fluo-
rine substituent and the vicinal nitrogen nucleophile.
[25] a) R. M. Suarez, J. P. Sestelo, L. A. Sarandeses, Synlett 2002, 1435–
1438; b) J. L. Luche, C. Allavena, C. Petrier, C. Dupuy, Tetrahedron
Lett. 1988, 29, 5373–5374.
[26] a) C. Hollingworth, A. Hazari, M. N. Hopkinson, M. Tredwell, E.
Benedetto, M. Huiban, A. D. Gee, J. M. Brown, V. Gouverneur,
Angew. Chem. 2011, 123, 2661–2665; Angew. Chem. Int. Ed. 2011,
50, 2613–2617; b) M. H. Katcher, A. Sha, A. G. Doyle, J. Am.
Chem. Soc. 2011, 133, 15902–15905; c) J. J. Topczewski, T. J.
Tewson, H. M. Nguyen, J. Am. Chem. Soc. 2011, 133, 19318–19321;
d) C. Hollingworth, V. Gouverneur, Chem. Commun. 2012, 48,
2929–2942.
[27] J. Cosier, A. M. Glazer, J. Appl. Crystallogr. 1986, 19, 105.
[28] Z. Otwinowski, W. Minor, Processing of X-ray Diffraction Data Col-
lected in Oscillation Mode, Methods Enzymol. (Eds.: C. W. Carter,
R. M. Sweet), Academic Press, New York, 1997, p. 276.
[29] A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, M. C.
Burla, G. Polidori, M. Camalli, J. Appl. Crystallogr. 1994, 27, 435.
[30] a) P. W. Betteridge, J. R. Carruthers, R. I. Cooper, K. Prout, D. J.
Watkin, J. Appl. Crystallogr. 2003, 36, 1487; b) A. L. Thompson, D. J.
Watkin, J. Appl. Crystallogr. 2011, 44, 1017–1022.
[31] R. I. Cooper, A. L. Thompson, D. J. Watkin, J. Appl. Crystallogr.
2010, 43, 1100–1107.
Received: May 6, 2012
Published online: && &&, 0000
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

T. D. W. Claridge, V. Gouverneur et al.
Synthetic Methods
L. E. Combettes, M. Schuler, R. Patel,
B. Bonillo, B. Odell, A. L. Thompson,
T. D. W. Claridge,*
V. Gouverneur* ............... &&&& — &&&&
The 5-exo-trig cyclisation of N-pro-
tected allylic fluorides leads to fluoro-
pyrrolidin(on)es in yields up to 97%.
The fluorine substituent is an efficient
stereodirecting group imposing a syn
relationship with the newly formed
iodomethylated stereogenic centre (see
scheme). The striking similarities, with
respect to stereocontrol, observed for
allylic fluorides bearing O- and N-tosyl
nucleophiles indicate that the cyclisa-
tions studied herein involve I2–p com-
plexes with the fluorine adopting the
inside position.
Synthesis of 3-Fluoropyrrolidines and
4-Fluoropyrrolidin-2-ones from Allylic
Fluorides
&
8
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!