Stereospecific benzylic dehydroxyfluorination reactions using Bio's TMS-amine additive approach with challenging substrates

Article abstract of DOI:10.1016/j.tetlet.2010.08.104

Reactions involving the conversion of a benzylic alcohol into a benzylic fluoride using RSF₃ reagents are notoriously difficult to achieve with high stereochemical inversion (S_N2 reaction) due to competing dissociative S_N1 reaction processes. This Letter develops the methodology of Bio et al., and reports that the addition of a preformed 1:TMS-amine 1:RSF₃ (fluorination reagent) complex as the reagent in these reactions significantly suppresses the S_N1 process and promotes a highly stereospecific reaction generating benzylic fluorination products of high %ee.

Full text of DOI:10.1016/j.tetlet.2010.08.104

Tetrahedron Letters 51 (2010) 5795–5797
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Stereospecific benzylic dehydroxyfluorination reactions using
Bio’s TMS-amine additive approach with challenging substrates
⇑
Stefano Bresciani, David O’Hagan
EastCHEM and Centre for Biomolecular Sciences, School of Chemistry, University of St Andrews, North Haugh, St Andrews, Fife, KY16 9ST, United Kingdom
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 June 2010
Revised 27 July 2010
Accepted 31 August 2010
Available online 6 September 2010
Reactions involving the conversion of a benzylic alcohol into a benzylic fluoride using RSF3 reagents are
notoriously difficult to achieve with high stereochemical inversion (SN2 reaction) due to competing dis-
sociative SN1 reaction processes. This Letter develops the methodology of Bio et al., and reports that the
addition of a preformed 1:TMS-amine 1:RSF3 (fluorination reagent) complex as the reagent in these reac-
tions significantly suppresses the SN1 process and promotes a highly stereospecific reaction generating
benzylic fluorination products of high %ee.
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
ment in a medicinal chemistry environment. Although mechanistic
details remain speculative regarding the influence of the TMS-
The introduction of fluorine into organic molecules is a widely
recognised strategy for imparting unique properties to perfor-
mance molecules,1–3 and particularly molecules going through
pharmaceutical development programmes have benefited dramat-
ically from the introduction of fluorine.4 An increasing range of re-
agents are being developed for the introduction of C–F bonds and
catalytic asymmetric fluorination strategies have attracted signifi-
cant attention,5–7 however the stereospecific conversion of an
enantiomerically pure alcohol into its corresponding organic fluo-
ride remains attractive due to the wide availability of appropriate
enantiopure alcohols. However, such dehydroxyfluorination reac-
tions are a considerable challenge, particularly if the alcohol is ben-
zylic and prone to an SN1 reaction course. (Diethylamino)sulfur
trifluoride (DAST)8 has been widely employed for such transforma-
tions, although modified reagents of this class such as Deoxo-
Fluor™,9 have been developed which are more stable to higher
reaction temperatures. Most recently, Fluolead™ has been intro-
duced10 for dehydroxyfluorination reactions, among other trans-
formations. The challenging aspect of these reagents involves
conducting dehydroxyfluorination reactions with high stereointeg-
rity (inversion of configuration) from a precursor enantiopure alco-
hol. Very little has been achieved to address this issue. Most
significantly, however Bio et al. reported11 in 2008 that the stereo-
specificity of the DAST-mediated dehydroxyfluorination of 1a and
1b was considerably improved by the addition of N-(trimethyl-
silyl)morpholine (3) or N-(trimethylsilyl)diethylamine to the reac-
tion as illustrated in Scheme 1.
amines, a working hypothesis suggests that, for example, a mor-
pholino intermediate such as 5 (Scheme 2), is less prone to an
SN1 dissociative reaction than the analogous intermediate 6 gener-
ated during a straightforward DAST reaction.
In the reaction with 1 (Scheme 1) the inductive effect of the
ortho-aryl fluorine is anticipated to assist in suppressing a dissocia-
tive SN1 reaction, and indeed without additive, the reaction gener-
ated a product of 50% ee, considerably higher than other benzylic
alcohols.11 We found that similar reactions of DAST with (R)-1-
phenylethanol (7) and methyl (R)-mandelate (9) (Tables 1 and 2)
were more prone to the SN1 reaction course and gave the corre-
sponding dehydroxyfluorination products with much lower %ees
(7% ee and 8% ee, respectively).
F
OH
O
O
NHR
N
1a = Cbz 1b = Boc
SiMe3
3
additive/reagent CH2Cl2
F
F
N
O
SiMe3
4
NHR
The %ees increased from 50% ee to 96–99% ee depending on the
additive. These observations were made during process develop-
DAST/TMS-NEt2
Deoxo-FluorTM/TMS-morpholine
99 %ee 2a
96 %ee 2b
Scheme 1. Previous observation of increased stereospecificity in the dehydroxy-
fluorination reaction of ortho-fluorobenzylic alcohols 1a and 1b with TMS-amine
additives.11 TMS-amines 3 and 4 used in this study.
⇑
Corresponding author. Tel.:+44 1334650708; fax: +44 1334463800.
E-mail address: do1@st-andrews.ac.uk (D. O’Hagan).
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.08.104

5796
S. Bresciani, D. O’Hagan / Tetrahedron Letters 51 (2010) 5795–5797
N lone pair
donation
Table 2
R
%Ees after reaction of (R)-9 with various fluorinating reagents and the TMS-amine
additives 3 and 4
R
F attractive
F
R N
F
R N
S
S
N
O
O
F
OH
CO2Me
F
O
additive/reagent
CH2Cl2, 24 h, RT
R'
F-
R'
F-
CO2Me
10
6
5
9
less dissociative SN2
(with additive 3)
more dissociative SN1
(without additive)
Entry
Additive:reagent (equiv)
%eea (% conv)b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Fluolead™ (1)
3:Fluolead™ (1)
3:Fluolead™ (3)
4:Fluolead™ (1)
4:Fluolead™ (3)
Deoxo-Fluor™ (1)
3:Deoxo-Fluor™ (1)
3:Deoxo-Fluor™ (3)
3:Deoxo-Fluor™ (4)
DAST (1)
3
64
87 (20)
89
94
Scheme 2. Rationale for improved stereospecificity where the amine additive
forms a complex and suppresses the dissociative reaction as shown in 5 against the
better leaving group as illustrated in 6.
23
99 (51)
98 (23)
98
Table 1
%Ees after reaction of (R)-7 with various fluorinating reagents and the TMS-amine
additives 3 and 4
8
3:DAST (1)
3:DAST (2)
3:DAST (3)
3:DAST (4)
99 (31)
98
97 (21)
97
OH
F
additive/reagent
CH2Cl2, 15 h, RT
a
Determined by GC/MS.
Selected conversions by GC/FID.
7
8
b
Entry
Additive:reagent (equiv)
%eea (% conv)b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Fluolead™ (1)
3:Fluolead™ (1)
3:Fluolead™ (3)
4:Fluolead™ (1)
4:Fluolead™ (3)
Deoxo-Fluor™ (1)
3:Deoxo-Fluor™ (1)
3:Deoxo-Fluor™ (3)
4:Deoxo-Fluor™: (1)
4:Deoxo-Fluor™: (3)
DAST (1)
3:DAST (1)
3:DAST (2)
3:DAST (3)
3:DAST (4)
3:DAST (9)
4:DAST (1)
4:DAST (3)
0
57
92 (34)
56
85
13
58
84 (96)
66
59
7
74
83
95 (92)
93
93
54
69
uct to generate (S)-8 of 95% ee, and with an excellent conversion
(Table 1, entry 14). The reaction with Fluolead™ under similar con-
ditions gave a product 8 of 92% ee but with only a modest conver-
sion (34%) (Table 1, entry 3). In the case of methyl (R)-mandelate
(9) reaction conversions were significantly lower due to the elec-
tron-withdrawing nature of the ester group, and without the addi-
tive the %ees are very low. So this is a particularly challenging
substrate. However, the transformations with all three reagents
demonstrated a significant improvement in stereospecificity, in
this case with only 1 equiv of additive/reagent with Deoxo-Fluor™
(Table 2, entry 7) and DAST (Table 2, entry 11) starting from a very
low %ee (Table 2, entries 1, 6 and 10). The data is presented fully in
Table 2. For Fluolead™, TMS-pyrrolidine 4 was the best additive at
3 equiv (Table 2, entry 5). The addition of one equivalent of the 1:1
complex clearly improved the %ee in these reactions, but optimal
performance appeared to be at about 3 equiv, particularly for 7,
which is the more challenging substrate in terms of suppressing
the SN1 process. This may be due to the equivalent of HF that is
a
Determined by GC/MS.
Selected conversions by GC/FID.
b
With this background we have now explored fluorination of
these substrates (7 and 9) to assess if the addition of TMS-amine
additives can be more generally applied to improve substantially
the stereointegrity of benzylic fluoride products from enantiomeri-
cally pure benzylic alcohols. Three dehydroxyfluorination reagents
of the RSF3 class have been explored with (R)-1-phenylethanol (7)
and methyl (R)-mandelate (9). In both of these cases straightfor-
ward treatment with DAST, Deoxo-Fluor™ and Fluolead™ in
CH2Cl2 lead to fluorinated products with low enantiopurity and
particularly in the case of 7 (see Table 1, entries 1, 6 and 11 for
7, and Table 2, entries 1, 6 and 10 for 9).
TMS-morpholine 3 and TMS-pyrrolidine 4 were explored as the
additives as they are commercially available, and they were added
to the reactions at various levels of equivalence. For the more
challenging substrate, in terms of achieving good stereospecificity,
(R)-1-phenylethanol (7) the data are presented for three reagents
in Table 1. This is a poor reaction in all cases without any additive.
However, in all cases, a substantial improvement in %ee is made
particularly by addition of 3 equiv for either 3 or 4. The most sub-
stantial improvement was found with TMS-morpholine addition to
DAST where the reaction improved from an almost racemic prod-
R2NSF3
+
TMS-N
O
O
R2NSF2-N
O + TMS-F
R'-OH
1st equivalent of adduct
+
R2NSF2-N
+ HF
O
R2NSF-N
R'O
promotes
SN1
R2NSF2-N
O + HF
R2N=SF-N
buffers HF
O + HF2
weak F-
2nd equivalent of adduct
nucleophile
R2N=SF-N
O + F
better F-
R2NSF2-N
O
3rd equivalent of adduct
nucleophile
Scheme 3. Rationale for the improved stereospecificity at 3 equiv of adduct to
benzylic alcohol. A second equivalent buffers HF produced from reaction of the
alcohol. The third equivalent generates a fluoride nucleophile

S. Bresciani, D. O’Hagan / Tetrahedron Letters 51 (2010) 5795–5797
5797
produced when the adduct reacts with the alcohol as illustrated in
Scheme 3. A second equivalent will buffer this HF as hydrogen
bifluoride, and then the third equivalent would generate a more
nucleophilic fluoride ion.
for the minor (R)-enantiomer and tR = 42.5 min for the major (S)-
enantiomer, indicated that the optical purity was 99% ee; ½a D22
ꢂ
+117.5 (c 1.4, CDCl3); mmax (NaCl)/cmꢀ1 3083, 3062, 3027, 2972,
2937, 2877, 1761, 1714, 1448, 1350, 1091; dH (400 MHz, CDCl3)
7.51–7.36 (m, 5H), 5.80 (d, J = 47.5, 1H), 3.78 (s, 3H); dC
(75.5 MHz, CDCl3) 169.1 (d, J = 27.7, C), 134.3 (d, J = 20.6, C),
129.8 (d, J = 2.1, CH), 128.9 (CH), 126.8 (d, J = 6.1, CH), 89.4 (d,
J = 185.6, CH), 52.7 (s, CH3); dF (282 MHz, CDCl3) ꢀ180.3 (d,
J = 47.5 Hz, 1F); dF{H} (282 MHz, CDCl3) ꢀ180.3 (s, 1F); GC/MS (EI-
SIM, +ve) m/z 168 [M+]; HRMS (ESI, +ve) C9H9FO2Na+ requires
m/z 191.0484, found 191.0482.
2. Experimental procedure and physical data for (S)-8 (95% ee,
entry 14 in Table 1)
N-(Trimethylsilyl)morpholine (neat, 447
added dropwise to a solution of DAST (312
l
l, 2.49 mmol) was
l
l, 2.43 mmol) in dry
CH2Cl2 (0.96 ml) at ꢀ78 °C. The resulting solution was stirred at
rt for 2.5 h. The reaction mixture was further cooled to ꢀ78 °C
and a solution of (R)-(+)-1-phenylethanol (7) (90 ll, 0.74 mmol)
Acknowledgement
in dry CH2Cl2 (2.24 ml) was slowly added via cannula. The result-
ing solution was stirred at rt for 15 h. The mixture was then slowly
poured into 15 ml of saturated aqueous NaHCO3 solution. The or-
ganic layer was separated and the aqueous layer was extracted
with CH2Cl2 (3 ꢁ 1.5 ml). The combined organic layers were dried
over anhydrous MgSO4 and the solvent was evaporated under re-
duced pressure. The residue was purified by silica gel chromatog-
raphy (100% pentane) to afford the desired product (S)-8 (65 mg,
71% yield) as a colourless oil. Chiral-phase GC/MS (EI) analysis of
the product, tR = 20.3 min for the major (S)-enantiomer and
tR = 21.3 min for the minor (R)-enantiomer, indicated that the opti-
We thank Dr. T. Umemoto and T. Toyokura, IM&T Research,
(Denver, CO, USA) for providing a sample of Fluolead™, and one
of us (S.B.) thanks EPSRC for the studentship funding.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2010.08.104.
cal purity was 95% ee; ½a D22
ꢂ
+35.6 (c 1.0, CHCl3);
mmax (NaCl)/cmꢀ1
References and notes
3083, 3064, 3034, 2980, 2926, 1451, 1211, 1063; dH (300 MHz,
CDCl3) 7.43–7.29 (m, 5H), 5.64 (dq, J = 47.6, 6.4, 1H), 1.65 (dd,
J = 23.9, 6.4, 3H); dC (75.5 MHz, CDCl3) 141.7 (d, J = 20.0, C), 128.6
(CH), 128.3 (d, J = 1.9, CH), 125.3 (d, J = 6.7, CH), 91.1 (d, J = 167.4,
CH), 23.1 (d, J = 25.3, CH3); dF (282 MHz, CDCl3) ꢀ167.5 (dq,
J = 47.6, 23.9, 1F); dF{H} (282 MHz, CDCl3) ꢀ167.5 (s, 1F); GC/MS
(EI, +ve) m/z 109, 124 [M+]; HRMS (EI, +ve) C8H9F requires m/z
124.0688, found 124.0689.
1. (a) Müller, K.; Boehm, H.-J. Chem. Biol. 2009, 16, 1130–1131; (b) Morgenthaler,
M.; Schweizer, E.; Hoffman-Röder, A.; Benini, F.; Martin, R. E.; Jaeschkee, G.;
Wagner, B.; Fischer, H.; Bendels, S.; Zimmerl, D.; Schneider, J.; Diederich, F.;
Kansy, M.; Müller, K. ChemMedChem. 2007, 2, 1100–1115; (c) Olsen, J.; Seiler,
P.; Wagner, B.; Fischer, H.; Tschopp, T.; Obst-Sander, U.; Banner, D. W.; Kansy,
M.; Müller, K.; Diederich, F. Org. Biomol. Chem. 2004, 2, 1339–1352.
2. O’Hagan, D. Chem. Soc. Rev. 2008, 37, 308–319.
3. Sparr, C.; Schweizer, W. B.; Senn, H. M.; Gilmour, R. Angew. Chem., Int. Ed. 2009,
48, 3065–3068.
4. (a) Haggmann, W. K. J. Med. Chem. 2008, 51, 4359–4369; (b) O’Hagan, D. J.
Fluorine Chem. 2010, in press; (c) Purser, S.; Moore, P. R.; Swallow, S.;
Gouverneur, V. Chem. Soc. Rev. 2008, 37, 320–330.
3. Experimental procedure and physical data for (S)-10 (99% ee,
5. Brunet, V.; O’Hagan, D. Angew. Chem., Int. Ed. 2008, 47, 1179–1182.
6. (a) Enders, D.; Huettle, M. R. M. Synlett 2005, 991–993; (b) Marigo, M.;
Fielenbach, D.; Braunton, A.; Kjærsaard, A.; Jørgenson, K. A. Angew. Chem., Int.
Ed. 2005, 44, 3703–3706; (c) Steiner, D. D.; Mase, N.; Barbas, C. F., III Angew.
Chem., Int. Ed. 2005, 44, 3706–3710; (d) Beeson, T. D.; McMillan, D. W. C. J. Am.
Chem. Soc. 2005, 127, 8826–8828.
7. (a) Reddy, D. S.; Shibata, N.; Nagai, J.; Nakamura, S.; Toru, T.; Kanemasa, S.
Angew. Chem., Int. Ed. 2008, 47, 164–168; (b) Moriya, K. I.; Hamashima, Y.;
Sodeoka, M. Synlett 2007, 1139–1142; (c) Suzuki, T.; Goto, T.; Hamashima, Y. J.
Org. Chem. 2007, 72, 246–250; (d) Cahard, D.; Audouard, C.; Plaquevent, J. C.;
Roques, N. Org. Lett. 2000, 2, 3699–3701; (e) Shibata, N.; Suzuki, E.; Takeuchi, Y.
J. Am. Chem. Soc. 2000, 122, 10728–10729; (f) Hintermann, L.; Togni, A. Angew.
Chem., Int. Ed. 2000, 39, 4342–4359.
8. Middleton, W. J. J. Org. Chem. 1975, 40, 574–578.
9. Lal, G. S.; Pez, G. P.; Pesaresi, R. J.; Prozonic, F. M.; Cheng, H. J. Org. Chem. 1999,
64, 7048–7054.
10. (a) Umemoto, T.; Xu, Y. U.S. Patent 7,265,247 (IM&T Research Inc.); Chem. Abstr.
2007, 147, 322721; (b) Umemoto, T.; Singh, R. P. U.S. Patent 20080039660
(IM&T Research Inc.) Feb 2008; Chem. Abstr. 2008, 148, 239352; (c) Bresciani, S.
J. Fluorine Chem. 2009, 130, 537–543.
11. Bio, M. M.; Waters, M.; Javadi, G.; Song, Z. J.; Zhang, F.; Thomas, D. Synthesis
2008, 891–896.
entry 7 in Table 2)
N-(Trimethylsilyl)morpholine (neat, 298
added dropwise to a solution of Deoxo-Fluor™ (50% in THF,
690
l, 1.62 mmol) in dry CH2Cl2 (2.20 ml) at ꢀ78 °C. The resulting
ll, 1.66 mmol) was
l
solution was stirred at rt for 2.5 h. The reaction mixture was fur-
ther cooled to ꢀ78 °C and a solution of methyl (R)-(ꢀ)-mandelate
(9) (248 mg, 1.48 mmol) in dry CH2Cl2 (4.48 ml) was slowly added
via cannula. The resulting solution was stirred at rt for 24 h. The
mixture was then slowly poured into 30 ml of saturated aqueous
NaHCO3 solution. The organic layer was separated and the aqueous
layer was extracted with CH2Cl2 (3 ꢁ 3.0 ml). The combined organ-
ic layers were dried over anhydrous MgSO4 and the solvent was
evaporated under reduced pressure. The residue was purified by
silica gel chromatography (9:1 hexane:EtOAc?4:1) to afford the
desired product (S)-10 (48 mg, 19% yield) as a colourless oil. Chi-
ral-phase GC/MS (EI-SIM) analysis of the product, tR = 42.2 min