Highly efficient diastereoselective reduction of α-fluoroimines

Article abstract of DOI:10.1021/ol100647b

A highly selective reduction of α-fluoroimines to the corresponding β-fluoroamines has been developed utilizing trichlorosilane as the reductant. The key aspect of this reaction is the ability of fluorine and nitrogen to activate organosilanes leading to high diastereoselectivity (>100:1) in the product distribution. This new method provides a new avenue for the diastereoselective synthesis of β-fluorinated amines in good yields and selectivity.

Full text of DOI:10.1021/ol100647b

ORGANIC
LETTERS
2010
Vol. 12, No. 10
2186-2189
Highly Efficient Diastereoselective
Reduction of r-Fluoroimines
Roy M. Malamakal,‡ Whitney R. Hess,‡ and Todd A. Davis*
Department of Chemistry, Idaho State UniVersity, Campus Box 8023,
Pocatello, Idaho 83209
daVitodd@isu.edu
Received November 2, 2009
ABSTRACT
A highly selective reduction of r-fluoroimines to the corresponding ꢀ-fluoroamines has been developed utilizing trichlorosilane as the reductant.
The key aspect of this reaction is the ability of fluorine and nitrogen to activate organosilanes leading to high diastereoselectivity (>100:1) in
the product distribution. This new method provides a new avenue for the diastereoselective synthesis of ꢀ-fluorinated amines in good yields
and selectivity.
Interest in fluorinated drug candidates has increased dramati-
cally over the past two decades. The introduction of fluorine
to a molecule can have a profound influence on its biological
activity altering properties including acidity, metabolic
stability, and binding affinity.1 Although many fluorinated
drug candidates have emerged with approximately 150
candidates in phase II and III clinical trials, methods for their
preparation are still lacking.2 Current methods for the
introduction of fluorine generally rely on nucleophilic
(DAST) or electrophilic fluorinating agents (Selectfluor or
Accufluor).3 Additionally, Jorgenson and MacMillan recently
reported the asymmetric synthesis of R-fluoroalcohols utiliz-
ing organocatalysts based on proline and imidazolidinone.4,5
Lindsley expanded on this methodology by utilizing Mac-
Millan’s imidazolidinone catalyst for the asymmetric syn-
thesis of ꢀ-fluoroamines.6 Although such methods are be-
ginning to emerge for the stereoselective introduction of
fluorine, methods for the derivitization of these substrates
remain limited.7 Motivated by the current interest in fluori-
nated drug candidates and the lack of synthetic methodology,
we initiated an investigation into the diastereoselective
reduction of R-fluoroimines to the corresponding ꢀ-fluoro-
amine in high yield and stereoselectivity. Our approach was
to utilize Lewis acid/base activation to mimic chelation
control, thus enhancing selectivity in the product distribu-
tion.8,9 Denmark and co-workers have pioneered the area of
Lewis base catalysis utilizing organosilanes and a variety of
catalysts.10 Our hypothesis in using Lewis acid/base activa-
‡ Undergraduate Researcher.
(1) (a) Kirk, K. L. Curr. Top. Med. Chem. 2006, 6, 1447–1456. (b)
Myers, A. G.; Barbay, J. K.; Zhong, B. J. Am. Chem. Soc. 2001, 123, 7207–
7219.
(2) Kirk, K. L. Curr. Top. Med. Chem. 2006, 6, 1013–1029.
(3) (a) Davis, F. A.; Han, W.; Murphy, C. K. J. Org. Chem. 1995, 60,
4730–4737. (b) Davis, F. A.; Han, W. Tetrahedron Lett. 1991, 32, 1631–
1634. (c) Enders, D.; Faure, S.; Potthoff, M.; Runsink, J. Synthesis 2001,
15, 2307–2319. (d) Davis, F. A.; Kasu, P., V. N. Tetrahedron Lett. 1998,
39, 6135–6138. (e) Davis, F. A.; Han, W. Tetrahedron Lett. 1992, 33, 1153–
1156. (f) Singh, R. P.; Shreeve, J. M. Synthesis 2002, 17, 2561–2576, and
references cited therein.
(4) Beeson, T. D.; MacMillan, D. W. C. J. Am. Chem. Soc. 2005, 127,
8826–8828
.
(5) Frazen, J.; Marigo, M.; Fielenbach, D.; Wabnitz, T. C.; Kjaersgaard,
A.; Jorgenson, K. A. J. Am. Chem. Soc. 2005, 127, 18296–18304
.
(6) Fadeyi, O. O.; Lindsley, C. W. Org. Lett. 2009, 11, 943–946.
(7) (a) Davis, F. A.; Srirajan, V.; Titus, D. D. J. Org. Chem. 1999, 64,
6931–6934. (b) Qiu, X.; Meng, Q. F. Tetrahedron 2004, 60, 6711–6745.
10.1021/ol100647b  2010 American Chemical Society
Published on Web 04/22/2010

Table 1. Effect of Molecular Additives on Rate of Imine Formation
entry
R-fluoroketone
amine
additive
no additive
Amberlyst-15
thiourea (10 mol %)
no additive
reaction time (h)
conversiona (%)
1
2
3
4
5
6
R-fluorocyclohexanone
R-fluorocyclohexanone
R-fluorocyclohexanone
R-fluoroindanone
R-fluoroindanone
R-fluoroindanone
p-anisidine
p-anisidine
p-anisidine
benzylamine
benzylamine
benzylamine
24
24
4
48
15
12
100
100
100
50
85
100
Amberlyst-15
thiourea (10 mol %)
a Imine conversion was determined by both GC and 19F NMR.
tion was based on the strong affinity of fluorine and nitrogen
for silicon, forming an ordered transition state, leading to
high diastereoselectivity in the product distribution.11 Our
method design was to use R-fluoroimine substrates (Lewis
base) to activate trichlorosilane (Cl3SiH) (Lewis acid),
increasing its potency as a reductant, and allowing the
production of ꢀ-fluoroamines in high yield and stereoselec-
tivity.12
Our initial approach was to perform this reaction in a
sequential method without the purification of the intermediate
imine, as R-fluoroimines are generally difficult to purify due
to their decomposition upon heating or standard flash
chromatography. In designing our method, an attempt to
accelerate the imine formation was required. We investigated
a variety of molecular additives and found that a catalytic
amount of thiourea (10 mol %) in toluene greatly decreased
the reaction time for the formation of the R-fluoroimine
(Table 1).13,14 In the case of R-fluorocyclohexanone and
p-anisidine, the reaction time decreased from 24 to 4 h upon
the addition of thiourea (Table 1, entry 3).
With an efficient method for the preparation of the imines
in hand, we began to investigate the reaction parameters
(solvent and temperature) for the diastereoselective reduction
of the imine derived from R-fluorocyclohexanone and
p-anisidine utilizing Cl3SiH. These studies showed that THF
and EtOAc provided the best diastereoselectivity (17:1 syn:
anti) in the reduction of the model imine at 0 °C (Table 2,
entries 3 and 7).15 EtOAc was chosen as the optimal solvent
due to its advantageous properties to industrial scale synthesis
in comparison to THF, as well as the modest increase in
product yield (57% to 73%). Next, we examined the effect
of temperature on the selectivity and yield utilizing EtOAc
Table 2. Reaction Optimization for the Reduction of
R-Fluoroiminesa b
,
(8) For reviews of chelation control see: (a) Reetz, M. Angew Chem.,
Int. Ed. 1984, 23, 556–569. (b) Mengel, A.; Reiser, O. Chem. ReV. 1999,
99, 1191–1223. (c) Reetz, M. T. Acc. Chem. Res. 1993, 26, 462–468. Also
see: (d) Cram, D. J.; Elhafez, F. A. A. J. Am. Chem. Soc. 1952, 74, 5828–
5835. (e) Cram, D. J.; Kopecky, K. J. Am. Chem. Soc. 1959, 81, 2748–
2755. (f) Evans, D. A.; Allison, B. D.; Yang, M. G.; Masse, C. E. J. Am.
Chem. Soc. 2001, 123, 10840–10852.
(9) For fluorine in chelation control see: (a) Mohanta, P. K.; Davis, T. A.;
Gooch, J. R.; Flowers, R. A., II J. Am. Chem. Soc. 2005, 127, 11896–
11897. (b) Ramachandran, P. V.; Gong, B.;Q.; Teodorovic, A. V. J. Fluorine
Chem. 2007, 128, 844-850.
(10) (a) Denmark, S. E.; Beutner, G. L. Angew. Chem., Int. Ed. 2008,
47, 1560–1638. (b) Malkov, A. V.; Stoncius, S.; MacDougall, K. N.;
Mariani, A.; McGeosh, G. D.; Kocovsky, P. Tetrahedron 2006, 62, 264–
284.
entry
solvent
temp (°C)
yield (%)
syn:anti
(11) Walsh, R. Acc. Chem. Res. 1981, 14, 246–252.
1
2
3
4
CH2Cl2
CH3CN
THF
toluene
acetone
DMSO
EtOAc
EtOAc
EtOAc
0
0
0
0
0
80
77
57
82
0
6:1
(12) For reductions of imines utilizing Cl3SiH see: (a) Zhou, L.; Wang,
Z.; Wei, S.; Sun J. J. Chem. Soc., Chem. Commun. 2007, 28, 2977–2979.
(b) Blackwell, J. M.; Sonmor, E. R.; Scoccitti, T.; Piers, W. R. Org. Lett.
2000, 2, 3921–3923. (c) Onomura, O.; Kouchi, Y.; Iwasaki, F.; Matsumura,
Y. Tetrahedron Lett. 2006, 47, 3751–3754. (d) Guizzetti, S.; Benaglia, M.;
Rossi, S. Org. Lett. 2009, 11, 2928–2931. (e) Wang, C.; Wu, Z.; Zhou, L.;
Sun, T. Chem.sEur. J. 2008, 29, 8789–8792. (f) Zheng, H.; Deng, J.; Lin,
W.; Zhang, X. Tetrahedron Lett. 2007, 45, 7934–7937. (g) Wang, Z.; Ye,
X.; Wei, S.; Zhang, A.; Sun, J. Org. Lett. 2006, 8, 999–1001. (h) Malkov,
A. V.; Stewart Liddon, A. J. P.; Ramirez-Lopex, P.; Bendova, L.; Haigh,
D.; Kocovsky, P. Angew. Chem., Int. Ed. 2006, 45, 1432–1435. (i) Giuzzetti,
S.; Benaglia, M.; Rossi, S. Org. Lett. 2009, 11, 2928–2931. (j) Malkov,
A. V.; Figlus, M.; Kocovsky, P. J. Org. Chem. 2008, 73, 3985–3995. (k)
Malkov, A. V.; Figlus, M.; Stoncius, S.; Kocovsky, P. J. Org. Chem. 2007,
72, 1315–1325. (l) Malkov, A. V.; Vrankova, K.; Stoncius, S.; Kocovsky,
P. J. Org. Chem. 2009, 74, 5839–5849.
11:1
17:1
4:1
N/A
N/A
17:1
20:1
24:1
5c
6c
7
0
0
0
73
81
80
8
9
-10
-78
a All reductions were performed on a 0.3 mmol scale. b All reactions
were initiated at the indicated temperature, allowed to slowly warm to room
temperature, and then stirred for 12 h. c Starting R-fluoroketone was
recovered.
Org. Lett., Vol. 12, No. 10, 2010
2187

Table 3. Examining the Substrate Scope for the Diastereselective Reduction of R-Fluoroimines
entry
R-fluoroketone
amine
imine conversiona (%)
yieldb (%)
syn:antic
1
2
3
4
5
6
7
8
9
10
11
12
13
R-fluorocyclohexanone
R-fluorocyclohexanone
R-fluorocyclohexanone
R-fluorocyclohexanone
R-fluorocyclohexanone
R-fluorocyclohexanone
R-fluoroindanone
R-fluoroindanone
R-fluoroindanone
R-fluoroindanone
p-anisidine
100
100
100
100
0
78
98
51
68
ND
ND
88
80
57
24:1
14:1
11:1
12:1
ND
benzylamine
cyclohexylamine
hexylamine
p-chloroaniline
aniline
25
ND
p-anisidine
100
100
100
10
75
91
>100:1
>100:1
78:1
ND
19:1
5:1
benzylamine
hexylamine
p-chloroaniline
p-anisidine
benzylamine
hexylamine
ND
51
R-fluoropropiophenone
R-fluoropropiophenone
R-fluoropropiophenone
68
100
51d
4:1
a Imine conversion was determined with GC. b Yield is based on purifed yield after chromatography. c Diastereoselectivity was determined by integration
of the 19F NMR and in select cases confirmed by GC. d Crude yield in 98% purity based on 1H and 19F NMR. Product decomposed upon attempted purification
procedures.
as the solvent of choice. As the temperature was lowered
from 0 to -78 °C, the diastereoselectivity increased from
17:1 to 24:1 syn:anti with comparable yields (Table 2, entries
7 and 9). On the basis of these results, optimal reaction
conditions favored the reduction of R-fluoroimines in EtOAc
at -78 °C.
A preliminary mechanistic investigation of this reaction
found that the absence of fluorine or nitrogen in the substrate
was sufficient to prevent the reduction of the imine or
R-fluoroketone (Scheme 1). As illustrated in Scheme 1,
neither the imine derived from cyclohexanone and p-anisidine
nor R-fluorocyclohexanone produced any of the desired
reduction products upon the treatment with Cl3SiH under
the conditions described above. However, the reduction of
the corresponding R-fluoroimines proceeds well in good yield
and diastereoselectivity favoring the syn-diastereomer. On
the basis of these results, it is imperative that the presence
of the fluorine and nitrogen are both required for the
reduction to occur.16 Our proposed mechanism involves an
initial activation of Cl3SiH (Lewis acid) by fluorine or
nitrogen (Lewis base), forming a pentacoordinate trigonal
bypyramidal silicon.17 Following the interaction of the
fluorine or nitrogen with silicon, the reaction proceeds
The substrate scope of the reduction was then investigated.
We discovered that the yield of the two-step process ((1)
imine formation followed by (2) reduction of the imine) was
highly dependent upon the formation of the imine. Electronic
effects played an important role in the formation of the imine,
as electron-poor, weakly nucleophilic amines led to minimal
imine formation (Table 3, entries 5, 6, and 10). Due to the
difficulty of forming these imines under these conditions,
we focused our attention on amines with electron-donating
substituents to foster complete conversion. The reaction
proceeded well for a variety of amines, including acyclic
(Table 3, entries 2, 4, 8, 9, 12, and 13), cyclic (Table 3,
entry 3), aromatic (Table 3, entries 1, 7, and 11), as well as
acyclic (Table 3 entries 11-13) and cyclic R-fluoroketones
(Table 3, entries 1-10). Diastereoselectivities ranged from
4:1 to >100:1 syn:anti. These substrate studies showed that
the reduction of R-fluoroimines proceeds in high yield and
selectivity accommodating a variety of amines with the
preparation of the subsequent R-fluoroimine being the
limiting factor of the two-step process.
Scheme 1
.
Preliminary Mechanistic Insights for the Reduction
of R-Fluoroimines
(13) Thiourea derivatives have been utilized extensively as organocata-
lysts for the activation of ketones and imines. For a recent review see:
Jacobsen, E. N.; Taylor, M. S. Angew. Chem., Int. Ed. 2006, 45, 1520–
1543.
(14) For thiourea activation see: (a) Menche, D.; Arikan, F. Synlett 2006,
6, 841–844. (b) Menche, D.; Hassfeld, J.; Menche, G.; Ritter, A.; Rudolph,
S. Org. Lett. 2008, 8, 740–744.
(15) The syn/anti stereochemistry was determined by 19F NMR and 1H
NMR. The 19F NMR signal for the syn stereoisomer appears upfield in
comparison to that of the anti stereoisomer. For examples of stereochemical
assignment of ꢀ-fluoroamines see: (a) Fan, R.; Zhou, Y.; Zhang, W.; Hou,
X.; Dai, L. J. Org. Chem. 2004, 69, 335–338. (b) Alvernhe, G. M.;
Ennakoua, C. M.; Lacombe, S. M.; Laurent, A. J. J. Org. Chem. 1981, 46,
4938–4948. (c) Wade, T. N. J. Org. Chem. 1980, 45, 5328–5333. (d)
Avernhe, G. M.; Lacombe, S. M.; Laurent, A. J. Tetrahedron Lett. 1980,
21, 289. (e) Toulgui, C.; Chaabouni, M. M.; Baklouti, A. J. Fluorine Chem.
1990, 46, 385–391.
2188
Org. Lett., Vol. 12, No. 10, 2010

through a highly ordered transition state featuring a five-
member ring between the octahedral silicon, fluorine, and
nitrogen atoms.18,19 In this transition state, the donation of
electron density from the fluorine and nitrogen to the silicon
produces an electron-rich silicon, which is now primed to
deliver a hydride to reduce the imine to the corresponding
ꢀ-fluoroamine in high diastereoselectivity. The high syn:anti
ratio is consistent with the transition state being under
chelation control (Scheme 1). The role of chelation is
supported by computational studies conducted by Paddon-
Row involving nucleophilic addition to acyclic R-fluorinated
aldehydes.20 On the basis of these computational studies in
the absence of metal chelation, the nitrogen and fluorine will
be antiperiplanar due to electron repulsion of the lone pairs,
and thus yield the anti-diastereomer.20 Our results for the
reduction of R-fluorinated imines (Scheme 1) display a
preference for the syn- diastereomer (Table 3, entries 1-13)
as the major product further suggesting the role of chelation
in the transition state leading to the observed syn-stereose-
lectivity.
We are currently conducting NMR studies to investigate
the interaction of fluorine, ketones, and imines with a variety
of organosilanes to begin to deduce the affinity of nitrogen
and fluorine for silicon, and to monitor the coordination
events around silicon. These results will be presented in due
course.
In conclusion, we have developed a new methodology for
the reduction of R-fluoroimines in high yield and diastereo-
selectivity. This new methodology is based on the activation
of organosilanes by fluorinated substrates, which results in
highly ordered transition states and enhanced reducing reac-
tivity. This approach is especially applicable for industrial-
scale synthesis because the reaction is sequential and elim-
inates the use of highly volatile ethereal solvents that are
generally used in transition metal based chelation control
synthesis.
Acknowledgment. The authors thank the College of Arts
& Sciences, Office of Research, and Department of Chem-
istry at Idaho State University for generous start-up funds.
T.A.D. is a recipient of the Ralph. E. Powe Junior Faculty
Enhancement Award administered by ORAU. R.M.M. was
supported by an Undergraduate Summer Fellow award from
Idaho INBRE (NIH Grant P20RR16454). We would also
like to thank Professors Karl De Jesus and Andy Holland at
Idaho State University for their insightful comments.
(16) For examples of nitrogen and fluorine interaction with silicon see:
(a) Nakash, M.; Gut, D.; Goldvaser, M. Inorg. Chem. 2005, 44, 1023–
1030. (b) Yamamura, M.; Kano, N.; Kawashima, T.; Matsumoto, T.; Harada,
J.; Ogawa, K. J. Org. Chem. 2008, 73, 8244–8249
.
(17) For mechanistic insights in the Lewis acid/base activation for
allylsilane additions see: (a) Denmark, S. E.; Fu, J. J. Am. Chem. Soc. 2000,
122, 12021–12022. (b) Denmark, S. E.; Fu, J.; Coe, D. M.; Su, X.; Pratt,
N. E.; Gridel, B. D. J. Org. Chem. 2006, 1513–1522
(18) Gordon, M. S.; Carroll, M. T.; Davis, L. P.; Burggaff, L. W. J.
Phys. Chem. 1990, 94, 8125–8128
(19) Fester, G. W.; Wagler, J.; Brendler, E.; Bohme, U.; Gerlach, D.;
Kroke, E. J. Am. Chem. Soc. 2009, 131, 6855–6864
.
.
Supporting Information Available: Experimental pro-
cedures and analytical data. This material is available free
of charge via the Internet at http://pubs.acs.org.
.
(20) (a) Wong, S. S.; Paddon-Row, M. N. J. Chem. Soc., Chem.
Commun. 1991, 327–330. (b) Wong, S. S.; Paddon-Row, M. N. Aust.
J. Chem. 1991, 44, 765–770.
OL100647B
Org. Lett., Vol. 12, No. 10, 2010
2189