Kilogram-scale asymmetric ruthenium-catalyzed hydrogenation of a tetrasubstituted fluoroenamide

Article abstract of DOI:10.1002/adsc.201100195

Ruthenium-catalyzed asymmetric homogeneous hydrogenation (AHH) is used as the key step of a multi-kilogram scale synthesis of an enantiomeric fluoropiperidine. The AHH of a tetrasubstituted β-fluoroenamide is carried out under mild conditions using a Ru/J

Full text of DOI:10.1002/adsc.201100195

COMMUNICATIONS
DOI: 10.1002/adsc.201100195
Kilogram-Scale Asymmetric Ruthenium-Catalyzed
Hydrogenation of a Tetrasubstituted Fluoroenamide
Andreas Stumpf,a,* Mark Reynolds,a Daniel Sutherlin,a Srinivasan Babu,a
Erhard Bappert,b Felix Spindler,b Michael Welch,c and John Gaudinoc
a
Genentech Inc., 1 DNA Way, South San Francisco, CA 94080, USA
Fax : (+1)-650-467-8382, phone: (+1)-650-467-7461; e-mail: stumpf.andreas@gene.com
Solvias AG, Mattenstrasse 22, 4002 Basel, Switzerland
Array Biopharma, 2620 Trade Center Avenue, Longmont, CO 80503, USA
b
c
Received: March 17, 2011; Revised: July 13, 2011; Published online: December 9, 2011
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adcs.201100195.
Abstract: Ruthenium-catalyzed asymmetric homo-
geneous hydrogenation (AHH) is used as the key
step of a multi-kilogram scale synthesis of an enan-
tiomeric fluoropiperidine. The AHH of a tetrasub-
stituted b-fluoroenamide is carried out under mild
conditions using a Ru/Josiphos catalyst with high ee
(98%).
The target compound 6 might be elaborated in mul-
tiple steps by conventional methods such as enantio-
selective fluorination,[2] catalytic enantioselective de-
carboxylation,[3] and diastereomeric salt resolution of
racemates.[4]
Due to the cis configuration of the piperidine sub-
stituents, AHH of a tetrasubstituted dehydropiperi-
dine was an attractive approach to 6. Thus, AHH of
the cyclic enamide 4 would allow introduction of the
two stereocenters in one step from an achiral starting
material. We hoped to access enamide 4 from the re-
gioselective reduction of commercially available 3-
fluoro-4-amino pyridine (1).
Keywords: amines; enantioselectivity; fluorine;
high-throughput screen (HTS); homogeneous catal-
ysis; hydrogenation; ruthenium; tetrasubstituted b-
fluoroenamides
Although the synthesis of chiral amines via asym-
metric hydrogenation of enamides, using Ru, Rh and
Ir catalysts has been widely investigated by numerous
Enantiomerically pure fluorine-containing molecules groups from both academia and industry,[5] we found
are useful intermediates for the synthesis of active no examples of enantioselective hydrogenation of a-
pharmaceutical ingredients (APIs).[1] As part of devel- or b-substituted fluoroenamides. Only a few examples
opment activities for a drug candidate, we needed of enantioselective hydrogenation of vinyl fluorides
kilogram amounts of enantiomerically enriched have been reported.[6] For example, an enantioselec-
(3S,4R)-1-ethyl-3-fluoropiperidin-4-amine (6) as a key tive Rh-catalyzed hydrogenation of a vinyl fluoride
API intermediate (Scheme 1).
possessing an allylic alcohol was described.[7] The al-
lylic alcohol presumably acts as metal coordinating
group to achieve excellent ee. In our case the carbonyl
function of the enamide would take over this coordi-
nating function to hopefully result in a high chiral in-
duction. However, enantioselective hydrogenation of
tetrasubstituted enamides has been challenging and
plagued by low conversion and enantioselectivity.[5a,b,8]
Furthermore it is known that fluoro-substituted ole-
fins can suffer from defluorination, leading to by-
products and the formation of fluoride ions, that
could poison the catalyst and decrease the yield.[6d,e]
These obstacles were considered as we attempted to
develop a successful protocol.
Scheme 1. Retrosynthetic analysis of 6.
Adv. Synth. Catal. 2011, 353, 3367 – 3372
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3367

Andreas Stumpf et al.
COMMUNICATIONS
Scheme 2. Synthesis of tetrasubstituted cyclic enamide 4.
We set out to synthesize our key intermediate 4 used as precursors in combination with diphosphine
from 4-amino-3-fluoropyridine (1) (Scheme 2). Usual- ligands (PP). The performance of these catalysts was
ly the N-acetyl group is used as protecting group for examined in both protic and aprotic solvents (metha-
enamines[5], but instead we decided to introduce a nol, THF, 1,2-dichloroethane) with TFA as an addi-
benzoyl group, because it provided a chromophore tive in some cases to improve solubility and catalyst
that would simplify HPLC analysis of the process.
performance.
Benzoylation of 1 with benzoyl chloride in THF
The best overall performances were obtained using
and triethylamine at 08C went to completion in 2 h in a Ru complex of the Josiphos ligands SL-J011-1 and
presence of triethylamine (TEA), and 2 was isolated SL-J002-1 (Figure 2) made by mixing [RuCl2A(p-
CHTUNGTRENNUNG
in 92% yield after crystallization from ethyl acetate cymene)]2 precursor with the ligand in MeOH. These
(EtOAc)/hexane. Ethylpyridinium species 3 was ob- two catalysts provided high levels of enantiomeric
tained from alkylation of compound 2 with iodo- excess (97%), small amounts of the de-fluoro com-
ethane in DMF. After reaction completion, 3 was pre- pound 5c (2–3%), and gratifyingly, no trans-5b was
cipitated by pouring the reaction mixture into EtOAc detected (Table 1, entries 7 and 8). The Ru/SL-J002-1
and then isolated after filtration in 97% yield. Inter- catalyst generated from [RuACHTUNTRGENNUG(COD)ACHUTTGNRNEN(GUN CF3CO2)2] under
mediate 3 was used without further purification into aprotic conditions in DCE (entry 9), in combination
the sodium borohydride reduction in MeOH.[7] We with TFA (0.5 equiv./4) as additive, delivered a com-
were delighted to find that treatment of 3 with parable conversion and selectivity, but the enantio-
NaBH4 in MeOH at 08C gave exclusively the desired meric excess was significantly lower (88%). In the re-
regioisomer 4 after reaction quench with aqueous action with catalyst Ru/SL-J301-1 in MeOH (Table 1,
NH4Cl. Extractive work-up and crystallization with entry 10) the de-fluorinated product 5c was formed in
EtOAc following NaHCO3 pH adjustment, furnished a significant amount (19%).
key intermediate 4 in 78% yield (99% purity by
The Rh and Ir catalysts were not as effective in hy-
HPLC). The reduction of 3 was completely selective drogenation of 4 (Table 1, entries 1–6) as the Ru cata-
and no regioisomers of 4 were detected. The overall lyst, resulting in lower ee and higher de-fluorination.
transformation of 1 to 4 proceeded in 68% yield over The Ir catalysts also suffered from low conversions.
3 steps and was successfully demonstrated on scale to Based on these results, a further scale-up using
provide a total of 20 kg of enamide 4.
2 mmol of 4 was undertaken with the Ru/SL-J011-1
Enamide 4 was then subjected to an asymmetric and Ru/SL-J002-1 catalyst systems, which were identi-
homogeneous hydrogenation micro-scale screen fied as the best metal-ligand combinations. Although
[0.04 mmol scale; substrate/catalyst ratio (S/C) of 25; the conversion (92% after 1 day) was lower for the
4 mol% catalyst loading; hydrogen pressure of 60 bar; Ru/SL-J011-1 catalyst (Table 2, entry 1), a high enan-
408C; 16 h].
tiomeric excess (96.5% ee) could be obtained, and
For these screening experiments, 5 types of Ru, Rh almost full conversion was achieved after 2 days. The
and Ir metal precursors in combination with 22 differ- Ru/SL-J002-1 catalyst (Table 2, entry 2) displayed
ent chiral diphosphine ligands to generate the active higher reactivity, but the enantioselectivity was lower
chiral catalysts and five isolated Rh and Ir catalyst (91.6% ee).
precursors were used (Figure 1). All evaluated Rh
With respect to the planned production, we desired
and Ir catalysts were of the type [M(diene)(Lig)]A the higher reactivity of the Ru/SL-J002-1 catalyst, but
A
ACHTUNGTRENNUNG
(M: Rh, Ir; diene COD, NBD; Lig: chiral diphos- needed to increase the enantioselectivity of the reac-
phine or phosphino-oxazoline; A: BF4, CF3SO3, tion. Further optimization raised the S/C ratio from
BARF). For the in-situ generation of the Ru catalysts, 25 to 100 (Table 2, entry 3), by lowering the hydroge-
[RuCl2ACHTUNGTRENNUNG(p-cymene)]2 and [RuACHUTNTRGEG(NNUN COD)ACTHNGUTREN(NUGN CF3COO)2] were nation pressure to 20 bar, and decreasing the reaction
3368
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 3367 – 3372

Kilogram-Scale Asymmetric Ruthenium-Catalyzed Hydrogenation of a Tetrasubstituted Fluoroenamide
Figure 1. Selected chiral diphosphine ligands used in HTS.
Figure 2. Asymmetric hydrogenation of 4.
temperature from 408C to room temperature. Conse- MeOH/DCE mixture (4/1) as solvent on a 10-mmol
quently, the enantioselectivity for 5a was improved to scale. Almost full conversion was reached even at
98.2% ee, and the reaction was complete after 16 h.
We then investigated [Ru(COD)(CF3CO2)2] as
room temperature after 16 h (Table 2, entry 4).
N
N
This difference can be rationalized by the fact that
metal precursor, which delivered a slightly higher re- different catalyst precursors are formed depending on
activity compared to [RuCl2A(p-cymene)]2. This al- the metal precursors. Specifically, treatment of
lowed us to increase the S/C ratio to 200, and use an [Ru(COD)(CF3CO2)2] with 1 equiv. of a josiphos
CTHUNGTRENNUNG
A
ACHTUNGTRENNUNG
Adv. Synth. Catal. 2011, 353, 3367 – 3372
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3369

Andreas Stumpf et al.
COMMUNICATIONS
Table 1. Asymmetric hydrogenation of enamide 4: selected results from micro-scale screening experiments.
Entry
Catalyst
[Rh (NBD)2]BF4
Ligand
Solvent
Conversion [%][b]
5a [%][a]
5a/5b
ee [%][b]
5c [%][c]
1
2
3
4
5
6
7
8
9
10
G
R
SL-J431-1
SL-J212-1
SL-J505-1
SL-J301-1
SL-J226-2
SL-A109-2
SL-J011-1
SL-J002-1
SL-J002-1
SL-J301-1
MeOH
MeOH
DCE[d]
THF
DCE
DCE
MeOH
MeOH
DCE[d]
MeOH
100[c]
100[c]
100[c]
50[c]
51
49
91
58
4
100/0
100/0
100/0
94/6
–/–
–/–
100/0
100/0
100/0
100/0
À87.7
63.0
À69.2
6.7
–
–
97.3
97.1
88.9
72.6
33
47
5
27
69
85
2
3
3
19
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
A
U
[Ir
[Ir
U
74[c]
42[c]
–
A
U
99[c]
97
95
93
77
A
U
100[c]
97[c]
[Ru
[Ru
N
N
100[c]
[a]
Determined by HPLC.
S/C=25, carried out at 408C and 60 bar H2 pressure, 16 h.
TFA additive (0.5 equiv./substrate).
[b]
[c]
Table 2. Selected results for catalyst optimization.
Entry Catalyst Ligand
S/C Scale [mmol] Conversion [%][a] 5a [%][a] 5a/5b ee [%][a] 5c [%][a]
1
2
3
4
N
G
SL-J011-1 25
SL-J002-1 25
SL-J002-1 100 10
2
2
92[b]
97
96
98
98
100/0 96.5
100/0 91.6
100/0 98.2
100/0 99.1
2
4
1
1
100[b]
100[c]
99[d,e]
N
[a]
[b]
[c]
[d]
[e]
Determined by HPLC.
Carried out at 408C and 60 bar H2 pressure, MeOH.
Carried out at 208C and 20 bar H2 pressure, MeOH.
Carried out at 208C and 20 bar H2 pressure, MeOH/DCE (4/1).
TFA additive (0.5 equiv./substrate).
ligand in the presence of methanol results in the for-
mation of a Ru complex with the generic structure enantioselective hydrogenation of a prochiral tetra-
[Ru(CF3CO2)2A substituted b-fluoroenamide, which was conducted by
(MeOH)2(PP)].[9]
In contrast, treatment of [RuCl (p-cymene)]2 with means of a Ru-Josiphos catalyst yielding the desired
The key step of this synthesis sequence was the
A
CHTUNGTRENNUNG
2A
josiphos affords a complex of the type [RuCl
ene)(PP)].
(p-cym- intermediate with high levels of enantiomeric purity
(98.7% ee) and small amount of de-fluorinated by-
With the optimized reaction conditions in hand the product (1.3%). To achieve this goal catalyst screen-
asymmetric hydrogenation of 4 was finally performed ing of Ru, Rh and Ir diphosphine catalysts, followed
on a 24-mol scale, using the [Ru
SL-J002-1 catalyst at an S/C ratio of 200 in a MeOH/ reaction represents a new approach to synthesize cis
DCE mixture (4/1) at 20 bar. vicinal aminofluoro compounds with 2 stereogenic
ACHTUTGNRENGU(N COD)ACHTUNGTRNEN(GUN CF3CO2)2]/ by a fast optimization program were necessary. This
The reaction was run for 20 h at room temperature centers in excellent ee and cis/trans selectivity. To the
and then heated to 408C for 7 h. The desired product best of our knowledge this is the first example for the
5a was obtained in excellent enantiomeric purity enantioselective hydrogenation of a b-fluoroenamide.
(98.1% ee) and 98.7% yield, while the de-fluorinated
Given the importance of this class of compound,
by-product 5c was generated in only 1.3% yield. Fi- this methodology has the potential to be of significant
nally, hydrolysis of 5a in refluxing 6M aqueous HCl relevance in the development of a variety of pharma-
for 9 h proceeded as expected furnishing (3R,4S)-6 ceuticals.
isolated as the dihydrochloride salt in 85% yield
(98.7% purity), 98.7% ee on a 24-mol scale. Overall
12 kg of enantiomerically enriched (3R, 4S)-6 were
made using our improved protocol.
Experimental Section
In summary, we have developed a straightforward,
scaleable, and highly enantioselective (98% ee) pro-
cess to synthesize a key pharmaceutical intermediate,
(3S,4R)-1-ethyl-3-fluoropiperidin-4-amine in 5 steps
and 65% yield.
Synthesis of N-(3-Fluoropyridin-4-yl)benzamide (2)
3-Fluoropyridin-4-amine (2.20 kg, 19.6 mol) was dissolved in
anhydrous THF (25 L and cooled to À58C. Under an atmos-
phere of nitrogen, triethylamine (3.99 kg, 5.5 L, 39.4 mol)
3370
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 3367 – 3372

Kilogram-Scale Asymmetric Ruthenium-Catalyzed Hydrogenation of a Tetrasubstituted Fluoroenamide
was charged followed by dropwise addition of benzoyl chlo-
ride (3.173 kg, 2.62 L, 22.6 mol) over 2 h while maintaining
the internal temperature between À58C to 58C. The reac-
tion was complete 2 h after addition. The reaction mixture
was then filtered and washed with dry THF (5ꢁ20 L). The
THF solution was concentrated to give crude compound 2.
Crude 2 was then dissolved in refluxing EtOAc (8.8 L) and
hexane (2.9 L) was added which caused the solution to
become cloudy. The mixture was cooled to 08C during 2 h,
filtered and dried to afford compound 2; yield: 3.90 kg
(92%); mp 135.98C. 1H NMR (400 MHz, CDCl3): d=8.51
(1H, t, J=5.46 Hz), 8.45 (1H, d, J=2.34 Hz), 8.39 (1H, d,
J=5.46 Hz), 8.30 (1H, s), 7.89 (2H, d, J=8.20 Hz), 7.62
(1H, t, J=7.03 Hz), 7.53 (2H, t, J=7.81 Hz); 13C NMR
(101 MHz, CDCl3): d=165.7, 149.4 (d, J=251.73 Hz), 147.1,
147.0, 137.2, 136.9, 133.6, 133.5, 132.8, 129.1, 127.3, 114.7;
19F NMR (400 MHz, CDCl3): d=À147.4 (1F, s); HR-MS:
m/z=217.0868, calcd. for C12H9FN2O; 216.0699.
CDCl3): d=165.4, 146.3, 143.8, 134.5, 131.8, 128.7, 127.1,
114.1 (d, 5.08 Hz), 51.5, 50.8, 50.5, 49.7, 25.7, 12.4; 19F NMR
(300 MHz, MeOD): d=À122.8 (1F, s); HR-MS: m/z=
249.1778, calcd. for C14H17FN2O 248.1325.
Preparation of Catalyst Solution
A
10-L Schlenk flask was charged with [Ru
ACHTUNGTRENNUNG
AHCTNUTGERG(NNNU CF3CO2)2] (52.6 g, 121 mmol), SL-J002–1
127 mmol), and the flask was vacuum inerted with argon 5
times. Dried and degassed methanol (4.0 L) and dichloro-
ethane (2.5 L) were added via cannula under argon. The sus-
pension was stirred for 45 min at 408C, and then cooled to
room temperature.
Synthesis of N-[(3S,4R)-1-Ethyl-3-fluoropiperidin-4-
yl)benzamide (5a)
A 50-L autoclave was pressurized with nitrogen to 10 bar
and checked for leakage, and subjected to a nitrogen iner-
tion 3 times via vacuum/pressure cycles. Compound 4
(6.00 kg, 24.16 mol) was dissolved in methanol (19 L) and
dichloroethane (1 L) in a 40-L reactor, and added to the au-
toclave. The autoclave was pressurized with nitrogen to
5 bar, depressurized, and this was repeated 3 times. The cat-
alyst solution (6.5 L) was added to the autoclave via canula,
pressurized with hydrogen gas to 20 bar, depressurized, re-
peated 3 times, then pressurized to 20 bar. This mixture was
stirred at 1000 rpm for 20 h at room temperature, while hy-
drogen uptake was monitored. The reactor was heated to
408C for 7 h, and then cooled to 258C. The reactor was
carefully depressurized, pressurized with nitrogen gas to
5 bar, depressurized, and repeated 3 times. The contents of
the autoclave were transferred to a 60-L reactor under a ni-
trogen pressure, and the autoclave rinsed with methanol
(2.0 L). Deloxan THPII (2.4 kg, 40 mass%) was charged and
stirred at 500 rpm for 3 days at room temperature. The sus-
pension was vacuum filtered over Arbocel (0.7 kg), the filter
cake was washed with methanol (5.0 L), and solvent evapo-
rated at 20 mbar and 458C to afford a brown solid that was
used as is in the amide hydrolysis; yield: 6.58 kg; mp
Synthesis of 4-Benzamido-1-ethyl-3-fluoropyridinium
Iodide (3)
Compound 2 (1.95 kg, 9.0 mol) was dissolved in anhydrous
DMF (10 L). The mixture was heated to 708C and iodo-
ethane was charged (1.55 kg, 0.795 L, 9.9 mol). The reaction
temperature rose to 1108C until addition was complete (ca.
1 h). The reaction was maintained at 1008C for a further
2 h, then cooled to room temperature and poured into 50 L
of ethyl acetate (pre-cooled to 58C in a 72-L reactor) caus-
ing immediate precipitation. This was aged for 1 h and the
precipitate was filtered and washed with ethyl acetate (3ꢁ
10 L) to give 2 as a yellowish solid. Product 2 was dried in a
tray vacuum oven at 508C, to provide pure compound 2;
yield: 3.25 kg (97%), mp 208.38C. 1H NMR (400 MHz,
DMSO): d=11.37 (1H, s), 9.41 (1H, dd, J=5.86, 1.56 Hz),
8.88 (1H, dd, J=7.03, 1.56 Hz), 8.73 (1H, t, J=7.42 Hz),
7.99 (2H, d, J=8.20 Hz), 7.72 (1H, t, J=7.42 Hz), 7.61 (2H,
t, J=8.20 Hz), 4.52 (2H, q, J=7.42 Hz), 1.54 (3H, t, J=
7.42 Hz); 13C NMR (101 MHz, DMSO): d=167.3, 149.6 (d,
J=252.57), 142.1, 141.6, 141.5, 134.1, 133.3, 132.6, 128.8,
128.6, 117.9, 117.9, 55.5, 15.9; 19F NMR (400 MHz, DMSO):
d=À130.0 (1F, s); HR-MS: m/z=245.1277, calcd. for
C14H14FN2O 245.1090.
1
154.38C. H NMR (400 MHz, MeOD): d=7.83 (2H, t, J=
9.1 Hz), 7.53 (1H, t, J=7.3 Hz), 7.45 (2H, t, J=7.6 Hz), 4.84
(1H, d, J=49.28 Hz), 4.25–3.91 (1H, m), 3.37–3.22 (2H, m),
3.10–2.95 (1H, m), 2.60–2.39 (2H, m), 2.32–2.02 (3H, m),
1.79 (1H, d, J=12.3 Hz), 1.11 (3H, t, J=7.2 Hz); 13C NMR
(101 MHz, MeOD): d=170.2, 135.6, 132.8, 129.1 (d, J=
176.65 Hz), 89.8, 88.1, 56.4, 56.2, 53.1, 52.8, 50.9, 50.7, 26.7,
11.7; 19F NMR (300 MHz, MeOD): d=À201.7; HR-MS:
Synthesis of N-(1-Ethyl-3-fluoro-1,2,5,6-tetrahydro-
pyridin-4-yl)benzamide (4)
Compound 3 (6.45 kg, 17.3 mol) was dissolved in methanol
(40 L) in a 72-L reactor. Sodium borohydride (1.82 kg,
43.3 mol) was charged portionwise (ca. 5–6 h) maintaining
the internal temperature between À58C and 108C. After re-
action completion, saturated aqueous ammonium chloride
(10 L) was added and th mixture stirred for 1 h, followed by
addition of saturated aqueous sodium bicarbonate (12 L).
This was aged overnight followed by removal of methanol
and extraction with ethyl acetate (3ꢁ30 L). The combined
organics were washed with water and brine, then dried over
sodium sulfate and concentrated to give compound 4; yield:
3.36 kg (78%); mp 137.18C. 1H NMR (400 MHz, CDCl3):
d=7.78 (2H, d, J=7.7 Hz), 7.48 (4H, dq, J=14.9, 7.4 Hz),
3.17 (2H, s), 2.89 (2H, s), 2.65 (2H, t, J=5.6 Hz), 2.55 (2H,
q, J=7.1 Hz), 1.13 (3H t, J=7.2 Hz); 13C NMR (101 MHz,
m/z=251.2293, calcd. for C14H19FN2O 250.1481; [a]26.7
:
+77.08 (589 nm, c 0.89, MeOH, 98% ee); HPLC (Chiralcel
Chiralpak AD-H, 25 cmꢁ4.6 mm, 5 mm; 208C; 220 nm;
1 mLminÀ1; n-heptane/ethanol 90:10): retention times (min):
11.921.
Synthesis of (3S,4R)-1-Ethyl-3-fluoropiperidin-4-
amine Dihydrochloride (6)
Aqueous hydrochloric acid (35 L, 209 mol) and 5a (6.58 kg,
24.16 mol) were charged to a 60-L enamel reactor. The reac-
tor was connected to a sodium hydroxide scrubber and
heated with vigorous stirring to 1308C jacket temperature.
After 50 min the mixture reached reflux (1068C internal
Adv. Synth. Catal. 2011, 353, 3367 – 3372
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3371

Andreas Stumpf et al.
COMMUNICATIONS
temperature). After 1.5 h a black precipitate formed. After
10 h heating was stopped and the supsension allowed to age
overnight at room temperature. The precipitate was collect-
ed on a glass frit and washed with water (2 L). The light
yellow mother liquor was divided between two 30-L rotary
evaporator bulbs and partially concentrated, removing a
total of 20 L solvent on a rotary evaporator at 30 mbar,
608C. Ethanol (10 L) was added to the rotary bulb, and a
further 13 L ethanol/water were distilled at 30 mbar at 608C.
The yellow oil was chased 2 more times with ethanol (10 L
each) as just described. Ethanol (20 L) was added to the re-
maining oil and heated to 608C. The solution was cooled to
room temperature overnight, causing crystallization. The
yellow suspension was cooled to 08C, aged for 1 h, then iso-
lated by filtration on a glass frit, and washed with cold etha-
nol (10 L, 08C), and diethyl ether (5 L). The tan solid was
dried in a vacuum oven for 24 h at 30 mbar, 808C giving 6
dihydrochloride as a tan, microcrystalline powder; yield:
4.70 kg (85%); mp 245.58C. 1H NMR (300 MHz, MeOD):
d=5.35 (d, J=47.2 Hz, 1H), 3.91 (m, 1H), 3.79–3.97 (m,
6H), 2.18 (m, 2H), 1.26 (t, 3H); 13C NMR (101 MHz,
MeOD): d=87.4, 87.2, 85.6, 85.4 (dd, J=177.02 Hz), 64.9,
64.7 (d), 53.8, 51.0, 50.9 (d), 25.4, 25.4 (d), 23.5, 9.7;
19F NMR (400 MHz, CDCl3): d=À204.4; HR-MS: m/z=
graph 187, American Chemical Society, Washington,
DC, 1995; b) T. D. Beeson, D. W. C. MacMillan, J. Am.
Chem. Soc. 2005, 127, 8826–8828; c) K. Uneyama, in:
Organofluorine Chemistry, Blackwell, Ames, IA, 2006;
d) D. W. C. Macmillan, T. D. Beeson, Enantioselective
alpha-fluorination of aldehydes using chiral organic cata-
lysts. U.S. Patent 7,265,249, 2007; e) K. Mueller, C. Faeh,
F. Diderich, Science 2007, 317, 1881–1886; f) D. S.
Reddy, N. Shibata, J. Nagai, S. Nakamura, T. Toru, S.
Kanemasa, Angew. Chem. 2008, 120, 170–174; Angew.
Chem. Int. Ed. 2008, 47, 164–168; g) J. A. Ma, D.
Cahard, Chem. Rev. 2008, 108, 1–43; h) V. A. Brunet, D.
OꢂHagan, Angew. Chem. 2008, 120, 1198–1201; Angew.
Chem. Int. Ed. 2008, 47, 1179–1182.
[3] M. Nakamura, A. Hajra, K. Endo, E. Nakamura,
Angew. Chem. 2005, 117, 7414–7417; Angew. Chem. Int.
Ed. 2005, 44, 7248–7251.
[4] a) J. K. Thottathil, Resolution of benzazepine intermedi-
ates. U.S. Patent 4,965,356, 1989; b) H. Nohira, S. Naka-
mura, M. Kai, Jpn. Kokai Tokyo Koho 6318863, 1989;
c) P. Bravo, G. Resnati, Tetrahedron: Asymmetry 1990,
1, 661–692.
[5] a) T. Nugent, M. El-Shazly, Adv. Synth. Catal. 2010, 352,
753–819; b) J. Meeuwissen, M. Kuil, A. M. Van der Burg,
A. J. Sandee, J. N. H. Reek, Chem. Eur. J. 2009, 15,
10272; c) J. H. Xie, S. F. Zhu, Q. L. Zhou, Chem. Rev.
2011, 111, 1713; d) Z. Zhang, G. Zhu, Q. Jiang, D. Xiao,
X. Zhang, J. Org. Chem. 1999, 64, 1774.
147.1303, calcd. for C7H15FN2: 146.1219; [a]25.3
(589 nm, c 1.09, DMSO, 98% ee).
: +17.0
Acknowledgements
[6] a) M. Saburi, L. Shao, T. Sakurai, Y. Uchida, Tetrahe-
dron Lett. 1992, 33, 7877–7880; b) M. Saburi, T. Sakurai,
H. Kumobayashi, Jpn. Kokai Tokyo Koho 05255177,
1993, Heisei; c) K. Mikami, Y. Itoh, M. Yamanaka,
Chem. Rev. 2004, 104, 1–16; d) M. Engman, J. S. Diesen,
A. Paptchikhine, P. G Andersson, J. Am. Chem. Soc.
2007, 129, 4536–4537; e) P. Kaukoranta, M. Engman, C.
Hedberg, J. Bergquist, P. G. Andersson, Adv. Synth.
Catal. 2008, 350, 1168–1176.
The author would like to thank Dr. Michael Dong and
Megan Sun for their assistance with obtaining HR-mass and
DSC spectra, Bob Garcia Jr. for obtaining optical rotation
data, and Dr. David Askin for helpful discussions during
preparation of this manuscript.
References
[7] S. W. Krska, J. V. Mitten, P. G. Dormer, D. Mowrey, F.
Machrouhi, Y. Sun, T. D. Nelson, Tetrahedron 2009, 65,
8987–8994.
[1] a) K. L. Kirk, J. Fluorine Chem. 2006, 127, 1013–1029;
b) W. K. Hagmann, J. Med. Chem. 2008, 51, 4359–4369;
c) R. Filler, R. Saha, Fut. Med. Chem. 2009, 1 (5), 777–
791; d) K. Mueller, ChemBioChem 2009, 10, 2114.
[8] P. Dupau, C. Bruneau, P. H. Dixneuf, Adv. Synth. Catal.
2001, 343, 331.
[9] N. C. Zanetti, F. Spindler, J. Spencer, A. Togni, G. Rihs,
Organometallics 1996, 15, 860.
[2] a) M. Hudlicky, A. E. Paylath, in: Chemistry of Organic
Fluorine Compounds II: A Critical Review, ACS Mono-
3372
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 3367 – 3372