Synthesis of a tertiary carbinamide via a novel Rh-catalyzed asymmetric hydrogenation

Article abstract of DOI:10.1021/jo702429u

(Chemical Equation Presented) Asymmetric hydrogenation of allylic dimethylcarbinamide 2 with 1 mol % of cationic Rh(I)-Josiphos complex in THF under 500 psi of H₂ generated the corresponding tertiary carbinamide 1 in 98.5% assay yield and a 94:

Full text of DOI:10.1021/jo702429u

SCHEME 1. Retrosynthetic Analysis of Carbinamide 1
Synthesis of a Tertiary Carbinamide via a Novel
Rh-Catalyzed Asymmetric Hydrogenation
John Limanto,* C. Scott Shultz, Benjamin Dorner,
Richard A. Desmond, Paul N. Devine, and Shane W. Krska
Department of Process Research, Merck Research Laboratories,
Merck & Co., Inc., Rahway, New Jersey 07065
SCHEME 2. Preparation of Intermediate 7
john_limanto@merck.com
ReceiVed NoVember 12, 2007
tuted allylic amides or phthalimides3 and a primary allylic
carbinamine4 has previously been reported. Due to the inherent
steric congestion at a tertiary allylic carbinamine, catalytic
asymmetric hydrogenation of this type of compound represents
a challenging task. Herein we report an efficient catalytic
enantioselective hydrogenation approach to 1, which is amenable
to preparative scale.
The requisite hydrogenation precursor 2 could be constructed
from the vinyl nitrile intermediate 3, which, in turn, would be
readily accessible from commercial N-Boc-4-piperidone. Hence,
ammonium acetate catalyzed Kno¨venagel condensation between
Meldrum’s acid and N-Boc-4-piperidone in EtOAc afforded
compound 5, which was crystallized directly from the reaction
mixture in 91% isolated yield (Scheme 2). Subsequent conjugate
addition was performed using Grignard reagent 6, in the absence
of any copper salts,5 in THF to afford compound 7 in 90%
isolated yield.
Initial attempts to generate indanone 4a from 7 under
lanthanide-catalyzed intramolecular Friedel-Crafts cyclization6
resulted only in the removal of the Boc group and decomposition
of the starting material. On the other hand, heating compound
7 either in a 2:1 mixture of pyridine/H2O at 90 °C or in wet
toluene (5-10% v/v H2O) in the presence of pyridine (2 molar
equiv) at 85-100 °C resulted in sequential hydrolysis and
decarboxylation to cleanly generate acid 8 in 90% yield (Scheme
3). While the product can be isolated as a pyridine complex,
the crude material was used directly in the next step.7 Conver-
sion of the acid to the corresponding acid chloride 9, followed
by treatment with AlCl3 (2-3 molar equiv) in either CH2Cl2 or
Asymmetric hydrogenation of allylic dimethylcarbinamide
2 with 1 mol % of cationic Rh(I)-Josiphos complex in THF
under 500 psi of H2 generated the corresponding tertiary
carbinamide 1 in 98.5% assay yield and a 94:6 enantiomeric
ratio. Upon crystallization, the product was isolated in 91%
isolated yield and 95:5 enantiomeric ratio.
Compounds containing spiropiperidine backbones have been
greatly utilized in pharmaceutical research due to their signifi-
cant biological properties.1 We recently required an efficient
and scalable enantioselective approach to a structurally complex
tertiary carbinamide 1, which could potentially be prepared via
an asymmetric hydrogenation of unsaturated tertiary carbinamide
2 (Scheme 1). While homogeneous catalytic asymmetric
hydrogenation of olefins has been a very active area of research
and has found abundant application in both academic and
industrial arena,2 hydrogenation involving R,R-disubstituted
allylic amines or amides, such as 2, has not yet been explored.
On the other hand, enantioselective hydrogenation of â-substi-
(1) For examples, see: (a) Lu, Z.; Tata, J. R.; Cheng, K.; Wei, L.; Chan,
W. W.-S.; Butler, B.; Schleim, K. D.; Jacks, T. M.; Hickey, G.; Patchett,
A. A. Bioorg. Med. Chem. Lett. 2007, 17, 3657. (b) Jia, L.; Zou, J.; So,
S.-S.; Sun, H. J. Chem. Inf. Mod. 2007, 47, 1545. (c) Bakshi, R. K.;
Dellureficio, J. P.; Dobbelaar, P. H.; Guo, L.; He, S.; Hong, Q.; Nargund,
R. P.; Ye, Z. PCT Int. Appl. 2007, 132. (d) Allerton, C. M. N.; Owen, D.
R.; Ryckmans, T.; Stammen, B. L. C. PCT Int. Appl. 2007, 197. (e) Ito,
F.; Koike, H.; Sudo, M.; Yamagishi, T.; Ando, K. PCT Int. Appl. 2003,
196. (f) Tata, J. R.; Lu, Z.; Jacks, T. M.; Schleim, K. D.; Cheng, K.; Wei,
L.; Chan, W. W.-S.; Butler, B.; Tsou, N.; Leung, K.; Chiu, S.-H. L.; Hickey,
G.; Smith, R. G.; Patchett, A. A. Bioorg. Med. Chem. Lett. 1997, 7, 2319.
(2) For reviews on metal-catalyzed asymmetric hydrogenations, see: (a)
Cui, X.; Burgess, K. Chem. ReV. 2005, 105, 3272. (b) Tang, W.; Zhang, X.
Chem. ReV. 2003, 103, 3029. (c) Blaser, H. U.; Malan, C.; Pugin, B.;
Spindler, F.; Steiner, H.; Studer, M. AdV. Synth. Catal. 2003, 345, 103. (d)
Ojima, I. Catalytic Asymmteric Synthesis; Wiley-VCH: New York, 2000.
(e) Noyori, R. Asymmetric Catalysis in Organic Synthesis; Wiley: New
York, 1994. For examples of industrial applications, see: (f) Lennon, I.
C.; Moran, P. H. Curr. Opin. Drug DiscoVery DeV. 2003, 6, 855. (g) Genet,
J.-P. Acc. Chem. Res. 2003, 36, 908.
(3) (a) Wang, C.-J.; Sun, X.; Zhang, X. Angew. Chem., Int. Ed. 2005,
44, 4933. (b) Saylik, D.; Campi, E. M.; Donohue, A. C.; Jackson, W. R.;
Robinson, A. J. Tetrahedron: Asymmetry 2001, 12, 657. (c) Brown, J. M;
Parker, D. J. Org. Chem. 1982, 47, 2722. For an example of diastereose-
lective reduction of chiral cyclic allylic amide under heterogeneous
conditions, see: Hu, X. E.; Kim, N. K.; Ledoussal, B. Org. Lett. 2002, 4,
4499.
(4) For example, see: Yamano, T.; Yamashita, M.; Adachi, M.; Tanaka,
M.; Matsumoto, K.; Kawada, M.; Uchikawa, O.; Fukatsu, K.; Ohkawa, S.
Tetrahedron: Asymmetry 2006, 17, 184.
(5) For examples of copper-salt-mediated conjugated additions to Mel-
drum’s alkylidenes, see: (a) Vogt, P. F.; Molino, B. F.; Robichaud, A. J.
Synth. Commun. 2001, 31, 679. (b) Davies, A. P.; Egan, T. J.; Orchard, M.
G.; Cunningham, D.; McArdle, P. Tetrahedron 1992, 48, 8725. (c) Huang,
X.; Chan, C.-C.; Wu, Q.-L. Tetrahedron Lett. 1982, 23, 75.
(6) Fillion, E.; Fishlock, D. Org. Lett. 2003, 5, 4653.
10.1021/jo702429u CCC: $40.75 © 2008 American Chemical Society
Published on Web 01/11/2008
J. Org. Chem. 2008, 73, 1639-1642
1639

SCHEME 3. Preparation of Indanone 4a
SCHEME 4. Synthesis of Vinyl Nitrile 3
PhCl at -25 °C, effected the Friedel-Crafts reaction to yield
an 80:20 regioisomeric mixture of indanones with concomitant
removal of the Boc group. For compatibility purposes in the
downstream chemistry, the crude amine was reprotected with
Boc2O to give regioisomeric products 4 in 80% combined assay
yield or 64% yield of the desired indanone 4a.
With indanone 4a in hand, generation of the vinyl nitrile
intermediate 3 was accomplished in a two-step one-pot process,
involving cyanophosphorylation, followed by Lewis acid in-
duced dehydrophosphorylation. Specifically, treatment of the
crude mixture of indanones 4a,b with diethyl cyanophosphate8
in the presence of 20 mol % of LiOEt9 or LiOMe in either
MTBE, THF, or toluene at rt afforded intermediates 10a and
10b, which upon subsequent exposure to BF3•Et2O13a (0.5 molar
equiv) at 0 °C f rt resulted in the chemoselective formation of
the desired vinyl nitrile 3a. Due to the slower elimination rate
observed for the undesired regioisomer 10b (k1 > k2),10 the
current protocol allowed for an efficient regioselectivity upgrade
to solely generate compound 3 in 81% isolated yield after
crystallization (Scheme 4). Subsequent incorporation of the gem-
dimethyl group was accomplished via a CeCl3-promoted double
addition of MeLi under non-cryogenic conditions11 to give
carbinamine 11, which upon acetylation in EtOAc generated
the desired carbinamide 2 in 83% isolated yield.12
TABLE 1. Initial Metal-Phosphine Screens: High Loadings
productf
entry metala ligand solventb S/Cc [SM]d % conve (LCAP) % eeg
1
2
3
4
5
6
7
8
9
10
11
12
13
Ru
Ru
Ru
Ru
Ru
Ru
Rh
Rh
Rh
Rh
Rh
Rh
Rh
12a MeOH
12b MeOH
12b TFE
12b EtOAc
12b THF
12b THF
12
12
12
12
12
20
12
12
12
12
20
20
20
20
20
20
20
20
40
20
10
20
20
20
20
20
10
34
94
5
27
64
62
77
83
78
84
75
28
83
71
86
64
74
40
53
59
44
49
21
67
86
78
82
81
97
100
100
90
100
83
36
93
91
97
(7) Attempts to perform Friedel-Crafts acylation directly from 8 using
various Brønsted acid or lanthanide triflates only resulted in Boc removal
and/or poor conversions. For MsOH-promoted process, see: (a) Premasagar,
V.; Palaniswamy, V. A.; Eisenbrau, E. J. J. Org. Chem. 1981, 46, 2974.
For PPA, see: (b) Guy, A.; Guette, J.-P. Synthesis 1980, 3, 222. For TfOH,
see: (c) Orita, A.; Yaruva, J.; Otera, J. Angew. Chem., Int. Ed. 1999, 38,
2267. (d) Quallich, G. J.; Woodall, T. M. Tetrahedron 1992, 48, 10239.
For ClSO3H, see: (e) Corey, E. J.; Gant, T. G. Tetrahedron Lett. 1994, 35,
30. For lanthanide-promoted processes, see: (f) Cui, D.-M.; Kawamura,
M.; Shimada, S.; Hayashi, T.; Tanaka, M. Tetrahedron Lett. 2003, 44, 4007
and references cited therein.
(8) Commercially available or generated in situ from ClPO(OEt)2: Shi,
E.; Xiao, J.; Pei, C. Phosphorous, Sulfur Silicon Relat. Phenom. 2004, 179,
1361.
(9) Treatment of (EtO)2POCN with LiOEt appears to generate LiCN and
(EtO)3PO as suggested by comparing the 31P NMR spectrum of the latter
to that of the authentic sample. For LiCN-induced cyanophosphorylation,
see: (a) Harusawa, S.; Yoneda, R.; Kurihara, T.; Hamada, Y.; Shioiri, T.
Tetrahedron Lett. 25, 4, 427. For similar LDA-induced process, see: (b)
Harusawa, S.; Yoneda, R.; Kurihara, T.; Hamada, Y.; Shioiri, T. Chem.
Pharm. Bull. 1983, 31, 2932.
(10) The conversions were monitored periodically by both NMR and
HPLC. After 18 h at rt, a full conversion of 10a to 3a was typically obtained,
while ∼90% of 10b remained unreacted under these conditions.
(11) Limanto, J.; Dorner, B.; Devine, P. N. Synthesis 2006, 24, 4143
and references cited therein.
(12) It is important to note that other protecting groups for the piperidine
nitrogen such as benzamide, acetamide, CBz, or ethyl carbamate are not
compatible with the organocerium chemistry, hence, justifying the necessity
to reincorporate the Boc protecting group in the piperidine ring after
Friedel-Crafts acylation.
(13) (a) McGarrity, J.; Spindler, F.; Fuchs, R.; Eyer, M. (LONZA-AG),
EP-A 624587 A2, 1995. (b) Togni, A.; Breutel, C.; Schnyder, A.; Spindler,
F.; Landert, H.; Tijani, A. J. Am. Chem. Soc. 1994, 116, 4062.
14
MeOH
13a MeOH
13b MeOH
12d MeOH
12d TFE
12d EtOAc
12d THF
a The Rh catalysts were formed by combining the metal precursor and
the ligands at rt for 1 h in MeOH or, in the case of Ru-methallyl precursor,
in DCE/MeOH along with 2 equiv of HBF4•Et2O for metal activation. b Dry
solvents were used for the studies (KF < 100 ppm). c Substrate to catalyst
ratio. d Concentration of starting olefin in g/L. e Uncorrected: [product A%/
SM A%] × 100%. f HPLC area percent, uncorrected for absorption factor.
g Determined by chiral HPLC analysis (Chiralcel OD-H). Absolute con-
figuration was obtained by X-ray single-crystal structure (see Supporting
Information).
With the requisite hydrogenation substrate 2a in hand, the
hydrogenation studies were initiated by screening a library of
chiral phosphine ligands in combination with Ru or Rh metal
precursors under high catalyst loading conditions (S/C ) 12-
20). From the initial screen, metal complexes of Josiphos-type
(12),13 BPE (13), and Norphos ligands (14) were found to exhibit
significant reactivities toward the olefin at 45-50 °C and 90
psi of H2 in the solvents presented in Table 1. Under Ru
catalysis, Josiphos-type ligands 12a,b offered modest enanti-
oselectivities and low conversions in MeOH (entries 1 and 2),
while the same transformations gave the highest conversions,
1640 J. Org. Chem., Vol. 73, No. 4, 2008

TABLE 2. Catalyst Loading and Pressure Screens
TABLE 3. Impurity Effects on the Hydrogenation of 2
producte
entry liganda solvent H2 (psi) S/Cb [SM]c % convd (LCAP) % eef
entry additives mol %a H2 (psi) T (°C) [SM]b % convc % eed
1
2
3
4
5
6
7
8
12d
12c
12c
12c
12c
12d
12d
12c
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
THF
90
90
90
500
500
500
500
500
20
40
20
25
50
70
70
70
70
70
91
29
21
87
30
99
100
100
71
23
19
74
27
85
82
79
82
87
84
87
89
88
1
2
3
4
5
6
7
8
none
n/a
10
5
5
1
1
5
3
500
90
40
25
40
40
40
40
25
40
100
50
100
1
8
90
7
1
86
nd
nd
81
58
nd
nd
77
NaOAc
NaOAc
Na2CO3
Et3N
Et3NHCl
Ac2O
100
100
130
100
100
100
500
500
500
500
500
500
100
100
100
100
50
93
1
33
THF
91 (98)g
H2O
100
a Catalyst was formed by mixing the ligand and precursor in the solvent
at rt for 1 h. b Substrate to catalyst ratio. c Concentration of starting material
in g/L. d Uncorrected: [product A%/SM A%] × 100%. e Uncorrected HPLC
area percent. f Determined by chiral SFC spectroscopy (Chiralcel OD-H).
g Number in parentheses represents HPLC assay yield.
a With respect to starting olefin. b Concentration of starting olefin in g/L.
c Uncorrected: [product A%/SM A%] × 100%. d Determined by chiral SFC
analysis.
TABLE 4. Hydrogenation of Amine 11
but lower selectivities in non-alcoholic solvents or TFE (entries
3-6). On the other hand, Rh complexes of BPE and Norphos
ligands exhibited high reactivities, promoting the hydrogenation
in high conversions, albeit in moderate enantioselectivities
(entries 7-9). Alternatively, Rh-Josiphos complex 12d cata-
lyzed the desired transformation in high conversions (91-97%)
and enantioselectivity (81%) when performed in non-alcoholic
solvents, such as THF or EtOAc (entries 10-13). Encouraged
by these promising results, further reaction optimization was
performed using this and other Rh-Josiphos catalyst systems.
Subsequent studies showed that, with ligand 12c, the reaction
conversions were attenuated significantly when the catalyst
loadings were reduced by more than 2-fold in EtOAc (entries
2 and 3, Table 2). However, with 1 mol % catalyst loading, the
high reactivity can be restored upon performing the reaction at
higher pressure (i.e., 500 psi), without sacrificing the enanti-
oselectivity (entry 4). With less than 1 mol % of Rh-12c at
500 psi H2, the reaction proceeded to give low conversion (entry
5), but still with high enantioselectivity. While other Josiphos
ligands (i.e., 12d) also effected the reaction efficiently at 1 mol
% loading (entries 6 and 7), due to its bulk availability and
lower cost, ligand 12c was selected for large scale productions.
Hence, subjection of 2 to 1 mol % of Rh-12c(BF4) in 10-14
mL/g THF at 40 °C under 500 psi of H2 afforded the saturated
product 1 in a 94:6 enantiomeric ratio and 98% assay yield.
Isolation by crystallization from MeCN afforded the desired
compound 1 in 91% isolated yield and in a satisfactory 95:5
enantiomeric ratio (90% ee).
Although Rh-12c proves to be an active catalyst, its catalytic
activity appears to be attenuated by the presence of impurities
in the starting material (Table 3). For example, under optimized
reaction conditions, a small amount of NaOAc (5-10 mol %)
strongly inhibited the catalyst, resulting in only negligible
conversions (entries 2 and 3). Such unfavorable reaction
outcomes, however, were not observed when 5 mol % of Na2-
CO3 was present at the beginning of the reaction (entry 4). On
the other hand, Et3N, Et3N•HCl salt, and Ac2O were all potent
inhibitors for the Rh-Josiphos catalyst system, even in very
small amounts (entries 5-7). In addition, performing the
hydrogenation in wet THF solution (i.e., 3 mol % of H2O)
resulted in much lower conversion and slightly reduced enan-
entry
liganda
H2 (psi)
S/Cb
[SM]c
% convd
% eee
1
2
3
4
5
6
7
8
15
15
15
15
16
16
16
16
90
90
90
500
90
90
6
30
100
100
6
12
30
65
4
100
100
100
10
20
100
100
100
72
4
87
46
0
5
0
98
98
64
2
93
90
53
0
90
500
a The catalysts were typically preformed by heating the metal precursors
and the ligands at 20-50 °C in DCE/EtOH or THF for 1 h. b Substrate to
catalyst ratio. c Concentration of starting olefin in g/L. d Uncorrected:
[product A%/SM A%] × 100%. e Determined by chiral SFC analysis.
tioselectivity (entry 8). For optimum results (entry 1), dry THF
(KF < 100 ppm) was routinely used and the solution was first
filtered through a 1 µm inline filter to remove any insoluble
salts or particulates prior to hydrogenation.
Virtually all of the catalyst poisons in Table 3 were used or
generated during the acetylation of 11 with AcCl or Ac2O with
Et3N and/or during reaction workup with Na2CO3. In order to
completely avoid these potential contamination issues, asym-
metric hydrogenation of the free amine 11 was also investigated
using our phosphine library. In this regard, a combination of
(COD)2RhBF4 and either (R)-Cl-OMe-Biphep 15 or (R)-
binaphane 16 was found to effect the desired transformation to
generate amine 17 in good yields and enantioselectivities (Table
4). Unfortunately, the best results could only be obtained under
high catalyst loadings (g10 mol %, entries 1, 5, and 6). Both
the conversion and enantioselectivity decreased significantly as
the amount of catalyst was reduced to as low as 1 mol % (entries
4 and 8). On the basis of these observations, asymmetric
J. Org. Chem, Vol. 73, No. 4, 2008 1641

assayed by HPLC to afford the crude product in 98% assay yield
and 88% ee. Crystallization of the product from MeCN afforded
(R)-1 in 91% isolated yield with a 95:5 enantiomeric ratio. While
chiral SFC analyses were performed for ee determination during
screens, chiral HPLC was utilized routinely during larger scale runs.
HPLC conditions: Chiralcel OD-RH, 150 × 4.6 mm i.d., isocratic
1:1 MeCN/0.1% H3PO4 in H2O, 1 mL/min, 20 °C, retention times:
10.90 min for desired and 11.78 min for undesired enantiomer. The
absolute configuration was assigned from the X-ray single-crystal
hydrogenation of carbinamide 2 with 1 mol % of Rh-12c
Josiphos system is a more feasible process for large-scale
productions provided that none of the aforementioned catalyst
inhibitors are present in the starting material.
In summary, we have developed a practical, enantioselective
approach to a pharmaceutically useful spiropiperidineindane-
containing tertiary carbinamide 1. The current route involves a
seven longest linear step, four isolation steps, and 30% overall
yield. The key transformation features a novel, unprecedented
Rh-Josiphos-catalyzed asymmetric hydrogenation of a sterically
hindered tertiary allylic carbinamide 2, which is easily accessible
from commercially available N-Boc-4-piperidone. Specifically,
subjection of 2 to 1 mol % of Rh-12c(BF4) in 10-14 mL/g
THF at 40 °C under 500 psi of H2 affords the corresponding
saturated carbinamide 1 in a 94:6 enantioselectivity, and upon
crystallization from MeCN, the desired product can be isolated
in a 95:5 enantioselectivity and 91% yield.
structure (see Supporting Information): [R]25 ) 33.8 (c 0.4,
D
1
MeOH); mp 213-214 °C (dec); H NMR (500 MHz, CDCl3) δ
7.17 (1H, s), 7.06 (1H, s), 5.72 (1H, br s), 4.31-3.87 (3H, m),
2.87 (2H, m), 2.32 (1H, m), 2.31 (3H, s), 2.00 (3H, s), 1.95 (1H,
dt, J ) 13.1, 4.4 Hz), 1.47 (9H, s), 1.52-1.37 (4H, m), 1.39 (3H,
s), 1.24 (3H, s); 13C NMR (125 MHz, CDCl3) δ 169.7, 155.0, 151.6,
141.4, 134.3, 133.1, 127.5, 123.5, 79.7, 56.3, 48.0, 44.4, 41.4, 37.7,
37.0, 28.6, 25.8, 24.8, 24.4, 20.6; IR (NaCl, cm-1) 3429, 3313 (br),
3056, 2976, 2931, 2865, 1652, 1548, 1429, 1366, 1276, 1171, 1094,
1045, 1003, 958, 870, 770, 737, 704, 614. Anal. Calcd for C24H35-
ClN2O3: C, 66.27; H, 8.11; N, 6.44. Found: C, 66.06; H, 8.28; N,
6.35.
Experimental Section
Rh-Josiphos (12c)-Catalyzed Asymmetric Hydrogenation of
2: In a nitrogen-filled glovebox (typically <10 ppm O2), ligand
12c (26.8 g, 0.042 mol, 0.011 equiv) was combined with
(COD)2RhBF4 (16.3 g, 0.040 mol, 0.01 equiv) in a round-bottom
flask. Dry THF (400 mL, N2 degassed) was added, and the slurry
was stirred for 60 min to give a homogeneous red-brown solution.
The catalyst solution was then poured into a hydrogenation
autoclave, containing a solution of the unsaturated carbinamide 1
(1.73 kg, 3.99 mol, 1 equiv) in THF (KF < 100 ppm). The resulting
mixture was then placed under H2 (500 psig), vented three times
for degassing, and pressurized to 500 psi of H2. The reaction
temperature was increased to 40 °C while maintaining a H2 pressure
of 500 psig. The mixture was agitated at this temperature for 18 h,
at which point no more starting material was observed by HPLC
analysis. The reaction solution was cooled to rt, vented, and then
Acknowledgment. We greatly acknowledge Mr. Bob Reamer
and Lisa DiMichelle for their assistance with NMR interpretation
and structural assignments, and Ms. Mirlinda Biba for develop-
ment of chiral assay. We also acknowledge Enrique Vazquez,
Jason Kowal, Joe Payack, Wenji Li, and Shawn Springfield for
their contributions to the racemic synthesis of the molecule, and
Jennifer Foley for her assistance with crystallographic data.
Supporting Information Available: Additional experimental
procedures, characterization data (PDF), and crystallographic data
for 1 (CIF). This material is available free of charge via the Internet
at http://pubs.acs.org.
JO702429U
1642 J. Org. Chem., Vol. 73, No. 4, 2008