Palladium-Catalyzed Ring Expansion of Spirocyclopropanes to Form Caprolactams and Azepanes

Article abstract of DOI:10.1021/acs.joc.5b01846

A palladium(0)-catalyzed rearrangement of piperidones and piperidines bearing a spirocyclopropane ring was developed. The ring expansion reaction led to a variety of functionalized caprolactam and azepane products in good to excellent yields. Experimental and computational mechanistic studies revealed an initial oxidative addition of the distal carbon-carbon bond of a cyclopropane ring to the palladium(0) catalyst and the relief of ring strain as a driving force for product formation.

Full text of DOI:10.1021/acs.joc.5b01846

Article
pubs.acs.org/joc
Palladium-Catalyzed Ring Expansion of Spirocyclopropanes to Form
Caprolactams and Azepanes
Olivier Rene,*,† Iain A. Stepek,†,§ Alberto Gobbi,† Benjamin P. Fauber,† and Simon Gaines‡
́
†Genentech, Incorporated, Discovery Chemistry, 1 DNA Way, South San Francisco, California 94080, United States
‡Argenta, Discovery Services, Charles River, 7-9 Spire Green Centre, Flex Meadow, Harlow, Essex CM19 5TR, United Kingdom
S
* Supporting Information
ABSTRACT: A palladium(0)-catalyzed rearrangement of piperidones and
piperidines bearing a spirocyclopropane ring was developed. The ring expansion
reaction led to a variety of functionalized caprolactam and azepane products in good
to excellent yields. Experimental and computational mechanistic studies revealed an
initial oxidative addition of the distal carbon−carbon bond of a cyclopropane ring to
the palladium(0) catalyst and the relief of ring strain as a driving force for product
formation.
INTRODUCTION
■
Seven-membered nitrogen-containing heterocycles are prom-
inently featured in biogically active natural products1 and marketed
drugs2 and play a critical role in material science.3 Despite the
prevalence of this structural motif, slow cyclization kinetics have
hindered the development of robust methods for the direct con-
struction of these medium-sized heterocycles.4 At present, these
systems are more commonly accessed via a select few ring-
expansion processes, such as migratory shift reactions, cleavage
of a shared bond in fused [4.1.0] or [3.2.0] bicyclic systems,
and electrocyclization reactions.5 Further exploration of novel
ring-expansion pathways leading to seven-membered nitrogen-
containing heterocycles is therefore of interest.
Because of their unique structural and electronic properties,
cyclopropanes may undergo transition metal-promoted cleavage
of the otherwise inert carbon−carbon bond.6,7 We envisaged that
disubstituted cyclopropane 1 (Figure 1A) could undergo C−C
bond cleavage in the presence of a transition metal via one of
two postulated mechanistic pathways, both leading to common
ring-open π-allyl intermediate 3. Path A would involve an initial
oxidative addition of the C−X bond of cyclopropane 1 to tran-
sition metal species [M], giving rise to metalated intermediate 2.
Subsequent C−C bond cleavage via β-carbon elimination8 would
afford π-allyl intermediate 3. Alternatively, in path B, direct
oxidative addition of the distal C−C bond of cyclopropane 1 to
the metal would lead to metallacycle intermediate 4. This inter-
mediate could then undergo rearrangement to π-allyl inter-
mediate 3 through β-elimination of X. In cases where X and Y
form part of a spirocyclic piperidone or piperidine ring system,
the resulting π-allyl intermediate would represent a ring expanded
metallacycle poised to undergo reductive elimination to the
corresponding caprolactam or azepane, respectively.
Figure 1. Mechanistic pathways for the transition metal-promoted C−C
bond cleavage of cyclopropane 1 and Pd(0)-catalyzed rearrangement of
lactams and piperidines bearing a spirocyclopropane ring.
late-stage diversification. Experimental and computational mech-
anistic studies were also performed to elucidate the mechanistic
pathway of this transformation. To our knowledge, this is the first
reported palladium(0)-catalyzed C−C bond cleavage reaction of
a spirocyclic cyclopropane that led to further functionalized ring-
expanded products.9−11
RESULTS AND DISCUSSION
■
Optimization and Scope. Initial reaction development
and optimization on the Pd(0)-catalyzed rearrangement of lactam
5 using Pd(OAc)2 as a precatalyst revealed the superiority
of electron-rich biarylphosphine ligands (Table 1).12 Although
MePhos (Table 1, entry 1) resulted in low conversion to
Herein, we report the palladium(0)-catalyzed rearrangement
of δ-lactams and saturated N-heterocycles bearing a spirocyclo-
propane ring, which furnished the corresponding 7-membered
heterocycles (Figure 1B). In addition, the ring-expanded products
featured a unique δ-exomethylene functionality primed for further
Received: August 10, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.5b01846
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
a
Table 1. Optimization of the Reaction Conditions
Table 2. Scope of Substituted oxo- and aza-Lactams
a
entry
ligand
base
solvent
conversion
1
2
3
4
5
6
7
8
9
MePhos
DavePhos
RuPhos
BINAP
PPh3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
no base
NaOt-Bu
Cs2CO3
Cs2CO3
Cs2CO3
DMF
12
79
97
2
DMF
DMF
DMF
DMF
0
RuPhos
RuPhos
RuPhos
RuPhos
RuPhos
DMF
38
0
DMF
toluene
t-AmOH
t-AmOH
94
100
b
10
0
a
b
Determined by 1H NMR analysis of the crude reaction mixture. No
Pd(OAc)2 added.
1
rearranged caprolactam 6 (12%), as determined by H NMR
analysis of the crude reaction mixture, significant improve-
ments were achieved with DavePhos (entry 2) and RuPhos
(entry 3), leading to conversions of 79 and 97%, respectively.
Conversely, bidentate ligand BINAP (entry 4) and triphenyl-
phospine (entry 5) resulted in only trace amounts or even no
observed formation of product 6.
Although not necessary for the generatation of product 6,
the addition of a base proved critical to achieve complete conver-
sion. For example, although the combination of Pd(OAc)2 and
RuPhos in conjunction with Cs2CO3 gave rise to almost
complete conversion (entry 3), the same reaction in the absence
of base provided rearranged product 6 in a more modest 38%
conversion (entry 6). From these data, it can be inferred that the
addition of excess carbonate favored the formation of a stabilized
Pd(0)LnX− anionic complex.13 This may represent the resting
state of the catalyst and would explain the increased observed
turnover in the presence of carbonate. Partial decomplexation to
a more active Pd(0)Ln−1X− complex presumably gave rise to the
active catalyst, which underwent oxidative insertion. Conversely,
in the presence of a strong yet sterically hindered base such as
NaOt-Bu (entry 7), the reaction was completely suppressed with
no product formation observed, highlighting the need for a more
coordinating base for the reaction to proceed efficiently.14
Several solvents proved to be tolerated in this reaction with
high conversion also achieved with toluene and DMF (entries 3
and 8, respectively). However, tert-amyl alcohol proved to be
optimal and furnished the rearranged product with complete
conversion in the most consistent manner (entry 9). In the
absence of palladium, the reaction did not proceed and complete
recovery of δ-lactam 5 was observed (entry 10).
Having established optimized conditions for the rearrange-
ment of N-benzyllactam 5, we evaluated the applicability of the
transformation using different substituted δ-lactams, as well as
oxo-lactams and aza-lactams. As illustrated in Table 2, different
N-alkyl substituents were tolerated, and N-benzyllactam 6
and N-methyllactam 7 were isolated in 82 and 75% yields,
respectively. Substitution at the lactam nitrogen was not required
for the Pd(0)-catalyzed ring expansion reaction to proceed
smoothly, as exemplified by the synthesis of NH-lactam 8 in 65%
isolated yield. N-arylated subtrates possessing a broad range of
electronic properties also gave rise to the rearranged caprolactam
a
b
All reactions were performed on 40 mg scale. Reaction ran for 20 min.
products in good to excellent yields. For example, both
N-phenyllactam 9 and electron-poor N-(4-nitrophenyl)lactam
10 were obtained in 74%, and electron-rich N-(4-dimethyl-
aminophenyl)lactam 11 was isolated in 99% yield. Furthermore,
oxo-lactam 12 was also compatible with the reaction conditions
and was delivered in 73% yield. Similarly, a Boc-protected aza-
lactam successfully provided desired heterocyclic product 13 in
56%. However, basic amines were not tolerated, and N-benzyl-
aza-lactam 14 was not formed under the reaction conditions.
N-Substituted piperidines also proved to be efficient sub-
strates for the Pd(0)-catalyzed rearrangement reaction (Table 3).
For example, N-benzoylpiperidine was effectively converted to
the corresponding azepane 15 in 70% yield. Sulfonamide and
urea functionalities were also compatible and resulted in the for-
mation of products 16 and 17 in 76 and 94% yields, respectively.
Lastly, homomorpholine 18 and homopiperazine 19 were
successfully obtained from the corresponding morpholine and
piperazine starting materials in 88 and 48% yields, respectively.
Overall, a broad range of δ-lactams and saturated N-heterocycles
reacted to yield the rearranged products, and the only observed
limitation was the incompatibility of basic amines. It can
therefore be inferred that the highly Lewis basic nature of these
systems inhibited the catalyst by occupying an open coordination
site required for the reaction to proceed, rendering it inactive
toward the desired transformation.15
Mechanistic Studies. There is limited available informa-
tion pertaining to the mechanism of the palladium-catalyzed
C−C bond cleavage of unactivated cyclopropanes. Quantum
mechanical studies reported by Blomberg16 revealed that the
C−C bond of unsubstituted cyclopropane was easier to break
when compared to ethane and cyclobutane. This was due to a
B
DOI: 10.1021/acs.joc.5b01846
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 3. Scope of N-Substituted Azepanes and Other
a
Saturated N-Heterocycles
a
All reactions were performed on 40 mg scale.
low activation barrier for the elimination reaction of metal-
lacyclobutane, which in turn led to a small energy barrier for the
reverse C−C bond oxidative addition process rather than
depending strictly on the innate C−C bond strength.
Activated methylenecyclopropane C−C bond cleavage was
also studied in the context of trimethylenemethane cyclo-
addition reactions.17 The mechanism proposed by Trost18 does
not invoke palladium-mediated C−C bond oxidative addi-
tion. Conversely, Binger19 proposed a mechanism including a
palladacyclobutane intermediate. Subsequent theoretical studies
by Fujimoto20 supported the involvement of a palladacyclobutane
intermediate, but no unifying mechanism had yet been esta-
blished. Yamamoto21 also demonstrated that methylenecyclopro-
panes may undergo ring opening via a palladium-catalyzed C−C
bond oxidative addition to form palladacyclobutane intermediates
in the presence of carbon-based nucleophiles. However, in the
presence of nitrogen nucleophiles, C−C bond cleavage was pro-
posed to proceed through β-carbon elimination.22 More recently,
Fu9 reported the Pd(0)-catalyzed C−C bond cleavage of elec-
tronically activated 1,1-difluoro-2-arylcyclopropanes. Preliminary
DFT calculations demonstrated that initial oxidative addition of a
cyclopropane C−C bond to form a palladacyclobutane inter-
mediate followed by β-fluorine elimination (Figure 1A, path B)
was lower in energy than an initial C−F oxidative addition
followed by β-carbon elimination (Figure 1A, Path A).
Similarly, we applied computational methods to investigate
the two proposed mechanistic pathways for the cyclopropane
ring expansion reaction of lactam 20 illustrated in Figure 2. All
calculations were performed using DFT with the B3LYP density
functional. The LANL2DZ basis set with the Hay−Wadt effec-
tive core potential was used for the Pd atom.23 For all other
atoms, the 6-31* basis set was used. All stationary points were
confirmed to be minima or transition states by computing the
frequencies on optimized structures.24
Figure 2. Computationally calculated mechanistic pathways for the
rearrangement of piperidone 20 to caprolactam 7. (a) Energies are
reported as the sum of the energy of the separated compounds 20 or 7
and catalyst 27.
ligands,25 such as RuPhos,26 which performed best in our reaction.
However, it has been shown that potentially significant differences
in calculated energies may occur when approximating the struc-
ture of ligands.27 With this consideration in mind, we decided to
elect biphenyldimethylphosphine (see Figure 2, complex 27) as
our model ligand for further DFT studies. We conducted initial
optimization on the active monoligated LPd(0) catalyst.28 The
lowest energy conformation of the catalyst displayed an η1 interac-
tion between the biphenyl substituent of the ligand and the
palladium(0) center. We then performed geometry optimizations
to identify low energy conformations of oxidative addition pro-
ducts 22 and 24 on the two pathways shown in Figure 2.29 Opti-
mizations were initiated from multiple input geometries to identify
the lowest energy conformations.24 The corresponding transition
states 21 and 23 were identified using the QST2 method30 in
Gaussian.31 The oxidative addition of the C−N bond of lactam 20
(Figure 2, path A) to palladium catalyst 27 was computed to
present a transition state energy barrier of +32.8 kcal/mol (TS 21),
yielding palladacycle 22 at +8 kcal/mol. Conversely, it was revealed
that the addition of the distal C−C bond of the cyclopropane ring
(Figure 2, path B) had a much lower transition state energy of
+3.9 kcal/mol (TS 23), giving rise to palladacyclobutane 24 at
−6.9 kcal/mol. On the basis of these results, it was concluded that
path B, which involved the formation of a palladacyclobutane
intermediate, was the operative reaction mechanism.32
Because of the more extensive computational power required
for large systems, few reported all-atom DFT studies have
been performed using the full structure of biarylphosphine
It was hypothesized that the relief of the cyclopropane ring
strain was the driving force that allowed for the low activation
C
DOI: 10.1021/acs.joc.5b01846
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
energy barrier at the C−C bond cleavage step in TS 23 and the
generation of energetically downhill palladacyclobutane inter-
mediate 24. In contrast, the high energy barrier associated with
the formation of palladacycle 22 in path A was hypothesized to
be the result of a more sterically encumbered C−N bond pre-
cluding the approach of the active palladium center during the
oxidative addition step. Additionally, intermediate 22 did not
benefit from the relief of cyclopropane ring strain.
Both palladacycle intermediates 22 and 24 could then
converge to common π-allyl complex 25 via a β-carbon elimina-
tion (Figure 2, path A) or a β-nitrogen elimination (Figure 2,
path B), respectively. It was calculated that π-allyl intermediate
25 possessed an energy of −6.6 kcal, similar to palladacyclo-
butane 24. Subsequent reductive elimination via TS 26 was judged
to be rate-determining with an energy barrier of +18.9 kcal/mol
leading to product 7 (−5.6 kcal/mol).33
barrier that would lead to a gem-dimethyl product from the
calculated palladacyclobutane intermediate. As such, only 30
formed in the reaction mixture.
CONCLUSIONS
■
In conclusion, we established Pd(0)-catalyzed conditions for
the ring opening of spirocyclopropanes that led to the forma-
tion of functionalized caprolactams and azepanes in good to
excellent yields. Mechanistic studies uncovered a low-energy
palladacyclobutane intermediate that was energetically favored
due to the relief of ring strain as a driving force. This represents the
first Pd(0)-catalyzed C−C bond cleavage reaction of a spirocyclic
cyclopropane that led to functionalized medium-sized nitrogen-
containing heterocycles. In addition to its synthetic utility, this
process also demonstrated the viability of cyclopropanes as reac-
tive coupling partners in transition-metal-catalyzed synthetic
transformations.
To further analyze our proposed mechanism, we attempted
to experimentally trap the various palladacycle intermediates
under hydrogenative conditions and isolate the hydrogenated
adducts for characterization (Figure 3). Treatment of lactam 28
EXPERIMENTAL SECTION
■
General Information. Reaction mixtures were analyzed on a
UPLC-MS system using formic acid/MeCN mobile phases and a C18
column (1.7 μm, 2.1 × 30 mm). 1H and 13C NMR spectra were
recorded in CDCl3 solutions at 400 MHz for 1H and 101 MHz
for 13C. The internal standard is TMS (0.00 ppm) for the resonance
of protons and residual chloroform (77.0 ppm) for the resonance of
carbons. High resolution mass spectra were obtained by positive EI
and orbitrap mass analysis. Flash column chromatography purification
was performed using disposable normal-phase 15−40 μm silica gel
columns. All reaction solvents were purchased anhydrous from commercial
sources and used as is. The following compounds were purchased from
commercial sources and were used as is 31 [546114-04-9], 32 [1100753-
07-8], 33 [1199794-51-8], 39 [1199794-52-9], 41 [1301739-56-9], 42
[218595-22-3], 43 [886766-28-5].35
Figure 3. Treatment of lactams 28 and 9 under hydrogenation
conditions.
General Procedure for Tables 2 and 3: Ring Expansion
Reactions. A sealable reaction vial was charged with the spirocyclopro-
pane starting material (40 mg), palladium(II) acetate (0.05 equiv),
RuPhos (0.10 equiv), and cesium carbonate (1.5 equiv) and then eva-
cuated under vacuum and backfilled with argon (3×). tert-Amyl alcohol
(0.3M) was added, and the resulting suspension was stirred and heated at
110 °C for 16 h. The reaction was then washed with CH2Cl2 (3 × 5 mL)
and filtered through a pad of Celite. Concentrating the resulting filtrate
followed by silica gel column chromatography (0−60% acetone in
heptane) afforded the corresponding caprolactam or azepane product.
1-Benzyl-6-methylene-azepan-2-one (6). Yield: 33 mg, 82%;
colorless oil; 1H NMR (400 MHz, CDCl3) δ 7.27 (m, 5H), 4.81
(s, 1H), 4.62 (m, 3H), 3.72 (s, 2H), 2.69 (m, 2H), 2.36 (t, J = 6.0 Hz,
2H), 1.83(m, 2H); 13C NMR (101 MHz, CDCl3) δ 174.8, 143.3,
137.5, 128.5, 128.2, 127.3, 113.9, 53.2, 50.1, 37.0, 36.1, 24.0; HRMS
calcd for C14H17NO [M + H]+ 216.1383, found 216.1391.
under our optimized reaction conditions, with the addition of
1 atm of hydrogen, resulted in the formation of hydrogenated
adduct 30 in 15% isolated yield. Other isolated products were
the ring expanded product (9, 45% yield) and the hydrogenated
olefin product (49, 35% yield, see Scheme 6). It was hypo-
thesized that the formation of acyclic amide 30 was the product
of homogeneous hydrogenation of π-allyl complex 29. This
result confirmed the presence of this key π-allyl intermediate
in the reaction mixture. Additionally, subjecting caprolactam
9 to the same reductive reaction conditions gave rise to the
same amount of hydrogenated product (30). The generation
of 30 from both cyclopropylpiperidone 28 and caprolactam 9
demonstrated that all reaction intermediates and products were
in equilibrium under the reaction conditions. This was in agree-
ment with the computed rate-determining energy barrier rela-
tive to the penultimate π-allyl intermediate (−6.6 kcal/mol)
and the final product (−5.6 kcal/mol) energies. Interestingly,
we did not observe a gem-dimethyl product, which would have
resulted from the homogeneous hydrogenation of the calcu-
lated low-energy palladacyclobutane intermediate. Because both
the palladacyclobutane (not shown) and the π-allyl complex
(29) were calculated to possess similar ground state energies
(−6.9 and −6.6 kcal/mol, respectively), it was hypothesized
that both species should also be in equilibrium and in at a
similar concentration in the reaction mixture. By applying the
Curtin−Hammett principle,34 it can be inferred that the cor-
responding barrier that produced hydrogenated amide product
30 from π-allyl complex 29 was lower in energy than the energy
1-Methyl-6-methylene-azepan-2-one (7). Yield: 30 mg, 75%;
colorless oil; 1H NMR (400 MHz, CDCl3) δ 4.88 (m, 2H), 3.80
(d, J = 0.7 Hz, 2H), 2.99 (s, 3H), 2.61(m, 2H), 2.37 (m, 2H), 1.78
(m, 2H); 13C NMR (101 MHz, CDCl3) δ 174.7, 143.4, 113.6, 56.0,
36.9, 35.8, 34.8, 24.0; HRMS calcd for C8H13NO [M + H]+ 140.1070,
found 140.1075.
6-Methyleneazepan-2-one (8). Yield: 26 mg, 65%; colorless oil; 1H
NMR (400 MHz, CDCl3) δ 6.28 (s, 1H), 4.88 (d, J = 12 Hz, 2H), 3.67
(d, J = 6.3 Hz, 2H), 2.54 (m, 2H), 2.42 (m, 2H), 1.79 (m, 2H); 13C
NMR (101 MHz, CDCl3) δ 177.6, 144.4, 114.0, 47.7, 37.4, 35.5, 23.5;
HRMS calcd for C7H11NO [M + H]+ 126.0914, found 126.0913.
6-Methylene-1-phenyl-azepan-2-one (9). Yield: 29.5 mg, 74%;
colorless oil; 1H NMR (400 MHz, CDCl3) δ 7.36 (m, 2H), 7.22
(m, 3H), 4.96 (s, 1H), 4.83(s, 1H), 4.20 (s, 2H), 2.78 (m, 2H), 2.49
(m, 2H), 1.92 (m, 2H); 13C NMR (101 MHz, CDCl3) δ 174.3, 144.0,
143.8, 129.1, 126.7, 126.6, 114.2, 57.5, 36.7, 36.4, 23.7; HRMS calcd
for C13H15NO [M + H]+ 202.1227, found 202.1233.
D
DOI: 10.1021/acs.joc.5b01846
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
6-Methylene-1-(4-nitrophenyl)azepan-2-one (10). Reaction was
run for 20 min instead of 16 h. Yield: 29.5 mg, 74%; colorless oil; H
167.1, 166.7, 150.3, 150.0, 139.6, 139.3, 138.3, 137.9, 137.4, 131.4,
131.3, 124.8, 123.8, 123.5, 121.9, 121.7, 110.2, 108.6, 107.1, 75.4, 75.3,
50.5, 48.0, 47.7, 47.0, 46.4, 46.1, 45.5, 44.9, 44.4, 44.0, 43.9, 43.5, 23.6;
HRMS calcd for C18H25N2O3 [M + H]+ 317.1860, found 317.1863.
8-Methyl-8-azaspiro[2.5]octan-7-one (20) (Scheme 1). A round-
bottomed flask was charged with 8-azaspiro[2.5]octan-7-one (200 mg,
1
NMR (400 MHz, CDCl3) δ 8.23 (d, J = 9.1 Hz, 2H), 7.44 (d, J =
8.9 Hz, 2H), 5.07−5.01 (m, 1H), 4.95−4.88 (m, 1H), 4.29 (s, 2H),
2.86−2.77 (m, 2H), 2.51 (td, J = 6.2, 1.1 Hz, 2H), 2.01−1.86 (m, 2H);
13C NMR (101 MHz, CDCl3) δ 174.2, 149.4, 145.4, 143.2, 126.7,
124.4, 114.7, 57.0, 36.4, 36.2, 23.3; HRMS calcd for C13H14N2O3
[M + H]+ 247.1077, found 247.1093.
Scheme 1. Preparation of Starting Materials 20 and 34-36
1-(4-(Dimethylamino)phenyl)-6-methyleneazepan-2-one (11).
1
Yield: 39.5 mg, >95%; colorless oil; H NMR (400 MHz, CDCl3) δ
7.10−6.97 (m, 2H), 6.78−6.63 (m, 2H), 4.98−4.90 (m, 1H), 4.87−
4.79 (m, 1H), 4.15 (s, 2H), 2.94 (s, 6H), 2.80−2.71 (m, 2H), 2.52−
2.44 (m, 2H), 1.97−1.85 (m, 2H); 13C NMR (101 MHz, CDCl3)
δ 174.6, 149.3, 144.0, 133.5, 127.1, 113.9, 112.9, 57.9, 40.7, 36.9,
36.4, 23.8; HRMS calcd for C15H20N2O [M + H]+ 245.1648, found
245.1667.
4-Benzyl-6-methylene-1,4-oxazepan-3-one (12). Yield: 29 mg,
1
73%; colorless oil; H NMR (400 MHz, CDCl3) δ 7.29 (m, 5H),
5.07 (s, 1H), 4.93 (s, 1H), 4.66 (s, 2H), 4.32 (s, 2H), 4.23(s, 2H),
3.85(s, 2H); 13C NMR (101 MHz, CDCl3) δ 171.0, 143.0, 136.7,
128.6, 128.1, 127.5, 116.3, 73.7, 71.8, 51.6, 50.5; HRMS calcd for
C13H15NO2 [M + H]+ 218.1176, found 218.1182.
tert-Butyl 4-Isopropyl-6-methylene-3-oxo-1,4-diazepane-1-car-
boxylate (13). Yield: 22.5 mg, 56%; colorless oil; 1H NMR
(400 MHz, CDCl3) δ 5.15 (m, 2H), 4.86 (m, 1H), 4.13 (m, 4H),
3.66 (m, 2H), 1.46 (s, 9H), 1.13 (d, J = 6.7 Hz, 6H); 13C NMR
(101 MHz, CDCl3) δ 168.7, 155.2, 142.3, 116.3, 80.8, 52.1, 51.4, 45.5,
44.9, 28.2, 19.6; HRMS calcd for C14H24N2O3 [M + H]+ 269.1860,
found 269.1867.
1.60 mmol) and N,N-dimethylformamide (5.3 mL). The resulting
solution was cooled to 0 °C, and sodium hydride (60% in mineral oil,
51 mg, 1.3 mmol) was added, forming a yellow suspension. The sus-
pension was then allowed to warm to room temperature, and
iodomethane (0.13 mL, 2.1 mmol) was added. The reaction mixture
was stirred for 16 h. Upon completion, the reaction was quenched with
H2O (1 mL), extracted with ethyl acetate (3 × 3 mL), and dried with
anhydrous MgSO4. Concentrating the organic layer followed by silica
gel column chromatography (0−60% acetone in heptane) afforded
8-methyl-8-azaspiro[2.5]octan-7-one (64.5 mg, 29% yield) as a colorless
1
gel. H NMR (400 MHz, CDCl3) δ 2.70 (s, 3H), 2.51 (d, J = 6.0 Hz,
(3-Methyleneazepan-1-yl)(phenyl)methanone (15). Yield: 28 mg,
70%; colorless oil; 1H NMR (400 MHz, CDCl3, reported a 2:1 mixture
of rotamers) δ 7.46−7.31 (m, 5H), 5.00 (d, J = 11.3 Hz, 2/3H), 4.86
(s, 2/3H), 4.78−4.66 (m, 2/3H), 4.34 (s, 2/3H), 4.02 (s, 4/3H),
3.67−3.56 (m, 4/3H), 3.38−3.24 (m, 2/3H), 2.41−2.29 (m, 2/3H),
2.28−2.17 (m, 4/3H), 1.91−1.77 (m, 4/3H), 1.75−1.59 (m, 2H),
1.59−1.48 (m, 2/3H); 13C NMR (101 MHz, CDCl3, reported as a
mixture of rotamers) δ 172.0, 171.1, 147.6, 146.4, 137.0, 136.7, 129.2,
129.1, 128.3, 126.7, 126.4, 114.0, 112.3, 56.8, 52.6, 49.0, 46.5, 34.9,
34.6, 30.6, 29.8, 28.3, 28.2; HRMS calcd for C14H17NO [M + H]+
216.1383, found 216.1388.
3-Methylene-1-(phenylsulfonyl)azepane (16). Yield: 30.5 mg,
76%; colorless oil; 1H NMR (400 MHz, CDCl3) δ 7.84−7.78
(m, 2H), 7.61−7.47 (m, 3H), 4.90−4.86 (m, 1H), 4.84−4.81 (m, 1H),
3.94−3.87 (m, 2H), 3.23−3.16 (m, 2H), 2.32−2.24 (m, 2H), 1.79−
1.69 (m, 2H), 1.64−1.53 (m, 2H); 13C NMR (101 MHz, CDCl3) δ
146.8, 139.4, 132.3, 129.0, 126.9, 113.2, 54.6, 48.1, 34.0, 30.5, 29.3;
HRMS calcd for C13H17NO2S [M + H]+ 252.1053, found 252.1058.
N-Methyl-3-methylene-N-phenylazepane-1-carboxamide (17).
Yield: 37.5 mg, 94%; colorless oil; 1H NMR (400 MHz, CDCl3)
δ 7.38−7.28 (m, 2H), 7.17−7.00 (m, 3H), 4.76 (d, J = 1.9 Hz, 1H),
4.70−4.62 (m, 1H), 3.65 (s, 2H), 3.26−3.12 (m, 5H), 2.24−2.07
(m, 2H), 1.67−1.57 (m, 2H), 1.57−1.48 (m, 2H); 13C NMR
(101 MHz, CDCl3) δ 161.8, 148.2, 147.2, 129.4, 124.3, 123.8, 112.3,
55.5, 48.0, 40.0, 34.4, 29.4, 28.6; HRMS calcd for C15H20N2O [M + H]+
245.1649, found 245.1658.
N-Methyl-3-methylene-N-phenylazepane-1-carboxamide (18).
Yield: 35 mg, 88%; colorless oil; 1H NMR (400 MHz, CDCl3,
reported as a 2:1 mixture of rotamers) δ 7.47−7.33 (m, 5H), 5.27
(br s, 1/3H), 5.11 (br s, 1/3H), 4.99 (br s, 2/3H), 4.72 (br s, 2/3H),
4.42 (br s, 2/3H), 4.29 (br s, 2/3H), 4.24 (br s, 4/3H), 4.11 (br s,
4/3H), 3.88 (br s, 2 2/3H), 3.64 (br s, 2/3H), 3.52 (br s, 2/3H); 13C
NMR (101 MHz, CDCl3, reported as a mixture of rotamers) δ 166.8,
140.6, 139.3, 131.5, 131.2, 124.9, 124.8, 123.8, 123.6, 122.0, 121.8,
109.9, 108.4, 69.8, 69.6, 67.8, 67.3, 50.1, 47.9, 45.5, 44.9; HRMS calcd
for C13H16NO2 [M + H]+ 218.1176, found 218.1179.
N-Methyl-3-methylene-N-phenylazepane-1-carboxamide (19).
Yield: 19 mg, 48%; colorless oil; 1H NMR (400 MHz, CDCl3,
reported as a mixture of rotamers) δ 7.49−7.29 (m, 5H), 5.38−4.47
(m, 2H), 4.29−3.79 (m, 5H), 3.71−3.25 (m, 3H), 1.46 (s, 9H); 13C
NMR (101 MHz, CDCl3, reported as a mixture of rotamers) δ 167.2,
2H), 1.86 (m, 2H), 1.69 (m, 2H), 1.01 (d, J = 6.0 Hz, 2H), 0.58
(m, 2H); 13C NMR (101 MHz, CDCl3) δ 172.3, 40.3, 33.3, 32.2, 27.5,
19.3, 9.8; HRMS calcd for C8H13NO [M + H]+ 140.1070, found
140.1070.
8-Benzyl-8-azaspiro[2.5]octan-7-one (34) (Scheme 1). A round-
bottomed flask was charged with 8-azaspiro[2.5]octan-7-one (200 mg,
1.60 mmol) and N,N-dimethylformamide (5.3 mL). The resulting
solution was cooled to 0 °C, and sodium hydride (60% in mineral oil,
51 mg, 1.3 mmol) was added, forming a yellow suspension. The sus-
pension was then allowed to warm to room temperature, and benzyl
bromide (0.25 mL, 2.1 mmol) was added. The reaction mixture was
stirred for 16 h. Upon completion, the reaction was quenched with
H2O (1 mL), extracted with ethyl acetate (3 × 3 mL), and dried with
anhydrous MgSO4. Concentrating the organic layer followed by silica gel
column chromatography (0−60% acetone in heptane) afforded 8-benzyl-
8-azaspiro[2.5]octan-7-one (26.5 mg, 77% yield) as a white amorphous
solid. 1H NMR (400 MHz, CDCl3) δ 7.35−7.09 (m, 5H), 4.46 (s, 2H),
2.63 (t, J = 7.1 Hz, 2H), 1.93 (p, J = 6.9 Hz, 2H), 1.63 (t, J = 6.8 Hz,
2H), 0.93−0.85 (m, 2H), 0.60−0.52 (m, 2H); 13C NMR (101 MHz,
CDCl3) δ 173.3, 138.7, 128.5, 126.8, 126.7, 44.8, 39.9, 33.4, 32.2, 19.3,
11.2; HRMS calcd for C14H17NO [M + H]+ 216.1383, found 216.1388.
8-Benzyl-5-oxa-8-azaspiro[2.5]octan-7-one (35) (Scheme 1). A
round-bottomed flask was charged with 5-oxa-8-azaspiro[2.5]octan-7-
one (200 mg, 1.57 mmol) and N,N-dimethylformamide (3.2 mL). The
resulting solution was cooled to 0 °C, and sodium hydride (60% in
mineral oil, 47 mg, 1.2 mmol) was added, forming a yellow suspension.
The suspension was then allowed to warm to room temperature, and
benzyl bromide (0.24 mL, 2.0 mmol) was added. The reaction mixture
was stirred for 16 h. Upon completion, the reaction was quenched with
H2O (1 mL), extracted with ethyl acetate (3 × 3 mL), and dried with
anhydrous MgSO4. Concentrating the organic layer followed by silica gel
column chromatography (0−60% acetone in heptane) afforded 8-benzyl-
8-azaspiro[2.5]octan-7-one (177 mg, 52% yield) as a white amorphous
1
solid. H NMR (400 MHz, CDCl3) δ 7.36−7.27 (m, 2H), 7.27−7.21
(m, 1H), 7.19 (dd, J = 7.5, 1.7 Hz, 2H), 4.45 (s, 2H), 4.43 (s, 2H), 3.69
(s, 2H), 1.02−0.95 (m, 2H), 0.71−0.65 (m, 2H); 13C NMR (101 MHz,
CDCl3) δ 168.5, 137.6, 128.7, 127.2, 126.4, 72.5, 68.2, 43.1, 39.0, 7.8;
HRMS calcd for C13H15NO2 [M + H]+ 218.1176, found 218.1185
tert-Butyl 8-Isopropyl-7-oxo-5,8-diazaspiro[2.5]octane-5-carbox-
ylate (36) (Scheme 1). A round-bottomed flask was charged with tert-
butyl 7-oxo-5,8-diazaspiro[2.5]octane-5-carboxylate (200 mg, 0.88 mmol)
E
DOI: 10.1021/acs.joc.5b01846
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
and N,N-dimethylformamide (2.9 mL). The resulting solution was cooled
to 0 °C, and sodium hydride (60% in mineral oil, 53 mg, 1.3 mmol) was
added, forming a yellow suspension. The suspension was then allowed to
warm to room temperature, and 2-iodopropane (0.26 mL, 2.7 mmol) was
added. The reaction mixture was stirred for 16 h. Upon completion, the
reaction was quenched with H2O (1 mL), extracted with ethyl acetate
(3 × 3 mL), and dried with anhydrous MgSO4. Concentrating the
organic layer followed by silica gel column chromatography (0−60%
acetone in heptane) afforded tert-butyl 8-isopropyl-7-oxo-5,8-
diazaspiro[2.5]octane-5-carboxylate (85 mg, 36% yield) as a white
amorphous solid. 1H NMR (400 MHz, CDCl3) δ 4.20 (m, 3H), 3.23
(m, 2H), 1.46 (s, 9H), 1.26 (d, J = 7.0 Hz, 6H), 1.15 (d, J = 6.0 Hz, 2H),
0.92 (s, 2H); 13C NMR (101 MHz, CDCl3, major rotamer reported) δ
170.1, 154.1, 80.3, 52.4, 49.0, 46.8, 37.1, 28.3, 20.8, 12.0; HRMS calcd for
C14H24N2O3 [M + H]+ 269.1860, found 269.1870.
slurry was heated at 100 °C and stirred for 20 h. The reaction was then
quenched with H2O (1 mL), diluted with EtOAc (3 × 5 mL), washed
with saturated aqueous NaHCO3 and brine, and dried with anhydrous
MgSO4. Concentrating the organic layer followed by silica gel column
chromatography (0−60% acetone in heptane) afforded 8-(4-dimethyla-
minophenyl)-8-azaspiro[2.5]octan-7-one (148.5 mg, 38% yield) as a
1
yellow amorphous solid. H NMR (400 MHz, CDCl3) δ 6.89 (d, J =
8.9 Hz, 2H), 6.68 (d, J = 8.9 Hz, 2H), 2.94 (s, 6H), 2.66 (t, J = 13.9 Hz,
2H), 2.05−1.96 (m, 2H), 1.88−1.81 (m, 2H), 0.66−0.60 (m, 2H), 0.60−
0.53 (m, 2H); 13C NMR (101 MHz, CDCl3) δ 172.7, 149.5, 129.2, 127.0,
112.6, 41.8, 40.6, 33.3, 32.7, 19.6, 11.4; HRMS calcd for C15H20N2O
[M + H]+ 245.1648, found 245.1660.
5-Benzyl-5,8-diazaspiro[2.5]octan-7-one (40) (Scheme 3). A
round-bottomed flask was charged with 5,8-diazaspiro[2.5]octan-7-
8-Phenyl-8-azaspiro[2.5]octan-7-one (28) (Scheme 2). A sealable
reaction vial was charged with 8-azaspiro[2.5]octan-7-one (200 mg,
Scheme 3. Preparation of Starting Material 40
Scheme 2. Preparation of Starting Materials 28, 37, and 38
one hydrochloride (200 mg, 1.59 mmol) and dichloromethane (5.3 mL),
and benzaldehyde (0.16 mL, 1.59 mmol) was added. Sodium triacetoxy-
borohydride (531 mg, 2.38 mmol) and acetic acid (0.14 mL, 2.38 mmol)
were then added, and the reaction was allowed to stir at room tem-
perature for 16 h. The reaction was then quenched with H2O (1 mL),
diluted with EtOAc (3 × 5 mL), washed with 1 M HCl, saturated
aqueous NaHCO3, and brine, and dried with anhydrous MgSO4. Con-
centrating the organic layer followed by silica gel column chromatog-
raphy (0−60% acetone in heptane) afforded 5-benzyl-5,8-diazaspiro-
[2.5]octan-7-one (123.5 mg, 36% yield) as a white amorphous solid. 1H
NMR (400 MHz, CDCl3) δ7.30 (m, 5H), 3.62 (s, 2H), 3.25 (s, 2H),
2.53 (s, 2H), 0.75 (m, 2H), 0.65 (m, 2H); 13C NMR (101 MHz,
CDCl3) δ 170.6, 137.1, 128.8, 128.4, 127.4, 61.1, 56.3, 56.1, 34.9, 11.7;
HRMS calcd for C13H16N2O [M + H]+ 217.1336, found 217.1345.
Phenyl(4-azaspiro[2.5]octan-4-yl)methanone (44) (Scheme 4).
8-Azaspiro[2.5]octane hydrochloride (0.450 g, 3.05 mmol) and potassium
1.6 mmol), copper(I) iodide (30 mg, 0.154 mmol), and potassium
phosphate (754 mg, 3.5 mmol) and then evacuated under vacuum and
backfilled with argon (3×). Toluene (3.2 mL), N,N-dimethylethyle-
nediamine (0.035 mL, 0.32 mmol), and iodobenzene (0.23 mL,
2.1 mmol) were then added, and the resulting slurry was heated at
90 °C and stirred for 16 h. The reaction was then quenched with H2O
(1 mL), diluted with EtOAc (3 × 5 mL), washed with 1 M HCl, satu-
rated aqueous NaHCO3, and brine, and dried with anhydrous MgSO4.
Concentrating the organic layer followed by silica gel column chro-
matography (0−60% acetone in heptane) afforded 8-phenyl-8-
azaspiro[2.5]octan-7-one (190 mg, 59% yield) as a white amorphous
solid. 1H NMR (400 MHz, CDCl3) δ 7.36 (m, 2H), 7.27 (m, 1H), 7.07
(m, 2H), 2.69 (d, J = 8.0 Hz, 2H), 2.03 (d, J = 7.0 Hz, 2H), 1.88 (d, J =
8.0 Hz, 2H), 0.64 (s, 4H); 13C NMR (101 MHz, CDCl3) δ 172.7, 138.4,
128.9, 128.5, 127.2, 41.6, 33.1, 32.7, 19.4, 12.1; HRMS calcd for
C13H15NO [M + H]+ 202.1227, found 202.1231.
Scheme 4. Preparation of Starting Materials 44−46
8-(4-Nitrophenyl)-8-azaspiro[2.5]octan-7-one (37) (Scheme 2). A
sealable reaction vial was charged with 8-azaspiro[2.5]octan-7-one
(200 mg, 1.6 mmol), copper(I) iodide (30 mg, 0.154 mmol), and
potassium phosphate (754 mg, 3.5 mmol) and then evacuated under
vacuum and backfilled with argon (3×). Toluene (3.2 mL), N,N-
dimethylethylenediamine (0.035 mL, 0.32 mmol), and 1-iodo-4-
nitrobenzene (522 mg, 2.1 mmol) were then added, and the resulting
slurry was heated at 100 °C and stirred for 16 h. The reaction was then
quenched with H2O (1 mL), diluted with EtOAc (3 × 5 mL), washed
with 1 M HCl, saturated aqueous NaHCO3, and brine, and dried with
anhydrous MgSO4. Concentrating the organic layer followed by silica gel
column chromatography (0−60% acetone in heptane) afforded 8-(4-
nitrophenyl)-8-azaspiro[2.5]octan-7-one (240 mg, 61% yield) as a white
phosphate dibasic (0.5 M in H2O, 10 mL) were combined at room
temperature. Benzoyl chloride (450 mg, 3.20 mmol) was added dropwise
to the stirred solution. A precipitate formed as the reagent was added. The
resulting slurry was stirred for 15 min at room temperature, and then
CPME (6 mL) was added. The organic fraction was isolated, and the
aqueous layer was rinsed with CPME (6 mL) a second time. The organic
fractions were combined, dried over Na2SO4, filtered, and concentrated
under reduced pressure. The resulting oil was purified by column chro-
matography (25% EtOAc in heptane) to yield phenyl(4-azaspiro[2.5]-
octan-4-yl)methanone as a white amorphous solid (510 mg, 74%).
1H NMR (400 MHz, CDCl3) δ 7.50−7.30 (m, 5H), 3.53 (br s, 2H), 1.80
(br s, 2H), 1.61 (br s, 4H), 0.58 (br s, 4H); 13C NMR (101 MHz, CDCl3,
reported as mixture of rotamers) δ 171.8, 169.7, 137.4, 129.6, 128.1,
127.1, 49.9, 44.8, 39.9, 38.3, 35.3, 32.2, 26.5, 24.1, 14.8; HRMS calcd for
C14H17NO [M + H]+ 216.1383, found 216.1388.
1
amorphous solid. H NMR (400 MHz, CDCl3) δ 8.22 (d, J = 9.1 Hz,
2H), 7.36 (d, J = 9.1 Hz, 2H), 2.74 (t, J = 7.1 Hz, 2H), 2.14−2.03 (m,
2H), 1.98−1.88 (m, 2H), 0.91−0.83 (m, 2H), 0.74−0.66 (m, 2H); 13C
NMR (101 MHz, CDCl3) δ 173.4, 145.5, 144.7, 127.8, 124.1, 41.47,
32.9, 32.1, 19.0, 13.9; HRMS calcd for C13H14N2O3 [M + H]+ 247.1077,
found 247.1093.
8-(4-Dimethylaminophenyl)-8-azaspiro[2.5]octan-7-one (38)
(Scheme 2). A sealable reaction vial was charged with 8-azaspiro[2.5]-
octan-7-one (200 mg, 1.6 mmol), copper(I) iodide (30 mg, 0.154 mmol),
and potassium phosphate (754 mg, 3.5 mmol) and then evacuated
under vacuum and backfilled with argon (3×). Toluene (3.2 mL), N,N-
dimethylethylenediamine (0.035 mL, 0.32 mmol), and 4-iodo-N,N-
dimethylaniline (518 mg, 2.1 mmol) were then added, and the resulting
Phenyl(7-oxa-4-azaspiro[2.5]octan-4-yl)methanone (45)
(Scheme 4). 8-Azaspiro[2.5]octane hydrochloride (0.50 g, 3.3 mmol)
and potassium phosphate dibasic (0.5 M in H2O, 10 mL) were combined
F
DOI: 10.1021/acs.joc.5b01846
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
at room temperature. Benzoyl chloride (0.43 mL, 3.7 mmol) was added
dropwise to the stirred solution. A precipitate formed as the reagent was
added. The resulting slurry was stirred for 15 min at room temperature,
and then CPME (6 mL) was added. The organic fraction was isolated,
and the aqueous layer was rinsed with CPME (6 mL) a second time.
The organic fractions were combined, dried over Na2SO4, filtered, and
concentrated under reduced pressure. The resulting oil was purified
by column chromatography (25% EtOAc in heptane) to yield phenyl-
(7-oxa-4-azaspiro[2.5]octan-4-yl)methanone as a white amorphous solid
(466 mg, 58%). 1H NMR (400 MHz, CDCl3) δ 7.52−7.31 (m, 5H), 3.88−
3.67 (m, 4H), 3.61 (s, 2H), 0.97−0.73 (m, 4H); 13C NMR (101 MHz,
CDCl3, reported as mixture of rotamers) δ 170.9, 136.6, 130.2, 129.6, 128.5,
128.3, 127.5, 126.9, 73.6, 68.9, 67.3, 67.1, 48.1, 39.3, 13.6, 10.5; HRMS
calcd for C13H16NO2 [M + H]+ 218.1176, found 218.1179.
Reaction of Piperidine 26 under Hydrogenation Conditions
(Scheme 6). A sealable reaction vial was charged with 4-phenyl-4-
Scheme 6. Reaction of Piperidine 26 under Hydrogenation
Conditions
tert-Butyl 4-benzoyl-4,7-diazaspiro[2.5]octane-7-carboxylate
(46) (Scheme 4). tert-Butyl 4,7-diazaspiro[2.5]octane-7-carboxylate
(0.537 g, 2.5 mmol) and potassium phosphate dibasic (0.5 M in
H2O, 10 mL) were combined at room temperature. Benzoyl chloride
(0.32 mL, 2.7 mmol) was added dropwise to the stirred solution.
A precipitate formed as the reagent was added. The resulting slurry
was stirred for 15 min at room temperature, and then CPME (6 mL)
was added. The organic fraction was isolated, and the aqueous layer was
rinsed with CPME (6 mL) a second time. The organic fractions were
combined, dried over Na2SO4, filtered, and concentrated under reduced
pressure. The resulting oil was purified by column chromatography (25%
EtOAc in heptane) to yield tert-butyl 4-benzoyl-4,7-diazaspiro[2.5]octane-
azaspiro[2.5]octan-5-one (50 mg, 0.248 mmol), palladium(II) acetate
(2.8 mg, 0.012 mmol), RuPhos (11.8 mg, 0.025 mmol), and cesium
carbonate (121 mg, 0.37 mmol) and then evacuated under vacuum
and backfilled with argon (3×). tert-Amyl alcohol (1 mL) was added,
and the reaction mixture was evacuated under vacuum and backfilled
with hydrogen (3×). A hydrogen-filled balloon was placed on top of
the reaction vessel. The resulting suspension was heated at 110 °C and
stirred for 16 h. The reaction was washed with CH2Cl2 (3 × 5 mL)
and filtered through Celite. Concentrating the filtrate followed by
silica gel column chromatography (0−60% acetone in heptane) afforded
5-methyl-N-phenylhex-5-enamide (7.5 mg, 15% yield) as a colorless oil,
6-methylene-1-phenylazepan-2-one (22.5 mg, 45% yield) as a color-
less oil, and 6-methyl-1-phenylazepan-2-one (17.5 mg, 35% yield) as a
colorless oil.
1
7-carboxylate as a white amorphous solid (720 mg, 91%). H NMR
(400 MHz, CDCl3) δ 7.51−7.29 (m, 5H), 3.77−3.60 (m, 2H), 3.54−3.45
(m, 2H), 3.43 (s, 2H), 1.46 (s, 9H), 0.90−0.72 (m, 4H); 13C NMR
(101 MHz, CDCl3, reported as a mixture of rotamers) δ 171.0, 154.8,
136.5, 133.0, 130.1, 129.9, 128.2, 127.3, 80.0, 67.5, 51.0, 46.9, 44.1, 38.7,
28.3, 21.7, 21.2, 14.1; HRMS calcd for C18H25N2O3 [M + H]+ 317.1860,
found 317.1863.
8-(Benzenesulfonyl)-8-azaspiro[2.5]octane (47) (Scheme 5). To a
solution of 4-azaspiro[2.5]octane hydrochloride (150 mg, 1.01 mmol)
5-Methyl-N-phenylhex-5-enamide (30). 1H NMR (400 MHz,
CDCl3) δ 7.56−7.46 (m, 2H), 7.36−7.29 (m, 2H), 7.13−7.07 (m, 1H),
4.77 (s, 1H), 4.72 (s, 1H), 2.41−2.29 (m, 2H), 2.12 (t, J = 7.5 Hz, 2H),
1.96−1.84 (m, 2H), 1.74 (s, 3H), 1.61 (s, 1H); 13C NMR (101 MHz,
CDCl3, CO carbon could not be differentiated from baseline) δ
144.9, 137.8, 129.0, 124.2, 119.6, 110.8, 37.0, 36.9, 23.1, 22.1; HRMS
calcd for C13H17NO [M + H]+ 204.1383, found 204.1388.
Scheme 5. Preparation of Starting Materials 47 and 48
6-Methyl-1-phenylazepan-2-one (49). 1H NMR (400 MHz,
CDCl3) δ 7.43−7.32 (m, 2H), 7.25−7.15 (m, 3H), 3.69 (dd, J =
15.0, 9.2 Hz, 1H), 3.46 (dt, J = 15.0, 1.5 Hz, 1H), 2.78−2.62 (m, 2H),
2.02−1.93 (m, 2H), 1.84−1.63 (m, 2H), 1.48−1.29 (m, 1H), 0.94
(d, J = 6.8 Hz, 3H); 13C NMR (101 MHz, CDCl3) δ 175.4, 144.7,
129.1, 126.4, 126.2, 59.1, 38.3, 37.5, 34.0, 22.7, 20.0; HRMS calcd for
C13H17NO [M + H]+ 204.1383, found 204.1388.
in dichloromethane (5 mL) was added triethylamine (0.43 mL,
3.04 mmol) and benzenesulfonyl chloride (0.19 mL, 1.52 mmol), and
the reaction was stirred at room temperature for 16 h. The reaction was
then concentrated on silica gel and purified by silica gel column chro-
matography (0−100% EtOAc in heptane) to give 8-(benzenesulfonyl)-8-
azaspiro[2.5]octane (239 mg, 94% yield) as a white amorphous solid.
1H NMR (400 MHz, CDCl3) δ 7.90−7.79 (m, 2H), 7.60−7.42 (m, 3H),
3.64−3.49 (m, 2H), 1.71−1.61 (m, 2H), 1.61−1.51 (m, 2H), 1.31−1.19
(m, 2H), 1.05−0.94 (m, 2H), 0.60−0.52 (m, 2H); 13C NMR (101 MHz,
CDCl3) δ 142.6, 132.2, 128.9, 127.1, 48.1, 39.3, 32.0, 24.3, 23.7, 13.9;
HRMS calcd for C13H17NO2S [M + H]+ 252.1053, found 252.1064.
N-Methyl-N-phenyl-8-azaspiro[2.5]octane-8-carboxamide (48)
(Scheme 5). To a solution of 4-azaspiro[2.5]octane hydrochloride
(150 mg, 1.01 mmol) in dichloromethane (5 mL) was added triethylamine
(0.43 mL, 3.04 mmol) and N-methyl-N-phenylcarbamoyl chloride
(258 mg, 1.52 mmol), and the reaction was stirred at room temperature
for 16 h. The reaction was then concentrated on silica gel and purified
by silica gel column chromatography (0−100% EtOAc in heptane) to give
N-methyl-N-phenyl-8-azaspiro[2.5]octane-8-carboxamide (222 mg, 90%
yield) as a white amorphous solid. 1H NMR (400 MHz, CDCl3) δ 7.32−
7.24 (m, 2H), 7.11−6.99 (m, 3H), 3.21 (s, 3H), 3.04−2.91 (m, 2H),
1.68−1.56 (m, 2H), 1.51−1.31 (m, 4H), 0.82−0.72 (m, 2H), 0.67−0.54
(m, 2H); 13C NMR (101 MHz, CDCl3) δ 161.0, 146.7, 129.0, 124.3,
124.1, 48.1, 38.8, 38.7, 31.6, 26.0, 24.0, 14.7; HRMS calcd for C15H20N2O
[M + H]+ 245.1649, found 245.1658.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.5b01846.
Copies of 1H, 13C NMR spectra and computational
methods (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*Phone: (650) 467-0236. E-mail: rene.olivier@gene.com.
Present Address
§I.A.S.: ETH Zurich Laboratory of Organic Chemistry, HCI F
306, Vladimir-Prelog-Weg 3, 8093 Zurich, Switzerland
Notes
The authors declare no competing financial interest.
G
DOI: 10.1021/acs.joc.5b01846
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(21) (a) Nakamura, I.; Saito, S.; Yamamoto, Y. J. Am. Chem. Soc.
2000, 122, 2661. (b) Nakamura, I.; Siriwardana, A. I.; Saito, S.;
Yamamoto, Y. J. Org. Chem. 2002, 67, 3445.
(22) (a) Nakamura, I.; Itagaki, H.; Yamamoto, Y. J. Org. Chem. 1998,
63, 6458. (b) Siriwardana, A. I.; Kamada, M.; Nakamura, I.;
Yamamoto, Y. J. Org. Chem. 2005, 70, 5932.
ACKNOWLEDGMENTS
■
We thank Baiwei Lin and Christopher Tom for HMRS support
and Yanzhou Liu for NMR support. We also thank Malcolm
Huestis and Lauren Sirois for sharing relevant literature reviews
in the preparation of this manuscript.
(23) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.
(24) Further details are available in the Supporting Information.
(25) For selected examples, see: (a) Hicks, J. D.; Hyde, A. M.;
Cuezva, A. M.; Buchwald, S. L. J. Am. Chem. Soc. 2009, 131, 16720.
REFERENCES
■
(1) (a) Heathcock, C. H.; Blumenkopf, T. A.; Smith, K. M. J. Org.
Chem. 1989, 54, 1548. (b) Lin, W.-H.; Xu, R.-S.; Wang, R.-J.; Mak, T.
C. W. J. Crystallogr. Spectrosc. Res. 1991, 21, 189. (c) Furstner, A.;
(b) Dufert, M. A.; Billingsley, K. L.; Buchwald, S. L. J. Am. Chem. Soc.
̈
̈
2013, 135, 12877. (c) Zhou, Y.; Zhang, X.; Liang, H.; Cao, Z.; Zhao,
X.; He, Y.; Wang, S.; Pang, J.; Zhou, Z.; Ke, Z.; Qiu, L. ACS Catal.
2014, 4, 1390.
Thiel, O. R. J. Org. Chem. 2000, 65, 1738.
(2) (a) Heck, H. d’A.; Buttrill, S. E., Jr; Flynn, N. W.; Dyer, R. L.;
Anbar, M.; Cairns, T.; Dighe, S.; Cabana, B. E. J. Pharmacokinet.
Biopharm. 1979, 7, 233. (b) Hou, F. F.; Zhang, X.; Zhang, G. H.; Xie,
D.; Chen, P. Y.; Zhang, W. R.; Jiang, J. P.; Liang, M.; Wang, G. B.; Liu,
Z. R.; Geng, R. W. N. Engl. J. Med. 2006, 354, 131. (c) Narasimhan,
M.; Bruce, T. O.; Masand, P. Neuropsychiatr. Dis. Treat. 2007, 3, 579.
(3) Carraher, C. E., Jr J. Chem. Educ. 1978, 55, 51.
(4) Casadei, M. A.; Galli, C.; Mandolini, L. J. Am. Chem. Soc. 1984,
106, 1051.
(5) For a review on ring expansion for the formation of seven-
membered rings, see: Kantorowski, E. J.; Kurth, M. J. Tetrahedron
2000, 56, 4317.
(6) For a review on transition metal chemistry of cyclopropenes and
cyclopropanes, see: Rubin, M.; Rubina, M.; Gevorgyan, V. Chem. Rev.
2007, 107, 3117.
(7) For selected reviews on the cleavage of carbon−carbon bonds by
transition metals, see: (a) Murakami, M.; Ito, Y. Top. Organomet.
Chem. 1999, 3, 97. (b) Murakami, M.; Matsuda, T. Chem. Commun.
2011, 47, 1100. (c) Seiser, T.; Cramer, N. Org. Biomol. Chem. 2009, 7,
2835. (d) Marek, I.; Masarwa, A.; Delaye, P.-O.; Leibeling, M. Angew.
Chem., Int. Ed. 2014, 53, 2. (e) Laetitia, S.; Cramer, N. Chem. Rev.
2015, 115, 9410.
(26) Charles, M. D.; Schultz, P.; Buchwald, S. L. Org. Lett. 2005, 7,
3965.
(27) Li, Z.; Fu, Y.; Guo, Q.; Liu, L. Organometallics 2008, 27, 4043.
(28) Previous computational and NMR studies identified mono-
ligated LPd(0) as the active catalyst for oxidative addition reactions
with biarylphoshine ligands SPhos and X-Phos. A similar η1 arene-Pd
interaction was calculated for these structures, see: Barder, T. E.;
Biscoe, M. R.; Buchwald, S. L. Organometallics 2007, 26, 2183.
(29) All reported energies are relative to the sum of the energy of the
separated educt 20 and the catalyst 27.
(30) Peng, C.; Ayala, P. Y.; Schlegel, H. B.; Frisch, M. J. J. Comput.
Chem. 1996, 17, 49.
(31) Frisch, M. J.; et al. Gaussian 03, revision D.01; Gaussian, Inc.:
Wallingford, CT, 2009.
(32) The calculated energy barriers that we have obtained for the C−
N bond and the C−C bond oxidative addition to the Pd(0) catalyst
are in agreement with the values obtained in a previous report
describing a related Pd(0)-catalyzed process that involved an initial
formation of a palladacyclobutane intermediate that subsequently
rearranged to a π-allyl complex, see ref 9.
(33) The rearrangement was attempted at room temperature to form
caprolactam 6, and no observable amount of product was observed
under catalytic conditions. However, under stoichiometric conditions
with 1 equiv of the palladium complex at room temperature, 6 was
(8) For a review on β-carbon elimination reactions, see: Aïssa, C.
Synthesis 2011, 2011, 3389.
(9) The Pd(0)-catalyzed carbon−carbon bond cleavage of electroni-
cally-activated 1,1-difluoro-2-arylcyclopropanes has been reported to
lead to palladacyclobutane intermediates in the synthesis of 2-
fluorinated allylic products; see: Xu, J.; Ahmed, E.-A.; Xiao, B.; Lu, Q.-
Q.; Wang, Y.-L.; Yu, C.-G.; Fu, Y. Angew. Chem., Int. Ed. 2015, 54,
8231.
(10) Palladium(0) has also been reported to effect carbon−carbon
bond cleavage in highly activated cyclopropanes, such as 1,1,2,2-
tetracyanocyclopropane, see: Lenarda, M.; Graziani, M.; Belluco, U. J.
Organomet. Chem. 1972, 46, C29.
(11) The platinum(0)-catalyzed rearrangement of highly strained
oxaspirohexanes led to 3-methylenetetrahydrofurans via a similar
proposed mechanistic pathway. Treatment of our reported substrates
to platinum catalysis led to no desired product formation; see: Malapit,
C. A.; Chitale, S. M.; Thakur, M. S.; Taboada, R.; Howell, A. R. J. Org.
Chem. 2015, 80, 5196.
1
formed in 7% conversion as determined by H NMR analysis of the
crude reaction mixture. A catalyst turnover number of 0.07 at room
temperature is in line with our calculated rate-determining energy
barrier.
(34) Anslyn, E. V.; Dougherty, D. A. Postulates and principles related
to kinetic analysis. Modern Physical Organic Chemistry, 1st ed.;
University Science Books: Sausalito, CA, 2006; pp 374−379.
(35) All azaspirocyclic lactam starting materials can be obtained in
one step from the corresponding γ-cyanoester using a chemoselective
titanium-mediated cyclopropanation reaction, see: Bertus, P.;
Szymoniak, J. Synlett 2003, 265.
(12) Surry, D. S.; Buchwald, S. L. Chem. Sci. 2011, 2, 27.
(13) For discussions on anionic palladium(0) complexes, see:
(a) Amatore, C.; Carre, E.; Jutand, A.; M’Barki, M. A.; Meyer, G.
́
Organometallics 1995, 14, 5605. (b) Amatore, C.; Jutand, A. Acc. Chem.
Res. 2000, 33, 314.
(14) Other carbonate, acetate, and phosphate bases also promoted
the reaction and gave rise to conversions ranging between 38 and 97%.
The use of NaOt-Bu with t-AmOH resulted in no product formation.
(15) For a review on catalyst deactivation, see: Crabtree, R. H. Chem.
Rev. 2015, 115, 127.
(16) Siegbahn, P. E.; Blomberg, M. R. A. J. Am. Chem. Soc. 1992, 114,
10548.
(17) For a review on trimethylenemethane cycloaddition chemistry,
see: Yamago, S.; Nakamura, E. Org. React. 2002, 61, 1.
(18) Trost, B. M. Angew. Chem., Int. Ed. Engl. 1986, 25, 1.
(19) Binger, P.; Schuchardt, U. Chem. Ber. 1980, 113, 3334.
(20) Suzuki, T.; Fujimoto, H. Inorg. Chem. 2000, 39, 1113.
H
DOI: 10.1021/acs.joc.5b01846
J. Org. Chem. XXXX, XXX, XXX−XXX