Directed ruthenium-catalyzed C(sp3)-H α-alkylation of cyclic amines using dioxolane-protected alkenones

Article abstract of DOI:10.1002/adsc.201400117

A catalytic system for ruthenium-catalyzed C(sp³)-H α-alkylation of piperidines with dioxolane-protected alkenones is reported. Dioxolane protection of the ketone proved crucial to obtain alkylation products. A diverse set of highly substituted piperidines was readily prepared in moderate to good yields via this methodology from easily accessible starting materials (C-2, C-3 and C-4 substituted piperidines). When the methodology is applied to C-3 substituted piperidines, featuring two α positions, only monoalkylated products (2,5-disubstituted) are observed. Even bicyclic amines, which feature a fused piperidine moiety, can be used. The successful directing group as well as protecting group (ketal) removal is also demonstrated. The methodology thus allows one to further derivatize and access hitherto unknown functionalized cyclic amine derivatives and will be useful in molecular library synthesis.

Full text of DOI:10.1002/adsc.201400117

FULL PAPERS
DOI: 10.1002/adsc.201400117
3
À
Directed Ruthenium-Catalyzed C(sp ) H a-Alkylation of Cyclic
N
Amines Using Dioxolane-Protected Alkenones
Artem A. Kulago,a Ben F. Van Steijvoort,a Emily A. Mitchell,a Lieven Meerpoel,b
and Bert U. W. Maesa,*
a
Organic Synthesis, Department of Chemistry, University of Antwerp, Groenenborgerlaan 171, B-2020 Antwerp, Belgium
Fax : (+32)-32-653-233; phone: (+32)-32-653-205; e-mail: bert.maes@uantwerpen.be
Janssen Research & Development, a division of Janssen Pharmaceutica N.V., Turnhoutseweg 30, B-2340, Beerse, Belgium
b
Received: January 30, 2014; Published online: && &&, 0000
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201400117.
Abstract: A catalytic system for ruthenium-catalyzed observed. Even bicyclic amines, which feature
3
À
C
G
protected alkenones is reported. Dioxolane protec- ful directing group as well as protecting group
tion of the ketone proved crucial to obtain alkylation (ketal) removal is also demonstrated. The methodol-
products. A diverse set of highly substituted piperi- ogy thus allows one to further derivatize and access
dines was readily prepared in moderate to good hitherto unknown functionalized cyclic amine deriva-
yields via this methodology from easily accessible tives and will be useful in molecular library synthe-
starting materials (C-2, C-3 and C-4 substituted pi- sis.
peridines). When the methodology is applied to C-3
substituted piperidines, featuring two a positions, Keywords: alkenes; C
only monoalkylated products (2,5-disubstituted) are groups; piperidines; ruthenium catalysis
3
À
G
Introduction
dines. This is not surprising when one considers the
inherently lower reactivity of six-membered rings
C-2 substituted saturated cyclic amines constitute an versus their five-membered counterparts.[7] In the past
important and valuable skeletal core in a number of years, our laboratory has developed a direct Ru-cata-
[8]
pharmaceuticals and natural products.[1] A variety of lyzed C H activation protocol for the C-2 aryla-
À
methodologies exists to synthesize these products di- tion[3g,i] and alkylation[3h] of substituted piperidines.
rectly from the corresponding cyclic amines.[2a,b] The The protocols have proved to be general, as justified
vast majority of these methods involve the use of stoi- by their applicability to other cyclic amine substrates.
chiometric quantities of an activating reagent; In 2012, we reported a C-2 alkylation strategy using
a strong base (a-anion intermediate) or an oxidant 1-hexene and 1-undecene as model reagents
(a-cation or a-radical intermediate).[2b] To address (Scheme 1).[3h]
this disadvantage, a transition metal catalyst can alter-
natively be used to activate the a-position of cyclic
amines.[3,4] However, the application of transition
metal catalysis in this area still remains limited.[3,4]
A
key feature of this approach is the use of an N-bound
directing group. A pyridin-2-yl group has been identi-
fied as a stable and efficient directing group for the
direct catalytic C-2 functionalization of cyclic am-
ACHTUNGTRENNUNG
ines.[3b–d,g–j] It is simple to install (SNAr or Pd cataly-
sis)[5] and our laboratory has recently demonstrated
that it can be easily removed after the a-functionali-
zation process.[6] Within the saturated cyclic amines,
pyrrolidines are far better explored for direct transi-
Scheme 1. Direct a-alkylation of piperidines using unfunc-
tion metal-catalyzed functionalization than piperi- tionalized alkenes.[3h]
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
ÞÞ
These are not the final page numbers!

FULL PAPERS
Artem A. Kulago et al.
Several challenges are associated with the use of 1-
alkenes for this direct alkylation process. Under the
applied reaction conditions, terminal alkenes can
polymerize and/or isomerize to non-terminal alkenes.
These internal alkenes are more stable than the 1-alk-
Figure 1. Design of the alkene.
ACHTUNGTRENNUNGenes and need to isomerize back to the terminal posi-
tion before direct functionalization can occur. In addi-
tion, direct alkylation is accompanied by a competitive
reduction reaction of the alkene.[3h] It was found that access the desired keto functionality. Ketal-protected
the rate of direct functionalization versus reduction methyl vinyl ketone, (2-methyl-2-vinyl-1,3-dioxolane,
could be increased by the use of a sterically hindered 2), under otherwise similar reaction conditions yield-
alcohol (2,4-dimethyl-3-pentanol) and the addition of ed 21% conversion of 3a (Table 1, entry 2). Changing
catalytic acid [trans-1,2-Cy
ACHTUNGTRENNUNG
were crucial to deliver a synthetically applicable [trans-Cy
N
ACHTUNGTRENNUNG
direct alkylation process for the less reactive piperi- (TFBA, “B”), resulted in a slight increase in conver-
dine system. Up to now, only unfunctionalized al- sion (entry 3). These data indicate that alkene 2 pos-
kenes have been applied using this protocol, so we sesses a much lower reactivity in comparison to the
wondered if the process would be compatible with previously reported 1-hexene.[10] Notably, as observed
oxygen functionalization in the alkene. If applicable, with our C-2 hexylation protocol, the presence of cat-
our expanded methodology would allow one to access alytic acid is important, as it suppresses the reduction
piperidines with valuable functionality in the alkyl of alkene (entries 1–3). Subsequently, the effect of the
side-chain. In this manuscript, we report our findings quantities of both alkene and alcohol were investigat-
on the introduction of 3-oxoalkyl moieties a to the ni- ed. For these experiments, dodecane was added as
trogen in piperidines and related cyclic amines using a co-solvent in order to compensate for differences in
Ru catalysis.
the reaction volume upon varying the loadings of
these reagents. The increase in the concentration of
2,4-dimethyl-3-pentanol from 10 equiv. to 15 equiv.,
and from 10 equiv. to 40 equiv., showed a pronounced
increase in the conversion of 3a from 29% (entry 4)
Results and Discussion
The reaction conditions disclosed previously for the to 42% and 54%, respectively (entries 5 and 6). Dou-
direct hexylation of piperidines with 1-hexene bling the concentration of alkene 2 from 10 equiv. to
(Scheme 1) were chosen as a starting point for alkyla- 20 equiv. further increased the conversion to 69% (en-
tion using piperidine 3a as a model substrate. The use tries 6 and 9). Additionally, it was noted that a high
of methyl vinyl ketone (1) as an alkylating reagent concentration of 2,4-dimethyl-3-pentanol is crucial to
was unfortunately not successful under these condi- obtain higher conversions of 3a when using this
tions and substrate 3a was quantitatively recovered alkene loading (entries 7–9). Increasing the TFBA/
(Scheme 2). Interestingly, GC-MS analysis also indi- Ru3(CO)12 loading to 8:8 mol% further improved the
cated that alkene 1 was completely consumed, result- conversion of 3a to 83% (entries 10 and 11). The
ing in 2-butanone (27%) and polymerized products. effect of the ratio of the loading of the TFBA with re-
We then envisaged a new strategy for successful alky- spect to Ru3(CO)12 was also investigated (entries 11–
lation based on ketone protection as a ketal (dioxo- 14). The smallest quantity of unreacted starting mate-
lane) (Figure 1). Ultimately, cleavage of this dioxo- rial 3a (8%) was observed when using 7 mol% TFBA
lane moiety after functionalization also allows one to and 8 mol% Ru3CO12 (entry 13). In this instance,
however, GC-MS data indicated the presence of side
products, giving a mass balance (sum of 3a, 4a and
5a) of only 81%. Shortening of the reaction time from
24 to 17 h allowed for the suppression of side product
formation, as is reflected in the improved mass bal-
ance, as well as in the isolated yields of the reaction
products (entry 15). At this stage, the re-screening of
trans-CyACTHNUGTRENUNG(COOH)2 as acid additive reconfirmed that
TFBA is superior as acid co-catalyst (90% vs. 75%
conversion of 3a, entries 15 and 16). When no acid
was used, after 17 h only 40% of 3a was converted to
products and only reduced 2 was observed (entry 17).
Even after 8 h reaction time, 76% of the alkene was
already reduced (entry 18). These experiments clarify
Scheme 2. Attempted direct a-alkylation of 1-pyridin-2-ylpi-
peridine (3a) with methyl vinyl ketone (1).[9]
2
asc.wiley-vch.de
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

3
À
FULL PAPERS
Directed Ruthenium-Catalyzed CACTHNUTRGNE(UNG sp ) H a-Alkylation of Cyclic Amines
Table 1. Optimization of the direct Ru-catalyzed a-alkylation of 1-pyridin-2-ylpiperidine (3a) using 2-methyl-2-vinyl-1,3-diox-
olane (2).[a]
Entry Acid[b]/[Ru]
[mol%]
Initial loadings
Time
[h]
GC yields [%][c]
Total
Yield
ACHTUNGTRENNUNG
2 [X equiv.] ROH [Y equiv.]
3a
4a
5a[d]
remaining 2[e] reduced 2[e] [%][f]
1
2
3
– (0/4)
A (4/4)
B (4/4)
B (4/4)
B (4/4)
B (4/4)
B (4/4)
B (4/4)
B (4/4)
B (6/6)
B (8/8)
B (9/8)
B (7/8)
B (6/8)
10
10
10
10
10
10
20
20
20
20
20
20
20
20
20
20
20
20
5
5
5
5
15
40
5
15
40
40
40
40
40
40
40
40
40
40
24
24
24
24
24
24
24
24
24
24
24
24
24
24
17
17
17
8
85
79
72
71
58
46
63
54
31
25
12
19
23
23
31
35
28
29
39
38
2
2
5
6
12
19
8
17
31
38
43
68
70
65
22
0
73
66
25
14
55
29
28
30
76
95
25
31
73
83
88
86
93
92
90
90
97
76
14
21
28
29
43
54
36
46
70
76
82
78
73
78
90
73
38
25
4[g]
5[g]
6[g]
7[g]
8[g]
9
10
11[h]
12
13[i]
14
17 (11) 38 (30) 44 (26) 10
20
8
20
35
30
38
43
43
40
11
0
5
2
5
15[h,j] B (7/8)
10 (4)
44 (39) 46 (35)
16[h]
17[h,j] – (0/8)
A (7/8)
25 (15) 40 (35) 33 (25)
60 (52) 31 (24) 7 (3)
73
0
22
18
– (0/8)
21
4
[a]
[b]
[c]
[d]
[e]
[f]
All reactions were performed on a 0.5-mmol scale of 3a.
A=trans-cyclohexane-1,2-dicarboxylic acid, B=3,4,5-trifluorobenzoic acid, [Ru]=Ru3(CO)12.
Crude reaction mixture using 1,3,5-trimethoxybenzene as an internal standard, isolated yields are given in parenthesis.
Sum of trans- and cis-5a isomers.
Values are given with respect to the initial loading of 2 (X).
Sum of the GC yields of 4a and 5a.
[g]
[h]
[i]
Dodecane was used as a co-solvent, in order to keep the reaction volume constant.
Reactions were performed on a 0.74-mmol scale of 3a to derive isolated yields.
The mass balance (81%) and GC chromatogram indicated decomposition.
GC yields represent the average of four experiments.
[j]
the role of the acid, which controls the rate of the thereby avoiding steric hindrance with the directing
functionalization of the piperidine substrate 3a versus group.[11] In the case of C-4 substituted piperidines
the reduction of the alkene 2, an effect which was (4g–i) a mixture of C-2 monoalkylated and C-2,6 bis-
also observed in our alkylation protocol with unfunc- alkylated piperidines was obtained, with the former
tionalized alkenes (Scheme 1).
being the major product observed. Both carbon-based
We subsequently evaluated the substrate scope with (ester 4h, phenyl 4i) as well as heteroatom-based
respect to the substitution pattern of the piperidine in (ketal, 4g) functional groups were tolerated at C-4.
the coupling with alkene 2. Gratifyingly, the synthetic The compatibility of the protocol with an ester in C-4
protocol could be successfully applied to a number of of the piperidine ring is particularly interesting, as its
piperidines featuring substituents in the 2-, 3-, and 4- presence makes the protons at C-4 quite acidic, creat-
positions, yielding compounds 4b–j in moderate to ing a site for competitive functionalization.[12] Re-
good yields (Table 2). In the C-2 position, both small markably, the protocol was found to be fully regiose-
(methyl, 4b) as well as large (phenyl, 4c) substituents lective for C-3 substituted piperidines (3d–f). The pi-
are well tolerated. This can be rationalized on the peridine derivatives possessing a C-3 trifluoromethyl
basis of pseudoallylic strain which places the C-2 sub- (3d), phenyl (3e) or methoxymethyl (3f) substituent
stituent in an axial position on the piperidine ring, all underwent smooth monoalkylation at the sterically
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3
ÞÞ
These are not the final page numbers!

FULL PAPERS
Artem A. Kulago et al.
Table 2. Reaction scope for the Ru-catalyzed direct a-alkylation of substituted piperidines 3.[a]
Entry
Product
Yield [%]
R1
4[b]
5
1
2
4b (Me)
4c (Ph)
84 (22/62)
82 (20/62)
–
–
3
4
5
4d (CF3)
4e (Ph)
4f (CH2OMe)
63 (40/23)
75 (55/20)
66 (39/27)
–
–
–
6
7
8
4g (O
4h (COOMe)
4i (Ph)
(CH2)2O)
38
33[c]
39 (26/13)
34 (23/11)
17[d]
18[e]
9
4j
74
–
[a]
Scale: 3b–f (0.5–0.75 mmol, 1 equiv.).
dr (cis:trans).
For 5g, cis:trans=6:27.
For 5h, dr 4:5:8.
[b]
[c]
[d]
[e]
For 5i dr 9:4:5.
less hindered a-position, with the resulting 2,5-disub- addition, ethyl 2,2-dimethyl-3-butenoate (7), featuring
stituted piperidines (4d–f) being isolated in 63–75% a ꢁlockedꢂ ester, also gave a high yield (87%) of the
yield. Regioselective monofunctionalization of C-3 desired product 4l (Table 3).
substituted piperidines was also observed in our previ-
Bicyclic amines, which comprise a fused piperidine
ously developed direct arylation protocol and there- moiety, are structural entities of significant interest.
fore seems to be substrate controlled (sterics) and in- In medicinal chemistry these scaffolds are used to
dependent of the reagents used.[3i] For the 2,5-disub- obtain constrained analogues of interesting pharma-
stituted piperidines (4d–f) resulting from this regiose- ceutical compounds.[13] Some have even been
lective monoalkylation, the cis-diastereoisomer was launched as APIs, as exemplified by the smoke cessa-
observed to be the major reaction product, in contrast tion agent Varenicline from Pfizer.[14] The locking of
to what was observed for the 2,6-disubstituted piperi- ring conformations and the steric hindrance associat-
dines (4b, c), where the trans-isomer was predomi- ed with the bridge make bicyclic piperidines chemical-
nant. In addition, the effect of benzoannulation at the ly challenging substrates for our direct functionaliza-
piperidine ring was also investigated, revealing that tion process. Two examples were chosen to challenge
C-2 alkylated 1,2,3,4-tetrahydroquinoline (4j) could our protocol: N-pyridin-2-yl-protected azabicyclo-
be obtained in very good yield (74%).
ACHTUNGTREN[NGNU 2.2.2]octane (isoquinuclidine) (8), which provides
In continuation of our efforts to explore the reac- a 2,5-disubstitution pattern and locks the piperidine in
tion scope, we studied variations in the alkene compo- a boat-like conformation, and N-pyridin-2-yl-protect-
nent on piperidine 3b under the conditions optimized ed 3-azabicycloACTHNUTRGENUG[N 3.2.1]octane (10) which features a 3,5-
for alkene 2. We discovered that ketal-protected disubstitution pattern and locks the piperidine in
phenyl vinyl ketone (6) was equally well tolerated in a chair conformation (Scheme 3). Starting from iso-
the protocol, yielding 4k in 87% yield (Table 3). In quinuclidine 8, alkylated 9 was derived in 47% yield.
4
asc.wiley-vch.de
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

3
À
FULL PAPERS
Directed Ruthenium-Catalyzed CACHTNUGTRENNUNG
Table 3. Alkene reaction scope for the Ru-catalyzed direct
a-alkylation of 3b.[a]
This is remarkable, since the formation of 2,6-bis-al-
kylated products from C-3 substituted piperidines
(3d–f) has never been observed before. Direct func-
tionalization of bicyclic amine 10 proved also success-
ful. Product 11 was obtained in 30% overall yield and
the exo- and endo-diastereoisomers of 11 were sepa-
rated by preparative HPLC.[15] To our delight, the use
of both sterically congested bicyclic piperidines gave
moderate yields under the developed reaction condi-
tions without any further additional optimization. Not
surprisingly, a significant amount of substrate was ob-
served in the GC chromatogram of the reaction mix-
ture of these challenging transformations. 2-Substitut-
Entry
1
Product
Yield [%]
(cis/trans)
R2
ed 3-azabicycloACTHNUTRGNEUNG[3.2.1]octanes like 11 are especially in-
87 (22/65)
87 (22/65)
teresting from an application point of view as they
have, to the best of our knowledge, only been poorly
explored.
2
[a]
As a final component to our study of the reaction
scope, the suitability of the direct Ru-catalyzed alky-
lation protocol to functionalize smaller (pyrrolidine)
and larger (azepane) saturated cyclic amines was also
examined (Scheme 4). The application of the opti-
mized reaction conditions on 1-pyridin-2-ylpyrrolidine
(12) and 1-pyridin-2-ylazepane (15) gave the antici-
pated alkylated products in synthetically attractive
yields. The high reactivity of 12 was manifested in
a high overall yield of alkylated reaction products 13
and 14 (84%) and a low mono to bis reaction product
ratio (1.2/1). This lends support to the proposed lower
reactivity of piperidines versus pyrrolidines observed
in other direct functionalization processes.[7] 1-Pyri-
din-2-ylazepane (15), gave a moderate overall yield
(49%) and revealed the presence of a significant
amount of starting material (28% GC yield) after
17 h, which suggests a lower reactivity of the azepane
versus the piperidine ring system.
Scale: 3b (0.57 mmol, 1 equiv.).
The assignment of the relative orientation (cis or
trans) of the substituents in all the piperidine deriva-
tives was done based on distinct differences in their
Scheme 3. Ru-catalyzed a-alkylation of bicyclic piperidines
8 and 10.
Scheme 4. Ru-catalyzed a-alkylation of 1-pyridin-2-ylpyrrolidine (12) and 1-pyridin-2-ylazepane (15).
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
5
ÞÞ
These are not the final page numbers!

FULL PAPERS
Artem A. Kulago et al.
formed on representative compounds. Crystals were
obtained for trans-4c, trans-5g, cis-4e and cis-14 by
slow evaporation of saturated solutions in heptane.[17]
From X-ray structures of these compounds (Figure 2
and Figure 3), it appeared that the N-pyridinyl-2-yl
moiety always adopts the thermodynamically favora-
ble pseudoequatorial orientation. The central piperi-
dine ring is significantly distorted to a twisted-boat
conformation in trans-4c and trans-5g and features
a pseudoaxial orientation of the substituents at the C-
2 and C-6 positions in order to minimize steric hin-
drance with the pyridin-2-yl group at the neighboring
nitrogen atom (Figure 2, a and b). In cis-4e (Figure 3,
a) the central piperidine ring adopts a chair-like con-
formation with an equatorial orientation for the sub-
stituent at the C-5 position and an axial one for the
substituent at C-2. For cis-14 (Figure 3, b), the central
pyrrolidine was noted to adopt an envelope-like con-
formation with pseudoaxial substituents at the corre-
sponding C-2 and C-5 centers.
With the new synthetic protocol for direct alkyla-
tion in hand, we became interested in the potential
for further synthetic transformations of the obtained
a-functionalized products. The C-2 alkylated piperi-
dine 4a was selected as a model compound for this
purpose. Application of our recently developed mild
one-pot sequential protocol for pyridine DG remov-
al[6] delivered piperidine 18 in an overall yield of 72%
(Scheme 5). Hydrolysis of the dioxolane protective
group in 4a proceeded smoothly under catalytic acidic
conditions delivering the 3-oxobutylpiperidine 19 in
very high yield (93%), which we failed to synthesize
directly from 3a and methyl vinyl ketone (1)
(Scheme 1).
Conclusions
3
À
An Ru-catalyzed protocol for the direct C
N
alkylation of piperidines has been developed. This
method permits one to efficiently introduce oxygen
functionality (3-oxoalkyl) into the a-alkyl chain. Al-
kenones proved not to be useful reagents under the
3
À
direct C
(sp ) H functionalization conditions previous-
Figure 2. X-ray structure of a) trans-4c, b) trans-5g Displace-
ment ellipsoids are drawn at the 50% probability level. Hy-
drogen atoms, distorted solvent and (minor) disorder com-
ponents are omitted for clarity.
ly reported by our group. In order to avoid the inher-
ent disadvantages of alkenes (i.e., reduction, polymer-
ization, isomerization) a rational design of the alkene
was required. By “masking” the keto functionality via
dioxolane protection and further optimization of the
1H NMR spectra.[16] The multiplets representing the reaction parameters an efficient protocol was
axial protons on the piperidine moiety were found achieved. Alternatively, locking of the conjugation of
shifted significantly upfield in comparison to the pro- the double bond with a carbonyl as exemplified by
tons adopting equatorial positions. Structural assign- ethyl 2,2-dimethyl-3-butenoate is also possible, thus
ment was additionally supported by the analysis of circumventing the undesired isomerization to the
the observed coupling constants. In order to unequiv- more stable a,b-unsaturated carbonyl and further re-
ocally confirm the relative orientation of the substitu- duction anticipated with ethyl-3-butenoate as alkylat-
ents, single crystal X-ray diffraction analysis was per- ing reagent. The versatility of the new protocol was
6
asc.wiley-vch.de
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

3
À
FULL PAPERS
Directed Ruthenium-Catalyzed CACTHNUTRGNE(UNG sp ) H a-Alkylation of Cyclic Amines
effectively demonstrated on a diverse set of piperi-
dines, equipped with a pyridin-2-yl directing group.
Piperidines with substituents in the C-2, C-4 positions
were well-tolerated and delivered the anticipated
monoalkylated and bisalkylated products in moderate
to good yields. Regioselective a-alkylation of C-3 sub-
stituted piperidines was observed as a result of steric
hindrance at the C-2 position. The assignment of the
relative orientation of the substituents for the derived
functionalized piperidines was done based on
1H NMR spectroscopy and X-ray analysis. Challeng-
ing bicyclic amines, isoquinuclidine and 3-azabicyclo-
AHCTUNGTREG[NNNU 3.2.1]octane, featuring a bis-substituted piperidine
entity were also effective substrates. In addition, the
a-alkylation methodology was successfully demon-
strated on other ring sizes (pyrrolidine and azepane
systems). The utility of the a-alkylation methodology
was demonstrated by the possibility to efficiently
remove the pyridin-2-yl directing group as well as the
ketal protective group in the reaction products. After
all, post-transformation of the secondary amine and
ketone functionalities will allow one to access a variety
of substituted cyclic amines hitherto unknown and
highly warranted for drug discovery purposes.
Experimental Section
General Procedure for Directed Ru-Catalyzed
3
À
C
G
Dioxolane Protected Alkenones
The reactions were performed on a 0.5–0.75-mmol scale of
cyclic amine in 10 mL microwave vials. Each vial was
equipped with a magnetic stirring bar and was charged with
the appropriate N-pyridin-2-yl-substituted cyclic amine
(1 equiv.), alkene (20 equiv.), a solution of 3,4,5-trifluoro-
benzoic acid (7.0 mol%) in 2,4-dimethyl-3-pentanol
(40 equiv.) and Ru3(CO)12 (8.0 mol%). The vials were
purged with argon and sealed by means of crimp caps. The
vials were placed in an oil bath preheated at 1408C. The cap
was secured with a top clamp. The reaction mixtures were
stirred for 17 h. After this time, the reaction vessels were
cooled to ambient temperature, opened and the contents of
all vials were combined. A commercially available rutheni-
um scavenger [Siliabond DMT, Silicycle, 0.2 g per 10 mg of
Ru3(CO)12] was added along with dichloromethane
(100 mL) and the resulting suspension was stirred at room
temperature for 16 h. Subsequently, the solids were removed
by filtration through a pad of Celite, which was additionally
rinsed with dichloromethane (2ꢃ50 mL). The combined fil-
trate was concentrated under reduced pressure and the resi-
due was subjected to column chromatography using an auto-
mated flash chromatography system.
Figure 3. X-ray structure of a) cis-4e, b) cis-14. Displace-
ment ellipsoids are drawn at the 50% probability level. Hy-
drogen atoms, distorted solvent and (minor) disorder com-
ponents are omitted for clarity.
Alkylation of 3a with Alkene 2
The reaction was performed in four separate vials according
to the general procedure described above. Each vial was
charged with 3a (120 mg, 0.74 mmol), a solution of 3,4,5-tri-
Scheme 5. Directing group removal and ketal deprotection
in compound 4a.
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
7
ÞÞ
These are not the final page numbers!

FULL PAPERS
Artem A. Kulago et al.
fluorobenzoic acid (9.1 mg, 0.05 mmol, 7 mol%) in 2,4-dime- References
thylpentan-3-ol (3440 mg, 29.6 mmol),
2
(1690 mg,
14.8 mmol) and Ru3(CO)12 (37.8 mg, 0.06 mmol, 8 mol%).
The crude products were purified by automated flash chro-
matography applying a heptane-ethyl acetate gradient (from
100% heptane to 50% heptane–50% ethyl acetate in
120 min, 40 mLminÀ1).
[1] a) Fundamentals of Heterocyclic Chemistry: Importance
in Nature and in the Synthesis of Pharmaceuticals,
(Eds.: L. D. Quin, J. Tyrell), Wiley, New Jersey, 2010;
b) Pharmaceutical Substances: Syntheses, Patents, Appli-
cations, 5th edn., (Eds.: A. Kleemann, J. Engel, B.
Kutscher, D. Reichert), Georg Thieme Verlag, Stutt-
gart, 2009.
[2] For recent reviews, see a) K. R. Campos, Chem. Soc.
Rev. 2007, 36, 1069–1084; b) E. A. Mitchell, A. Pe-
schiulli, N. Lefꢄvre, L. Meerpoel, B. U. W. Maes, Chem.
Eur. J. 2012, 18, 10092–10142.
[3] For cyclic amines, see a) Y. Ishii, N. Chatani, F. Kakiu-
chi, S. Murai, Tetrahedron Lett. 1997, 38, 7565–7568;
b) Y. Ishii, N. Chatani, F. Kakiuchi, S. Murai, Organo-
metallics 1997, 16, 3615–3622; c) N. Chatani, T. Asaumi,
T. Ikeda, S. Yorimitsu, Y. Ishii, F. Kakiuchi, S. Murai, J.
Am. Chem. Soc. 2000, 122, 12882–12883; d) N. Chatani,
T. Asaumi, S. Yorimitsu, T. Ikeda, F. Kakiuchi, S.
Murai, J. Am. Chem. Soc. 2001, 123, 10935–10941;
e) C. S. Yi, S. Y. Yun, Organometallics 2004, 23, 5392–
5395; f) S. J. Pastine, D. V. Gribkov, D. Sames, J. Am.
Chem. Soc. 2006, 128, 14220–14221; g) H. Prokopcovꢅ,
S. D. Bergman, K. Aelvoet, V. Smout, W. Herrebout, B.
Van der Veken, L. Meerpoel, B. U. W. Maes, Chem.
Eur. J. 2010, 16, 13063–13067; h) S. D. Bergman, T. E.
Storr, H. Prokopcovꢅ, K. Aelvoet, G. Diels, L. Meer-
poel, B. U. W. Maes, Chem. Eur. J. 2012, 18, 10393–
10398; i) A. Peschiulli, V. Smout, T. Storr, E. A. Mitch-
2-{2-[2-(2-Methyl-1,3-dioxolan-2-yl)ethyl]piperidin-1-yl}-
pyridine (4a): yield: 320 mg (1.16 mmol, 39%); colorless vis-
cous oil. 1H NMR (CDCl3): d=8.10 (ddd, J=4.9, 2.0,
0.8 Hz, 1H), 7.38 (ddd, J=8.9, 7.1, 2.0 Hz, 1H), 6.56 (d, J=
8.7 Hz, 1H), 6.45 (dd, J=5.0, 7.0 Hz, 1H), 4.44–4.42 (m,
1H), 4.22–4.11 (m, 1H), 3.93–3.85 (m, 4H), 2.95 (td, J=
13.2, 2.8 Hz, 1H), 1.86–1.46 (m, 10H), 1.29 (s, 3H);
13C NMR (CDCl3): d=159.0, 147.7, 137.4, 111.5, 110.0,
106.7, 64.6, (2C) 51.4, 39.3, 36.0, 27.8, 25.3, 23.8, 23.0, 19.2;
+
HR-MS (ESI): m/z=277.1909, calculated for C16H25N2O2
[M+H]+: 277.1916.
trans-2-{2,6-Bis[2-(2-methyl-1,3-dioxolan-2-yl)ethyl]piperi-
din-1-yl}pyridine (trans-5a): yield: 340 mg (0.87 mmol,
1
29%); colorless viscous oil. H NMR (CDCl3): d=8.15 (dd,
J=1.3, 5.0 Hz, 1H), 7.40 (ddd, J=8.9, 7.1, 2.0 Hz, 1H), 6.52
(dd, J=6.5, 5.0 Hz, 1H), 6.47 (d, J=8.6 Hz, 1H), 3.97–3.80
(m, 10H), 1.90–1.54 (m, 14H), 1.29 (s, 6H); 13C NMR
(CDCl3): d=158.7, 147.6, 136.6, 111.9, 110.0, 108.9, 64.5
(4C), 52.7, 36.2, 26.8, 24.5, 23.7, 15.0; HR-MS (ESI): m/z=
+
391.2602, calculated for C22H35N2O4 [M+H]+: 391.2597.
cis-2-{2,6-Bis[2-(2-methyl-1,3-dioxolan-2-yl)ethyl]piperi-
din-1-yl}pyridine (cis-5a): yield: 65 mg (0.17 mmol, 6%);
1
white solid; mp 125–1278C. H NMR (CDCl3): d=8.10 (d,
J=3.6 Hz, 1H), 7.37 (t, J=7.6 Hz, 1H), 6.44 (d, J=7.8 Hz,
2H), 4.31 (m, 2H), 3.90–3.87 (m, 8H), 1.79–1.46 (m, 14H),
1.28 (s, 6H); 13C NMR (CDCl3): d=158.4, 147.8, 137.2,
111.2, 110.0, 106.2, 64.5 (4C), 50.0, 37.0, 27.5, 23.8, 14.8;
ˇ
ell, Z. Eliꢅs, W. Herrebout, D. Berthelot, L. Meerpoel,
B. U. W. Maes, Chem. Eur. J. 2013, 19, 10378–10387;
j) M. C. Schwarz, N. Dastbaravardeh, K. Kirchner, M.
Schnꢆrch, M. D. Mihovilovic, Monatsh. Chem. 2013,
144, 539–552.
+
HR-MS (ESI): m/z=391.2602, calculated for C22H35N2O4
[M+H]+: 391.2597.
[4] For acyclic amines, see a) C.-H. Jen, D. C. Hwang, S.-J.
Na, Chem. Commun. 1998, 1405–1406; b) S. Pan, K.
Endo, T. Shibata, Org. Lett. 2011, 13, 4692–4695; c) S.
Pan, T. Matsuo, K. Endo, T. Shibata, Tetrahedron 2012,
68, 9009–9015; d) N. Dastbaravardeh, M. Schnꢆrch,
M. D. Mihovilovic, Org. Lett. 2012, 14, 1930–1933; e) N.
Dastbaravardeh, K. Kirchner, M. Schnꢆrch, M. D. Mi-
hovilovic, J. Org. Chem. 2013, 78, 658–672; f) N. Dast-
baravardeh, M. Schnꢆrch, M. D. Mihovilovic, Org. Lett.
2012, 14, 3792–3795; g) N. Y. Phani Kumar, R. Jeya-
chandran, L. Ackermann, J. Org. Chem. 2013, 78,
4145–4152.
[5] For Buchwald–Hartwig reactions involving halopyri-
dines, see, for example: a) T. H. M. Jonckers, B. U. W.
Maes, G. L. F. Lemiꢄre, R. Dommisse, Tetrahedron
2001, 57, 7027–7034; b) B. U. W. Maes, K. T. J. Loones,
T. H. M. Jonckers, G. L. F. Lemiꢄre, R. A. Dommisse,
A. Haemers, Synlett 2002, 1995–1998; c) B. U. W. Maes,
K. T. J. Loones, G. L. F. Lemiꢄre, R. A. Dommisse, Syn-
lett 2003, 1822–1824. For a review, see d) D. Maiti, B. P.
Fors, J. L. Henderson, Y. Nakamura, S. L. Buchwald,
Chem. Sci. 2011, 2, 57–68.
Acknowledgements
This work was financially supported by the University of
Antwerp (IOF, BOF), the Marie Curie action Grant PIIF-
GA-2012-331366, Janssen Research & Development and the
Hercules Foundation (AUHA 09/014, AUHA 09/008,
AUHA-002, AUHA-008). The authors thank Philippe Frank
for technical assistance, Heidi Seykens and Veerle Smout for
NMR assistance, and Norbert Hancke for HR-MS measure-
ments. The authors are grateful to Matthias Zeller (Youngs-
town State University, Ohio, USA) for the collection of the
X-ray data sets of trans-4c, cis-4e and cis-14. The X-ray dif-
fractometer was funded by NSF Grant 0087210, Ohio Board
of Regents Grant CAP-491, and by Youngstown State Uni-
versity. The authors also thank Kristof Van Hecke (Ghent
University, Belgium) for the collection of the X-ray data of
trans-5g and the Hercules Foundation (AUGE 11/029) for
funding the diffractometer and Christophe Vande Velde (Uni-
versity of Antwerp, Belgium) for the refinement of the struc-
tures.
[6] V. Smout, A. Peschiulli, S. Verbeeck, E. A. Mitchell, W.
Herrebout, P. Bultnick, C. M. L. Vande Velde, D. Ber-
thelot, L. Meerpoel, B. U. W. Maes, J. Org. Chem. 2013,
78, 9803–9814.
8
asc.wiley-vch.de
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

3
À
FULL PAPERS
Directed Ruthenium-Catalyzed CACTHNUTRGNE(UNG sp ) H a-Alkylation of Cyclic Amines
[7] H. M. L. Davies, R. E. J. Beckwith, Chem. Rev. 2003,
103, 2861–2903.
[8] For reviews on ruthenium-catalyzed C H activation
Wilson, R. F. Borne, Studies in Natural Products
Chemistry 2008, 34, 753–787.
[14] R. Niaura, C. Jones, P. Kirkpatrick, Nat. Rev. Drug Dis-
covery 2006, 5, 537–538.
[15] See the Supporting Information for details regarding
HPLC separation.
À
see a) P. B. Acrockiam, C. Bruneau, P. H. Dixneuf,
Chem. Rev. 2012, 112, 5879–5918; b) L. Ackermann,
Top. Curr. Chem. 2010, 292, 211–229.
[9] See the Supporting Information for details of experi-
[16] For 2,6-disubstituted piperidines, see: a) G. Zheng, L. P.
Dwoskin, A. G. Deaciuc, P. A. Crooks, J. Org. Chem.
2009, 74, 6072–6076. For 2,5-disubstituted piperidines,
see: b) S. Seel, T. Thaler, K. Takatsu, C. Zhang, H.
Zipse, B. R. Straub, P. Mayer, P. Knochel, J. Am. Chem.
Soc. 2011, 133, 4774–4777; c) P. A. Vadola, I. Carrera,
S. Sames, J. Org. Chem. 2012, 77, 6689–6702. For bicy-
clics, see: d) J. Quirante, F. Diaba, X. Vila, J. Bonjoch,
Magn. Reson. Chem. 2005, 43, 599–601.
[17] CCDC 983854, CCDC 983855, CCDC 983856 and
CCCDC 983857 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
ments with methyl vinyl ketone.
[10] In our previous work (see ref.[3h]) with 3a, we isolated
69% of hexylated products under the same reaction
conditions.
[11] P. Beak, A. Basu, D. J. Gallagher, Y. S. Park, S. Thayu-
manavan, Acc. Chem. Res. 1996, 29, 552–560.
[12] F. Bellina, R. Rossi, Chem. Rev. 2010, 110, 1082–1146.
[13] For examples, see: a) M. McLaughlin, S. Garcꢇa Rubio,
U. Tilstam, O. A. C. Antunes, T. Laird, G. D. Yadav, A.
Zlota, Org. Process Res. Dev. 2007, 11, 2–12; b) J. Car-
reras, A. Avenoza, J. H. Busto, J. M. Peregrina, J. Org.
Chem. 2007, 72, 3112–3115; c) M. O. Faruk Khan, M. S.
Levi, C. R. Clark, S. Y. Ablordeppey, S.-J. Law, N. H.
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
9
ÞÞ
These are not the final page numbers!

FULL PAPERS
3
À
10 Directed Ruthenium-Catalyzed CACTHNUTRGNEUNG(sp ) H a-Alkylation of
Cyclic Amines Using Dioxolane-Protected Alkenones
Adv. Synth. Catal. 2014, 356, 1 – 10
Artem A. Kulago, Ben F. Van Steijvoort, Emily A. Mitchell,
Lieven Meerpoel, Bert U. W. Maes*
10
asc.wiley-vch.de
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!