Diaminophosphine oxide ligand enabled asymmetric nickel-catalyzed hydrocarbamoylations of alkenes

Article abstract of DOI:10.1021/ja406730t

Chiral trivalent phosphorus species are the dominant class of ligands and the key controlling element in asymmetric homogeneous transition-metal catalysis. Here, novel chiral diaminophosphine oxide ligands are described. The arising catalyst system with nickel(0) and trimethylaluminum efficiently activates formamide C-H bonds under mild conditions providing pyrrolidones via intramolecular hydrocarbamoylation in a highly enantioselective manner with as little as 0.25% mol catalyst loading. Mechanistically, the secondary phosphine oxides behave as bridging ligands for the nickel center and the Lewis acidic organoaluminum center to give a heterobimetallic catalyst with superior reactivity.

Full text of DOI:10.1021/ja406730t

Communication
pubs.acs.org/JACS
Diaminophosphine Oxide Ligand Enabled Asymmetric Nickel-
Catalyzed Hydrocarbamoylations of Alkenes
Pavel A. Donets and Nicolai Cramer*
Laboratory of Asymmetric Catalysis and Synthesis, Institute of Chemical Sciences and Engineering, Ecole Polytechnique Fed
́ ́
erale de
Lausanne (EPFL), Lausanne, Switzerland
S
* Supporting Information
tions with formamides. The first one, involves a ligand-free
ruthenium−carbonyl complex and requires a CO atmospher-
e.9a−c Carreira demonstrated its utility for endo-cyclizations of
allyl formamides to access α-unsubstituted pyrrolidinones.9c
Alternatively, Nakao and Hiyama developed a nickel/Lewis acid-
catalyzed process. With the exception of two examples, this
method is largely limited to alkyne substrates9c and provides
simple alkene linear products.9d With either catalyst system,
asymmetric reactions are completely elusive. Furthermore, both
processes suffer from low catalyst efficiencies and/or harsh
reaction conditions. Driven by our interest in the asymmetric
functionalization of C−H bonds,11 we report a novel class of
chiral diaminophosphineoxides ligands enabling asymmetric
hydrocarbamoylations of homoallylic formamides by a hetero-
bimetallic activation mode (Scheme 1).
ABSTRACT: Chiral trivalent phosphorus species are the
dominant class of ligands and the key controlling element
in asymmetric homogeneous transition-metal catalysis.
Here, novel chiral diaminophosphine oxide ligands are
described. The arising catalyst system with nickel(0) and
trimethylaluminum efficiently activates formamide C−H
bonds under mild conditions providing pyrrolidones via
intramolecular hydrocarbamoylation in a highly enantio-
selective manner with as little as 0.25% mol catalyst
loading. Mechanistically, the secondary phosphine oxides
behave as bridging ligands for the nickel center and the
Lewis acidic organoaluminum center to give a hetero-
bimetallic catalyst with superior reactivity.
hiral ligands are the key controlling element in asymmetric
Scheme 1. Asymmetric Hydrocarbamoylations Enabled by a
Chiral Heterobimetallic Catalyst
C
catalysis with transition metals. Trivalent phosphorus
species are arguably the most dominant class of ligands for
homogeneous transition-metal catalysis.1 Whereas the vast
majority belongs to the class of triaryl- or alkylphosphines,
secondary phosphine oxides (SPOs) have only recently attracted
more attention as air-stable and robust preligands for transition-
metal catalysis.2 This stability largely stems from its unique
tautomerism between the pentavalent and the trivalent tautomer
(eq 1).3 By and large, the PV form dominates. In the presence of
Representing a complementary disconnection strategy for
synthetically valuable chiral pyrrolidones and pyrrolidines,12 we
recognized the importance of an efficient enantioselective
process. Working toward this goal, a large screen of a broad
array of chiral phosphine ligands revealed their global failure for
the hydrocarbamoylation of formamide 1a (Table 1, entries 1−
9). Solely methyl phospholane L9 (entry 9 and 10) seemed to
show promising results, although the reproducibility was strongly
dependent on the batch of L9. We could trace this inconsistency
to an impurity (5%) identified as the secondary phosphine oxide
SPO1. Alone SPO1 was poorly active, presumably due to
kinetically disfavored coordination to the nickel (entry 11). The
best reactivity was achieved in conjunction with an additional
phosphine helping to displace cyclooctadiene from the nickel
center, thus enabling the formation of the catalytically competent
species.13 With added PPh3, which alone does not catalyze the
reaction at all, a very robust catalytic system was found with a
1:1:1 ratio of [Ni(COD)2], SPO, and PPh3. SPO1 provided a
promising selectivity of 86:14 er (entry 12).14
bases or transition metals, the equilibrium can be shifted toward
to the trivalent phosphinous acid form.4 Whereas its phosphorus
atom coordinates well to late-transition metals (M1), the oxygen
atom represents a good ligand for early transition metals (M2).
The coordination of two different metals on the same secondary
phosphineoxide for defined early/late heterobimetallic catalysis
is largely unexplored.5−7
Despite the tremendous progress of carbon−hydrogen (C−
H) bond functionalizations over the past decade,8 intrinsic
reactivity boundaries of substrates often prevent the desired
transformation. For instance, formamide C−H bonds are largely
less reactive toward transition-metal insertions9 compared to
aldehydes commonly used for hydroacylations.10 Only two sets
of conditions have been developed to effect hydrocarbamoyla-
Received: July 2, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja406730t | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Table 1. Optimization of the Enantioselective Formamide C−
Scheme 2. Synthesis and X-Ray Crystal Structure of
Diaminophosphine Oxide SPO5
a
H Functionalization
b
c
entry
x (mol %)
SPO
PAr3
yield (%)
er
1
2
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
0.5
0.25
1
L1 (2 x)
L2 (2 x)
L3 (2 x)
L4 (2 x)
L5 (2 x)
L6 (2 x)
L7 (2 x)
L8 (2 x)
L9 (2 x)
SPO1
−
−
−
−
−
−
−
−
−
50
5
57:43
−
−
3
2
4
10
3
51:49
5
−
6
41
26
10
14
83
18
87
40
83
82
88
90 (90)
61
19
0
60:40
72:28
52:48
68:32
19:81
61:39
14:86
62:38
84:16
40:60
96.5:3.5
96.5:3.5
96.5:3.5
96:4
7
8
9
10
11
12
13
14
15
16
17
18
19
20
L9
SPO1
−
SPO1
PPh3
PPh3
PPh3
PPh3
PPh3
PPh3
PPh3
PPh3
PPh3
PPh3
SPO2
SPO3
SPO4
severe syn-pentane strain and thus fixing conformation and
efficiently shielding two quadrants.
SPO5
SPO5
To our delight, the diaminophosphine oxide SPO2-SPO5 is
well stable to AlMe3, and bulky SPO5 turned out to be a
particularly reactive and selective catalyst, yielding 2a in 88% and
96.5:3.5 er (entry 16).20 Most notably, the catalyst loading could
be reduced to 0.5 mol % without affecting the reaction outcome
at all (entry 17).21 Even at 0.25 mol % loading 2a is obtained,
though no full conversion and some double-bond isomerization
proceeds (entry 18). To explore the impact of the hetero-
bimetallic assembly of the catalyst for the reactivity, structurally
similar trivalent P-ligand L10 which is not able to form a covalent
adduct with the aluminum Lewis acid was tested.22 Compared to
SPO5, it exhibits even at double catalyst loading a drastically
reduced activity, albeit comparable enantioselectivity (entry 19).
Ligand L11, not able to coordinate at all to AlMe3, did not give
even trace amounts of product (entry 20). Additionally, BPh3, a
Lewis acid which is unable to form a bridging complex with the
SPO failed as well completely (entry 21).
SPO5
L10
5
L11
−
−
d
21
10
SPO5
0
a
Conditions: 0.10 mmol 1a, x mol % catalyst (L/PAr3/[Ni(COD)2] =
b
1:1:1), 40 mol % AlMe3, 0.25 m in toluene at 40 °C for 24 h. NMR
yields (isolated yields). Determined by HPLC with a chiral stationary
phase. with 40 mol % of BPh3 instead of AlMe3.
c
d
With the optimized reaction conditions in hand, the scope of
the activation/cyclization was evaluated (Table 2). As expected,
deuterated formamide 2b led to a complete deuterium transfer to
the terminal carbon atom of substrate 1b. The selectivity of the
reaction is largely independent of the substitution of the nitrogen
atom, allowing for alkyl, aryl, and benzyl groups as well as esters.
Bis-allylated formamides are competent substrates giving lactams
2h−2n in high yields and enantioselectivities. The observed
diastereoselectivity ranges from 4:1 to >20:1, evidencing a good
enantiotopic allyl group selection. Such bicyclic pyrrolidone
framework is a ubiquitous structural feature in important
pyrrolizidine and indolizidine alkaloids.12 Besides cyclization to
γ-lactams,23 the reaction concept is suitable for larger ring sizes
(2o), though competing endo-cyclization to 4 becomes a side
reaction.
The described reaction can be used as well to override intrinsic
substrate control of a preexisting stereogenic center (Scheme 3).
Exposing enantiopure formamide (S)-1p to the reaction
conditions with Cy2POH as prototypical achiral ligand, lactam
2p is formed in moderate yield and 4:1 diastereomeric ratio in
favor of the cis-isomer cis-2p. Whereas cis-2p was produced in
high diastereoselectivity (16:1) using (R)-SPO5 in the matched
However, the phospholane scaffold is not very user-friendly in
terms of simple and modular ligand modifications. In contrast,
heteroatom-substituted secondary phosphine oxides are easily
accessible and offer this synthetic advantage.15 Surprisingly, with
the exception of reports from Hamada,16 diaminophosphosphine
oxides have been largely neglected for asymmetric catalysis. Our
herein envisioned ligands class is readily prepared from the
corresponding widely used chiral alkylbenzylamine precursors 3
(Scheme 2).17 To the ethylenediamines 4, PCl3 is added, and
after cyclization proceeded, the remaining P−Cl bond is
hydrolyzed giving directly the air and moisture stable ligands
SPO2-SPO5. The salient structural features of this new ligand
class can be visualized by the X-ray crystal structure of SPO5.18 In
contrast to SPO1, the chirality is shifted outside of the cycle. The
large 1-naphthyl and t-butyl substituents arrange in a pseudo C2-
symmetric fashion. In analogy to the Kundig carbenes built from
̈
the same diimino intermediates,19 this arrangement minimizes
B
dx.doi.org/10.1021/ja406730t | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Table 2. Substrate Scope of the Enantioselective
Hydrocarbamoylation
Scheme 3. Matched/Mismatched Stereocontrol of Chiral
Substrate 1p
a
Taking the unique reactivity of secondary phosphine oxides for
this type of transformation into account, the following
mechanism is suggested: The secondary phosphine oxide L
tautomerizes to its phosphinous acid form L′ (Scheme 4). The
Scheme 4. Mechanistic Proposal of the Bimetallic Activation
with SPO Ligands
acidic hydroxyl group of L′ reacts with AlMe3 liberating methane
and gives adduct A,24 while retaining the Lewis acidic character of
the aluminum center. [Ni(COD)2] coordinates to the
phosphorus atom to deliver B. The Lewis acidic aluminum
center activates the carbonyl group of 1a giving C. Subsequently,
the nickel center could oxidatively insert into the now activated
C−H bond in a favorable intramolecular fashion forming the six-
membered bimetallic heterocycle D. Migratory insertion leads to
metallocycle E, which in turn reductively eliminates releasing
lactam 2a and regenerates the bimetallic catalyst B.
To support our hypothesis of the linked bimetallic species, we
independently prepared putative intermediate SPO5-Al by
stoichiometric deprotonation with butyllithium, followed by
salt metathesis with ClAlMe2. Though SPO5-Al THF adduct is
sensitive and only moderately stable, it is able to catalyze the
process with slightly superior results (96% yield and 97:3 er) at
0.25 mol % catalyst loading without any added additional AlMe3
and PPh3 (Scheme 5). These findings clearly indicate that neither
free AlMe3 nor PPh3 are active participants in the catalytic cycle.
Their requirements in the instant protocol point toward their
importance in entering the catalytic cycle.
Scheme 5. Catalytic Performance of the Preformed Lewis
Acid/Ligand SPO5-Al
a
0.10 mmol 1, 5 mol % [Ni(COD)2], 5 mol % SPO5, 5 mol % PPh3,
b
40.0 μmol AlMe3, 0.25 m in toluene at 40 °C for 12 h. Isolated yields.
c
d
Determined by HPLC with a chiral stationary phase.19 With 10 mol
e
f
% SPO1, 10 mol % [Ni(COD)2], 10 mol % PPh3. At 60 °C. 0.5
mmol scale with 1 mol % catalyst loading.
chirality scenario, trans-2p could be selectively obtained in a
moderate inverse dr of 3:1 with (S)-SPO5.
C
dx.doi.org/10.1021/ja406730t | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Chem. Rev. 2010, 110, 624. (h) Lyons, T. W.; Sanford, M. S. Chem. Rev.
2010, 110, 1147. (j) Jazzar, R.; Hitce, J.; Renaudat, A.; Sofack- Kreutzer,
J.; Baudoin, O. Chem.Eur. J. 2010, 16, 2654. (k) Chen, X.; Engle, K.
M.; Wang, D.-H.; Yu, J.-Q. Angew. Chem., Int. Ed. 2009, 48, 5094.
(9) (a) Kondo, T.; Okada, T.; Mitsudo, T.-A. Organometallics 1999, 18,
4123. (b) Ko, S.; Han, H.; Chang, S. Org. Lett. 2003, 5, 2687.
(c) Armanino, N.; Carreira, E. M. J. Am. Chem. Soc. 2013, 135, 6814.
(d) Nakao, Y.; Idei, H.; Kanyiva, K. S.; Hiyama, T. J. Am. Chem. Soc.
2009, 131, 5070. (e) Miyazaki, Y.; Yamada, Y.; Nakao, Y.; Hiyama, T.
Chem. Lett. 2012, 41, 298.
In conclusion, we report a new class of air and moisture stable
and readily accessible chiral bulky diaminophosphine oxide
preligands. We demonstrate their potential for asymmetric early/
late heterobimetallic catalysis with nickel(0)/Lewis acid
catalyzed C−H activations of formamides providing a comple-
mentary access to chiral γ-lactams. The diaminophosphineoxide
ligand is believed to simultaneously bond to the nickel center and
the aluminum Lewis acid resulting in superior reactivity and
performance.
(10) Reviews on hydroacylations: (a) Jun, C.-H.; Jo, E.-A.; Park, J.-W.
Eur. J. Org. Chem. 2007, 1869. (b) Willis, M. C. Chem. Rev. 2010, 110,
725.
ASSOCIATED CONTENT
* Supporting Information
Procedures and characterization data. This material is available
free of charge via the Internet at http://pubs.acs.org.
■
S
(11) Recent examples of in asymmetric C−Het, C−C, and C−H
functionalizations: (a) Seiser, T.; Roth, O. A.; Cramer, N. Angew. Chem.,
Int. Ed. 2009, 48, 6320. (b) Albicker, M. R.; Cramer, N. Angew. Chem.,
Int. Ed. 2009, 48, 9139. (c) Tran, D. N.; Cramer, N. Angew. Chem., Int.
Ed. 2011, 50, 11098. (d) Saget, T.; Lemouzy, S.; Cramer, N. Angew.
Chem., Int. Ed. 2012, 51, 2238. (e) Ye, B.; Cramer, N. Science 2012, 338,
504. (f) Saget, T.; Cramer, N. Angew. Chem., Int. Ed. 2012, 51, 12842.
(g) Ye, B.; Cramer, N. J. Am. Chem. Soc. 2013, 135, 636.
AUTHOR INFORMATION
Corresponding Author
nicolai.cramer@epfl.ch
Notes
■
(12) Daly, J. W.; Garraffo, H. M.; Spande, T. F. In The Alkaloids:
Chemical and Biological Perspectives; Pelletier, S. W., Ed.; Pergamon
Press: Amsterdam, 1990; pp 1−161.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(13) Other Ni(0) or Ni(II) sources were not successful.
This work is supported by the European Research Council under
the European Community’s Seventh Framework Program (FP7
2007−2013)/ERC grant agreement no. 257891 and the Swiss
National Science Foundation (no. 137666). We thank Dr. R.
Scopelliti for X-ray crystallographic analysis of SPO5.
(14) Combinations with several chiral phosphines were evaluated but
were found to be not superior to combinations with achiral phosphines.
(15) Representative examples of secondary heteroatom-substituted
phosphine oxides as ligands in transition-metal catalysis: (a) Ackermann,
L.; Born, R. Angew. Chem., Int. Ed. 2005, 44, 2444. (b) Ackermann, L.;
Born, R.; Spatz, J. H.; Meyer, D. Angew. Chem., Int. Ed. 2005, 44, 7216.
(c) Ackermann, L.; Kapdi, A. R.; Fenner, S.; Kornhaaß, C.; Schulzke, C.
Chem.Eur. J. 2011, 17, 2965. for an overview, see: (d) Ackermann, L.
Isr. J. Chem. 2010, 50, 652.
(16) (a) Nemoto, T.; Matsumoto, T.; Masuda, T.; Hitomi, T.; Hatano,
K.; Hamada, Y. J. Am. Chem. Soc. 2004, 126, 3690. (b) Nemoto, T.;
Masuda, T.; Matsumoto, T.; Hamada, Y. J. Org. Chem. 2005, 70, 7172.
(c) Nemoto, T.; Sakamoto, T.; Matsumoto, T.; Hamada, Y. Tetrahedron
Lett. 2006, 47, 8737. (d) Nemoto, T.; Hamada, Y. Chem. Rec. 2007, 7,
150. (e) Nemoto, T.; Hamada, Y. Tetrahedron 2011, 67, 667.
(17) Breuer, M.; Ditrich, K.; Habicher, T.; Hauer, B.; Kesseler, M.;
REFERENCES
■
(1) Phosphorous Ligand in Asymmetric Catalysis; Borner, A., Ed; Wiley-
VCH: Weinheim, 2008.
(2) Reviews: (a) Dubrovina, N. V.; Borner, A. Angew. Chem., Int. Ed.
2004, 43, 5883. (b) Ackermann, L.; Born, R.; Spatz, J. H.; Althammer,
A.; Gschrei, C. J. Pure Appl. Chem. 2006, 78, 209. (c) Ackermann, L.
Synthesis 2006, 1557. (d) Shaikh, T. M.; Weng, C.-M.; Hong, F.-E.
Coord. Chem. Rev. 2012, 256, 771.
(3) Christiansen, A.; Li, C.; Garland, M.; Selent, D.; Ludwig, R.;
Spannenberg, A.; Baumann, W.; Franke, R.; Borner, A. Eur. J. Org. Chem.
̈
̈
̈
Sturmer, R. Angew. Chem., Int. Ed. 2004, 43, 788.
̈
2010, 2733.
(18) Crystallographic data for SPO5: CCDC 925530 contains
supplementary crystallographic data for this paper and can be obtained
free of charge via The Cambridge Crystallographic Data Centre (www.
ccdc.cam.ac.uk/data_request/cif).
(4) (a) Achard, T.; Giordano, L.; Tenaglia, A.; Gimbert, Y.; Buono, G.
Organometallics 2010, 29, 3936. (b) Christiansen, A.; Selent, D.;
Spannenberg, A.; Baumann, W.; Franke, R.; Borner, A. Organometallics
̈
2010, 29, 3139.
(19) Synthesis and use of chiral carbene ligands derived from the same
(5) Bimetallic catalysis overviews: (a) Bullock, R. M.; Casey, C. P. Acc.
Chem. Res. 1987, 20, 167. (b) Stephan, D. W. Coord. Chem. Rev. 1989,
95, 41. (c) Wheatley, N.; Kalck, P. Chem. Rev. 1999, 99, 3379.
(d) Cooper, B. G.; Napoline, J. W.; Thomas, C. M. Catal. Rev. Sci. Eng.
2012, 54, 1.
(6) Nickel/magnesium bimetallic cooperation has been suggested with
secondary phosphine oxides for Kumada-Corriu cross-couplings of aryl
fluorides: (a) Ackermann, L.; Althammer, A. Chem. Unserer Zeit 2009,
43, 74. (b) Jin, Z.; Li, Y.-J.; Ma, Y.-Q.; Qiu, L.-L.; Fang, J.-X. Chem.Eur.
J. 2012, 18, 446.
(7) For suggested nickel/magnesium bimetallic cooperation in
Kumada-Corriu cross-couplings with a different ligand class: (a) Yoshi-
kai, N.; Mashima, H.; Nakamura, E. J. Am. Chem. Soc. 2005, 127, 17978.
(b) Yoshikai, N.; Matsuda, H.; Nakamura, E. J. Am. Chem. Soc. 2009,
131, 9590.
chiral amines: (a) Kundig, E. P.; Seidel, T. M.; Jia, Y.-X.; Bernardinelli, G.
̈
Angew. Chem., Int. Ed. 2007, 46, 8484. (b) Nakanishi, M.; Katayev, D.;
Besnard, C.; Kundig, E. P. Angew. Chem., Int. Ed. 2011, 50, 7438.
̈
(c) Banerjee, D.; Besnard, C.; Kundig, E. P. Organometallics 2012, 31,
709.
̈
(20) For determination of the absolute configuration, see Supporting
Information.
(21) Lower amounts of AlMe3 reduced conversion without influencing
enantioselectivity. Employed amounts of AlMe3 compensate for residual
moisture in the reaction mixture and binds to the pyrrolidone product.
(22) Possible activation via five-coordinate aluminum species:
(a) Elms, F. M.; Gardiner, M. G.; Koutsantonis, G. A.; Raston, C. L.;
Atwood, J. L.; Robinson, K. D. J. Organomet. Chem. 1993, 449, 45.
(b) Mitra, A.; DePue, L. J.; Parkin, S.; Atwoodm, D. A. J. Am. Chem. Soc.
2006, 128, 1147.
(8) Reviews on C−H functionalization: (a) Kuhl, N.; Hopkinson, M.
N.; Wencel-Delord, J.; Glorius, F. Angew. Chem., Int. Ed. 2012, 51,
10236. (b) Yamaguchi, J.; Yamaguchi, A. D.; Itami, K. Angew. Chem., Int.
Ed. 2012, 51, 8960. (c) Ackermann, L. Chem. Rev. 2011, 111, 1315.
(d) McMurray, L.; O’Hara, F.; Gaunt, M. J. Chem. Soc. Rev. 2011, 40,
1885. (e) Gutekunst, W. R.; Baran, P. S. Chem. Soc. Rev. 2011, 40, 1976.
(f) Yu, J.-Q.; Shi, Z.-J. In Topics in Current Chemistry 292; Springer-
Verlag: Berlin, 2010. (g) Colby, D. A.; Bergman, R. G.; Ellman, J. A.
(23) Preliminary results with nonterminal alkene substrates provide
hydrocarbamoylated products in lower selectivities.
(24) Nifant’ev, E. E.; Trofimtsova, I. A.; Chekulaeva, L. A.;
Chechegoeva, E. V.; Gavrilenko, V. V. Zh. Obshch. Khim. 1992, 62, 1027.
D
dx.doi.org/10.1021/ja406730t | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX