Development of Non-C2-symmetric ProPhenol Ligands. the Asymmetric Vinylation of N-Boc Imines

Article abstract of DOI:10.1021/acs.orglett.5b01755

The development and application of a new generation of non-C2-symmetric ProPhenol ligands is reported herein. Rational design of the ProPhenol ligand paved the way to the first catalytic and asymmetric vinylation of N-Boc imines via hydrozirconation giving rise to valuable allylic amines in excellent yields and enantioselectivities. The utility of this method was demonstrated by developing the shortest reported asymmetric synthesis of the selective serotonine reuptake inhibitor (SSRI) (-)-dapoxetine.

Full text of DOI:10.1021/acs.orglett.5b01755

Letter
pubs.acs.org/OrgLett
Development of Non‑C2‑symmetric ProPhenol Ligands. The
Asymmetric Vinylation of N‑Boc Imines
Barry M. Trost,* Chao-I (Joey) Hung,‡ Dennis C. Koester,‡ and Yan Miller
Department of Chemistry, Stanford University, Stanford, California 94305-5080, United States
S
* Supporting Information
ABSTRACT: The development and application of a new generation of
non-C2-symmetric ProPhenol ligands is reported herein. Rational design of
the ProPhenol ligand paved the way to the first catalytic and asymmetric
vinylation of N-Boc imines via hydrozirconation giving rise to valuable
allylic amines in excellent yields and enantioselectivities. The utility of this
method was demonstrated by developing the shortest reported asymmetric
synthesis of the selective serotonine reuptake inhibitor (SSRI) (−)-dapox-
etine.
rafting chiral space is a major challenge for synthetic
organic chemists in order to develop highly enantiose-
Scheme 1. Synthetic Approach to a New Family of Non-C2-
symmetric ProPhenol Ligands
C
lective, catalytic processes. By learning from biological
mechanisms for asymmetric catalysis in nature, we set out to
develop a biomimetic catalytic system tailored for asymmetric
addition reactions. We and others have reported on several
successful applications of the commercially available C2-
symmetric ProPhenol ligand.1 The versatility of this catalytic
system propelled our interest in further elaborating its properties.
We were driven by the question, whether electronically distinct
aryl groups on each of the diphenyl prolinol subunits could affect
the performance of the dinuclear catalyst. In other words, can one
induce differences in the electronic nature of the zinc centers by
perturbation of electronics in a remote manner?
The C2-symmetric ProPhenol ligand relies upon steric
differentiation of the chiral space to effect asymmetric induction.1
The ability of the ProPhenol ligand to self-assemble a dinuclear
catalyst sets the stage for an interaction with both addition
reaction partners through a Lewis acidic and a Lewis basic site
and makes it highly attractive for further engineering. Not only
the Zn-ProPhenol complex but also dinuclear ProPhenol
complexes of other metals such as magnesium have been
developed and proved superior in some of the investigated
reactions.2 Recently, additional steric encumbrance in the
prolinol core proved to be of importance for the diastereo- and
enantioselective preparation of β-hydroxy-α-amino esters
through a Zn-ProPhenol catalyzed aldol reaction.1a In this
transformation we demonstrated the differential effect of the
nature of the ProPhenol ligand L4 on the enantioselectivity of
the reaction (Scheme 1). This second generation of ProPhenol
adds stereocenters to the prolinol core, but still leaves the C2-
symmetry untouched. In further pursuing the development of
the ProPhenol ligand, we sought to achieve distinction of the two
zinc atoms in the catalyst through electronic desymmetrization of
the chiral space. We designed a non-C2-symmetric ligand by
attaching electronically different aromatics to either prolinol
(Scheme 1). The result is a push−pull system to generate a more
electrophilic, Lewis acidic and a more nucleophilic, Lewis basic
zinc atom in the same dinuclear catalyst. By tuning the electronic
properties of the zinc centers, hence differentiating the chiral
space, we sought to enhance enantioselectivity as well as
reactivity in the catalytic transformation.
The synthesis of the desired ligands was achieved expediently
from salicylaldehyde 1, which can be prepared from
commercially available 2-hydroxy-5-methylbenzaldehyde in one
step, or more economically from p-cresol in two steps.3
Nucleophilic substitution of the benzyl bromide under basic
conditions with the first prolinol unit 2 followed by a reductive
amination of the aldehyde 3 with the second prolinol unit 4
furnished the desired non-C2-symmetric ProPhenol ligands L1−
3 in 62−76% yields over two steps. The major assets of this
Received: June 16, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.5b01755
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
approach are its convergence, generality, robustness, reliability,
and scalability.
Table 1. Reaction Development
The application of this new generation of ProPhenol ligands is
demonstrated through the asymmetric vinylation of N-Boc
imines giving rise to valuable allylic amines, a process that
heretofore has not been demonstrated. Allylic amines are
important in their own right, but are also intermediates to
more elaborate saturated amines. They occur in numerous
pharmaceuticals with a broad range of biological activity such as
the antiepileptic drug vigabatrin (S-enantiomer is pharmacolog-
ically active)4 or the antiparkinson agent lisuride.5 The catalytic,
asymmetric synthesis of allylic amines typically involves one of
two strategies, asymmetric formation of the C−N6 or the C−C
bond.7 The utilization of an inexpensive main-group metal has
economic advantages over the use of expensive transition metal
catalysts. The high costs of Rh- or Ir-catalysts7,8 previously
employed as well as the requirement for the N-tosyl imines, a
notorious amine protecting group because of the difficulties in its
removal, are major drawbacks of these methods. Further the Ir-
catalyzed process requires the use of disubstituted alkynes.
Interestingly, the vinylation of N-Boc imines has never been
reported to date. Given the benefits of Boc as the nitrogen
protecting group (vide infra) the direct synthesis of allylic N-Boc
amines became an attractive goal.
The nucleophiles of choice were vinylzirconium species
resulting from the hydrozirconation of alkynes employing
Schwartz’ reagent deriving from DIBAL and Cp2ZrCl2, a gasoline
additive.9,10 The easy accessibility and the stability of
alkenylzirconium species is a great asset of zirconium chemistry.
However, previous reports utilizing vinylzinc species resulting
from vinyl zirconium species relied on stoichiometric use of
organozinc compounds in order to achieve the transmetalation
from zirconium to zinc.7,10 This prompted us to explore our
newly designed Zn-ProPhenol system in the first catalytic,
asymmetric transmetalation from zirconium to zinc in a CX
addition reaction to form a C−C bond. The presented method
obviates the use of expensive transition metal catalysts such as Rh
or Ir, but rather makes use of a biomimetic approach employing
catalytic amounts of Zn in a dinuclear Zn-ProPhenol catalyst to
allow for the first asymmetric vinylation of N-Boc imines.1 The
design of a new generation of ProPhenol ligands proved to be
crucial for the success of this process.
b
c
entry
1
catalyst system (mol %)
ee (%)
yield (%)
ProPhenol 7a (15 mol %)
ProPhenol 7b (15 mol %)
ProPhenol 7c (15 mol %)
ProPhenol 7d (15 mol %)
ProPhenol 7e (15 mol %)
ProPhenol 7f (15 mol %)
ProPhenol 7f (7.5 mol %)
ProPhenol 7f (15 mol %)
ProPhenol 7g (15 mol %)
ProPhenol 7g (15 mol %)
82
80
85
42
30
81
42
87
45
51
d
2
e
3
4
5
6
n.d.
81
80
94
94
85
93
81
f
7
g
8
h
9
52
i
10
20
a
Conditions: 5a (0.16 mmol), 6a (0.24 mmol), Zn-ProPhenol 7 (15
mol %), DME/THF (4:1, 1.0 mL), 4 °C, 48 h, under Ar. Ee
determined by HPLC. Isolated yield. Reaction was performed with
1-hexyne instead of phenylacetylene. Not determined due to
overlapping impurity. The reaction performed at rt instead of 4 °C.
Only DME was used as solvent. Material was not entirely pure.
b
c
d
e
f
g
h
i
Additive: Methyl hydroxypivalate (15 mol %).
electronic differentiation. Surprisingly, the selectivity (entry 4)
was similar to the parent ProPhenol 7a. On the other hand, using
the combination of phenyl and p-trifluoromethylphenyl in the
prolinol backbone (7f) was found to be optimal to provide both
excellent yields and enantioselectivities. The desired allyl amine
was isolated in a good yield of 87% and an excellent
enantioselectivity of 94% (entry 6). Control experiments
revealed that only the dinuclear Zn-ProPhenol complex is an
effective catalyst in this transformation. A catalyst loading of 15
mol % was required to ensure high conversions; the high
enantioselectivity, however, was retained even with catalyst
loadings as low as 7.5 mol % (entry 7).
In contradistinction to all previous reports,7,10 we identified N-
Boc imines as the electrophiles of choice in the presented
asymmetric vinylation. Their easy synthesis and handling as well
as their excellent properties as a nitrogen protecting group are
major advantages of the strategy. The Boc-group can easily be
removed under slightly acidic conditions, compatible with a
broad range of functional groups,11 whereas the deprotection of
sulfonyl amines usually requires harsh reaction conditions
hampering the further modification of the products.12
An extensive screening of reaction conditions was undertaken
(Table 1 and Supporting Information (SI) for further details).
The parent ProPhenol 7a (Table 1, entry 1) gave promising
results; however, further improvement by varying reaction
conditions was not achieved. Increasing the Lewis basicity as in
the tetrakis(p-methoxyphenyl) analog 7b (entry 2) furnished the
desired product in 42% yield. An increased Lewis acidity of both
zinc atoms as in tetrakis(p-trifluoromethylphenyl) analog 7c
(entry 3) led to even worse results. Turning to electronically
desymmetrizing the complex to enhance the differentiation of
the chiral space led us to the bis(p-methoxyphenyl)-bis(p-
trifluoromethylphenyl) analog 7d wherein we maximized the
Further, the new ProPhenol ligand is easily recovered in 92%
yield by employing a simple acid−base extraction as we have
reported for the standard ProPhenol ligand previously.13
A
mixture of DME/THF = 4:1 proved to be the solvent of choice.
Only the combination of both solvents led to a high yield and
enantioselectivity. The ideal reaction temperature was identified
as 4 °C. Standard additives,1,2 including triphenylphosphine
oxide, cis-1,2-cyclopentanediol, and methyl hydroxypivalate,
which are crucial in numerous reported Zn-ProPhenol catalyzed
reactions were found to be counterproductive in terms of both
reactivity and enantioselectivity (for a detailed screening table
see SI).
With the optimized conditions in hand, we examined the scope
of this catalytic, asymmetric vinylation of N-Boc imines (Scheme
2). To our delight, we found that a variety of electronically and
sterically distinct aromatic N-Boc imines were competent
substrates in the described transformation. Using many of
these same N-Boc imines with the parent ProPhenol ligand gave
lower enantioselectivities in the 70−81% range. Substituents in
the para, meta, and even ortho positions of the aryl imines were
successful. The electronic nature of the imine seems to have no
B
DOI: 10.1021/acs.orglett.5b01755
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
The unprecedented application of an alkynyl-aryl ether in a
hydrozirconation reaction allowed us to report the shortest
enantioselective synthesis of the selective serotonin reuptake
inhibitor (SSRI) (−)-dapoxetine.15 The drug was synthesized
from 1-naphthol and trichloroethene in four simple synthetic
operations (Scheme 4).
Scheme 2. Catalytic Asymmetric Vinylation of Imines under
Dinuclear Zn-ProPhenol Catalysis
a b
,
Scheme 4. Shortest Enantioselective Synthesis of
(−)-Dapoextine
The alkynyl-aryl ether 12 was synthesized by a literature
known method.16 Allylic amine 8ae was accessed in 90% yield
and 93% ee by applying the optimized reaction conditions for the
asymmetric vinylation. A one-pot approach making use of Pd/C
catalyzed hydrogenation, in situ deprotection of the N-Boc group
mediated by TFA, and final in situ reductive amination with
formaldehyde furnished the desired (−)-dapoxetine in 92% yield.
Vinyldimethylbenzyl silane 8ca was synthesized in 80% yield
with 91% ee utilizing the optimized reaction conditions. To
introduce functional groups, such as esters and ketones, that are
generally incompatible with the hydrozirconation process, a
Hiyama coupling with aryl iodides could be readily performed
(Scheme 5). Employing ester substituted aryl iodide 14, we were
a
Conditions: 1 (0.16 mmol), 2 (0.24 mmol), 30 mol % ZnEt2, 15.0
mol % L3 in DME/THF (4:1, 1.0 mL) at 4 °C, 48 h, under Ar;
b
isolated yield. Reactions employing 7a generally give ee’s in the range
of 70−81%.
major impact on the enantioselectivity of the reaction. In most
cases the ee is above 90%. We found electron-poor aromatic
imines give slightly higher yields (80% to 95%) than electron-
neutral and -rich aromatic imines. On the other hand, the
electron-rich furan- and thiophene-derived imines performed
well under the optimized reaction conditions.
Scheme 5. Synthetic Application of the Synthesized Allyl
Amines
The scope of the alkynes was also thoroughly investigated.
Both, aromatic and aliphatic acetylenes were equally viable
substrates (phenylactylene 87% yield, 94% ee; 1-hexyne 82%
yield, 94% ee; see Scheme 2). Protected propargyl alcohols are
effective substrates, and distant halides in an aliphatic acetylene
are well tolerated. Most notably, vinyl benzyldimethylsilyl amines
can be accessed setting the stage for further functionalization of
the vinyl unit via Hiyama cross-coupling.14 Gratifyingly, enynes
could be employed in the desired transformation without loss of
reactivity and enantioselectivity, giving rise to chiral substituted
(E,E)-α,β,γ,δ-unsaturated amines never obtained by any other
vinylation method before.
The reaction was also applicable to α,β-unsaturated imines,
albeit with slightly lower enantioselectivities (Scheme 3).
Unprecedented chiral diallylic amines can be readily synthesized
in high yields of 87−95% with good enantioselectivities from 74
to 77% ee.
able to synthesize allylic amine 8cb in 82% yield. This compound
can readily be converted to benzazepinones which possess
profound chemotherapeutic properties.17
The proposed catalytic cycle is depicted in Scheme 6. After the
formation of the dinuclear Zn-ProPhenol complex 7, trans-
metalation from the vinylzirconium species to the zinc catalyst
occurs and places the nucleophile in a chiral environment (15). A
bidentate coordination of the N-Boc group to both zinc atoms
and the electronic differentiation by the introduction of an
electron-withdrawing group to one side of the ProPhenol ligand
may account for the high enantioselectivities achieved with N-
Boc imines and the employed non-C2-symmetric ligand L3. The
nucleophilic addition of the vinylzinc species into the
coordinated imine takes place in a six-membered transition
state leading to the enantiomerically enriched allylic amine.
Finally, a transmetalation of the vinylzirconium species will then
liberate carbonimidate 18, which will later be transformed to the
desired allylic amine 8 upon protonation during workup. The
catalytic cycle starts again with the coordination of an additional
imine 5.
Scheme 3. Application of α,β-Unsaturated Imines for the
Asymmetric Synthesis of Diallylic Amines
C
DOI: 10.1021/acs.orglett.5b01755
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 6. Proposed Catalytic Cycle
ACKNOWLEDGMENTS
■
We thank the U.S. National Science Foundation (CHE-
1145236) for generous support of our program and the
Alexander von Humboldt Foundation for fellowship support to
D.C.K.
REFERENCES
■
(1) (a) Trost, B. M.; Miege, F. J. Am. Chem. Soc. 2014, 136, 3016−3019
and references therein. (b) Wang, B.-L.; Li, N.-K.; Zhang, J.-X.; Liu, G.-
G.; Liu, T.; Shen, Q.; Wang, X.-W. Org. Biomol. Chem. 2011, 9, 2614−
2617. (c) Smirnov, P.; Mathew, J.; Nijs, A.; Katan, E.; Karni, M.; Bolm,
C.; Apeloig, Y.; Marek, I. Angew. Chem., Int. Ed. 2013, 52, 13717−13721.
(d) Smirnov, P.; Katan, E.; Mathew, J.; Kostenko, A.; Karni, M.; Nijs, A.;
Bolm, C.; Apeloig, Y.; Marek, I. J. Org. Chem. 2014, 79, 12122−12135.
(2) (a) Trost, B. M.; Malhotra, S.; Fried, B. A. J. Am. Chem. Soc. 2009,
131, 1674−1675. (b) Trost, B. M.; Hisaindee, S. Org. Lett. 2006, 8,
6003−6007.
(3) Hofsløkken, N. U.; Skattebøl, L. Acta Chem. Scand. 1999, 53, 258−
262.
(4) (a) Gram, L.; Larsson, O. M.; Johnsen, A.; Schousboe, A. Br. J. Clin.
Pharmacol. 1989, 27, 13S−17S. (b) Sheean, G.; Schramm, T.; Anderson,
D. S.; Eadie, M. J. Clin. Exp. Neurol. 1992, 29, 107−116.
(5) Hofmann, C.; Penner, U.; Dorow, R.; Pertz, H. H.; Jahnichen, S.;
̈
Horowski, R.; Latte, K. P.; Palla, D.; Schurad, B. Clin. Neuropharmacol.
2006, 29, 80−86.
In summary, we have developed a new generation of non-C2-
symmetric ProPhenol ligands that proved to enhance the
selectivity by electronically differentiating the chiral space in
the dinuclear zinc ProPhenol catalyst. It is remarkable that such a
long-range effect can have a profound impact on the selectivity.
These readily prepared and quantitatively recoverable ligands
allowed the first enantioselective addition of vinylzirconocenes to
N-Boc imines. Ease of recovery and recycling the ProPhenol
ligand mitigate the loading levels for a single run wherein the net
effective loading becomes much less. This report displays the first
asymmetric vinylation of an imine in the absence of a precious
metal or a chiral auxiliary. Invoking the first catalytic trans-
metalation from zirconium to zinc obviating the need for
preformation of a vinylzinc species provides a facile way to access
valuable allylic amines in excellent yields and enantioselectivities.
The scope of the reaction comprises not only aromatic but also
α,β-unsaturated N-Boc imines. A wide range of different
acetylenes, including enynes and an alkynyl ether, were
employed successfully. We were able to demonstrate the
synthetic utility of the generated allylic amines by presenting
the shortest asymmetric synthesis of (−)-dapoxetine. The
differential effect of the new ProPhenol generation on various
reactions is currently being investigated in our laboratory, and
results will be published in due course.
(6) Review on allylamine synthesis: (a) Johannsen, M.; Jørgensen.
Chem. Rev. 1998, 98, 1689−1708. C−N-bond formation, asymmetric:
(b) Chen, Y. K.; Lurain, A. E.; Walsh, P. J. J. Am. Chem. Soc. 2002, 124,
12225−12231. (c) Arnold, J. S.; Nguyen, H. M. J. Am. Chem. Soc. 2012,
134, 8380−8383. (d) Nagano, T.; Kobayashi, S. J. Am. Chem. Soc. 2009,
131, 4200−4201.
(7) (a) Wipf, P.; Kendall, C.; Stephenson, C. R. J. J. Am. Chem. Soc.
2003, 125, 761−768. (b) Patel, S. J.; Jamison, T. F. Angew. Chem., Int. Ed.
2004, 43, 3941−3944. (c) Ngai, M.-Y.; Barchuk, A.; Krische, M. J. J. Am.
Chem. Soc. 2007, 129, 12644−12645. (d) Gopula, B.; Chiang, C.-W.;
Lee, W.-Z.; Kuo, T.-S.; Wu, P.-Y.; Henschke, J. P.; Wu, H.-L. Org. Lett.
2014, 16, 632−635.
(8) Skucas, M. Y.; Ngai, V.; Komanduri, V.; Krische, M. J. Acc. Chem.
Res. 2007, 40, 1394−1401.
(9) For reviews on hydrozirconation, see: (a) Song, Z.; Takahashi, T.
Comprehensive Organic Synthesis II 2014, Vol. 8, 838−876. (b) Wipf, P.;
Kendall, C. Top. Organomet. Chem. 2004, 8, 1−25.
(10) (a) Li, H.; Walsh, P. J. J. Am. Chem. Soc. 2004, 126, 6538−6539.
(b) Wipf, P.; Pierce, J. G. Org. Lett. 2005, 7, 3537−3540.
(11) Shendage, D. M.; Frohlich, R.; Haufe, G. Org. Lett. 2004, 6, 3675−
̈
3678.
́
(12) (a) Alonso, E.; Ramon, D. J.; Yus, M. Tetrahedron 1997, 53,
14355−14368. (b) Shohji, N.; Kawaji, T.; Okamoto, S. Org. Lett. 2011,
13, 2626−2629.
(13) Trost, B. M.; Bartlett, M. J. Acc. Chem. Res. 2015, 48, 688−701.
(14) Trost, B. M.; Ball, Z. T. J. Am. Chem. Soc. 2005, 127, 17644−
17655.
ASSOCIATED CONTENT
* Supporting Information
■
(15) (a) Torre, O.; Gotor-Fernandez, V.; Gotor, V. Tetrahedron:
Asymmetry 2006, 17, 860−866. (b) Alt, O. A.; Robey, R. L.; Meter, E. E.
V. U.S. Pat. 5292962, 1994. (c) Robertson, D. W.; Thompson, D. C.;
Wong, D. T. Eur. Pat. Appl. EP 288188A1 19881026, 1988. (d) Siddiqui,
S. A.; Srinivasan, K. V. Tetrahedron: Asymmetry 2007, 18, 2099−2103.
(e) Chincholkar, P. M.; Kale, A. S.; Gumaste, V. K.; Rakeeb, A.;
Deshmukh, A. S. Tetrahedron 2009, 65, 2605−2609. (f) Kang, S.; Lee,
H.-K. J. Org. Chem. 2010, 75, 237−240. (g) Mahale, R. D.; Chaskar, S. P.;
Patil, K. E.; Maikap, G. C.; Gurjar, M. K. Org. Process Res. Dev. 2012, 16,
710−713. (h) You, P.; Qiu, J.; Su, E.; Wei, D. Eur. J. Org. Chem. 2013,
2013, 557−565.
S
Experimental procedures and full spectroscopic data for all new
compounds. The Supporting Information is available free of
charge on the ACS Publications website at DOI: 10.1021/
acs.orglett.5b01755.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: bmtrost@stanford.edu.
(16) Komine, Y.; Tanaka, K. Org. Lett. 2010, 12, 1312−1315.
(17) Lebrun, S.; Couture, A.; Deniau, E.; Grandclaudon, P. Synthesis
2012, 44, 1410−1416.
Author Contributions
‡C.-I.H. and D.C.K. contributed equally.
Notes
The authors declare no competing financial interest.
D
DOI: 10.1021/acs.orglett.5b01755
Org. Lett. XXXX, XXX, XXX−XXX