Umpolung Synthesis of Diarylmethylamines via Palladium- Catalyzed Arylation of N-Benzyl Aldimines

Article abstract of DOI:10.1002/adsc.201600075

An umpolung synthesis of diarylmethylamine derivatives is presented. This reaction entails a palladium-catalyzed arylation of 1,3-diaryl-2-azaallyl anions, in situ generated from N-benzyl aldimines. A Pd(NIXANTPHOS)-based catalyst together with hindered silylamide bases enabled the coupling of aldimines with aryl bromides in good to excellent yields without product isomerization. Moreover, regioselectivity in the arylation of unsymmetrical 1,3-diaryl-2-azaallyl anions was studied. This method is suitable for a gram scale synthesis of diarylmethylamine derivatives at room temperature without use of a glove box. (Figure presented.) .

Full text of DOI:10.1002/adsc.201600075

COMMUNICATIONS
DOI: 10.1002/adsc.201600075
Umpolung Synthesis of Diarylmethylamines via Palladium-
Catalyzed Arylation of N-Benzyl Aldimines
Minyan Li,+a Baris Yucel,+a,b Jacqueline JimØnez,a,c Madeline Rotella,a Yue Fu,a
and Patrick J. Walsha,*
a
Roy and Diana Vagelos Laboratories, Department of Chemistry, University of Pennsylvania, 231 South 34th Street,
Philadelphia, Pennsylvania 19104-6323, USA
Fax: (+1)-215-573-6743; e-mail: pwalsh@sas.upenn.edu
Istanbul Technical University, Science Faculty, Department of Chemistry, Maslak 34469, Istanbul, Turkey
Facultad de Ciencias Químicas, BenemØrita Universidad Autónoma de Puebla, Av. San Claudio y 14 Sur, Puebla, Pue
b
c
72570, MØxico
+
M. Li and B. Yucel contributed equally.
Received: January 18, 2016; Revised: March 24, 2016; Published online: May 12, 2016
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201600075.
Abstract: An umpolung synthesis of diarylmethyl-
amine derivatives is presented. This reaction entails
ed into diarylmethylamine derivatives.[4] Recently,
Nolan and co-workers showed that this chemistry can
also be performed with Ni(NHC)-based catalysts
(Scheme 1d).[5]
a
palladium-catalyzed arylation of 1,3-diaryl-2-
azaallyl anions, in situ generated from N-benzyl ald-
imines. A Pd(NIXANTPHOS)-based catalyst to-
gether with hindered silylamide bases enabled the
coupling of aldimines with aryl bromides in good to
excellent yields without product isomerization.
Moreover, regioselectivity in the arylation of un-
symmetrical 1,3-diaryl-2-azaallyl anions was studied.
This method is suitable for a gram scale synthesis of
diarylmethylamine derivatives at room temperature
without use of a glove box.
Despite the broad scope and high yields of the ary-
lated products, the arylation of ketimines and ald-
imines in Scheme 1c leaves room for improvement.
First, formation of two by-products detracts from the
methodꢀs utility. As shown in Scheme 2, the starting
ketimine undergoes deprotonation and isomerization
to the aldimine, which requires more forcing condi-
tions to deprotonate and regenerate the 2-azaallyl
anion. Additionally, the 2-azaallyl anion forms
a dimer, presumably through reaction with an equiva-
lent of ketimine (Scheme 2). This process consumes
two equivalents of the starting material. To circum-
vent these side reactions and maintain high yields, an
excess of the 2-azaallyl anion precursor was necessary
(usually 2 equiv.).[6]
Keywords: 2-azaallyl anion; diarylmethylamines;
NIXANTPHOS; regioselectivity; umpolung
To improve our method to prepare diarylmethyl-
In the past decade, considerable effort has been de- amine derivatives, we envisioned an approach that
voted to the synthesis of diarylmethylamines due to could avoid the side reactions in Scheme 2 while also
their common occurrence in bioactive compounds.[1] increasing the overall synthetic efficiency of the pro-
Among the methods for their synthesis,[2] 2-azaallyl cess. Herein, an effective umpolung synthesis of
anions derived from either ketimines or aldimines diarylmethylamines via palladium-catalyzed arylation
have recently emerged as a valuable umpolung ap- of 1,3-diaryl-2-azaallyl anions is described. Under
proach to prepare both diarylmethylamine and ben- these reaction conditions, no dimerization or isomeri-
zylic amine derivatives, as demonstrated by Oshima, zation of the starting materials was observed. Addi-
Buchwald, Nolan and us (Scheme 1a–d).[3] Recently, tionally, good to excellent yields of the triaryl aldi-
we reported the regioselective arylation of 1,1,3-triar- mine products were obtained without product isomer-
yl-2-azaallyl anions generated from N-benzyl ket- ization or further arylation of the more acidic prod-
imines and aldimines (Scheme 1c). In the presence of uct.
a Pd(NIXANTPHOS)-based catalyst and hindered
To render the arylation of 2-azaallyl anions more
silylamide bases, 2-azaallyl anions are readily convert- efficient we focused on 1,3-diaryl-2-azaallyl anions 7
Adv. Synth. Catal. 2016, 358, 1910 – 1915
1910
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

COMMUNICATIONS
asc.wiley-vch.de
Scheme 1. Synthesis of diarylmethylamine and benzylic amine derivatives from arylation of 2-azaallyl anions.
Scheme 2. Side reactions in arylation of N-benzyl ketimines 1 and aldimines 3.
(Scheme 3). The challenge of this approach is that the and Scheme 3b). Therefore, it is important that the
product 8 also readily undergoes arylation (Scheme 1c catalyst promote the selective formation of the mono-
Adv. Synth. Catal. 2016, 358, 1910 – 1915
1911
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

COMMUNICATIONS
asc.wiley-vch.de
Table 1. Scope of aryl bromides in the arylation of aldimine
6a.[a]
Scheme 3. Palladium-catalyzed arylation of 1,3-diaryl-2-
azaallyl anions 7.
arylation product. Moreover, the conditions must be
sufficiently mild such that the product does not under-
go deprotonation, as outlined above.
To identify conditions for the arylation in
Scheme 3a, we studied the reaction of 1.0 equivalent
of aldimine 6 with 1.5 equivalents of 1-bromo-4-tert-
butylbenzene (10a).
A Pd(NIXANTPHOS)-based
catalyst, which exhibited high efficiency in our previ-
ous deprotonative cross-coupling processes (DCCP),
including Scheme 1c, was examined.[4,7] We tested dif-
ferent silylamide bases, MN(SiMe3)2 (M=Li, Na and
K) leading to up to 93% isolated yield of the desired
product 8aa. To prevent the bis-arylation and achieve
this high yield, it was necessary to add the 1.5 equiva-
lents of LiN(SiMe3)2 over 1 h in THF at room temper-
ature (Table 1). It is surprising that the reaction could
be stopped at the monoarylation stage. Comparison
of the reaction parameters between Table 1 and
Scheme 1c highlights the importance of the base. In
both reactions, the same catalyst system was used, but
the more reactive NaN(SiMe3)2 is employed with the
more sterically conjested ketimine substrates, while
the LiN(SiMe3)2 is used in Table 1.
[a]
Reactions were conducted on a 0.2 mmol scale at 0.1M
and were stirred for 1 h after slow addition (1 h) of
LiN(SiMe3)2 into the reaction mixtures.
Under same reaction conditions, changing the base to
NaN(SiMe3)2 and KN(SiMe3)2 led to 53% and 6% assay
yields.
[b]
(Scheme 4). The reaction was conducted without use
The scope of the reaction was next briefly exam- of a glove box and employed a commercially avail-
ined. The reactions of the parent aldimine 6a with able 1.0M solution of LiN(SiMe3)2 (15.0 mmol) in
bromobenzene 10b and 3-bromotoluene 10c furnished THF. After 1.5 h slow addition of the base by syringe,
the corresponding products 8ab and 8ac in 75 and the reaction was stirred for an additional 2 h. After
88% yield, respectively. Aryl bromides bearing elec- work-up, 85% yield (2.67 g) of product was obtained
tron-donating 4-methoxy (10d) and 4-bromo-N,N-di- (see the Supporting Information for details). The
methylaniline (10e) groups produced the desired imine 8ae was then hydrolyzed with 1N HCl in
imines 8ad and 8ae in excellent yields (85% and 93%, MeOH, free-based with NaOH, and the amine 11ae
respectively). The reaction of imine 6a with 1-bromo- was isolated in 98% yield.
4-fluorobenzene (8af) gave 52% yield of the mono-ar-
Although the chemistry in Table 1 is useful, in
ylation product with 16% bis-arylated product 9af. some cases the synthesis of symmetric aldimines may
Although the yield of 8af was diminished due to bis- be impractical due to availability or cost of the amine
arylation, it is surprising that no isomerizaiton or di- or aldehyde precursors. We therefore turned our at-
merization of 8af was observed.
tention to the arylation of unsymmetrical 1,3-diaryl-2-
The scalability of this umpolung approach was next azaallyl anions. We wanted to explore the impact of
studied. We performed a 10.0 mmol scale reaction of electronic and steric effects on the distribution of
imine 6a with 15.0 mmol 4-bromo-N,N-dimethylani- product isomers (Scheme 5a). In order to develop the
line (10e) in the presence of 2.5 mol% Pd(OAc)2 and arylation of unsymmetrical 1,3-diaryl-2-azaallyl anions
5.0 mol% NIXANTPHOS at room temperature into a synthetically useful method, we hypothesize
Adv. Synth. Catal. 2016, 358, 1910 – 1915
1912
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

COMMUNICATIONS
asc.wiley-vch.de
substituted aldimine 6c. Under the reaction conditions
listed in Table 2, coupling of 6c with 1-bromo-4-tert-
butylbenzene (10a) afforded product 8ca as a single
isomer in 77% yield. We therefore examined the sub-
strate scope based on mesityl-substituted aldimine nu-
Table 2. Scope of aryl bromides in the arylation of mesityl-
substituted aldimines.[a]
Scheme 4. Large scale synthesis of imine 8ae and its hydrol-
ysis to produce diarylmethylamine 11ae.
Scheme 5. Palladium-catalyzed arylation of o-anisyl-substi-
tuted imine 6b.
[a]
that increasing steric hinderance[8] of one aryl would
enhance the regioselectivity. Hence, we studied the ar-
ylation of o-anisyl substituted imine 6b with 1-bromo-
4-tert-butylbenzene (10a). Under the reaction condi-
tions in Scheme 5b, arylation mainly occurred at the
sterically less hindered position with a 4:1 ratio of iso-
mers 8ba and 8ba’ in 76% overall yield.
Reactions conducted on a 0.2 mmol scale by using
1.0 equiv. aldimine, 1.5 equiv. aryl bromide and 1.5 equiv.
LiN(SiMe3)2 at 0.1M. Slow addition of LiN(SiMe3)2 over
1 h followed by 4 h stirring.
2.0 equiv. aryl bromide, 6 h, 508C.
2.0 equiv. aryl bromide, 6 h.
2.0 equiv. aryl bromide, slow addition of LiN(SiMe3)2 in
1.5 h followed by 6 h reaction time.
2.0 equiv. aldimine, 1.0 equiv. aryl bromide and 2.0 equiv.
LiN(SiMe3)2 at 0.05M.
[b]
[c]
[d]
[e]
The moderate selectivity in the arylation of 6b led
us to examine the sterically more demanding mesityl-
Adv. Synth. Catal. 2016, 358, 1910 – 1915
1913
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

COMMUNICATIONS
asc.wiley-vch.de
cleophiles. Arylation of 6c with aryl bromides con- and unsymmetrical 1,3-diaryl-2-azaallyl anions de-
taining alkyl (4-C6H4-t-Bu, Ph, 3-C6H4-Me), electron rived from N-benzyl aldimines under mild conditions.
donating (4-C6H4-OMe, 4-C6H4-NMe2) and withdraw- The reaction is promoted by a Pd(NIXANTPHOS)-
ing (4-C6H4-F) groups produced the desired products based catalyst. Interestingly, this catalyst exhibits ex-
8cb–8cf in good yields (67–82%). The yield of cou- cellent selectivity in the mono-arylation of 1,3-diaryl-
pling the mesityl-substituted aldimine 6c and 1- 2-azaallyl anions, despite its known ability to promote
bromo-4-fluorobenzene (10f) was 68%. This yield can the arylation of the product to generate tetraaryl ket-
be compared with coupling of the symmetrical 6a imines. Selectivity in these reactions is controlled by
with 1-bromo-4-fluorobenzene, which provided only careful choice of reaction conditions. It is also note-
52% yield of the mono-arylation product (Table 1).
worthy that the arylation of 1,3-diaryl-2-azaallyl
Next, the substituents on the aldimine were varied. anions provides various diarylmethylamine derivatives
Coupling of 4-methoxy-substituted aldimine 6d with without product isomerization or dimerization. This
1-bromo-4-N,N-dimethylaniline (10e) was accom- method is also amenable to gram scale synthesis in
plished by slow addition (1.5 h) of LiN(SiMe3)2 to the a time efficient manner.
reaction mixture at room temperature followed by
heating to 508C for 5 h. The arylation of fluoro-substi-
tuted aldimines bearing either 4-fluoro (6e) or 2- Experimental Section
fluoro (6f) substituents with 4-bromoanisole (10d)
and 4-bromo-N,N-dimethylaniline (10e) required General Procedure and Characterization for Pd-
longer reaction times (6 h) at room temperature for Catalyzed Reactions of Aldimines
complete consumption of aldimines and gave the
An oven-dried microwave vial equipped with a stir bar was
products 8ed, 8ee, and 8fe in 60–62% yield. To access
charged with aldimine 6a (39.0 mg, 0.20 mmol) and 1-
heteroaryl-substituted diarylmethylamines, we per-
bromo-4-tert-butylbenzene 10a (51.9 mL, 0.30 mmol) under
form the reaction of 3-pyridyl-substituted aldimine 6g
with 4-bromo-N,N-dimethylaniline 10e. The product
8ge was obtained in 80% yield with aryl bromide 10e
as limiting reagent.
a nitrogen atmosphere in a glove box. A stock solution of
Pd(OAc)2 (2.25 mg, 0.010 mmol, 5 mol%) and NIXANT-
PHOS (11.10 mg, 0.020 mmol, 10 mol%) in 0.5 mL of dry
THF was taken up by syringe and added to the reaction vial
under nitrogen. The vial was sealed with a cap (with rubber
septum) and removed from the glove box. A solution of
LiN(SiMe3)2 (50.2 mg, 0.30 mmol) in 1.5 mL THF was added
portionwise by syringe at 238C over 1 h. The reaction mix-
ture was stirred for 1 h after the addition (2 h in total) at
238C then opened to air, quenched with two drops of H2O,
diluted with 3 mL of ethyl acetate, and filtered over a pad
(a 6 mL Syringe) of MgSO4 and silica. The pad was rinsed
with ethyl acetate (3”2 mL), and the combined solutions
were concentrated under vacuum. The crude material was
loaded onto a deactivated silica gel column and purified by
flash chromatography using 20:1 hexanes/ethyl acetate as
eluent to yield the product 8aa as a white solid; yield: 61 mg
(0.19 mmol, 93%).
Finally, given Eastgate and Blackmondꢀs recent
report of reactions catalyzed by mono-oxidized biden-
tate XANTPHOS and palladium,[9] we wanted to de-
termine if such a species might be involved in the ary-
lation reactions outlined herein. To address this issue,
we employed the Buchwald pre-catalyst[10] with NIX-
ANTPHOS ligand, which generates Pd(0) by reduc-
tive elimination (Scheme 6). With 5 mol% of the pre-
catalyst, coupling of 6a with 4-tert-butyl bromoben-
zene resulted in 99% assay yield of the coupling prod-
uct, suggesting that the reaction does not proceed via
the oxidized phosphine ligand.
In summary, we present an umpolung approach to
diarylmethylamine derivatives from both symmetrical
Acknowledgements
P.J.W. acknowledges the NIH (National Institute of General
Medical Sciences NIGMS 104349). B.Y thanks the Scientific
and Technical Research Council of Turkey for a TUBITAK-
2219 fellowship. J. J. thanks CONACyT (MØxico) for fellow-
ships.
References
[1] a) G. X. Ayala, R. Tapia, Eur. J. Neurosci. 2005, 22,
3067–3076; b) M. P. Curran, L. J. Scott, C. M. Perry,
Drugs 2004, 64, 523–561; c) H. Hara, N. Toriu, M. Shi-
mazawa, Cardiovasc. Drug Rev. 2004, 22, 199–214;
d) M. A. Rogawski, W. Lçscher, Nat. Rev. Neurosci.
Scheme 6. Use of Buchwaldꢀs pre-catalyst in the arylation of
azaallyl anions.
Adv. Synth. Catal. 2016, 358, 1910 – 1915
1914
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

COMMUNICATIONS
asc.wiley-vch.de
2004, 5, 553–564; e) H. Mouridsen, M. Gershanovich,
Y. Sun, R. PØrez-Carriòn, C. Boni, A. Monnier, J. Apf-
felstaedt, R. Smith, H. P. Sleeboom, F. Jaenicke, A. Plu-
zanska, M. Dank, D. Becquart, P. P. Bapsy, E. Salmi-
nen, R. Snyder, H. Chaudri-Ross, R. Lang, P. Wyld, A.
Bhatnagar, J. Clin. Oncol. 2003, 21, 2101–2109.
Walsh, Org. Lett. 2014, 16, 4312–4315; c) M. Li, M.
Gonzµlez-Esguevillas, S. Berritt, X, Yang, A. Bellomo,
P. J. Walsh, Angew. Chem. 2016, 128, 2875–2879;
Angew. Chem. Int. Ed. 2016, 55, 2825–2829.
[5] J. A. Fernµndez-Salas, E. Marelli, S. P. Nolan, Chem.
Sci. 2015, 6, 4973–4977.
[2] a) A. Desmarchelier, P. Ortiz, S. R. Harutyunyan,
Chem. Commun. 2015, 51, 703–706; b) C.-C. Chen, B.
Gopula, J.-F. Syu, J.-H. Pan, T.-S. Kuo, P.-Y. Wu, J. P.
Henschke, H.-L. Wu, J. Org. Chem. 2014, 79, 8077–
8085; c) L. Chu, X.-C. Wang, C. E. Moore, A. L. Rhein-
gold, J.-Q. Yu, J. Am. Chem. Soc. 2013, 135, 16344–
16347; d) R. H. Crampton, M. Fox, S. Woodward, Tetra-
hedron: Asymmetry 2013, 24, 599–605; e) N. Dastbara-
vardeh, M. Schnürch, M. D. Mihovilovic, Org. Lett.
2012, 14, 3792–3795; f) A. Hamze, B. TrØguier, J.-D.
Brion, M. Alami, Org. Biomol. Chem. 2011, 9, 6200–
6204; g) J. Clayden, J. Dufuor, D. M. Grainger, M. Hel-
liwell, J. Am. Chem. Soc. 2007, 129, 7488–7489.
[6] a) An example for base-catalyzed isomerization of 2-
azaallyl anions, see; G. Cainelli, D. Giacomini, A.
Trer›, P. P. Boyl, J. Org. Chem. 1996, 61, 5134–5139;
b) for the formation of piperazine derivatives via cyclo-
dimerization of 2-azaallyl anions, see: Z. Chen, J. Wu,
Y. Chen, L. Li, Y. Xia, Y. Li, W. Liu, T. Lei, L. Yang,
D. Gao, W. Li, Organometallics 2012, 31, 6005–6013.
[7] a) B. Yucel, P. J. Walsh, Adv. Synth. Catal. 2014, 356,
3659–3667; b) J. Zhang, A. Bellomo, A. D. Creamer,
S. D. Dreher, P. J. Walsh, J. Am. Chem. Soc. 2012, 134,
13765–13772; c) X. Cao, S.-C. Sha, M. Li, B.-S. Kim, C.
Morgan, R. Huang, X. Yang, P. J. Walsh, Chem. Sci.
2016, 7, 611–618.
[3] a) T. Niwa, H. Yorimitsu, K. Oshima, Org. Lett. 2008,
10, 4689–4691; b) T. Niwa, T. Suehiro, H. Yorimitsu, K.
Oshima, Tetrahedron 2009, 65, 5125–5131; c) Y. Zhu,
S. L. Buchwald, J. Am. Chem. Soc. 2014, 136, 4500–
4503; d) E. C. Burger, J. A. Tunge, J. Am. Chem. Soc.
2006, 128, 10002–10003; e) W. H. Fields, J. J. Chruma,
Org. Lett. 2010, 12, 316–319; f) G. Wolf, E.-U.
Würthwein, Tetrahedron Lett. 1988, 29, 3647–3650;
g) M. J. OꢀDonnell, X. Yang, M. Li, Tetrahedron Lett.
1990, 31, 5135–5138.
[8] The effect of para-methoxy group on the aryl ring in an
azaallyl anion intermediate was observed before, see:
S. Lee, N. A. Beare, J. F. Hartwig, J. Am. Chem. Soc.
2001, 123, 8410–8411.
[9] Y. Ji, R. E. Plata, C. S. Regens, M. Hay, M. Schmidt, T.
Razler, Y. Qiu, P. Geng, Y. Hsiao, T. Rosner, M. D.
Eastgate, D. G. Blackmond, J. Am. Chem. Soc. 2015,
137, 13272–13281.
[10] N. C. Bruno, M. T. Tudge, S. L. Buchwald, Chem. Sci.
2013, 4, 916–920.
[4] a) M. Li, B. Yücel, J. Adrio, A. Bellomo, P. J. Walsh,
Chem. Sci. 2014, 5, 2383–2391; b) M. Li, S. Berritt, P. J.
Adv. Synth. Catal. 2016, 358, 1910 – 1915
1915
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim