Native amine-directed site-selective C(sp3)-H arylation of primary aliphatic amines with aryl iodides

Article abstract of DOI:10.1016/j.cclet.2019.10.034

Direct C(sp³)-H functionalization of N-unprotected aliphatic amines represents one of the most efficient and straightforward strategies for amine synthesis. Despite some recent progress in this field, the NH₂-directed γ-C(sp³)-H arylation of primary aliphatic amines except α-amino esters remained an unmet challenge. In this report, we established a simple and efficient method for site-selective C(sp³)-H arylation of primary aliphatic amines by aryl iodides. In the presence of only 5 mol% Pd(OAc)₂, a wide range of aliphatic amines including O-benzyl and O-silyl amino alcohols were arylated at γ- or δ-positions by aryl iodides containing a broad scope of functional groups. The synthetic application of this method had also been demonstrated by large-scale synthesis, the synthesis of a fingolimod analogue, and the conjugation with natural D-menthol and fluorescent 1,8-naphthalimide.

Full text of DOI:10.1016/j.cclet.2019.10.034

Journal Pre-proof
Native amine-directed site-selective C(sp3)-H arylation of primary
aliphatic amines with aryl iodides
Pranab K. Pramanick, Zhibing Zhou, Zhenlin Hou, Yufei Ao, Bo Yao
PII:
S1001-8417(19)30638-2
https://doi.org/10.1016/j.cclet.2019.10.034
CCLET 5298
DOI:
Reference:
To appear in:
Chinese Chemical Letters
Received Date:
Revised Date:
Accepted Date:
11 September 2019
23 October 2019
28 October 2019
Please cite this article as: { doi: https://doi.org/
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
© 2019 Published by Elsevier.

Chinese
Chemical
Letters
journal homepage: www.elsevier.com
Communication
Native amine-directed site-selective C(sp3)-H arylation of primary aliphatic amines
with aryl iodides
Pranab K. Pramanick a,§, Zhibing Zhou a,§, Zhenlin Hou a, Yufei Aob, Bo Yao*,a,b
a MOE Key Laboratory of Cluster Science, Beijing Key Laboratory of Photoelectronic-Electrophotonic Conversion Materials, School of Chemistry and Chemical
Engineering, Beijing Institute of Technology, Beijing 102488, China
b Beijing National Laboratory for Molecular Sciences, CAS Key Laboratory of Molecular Recognition and Function, Institute of Chemistry, Chinese Academy of
Sciences, Beijing 100190, China
 Corresponding author.
E-mail address: yaobo@bit.edu.cn
§ The authors contributed equally to the work.
Graphical abstract
The first NH2-directed γ-C(sp3)-H arylation of primary aliphatic amines with aryl iodides was developed.
ARTICLE INFO
Article history:
Received 11 September 2019
Received in revised form 23 October 2019
Accepted 24 October 2019
Available online
ABSTRACT
Direct C(sp3)-H functionalization of N-unprotected aliphatic amines represents one of the most efficient and straightforward strategies for amine synthesis.
Despite some recent progress in this field, the NH2-directed γ-C(sp3)-H arylation of primary aliphatic amines except α-amino esters remained an unmet
challenge. In this report, we established a simple and efficient method for site-selective C(sp3)-H arylation of primary aliphatic amines by aryl iodides. In the
presence of only 5 mol% Pd(OAc)2, a wide range of aliphatic amines including O-benzyl and O-silyl amino alcohols were arylated at γ- or δ-positions by aryl
iodides containing a broad scope of functional groups. The synthetic application of this method had also been demonstrated by large-scale synthesis, the
synthesis of a fingolimod analogue, and the conjugation with natural D-menthol and fluorescent 1,8-naphthalimide.
Keywords:
Primary aliphatic amine,
Amino ether

C(sp3)-H arylation,
Palladium catalysis,
Free amino group
Aliphatic amines are an important class of structural motifs ubiquitously found in pharmaceuticals and agrochemicals [1]. Among
various approaches for amine synthesis [2], direct C(sp3)-H functionalization of readily available amines provides one of the most
efficient and straightforward strategies for efficient construction of complex amine targets. However, the existing methods on remote
C(sp3)-H functionalization [3-6] often require pre-installation of various protecting groups or directing auxiliaries to improve the
stability and reactivity, and the removal of these groups after functionalization is also needed to release the functionalized amines.
Although being atom- and step-economical, native amine-directed C(sp3)-H functionzalization of aliphatic amines has been far less
explored [7]. Successful examples were mainly restricted to β- and γ-C(sp3)-H functionalization of sterically bulky secondary amines
[8]. Thus far, NH2-directed remote C(sp3)-H functionalization of primary aliphatic amines has been a long-standing challenge in the
past decade. The main reason for this limitation might be that primary amines have a higher propensity to forming less active bis-
amine/metal complexes with metal catalysts than secondary and tertiary amines [9]. To overcome this limitation, Shi adopted a
protonation strategy [10] by using AcOH as the solvent and realized the first palladium-catalyzed γ-C(sp3)-H acetoxylation of primary
aliphatic amines with PhI(OAc)2 [11].
Scheme 1. Remote C(sp3)-H arylation of primary aliphatic amines.
Despite the pioneering work by Shi, efforts on NH2-directed remote C(sp3)-H arylation of primary aliphatic amines met with
unsatisfactory results. According to the reports on transient ligand enabled γ- and δ-C(sp3)-H arylation of primary aliphatic amines by
Dong [12], Yu [13], Ge [14] and others [15-20], transient ligands were very crucial for the high yield of these reactions (eq 1, Scheme
1). Therefore, the challenge in the NH2-directed remote C(sp3)-H arylation of aliphatic amines was not conquered until two recent
reports. One was the Pd-catalyzed γ-C(sp3)-H arylation of α,α-dialkyl-α-amino esters with diaryliodonium triflates developed by our
group [21] (eq 2, Scheme 1). However, further applying this protocol to the C-H arylation of other primary aliphatic amines without α-
CO2R (for example, tert-amylamine) led to poor yield (< 40%). In addition, the harsh conditions in the preparation of diaryliodonium
triflates also limited the broad application of this protocol. Another report was the Pd-catalyzed o-nitrobenzoic acid-promoted δ-C(sp3)-
H and C(sp2)-H arylation with aryl iodides developed by Bannister and his co-workers [22] (eq 3, Scheme 1). Although two examples
on the γ-C(sp3)-H arylation were also provided, the yield was very low (38% and 48%). Therefore, development of new protocols for
efficient γ-C(sp3)-H arylation of primary aliphatic amines with easily available aryl iodides is very desirable. In this context and in
continuation with our previous work, we developed and herein reported a simple but efficient method for palladium-catalyzed site-
selective C(sp3)-H arylation of primary aliphatic amines with aryl iodides (eq 4, Scheme 1). A series of α-branched primary aliphatic
amines including β-/γ-amino ethers and silylethers were arylated efficiently at the γ- or δ-positions [23] by a wide range of aryl iodides.
And the synthetic application of the new method was exemplified by the synthesis of fingolimod analogue, and the conjugation with
natural D-menthol and fluorescent 1,8-naphthalimide.
The study was commenced by conditions optimization of the reaction between tert-amylamine 1a and para-iodoanisole 2a. Initially,
the treatment of 1a with 2.0 equiv. 2a, 5 mol% Pd(OAc)2, and 1.5 equiv. AgTFA in AcOH as the solvent at 120 ºC for 18 h delivered
the desired γ-arylation product 3a in 22% yield (Table 1, entry 1). However, the reaction in pure HFIP also gave the product in 17%
yield (entry 2). Then, acid effect was investigated by performing the reaction in HFIP using 1.0 equiv. acid as the additive (entry 3).
Among a variety of protic acids, AcOH showed strongly positive effect by improving the yield to 42%. Inspired by our previous
discovery in the γ-arylation of α-amino esters by diaryliodonium triflates, we envisioned that co-solvent systems might also be

beneficial to the reaction with aryl iodides. Among a variety of co-solvent systems tested, the combination of HFIP/AcOH in a ratio of
85/15 afforded the optimal result (entry 6). Either changing the ratios of HFIP/AcOH (entries 4-7) or replacing HFIP by other solvents
such as tBuOH, DMF, CH3CN, dichloroethane (DCE) and THF (entries 8-12), led to reduced yield. In order to further increase the
yield, a lot of silver and copper salts were tested as the additive to replace AgTFA (Table S3 in Supporting information). Silver salts
with basic counteranions were all effective, among which 0.75 equiv. Ag2O gave the best result (entry 19). But the use of other silver
salts having neutral counteranions resulted in much lower yield (entries 21 and 22), while all the tested copper salts showed no activity
(Table S3 in Supporting information). Finally, the optimal conditions were determined as: 1a (0.2 mmol), 2a (0.6 mmol), 5 mol%
Pd(OAc)2, 0.75 equiv. Ag2O, HFIP/AcOH (85/15, 1 mL), 120 ºC, air for 18 h. The reaction under these conditions gave 3a as the single
product in 67% yield without chromatographic purification (entry 19).
Table 1
Conditions optimization for γ-C(sp3)-H arylation of tert-amylamine (1a) with 4-iodoanisole (2a). a
2a
(equiv.) (equiv.)
AgTFA (1.5)
Additive
Yield
(%)b
22
Entry
Solvent (ratio)
1
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
1.0
3.0
5.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
HOAc
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
AgTFA (1.5)
AgTFA (1.5)
AgTFA (1.5)
AgTFA (1.5)
AgTFA (1.5)
AgTFA (1.5)
AgTFA (1.5)
AgTFA (1.5)
AgTFA (1.5)
AgTFA (1.5)
AgTFA (1.5)
AgTFA (1.5)
AgTFA (1.5)
AgTFA (1.5)
AgOAc (1.5)
Ag2CO3 (1.5)
Ag2O (1.5)
HFIP
17
0-42
21
43
57
40
16
25
trace
37
21
52
61
48
55
52
43
67
HFIP (1.0 equiv. acid)
HFIP/HOAc (10/90)
HFIP/HOAc (50/50)
HFIP/HOAc (85/15)
HFIP/HOAc (95/5)
tBuOH/AcOH (85/15)
DMF/AcOH (85/15)
CH3CN/AcOH (85/15)
DCE/AcOH (85/15)
THF/AcOH (85/15)
HFIP/AcOH (85/15)
HFIP/AcOH (85/15)
HFIP/AcOH (85/15)
HFIP/AcOH (85/15)
HFIP/AcOH (85/15)
HFIP/AcOH (85/15)
HFIP/AcOH (85/15)
HFIP/AcOH (85/15)
HFIP/AcOH (85/15)
HFIP/AcOH (85/15)
Ag2O (0.75)
Ag2O (0.5)
AgOTf (1.5)
AgNO3 (1.5)
54
21
trace
a
Conditions: tert-amylamine 1a (0.2 mmol), 4-iodoanisole 2a (1.0-5.0 equiv.), Pd(OAc)2 (5 mol%, 0.01 mol), additive (0.15-0.3 mmol),
solvent (1 mL), 120 ºC, air, 18 h.
b Yield obtained by workup without chromatographic purification.
With the optimal conditions in hand, the scope of aryl iodides in the γ-arylation of tert-amylamine 1a was first investigated (Scheme
2). A large number of aryl iodides having substituents at the ortho-, meta-, and para-positions were tested. The reaction of ortho- and
meta-substituted aryl iodides required longer reaction time (36 h) and/or a higher catalyst loading (10 mol%) to reach similar
conversion than para-substituted counterparts (3f vs. 3r; 3e vs. 3q). Meanwhile, both electron-rich (OMe, 3a) and electron-poor aryl
groups (F, 3f; CO2Me, 3l) were successfully installed to the γ-position of tert-amylamine in similar yield. The result indicated that the
reaction was sensitive to the steric effect, but not to the electronic property of aryl groups. Furthermore, a wide range of functional
groups such as CF3, F, Cl, Br, OTs, Ac, CONHPh, CO2Me, NO2 and CN were well tolerated to the conditions. The compatibility of
these functional groups not only provided useful handles for further transformations of the products, but also gave opportunities for late-
stage modification of complex amines. It is worth noting that there were no N-arylation and other C-H arylation products detected in all
these reactions. Therefore, chromatographic purification is not required. Besides aryl groups, a Ts-protected indole ring (3u) was also
connected to the amine successfully by this transformation, which might find applications in the synthesis of indole-embedded natural
products in the future.

Scheme 2. Scope of aryl iodides for the γ-C(sp3)-H arylation of tert-amylamine (1a). Conditions: 1a (0.2 mmol), 2 (0.6 mmol), Pd(OAc)2 (5
mol%, 0.01 mol), Ag2O (0.75 equiv., 0.15 mmol), HFIP/AcOH (1 mL, 85/15, v/v), 120 ºC, air, 18 h.
After testing the scope of aryl iodides, we continued to examine the scope of primary aliphatic amines (Scheme 3). Firstly, α-tertiary
carbons were found to be indispensible for the transformation because amines having α-H such as 1p and 1q underwent facile oxidation
followed by hydrolysis to afford aldehydes. A series of α,α-dialkylpropylamines were arylated at the 1° γ-carbons in moderate to good
yields (4b-4f, 46%-74%). In these reactions, none of the 2° γ-C-H, benzylic C-H, 1° β-C-H and 1° δ-C-H bonds were functionalized.
The excellent regioselectivity ruled out that the reaction followed a Shilov-type oxidation pathway [3], but supported that the amino
group acted as the directing group for C-H functionalization. Interestingly, when there was no 1° γ-C-H bond in the substrate, 1° δ-C-H
arylation could take place under the same conditions. The reaction of 2,4,4-trimethylpentan-2-amine with PhI provided the δ-
monoarylation product 4g as the single product in 44% yield. Secondly, O-benzyl and O-silyl ethers were well compatible to the
reaction conditions. The reaction of O-benzyl amino ethers afforded the γ-arylation products (4h and 4i) in moderate yields with aryl
C(sp2)-H bonds and α-C(sp3)-H bonds of oxygen unaffected. Probably due to the steric effect of silyl groups, O-TBS amino ethers
displayed higher reactivity than O-benzyl ethers. Two α-ethyl-α-alkyl substituted amino silylethers underwent γ-arylation to give the
products 4j and 4k in much higher yields than 4i (63% and 68% vs. 44%). Also, in the reaction of α-ethyl-α-isobutyl substituted amino
silylether 1m, both of γ-monoarylation and δ-monoarylation products (4m) were formed in 58% total yield and 3:1 ratio. With a higher
catalyst loading (10 mol%), the α-methyl-α-isobutyl amino TIPS-ether 1n was even arylated at the δ-positions to yield the mono- and
di-arylation products in 60% total yield and 2.3:1 ratio. The steric bulk of silyl groups also resulted in excellent diastereoselectivity. In
our previous report, the γ-arylation of α-methyl-valine ester gave the product in very poor diastereoselectivity (dr 1.2:1) [21].
Surprisingly, in the reaction of valine-derived silylether 1l, only one diastereoisomer 4l was observed by1H NMR analysis. Thirdly, the
proximal COOEt might be detrimental to the reaction as the reaction of α-amino ester 1o with PhI afforded the arylation product 4o in
just 16% yield. In addition, the conditions were also not suitable for secondary amines. The reaction of N-pentyl-tert-amylamine 1r did
not give any C-H arylation products, but left the starting material mainly recovered.

Scheme 3. Scope of primary aliphatic amines. Conditions: 1 (0.2 mmol), 2 (3.0 equiv.), Pd(OAc)2 (5 mol%), Ag2O (0.75 equiv.), HFIP/AcOH
(1 mL, 85/15, v/v), 120 ºC, air, 18 h.
Finally, the synthetic utility of the new protocol was explored. First, the γ-arylation reaction of 1a was conducted at 5 or 10 times the
scale for the optimal conditions and delivered the product 3a in 60% and 48% yield (Scheme 4a). Furthermore, using the γ-arylation
reaction as the key step, a fingolimod analogue 6 was also prepared from the commercial 2-amino-2-ethylpropane-1,3-diol in 41% yield
via a three-step synthesis (Scheme 4b). In addition, with ester and imide groups as the linkers, natural D-menthol and fluorescent 1,8-
naphthalimide were connected to the aryl moiety of PhI. Applying the conditions to the reaction of tert-amylamine with the two
modified aryl iodides resulted in the formation of two conjugation products 3v and 3w in 39% and 45% yield respectively (Scheme 4c).

Scheme 4. Application of the method.
In conclusion, we have developed a simple but efficient protocol for site-selective arylation of primary aliphatic amines by aryl
iodides. With only 5 mol% Pd(OAc)2, a wide range of primary amines including β-/γ-amino ethers were efficiently arylated at the γ- or
δ-C-H bonds by a large number of aryl iodides with diverse functional groups. The synthetic utility of this method was also explored by
the conjugation of tert-amylamine with natural D-menthol and fluorescent 1,8-naphthalimide, and the synthesis of a fingolimod
analogue.
Acknowledgments
We thank the National Natural Science Foundation of China (No. 21502006), Beijing National Laboratory for Molecular Sciences
and Beijing Institute of Technology (BIT) for financial support. The Analysis and Testing Centre of Beijing Institute of Technology is
highly appreciated for their instrument support.

References
[1] (a) S. Jeanmart, A.J.F. Edmunds, C. Lamberth, M. Pouliot, Bioorg. Med. Chem. 24 (2016) 317-341.
(b) P. Das, M.D. Delost, M.H. Qureshi, D.T. Smith, J.T. Njardarson, J. Med. Chem. 62 (2019) 4265-4311.
(c) F.Z. Dörwald, Aliphatic amines, in: F.Z. Dörwald (Ed.), Lead Optimization for Medicinal Chemists: Pharmacokinetic Properties of Functional
Groups and Organic Compounds, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2012, pp. 107-117.
[2] (a) J.F. Knifton, Catal. Today. 36 (1997) 305-310.
(b) Y. Park, Y. Kim, S. Chang, Chem. Rev. 117 (2017) 9247-9301.
(c) P. Kalck, M. Urrutigoity, Chem. Rev. 118 (2018) 3833-3861.
[3] (a) B.D. Dangel, J.A. Johnson, D. Sames, J. Am. Chem. Soc. 123 (2001) 8149-8150.
(b) M. Lee, M.S. Sanford, J. Am. Chem. Soc. 137 (2015) 12796-12799.
(c) C.T. Mbofana, E. Chong, J. Lawniczak, M.S. Sanford, Org. Lett. 18 (2016) 4258-4261.
(d) M. Lee, M.S. Sanford, Org. Lett. 19 (2017) 572-575.
[4] (a) C. Martínez, K. Muñiz, Angew. Chem. Int. Ed. 54 (2015) 8287-8291.
(b) E.A. Wappes, S.C. Fosu, T.C. Chopko, D.A. Nagib, Angew. Chem. Int. Ed. 55 (2016) 9974-9978.
(c) G.J. Choi, Q. Zhu, D.C. Miller, C.J. Gu, R.R. Knowles, Nature 539 (2016) 268.
(d) J.C.K. Chu, T. Rovis, Nature 539 (2016) 272.
(e) M.Y. Yang, B. Su, Y. Wang, et al., Nat. Commun. 5 (2014) 5707.
(f) Z. Li, Q. Wang, J. Zhu, Angew. Chem. Int. Ed. 57 (2018) 13288-13292.
[5] S. Munnuri, A.M. Adebesin, M.P. Paudyal, et al., J. Am. Chem. Soc. 139 (2017) 18288-18294.
[6] (a) G. Rouquet, N. Chatani, Angew. Chem. Int. Ed. 52 (2013) 11726-11743.
(b) O. Daugulis, J. Roane, L.D. Tran, Acc. Chem. Res. 48 (2015) 1053-1064.
(c) B. Liu, Y.Q. Han, F. Hu, B.F. Shi, Palladium-catalyzed directed arylation of unactivated C(sp3)-H bonds, in: A.R. Kapdi, D. Maiti (Eds.),
Strategies for Palladium-Catalyzed Non-Directed and Directed C-H Bond Functionalization, Elsevier, 2017, pp. 167-203. (d) C. He, W. G.
Whitehurst, M. J. Gaunt, Chem. 5 (2019) 1031-1058.
[7] (a) Z. Liang, L. Ju, Y. Xie, L. Huang, Y. Zhang, Chem. Eur. J. 18 (2012) 15816-15821.
(b) C. Suzuki, K. Hirano, T. Satoh, M. Miura, Org. Lett. 15 (2013) 3990-3993.
(c) A. Boelke, L.D. Caspers, B.J. Nachtsheim, Org. Lett. 19 (2017) 5344-5347.
(d) K. Orito, A. Horibata, T. Nakamura, et al., J. Am. Chem. Soc. 126 (2004) 14342-14343.
(e) A. Lazareva, O. Daugulis, Org. Lett. 8 (2006) 5211-5213.
(f) B. Haffemayer, M. Gulias, M.J. Gaunt, Chem. Sci. 2 (2011) 312-315.
(g) M. Miura, C.G. Feng, S. Ma, J.Q. Yu, Org. Lett. 15 (2013) 5258-5261.
(h) A. Mancinelli, C. Alamillo, J. Albert, et al., Organometallics 36 (2017) 911-919.
(i) C.H. Zhang, Y.Z. Ding, Y.Z. Gao, S.D. Li, G. Li, Org. Lett. 20 (2018) 2595-2598.
[8] (a) A. McNally, B. Haffemayer, B.S.L. Collins, M.J. Gaunt, Nature 510 (2014) 129-133.
(b) J. Calleja, D. Pla, T.W. Gorman, et al., Nat. Chem. 7 (2015) 1009-1016.
(c) C. He, M.J. Gaunt, Angew. Chem. Int. Ed. 54 (2015) 15840-15844.
(d) D. Willcox, B.G.N. Chappell, K.F. Hogg, et al., Science 354 (2016) 851.
[9] A.D. Ryabov, Chem. Rev. 90 (1990) 403-424.
[10] A.P. Smalley, M.J. Gaunt, J. Am. Chem. Soc. 137 (2015) 10632-10641.
[11] K. Chen, D. Wang, Z.W. Li, et al., Org. Chem. Front. 4 (2017) 2097-2101.
[12] Y. Xu, M.C. Young, C. Wang, D.M. Magness, G. Dong, Angew. Chem. Int. Ed. 55 (2016) 9084-9087.
[13] (a) Y. Wu, Y.Q. Chen, T. Liu, M.D. Eastgate, J.Q. Yu, J. Am. Chem. Soc. 138 (2016) 14554-14557.
(b) Y.Q. Chen, Z. Wang, Y. Wu, et al., J. Am. Chem. Soc. 140 (2018) 17884-17894.
[14] Y. Liu, H. Ge, Nat. Chem. 9 (2017) 26-32.
[15] A. Yada, W. Liao, Y. Sato, M. Murakami, Angew. Chem. Int. Ed. 56 (2017) 1073-1076.
[16] M. Kapoor, D. Liu, M.C. Young, J. Am. Chem. Soc. 140 (2018) 6818-6822.
[17] H. Lin, C. Wang, T.D. Bannister, T.M. Kamenecka, Chem. -Eur. J. 24 (2018) 9535-9541.
[18] S. St John-Campbell, A.K. Ou, J.A. Bull, Chem. -Eur. J. 24 (2018) 17838-17843.
[19] J.T. Ye, I. Kalvet, F. Schoenebeck, T. Rovis, Nat. Chem. 10 (2018) 1037-1041.
[20] X.X. Hu, J.B. Liu, L.L. Wang, et al., Org. Chem. Front. 5 (2018) 1670-1678.
[21] P.K. Pramanick, Z. Zhou, Z.L. Hou, B. Yao, J. Org. Chem. 84 (2019) 5684-5694.
[22] H. Lin, X. Pan, A.L. Barsamian, T.M. Kamenecka, T.D. Bannister, ACS Catal. 9 (2019) 4887-4891.
[23] (a) S. Guin, P. Dolui, X. Zhang, et al., Angew. Chem. Int. Ed. 58 (2019) 5633-5638.
(b) B.B. Zhan, Y. Li, J.W. Xu, et al., Angew. Chem. Int. Ed. 57 (2018) 5858-5862.
(c) J.W. Xu, Z.Z. Zhang, W.H. Rao, B.F. Shi, J. Am. Chem. Soc. 138 (2016) 10750-10753.