Pivotal role of intramolecular catalysis in the selective acetylation of alkyl amines

Article abstract of DOI:10.1002/poc.3842

Preparation of amides by the use of esters as the "acyl donor" is less explored because they are less reactive and usually more steric demanding than conventional acid halides and anhydrides. Here, we report that 3-acetoxy-2-naphthoic acid, an aspirin analogue, can be used as a mild amine acetylating agent in ethanol at 25°C. The reaction is sensitive to steric and polar effects of the attacking amine, and the rate constants can be appropriately fitted by the Pavelich-Taft correlation. Density functional theory calculations used to study all reaction steps indicate that the o-carboxy group plays a pivotal role, guiding the attacking amine and accelerating the reaction. The reaction can be conveniently used for the acylation of a variety of primary and secondary amines.

Full text of DOI:10.1002/poc.3842

Received: 6 December 2017
DOI: 10.1002/poc.3842
Revised: 14 February 2018
Accepted: 15 February 2018
S P E C I A L I S S U E A R T I C L E
Pivotal role of intramolecular catalysis in the selective
acetylation of alkyl amines
Elvis N. Nishida
| Ramon Vitto | Rômulo C.R. Peixoto
| Faruk Nome
| Bruno S. Souza
Department of Chemistry, Federal
University of Santa Catarina,
Abstract
Preparation of amides by the use of esters as the “acyl donor” is less explored
because they are less reactive and usually more steric demanding than conven-
tional acid halides and anhydrides. Here, we report that 3‐acetoxy‐2‐naphthoic
acid, an aspirin analogue, can be used as a mild amine acetylating agent in eth-
anol at 25°C. The reaction is sensitive to steric and polar effects of the attacking
amine, and the rate constants can be appropriately fitted by the Pavelich‐Taft
correlation. Density functional theory calculations used to study all reaction
steps indicate that the o‐carboxy group plays a pivotal role, guiding the
attacking amine and accelerating the reaction. The reaction can be conve-
niently used for the acylation of a variety of primary and secondary amines.
Florianópolis, Santa Catarina, Brazil
Correspondence
Bruno S. Souza, Department of Chemistry,
Federal University of Santa Catarina,
Florianópolis, Santa Catarina 88040‐900,
Brazil.
Email: bruno.souza@ufsc.br
Funding information
INCT‐Catálise; CAPES; CNPq; FAPESC
KEYWORDS
acetylation, amide, chemoselectivity, kinetics, regioselectivity
1 | INTRODUCTION
effective for acetylation of a less hindered amino group
amongst more hindered amino groups.[5] Other examples
Amides are ubiquitous compounds present in the struc-
ture of drugs, biomolecules, and polymers and are usu-
ally prepared using acid anhydrides or acyl halides as
the acylating agent.[1] Coupling agents, such as 1,1′‐
carbonyldiimidazole and N,N'‐dicyclohexylcarbodiimide,
have also been extensively employed, but such methodol-
ogies suffer from the need of a stoichiometric quantity of
the external reagent. A common concern involving acyl-
ation is to avoid the reaction of other functional groups
such as hydroxyl and thiol. Additionally, regioselective
acylation is an important issue when dealing with di‐ or
polyamino compounds.[2] Thus, there is a need to
develop chemoselective and regioselective N‐acylation
methods.[2,3]
Some selective acylating agents have been developed
and usually involve highly reactive agents bearing a trans-
ferable acyl group. For example, N‐methoxydiacetamide
was shown to acetylate primary amines in the presence
of alcohols or secondary amines.[4] Similarly, o‐
substituted N,N‐diacetylanilines were shown to be
include the heterocycle 3‐acyl‐1,3‐thiazolidin‐2‐thione,[6]
potassium acyltrifluoroborates,[7] destabilized amides,[8]
and a magnetic recoverable organogel containing a N‐
hydroxysuccinimide activated ester.[9]
The reaction between amines and unactivated esters
can be used to prepare amides. However, high tempera-
ture or additions of external agents are needed to acceler-
ate the reaction. Morimoto et al[10] showed that
lanthanum (III) triflate catalyze the amidation of esters,
possibly through a Lewis acid carbonyl activation. Cald-
well et al[11] reported that 2,2,2‐trifluorethanol facilitates
the condensation of unactivated ester and amines through
nucleophilic catalysis by the added alcohol.
Although intramolecular catalysis is efficient and fre-
quently observed in enzyme catalysis and in biomimetic
models which include the reactions of amides and carbox-
ylic and phosphate esters,[12–17] it is less explored in syn-
thetic preparations. In this regard, it has been shown
that N‐acetylaminoacids can be conveniently prepared
using aspirin as the acetylating agent of amino acids in
J Phys Org Chem. 2018;e3842.
wileyonlinelibrary.com/journal/poc
Copyright © 2018 John Wiley & Sons, Ltd.
1 of 6
https://doi.org/10.1002/poc.3842

2 of 6
NISHIDA ET AL.
aqueous medium.[18] Here, we demonstrate that the
related 3‐acetoxy‐2‐naphthoic acid (3Ac) can be employed
in the selective acetylation of primary amines in ethanol.
Kinetic evidence as well as theoretical calculations sup-
ports the essential role of the neighboring carboxylate
group in this reaction.
2.4 | Computational calculations
Density functional theory (DFT) calculations were per-
formed at the B3LYP[19–22] and M06[23] with the
6‐31 + G(d,p) basis set using GAUSSIAN 09 package[24]
implemented in Linux operating systems. The default
parameters for convergence were used; convergence on
the density matrix was 10−9 atomic units, the threshold
value for maximum displacement was 0.0018 Å, and the
maximum force was 0.00045 Hartree/Bohr. Stationary
points on the potential energy surface were identified by
frequency calculations at 1 atm and 298.15 K. The polariz-
able continuum model and the solvation model density
(SMD) of Truhlar and coworkers were used in all calcula-
tions.[25] The transition states (TS) were identified by their
single imaginary frequencies, whereas reactants, interme-
diates, and products showed no imaginary frequency. Free
energies of transition states and intermediates were calcu-
lated taking 3Ac and hexylamine, optimized separately, as
the reference. For conversion from 1 atm standard state to
1 mol L−1 standard state, ΔG values were corrected by
subtracting 1.90 kcal mol‐1 to the calculated value accord-
ing to an A + B → C reaction as recommended.[26]
2 | EXPERIMENTAL
2.1 | Materials
All materials were purchased from Sigma Aldrich.
Amines were distilled before use and kept under inert
atmosphere. Ethanol was dried over molecular sieves
and distilled prior to use. 3Ac was prepared as previously
described.[15]
2.2 | Kinetic measurements
3 | RESULTS AND DISCUSSION
Reaction was monitored by UV‐Vis spectrophotometry by
following the appearance of 3‐hydroxy‐2‐naphthoate
(3OH) at 352 nm in closed quartz cuvettes at the desired
temperature accurate to 0.1°C. The reactions were initi-
ated by injecting 15 μL of a stock solution of the substrate
(10−2 mol L−1 in acetonitrile) into 3 mL of the solution of
the amine in ethanol. The reactions were followed for at
least 5 half‐lives, and in all cases pseudo‐first order plots
were obtained. The observed rate constants kobs were cal-
culated by the nonlinear fitting of absorbance versus time
plots with correlation coefficients >0.999. Second‐order
rate constants kN for aminolysis were obtained from lin-
ear plots of kobs against amine concentration.
Reactions of amines with 3Ac, in ethanol, result in the
formation of the corresponding amides and 3‐hydroxy‐2‐
naphthoate (3OH) (Scheme 1) and can be conveniently
monitored by UV‐Vis following the appearance of 3OH
at 352 nm (typical spectral changes shown in Figure S1),
allowing to calculate the observed rate constants as a
function of amine concentration shown in Figure 1.
As can be seen in Figure 1, in the concentration
range studied, the plots are linear, and the data can be
fitted by Equation 1 with values of second‐order rate
constants kN given in Table 1. The intercept ko can be
interpreted as the transesterification reaction, ie, attack
of ethanol on 3Ac in the absence of amine, which is
probably intramolecularly catalyzed by the neighboring
carboxylate anion.
2.3 | Product characterization
The products from the reactions of 3Ac with
dodecylamine, N‐ethylethylenediamine, and ethanol-
amine were identified by GC‐MS after incubating
0.6 mmol of 3Ac with 5 eq. of the respective amine in
2 mL of ethanol overnight at 25°C. Ethanol was removed
in vacuo, and the residue extracted with NaHCO3 5% and
ethyl acetate. Dodecylacetamide was found in the organic
phase, while the more polar N‐[2‐(ethylamino)ethyl]‐acet-
amide and N‐(2‐hydroxyethyl)‐acetamide were found in
the aqueous phase. In all cases, the fragmentograms were
consistent with those expected for the monoacetylated
products and are given in the ESI, Figures S2 to S4.
kobs ¼ k0 þ kN ½amineꢀ
(1)
Because for all amines, values of kN are considerably
larger than ko, it is clear that aminolysis is faster than
SCHEME 1 Reaction of 3Ac with alkylamines affording the
corresponding acetamides

NISHIDA ET AL.
3 of 6
FIGURE 1 Observed rate constant as a
function of amine concentration in
ethanol, 25°C. Data were fitted to
Equation 1. (A) Primary non‐substituted
alkylamines; (B) substituted and secondary
amines
TABLE 1 Second‐order rate constants for reactions of amines
with 3‐acetoxy‐2‐naphthoic acid at 25°C, k0 = 5.66 2.95 × 10‐5 s‐1
It is worth mentioning that tertiary amines catalyze
the degradation of 3Ac and the rate constant for the
ethanolysis of 3Ac showed linear increase as a function
of the concentration of triethylamine, with a rate constant
1.54 × 10−5 M−1 s−1 (data given in Figure S5 and included
in Table 1). This rate constant probably corresponds to the
effect of the tertiary amine acting as a general base
(Scheme 2). The fact that this rate constant is considerably
smaller than those determined for other amines indicates
that the experimental kN values correspond to the
aminolysis reaction and that the general base catalyzed
transesterification path is unimportant.
The sensitivity of the reaction to polar and steric effects
was examined applying the Taft equation (Equation 2),
where the parameters δ and ρ* indicate how sensitive these
reactions are to steric and electronic effects. Values of the
used polar substituent constant σ* and steric substituent
constant Es are given in Table 1 (the methyl group is taken
as reference). Figure 2 shows a plot of the experimental
versus calculated values obtained using Equation 2 and a
multilinear regression routine. The calculated ρ* value is
−1.39 which indicates that electron donation through
sigma bonds is important, while the value for δ is 1.92, indi-
cating that the reaction center is more sterically crowded
than that for hydrolysis of methyl esters.[28] Thus, the reac-
tion is largely influenced by steric and electronic effects,
being fast for linear primary amines.
a
Amine
kN, M−1 s−1
σ*a
Es
Methylamine
5.22 0.37 × 10−2
1.10 0.07 × 10−2
1.30 0.03 × 10−2
1.08 0.05 × 10−2
1.12 0.09 × 10−3
1.01 0.02 × 10−3
1.37 0.03 × 10−4
3.56 0.10 × 10−3
3.53 0.04 × 10−3
5.99 0.14 × 10−3
1.54 0.04 × 10−5
0.98
0.85
0.73
0.70
0.83
0.29
0.36
NA
2.48
2.09
2.18
2.15
1.69
1.10
0.85
NA
Butylamine
Hexylamine
Dodecylamine
Cyclohexylamine
Diethylamine
N‐methylbutylamine
Ethanolamine
Histamine
NA
NA
N‐ethylethylenediamine
Triethylamineb
NA
NA
aValues from Ref. 27.
bSecond‐order constant for the general base catalysis of the ethanolysis
reaction.
ethanolysis. It should be noted that the contribution of
ethanolysis to the overall reaction depends on the concen-
tration of amine used. For example, at 0.05 mol.L‐1
butylamine, ethanolysis contributes to 9% of the overall
reaction, while at 0.1 mol.L‐1 it is 5% only.
As can be seen, the reaction of 3Ac with methylamine
is approximately 5‐fold faster than with other linear pri-
mary amines, with alkyl groups butyl, hexyl, and dodecyl.
Cyclohexylamine, a hindered primary amine, reacts 10‐
fold slower than hexylamine and at about the same rate
than diethylamine. Additionally, amines containing
hydroxyl and amino groups (N‐methylbutylamine, etha-
nolamine, and histamine) are slightly less reactive than
the unsubstituted linear alkylamines.
ꢀ
ꢁ
kN
log
¼ ρ* σ* þ δEs
(2)
kMe
SCHEME 2 Triethylamine catalyzed ethanolysis of 3Ac

4 of 6
NISHIDA ET AL.
regioselectivity of the reaction to the primary center. As
depicted in Scheme 3, only N‐[2‐(ethylamino)ethyl]‐acet-
amide 1 was detected, indicating excellent regioselectivity
to acetylation of primary amines. Similarly, reaction of
3Ac with ethanolamine yields only N‐acetylethanolamine
2, demonstrating good chemoselectivity in the presence of
hydroxyl group.
3.3 | Computational calculations
To gain a deeper understanding of the role of the carbox-
ylate group of 3Ac on the aminolysis reactions, we per-
formed DFT calculations at the SMD/B3LYP/6‐
31 + G(d,p) in ethanol. Using hexylamine as the selected
amine, 3 transition states were found which are given in
Figure 3. A better understating of the reaction can be visu-
alized by following the intrinsic reaction coordinate IRC
as shown in Figure 4, while the most important inter-
atomic distances are given in Table 2. The structure of
reactions shown in Figure 3 corresponds to the structure
found at IRC = −7 amu1/2.bohr in the reverse direction
starting from TS1. In this structure, the NH2 group is
2.73 Å from the ester carbon (N···C1) while the
FIGURE 2 Pavelich‐Taft plot for aminolysis of 3‐acetoxy‐2‐
naphthoic acid in ethanol at 25°C
3.1 | Effect of the o‐carboxylate group
Aspirin is a classic example of rate enhancement due to
the intramolecular assistance by the carboxylate group
acting as a general base.[29] Similarly, the effect of the car-
boxylate group on 3Ac was evaluated by comparing the
aminolysis rate of 3Ac with that of 2‐naphthylacetate
(2AcN) under the same experimental conditions. Values
of kN for hexylamine and N‐methylbutylamine are
1.15 × 10−3 M−1 s−1 and 2.13 × 10−4 M−1 s−1, respectively
(data in Figure S6 of the ESI) indicating that aminolysis of
3Ac is faster than that of 2AcN. Additionally, the observed
rate enhancement is larger for hexylamine, indicating
that the neighboring group participation is more efficient
for the less crowded amines. The fact that both primary
and secondary amines react faster with 3Ac than 2AcN
lends credence to the intramolecular assistance of the car-
boxylate group. These results indicate that the carboxylate
group in 3Ac is directly involved in these reactions, in
accordance with previous observations that hydrolysis of
3Ac anion involves intramolecular general base catalysis
and is faster than that of 2AcN at neutral pH.[15]
‐
CO2 ···H‐NHR interatomic distance (O3···H) is 2.05 Å.
On going from Reactants to TS1, the amine approaches
C1 accompanied by the decrease of O3‐H interatomic dis-
tance. The intermediate species found can be character-
ized as a tetrahedral addition intermediate where the
carbonyl ester group is 1.56 Å from the naphthoate leav-
ing group (C1‐O2) and 1.61 Å (C1‐N) from the attacking
amino group. The intermediate also shows a longer C1‐
O1 bond length (1.27 Å versus 1.22 Å in 3Ac) and pyrami-
dal geometry (O2‐C1‐O1‐CH3 dihedral angle is 109.3° ver-
sus 179.9° in 3Ac). The breakdown of the intermediate
occurs through TS2 and the most important structural
variation in this step is the elongation of C1‐O2 and
decrease of N‐C1 and C1‐O1, as well as a steady decrease
in O3‐H1 interatomic distance. Optimization of the last
point in the forward direction from TS2 leads to a charged
complex in which the charged amide group interacts with
the carboxylate moiety. Finally, the last reaction step
involves proton transfer from the charged amide to the
carboxylate group in TS3 (O3‐H1 = 1.34 Å and N‐
H = 1.19 Å) along with elongation of O2‐C1 interatomic
distance. This step is followed by the intramolecular
3.2 | Selectivity to monoacetylated
products
From the Pavelich‐Taft analysis, it is clear that steric
effects play a crucial role in the aminolysis of 3Ac. Thus,
products from the reaction of N‐ethylethylenediamine
were analyzed by GC‐MS in order to evaluate the
SCHEME 3 Aminolysis of 3Ac with N‐
ethylethylenediamine and ethanolamine

NISHIDA ET AL.
5 of 6
FIGURE 3 Calculated structures of Reactants, TS1, TI, TS2, TS3, and Products optimized at the SMD/B3LYP/6‐31 + G(d,p) level. Reactants
½
½
correspond to the structure found at IRC = −7.5 amu .bohr from TS1 in the reverse direction, while Products is found at IRC = 13.5 amu
.
bohr starting from TS3 in the forward direction
FIGURE 4 (A) Evolution of interatomic distances for reaction of 3Ac with hexylamine calculated by following the IRC starting from TS1
(left) and TS2 (right). Also given are the interatomic distances for the intermediate formed. (B) IRC plot calculated for TS3 in the forward and
reverse directions. All calculations performed at the SMD/B3LYP/6‐31 + G(d,p) level. Numerations according to TS2 in Figure 3
proton transfer from O3 to O2 giving 3OH and N‐
TABLE 2 Selected distances (Å) for optimized structures involved
hexylacetamide as shown in Products in Figure 4B.
The DFT calculations indicate that breakdown of TI
through TS2 is rate determining, and the calculated
in the reaction of 3Ac and hexylamine calculated at the SMD/
B3LYP/6‐31 + G(d,p) level. Numbering according to TS2 in Figure 3
Structure
C1‐O1
C1‐O2
C1‐N
O3‐H
N‐H
activation free energy obtained at the SMD/B3LYP/
6‐31 + G(d,p) is 23.86 kcal.mol−1, slightly higher than
the experimental value of 20.5 kcal.mol−1 determined by
the Eyring plot (graph given in Figure S7 of the ESI).
The Minnesota functional M06[23] was also used to model
this reaction, and similar geometries were obtained, while
the calculated activation free energy for TS2 falls to
20.72 kcal.mol−1, largely improving the agreement with
the experimental value. It should be noted that although
calculations indicate formation of a tetrahedral intermedi-
ate, the free energy differences among all stationary
points were found to be small (within 3 kcal.mol−1). It is
Reactanta
TS1
1.22
1.24
1.27
1.23
1.22
1.25
1.36
1.45
1.56
2.02
2.67
3.16
2.73
1.96
1.61
1.52
1.47
1.35
2.05
1.90
1.74
1.63
1.34
1.53
1.02
1.03
1.05
1.07
1.19
‐
TI
TS2
TS3
Productsb
aCorresponds to the structure found at IRC = ‐7.5 amu1/2.bohr by following
the IRC from TS1 in the reverse direction.
bCorresponds to the structure found at IRC = 13.5 amu1/2.bohr by following
the IRC from TS3 in the forward direction.

6 of 6
NISHIDA ET AL.
[9] Q. M. Kainz, R. Linhardt, P. K. Maity, P. R. Hanson, O. Reiser,
therefore likely that the reaction proceeds in a concerted
rather than in stepwise fashion.
ChemSusChem 2013, 6, 721.
[10] H. Morimoto, R. Fujiwara, Y. Shimizu, K. Morisaki, T.
Ohshima, Org. Lett. 2014, 16, 2018.
4 | CONCLUSIONS
[11] N. Caldwell, C. Jamieson, I. Simpson, A. J. B. Watson, Chem
Commun 2015, 51, 9495.
The use of 3Ac as a mild amine acetylating agent was
examined, and the reactions proved to be sensitive to ste-
ric and polar effects of the attacking nucleophile.
Aminolysis of 3Ac is somewhat faster than the corre-
sponding reactions with 2‐naphthylacetate, which is most
probably due to the intramolecular assistance of the car-
boxylate group. This hypothesis is supported by DFT cal-
culations which indicate that the o‐carboxy group plays
a pivotal role, guiding the attacking amine and accelerat-
ing the reaction. The reaction of 3Ac can be used for the
selective acetylation of primary amino groups in diamines
as well in amino alcohols, and we are in the process of
examining the wider scope of this intramolecularly cata-
lyzed reaction.
[12] B. S. Souza, J. R. Mora, E. H. Wanderlind, R. M. Clementin, J.
C. Gesser, H. D. Fiedler, F. Nome, F. M. Menger, Angew Chem
Int Ed 2017, 56, 5345.
[13] B. S. Souza, S. F. Yunes, M. F. Lima, J. C. Gesser, M. M. Sá, H.
D. Fiedler, F. Nome, Org. Biomol. Chem. 2011, 9, 6163.
[14] E. S. Orth, T. A. S. Brandão, B. S. Souza, J. R. Pliego, B. G. Vaz,
M. N. Eberlin, A. J. Kirby, F. Nome, J. Am. Chem. Soc. 2010,
132, 8513.
[15] B. S. Souza, F. Nome, J Org Chem 2010, 75, 7186.
[16] A. J. Kirby, M. F. Lima, D. da Silva, C. D. Roussev, F. Nome, J
Org Chem 2006, 128, 16944.
[17] M. S. Pereira, B. Murta, T. C. F. Oliveira, A. M. Manfredi, F.
Nome, A. C. Hengge, T. A. S. Brandão, J Org Chem 2016, 81,
8663.
[18] E. S. Orth, B. S. Souza, M. Medeiros, N. B. Caon, A. J. Kirby, F.
Nome, J. Phys. Org. Chem. 2012, 25, 939.
ACKNOWLEDGEMENTS
[19] A. D. Becke, Phys Rev A 1988, 38, 3098.
We are grateful to INCT‐Catálise, FAPESC, CNPq, and
CAPES for financial support.
[20] C. W. Lee, W. Yang, R. G. Parr, Phys Rev B 1988, 37, 785.
[21] A. D. Becke, J Chem Phys 1993, 98, 5648.
[22] P. J. Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J
Phys Chem 1994, 98, 11623.
ORCID
[23] Y. Zhao, D. G. Truhlar, Theor Chem Account 2008, 120, 215.
Elvis N. Nishida http://orcid.org/0000-0002-1482-4006
Rômulo C.R. Peixoto http://orcid.org/0000-0001-6319-9668
Faruk Nome http://orcid.org/0000-0001-8864-6807
Bruno S. Souza http://orcid.org/0000-0001-7030-9271
[24] M. J. Frisch, Gaussian 09, Revision A.02, Gaussian, Inc, Walling-
ford CT 2009 (Supporting Information).
[25] A. V. Marenich, C. J. Cramer, D. G. Truhlar, J Phys Chem B
2009, 113, 6378.
[26] T. Petrova, S. Okovytyy, L. Gorb, J. Leszczynski, J Phys Chem A
2008, 112, 5224.
REFERENCES
[27] A. Williams, Free Energies Relationships in Organic and Bio‐
Organic Chemistry, RSC, Cambridge 2003.
[1] S. Aerry, A. Kumar, A. Saxena, A. De, S. Mozumdar, Green
Chem. Lett. Rev. 2013, 6, 183.
[28] W. A. Pavelich, R. W. Taft Jr., J. Am. Chem. Soc. 1957, 79, 4935.
[29] A. R. Fersht, A. J. Kirby, J. Am. Chem. Soc. 1967, 89, 4857.
[2] I. H. Chung, K. S. Cha, J. H. Seo, J. H. Kim, B. Y. Chung, C. S.
Kim, Heterocycles 2000, 53, 529.
[3] S. M. Coman, M. Florea, V. I. Parvulescu, V. David, A.
Medvedovici, D. De Vos, P. A. Jacobs, G. Poncelet, P. Grange,
J. Catal. 2007, 249, 359.
SUPPORTING INFORMATION
Additional Supporting Information may be found online
in the supporting information tab for this article.
[4] Y. Kikugawa, K. Mitsui, T. Sakamoto, M. Kawase, H. Tamiya,
Tetrahedron Lett. 1990, 31, 243.
[5] Y. Murakami, K. Kondo, K. Miki, Y. Akiyama, T. Watanabe, Y.
Yokoyama, Tetrahedron Lett. 1997, 38, 3751.
How to cite this article: Nishida EN, Vitto R,
Peixoto RCR, Nome F, Souza BS. Pivotal role of
intramolecular catalysis in the selective acetylation
of alkyl amines. J Phys Org Chem. 2018;e3842.
https://doi.org/10.1002/poc.3842
[6] Y. Nagao, K. Seno, K. Kawabata, T. Miyasaka, S. Takao, E.
Fujita, Chem. Pharm. Bull. 1984, 32, 2687.
[7] A. O. Gálvez, C. P. Schaack, H. Noda, J. W. Bode, J. Am. Chem.
Soc. 2017, 139, 1826.
[8] Y. Liu, S. Shi, M. Achtenhagen, R. Liu, M. Szostak, Org. Lett.
2017, 19, 1614.