Electrochemical, Iodine-Mediated α-CH Amination of Ketones by Umpolung of Silyl Enol Ethers

Article abstract of DOI:10.1021/acs.orglett.0c02068

The electrochemical, oxidative Umpolung reaction of silyl enol ethers utilizing simple iodide salts for the synthesis of α-amino ketones is described. The products were isolated in excellent yields of up to 100percent, and various functionalized starting materials were accepted in an undivided electrochemical cell design. Moreover, a sensitivity assessment to ensure an improved reproducibility of the reaction and cyclic voltammetry experiments were performed to postulate a plausible reaction mechanism on their basis.

Full text of DOI:10.1021/acs.orglett.0c02068

pubs.acs.org/OrgLett
Letter
Electrochemical, Iodine-Mediated α‑CH Amination of Ketones by
Umpolung of Silyl Enol Ethers
Julia Strehl and Gerhard Hilt*
Cite This: https://dx.doi.org/10.1021/acs.orglett.0c02068
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: The electrochemical, oxidative Umpolung reaction
of silyl enol ethers utilizing simple iodide salts for the synthesis of
α-amino ketones is described. The products were isolated in
excellent yields of up to 100%, and various functionalized starting
materials were accepted in an undivided electrochemical cell
design. Moreover, a sensitivity assessment to ensure an improved
reproducibility of the reaction and cyclic voltammetry experiments
were performed to postulate a plausible reaction mechanism on their basis.
α-Amino ketones are a very important structure motif in organic
chemistry because they represent the key structure of many
natural products and of biological and pharmacological active
compounds, such as the psychoactive cathinone and amfepra-
mone or the antiplatelent prasugrel.1 Moreover, they act as
important building blocks for interesting structures like hetero-
cycles or 1,2-amino alcohols, e.g., spisulosine (Figure 1).2
and tBuOOHasoxidant(Scheme1, pathb).9 Zengand Sunwere
able to develop an electrochemical alternative avoiding further
chemical oxidants (Scheme 1, path c).10 However, very long
reactiontimesandanexcessofelectriccurrent(14F·mol−1)were
needed, and the reaction seems to be limited to acetophenone
derivatives which could be converted to the desired products in
good yields of up to 75%. A very innovative approach was
described by Wirth, utilizing the Umpolung of silyl enol ethers
with stoichiometric amounts of PIDA (PhI(OAc)2) and the
intramolecular reaction of a preinstalled nucleophile (path d).11
Scheme 1. Previously Reported Procedures for the Synthesis
of α-Amino Ketones
Figure 1. Selected examples of biologically active α-amino ketones and
α-amino alcohols.
Due to the significance of this substance class the development
and improvement of their synthesis has been of high interest
since the 19th century.3 One method for the synthesis of these
compounds is the oxidative amination reaction.4 For example, a
Cu(II)-catalyzedaminationofketones, esters, and aldehydeswas
describedbyMacMillan.4a Commonly,theformationofα-amino
ketones can be achieved either by α-amination of carbonyl
derivatives, e.g., silyl enol ethers, with electrophilic N-
compounds, like α-haloamines, hydroxylamines, or nitro-
soformates,5 or by reaction of carbonyl groups with reversed
polarity with nucleophilic N-compounds.3a,6−11 The Umpolung
reaction can be performed with, e.g., tosylazoalkanes, nitro-
soalkenes, or enamine derivatives.7 Guo developed a NH4I-
catalyzed method for the reaction of acetophenone derivatives
with secondary amines using sodium percarbonate as external
oxidant (Scheme 1, path a).8 This conversion can be
accomplished as well by applying N-iodosuccinimide (NIS)
Received: June 23, 2020
https://dx.doi.org/10.1021/acs.orglett.0c02068
© XXXX American Chemical Society
Org. Lett. XXXX, XXX, XXX−XXX
A

Organic Letters
pubs.acs.org/OrgLett
Letter
The complex prefunctionalization of the starting materials,
stoichiometric or even excessive usage of co-oxidants, limited
substrate scopes and/or the addition of toxic heavy metal
catalysts are still drawbacks of these reactions. For this reason, we
envisioned an electrochemical oxidative method, which enables
the substoichiometric use of nontoxic and simple iodide salts as
Umpolung reagent for silyl enol ethers. The use of silyl enol
ethers is advantageous compared to ketones because they cannot
condense with amines forming enamines. We intended to
simplify the elaborate prefunctionalization of the starting
materials as demonstrated by Wirth to regain flexibility in the
approach by adding the N-nucleophile separately to the reaction
mixture and performing an intermolecular reaction.
Accordingly, the synthesis of α-amino ketone 3a by the
reaction of silyl enol ether 1a with dibenzylamine 2a under
electrochemical conditions was examined. We started our
investigation adopting the conditions described by Zeng and
Sun (NH4I, 0.5 equiv), LiClO4 (0.1 M), amine (3.0 equiv), and
anhydrous CH3CN as solvent)10 and were able to isolate the
desired product 3a in 59% yield. During a screening of different
iodide salts, the yield could be raised to 78% by using sodium
iodide (Table 1, entry 1). Based on these conditions, we started
thescreeningprogramfortheoptimizationofthereaction(Table
1).
experiments. The variation of the reaction temperature (0 °C, 40
°C), electrolyte (nBu4NBr, nBu4NBF4, none), or of the solvent
(dichloromethane, methanol, THF) led to no further improve-
ments. By varying the amount of the amine (1.1−2.5 equiv) 3a
was isolated in quantitative yield when 2.0 equiv of amine 2a was
applied. Further variations of the amount of NaI (0.1−0.3 equiv)
and the electrode material (glassy carbon, graphite) downgraded
the yield for which reason the former determined reaction
conditions were chosen (for more information, see the SI). At
last, a comparative experiment was done by using equimolar
amounts of iodine instead of electricity. The product 3a could be
isolated in 64% yield (Table 1, entry 12), which exemplified the
great potential of this reaction.
Despite the many advantages of electroorganic synthesis (e.g.,
less waste, the use of renewable energy and electrons as
reagent)12 its application is underrepresented, which might be
rooted in reproducibility issues. To avoid this problem and assess
systematic and random errors, a sensitivity test was performed
which is based on the assumption that already small variations of
the reaction parameters, that might occur accidentally by
reproduction of the experiment by other researchers, could
have an impact on the reproducibility.13 The sensitivity of the
reaction to fluctuations of typical reaction parameters (e.g.,
concentration or scale) but also a discrepancy of the setup was
investigated. Especially the setup is a major factor in electro-
chemical reactions because most working groups use their
individually manufactured equipment. The amination reaction
was insensitive against many parameters but showed a medium
sensitivity toward a lower concentration, higher temperatures
andincreasedcurrents(Figure 2). The reductionofthe electrode
distance (−30%) led to the bisection of the yield. An undersized
electrode distance seems to lead to a “chemical shortcut”.
Meaning, that the oxidized iodine species is directly reduced at
the cathode without undergoing the desired reaction. So special
attention should be paid to this parameter.
Table 1. Selected Examples of Variation of the Reaction
Conditions*
*
Reactions were performed in an undivided cell on a 0.5 mmol scale.
a
Conditions: 0.5 equiv of NaI, 3.0 equiv of amine 2a, LiClO4 (0.1 M),
Pt electrodes (1.0 × 3.0 cm2), anhydrous CH3CN, rt, 10 mA, 4 F·
Figure 2. Sensitivity assessment of the electrochemical amination (see
the SI for more information).
b
mol−1. The yield was determined by GC analysis of the unpurified
c
reaction mixture with mesitylene as internal standard. This change
d
was kept for entries 5−11. This change was kept for entry 11.
e
After the optimization of the reaction conditions and the
sensitivity assessment, the starting materials were varied to
examine scope and limitations of the electrochemical amination
reaction (Scheme 2). First, the tolerance of the reaction against
functional groups should be tested by applying different
substituted dibenzylamines. Fortunately, electron-deficient as
well as electron-rich arene moieties were tolerated, but the
introduction of electron-deficient groups was associated with
lower yields. While the 4-methoxy derivative 3b was generated
Isolated yield.
When the reaction was performed under inert conditions the
yield of 3a dropped to 2%. Notably, the water amount seemed to
have a profound effect on the yield and was therefore varied
subsequently (0.5−50 equiv. H2O, anhydrous CH3CN). The
addition of 20.0 equiv of H2O increased the yield to 89%;
consequently, this parameter was maintained for the following
B
https://dx.doi.org/10.1021/acs.orglett.0c02068
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
quantitatively, the 4-cyano and 4-bromo derivatives 3c and 3d
were isolated in moderate yields (51−55%). As expected,
compound 3e was isolated with the lowest yield (27%) probably
due to the highly deactivation and redox lability of the nitro
group. Subsequently, the substituents of the nitrogen were
varied. N-Methylbenzylamine as well as the sterically highly
demanding N-benzylcyclohexylamine could be converted to the
products 3h and 3i (90% and 39%). The conversion of the
primary amine N-benzylamine 2j failed. Moreover, amine 2k
(with a substituent in 1-position) was converted successfully and
the product was isolated in a good yield of 91%. Also, cyclic
amines were tested as well: While the tetrahydroisoquinoline
product 3l was isolated in 82% yield, the application of
morpholine2nonlyledto3nin38%yieldandthetransformation
with piperidine for the synthesis of 3m failed completely for no
obvious reason. The same was observed for N-methylbenzamide
(2o), which was anticipated due to the lower nucleophilicity of
amides. Nevertheless, di-n-butylamine and diallylamine yielded
the products 3p and 3q in 53−56%, respectively. After variation
of the amines, different silyl enol ethers were reacted. The
conversion of sterically more demanding alkyl-substituted silyl
enol ethers to products 3r and 3s succeeded in 71−73% yield,
respectively. But the reaction failed with cyclic silyl enol ethers
toward the α-amino ketone 3t.14 Fortunately, the generation of
product 3u, where a tertiary carbon atom was generated, worked
excellently(97%yield).Atlast,thediastereomericproduct3v(dr
= 1.4:1, 22%) could be isolated.
in 75% (Table 2, entry 2). The utilization of other functionalized
styrene-type silyl enol ethers led to overoxidation and generation
of the α-amidated products 4y, 4z, and 4aa (entries 3−5). The
reaction was repeated and stopped after a consumption of 2 F·
mol−1 utilizing ((1-mesitylvinyl)oxy)trimethylsilane. Although
the reaction was not complete only the α-ketoamide 4y was
isolated (40% yield). In the literature, the same observation was
made in case of the direct oxidative conversion (tBuOOH/I2 or
electrochemical) of acetophenone derivatives.9,15 By introduc-
tion of a substituent in α-position of the carbonyl group, the
isolation of α-amino ketones was enabled. This approach also
worked well for the electrochemical conversion, since the
reaction of 1-phenyl-1-trimethylsilyloxypropene led to product
3aa which was isolated in 62% yield (Table 2, entry 5). Finally,
the psychoactive drug amfepramone 3ab could be generated as
well although only in moderate yield (Table 2, entry 6).
Table 2. Aromatic Silyl Enol Ethers as Starting Materials
In the same manner, different substituted aromatic silyl enol
ethers could be converted. While no product was observed in the
reaction of α-trimethylsilyloxystyrene with N-methylbenzyl-
amine (Table 2, entry 1), the conversion of ((1-(4-methoxy-
phenyl)vinyl)oxy)trimethylsilane gave the amination product 3x
Scheme 2. Substrate Scope of the Electrochemical Amination
a
of Silyl Enol Ethers
a
Consumption of 2 F·mol−1.
For clarification of the reaction mechanism, electroanalytical
measurements were performed. The cyclic voltammogram (CV)
of NaI showed two reversible oxidation peaks (peak potentials:
+341 and +677 mV vs Ag/AgCl as reference electrode). The first
peak can be assigned to the oxidation of iodide to triiodide and
the second one to the oxidation of triiode to iodine.16 When
another cyclic voltammogram of NaI in the presence of silyl enol
ether 1a was run a great peak enhancement of the second
oxidation peak appeared (Figure 3, left CV). Moreover, the
reduction peaks diminished. This indicates that a reaction of the
silyl enol ether with iodine takes place forming a α-iodoketone
and iodide. The peak enhancement can be explained by the fact,
that the iodide is directly oxidized with triiodide generating new
iodine.
a
Yields refer to isolated, purified products.
C
https://dx.doi.org/10.1021/acs.orglett.0c02068
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
Figure 3. (Left) Cyclic voltammogram of: black, NaI (5 mM); red, NaI (5 mM) and silyl enol ether 1a (2.0 equiv) in CH3CN/LiClO4 (10 mL, 0.1 M).
(Right)Key:black,NaI(5mM);red,NaI(5mM),silylenolether1a(2.0equiv)andamine2a(4.0equiv);blue,NaI(5mM)andamine2a(4.0equiv)in
CH3CN/LiClO4 (10 mL, 0.1 M). Reference electrode: Ag/AgCl (3 M NaCl), scan rate: 200 mV/s.
The CV of NaI and amine 2a showed an enhancement and a
small shift of all peaks, but form and appearance stayed the same
(Figure 3, right CV). In particular, the second oxidation peak is
shifted and the CV looks more reversible. This indicates the
reversible formation of a complex between the iodine and the
amine, which has been described in literature before.17 By
generating this complex, iodide is formed, which can be
reoxidized and leads to the peak enhancement. At last, the cyclic
voltammogram of NaI, amine 2a, and silyl enol ether 1a showed
only a small enhancement of the second oxidation peak, but all
reduction peaks disappeared completely. This indicates the
reaction of silyl enol ether 1a with the cationic iodonium species
and the further SN reaction of the in situ generated compound
with amine 2a. To prove this assumption, 1-iodopropan-2-one
was reacted with dibenzylamine 2a. Fortunately, the product 3a
was formed in 93% yield. On the basis of these results a reaction
mechanism was postulated (Scheme 3).
intermediate B via nucleophilic substitution and product 3a is
generated. The nascent iodide can be reoxidized at the anode,
and the proton is reduced to molecular hydrogen at the cathode.
In conclusion, an electrochemical, NaI-assisted oxidative
Umpolung reaction of silyl enol ethers for the formation of α-
amino ketones is described. The scope and limitations of the
reaction were investigated, and to enable an improved
reproducibility of the amination reaction a sensitivity assessment
was done. Moreover, the synthesis of the psychoactive drug
amfepramone succeeded. Compared to the work of Zeng and
Sun, where only propiophenone derivatives could be converted
to α-amino ketones,10,18 we were able to enlarge the substrate
scope so that different alkyl-substituted amino ketones were
generated in addition to aromatic ones. Moreover, a distinct
reduction of the reaction time from 14 to 4 F/mol was achieved.
Lastly, cyclic voltammetry measurements and control experi-
ments led to the formulation of a plausible reaction mechanism.
ASSOCIATED CONTENT
Scheme 3. Proposed Reaction Mechanism
■
sı
* Supporting Information
The SupportingInformationisavailablefree ofchargeathttps://
pubs.acs.org/doi/10.1021/acs.orglett.0c02068.
Synthesis, sensitivity assessment, analytical data, NMR,
and CV spectra (PDF)
AUTHOR INFORMATION
■
Corresponding Author
Gerhard Hilt − Institut fur Chemie, Universitat Oldenburg, D-
̈
̈
26111 Oldenburg, Germany; orcid.org/0000-0002-5279-
3378; Email: gerhard.hilt@uni-oldenburg.de; Fax: +49 441
798 3329
Author
JuliaStrehl−Institutfu
Oldenburg, Germany
̈
̈
rChemie, UniversitatOldenburg, D-26111
We postulate the anodic oxidation of iodide as the first step of
thereaction. Ononehand, thegeneratediodinecandirectlyreact
with silyl enol ether 2a to form α-iodoketone B. On the other
hand, the iodine can react with amine 2a and the complex Α is
generated. Afterward, the silyl enol ether 1a reacts with A to form
the α-iodoketone B as well. Finally, the amine 2a attacks the
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.orglett.0c02068
Notes
The authors declare no competing financial interest.
D
https://dx.doi.org/10.1021/acs.orglett.0c02068
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
slow formation of the α-iodocyclohexan-1-one was detected. For more
information, see the Supporting Information.
(15) (a) Zhang, X.; Wang, L. Green Chem. 2012, 14, 2141−2145.
(b) Zhang, Z.; Zha, Z.; Wang, Z. Chem. Commun. 2013, 49, 8982−8984.
(16) (a) Bentley, C. L.; Bond, A. M.; Hollenkamp, A. F.; Mahon, P. J.;
Zhang, J. J. Phys. Chem. C 2015, 119, 22392−22403. (b) Iwamoto, R. T.
Anal. Chem. 1959, 31, 955.
(17) (a) Schug, J. C.; Kogan, M. J. J. Magn. Reson. 1973, 11, 406−415.
(b)Hassel, O.Science1970,170,497−502.(c)Schmulbach,C. D.;Hart,
D. M. J. Am. Chem. Soc. 1964, 86, 2347−2351. (d) Yada, H.; Tanaka, J.;
Nagakura, S. Bull. Chem. Soc. Jpn. 1960, 33, 1660−1667.
(18) Except for one example using 2-cyclopropyl-1-phenylethanone as
starting material.
ACKNOWLEDGMENTS
■
We thank the German Science Foundation (DFG) for its
financial support (Hi655 - GRK 2226).
REFERENCES
■
(1) (a) Kalix, P. Pharmacol. Toxicol. 1992, 70, 77−86. (b) Meltzer, P.
C.; Butler, D.; Deschamps, J. R.; Madras, B. K. J. Med. Chem. 2006, 49,
1420−1432. (c) Blough, B. E.; Landavazo, A.; Partilla, J. S.; Baumann,
M. H.; Decker, A. M.; Page, K. M.; Rothman, R. B. ACS Med. Chem. Lett.
2014, 5, 623−627. (d) Garcia-Mijares, M.; Martins Torres Bernardes,
A.; Araujo Silva, M. T. Pharmacol., Biochem. Behav. 2009, 91, 621−628.
(e) Akbari, S. H.; Reynolds, M. R.; Kadkhodayan, Y.; Cross, D. T.;
Moran, C. J. J. NeuroIntervent. Surg. 2013, 5, 337−343.
(2)(a)Franzen, R. G. J. Comb.Chem. 2000, 2, 195−214. (b)Bouteiller,
C.; Becerril-Ortega, J.; Marchand, P.; Nicole, O.; Barre, L.; Buisson, A.;
́
Perrio, C. Org. Biomol. Chem. 2010, 8, 1111−1120. (c) Hill, R. K.;
Nugara, P. N. J. Org. Chem. 1992, 57, 1045−1047. (d) Cuadros, R.;
́
Montejo de Garcini, E.; Wandosell, F.; Faircloth, G.; Fernandez-Sousa,
J. M.;Avila, J. CancerLett. 2000, 152, 23−29. (e)Carroll, F. I.;Blough, B.
E.; Abraham, P.; Mills, A. C.; Holleman, J. A.; Wolckenhauer, S. A.;
Decker, A. M.; Landavazo, A.; McElroy, K. T.; Navarro, H. A.; Gatch, M.
B.; Forster, M. J. J. Med. Chem. 2009, 52, 6768−6781.
(3) (a) Fisher, L. E.; Muchowski, J. M. Org. Prep. Proced. Int. 1990, 22,
399−484. (b)Tian, J.S.;Ng,K.W.;Wong,J.R.;Loh,T.P. Angew.Chem.,
Int. Ed. 2012, 51, 9105−9109. (c) Pfaltz, A.; Anwar, S. Tetrahedron Lett.
1984, 25, 2977−2980. (d) McClure, D. E.; Arison, B. H.; Jones, J. H.;
Baldwin, J. J. J. Org. Chem. 1981, 46, 2431−2433. (e) Fisher, M. J.;
Overman, L. E. J. Org. Chem. 1990, 55, 1447−1459. (f) End, N.;
Sommerlade, R. H.; Richter, Y. U.S. Patent 8252959B2, 2012.
(g) Stevens, C. L. U.S. Patent 3254124, 1966.
(4) For recent examples, see: (a) Evans, R. W.; Zbieg, J. R.; Zhu, S.; Li,
W.; MacMillan, D. W. C. J. Am. Chem. Soc. 2013, 135, 16074−16077.
(b) Rajeshkumar, V.; Chandrasekar, S.; Sekar, G. Org. Biomol. Chem.
2014, 12,8512−8518. (c)Lv, Y.;Li,Y.;Xiong, T.;Lu, Y.;Liu, Q.;Zhang,
Q. Chem. Commun. 2014, 50, 2367−2369. (d) Tian, J. S.; Loh, T. P.
Angew. Chem., Int. Ed. 2010, 49, 8417−8020. (e) Tian, J. S.; Loh, T. P.
Angew. Chem., Int. Ed. 2011, 50, 5458−5460.
(5) (a) Selig, P. Angew. Chem., Int. Ed. 2013, 52, 7080−7082.
(b) Matsuda, N.; Hirano, K.; Satoh, T.; Miura, M. Angew. Chem., Int. Ed.
2012, 51, 11827−11831. (c) Miura, T.; Morimoto, M.; Murakami, M.
Org. Lett. 2012, 14, 5214−5217. (d) Adam, W.; Roschmann, K. J.; Saha-
̈
Moller, C. R. Eur. J. Org. Chem. 2000, 2000, 557−561. For a review, see:
(e) Erdik, E. Tetrahedron 2004, 60, 8747−8782.
(6) (a) Vander Wal, M. N.; Dilger, A. K.; MacMillan, D. W. C. Chem.
Sci. 2013, 4, 3075−3079. (b) Guha, S.; Rajeshkumar, V.; Kotha, S. S.;
Sekar, G. Org. Lett. 2015, 17, 406−409. (c)Wei, Y.;Lin, S.;Liang, F. Org.
Lett. 2012, 14, 4202−4205.
(7) Miyata, O.; Miyoshi, T.; Ueda, M. ARKIVOC 2013, No. ii, 60−81.
See also references cited therein.
(8) Jiang, Q.; Xu, B.; Zhao, A.; Jia, J.; Liu, T.; Guo, T. J. Org. Chem.
2014, 79, 8750−8756.
(9) Lamani, M.; Prabhu, K. R. Chem. - Eur. J. 2012, 18, 14638−14642.
(10)Liang, S.;Zeng, C.C.;Tian, H.Y.;Sun, B. G.;Luo,X. G.;Ren, F. Z.
J. Org. Chem. 2016, 81, 11565−11573.
(11) Mizar, P.; Wirth, T. Angew. Chem., Int. Ed. 2014, 53, 5993−5997.
(12) (a) Meyer, T. H.; Finger, L. H.; Gandeepan, P.; Ackermann, L.
̈
Trends Chem. 2019, 1, 63−76. (b) Wiebe, T.; Gieshoff, T.; Mohle, S.;
Rodrigo, E.;Zirbes, M.;Waldvogel, S. R. Angew. Chem., Int. Ed. 2018, 57,
̈
5594−5619. (c) Mohle, S.; Zirbes, M.; Rodrigo, E.; Gieshoff, T.; Wiebe,
A.; Waldvogel, S. R. Angew. Chem., Int. Ed. 2018, 57, 6018−6041.
(d)Frontana-Uribe,B.A.;Little,R. D.;Ibanez, J.G.;Palma,A.;Vasquez-
Medrano, R. Green Chem. 2010, 12, 2099−2119. (e) Steckhan, E.; Arns,
̈
̈
T.; Heineman, W. R.; Hilt, G.; Hoormann, D.; Jorissen, J.; Kroner, L.;
Lewall, B.; Putter, H. Chemosphere 2001, 43, 63−73.
̈
̈
(13) Pitzer, L.; Schafers, F.; Glorius, F. Angew. Chem., Int. Ed. 2019, 58,
8572−8576.
(14) When using 1-trimethylsiloxycyclohexene as starting material no
conversion to the corresponding α-amino ketone was observed, but the
E
https://dx.doi.org/10.1021/acs.orglett.0c02068
Org. Lett. XXXX, XXX, XXX−XXX