Why Cyclopropanation is not Involved in Photoinduced α-Alkylation of Ketones with Diazo Compounds

Article abstract of DOI:10.1002/ejoc.201800542

Diazo compounds are widely used reagents in organic synthesis. Direct photolysis leads to singlet carbenes whereas, in the presence of a photosensitizer, carbenes in the triplet state are generated. However, their reactivity under light irradiation remains poorly understood. Herein, we report photocatalytic, direct alkylation of enamines, generated from ketones, with diazo esters in the presence of free-base porphyrins acting as photoredox catalysts. Based on the experimental and theoretical studies, we propose a plausible mechanism involving generation of an enamine radical cation that subsequently reacts with diazo compounds giving the radical thus excluding the controversial cyclopropanation–ring opening pathway.

Full text of DOI:10.1002/ejoc.201800542

A Journal of
Accepted Article
Title: Why Cyclopropanation is not involved in Photoinduced α-
Alkylation of Ketones with Diazo Compounds?
Authors: Dorota Gryko, Katarzyna Rybicka-Jasińska, Katarzyna
Orłowska, Maksymilian Karczewski, and Katarzyna Zawada
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201800542
Link to VoR: http://dx.doi.org/10.1002/ejoc.201800542
Supported by

10.1002/ejoc.201800542
European Journal of Organic Chemistry
FULL PAPER
Why Cyclopropanation is not involved in Photoinduced
α-Alkylation of Ketones with Diazo Compounds?
Katarzyna Rybicka-Jasińska,[a] Katarzyna Orłowska,[a] Maksymilian Karczewski,[a] Katarzyna Zawada,*[b]
Dorota Gryko*[a]
Abstract: Diazo compounds are widely used reagents in organic
synthesis. Direct photolysis leads to singlet carbenes while in the
presence of a photosensitizer carbenes in the triplet state are
generated. However, their reactivity under light irradiation remains
poorly understood. Herein, we report photocatalytic, direct alkylation
of enamines, generated from ketones, with diazo esters in the
presence of free base porphyrins acting as photoredox catalysts.
Based on the experimental and theoretical studies, we propose a
plausible mechanism involving generation of an enamine radical
cation that subsequently reacts with diazo compounds thus
excluding the controversial cyclopropanation-ring opening pathway.
carbonyl compounds under light irradiation. The reaction of
dimethyl diazo malonate with carbonyl compounds in the
presence/absence of benzophenone as a sensitizer furnishes a
mixture of products. According to the mechanism, an ylide
intermediate predominates upon direct irradiation while in the
presence of a sensitizer insertion of a triplet carbene to the
secondary C-H is also observed.5 This clearly shows that the
reaction pathway is determined by a competition between the
singlet and triplet carbene formation. Recently, we have reported
that in the presence of either porphyrin or [Ru(bpy)3]Cl2 as the
photoredox catalyst, ethyl diazoacetate acts selectively as the
alkylating agent of aldehydes (precisely enamines) affording
products in high yields.6a Mechanistic studies confirmed the
radical nature of the process that presumably involves the
reaction of an enamine radical cation with a triplet carbene.
Later, Meggers et al reported α-alkylation of 2-acyl imidazoles
with diazo compounds in the presence of chiral Ru-photoredox
catalyst and Rh-Lewis acid.7 It was hypothesized that generated
in situ [Ru(bpy)3]+ transfers an electron to diazo ester and after
extrusion of nitrogen and protonation produces carbon-centred
radicals. Interestingly, only Rh-enolate quenches the excited
state of [Ru(bpy)3]2+.
As far as the reaction with enamines and enols is concerned,
an alternative pathway - the cyclopropanation of enamines/enols
followed by the ring-opening, should be considered. It is known
that cyclopropanes bearing both electron-donating and
withdrawing substituents are easily cleaved.8 For example, the
reaction of enamines with diazoethane in the presence of copper
chloride leads to cyclopropylamines, which after cleavage under
thermal hydrolysis furnish a mixture of alkylated carbonyl
compounds (Scheme 2a).9
Introduction
Diazo compounds are one of the most versatile chemical
feedstocks and key components in numerous synthetic
transformations,
particularly
in cyclopropanation,
Wolf
rearrangements, C-H or heteroatom-H insertions, and
polymerization.1 Over the last decade much effort has been also
devoted to the reactivity of these compounds under light
irradiation.2 It was found that direct irradiation gives singlet
carbenes while photosensitization generates carbenes in the
triplet state (Scheme 1).3 As a consequence, their reactions with
olefins afford cis- or cis- and trans-cyclopropanes respectively.4
Scheme 1. The reactivity of diazo compounds under light irradiation.
Although these reactions have been extensively studied,
little is known about the reactivity of diazo reagents towards
[a]
[b]
K. Rybicka-Jasińska, K. Orłowska, M. Karczewski, D. Gryko*
Institute of Organic Chemistry
Polish Academy of Sciences
Kasprzaka 44/52, 01-224 Warsaw, Poland
E-mail: dorota.gryko@icho.edu.pl
K. Zawada*
Faculty of Pharmacy with the Laboratory Medicine Division,
Department of Physical Chemistry
Medical University of Warsaw
Banacha 1, 02-097 Warsaw, Poland
E-mail:katarzyna.zawada@wum.edu.pl
Scheme 2. Reactions involving cyclopropanation followed by ring-opening.
Moreover, copper and rhodium complexes catalyse the
reaction of enamines with aryldiazoacetates leading to
-ketoesters in moderate yield via cyclopropanation/ring-opening
pathway.10 Similarly, the Rh(II)-catalysed reaction of
α-diazoketones with vinyl ethers affords
a cyclopropane
derivative which subsequently undergoes the C-C bond
cleavage yielding 1,4-dicarbonyl compounds (Scheme 2b).11
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800542
European Journal of Organic Chemistry
FULL PAPER
Table 1. Initial Studies.[a]
Differences in the proposed mechanisms for light-induced
and metal-catalysed reactions involving diazo compounds arisen
an important question – If photocatalyzed reaction of enamines
with diazo reagents proceed via the cyclopropanation/ring
opening pathway?
Entry
Amine
morpholine
pyrrolidine
pyrrolidine
pyrrolidine
pyrrolidine
-
Light
LED525nm
LED525nm
LED525nm
-
Cat. 11
Yield[b] [%]
1
2
a
b
-
To answer this question we resolved to continue our studies
on photocatalytic reactions involving diazo reagents. In addition
to exploring mechanistic aspects, we wondered whether their
reactivity could be shifted also toward α-functionalization of
more-challenging ketones. Consequently, we have developed
photoinduced alkylation of simple ketones with diazo
compounds in the presence of a free-base porphyrin and a
secondary amine is reported. Interestingly, for certain substrates
the reaction proceeds in the absence of the porphyrin. Extensive
mechanistic studies supported by DFT calculations enabled us
to present a comprehensive view of the mechanism. The
reaction proceeds via the radical pathway and does not involve
the assumed formation of a cyclopropane intermediate.
3
51
traces
68
4
b
b
b
5[c]
6
-
LED525nm
0
[a] Reaction conditions: ketone (8a, 0.5 mmol), diazo ester (9a, 0.5 mmol),
porphyrin 11a or 11b (1.0 mol%), amine (20 mol %), DMSO/buffer pH = 4.0,
9/1 (v/v) 5 mL, 8 h, rt. [b] Isolated yields. [c] Reaction was performed at 45C.
To explain why the reaction proceeds regardless of the
presence of a photocatalyst further experimental work was
required.
Results and Discussion
Optimization studies
Photoredox catalysis enables introduction of functional
groups (e. g. trifluoromethyl13 benzyl,14 alkyl15) at the α-position
to the aldehyde moiety.16 Due to lower reactivity of ketones, only
few methods for the α-alkylation of ketones under light irradiation
were reported though each possessing certain limitations.17 In
the presence of Ru(bpy)3Cl2 as a photoredox catalyst, enamines,
generated in situ from a primary amine and -ketocarbonyls,
react with activated organic bromides giving functionalized -
ketoesters.18 When enamine intermediates (from cyclic ketones
or β-ketoesters) form with alkyl halides photoactive electron
donor-acceptor (EDA) complexes, alkylation does not require
any photocatalyst.19 On the other hand, Meggers et al showed
that with a photocatalyst absorbing light and at the same time
activating the carbonyl group, certain carbonyl compounds can
be functionalized with benzyl and phenacyl bromides.20 Though
stereoselective and high yielding, these catalytic procedures are
limited to specifically designed either carbonyl compounds or
alkylating reagents, or require the addition of co-catalysts.
Hence, we resolved to optimize our newly developed method
focusing on identifying the optimal reaction conditions giving
product 10a in high yields (for details see SI). Substituents
present at the periphery of the porphyrin macrocycle have a
strong impact on the redox properties,21 therefore various
porphyrins with electron-withdrawing and electron-donating
substituents were tested (Figure 1).
Initial Studies
In the initial experiment 4-oxotetrahydropyran (8a) was
reacted with benzyl diazoacetate (BDA, 9a) in the presence of
morpholine
tetraphenylporphyrin (H2TPP, 11a) as a photoredox catalyst
under our previously developed conditions (Scheme 3).6b The
reaction led only to the recovery of the starting material.
However, as the reactivity of enamines follows the order
pyrrolidine > acyclic amine > piperidine > morpholine- derived
enamines,12 we assumed that for ketones, a more reactive
enamine should be generated as an intermediate.
as
an
organocatalyst
and
free-base
All of them catalysed the model reaction of
4-oxotetrahydropyran (8a) with BDA (9a) giving the desired
product 10a, with free base porphyrin 11b being the most
effective.
Expectedly, only in the presence of primary and secondary
amines the reaction furnished the desired product, reaching 68%
for pyrrolidine (Table 2, entries 1-4). Tertiary amine NEt3 did not
catalyse the reaction, thus confirming the assumed role of an
amine to form enamine in the catalytic cycle.
Scheme 3. Light-Induced reaction of ketones with BDA (9a).
Indeed, the use of pyrrolidine enabled the formation of
α-alkylated product 10a in reasonable yield (Scheme 3, Table 1,
entry 2). Control experiments revealed that both light and
pyrrolidine are essential as their exclusion halted the reaction
completely (Table 1, entries 4-6). Surprisingly, the reaction gave
product even in the absence of porphyrin 11b (Table 1, entry 3).
Further optimization studies comprised experiments with
different diazo ester 9a/ketone 8a ratios in relation to
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800542
European Journal of Organic Chemistry
FULL PAPER
H2T(p-CO2MeP)P (11b) as a photocatalyst and pyrrolidine as an
organocatalyst. Importantly, the use of starting materials 8a/9a
in 1:1.2 molar ratio led to a significant increase in the reaction
yield up to 83ꢀ% (Table 2, entry 7). In the case of
4-oxotetrahydropyran (8a) only mono-substituted product 10a
formed but alkylation of cyclohexanone (8b) led to a mixture of
mono- 10b and di-substituted 10bb products.
Consequently, the conditions predominantly yielding either
mono- or di-product were defined (Table 3, for details see SI).
Table 3. Optimization studies.[a]
Amine
[mol%]
10b[b]
[%]
10bb[b]
[%]
Entry
1[c]
9a [eq.]
Time [h]
5
1
20
-
-
2[d]
3
1
1.2
1
20
20
20
40
5
5
5
7
10
30
48
10
2
30
20
80
4[e]
5
3
[a] Reaction conditions: ketone (8b, 0.25 mmol, 1 equiv.), BDA (9a, X equiv.),
photocatalyst (11b, 1.0 mol%), pyrrolidine (X mol%), DMSO/buffer pH = 4.0,
9:1 (v/v) 5 mL, LED525nm, 5 h, rt. [b] Isolated yields. [c] Without light. [d] Without
photocatalyst. [e] Ketone (8b, 0.25 mmol, 1 equiv.), BDA (9a, X equiv.),
photocatalyst (11b, 1.0 mol%), pyrrolidine (X mol%), DMSO/buffer pH 7.0 =
9:1 (v/v) 2.5 mL, LED525nm, 5 h, rt.
Reaction scope and limitations
Under the developed conditions the scope and limitations of
the -functionalisation of ketones were explored employing two
sets of conditions (with and without porphyrin 11b) for mono-
alkylated product (Figure 2). Method A for mono-product:
ketone (0.5 mmol), pyrrolidine (20 mol%), diazo ester (1.2
equiv.), H2T(p-CO2MeP)P (11b, 1 mol%) a DMSO/buffer mixture
(9/1 (v/v), 5 mL, buffer pH = 4, c = 0.1 M), 5 h, LED525nm, rt.
(Method A’ – same as A but without photocatalyst added).
Additionally, -difunctionalisation of ketones was explored
employing two sets of conditions (with and without porphyrin):
Method B for di-product: ketone (0.25 mmol), pyrrolidine
(40 mol%), diazo ester (3 equiv.), H2T(p-CO2MeP)P (11b,
1 mol%) a DMSO/buffer mixture (9/1 (v/v), 2.5 mL, buffer pH = 7,
c = 0.1 M), 7 h, LED525nm, rt (Figure 3, for optimization see SI).
Dialkylated products were only obtained in porphyrin-
catalysed reaction (Chart 3). In general, ketones that easily form
enamines were reactive under the developed conditions.22 The
mono-alkylation reaction of ketones with various diazo esters
(CO2Bn (9a), CO2Et (9b), CO2t-Bu (9c), CO2CH2CH2Ph (9d),
CO2CH2(p-OMePh) (9e)) gave products in decent yields.
However, the reaction with N2CHCO2CH2(p-NO2Ph) (9f) proved
difficult under developed conditions presumably due to the
strong electron-withdrawing character of the NO2 group.23 The
desired product 10m was even obtained from the aliphatic
ketone – acetone, but acetophenone remains unreacted as with
pyrrolidine it did not generate enamine.
Figure 1. Porphyrins tested as photocatalysts in alkylation of ketone 8a. Yields
of product 10a.
Table 2. Optimization studies.[a]
Entry
1
8a/9a [equiv.]
Amine
morpholine
piperazine
NEt3
Yield[c] [%]
1/1
1/1
21
21
0
2
3
1/1
4
1/1
pyrrolidine
pyrrolidine
pyrrolidine
pyrrolidine
71
74
65
83
5[b]
6[b]
7[b]
1/1
1.2/1
1/1.2
[a] Reaction conditions: ketone (8a, 0.5 mmol), BDA (9a, 0.5 mmol), porphyrin
(11b, 1.0 mol%), amine (20 mol%), DMSO/buffer pH 4.0 = 9:1 (v/v) 5 mL,
LEDs525nm, 7 h. [b] Reaction finished after 5 h, LED525nm. [c] Isolated yields.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800542
European Journal of Organic Chemistry
FULL PAPER
better catalytic systems’24 we resolved to perform detailed
mechanistic
studies.
The
model
reaction
of
4-oxotetrahydropyran (8a) with BDA (9a) using pyrrolidine as an
organocatalyst afforded products regardless of the presence of
the porphyrin though in various yields. Consequently, based on
the CV, radical trapping, EPR, UV-Vis experiments supported by
theoretical calculations, two plausible mechanistic pathways are
proposed (Scheme 4).
Scheme 4. Proposed mechanisms for light-induced functionalization of
ketones with BDA.
Figure 2. Scope and limitations – mono-alkylated products.
In both reaction pathways I and II, the initial step involves the
reaction of the ketone with pyrrolidine leading to enamine A
(detected by 1H NMR and ESI-MS analyses, see SI). Under light
irradiation, it (0.49 V vs SCE, MeCN)25 is oxidized to radical
cation B by H2T(p-CO2MeP)P in the excited state (11b,
Ered*[Por*/Por•-] = 1.03 V vs SCE, DMSO), if present (pathway I).
Otherwise, it is enamine A that absorbs light and in the exited
state is assumed to be oxidized by oxygen (pathway II, 0.99 V
vs SCE, solvent). Subsequently, radical cation B reacts with
protonated diazo reagent C with the simultaneous extrusion of
nitrogen. The resultant adduct D undergoes reduction to betaine
Figure 3. Scope and limitations – dialkylated products.
E
through a PET process with the regeneration of the
•-
photocatalyst or is reduced by O2 . The catalytic cycle ends with
the hydrolysis of the imine (in acidic conditions) affording the
desired product 10 and regenerating the amine. In addition,
chain propagation reactions may likely be also involved as the
quantum yield of the reaction is 20% when the porphyrin is
added, and 7% when there is no photocatalyst in the reaction
mixture.26
In the dialkylation reaction, a variety of six-membered carbo-
and heterocycles furnished products in high yields and with
diastereoselectivity up 4 : 1. Interestingly, N-Boc-piperidone
(10o)
underwent
the
exclusive
formation
of
2,5-dialkylated product (10oo).
In the following sections we discuss the experiments
supporting the proposed mechanistic pathways.
Mechanistic investigations
As stated by prof. König, ‘Photoredox reactions with visible
light have been used in a broad range of transformations,
though the mechanism needs to be fully understood to design
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800542
European Journal of Organic Chemistry
FULL PAPER
Verification of cyclopropane - ring opening mechanism.
As an alternative pathway, cyclopropanation of an enamine
followed by ring-opening was considered. Kuehne and King
reported that the reaction of enamines with diazoethane in the
presence of copper chloride gives cyclopropylamines.9 To
ascertain the involvement of a cyclopropane intermediate or the
formation of regioisomeric products, the progress of the reaction
was monitored by 1H NMR, but we were not able to detect either
of them (for details see SI). As cyclopropanes bearing both
electron-donating and electron-withdrawing groups are easily
cleaved,8 their detection in the reaction mixture might be difficult.
Therefore, olefins (12a-c)
-
common substrates for
cyclopropanation were subjected to our model, porphyrin-
catalysed reaction conditions. They remained unreactive (were
recovered), thus supporting the lack of the cyclopropane-
intermediate pathway in the considered reaction mechanism
(Scheme 5).
Figure 4. Potential energy surface (PES) for transformation of mono-
substituted enamine to cyclopropane.
could originate from carbenes generated from direct or
photosensitized photolysis of EDA (9b) but current data suggest
that in our case, they are rather not involved in the C-C bond
forming reaction. In addition, in the presence of water, triplet
carbenes may abstract hydrogen atom and the hydroxyalkyl
radicals generated may induce the decomposition of EDA by
chain processes.3
Scheme 5. Verification of cyclopropane-intermediate mechanism.
Intrigued by the question why cyclopropanation is not
observed in the alkylation of ketones, theoretical studies were
pursued. The potential energy surface (PES) was scanned for
both the enamine and imine cationic radical (Figure 4).
Furthermore, two parameters were studied at the M062x/6-
31G(d) level of theory – the angle (α) between the newly formed
C-C bond (blue-red dotted) and the cyclohexyl ring (blue) and
the dihedral angle (Θ) between the mentioned bond and the
ester moiety (red).27 In each case the potential energy surface is
the same with respect to its shape implying a spontaneous
process in which once formed, cyclopropane should stay as the
final product. As in our case only the alkylated derivative is
observed, so the lack of cyclopropanation is supported. It is
known that in acidic solutions diazo acetates are protonated.28 In
our reaction proceeding at pH = 4, EDA (9b) should be
protonated and as such reacts with radical cation B forming
intermediate D. This, already bears two hydrogen atoms at the
α-position to the ester group, and therefore is not able to form
cyclopropane intermediate.
Verification of a radical mechanism. In accordance with the
proposed mechanism reactive radicals are formed and indeed
the addition of TEMPO (a radical scavenger) stopped the
reaction completely. The reaction involves reactive species both
in triplet and singlet excited states as the addition of either
benzoquinone29 or cycloheksan-1,3-dien30 (singlet and triplet
exited states quenchers) stopped model reactions (with and
without porphyrin) completely (for details see SI). Two radical
species as TEMPO adducts 14, 15 were detected by MS further
evidencing the involvement of radical B (Scheme 6). Adduct 15
Scheme 6. TEMPO trapping.
Additionally, the use of two different spin traps (PBN and
DMPO) in EPR experiments allowed for the detection of two
paramagnetic species B and D, thus supporting the proposed
radical mechanism. In the presence of PBN, the EPR spectrum
recorded for the reaction of 4-oxotetrahydropyran (8a) with EDA
(9b) catalysed by porphyrin 11b and pyrrolidine, is the
superposition of two very similar components (triplets of
doublets with hyperfine splitting constants: aN = 1.49 mT, aH β
=
0.44 mT and aN = 1.51 mT, aH β = 0.41 mT), of relative intensity
53 and 47%, respectively (Figure 5a). Their hyperfine splitting
constants suggest the presence of carbon-centred radicals, as
they are similar to values obtained for PBN-benzoyl radical
adduct in DMSO solution (aN = 1.45 mT and aH = 0.47 mT).
When the reaction was conducted in the presence of DMPO
(Figure 5b) two components are also seen, the dominating one
(90% of total intensity) with hyperfine splitting constants of 1.47
mT (aN) and 2.16 mT (aHβ) and a second one with aN = 1.60 mT
and aH = 2.46 mT. In this case, both components also indicate
the formation of carbon-centred radicals, as the value of aHβ is
higher than that of aN.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800542
European Journal of Organic Chemistry
FULL PAPER
2000
1000
0
3000
1000
-1000
-3000
-5000
-7000
-1000
-2000
-3000
-4000
329
330
331
332
333
334
B [mT]
329
330
331
332
333
334
B [mT]
measured
simulated
measured
simulated: component 1
simulated
simulated: component 1
simulated: component 2
simulated: component 2
4000
2000
0
Figure 6. EPR spectra of the mixture of 4-oxotetrahydropyran (8a), pyrrolidine,
EDA (9b) in DMSO/buffer with PBN. Reaction conditions: ketone 8a (1 equiv.,
0.25 mmol), pyrrolidine (0.2 equiv.), EDA (9b, 1 equiv.) DMSO/buffer pH = 4
(5 mL, 9/1 (v/v) mixture), spin trap PBN after 10 min of irradiation with LED.
-2000
-4000
-6000
-8000
-10000
-12000
The same pattern, but with a significantly lower total intensity, is
also observed in the absence of EDA (9b).
Furthermore, to confirm the formation of the oxygen
derived radical, the reaction of 4-oxotetrahydropyran (8a) with
pyrrolidine was conducted in aerated conditions, under green
light irradiation. In these conditions only one component is seen
328
329
330
331
332
333
334
335
B [mT]
measured
simulated
simulated: component 1
simulated: component 2
Figure 5. EPR spectra for the mixture of 4-oxotetrahydropyran (8a),
pyrrolidine, EDA (9b), porphyrin 11b in DMSO/buffer: a) with PBN, b) DMPO.
in the EPR spectrum with PBN, with aN = 1.39 mT and aH β
=
0.23 mT – which is also observed as a minor one in the
spectrum of 4-oxotetrahydropyran (8a), pyrrolidine, and EDA
(2b) system in degassed solvent. When DMPO was used as a
spin trap with the mixture of 4-oxotetrahydropyran (8a), EDA
(9b), and pyrrolidine was irradiated with green light, the
dominating component (90% of total intensity) is the same as in
the presence of the porphyrin (aN = 1.47 mT, aH β = 2.16 mT),
arising from the carbon-centred radical (B). In accordance with
the proposed mechanism reactive radicals are formed. EPR
measurements proved the formation of two different carbon-
centred radicals, which after analysing simulated hyperfine
splitting constants of spin adducts can be assigned to radical
cation B and radical cation D. Additionally, no signal from a
carbene radical was found in the reaction mixture or background
reactions.
Stern-Volmer analysis. Expectedly, NMR experiments
confirm that enamine forms under developed conditions (see SI).
The quenching rates (kq) for each of the reaction components
were measured using standard Stern−Volmer analyses. Only
enamine A (kq = 4.1 x 1010 [M-1s-1]) strongly quenches the
luminescence of porphyrin 11b while for EDA (9b) (kq = 4.5 x 109
[M-1s-1]), ketone 8a (kq = 1.2 x 108 [M-1s-1]), and pyrrolidine
(kq ~ 8.2 x 107 [M-1s-1])31 quenching rates are much smaller
(Figure 7). This is in contrast to the reaction with aldehydes
where the highest quenching rate was measured for EDA (9b)6a
thus suggesting differences in the mechanistic pathways.
Reaction conditions: 4-oxotetrahydropyran (8a,
1 equiv., 0.25 mmol),
pyrrolidine (0.2 equiv.), EDA (9b, equiv.), porphyrin 11b (1 mol%),
1
DMSO/buffer pH = 4 (5 mL, 9/1 (v/v), mixture), spin trap PBN/DMPO after 10
min of irradiation with green LED.
In the EPR spectrum registered after the visible light
irradiation of a mixture of porphyrin 11b, pyrrolidine, and
4-oxotetrahydropyran (8a) (no EDA added) in the presence of
PBN a triplet of doublets is observed. The best fit is obtained for
two very similar components (hyperfine splitting constants: aN =
1.51 mT, aH β = 0.33 mT and aN = 1.50 mT, aH β = 0.35 mT).
The spectrum is very similar to the one measured for the whole
reaction, differing only with respect to intensity thus strongly
supporting the formation of radical B.
Under porphyrin-free conditions, the mixture of ketone (8a),
EDA (9b) and pyrrolidine, in the presence of PBN, a two-
component EPR spectrum is observed (aN = 1.51 mT, aH β =
0.40 mT, I = 62% and aN = 1.38 mT, aH β = 0.24 mT, I = 38%)
(Figure 6). The first component is the same as one of the
components in the EPR spectrum of a reaction mixture
(compare with Figure 5b) and corresponds to a carbon-centred
radical. The second component is similar to the one present in
the Fenton reaction (hydrogen peroxide and ferrous sulphate) in
the DMSO:buffer mixture and it can be assigned to PBN-CH3 or
PBN-OH adducts thus suggesting the presence of superoxide
anion radical formed in the reaction with molecular oxygen.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800542
European Journal of Organic Chemistry
FULL PAPER
a) 4-oxotetrahydropyran+pyrrolidine
19
14
9
2
1
0
y = 223,06x + 0,2398
y = 25,58x + 0,921
y = 0,9105x + 0,9908
y = -0,293x + 0,9945
4
300
350
400
450
500
550
Wavelenght [nm]
4-oxotetrahydropyran
pyrrolidine
-1
0
0,02
0,04
0,06
0,08
0,1
0,12
C [M]
0,14
pyrrolidine
4-oxotetrahydropyran+pyrrolidine 0 min
4-oxotetrahydropyran+pyrrolidine 20 min
4-oxotetrahydropyran+pyrrolidine 40 min
4-oxotetrahydropyran+pyrrolidine 60 min
4-oxotetrahydropyran
EDA
4-oxotetrahydropyran+pyrrolidine
b) 4-oxotetrahydropyran+pyrrolidine+BDA
3,5
3
2
1
0
2,5
2
1,5
1
0,5
0
620
630
640
650
Wavelenght [nm]
pyrrolidine (0 mmol) + 4-oxotetrahydropyran (0 mmol)
660
670
680
390
490
0 h
590
Wavelenght [nm]
1 h
690
790
3 h
5 h
pyrrolidine (0.05 mmol) + 4-oxotetrahydropyran (0.05 mmol)
pyrrolidine (0.1 mmol) + 4-oxotetrahydropyran (0.1 mmol)
pyrrolidine (0.15 mmol) + 4-oxotetrahydropyran (0.15 mmol)
pyrrolidine (0.2 mmol) + 4-oxotetrahydropyran (0.2 mmol)
c) cyclohexanone+pyrrolidine+BDA
3
2,5
2
Figure 7. Up: Stern−Volmer quenching experiment for H2T(p-CO2MeP)P
(11b). Experimental conditions: for pyrrolidine, EDA (9b), enamine (formed in
situ), and 4-oxotetrahydropyran (8a) samples were prepared by adding the
solutions of substrates (2.5 x 10-3 M) to H2T(p-CO2MeP)P (11b) in DMSO
1,5
1
(total volume
2 mL) and degassed with Ar. The concentration of
0,5
0
H2T(p-CO2MeP)P (11b) in DMSO was 7.2 × 10-7 M. Down: Fluorescence
quenching of H2T(p-CO2MeP)P (11b) (7.2 × 10-7 M in DMSO) upon titration
with pyrrolidine and 4-oxotetrahydropyran (8a).
390
440
490
540
590
640
5 h
690
Wavelenght [nm]
1 h
3 h
Presumably, these differences result from the fact that enamines
formed from cyclic ketones and pyrrolidine are easier to
oxidize32 than enamines generated from aldehydes and
morpholine.33
2
d) acetone+pyrrolidine+BDA
1
0
The main reaction pathway involves the oxidation of enamine A
to the radical cation. Free-base porphyrins can act as both
reducing and oxidizing agents, and the driving force for the
electron transfer relates to the standard potential of oxidation of
the donor the standard potential of the acceptor, and the energy
of excited state. In the excited state reduction potential of
H2T(p-CO2MeP)P (11b, 1.03 V vs SCE, DMSO) is high enough
to oxidize enamine A (0.49 V vs SCE, MeCN) to radical cation B.
Moreover, the Rehm−Weller formalism allows for estimating the
thermodynamic driving force, -GPET(0), for PET between the
enamines and porphyrins in the excited-state, (-GPET(0)  -0.6 V).
Because of the irreversible electrochemical oxidation of
enamines and the solvents used (DMSO/buffer), we do not have
exact values for their oxidation potentials.
300
400
0 h
500
600
700
Wavelenght [nm]
1 h
3 h
5 h
Figure 8. UV-Vis Experiment. Reaction conditions: a) control experiments: 1)
ketone (8a) (0.25 mmol, DMSO/buffer pH = 4, 1.5 mL, 9/1 (v/v) mixture); 2)
pyrrolidine (20 mol%), DMSO/buffer pH = 4 (1.5 mL, 9/1 mixture); 3) ketone
(8a) (0.25 mmol), pyrrolidine (20 mol%), DMSO/buffer pH = 4 (1.5 mL, 9/1 v/v
mixture) measured in 0, 20, 40, 60 min after mixing. b) ketone (8a, 0.25
mmol), pyrrolidine (20 mol %), BDA (9a, 0.25 mmol), DMSO/buffer pH = 4, 1.5
mL, 9/1 (v/v) mixture), LED525nm. c) cyclohexanone (8b, 0.25 mmol), pyrrolidine
(20 mol%), BDA (9a, 0.25 mmol), DMSO/buffer pH = 4, 1.5 mL, 9/1 (v/v)
mixture), LED525nm. d) acetone (8m, 0.25 mmol), pyrrolidine (20 mol%), BDA
(9a, 0.25 mmol), DMSO/buffer pH = 4, 1.5 mL, 9/1 (v/v) mixture), LED525nm
.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800542
European Journal of Organic Chemistry
FULL PAPER
For acetonitrile, the voltammograms show peak potentials about
0.49 V vs SCE for oxidation of enamines. Therefore, for PET
initiated from the singlet excited state of the porphyrins, ΔG most
likely assumes negative values, making it thermodynamically
favourable.
Conversely, the obtained Gibbs energies tentatively indicate that
singlet oxygen spontaneously oxidize enamine A (GB3LYP/6-
= -11.72 kcal/mol, GM06-2X/6-311++G(2d,2p), DMSO = -
311++G(2d,2p), DMSO
2.30 kcal/mol). Hence, the most probable path for porphyrin-free
reaction is the oxidation of enamine A by singlet oxygen. Even
more preferred would be – according to the calculated Gibbs
energy values – the oxidation of enamine A* in the excited state,
as these values are negative for both triplet (GM06-2X/6-311++G(2d,2p),
However, in the porphyrin-free reaction, it is enamine A that
absorbs light and in the excited state (A*) is oxidized
presumably by molecular oxygen. It is consistent with the
observation that the enamine generated in situ from 4-
oxotetrahydropyran (8a) and pyrrolidine absorbs visible light at
the maximum 437 nm with its absorption increasing over time
(Figure 8) (for details see SI). However, for enamines (from
cyclohexanone (8b) and acetone (8m), for details see SI) the
new absorption band is not present which is in agreement with
the fact that these ketones afford products only in the porphyrin-
catalysed reaction. Thus, in porphyrin-free reactions the
absorption of visible light by enamine is crucial for the product
formation. The mechanism for the porphyrin-free reaction
assumes oxidation of excited enamine A* by molecular oxygen.
Indeed, the yield increases when the reaction was exposed to
the air, opposite to the reaction with the porphyrin. This
photocatalyst is able to generate singlet oxygen and/or reactive
oxygen species ROS which by reacting with both the catalyst
and substrates diminished the reaction yield (Figure 6).
Furthermore, the addition of K2S2O8 as an oxidant (1.85 V vs
SCE, MeCN)34 to the degassed porphyrin-free reaction caused
an increase in the reaction yield from traces to 25%. Hence, we
concluded that molecular oxygen can oxidize enamine A in the
excited state to radical cation B if the porphyrin is not present.
Next, we focused on elucidating whether the second
alkylation occurs before or after the hydrolysis of iminium E to
the desired mono-alkylated product. To this end, we subjected
mono-alkylated ketones 10a and 10l to the developed reaction
conditions, however no conversion was observed (Scheme 7).
As a result, we concluded that dialkylated ketones form if
= -60.29 kcal/mol) and singlet (GM06-2X/6-311++G(2d,2p), DMSO
=
DMSO
-97.99 kcal/mol) oxygen.
For the reaction of porphyrin in the excited state (Por*) with
enamine A, the values of Gibbs energy are also negative (GM06-
= -7.55 kcal/mol) indicating that the oxidation of
2X/6-31G, DMSO
enamine A to radical cation B is favoured thermodynamically.
The consequent reduction of radical adduct D by porphyrin
radical anion is also favourable (GM06-2X/6-31G,
= -29.12
DMSO
kcal/mol).
•-
For the reduction of radical cation D by O2 , the B3LYP
functional gives negative value (GB3LYP/6-31G, DMSO = -
22.93 kcal/mol). However, by applying M06-2X and a larger
basis set functional positive value was obtained (GM06-2X/6-
311++G(2d,2p), DMSO = +8.95 kcal/mol). This might indicate
why in the absence of the porphyrin this step is less favoured
thermodynamically and is in agreement with the decreased
yields in porphyrin-free reaction in comparison to photocatalysed
reaction.
Conclusions
Developed, photoinduced -alkylation of ketones with diazo
esters gives access to mono- and dialkylated products. For
ketones forming enamines that absorb visible light the reaction
does not require the addition of the photocatalyst though yields
are diminished.
iminium
E directly undergoes second oxidation-alkylation
Extensive mechanistic studies supported by theoretical
calculations provide sufficient data to corroborate the proposed
radical mechanism. The reaction does not involve expected
cyclopropanation of enamine instead enamine radical cation B
reacts with diazo acetate leading to iminium radical.
sequence faster than hydrolysis to the mono-alkylated product.
These findings demonstrate venues in both diazo compound
chemistry and its utility in photoredox catalysis.
Scheme 7. Alkylation of product 10a.
Experimental Section
Theoretical studies. For catalytic cycle steps including
porphyrin in the excited state we have been able to conduct
calculations using only a very small basis set (B3LYP/6-31G35
and M06-2X/6-31G36). Thus, for all reactions in the presence of
porphyrin as a catalyst the data obtained from the small basis
set calculations are taken into account while the values for
reactions without porphyrin were obtained with the 6-
311++G(2d,2p) basis set.
General procedure for α-mono-functionalization of ketones –
method A:
To a 10 mL vial equipped with stir bar a photocatalyst (1 mol%) was
added and dissolved in a mixture of DMSO and buffer pH 4 (mixture 9:1,
5 mL). The vial was sealed and purged with argon for 5 min. Then the
ketone (0.5 mmol), pyrrolidine (0.2 equiv., 0.1 mmol) and diazo ester (1.2
equiv., 0.6 mmol) were added. The reaction mixture was stirred under
light irradiation (LED525nm, 25 ̊C) for 5 h. The light was turned off and the
The oxidation of the enamine by triplet oxygen seems to be
reaction mixture was diluted with Et2O, and washed with 1N HCl. The
aqueous phase was extracted with Et2O three times. The combined
organic phases were washed with saturated NaHCO3aq, brine, dried over
not allowed thermodynamically (GB3LYP/6-311++G(2d,2p),
=
DMSO
+27.02 kcal/mol, GM06-2X/6-311++G(2d,2p), DMSO
= +35.40 kcal/mol).
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800542
European Journal of Organic Chemistry
FULL PAPER
MgSO4, filtered, and concentrated. The crude product was purified by
flash chromatography using silica gel (hexanes/Et2O).
m) S. Das, C. G. Daniliuc, A. Studer, Angew. Chem. Int. Ed. 2017, 56,
11554-11558.
[9]
M. E. Kuehne, J. C. King, J. Org. Chem. 1973, 38, 304-311.
[10] A. Pereira, Y. Champouret, C. Martín, E. Álvarez, M. Etienne, T. R.
Belderraín, P. J. Perèz, Chem. Eur. J. 2015, 21, 9769-9775.
[11] S. Muthusamy, P. Srinivasan, Tetrahedron Lett. 2006, 47, 6297-6300.
[12] P. M. Pihko, I. Majander, A. Erkkilä, Asymmetric Organocatalysis, (Ed.:
B. List), 1st edn., Springer, 2010, pp. 145-200.
General procedure for α-di-functionalization of ketones – method B:
To a 10 mL vial equipped with stir bar a photocatalyst (1 mol%) was
added and dissolved in a mixture of DMSO and buffer pH 7 (mixture 9:1,
5 mL). The vial was sealed and purged with argon for 5 min. Then ketone
(1 equiv., 0.25 mmol), pyrrolidine (0.4 equiv., 0.1 mmol) and EDA (3
equiv., 0.75 mmol) were added. The reaction mixture was stirred under
[13] D. A. Nagib, M. E. Scott, D. W. C. McMillan, J. Am. Chem. Soc. 2009,
131, 10875-10877.
[14] H.-W. Shih, M. N. Vander Wal, R. L. Grange, D. W. C. MacMillan, J. Am.
Chem. Soc. 2010, 132, 13600-13603.
light irradiation (LED525nm, 25 ̊C) for 7 h. The light was turned off, the
reaction mixture was diluted with Et2O, and washed with 1N HCl. The
aqueous phase was extracted with Et2O three times. The combined
organic phases were washed with saturated NaHCO3aq, brine, dried over
MgSO4, filtered, and concentrated. The crude product was purified by
flash chromatography using silica gel (hexanes/Et2O).
[15] D. A. Nicewicz, D. W. C. MacMillan, Science 2008, 322, 77-80.
[16] For Photoredox catalysis see reviews: a) C. K. Prier, D. A. Rankic, D.
W. C. MacMillan, Chem. Rev. 2013, 113, 5322-5363. b) K. L. Skubi, T.
R. Blum, T. P. Yoon, Chem. Rev. 2016, 116, 10035-10074. c) J. W.
Tucker, C. R. J. Stephenson, J. Org. Chem. 2012, 77, 1617-1622. d) J.
J. Douglas, M. J. Sevrin, C. R. J. Stephenson, Org. Process Res. Dev.
2016, 20, 1134-1147. e) D. Staveness, I. B. Bosque, C. R. J.
Stephenson, Acc. Chem. Res. 2016, 49, 2295-2306. f) K. Teegardin, J.
I. Day, J. Chan, J. Weaver, Org. Process Res. Dev. 2016, 20, 1156-
1163. g) D. A. Nicewicz, T. M. Nguyen, ACS Catal 2014, 4, 355-360. h)
M. H. Shaw, J. Twilton, D. W. C. MacMillan, J. Org. Chem., 2016, 81,
6898-6926. h) J. Twilton, C. C. Le, P. Zhang, M. H. Shaw, R. W. Evans,
D. W. C. MacMillan, Nature Reviews Chemistry 2017, 1, 0052. i) Y.-Q.
Zou, F. M. Hörmann, T. Bach, Chem. Soc. Rev. 2018, DOI:
10.1039/C7CS00509A.
Acknowledgements
The support of the National Science Center (K.R.-J.,
PRELUDIUM Grant UMO-2016/21/N/ST5/03353, D.G., M. K.
OPUS Grant no. 2016/21/B/ST5/03169 is gratefully
acknowledged. The computational resources used in this work
were provided by ICM UW as a part of the G14-6 grant.
Calculations have been carried out using resources provided by
Wroclaw Centre for Networking and Supercomputing
(http://wcss.pl), grant No. 432.
[17] R. Cano, A. Zakarian, G. P. McGlacken, Angew. Chem. Int. Ed. 2017,
56, 9278-9290.
[18] Y. Zhu, L. Zhang, S. Luo, J. Am. Chem. Soc. 2014, 136, 14642-14645.
[19] a) L. Woźniak, J. J. Murphy, P. Melchiorre, J. Am. Chem. Soc. 2015,
137, 5678-5681. b) E. Arceo, A. Bahamonde, G. Bergonzini, P.
Melchiorre, Chem. Sci. 2014, 5, 2438-2442.
Keywords: diazo compounds • photoredox catalysis • alkylation
of ketones • porphyrins
[20] a) H. Huo, X. Shen, C. Wang, L. Zhang, P. Rëse, L.-A. Chen, K. Harms,
M. Marsch, G. Hilt, E. Meggers, Nature 2014, 515, 100-103. b) C.
Wang, Y. Zheng, H. Huo, P. Rëse, L. Zhang, K. Harms, G. Hilt, E.
Meggers, Chem. Eur. J. 2015, 21, 7355-7359.
[1]
a) A. Ford, H. Miel, A. Ring, C. N. Slattery, A. R. Maguire, M. A.
McKervey, Chem. Rev. 2015, 115, 9981-10080. b) T. Ye, M. A.
McKervey, Chem. Rev. 1994, 94, 1091-1160.
[21] a) K. M. Kadish, M. M. Morrison, Bioinorg. Chem. 1977, 7, 107-115. b)
K. M. Kadish, E. Van Caemelbecke, J. Solid State Electrochem. 2003,
7, 254-258. c) R. F. X. Williams, P. Hambright, Bioinorg. Chem. 1978, 9,
537-544. d) Y. Cui, L. Zeng, Y. Fang, J. Zhu, C. H. Devillers, D. Lucas,
N. Desbois, C. P. Gros, K. M. Kadish, ChemElectroChem 2016, 3, 228-
241. e) Y.-J. Tu, H. C. Cheng, I. Chao, C.-R. Cho, R.-J. Cheng, Y. O.
Su, J. Phys. Chem. A 2012, 116, 1632-1637. f) S. Xue, Z. Ou, L. Ye, G.
Lu, Y. Fang, X. Jiang, K. M. Kadish, Chem. Eur. J. 2015, 21, 2651-2661.
g) Y. Fang, P. Bhyrappa, Z. Ou, K. M. Kadish, Chem. Eur. J. 2014, 20,
524-532. h) K. M. Kadish, M. M. Morrison, J. Am. Chem. Soc. 1976, 98,
3326-3328.
[2]
[3]
Z. Wang, A. G. Herraiz, A. M. Del Hoyo, M. G. Suero, Nature 2018, 554,
86-91.
a) H. D. Roth, M. L. Manion, J. Am. Chem. Soc. 1975, 97, 779-783. b)
H. D. Roth, Acc. Chem. Res. 1977, 10, 86-91.
[4]
[5]
R. A. Moss, U.-H. Dolling, J. Am. Chem. Soc. 1971, 93, 951-960.
R. P. L’ꢁsperance, T. M. Ford, M. Jones, J. Am. Chem. Soc. 1988, 110,
209-213.
[6]
a) K. Rybicka-Jasińska, Ł. W. Ciszewski, D. Gryko, Adv. Synth. Catal.
2016, 358, 1671-1678. b) K. Rybicka-Jasińska, W. Shan, K. Zawada, K.
M. Kadish, D. Gryko, J. Am. Chem. Soc. 2016, 138, 15451-15458.
X. Huang, R. D. Webster, K. Harms, E. Meggers, J. Am. Chem. Soc.
2016, 138, 12636-12642.
[22] D. Sánchez, D. Bastida, J. Burés, C. Isart, O. Pineda, J. Vilarrasa, Org.
Lett. 2012, 14, 536-539.
[7]
[8]
[23] T. Bug, M. Hartnagel, C. Schlierf, H. Mayr, Chem. Eur. J. 2003, 9,
4068-4076.
a) H.-U. Reissig, R. Zimmer, Chem. Rev. 2003, 103, 1151-1196. b) M.
Yu, B. L. Pagenkopf, Tetrahedron, 2005, 61, 321-347. c) F. De Simone,
J. Waser, Synthesis 2009, 20, 3353-3374. d) F. De Nanteuil, F. De
Simone, R. Frei, F. Benfatti, E. Serrano, J. Waser, Chem. Commun.
2014, 50, 10912-10928. e) T. Schneider, J. Kaschel, D. B. Werz,
Angew. Chem. Int. Ed. 2014, 53, 5504-5523. f) M. A. Cavitt, L. H. Phun,
S. France, Chem. Soc. Rev. 2014, 43, 804-818. g) H. K. Grover, M. R.
Emmett, M. A. Kerr, Org. Biomol. Chem. 2015, 13, 655-671. h) L. K. B.
Garve, P. Barkawitz, P. G. Jones, D. B. Werz, Org. Lett. 2014, 16,
5804-5807. i) S. Das, C. G. Daniliuc, A. Studer, Org. Lett. 2016, 18,
5576-5579. j) J. Wallbaum, L. K. B. Garve, P. G. Jones, D. B. Werz,
Chem. Eur. J. 2016, 22, 18756-18759. k) J. Wallbaum, L. K. B. Garve,
P. G. Jones, D. B. Werz, Org. Lett. 2017, 19, 98-101. l) L. K. B. Garve,
P. G. Jones, D. B. Werz, Angew. Chem. Int. Ed. 2017, 56, 9226-9230.
[24] B. König, Eur. J. Org. Chem. 2017, 15, 1979-1981.
[25] Because of the irreversible electrochemical oxidation of the enamines
and the solvents used (DMSO/buffer), we do not have exact values for
the oxidation potentials in adopted in the reaction solvent
(DMSO/buffer). For acetonitrile, the voltammograms show peak
potentials about 0.49 V vs SCE for oxidation of enamines.
[26] M. A. Cismesia, T. P. Yoon, Chem. Sci. 2015, 6, 5426-5434.
[27] E. F. Petterson, T. D. Goddard, C. C. Huang, G. S. Couch, D. M.
Greenblatt, E. C. Meng, T. E. Ferrin, J. Comput. Chem. 2004, 25, 1605-
1612.
[28] S. Cabani, G. Conti, L. Lepordi, J. Chem. Trans. Faraday Soc. 1971, 67,
1933-1948.
[29] A. Osuka, K. Maruyama, J. Chem. Res. 1987, 9, 2401-2409.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800542
European Journal of Organic Chemistry
FULL PAPER
[30] H. Tomioka, M. Itoh, S. Yamakawa, Y. Izawa, J. Chem. Soc. Perkin
Trans. 2 1980, 603-609.
[34] T. Hering, A. U. Meyer, B. König, J. Org. Chem. 2016, 81, 6927-6936.
[35] a) C. Lee, W. Yang, R. G. Parr, Phys. Rev B: Condens. Matter 1988, 37,
785-789. b) A. D. Becke, J. Chem. Phys. 1993, 98, 5648-5653.
[31] Estimation due to random divergence of results – amine do not quench
poprhyrin luminescence.
[36] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215-41.
[32] P. Renaud, S. Schubert, Angew. Chem. 1990, 102, 416-417.
[33] W. W. Schoeller, J. Niemann, P. Rademacher, J. Chem. Soc. Perkin
Trans. 2 1988, 369-373.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800542
European Journal of Organic Chemistry
FULL PAPER
Layout 1:
FULL PAPER
Photoredox Catalysis
Katarzyna Rybicka-Jasińska,[a]
Katarzyna Orłowska,[a]
Maksymilian Karczewski,[a]
Katarzyna Zawada,*[b] Dorota
Gryko*[a]
Page No. – Page No.
Why Cyclopropanation is not
involved in Photoinduced α-
Alkylation of Ketones with
Diazo Compounds?
This article is protected by copyright. All rights reserved.