Harnessing Energy-Transfer in N-Centered Radical-Mediated Synthesis of Pyrrolidines

Article abstract of DOI:10.1002/ejoc.202000537

Atom transfer radical addition (ATRA), cyclization (ATRC), and polymerization (ATRP) are valuable synthetic methods for the functionalization of olefins. With the advent of photoredox catalysis, visible-light became a popular tool for the initiation of these reactions. We have developed a protocol that enables easy access to distally functionalized pyrrolidines employing blue-light mediated atom transfer radical [3+2] cyclization. The reaction is scalable, proceeds at very mild conditions. tolerates various functional groups, and provides the corresponding products in good to excellent yields. If rigid olefins are utilized as the reaction partners, the products can be isolated as single diastereomers. The mechanistic investigations provide strong support for an energy-transfer mechanism.

Full text of DOI:10.1002/ejoc.202000537

European Journal of Organic Chemistry
Accepted Article
Accepted Article
Title: Harnessing Energy-Transfer in N-centered Radical-Mediated
Synthesis of Pyrrolidines
Authors: Peter Fodran and Carl-Johan Wallentin
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.202000537
Link to VoR: https://doi.org/10.1002/ejoc.202000537
01/2020

10.1002/ejoc.202000537
European Journal of Organic Chemistry
COMMUNICATION
Harnessing Energy-Transfer in N-centered Radical-Mediated
Synthesis of Pyrrolidines
Peter Fodran,*[a] and Carl-Johan Wallentin*[a]
Dedicated to Professor R.M. Kellogg on the occasion of his 80th birthday.
[a]
Dr. P. Fodran(s), C-J Wallentin
Department of Chemistry and Molecular Biology
University of Gothenburg
Kemivägen 10
E-mail: peter.fodran@gmail.com, carl.wallentin@chem.gu.se
Supporting information for this article is given via a link at the end of the document.
Abstract: Atom transfer radical additions (ATRA), cyclizations
(ATRC), and polymerization (ATRP) are valuable synthetic methods
for the functionalization of olefins. With the advent of photoredox
catalysis, visible-light became a popular tool for the initiation of these
reactions. We have developed a protocol that enables easy access to
distally functionalized pyrrolidines employing blue-light mediated
atom-transfer radical [3+2] cyclization. The reaction is scalable,
proceeds at very mild conditions tolerates various functional groups,
and provides the corresponding products in good to excellent yields.
If rigid olefins are utilized as the reaction partners, the products can
be isolated as single diastereomers. The mechanistic investigations
provide strong support for an energy transfer mechanism.
quencher. In a pioneering report the group of Melchiorre reported
that 20 mol% of p-anisaldehyde, together with a stoichiometric
amount of 2,6-lutidine, constitute a competent system for an
energy transfer-based ATRA between electron-deficient
haloalkanes and electron-rich alkenes.[21]
Introduction
Atom transfer radical additions (ATRA), cyclizations (ATRC),[1]
and polymerization (ATRP)[2] are valuable synthetic methods for
the functionalization of olefins.[3] Since the introduction in 1945 by
Kharasch, Jensen, and Urry,[1] ATRA underwent several notable
improvements and modifications. Most recently, the groups of
[5]
Stephenson[4]
and Reiser[6] independently reported that
Figure 1. Developments in ATRA. 1A) Initial development
by Kharash and Photoredox-catalyzed ATRA developed by
Stephenson. 1B) Energy transfer ATRA/ATRC reported in this
work.
photoredox catalysis is a viable activation mode for atom-transfer
transformations. (Figure 1A). Currently, these initiation protocols
are broadly exploited in numerous photoredox mediated
processes.[3, 7] [8] [9] [10] [11-12] [13] [14]
.
In contrast to the electron-transfer ATRA, energy-transfer [15]
[16-17] [18] [19] [20] based ATRA received significantly less attention.
This discrepancy can be explained by the difference in the
physical data required for the rational design of the corresponding
processes. While the collection of electrochemical data and
construction of the Stern-Volmer plots of the reagents required for
Strain-release promoted processes are frequently exploited in
radical chemistry. For example, the well-documented behavior of
substituted cyclopropyl methyl radicals serves as a radical
clock[22] and as verification of intermediacy of radicals. The
corresponding heterosubstituted small rings as aziridines[23-24]
and oxiranes[25] have received less attention in the context of
radical chemistry. Aziridines have been applied in photoredox
electron-transfer
processes
are
straightforward,
the
catalysis, but mainly in the generation of C-centered radicals [26]
,
photoacoustic calorimetry or transient absorption spectroscopy,
which are necessary for obtaining triplet states energy (TSE)
necessary for the Dexter-type energy-transfer are not commonly
found in the organic laboratories. This downside is partially
remedied by an approximation, which utilizes the triplet-state
energy of photosensitizers and bond-dissociation energy of the
which could be harnessed in various transformations. However,
the application of aziridines in the generation of heteroatom-
centered radicals is rare and typically limited to intramolecular
reactions.[27]
1
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000537
European Journal of Organic Chemistry
COMMUNICATION
Taguchi has reported an interesting ATRA/ATRC[28-29]
,
just 24 hours. Both increase (entry 6) and decrease (entry 7) in
the concentration was detrimental to the yield. An increase to 1.0
M led to a drop in the yield to 45% and incomplete, 50%
conversion. A decrease in concentration to 0.1 M led to a
decrease in the reaction rate, with incomplete, 84% conversion
after 48 h
where the authors utilized aziridinylmethyl iodides as precursors
of azahomoallyl radicals in the synthesis of pyrrolidines. The
optimized conditions required either relatively high loading
(20 mol%) of toxic (Bu3Sn)2 or superstoichiometric amounts of
Et3B (1.5 equiv) in oxygen atmosphere. These reaction conditions
demonstrated a relatively narrow scope, moderate yields, and the
authors reported explosions. Given that radical chemistry without
the use of tin was proclaimed as one of the key green chemistry
areas [30], we aimed to develop a practical tin-free [3+2] approach
to pyrrolidines.
With the optimized conditions established, we explored the
scope of the reaction (Figure 2). Given that the substituents on
the N-arylsulfonyl groups can modulate the electrophilicity of the
generated N-centered radicals (see supporting information), we
first examined their influence on the reactivity. In general, N-
arylsulfonyl
groups
substituted
with
electron-donating
substituents, e.g., 4-methyl and 4-methoxy, provided the [3+2]
adducts 3 and 4 in the highest isolated yields – 87% and 85%,
respectively. Unsubstituted N-arylsulfonyl such as phenyl and
fused polycyclic aromates such as 2-naphthyl provided the
corresponding products 5 and 6 in lower yields – 65% and 66%,
respectively. Substitution of the N-arylsulfonyl group with
electron-withdrawing groups such as 4-bromo and 4-cyano led to
lower but still synthetically useful yields of the corresponding
cycloadducts 7 and 8 – 58% and 50%, respectively. The aryl
group can also bear two substituents — the 2,4-dimethoxyphenyl
sulfonyl substituted aziridine, provided 9 in 54% yield. The
arylsulfonyl group can be replaced with an alkylsulfonyl group.
The corresponding mesyl-substituted aziridine provides the
desired product 10 in 63% yield, albeit with longer reaction time
(72 h). In summary, we observed that the N-substitution
influences the yield, but it does not have any effect on the
diastereoselectivity.
Results and Discussion
Our initial efforts towards the development of visible-light
mediated [3+2]-cycloaddition protocol involved tosyl iodomethyl
aziridine 1 and ethyl vinyl ether as model substrates (Table 1). A
comprehensive screening of photocatalysts, solvents, and
stoichiometry of the reagents (see supporting information)
identified Ir(ppy)2dtbpy.BF4 (2.0 mol%) as the competent catalyst
and CH3CN (0.5 M) as the optimal solvent. These conditions
yielded the corresponding pyrrolidine derivative in 96% (approx.
1:1 d.r.) upon 36 hours of blue-light (460 nm) irradiation (entry 1).
Control experiments (entries 2 and 3) confirmed that light and
photocatalyst are essential for the reaction. In their absence, we
did not observe any formation of products and no conversion of 1.
Table 1. Optimization of the conditions and control experiments
Next, we focused our attention on the scope of aziridines,
which were obtained by modification of the existing methods.[31-
32]
We evaluated the reactivity of aziridines together with the t-
butyl vinyl ether because the corresponding products display
singlets in the 1H NMR. Thus, the evaluation of the diastereomeric
ratios is convenient and precise. The substitution patterns of the
aziridines can easily be recognized in the corresponding
pyrrolidine products. The substituent introduced to position 1
(arbitrary numbering, see Figure 2A) is reflected as an additional
exocyclic stereogenic center, as evident in 11. All four
diastereomers of 11 were detected in the reaction mixture in the
ratio 41:23:18:18. The introduction of a methyl substituent at
position 3 also leads to the formation of all four diastereomers of
12 in the ratio 31:31:24:14 and 52% yield. When increasing the
steric demand of the substituent going from methyl to isopropyl as
represented by product 13, changes the ratio to 53:36:6:5. While
the diastereomers were inseparable, their structures can be
assigned on several assumptions, observations and experiments.
First of all, the reaction proceeds via the early Beckwith-Houk type
Entr
y
Deviation from the optimized
conditions[a]
Conversion 1[b]
yield 2a + 2b[b,c]
1
2
3
4
No deviation
No light
100%
0%
96%
0%
No catalyst
0%
0%
5.0 equiv of nbutyl vinyl ether
as reaction partner[d]
100%
96% 3a + 3b
5
2.0 equiv. of nbutyl vinyl
ether [d]
100%
95% 3a + 3b
6
7
1.0 M concentration
0.1 M concentration
50%
84%
45%
75%
transition state, where the substituents try to adopt
[a] Optimized conditions: 50 µmol scale with Ir(ppy)2dtbpy.BF4 (2.0 mol%),
5.0 equiv. of n-Butyl vinyl ether irradiated for 24-36 h in CH3CN (0.5 M), [b]
determined by 1H NMR using dimethyl sulfone as the internal standard, [c]
combined yield of both stereoisomers [d] 24 hours reaction time
[34]
pseudoequatorial conformation.[33]
In this conformation, the
A1,3 strain in the aziridine-derived portion (purple) of the molecule
is minimal, and the t-butoxy group is anti to the sulfonamide to
minimize the torsional strain in the vinyl ether derived portion
(green) of the molecule. Given that the transition-state is early,
the eclipsing interaction between the t-butoxy substituent and the
olefinic moiety is insignificant. Elimination of HI with DBU (1.0
equiv) in DMF from 12 and 13 provides the corresponding 3-
methylenepyrrolidines in identical 40:60 cis:trans ratio, which
roughly corresponds to the syn: anti conformational preference of
the t-butoxy substituent in the ring-closing event. This preference
Given the low boiling point of ethyl vinyl ether (bp 33 °C), it is
expected that a considerable amount of this reagent will be
allocated to the headspace of the reaction vessel. Consequently,
we reasoned that a vinyl ether with a higher boiling point would
allow us to decrease its stoichiometry. In line with this argument,
we subjected the less volatile n-butyl vinyl ether (bp 94 °C) to the
reaction conditions, which indeed provided the product in an
excellent 96% yield using only 2 equivalents (entry 5) of olefin in
2
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000537
European Journal of Organic Chemistry
COMMUNICATION
is further corroborated by the bicyclic example 14 (Figure 2B),
where the aziridine-derived portion of the molecule is rigid,
resulting in two stereoisomers of 14 also obtained in 60:40 d.r.
Figure 2. Scope of photoredox catalyzed [3+2] cycloaddition. A) Scope in N-
arylsulfonyl groups. B) Scope in aziridines. Conditions[a] 200 µmol scale with
Next, we explored 2-substituted vinyl ethers. The
commercially available 2-methoxyprop-2-ene provided the
corresponding tertiary ether 20 in a reasonable 68% with 55:45 d.r.
Readily available[37] silyl ketene acetal 1-ethoxy-1-(TBS)ethene
provided the cycloadduct 21 in 58% yield with a d.r. of 60:40. The
homologous E-silyl ketene acetal and Z- silyl ketene acetal
provide product 22 in identical d.r. and almost identical yields
(48% vs. 52% yield), which suggests that this [3+2] addition is a
stepwise process. Although silyl ketene acetals are rarely used in
radical reactions, they are suitable partners in our [3+2] conditions
[38]
,
thus providing
a
smooth entry into valuable 3-
oxopyrrolidines.[39] [40]
Ir(ppy)2dtbpy.BF4 (2 mol%), 2.0 equiv of n-Butyl vinyl ether irradiated for 48 h in
CH3CN (0.5 M), yield refer to isolated products; [b] irradiated for 72 h; [c] 5.0
equiv of olefin was used.
The influence of the A1,3 strain is notable in the proportion of the
major diastereomers in 12 and 13. The increase of the steric
demand in going from methyl to isopropyl leads to an increase in
the proportion of the presumed all-pseudoequatorial major
stereoisomer (as depicted in Figure 2) from 31% to 53%. Worth
mentioning is that since 13 is a derivative of L-valine, our protocol
can be utilized in the synthesis of enantiomerically pure, highly
substituted pyrrolidines.
Figure 3. Scope of photoredox catalyzed [3+2] cycloaddition. A) Scope in
olefins. B) Unsuccessful substrates. Conditions: 200 µmol scale with
Ir(ppy)2dtbpy.BF4 (2 mol%), 3.0 equiv of vinyl ether irradiated for 48 h in CH3CN
(0.5 M) yield refer to isolated products; [b] irradiated for 72 h; [c] 5.0 equiv of
olefin was used
A methyl substituent at position 2 of the aziridines leads to
the formation of 15 bearing an all-carbon quaternary stereocenter
in 58% yield with a 50:50 d.r. Extension of the alkyl substituent at
the 3-position to n-butyl leads to a competing intramolecular 1,5-
hydrogen atom abstraction of the nitrogen-centered radical, which
ultimately suppresses the formation 16.
We also explored cyclic olefins as possible reaction partners.
Commercially available 2,3-dihydrofurane yielded a medicinally
relevant[41] bicyclic product 23 as a single stereoisomer, which
demonstrates that our protocol provides access to tri-substituted
pyrrolidines in synthetically useful yields and stereoselectivities if
the olefinic partner is rigid. 1D and 2D NMR spectroscopy
confirmed the stereochemistry of 23, which is in agreement with
the stereoelectronic requirements of radical bicyclizations.[42]
Alkenes such as methylene cyclohexane participate in the
reaction affording spirocyclic pyrrolidone 24 in relatively modest
yield (35%) after an increase in the catalyst loading (7.5 mol%).
Further expanding the scope beyond vinyl ethers, we next
pursued a variety of other olefinic substrates. To our dismay,
substrates such as vinyl thioethers 25, enamines 26, vinyl silanes
27, and styrenes 28 (Figure 3B) did not afford any [3+2]
cycloadducts. However, many of these substrates were reported
unreactive towards the addition of N-sulfonamidyl radicals.[43]
In addition to investigating the scope using a substrate
derivative library, Glorius and co-workers recently formalized
means to assess the robustness of novel methods by evaluating
the outcome of a reaction in the presence of various additives. [44]
Accordingly, we selected and subjected ten different additives
(1.0 equivalent) to the optimized reaction conditions.
Subsequently, we followed the yield of 17 and the recovery of the
additive. The results are summarized in Figure 4. The reaction
conditions tolerate esters (ethyl benzoate), diaryl sulfides
(diphenyl sulfide), sulfonamides (tosylamide), electron-poor
heterocycles (pyrimidine) and halogenated anisoles. In these
After exploring the scope in aziridines, we turned our
attention to the evaluation of different olefins (Figure 3A). Ethyl
vinyl ether and t-butyl vinyl ether provided the corresponding
products 2 and 17 in similar yields (83% vs. 76%) with a
diastereomeric ratio of 50:50 and 60:40 respectively. In the case
of 17, the sterically more hindered cis-isomer is the major isomer,
which is again consistent with the Beckwith-Houk model.[34] [33] [35]
It is noteworthy that in a related [3+2] cycloaddition reported by
Tsuritani, Shinokubo, and Oshima[36], the authors reported a
68:32 d.r. in a similar reaction albeit with a considerable lower
yield of 28%. The cyclohexyl substituted vinyl ether provided 18
in 55% yield and 50:50 d.r. The halogen-substituted (2-
chloroethoxy)ethene is also a viable substrate providing 19 in
moderate 41% yield and 50:50 d.r., illustrating that the reaction
conditions tolerate potential synthetic handles on the vinyl ether.
Comparing various steric demands of the alkyl groups of vinyl
ethers (3, 17-19), it is evident that the bulk of the alkoxy group
effects the yield and has
a modest influence on the
diastereoselectivity of the reaction. More sterically demanding
groups (larger A-values) on the oxygen atom provide the
pyrrolidines in lower yields, but in slightly higher d.r.
3
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000537
European Journal of Organic Chemistry
COMMUNICATION
cases, the additive did not affect the yields, and the recovery of
additive was >90% in all cases. In the presence of decanal, the
yield of 17 decreased to 79%, but the recovery of the additive was
complete (100%). In the case of benzyl alcohol, the yield of 17
was 75% yield, but the recovery of benzyl alcohol was only 47%.
It is noteworthy that a mixed benzyl-t-butyl acetal formed as a by-
product in significant amounts (50%). Submission of 4-t-
butylphenylacetylene in the optimized conditions resulted in a
decrease of the yield of 17 to 48%, but a significant amount (75%)
of the additive could be recovered. Electron-rich heterocycles as
BOC protected indole and dimethylaniline result in a further
decrease in the yield. Submitting N-Boc-indol led to a 25% yield
of 17 and mediocre recovery (52%). N, N-dimethylaniline halts the
reaction completely, but almost all of it (90%) can be recovered.
which presumably comes from ring-opening and subsequent HAT
of the azahomoallyl radical. These results suggest that the
formation of product in the reaction vessel has a non-productive
parasitic effect on the overall reaction.
Next, we evaluated the reduction potentials of aziridinyl
iodide and pyrrolidine, and we found that both compounds display
irreversible two-electron reductions. The reduction of
aziridinylmethyl iodide 1 occurs at -1.88 V vs. SCE (Saturated
Calomel Electrode) and the reduction of the products 2a and 2b
at even lower -2.51 V vs. SCE. Given that the reduction potential
of [Ir(ppy)2dtbpy]+ is E1/2[*IrIII/IrIV]=-0.96 V vs. SCE in CH3CN[45], a
possible redox process would be endergonic by more than 21.0
kcal mol-1.
Another observation that indicates the absence of redox
events comes from a careful analysis of the reaction mixture. If
the reaction was triggered by a single electron reduction of
aziridinylmethyl iodide, the oxidatively quenched [Ir(ppy)2dtbpy]2+
would have a sufficient reduction potential Ered[IrIII/IrIV] = +1.21 V
vs. SCE in CH3CN[46] to oxidize the primary radical
Figure 4 Robustness of the reaction.
(E(1/2)(prim°/prim+)=0.99
V vs. SCE) to the corresponding
carbocation, which would lead to the formation of a Ritter
product.[47] However, we did not detect any Ritter products even
after an independent synthesis of the corresponding compound
(see supporting information) and spiked them into the crude
reaction mixture.
The quantum yield of the reaction determined by the
procedure reported by Cismesia and Yoon[48] is Φ[3+2]=6.5. A value
higher than one is evidence for a productive chain propagation
mechanism.
After considering the electrochemical, photophysical, and
chemical experiments, we can rule out that a redox process is
operating in this [3+2] cycloaddition, and we propose (Figure 5)
that the reaction is initiated by an energy-transfer process and
mainly propagated by iodine transfer. The initial step is the
excitation of Ir(ppy)2dtbpy+ by a visible-light photon. Upon collision,
the energy is transferred via a Dexter mechanism to the
aziridinylmethyl iodide 1, which undergoes homolysis affording
iodine radical and aziridinyl methyl radical II, which in turn
undergoes a rapid ring opening providing azahomoallyl radical III.
Radical III adds to the olefin, providing a translocated radical IV,
which undertakes a 5-endo-trig cyclization yielding V.
Conditions[a] 200 µmol scale with Ir(ppy)2dtbpy.BF4 (2 mol%), 2.0 equiv of t-
Butyl vinyl ether and 1.0 equiv of additive irradiated for 48 h in CH3CN (0.5 M).
Lastly, we explored the scalability of the reaction
(Figure 3A). The reaction of 1 with t-butyl vinyl ether could be
scaled to 2.0 mmol with a slight increase in the yield of 17
(0.2 mmol – 76%, 2.0 mmol 84%). With 2,3-dihydrofuran, the
reaction could be scaled-up to gram scale (3.0 mmol), together
with a slight decrease of the catalyst loading (1.5 mol%) while
providing 24 without any notable deterioration of the yield (56%).
Mechanism
In order to elucidate the mechanism, we first determined by
Stern-Volmer titration (see supporting information) which
components of the reaction mixture quench the excited
[Ir(ppy)2dtbpy]+. We observed that the vinyl ether does not quench
the photocatalysts excited state, while aziridinylmethyl iodide
does with a rather low k =422 M-1s-1. Furthermore, the Stern-
1
q
Volmer titrations revealed that product 18 is also a quencher, with
more than two times larger quenching constant kq18=1023 M-1 s-1.
Accordingly, we evaluated the possibility that the product
participates in
a productive auto-catalytic process. After
subjecting a mixture of cis and trans 17 to the reaction conditions,
we did not observe any increase in the rate of the [3+2]
cycloaddition. Irradiation of 17, together with the catalyst and
TEMPO, did not lead to the corresponding adducts, suggesting
that even if the C-I bond is homolyzed, the recombination of the
generated radical is faster than a potential escape from the
solvent cage. Subjecting TEMPO to the reaction conditions did
not afford any 17, but instead yielded 20% of N-allyltosylamide,
Figure 5 – Proposed mechanism
Radical V can recombine with the iodine radical, which leads to
termination of the chain process or it can abstract iodine[49] from
1 thus propagating the reaction.
4
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000537
European Journal of Organic Chemistry
COMMUNICATION
Subjecting commonly utilized photoredox catalysts[46] to the
optimized conditions did not reveal any trend between their triplet
state energy (TSE) and yield of the [3+2] product 17 (Table 2).
Ir(dF-CF3-ppy)2dtbpy+ with TSE of 60.1 kcal.mol-1[46] is frequently
Keywords: energy-transfer • photocatalysis • ATRA • ATRC •
pyrrolidines
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
M. S. Kharasch, E. V. Jensen, W. H. Urry, Science 1945,
102, 128.
J.-S. Wang, K. Matyjaszewski, J. Am. Chem. Soc. 1995,
117, 5614-5615.
T. Courant, G. Masson, J. Org. Chem. 2016, 81, 6945-
6952.
J. D. Nguyen, E. M. D'Amato, J. M. R. Narayanam, C. R.
J. Stephenson, Nature Chemistry 2012, 4, 854-859.
C.-J. Wallentin, J. D. Nguyen, P. Finkbeiner, C. R. J.
Stephenson, J. Am. Chem. Soc. 2012, 134, 8875-8884.
M. Pirtsch, S. Paria, T. Matsuno, H. Isobe, O. Reiser,
Chem. Eur. J. 2012, 18, 7336-7340.
T. Rawner, E. Lutsker, C. A. Kaiser, O. Reiser, ACS
Catalysis 2018, 8, 3950-3956.
M. Knorn, T. Rawner, R. Czerwieniec, O. Reiser, ACS
Catalysis 2015, 5, 5186-5193.
Q. Qin, D. Ren, S. Yu, Org. Biomol. Chem. 2015, 13,
10295-10298.
N. Iqbal, J. Jung, S. Park, E. J. Cho, Angew. Chem. Int.
Ed. 2014, 53, 539-542.
X. Sun, Y. He, S. Yu, J. Photochem. Photobiol. A 2018,
355, 326-331.
utilized in energy-transfer processes. In our [3+2] cycloaddition,
[39]
this catalyst provided 17 in 54% yield. Ir(ppy)3
is another
catalyst with relatively high TSE of 55.8 kcal.mol-1but it provided
17 in only 11% yield. Ru(bpy)3 with TSE of 45.0 kcal.mol-1
2+
provided 17 in 44% yield and Eosin Y[50] with slightly lower TSE of
43.6 kcal.mol-1 provided the products in a 9% yield. Ir(ppy)2dtbpy+,
which was identified as the optimal catalyst, provided the highest,
100% yield in the given set of experiments. These results illustrate
that similar processes initiated via energy transfer are more
dependent on matched energy levels between reaction partners
rather than the TSE of the photosensitizer and BDE of the
homolyzed bond.
Table 2. Influence of TSE on the yield of the [3+2] cycloaddition.
Entry[a]
catalyst
TSE
(kcal.mol-1)
yield of 17a
1
2
3
4
5
Ir(dF-CF3-ppy)2dtbpy.PF6
Ir(ppy)3
60.1
55.8[46]
49.3
54%
11%
100%
44%
9%
[12]
[13]
T. M. Williams, C. R. J. Stephenson.
X. Gu, X. Li, Y. Qu, Q. Yang, P. Li, Y. Yao, Chem. Eur. J.
2013, 19, 11878-11882.
Ir(ppy)2dtbpy.BF4
Ru(bpy)3.(PF6)2
Eosin Y
[14]
T. M. Williams, C. R. J. Stephenson, in Visible Light
Photocatalysis in Organic Chemistry (Eds.: C. R. J.
Stephenson, T. P. Yoon, D. W. C. MacMillan), Wiley-VCH,
2018, pp. 73-92.
45.0
43.6[50]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
F. Strieth-Kalthoff, M. J. James, M. Teders, L. Pitzer, F.
Glorius, Chem. Soc. Rev. 2018, 47, 7190-7202.
K. L. Skubi, T. R. Blum, T. P. Yoon, Chem. Rev. 2016,
116, 10035-10074.
C. Michelin, N. Hoffmann, ACS Catalysis 2018, 8, 12046-
12055.
A. E. Hurtley, Z. Lu, T. P. Yoon, Angew. Chem. Int. Ed.
2014, 53, 8991-8994.
E. R. Welin, C. Le, D. M. Arias-Rotondo, J. K. McCusker,
D. W. C. MacMillan, Science 2017, 355, 380.
T. R. Blum, Z. D. Miller, D. M. Bates, I. A. Guzei, T. P.
Yoon, Science 2016, 354, 1391.
E. Arceo, E. Montroni, P. Melchiorre, Angew. Chem. Int.
Ed. 2014, 53, 12064-12068.
D. Griller, K. U. Ingold, Acc. Chem. Res. 1980, 13, 317-
323.
[a] determined by calibrated GC/MS analysis using dimethyl sulfone as the
internal standard. [a] Conditions 200 µmol scale with catalyst (2 mol%), 2.0
equiv of t-Butyl vinyl ether irradiated for 48 h in CH3CN (0.5 M);
Conclusion
In conclusion, we have developed a protocol for the formal
[3+2] cycloaddition between aziridinylmethyl iodides and electron-
rich olefins. The conditions grant access to all substitution
patterns of trisubstituted pyrrolidines. While control of
stereochemistry is challenging, we observed that the substitution
pattern for the olefins plays a more important role than that of the
aziridines. In general, the reaction conditions tolerate a variety of
functional groups, and the corresponding products are isolated in
useful chemical yields. We have also demonstrated that the
energy transfer triggers the initiation of the reaction and that the
operating mechanism is a chain transfer process.
[23]
[24]
D. J. Pasto, J. Org. Chem. 1996, 61, 252-256.
B. A. Marples, R. C. Toon, Tetrahedron Lett. 1999, 40,
4873-4876.
[25]
[26]
N. Prévost, M. Shipman, Org. Lett. 2001, 3, 2383-2385.
M.-H. Larraufie, R. Pellet, L. Fensterbank, J.-P. Goddard,
E. Lacôte, M. Malacria, C. Ollivier, Angew. Chem. Int. Ed.
2011, 50, 4463-4466.
[27]
[28]
[29]
[30]
W. R. Bowman, M. J. Broadhurst, D. R. Coghlan, K. A.
Lewis, Tetrahedron Lett. 1997, 38, 6301-6304.
O. Kitagawa, Y. Yamada, H. Fujiwara, T. Taguchi, Angew.
Chem. Int. Ed. 2001, 40, 3865-3867.
O. Kitagawa, S. Miyaji, Y. Yamada, H. Fujiwara, T.
Taguchi, J. Org. Chem. 2003, 68, 3184-3189.
D. J. C. Constable, P. J. Dunn, J. D. Hayler, G. R.
Humphrey, J. J. L. Leazer, R. J. Linderman, K. Lorenz, J.
Manley, B. A. Pearlman, A. Wells, A. Zaks, T. Y. Zhang,
Green Chemistry 2007, 9, 411-420.
Acknowledgements
The authors would like to acknowledge the European
Commission for a Marie-Curie individual fellowship to P. F. (grant
number 799943 SUPER) and the Swedish research council for a
grant to C.J.W (grant number 2018-04871). Professor Elisabeth
Ahlberg is gratefully acknowledged for fruitful discussions on
electrochemistry. Professor Karl Börjesson and Manuel Hertzog
are acknowledged for discussions on energy transfer
mechanisms and assistance with quenching assays, and Dr.
Leticia Monjas is acknowledged for proof-reading of the
manuscript.
[31]
[32]
[33]
[34]
O. Kitagawa, T. Suzuki, T. Taguchi, J. Org. Chem. 1998,
63, 4842-4845.
Y. Morino, I. Hidaka, Y. Oderaotoshi, M. Komatsu, S.
Minakata, Tetrahedron 2006, 62, 12247-12251.
A. L. J. Beckwith, C. J. Easton, A. K. Serelis, J. Chem.
Soc., Chem. Commun. 1980, 482-483.
A. L. J. Beckwith, T. Lawrence, A. K. Serelis, J. Chem.
Soc., Chem. Commun. 1980, 484-485.
5
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000537
European Journal of Organic Chemistry
COMMUNICATION
[35]
[36]
[37]
[38]
[39]
[40]
D. C. Spellmeyer, K. N. Houk, J. Org. Chem. 1987, 52,
959-974.
T. Tsuritani, H. Shinokubo, K. Oshima, Org. Lett. 2001, 3,
2709-2711.
R. E. Ireland, R. H. Mueller, A. K. Willard, J. Am. Chem.
Soc. 1976, 98, 2868-2877.
P. V. Pham, D. A. Nagib, D. W. C. MacMillan, Angew.
Chem. Int. Ed. 2011, 50, 6119-6122.
K. Lv, M.-L. Liu, L.-S. Feng, L.-Y. Sun, Y.-X. Sun, Z.-Q.
Wei, H.-Q. Guo, Eur. J. Med. Chem. 2012, 47, 619 - 625.
M.-L. Liu, L.-S. Feng, K. Lv, Y. Chai, S. Wang, J. Cao, H.-
Y. Guo, Asian J. Chem. 2013, 25, 2327 - 2328.
M. Quibell, J. P. Watts, Vol. PCT/GB2007/002624, 2008.
D. P. Curranl, D. M. Rakiewicz, Tetrahedron 1985, 41,
3943-3958.
[41]
[42]
[43]
[44]
[45]
Q. Zhu, D. E. Graff, R. R. Knowles, J. Am. Chem. Soc.
2018, 140, 741-747.
K. D. Collins, F. Glorius, Nature Chemistry 2013, 5, 597-
601.
M. S. Lowry, J. I. Goldsmith, J. D. Slinker, R. Rohl, R. A.
Pascal, G. G. Malliaras, S. Bernhard, Chem. Mater. 2005,
17, 5712-5719.
[46]
A. Singh, K. Teegardin, M. Kelly, K. S. Prasad, S.
Krishnan, J. D. Weaver, J. Organomet. Chem. 2015, 776,
51-59.
[47]
[48]
[49]
[50]
D. Prasad Hari, T. Hering, B. König, Angew. Chem. Int.
Ed. 2014, 53, 725-728.
M. A. Cismesia, T. P. Yoon, Chemical Science 2015, 6,
5426-5434.
D. P. Curran, E. Bosch, J. Kaplan, M. Newcomb, J. Org.
Chem. 1989, 54, 1826-1831.
M. Majek, F. Filace, A. J. v. Wangelin, Beilstein Journal of
Organic Chemistry 2014, 10, 981-989.
6
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000537
European Journal of Organic Chemistry
COMMUNICATION
Entry for the Table of Contents
Energy Transfer provides a route to nitrogen centered radicals, which can be harnessed in a formal [3+2] cycloaddition. The
corresponding trisubstituted pyrrolidines are obtained in moderate to high yields.
7
This article is protected by copyright. All rights reserved.