Aminoalkyl radicals as halogen-atom transfer agents for activation of alkyl and aryl halides

Article abstract of DOI:10.1126/science.aba2419

Organic halides are important building blocks in synthesis, but their use in (photo)redox chemistry is limited by their low reduction potentials. Halogen-atom transfer remains the most reliable approach to exploit these substrates in radical processes despite its requirement for hazardous reagents and initiators such as tributyltin hydride. In this study, we demonstrate that a-aminoalkyl radicals, easily accessible from simple amines, promote the homolytic activation of carbon-halogen bonds with a reactivity profile mirroring that of classical tin radicals. This strategy conveniently engages alkyl and aryl halides in a wide range of redox transformations to construct sp³-sp³, sp³-sp², and sp²-sp² carbon-carbon bonds under mild conditions with high chemoselectivity.

Full text of DOI:10.1126/science.aba2419

RESEARCH
ORGANIC CHEMISTRY
metallaphotoredox catalysis to overcome slug-
gish carbon-halogen oxidative additions with
transition metals (16, 17).
Aminoalkyl radicals as halogen-atom transfer agents
for activation of alkyl and aryl halides
We questioned whether a-aminoalkyl radi-
cals could serve as a distinct class of halogen-
abstracting reagents (Fig. 1B). Our idea for this
reactivity stemmed from the fact that although
classical XAT processes benefit from the for-
mation of strong halogen-tin or halogen-silicon
bonds, it is the high degree of charge transfer in
the transition state that facilitates halogen-
atom abstraction by these nucleophilic radi-
cals (18). We therefore reasoned that strongly
nucleophilic a-aminoalkyl radicals might ben-
efit from related kinetic polar effects and man-
ifest the same reactivity. Such radicals can be
easily generated from simple amines, a class
of abundant and inexpensive reagents that
would offer ample opportunity for fine steric
and electronic tuning.
Here we report the successful realization of
this concept and its implementation as part of a
mild and general strategy for the engagement
of unactivated alkyl and aryl halides in redox
chemistry (Fig. 1C). Because a-aminoalkyl radi-
cals display a reactivity profile similar to that
of tin radicals, their capacity to abstract iodine
and bromine atoms has enabled the develop-
ment of deuteration, cross-electrophile cou-
pling, Heck-type olefination, and aromatic
C–H alkylation protocols.
Timothée Constantin1, Margherita Zanini1, Alessio Regni1, Nadeem S. Sheikh2,
Fabio Juliá1*, Daniele Leonori1*
Organic halides are important building blocks in synthesis, but their use in (photo)redox chemistry
is limited by their low reduction potentials. Halogen-atom transfer remains the most reliable approach
to exploit these substrates in radical processes despite its requirement for hazardous reagents and
initiators such as tributyltin hydride. In this study, we demonstrate that a-aminoalkyl radicals, easily
accessible from simple amines, promote the homolytic activation of carbon-halogen bonds with a
reactivity profile mirroring that of classical tin radicals. This strategy conveniently engages alkyl and
aryl halides in a wide range of redox transformations to construct sp3-sp3, sp3-sp2, and sp2-sp2
carbon-carbon bonds under mild conditions with high chemoselectivity.
arbon radicals are versatile synthetic in-
termediates central to the preparation of
high-value compounds (1, 2). The advent
of visible-light photoredox catalysis (3)
has offered a broadly applicable radical
tems (11, 12) (Fig. 1A). Furthermore, the mech-
anisms involved in photoredox reactions are
often uncertain (9), displaying large redox mis-
matches (>1 V) for SET activation and thus
thwarting the exploitation of the carbon radi-
cals accessed in this manner.
C
generation strategy, transforming a variety of
redox-active precursors into open-shell inter-
mediates by single-electron transfer (SET) and
fragmentation (4–6). However, photoredox ac-
tivation has thus far rarely extended to organic
halides, one of the largest classes of building
blocks available to organic chemists. The cur-
rent synthetic gap is especially evident in the
case of unactivated alkyl halides, for which
only dehalogenation and intramolecular cycli-
zation of iodides have been reported (7–10).
The difficulties in engaging these feedstocks in
redox chemistry arise from their highly nega-
tive reduction potentials [Ered < –2 V versus
saturated calomel electrode (SCE) for unac-
tivated alkyl and aryl iodides], which in turn
necessitate the use of strongly reducing sys-
This lack of synthetic applicability stands in
stark contrast to the fundamental role alkyl
and aryl halides have played in the develop-
ment of radical chemistry. Methods based on
tin or silicon reagents and trialkylborane–O2
systems have proven to be highly reliable in
accessing carbon radicals from organic hal-
ides, generating the open-shell intermediate
by homolytic carbon-halogen bond cleavage via
halogen-atom transfer (XAT) (13–15). However,
the toxic, hazardous nature of these reagents
and initiators is problematic and has been one
of the main drivers toward the identification
of alternative precursors and chemical strate-
gies for carbon radical generation. Neverthe-
less, silicon radicals have been recently used in
We initiated our study by evaluating the
iodine-atom transfer reaction from cyclohexyl
iodide 2 to the a-aminoalkyl radical I-a, de-
rived from triethylamine (Et3N, 1a) (Fig. 2A).
1Department of Chemistry, University of Manchester,
Manchester M13 9PL, UK. 2Department of Chemistry, College of
Science, King Faisal University, Al-Ahsa 31982, Saudi Arabia.
*Corresponding author. Email: fabio.juliahernandez@manchester.
ac.uk (F.J.); daniele.leonori@manchester.ac.uk (D.L.)
Fig. 1. Homolysis of carbon-halogen bonds by a-aminoalkyl radicals. (A) Activation modes for the generation of carbon radicals from alkyl and aryl halides. e–,
electron. (B) Nucleophilic a-aminoalkyl radicals abstract halogen atoms (X) through polarized transition states, in analogy to tin and silicon radicals. Me, methyl;
Bu, butyl; R, alkyl, aryl. (C) Outline of the transformations possible using alkyl and aryl halides activated via a-aminoalkyl radical-mediated XAT. Ar, aryl.
Constantin et al., Science 367, 1021–1026 (2020)
28 February 2020
1 of 6

RESEARCH
| REPORT
Density functional theory calculations pre-
dicted this XAT to be kinetically feasible,
involving a polarized transition state with a
notable charge-transfer character (dTS = 0.42),
which supports the anticipated interplay of
polar effects. Although the XAT is only slight-
ly exothermic (19), the fast and irreversible
dissociation of the resulting a-iodoamine III-a
into the iminium iodide IV-a provides the
thermodynamic driving force to the process.
To gather direct experimental evidence, we
generated and monitored I-a using laser flash
photolysis (20, 21) and observed a noticeable
reactivity toward 2. Data analysis provided a
fast rate constant (kXAT = 3.6 108 M–1 s–1) that
is only one order of magnitude slower than
reported rates for I-abstraction by Bu3Sn• and
(Me3Si)3Si• (~109 M–1 s–1) (22), showing pro-
mise for implementation in synthetic radical
chemistry.
redox catalysis (23), triplet benzophenone (24),
and SO4 (25)] or direct H-atom transfer
•–
(HAT) [Et3N: a-N-C–H bond dissociation en-
ergy (BDE) = 91 kcal mol–1] using t-BuO• (26).
The desired product 4 was obtained in all
cases in excellent to good yields, exemplifying
the variety of conditions for a-aminoalkyl radi-
cal generation and ensuing XAT.
The proposed mechanism under photoredox
conditions is depicted in Fig. 2C. Upon blue
light irradiation, the excited organic photo-
catalyst 4CzIPN (*Ered = +1.35 V versus SCE)
oxidizes 1a, which, after subsequent deproto-
nation, furnishes the key a-aminoalkyl radi-
cal I-a. This species undergoes XAT with 3,
and the resulting alkyl radical V provides the
To explore the applicability of this strategy in
radical reactions, we chose the dehalogenation
of 4-iodo-N-Boc-piperidine 3, using Et3N as the
XAT-agent precursor and methyl thioglycolate–
H2O as the H-atom donor (Fig. 2B). At the
outset, we were particularly interested to eval-
uate whether the photochemical or thermal
modes for a-aminoalkyl radical generation could
be recruited for XAT reactivity. We therefore
began by testing four known systems based on
amine SET oxidation (Et3N: Eox = +0.77 V ver-
sus SCE) followed by deprotonation [i.e., photo-
Fig. 2. Mechanistic analysis and application to dehalogenation and
deuteration reactions. (A) Computational [B3LYP-D3/def2-TZVP] and laser
flash photolysis studies on a model XAT reaction with an alkyl iodide. Et,
ethyl; DFT, density functional theory; DG‡, Gibbs energy of activation; DG°,
dehalogenation of alkyl iodide 3. Mechanistic studies support the intermediacy
of an a-aminoalkyl radical in the activation of the C–I bond. h, Planck’s constant;
n, photon frequency; dtbbpy, 4,4′-di-tert-butyl-2,2′-dipyridyl; ppy, 2-phenylpyridyl;
dF, difluoro; bpy, 2,2′-bipyridine; Mes, mesityl; Acr, acridinium; n/d,
Gibbs free energy; lmax, wavelength of maximum absorption. (B) Evaluation of not determined. (D) Application of the XAT methodology in deuteration of
photochemical and thermal strategies for a-aminoalkyl radical generation and
their use in the dehalogenation of alkyl iodide 3. Boc, tert-butoxycarbonyl;
alkyl halides. All yields are isolated. Deuteration was determined by gas
chromatography–mass spectrometry/quantitative 13C nuclear magnetic resonance
LEDs, light-emitting diodes; r.t., room temperature; Ph, phenyl; UV, ultraviolet; spectroscopy. *Tribenzylamine 1b was used as the amine. Ac, acetyl; Bn, benzyl;
t-Bu, tert-butyl. (C) Proposed mechanism for the photoredox-based dr, diastereomeric ratio.
Constantin et al., Science 367, 1021–1026 (2020)
28 February 2020
2 of 6

RESEARCH
| REPORT
Fig. 3. Application to hydroalkylation and allylation. (A) Scope for the alkylation of alkyl iodides, alkyl bromides, and aryl iodides. *1a was used as the amine.
†1b was used as the amine. ‡1c was used as the amine. §1d was used as the amine. EWG, electron withdrawing group; quant., quantitative; pin, pinacolato; Ac, acetyl;
i-Bu, iso-butyl. (B) Tailoring XAT reactivity by modifying the a-aminoalkyl radical structure. (C) Scope for the allylation of alkyl iodides, alkyl bromides, and aryl
iodides. All yields are isolated. ¶The corresponding allyl sulfone was used. Ts, tosyl.
product 4 by favorable HAT from methyl thio-
glycolate (S–H BDE = 87 kcal mol–1). Lastly,
SET between the thiyl radical and 4CzIPN•–,
followed by protonation with H2O, regener-
ates the thiol along with the ground-state
photocatalyst. The choice of 4CzIPN and Et3N
is relevant to our mechanistic hypothesis be-
cause neither the excited nor the reduced state
cal generation is now dissected by the redox
requirements of the system, and therefore the
reductive ability of the photocatalyst is not
crucial to the outcome of the reaction. Indeed,
this process can be achieved with a diverse range
of photocatalysts, including those of limited
reductive power (e.g., Fukuzumi’s acridinium;
Ered = –0.57 V versus SCE). The replacement of
Et3N with other common electron donors
[e.g., Ph2N(PMP), sodium ascorbate, or Hantzsch
ester] suppressed the reactivity, despite all com-
pounds effectively quenching the excited pho-
tocatalyst (19). Moreover, other alkyl amines
were tested, but only those able to generate an
a-aminoalkyl radical promoted the desired re-
activity (19). These results suggest that alkyl
iodide activation via a reductive-quenching
photoredox cycle is not operative and that the
amine plays a fundamental role in the C–I
bond cleavage that goes beyond its capacity
to act as an electron donor.
The high yields obtained with the photo-
redox system, along with the use of H2O as a
stoichiometric H-atom source, prompted ex-
ploration of dehalogenation-deuteration reactions
using D2O (Fig. 2D). After optimization, we
of the photocatalyst [*Eox = –1.04 V; Ered
=
–1.21 V versus SCE (27)] or I-a [Eox = –1.12 V
versus SCE (27)] is strong enough to promote
direct SET reduction of 3 (Ered = –2.35 V
versus SCE). This means that the carbon radi-
Constantin et al., Science 367, 1021–1026 (2020)
28 February 2020
3 of 6

RESEARCH
| REPORT
Fig. 4. Application to olefinations and arylations. (A) Scope for olefination of alkyl iodides and alkyl bromides. dmg, dimethylglyoximate; DMF, dimethylformamide.
(B) Scope for the C–H alkylation and arylation of aromatics. All yields are isolated. *1c was used as the amine. †Me3N was used as the amine. ‡Bu3N was used as the amine.
§The reaction was run with 50 equiv of the arene. DMSO, dimethyl sulfoxide. Asterisks in structures indicate the position of the minor constitutional isomer.
achieved efficient deuteration of primary, sec-
ondary, and tertiary alkyl iodides in nearly
quantitative yields (5 to 12). The mild reac-
tion conditions tolerated multiple functional
groups, showcasing the strong chemoselectiv-
ity of this XAT approach. Activation of alkyl
bromides is still a challenging task in radi-
cal chemistry and is considered unfeasible
using trialkylborane–O2 systems (28). Our
a-aminoalkyl radical–based XAT strategy is
applicable to bromides, albeit in lower con-
version compared with the results obtained
for iodides.
The XAT strategy oxidatively generates car-
bon radicals from organic halides, represent-
ing an umpolung approach relative to the
natural redox requirement for SET activation
of these building blocks. We posited that the
generated radicals could therefore be used in
similar mechanistic scenarios to those involv-
ing carboxylic acids or potassium trifluorobo-
rates, allowing their modular application in net
reductive processes such as cross-electrophile
couplings (29, 30).
yields (13 to 23). A variety of functionalities—
including polar groups such as free carboxylic
acid, primary amide, pyridine, and boronic
ester—were readily accommodated. When the
same reactions were attempted using 3-bromo-
N-Boc-azetidine, no desired product was ob-
tained and a substantial amount of the adduct
arising from direct addition of I-a to the olefin
acceptor was identified (Fig. 3B). In this case,
XAT is slower owing to the stronger nature of
the C–Br bond, thus rendering the direct
Giese reaction of I-a with the acceptor
competitive [observed rate constant kobs
~ 107 M–1 s–1 (21)]. We therefore reasoned
that the modulation of the electronic and
steric properties of the a-aminoalkyl radical
could be used to tune its reactivity. Indeed,
we used tribenzylamine (1b) to restore XAT
as the favored pathway for reactions of un-
activated alkyl bromides in these hydro-
alkylations. Because the stabilized a-aminoalkyl
We explored this premise by developing
Giese-type hydroalkylation of electron-poor
olefins. Although these transformations have
been performed with the aid of nickel catal-
ysis, they typically require the use of stoi-
chiometric metal reductants (e.g., Mn0, Zn0)
or silane H-donors (31, 32). In our case, be-
cause a-aminoalkyl radicals have been used
as substrates in Giese additions (33), the suc-
cess of this strategy hinged on their capacity
to undergo XAT preferentially over their
known reaction with the olefin. Exploration
began with 3-iodo-N-Boc-azetidine in the pres-
ence of Et3N and 4CzIPN under blue light
irradiation (Fig. 3A; see fig. S10 for a pro-
posed mechanism). A diverse range of electron-
poor olefins were efficiently converted to the
corresponding products in high to excellent
Constantin et al., Science 367, 1021–1026 (2020)
28 February 2020
4 of 6

RESEARCH
| REPORT
radical I-b was essentially unreactive toward
electron-poor olefins [calculated rate constant
kcalc ~ 10−1 M–1 s–1 (21)], bromine abstraction
became possible, providing the desired prod-
ucts in good yields.
formation (via VIII) without the need for pre-
cious metals (see fig. S14 for a proposed mech-
anism). As shown in Fig. 4A, we found this
dual XAT–[Co] protocol feasible, thus allowing
direct olefination of primary, secondary, and
tertiary alkyl iodides and bromides exclusively
as the E isomers (51 to 74, with the exception
of 54 and 62). The broad functional group
compatibility was demonstrated with the suc-
cessful engagement of substrates containing
phenol, aniline, and benzoic acid moieties, as
well as aryl bromide, boronic acid, and phos-
phine groups that could limit application under
transition metal catalysis. The olefination was
also very effective in intramolecular settings,
as showcased by the construction of tricyclic
75 in good yield. Couplings with aryl iodides
were attempted but generally resulted in low
yields.
In a final effort to establish the generality
of this XAT strategy, we turned our attention
to the direct aromatic C–H alkylation via rad-
ical intermediates (Fig. 4B; see fig. S15 for a
proposed mechanism). Recently, the use of
zinc alkylsulfinates has provided a powerful
and effective solution to this synthetic chal-
lenge (41, 42). Because these reagents are often
prepared from the corresponding halides, a
methodology that directly uses these building
blocks would obviate multistep synthesis of
any reactive intermediate. In this case, how-
ever, a photoredox system for a-aminoalkyl
radical generation is difficult to implement,
owing to the mechanistic requirement of a
second oxidation after radical addition to the
arene to allow rearomatization (IX→X). The
broad set of reactivity modes for a-aminoalkyl
radical generation enabled identification of
simple thermal, net oxidative conditions for
the direct alkylation of caffeine with alkyl
iodides, without the need for light or catalysts
(76 to 80). This manifold for aromatic C–H
alkylation was compatible with the installa-
tion of primary, secondary, and tertiary alkyl
groups and could be extended to other hete-
roarenes commonly found in bioactive mol-
ecules, such as indoles and azoles as well as
benzenoids (43) (81 to 87). Furthermore, we
demonstrated that aryl iodide activation and
subsequent sp2-sp2 coupling (44) is also pos-
sible, as shown by the successful preparation
of 88 to 91.
2. S. P. Pitre, N. A. Weires, L. E. Overman, J. Am. Chem. Soc. 141,
2800–2813 (2019).
3. C. K. Prier, D. A. Rankic, D. W. C. MacMillan, Chem. Rev. 113,
5322–5363 (2013).
4. J. Xuan, Z.-G. Zhang, W.-J. Xiao, Angew. Chem. Int. Ed. 54,
15632–15641 (2015).
5. J. K. Matsui, S. B. Lang, D. R. Heitz, G. A. Molander, ACS Catal.
7, 2563–2575 (2017).
6. A. Togni et al., Angew. Chem. Int. Ed. 10.1002/anie.201911660
(2019).
7. J. D. Nguyen, E. M. D’Amato, J. M. R. Narayanam,
C. R. J. Stephenson, Nat. Chem. 4, 854–859 (2012).
8. H. Kim, C. Lee, Angew. Chem. Int. Ed. 51, 12303–12306
(2012).
9. Y. Shen, J. Cornella, F. Juliá-Hernández, R. Martin, ACS Catal.
7, 409–412 (2017).
10. D. Alpers, M. Gallhof, J. Witt, F. Hoffmann, M. Brasholz,
Angew. Chem. Int. Ed. 56, 1402–1406 (2017).
11. J. L. Kuo, C. Lorenc, J. M. Abuyuan, J. R. Norton, J. Am. Chem. Soc.
140, 4512–4516 (2018).
12. G. Nocera et al., J. Am. Chem. Soc. 140, 9751–9757
(2018).
13. W. P. Neumann, Synthesis 1987, 665–683 (1987).
14. C. Chatgilialoglu, C. Ferreri, Y. Landais, V. I. Timokhin, Chem.
Rev. 118, 6516–6572 (2018).
15. H. Yorimitsu, K. Oshima, in Radicals in Organic Synthesis,
P. Renaud, M. P. Sibi, Eds. (Wiley, 2008), pp. 11–27.
16. C. Le, T. Q. Chen, T. Liang, P. Zhang, D. W. C. MacMillan,
Science 360, 1010–1014 (2018).
17. P. Zhang, C. C. Le, D. W. C. MacMillan, J. Am. Chem. Soc. 138,
8084–8087 (2016).
We next explored the alkyl iodide scope
by using Boc-protected dehydroalanine as an
olefin acceptor, thus providing convenient ac-
cess to unnatural amino acids (24 to 35). In
this case, a variety of organyl groups bearing
common functionalities (e.g., free alcohol, alkyl
chloride, silane, and terminal alkyne) were
compatible, reflecting the mildness of the re-
action conditions. This protocol has also been
carried out at gram scale without erosion in
yield. The ability to generate primary alkyl
radicals complements approaches that use
oxalates and trifluoroborates, which are
known to experience sluggish fragmentations
(34, 35). When alkyl halides activated toward
SN2 (second-order nucleophilic displacement)
attack by Et3N (e.g., 29 and 32) were em-
ployed, the desired products were, unsurpris-
ingly, obtained in low yields. This hurdle was
addressed by adjusting the steric properties
of the XAT reagent: Efficient couplings were
achieved with the use of the bulkier amine
1,2,2,6,6-pentamethylpiperidine (PMP; 1c). We
have also been able to extend this methodol-
ogy to unactivated aryl iodides by using the
more hindered but less stabilized a-aminoalkyl
radical derived from triisobutylamine (1d).
These conditions enabled direct access to aryl
radicals by sp2 C–I bond cleavage and were
applied to the one-pot transformation of to-
sylated serine into phenylalanine derivatives
(36 to 39). Overall, these results illustrate how
the large structural diversity of available ter-
tiary amines facilitates the rational tailoring
of the a-aminoalkyl radical reactivity to ad-
dress different challenges in carbon-halogen
bond activation.
The XAT strategy for cross-electrophile cou-
pling is not restricted to electron-poor olefins.
We also achieved efficient allylation of alkyl
and aryl halides by using simple allyl chlorides
and other pseudohalides (40 to 50) (Fig. 3C;
see fig. S12 for a proposed mechanism). This
approach bypasses the conventional conver-
sion of one of the two coupling partners into a
Grignard or organozinc reagent (36) and there-
fore tolerates functionalities, such as free alco-
hol and ketone, that are often troublesome
with organometallics.
To further demonstrate the versatility of
this activation mode, we sought to adapt it to
target the use of alkyl halides in Heck-type
olefinations, a long-standing challenge in con-
ventional palladium catalysis, owing to unde-
sired b-hydride elimination (37–39). Specifically,
we questioned whether, after addition of alkyl
radicals to suitable olefins (VII), a cobaloxime
cocatalyst might trigger a dehydrogenation re-
action (40), thus leading to sp3-sp2 C–C bond
18. W. H. Tamblyn, E. A. Vogler, J. K. Kochi, J. Org. Chem. 45,
3912–3915 (1980).
19. See the supplementary materials for more information.
20. J. C. Scaiano, in Reactive Intermediate Chemistry, R. A. Moss,
M. S. Platz, M. Jones, Eds. (Wiley, 2005), pp. 847–871.
21. J. Lalevée, B. Graff, X. Allonas, J. P. Fouassier, J. Phys. Chem. A
111, 6991–6998 (2007).
22. K. U. Ingold, J. Lusztyk, J. C. Scaiano, J. Am. Chem. Soc. 106,
343–348 (1984).
23. K. Nakajima, Y. Miyake, Y. Nishibayashi, Acc. Chem. Res. 49,
1946–1956 (2016).
24. S. G. Cohen, A. Parola, G. H. Parsons, Chem. Rev. 73, 141–161
(1973).
25. J. K. Laha, K. S. S. Tummalapalli, A. Nair, N. Patel, J. Org.
Chem. 80, 11351–11359 (2015).
26. D. D. M. Wayner, J. J. Dannenberg, D. Griller, Chem. Phys. Lett.
131, 189–191 (1986).
27. J. Luo, J. Zhang, ACS Catal. 6, 873–877 (2016).
28. V. Soulard, G. Villa, D. P. Vollmar, P. Renaud, J. Am. Chem. Soc.
140, 155–158 (2018).
29. D. A. Everson, D. J. Weix, J. Org. Chem. 79, 4793–4798
(2014).
30. E. Richmond, J. Moran, Synthesis 50, 499–513 (2018).
31. K. M. M. Huihui, R. Shrestha, D. J. Weix, Org. Lett. 19, 340–343
(2017).
32. F. Zhou, J. Zhu, Y. Zhang, S. Zhu, Angew. Chem. Int. Ed. 57,
4058–4062 (2018).
33. R. A. Aycock, C. J. Pratt, N. T. Jui, ACS Catal. 8, 9115–9119
(2018).
34. C. C. Nawrat, C. R. Jamison, Y. Slutskyy, D. W. C. MacMillan,
L. E. Overman, J. Am. Chem. Soc. 137, 11270–11273
(2015).
35. C. Lévêque, L. Chenneberg, V. Corcé, C. Ollivier, L. Fensterbank,
Chem. Commun. 52, 9877–9880 (2016).
36. H. Yorimitsu, K. Oshima, Angew. Chem. Int. Ed. 44, 4435–4439
(2005).
37. M. R. Netherton, G. C. Fu, in Palladium in Organic Synthesis,
J. Tsuji, Ed. (Springer, 2005), pp. 85–108.
38. G.-Z. Wang, R. Shang, W.-M. Cheng, Y. Fu, J. Am. Chem. Soc.
139, 18307–18312 (2017).
The results presented here demonstrate that
alkyl and aryl halides can be converted to
carbon radicals by XAT using a-aminoalkyl
radicals. We believe that the broad scope, func-
tional group tolerance, and modularity of this
approach for carbon-halogen bond activation
will likely be of great utility to chemists in both
academia and industry.
39. D. Kurandina, M. Parasram, V. Gevorgyan, Angew. Chem. Int.
Ed. 56, 14212–14216 (2017).
40. X. Sun, J. Chen, T. Ritter, Nat. Chem. 10, 1229–1233
(2018).
41. Y. Fujiwara et al., Nature 492, 95–99 (2012).
42. R. Gianatassio et al., Angew. Chem. Int. Ed. 53, 9851–9855
(2014).
43. Z. Jiao, L. H. Lim, H. Hirao, J. S. Zhou, Angew. Chem. Int. Ed.
57, 6294–6298 (2018).
REFERENCES AND NOTES
44. I. Ghosh, T. Ghosh, J. I. Bardagi, B. König, Science 346,
725–728 (2014).
1. J. M. Smith, S. J. Harwood, P. S. Baran, Acc. Chem. Res. 51,
1807–1817 (2018).
Constantin et al., Science 367, 1021–1026 (2020)
28 February 2020
5 of 6

RESEARCH
| REPORT
ACKNOWLEDGMENTS
analyzed the results. Competing interests: The authors
declare no competing interests. Data and materials
availability: All data are available in the main text or the
supplementary materials.
Figs. S1 to S24
Tables S1 to S16
NMR Spectra
We thank D. Heyes for help with laser flash photolysis
studies. Funding: D.L. thanks EPSRC for a Fellowship
(EP/P004997/1) and the European Research Council for
a research grant (758427). Author contributions: F.J.
and D.L. designed the project and wrote the manuscript.
T.C., M.Z., A.R., and F.J. performed all experiments.
N.S.S. performed the computational studies. All authors
References (45–101)
SUPPLEMENTARY MATERIALS
science.sciencemag.org/content/367/6481/1021/suppl/DC1
Materials and Methods
16 November 2019; accepted 31 January 2020
10.1126/science.aba2419
Constantin et al., Science 367, 1021–1026 (2020)
28 February 2020
6 of 6

Aminoalkyl radicals as halogen-atom transfer agents for activation of alkyl and aryl halides
Timothée Constantin, Margherita Zanini, Alessio Regni, Nadeem S. Sheikh, Fabio Juliá and Daniele Leonori
Science 367 (6481), 1021-1026.
DOI: 10.1126/science.aba2419
Amines as a gateway to alkyl radicals
In recent years, photoredox catalysis driven by blue light has often been used to oxidize carbon centers adjacent
to nitrogen. Constantin et al. now show that these aminoalkyl radicals can, in turn, conveniently strip iodine atoms from a
variety of alkyl carbons. The new alkyl radicals that result readily undergo deuteration and couplings such as alkylation,
allylation, and olefination. The upshot is that simple amines can replace more hazardous conventional reagents such as
trialkyltin compounds in the homolytic activation and functionalization of halocarbons.
Science, this issue p. 1021
ARTICLE TOOLS
http://science.sciencemag.org/content/367/6481/1021
SUPPLEMENTARY
MATERIALS
http://science.sciencemag.org/content/suppl/2020/02/26/367.6481.1021.DC1
REFERENCES
This article cites 93 articles, 4 of which you can access for free
http://science.sciencemag.org/content/367/6481/1021#BIBL
PERMISSIONS
http://www.sciencemag.org/help/reprints-and-permissions
Use of this article is subject to the Terms of Service
Science (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement of
Science, 1200 New York Avenue NW, Washington, DC 20005. The title Science is a registered trademark of AAAS.
Copyright © 2020 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of
Science. No claim to original U.S. Government Works