Carbonyl-Directed Aliphatic Fluorination: A Special Type of Hydrogen Atom Transfer Beats out Norrish II

Article abstract of DOI:10.1021/jacs.0c07004

Recently, our group reported that enone and ketone functional groups, upon photoexcitation, can direct site-selective sp3 C-H fluorination in terpenoid derivatives. How this transformation actually occurred remained mysterious, as a significant number of mechanistic possibilities came to mind. Herein, we report a comprehensive study describing the reaction mechanism through kinetic studies, isotope-labeling experiments, 19F NMR, electrochemical studies, synthetic probes, and computational experiments. To our surprise, the mechanism suggests intermolecular hydrogen atom transfer (HAT) chemistry is at play, rather than classical Norrish hydrogen atom abstraction as initially conceived. What is more, we discovered a unique role for photopromoters such as benzil and related compounds that necessitates their chemical transformation through fluorination in order to be effective. Our findings provide documentation of an unusual form of directed HAT and are of crucial importance for defining the necessary parameters for the development of future methods.

Full text of DOI:10.1021/jacs.0c07004

Subscriber access provided by CARLETON UNIVERSITY
Article
Carbonyl-Directed Aliphatic Fluorination: A Special
Type of Hydrogen Atom Transfer Beats Out Norrish II
Fereshte Ghorbani, Stefan Andrew Harry, Joseph N. Capilato, Cody Ross Pitts, Jacob Joram, Garvin
N. Peters, John D. Tovar, Ivor Smajlagic, Maxime A. Siegler, Travis Dudding, and Thomas Lectka
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.0c07004 • Publication Date (Web): 03 Aug 2020
Downloaded from pubs.acs.org on August 3, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 19
1
2
3
Journal of the American Chemical Society
Carbonyl-Directed Aliphatic Fluorination: A Special Type of Hydro-
gen Atom Transfer Beats Out Norrish II
4
5
6
7
8
9
Fereshte Ghorbani,‡a Stefan Andrew Harry,‡a Joseph N. Capilato,a Cody Ross Pitts,a Jacob Joram,a
Garvin N. Peters,a John D. Tovar,a Ivor Smajlagic,b Maxime A. Siegler,a Travis Dudding,*b and Thomas
Lectka*a
aDepartment of Chemistry, Johns Hopkins University, 3400 N. Charles St., Baltimore, MD, 21218, United States
bDepartment of Chemistry, Brock University, 1812 Sir Isaac Brock Way, St. Catharines, ON L2S 3A1, Canada
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
‡ F.G. and S.A.H. contributed equally to this work.
ABSTRACT: Recently, our group reported that enone and ketone functional groups, upon photoexcitation, can direct site-
selective sp3 C-H fluorination in terpenoid derivatives. How this transformation actually occurred remained mysterious, as a
significant number of mechanistic possibilities came to mind. Herein, we report a comprehensive study describing the reac-
tion mechanism through kinetic studies, isotope labeling experiments,19F NMR, electrochemical studies, synthetic probes, and
computational experiments. To our surprise, the mech-
OAc
anism suggests intermolecular hydrogen atom transfer
Me
photo-, electro-, or chemical conditions
site-selective sp3 fluorination
radical chain mechanism
coordinated HAT
(HAT) chemistry at play, rather than classical Norrish
hydrogen atom abstraction as initially conceived. What
is more, we discovered a unique role for photopromot-
ers such as benzil and related compounds that necessi-
tates their chemical transformation through fluorina-
tion in order to be effective. Our findings provide doc-
umentation of an unusual form of directed HAT and are
of crucial importance for defining the necessary param-
eters for the development of future methods.
Me
H
H
H
H
H
AcO
O
H
R
F
R
N
O
O
H
H
Selectfluor
N
conditions
CH2Cl
P1
P2
R1
R2
F
R
F
R
O
O
CH2Cl
H
R
H
R
O
O
N
2BF4
(2.2 equiv.)
N
mode I
(distal-6)
mode II
(proximal-6)
(Selectfluor)
F
P3
P4
R3
R4
MeCN, hv (300 nm)
O
F
O
F
O
H
O
H
four previously defined
reactivity modes for
enone-directed C-H
fluorination
mode III
(distal-5)
mode IV
(proximal-5)
Scheme 1. Initial discovery of enone-directed fluorination reactivity modes, classified based on 1) proximity of C=C bonds to
the reactive site and 2) number of bonds between the carbonyl oxygen atom and the abstracted hydrogen atom.
Introduction. Putative single-electron transfer (SET) and
hydrogen atom transfer (HAT) processes underpin much re-
cent and remarkable synthetic chemistry.1 The detailed
mechanisms by which these reactions occur are generally
much less well understood.2 Under the umbrella of HAT,
proton transfers play an additional role, giving rise to a spec-
trum of mechanistic scenarios: concerted proton-coupled
electron transfer (CPET), sequential ET/PT, etc.3 We re-
cently reported that enone-containing rigid terpenoid deriv-
atives, in which the carbonyl group is positioned to interact
through potential 5- or 6-membered transition states with
proximate C−H bonds, afforded alkyl fluorides regioselec-
tively in moderate to high yields upon irradiation at 300 nm
in the presence of Selectfluor (SF)(Scheme 1).4
ethers, and benzylic positions activated by carbonyl groups
(Figure 1).5
C8H17
Me
Me
Me
Me
Me
Me
H
H
H
H
AcO
O
O
ketones
enones
O
Me
O
O
Me
Me
H
O
O
Me
Me
O
H
H
O
TFAO
keto-ethers
benzylic carbonyls
Figure 1. Expanded scope of study for carbonyl-directed
fluorination in our recent work.
We soon extended the scope of this work to include site-
selective fluorination of substrates containing ketones, keto
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 19
1
Scenario I: Norrish
2
3
*
H
R
H
F
R
F
R
H
R
H
R
•
O
SF
O
O
O
R3N
hv
O
4
5
R
R
R
R
R
6
7
Scenario II: ET-PT (McLafferty-like)
8
9
•
H
R
H
R
R
H
R
H
F
R
O
O
O
O
SF
O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
R
R
R
R
R
Scenario III: PT-ET
F
R
H
R
H
R
R
O
O
O
O
SF
R
R
R
R
Scenario IV: Intramolecular CPET
F
R
R
H
R
H
R
O
O
O
O
SF
R
R
R
R
Scenario V: HAT
F
R
H
R
R
O
O
O
SF
R
R
R
Figure 2. Five possible mechanistic hypotheses.
In each successive case, a proximate carbonyl group is
poised to interact with a C-H bond through a 5- or 6-mem-
bered ring. On the other hand, how this remarkable trans-
formation occurred was shrouded in mystery; aside from
the fact that a proximate carbonyl group exerted a key di-
recting effect, mechanistic details remained speculative.
We settled on five possible mechanistic hypotheses (Fig-
ure 2). Initially, our data seemed to comport with an inter-
rupted Norrish II process involving intramolecular HAT
(Scenario I). Electron transfer may play a pivotal role – in
either sequential electron transfer-proton transfer (ET-PT,
Scenario II) or PT-ET, which seems thermodynamically un-
reasonable but is included for the sake of completeness
(Scenario III). Proton-coupled electron transfer, which may
represent a point on the mechanistic continuum of HAT,
must be considered a serious alternative (PCET, Scenario
IV). On the other hand, direct, rate-determining HAT re-
mained a possibility, although this scenario begs the ques-
tion of what precise role the carbonyl group plays in direct-
ing the reaction (Scenario V).
Given that the only chromophore present in the reaction
mixture absorbing in the 300 nm region was the enone moi-
ety of the substrate, we originally surmised the main viable
initiation pathway following its excitation could be a Nor-
rish type II intramolecular process.6 This initial supposition
turned out to be overly simplistic, if not incorrect, and was
surprisingly clarified by the subsequent use of longer wave-
length absorbing photoinitiators such as benzil, thereby ob-
viating Norrish chemistry. After wending our way down a
tortuous path, we found that the mechanism instead indi-
cates intermolecular HAT chemistry at play, rather than clas-
sical intramolecular Norrish hydrogen atom abstraction.
This HAT can also be thought of potentially as a limiting
PCET termed multi-site concerted proton-coupled electron
transfer: MS-CPET.7
In this article, we wish to present our detailed mechanis-
tic findings of this unusual and timely reaction. We also doc-
ument the key role played by the remarkable chemical
transformations of the photoinitiators (affecting the para-
digm or supposition that such compounds often act as pho-
tocatalysts). Mechanistic clues accumulated along the way
were bolstered by alternative initiation of the reaction
through chemical (BEt3/O2) and electrochemical means.
Our findings provide an interesting case of "directed" HAT
in a general synthetic method and hopefully will prove to be
of crucial importance for defining parameters for the devel-
opment of related methods.
Background. In earlier work, we developed a series of rad-
ical-based fluorination methods that highlight different po-
tential mechanisms. For example, the Cu(I)-promoted fluor-
ination of aliphatic substrates was shown (in at least some
cases) most likely to proceed through a key HAT step.8
Given the involvement of the putative Selectfluor radical di-
cation (SRD) in published work and the present reaction,
HAT became a logical mechanistic candidate, although we
were quite skeptical as this pathway generally leads to
"scattershot" fluorination at a large number of sites within a
ACS Paragon Plus Environment

Page 3 of 19
Journal of the American Chemical Society
1
2
3
4
5
6
prototype reaction
CH2Cl
prior work
OAc
OAc
Me
Me
N
2BF4
Selectfluor
Me
H
Me
H
F
70%
2
Cu(I)
N
SRD involved
in HAT
MeCN
hv (300 nm)
H
H
Selectfluor
K2CO3, MeCN
H
F
Selectfluor radical
dication (SRD)
AcO
O
AcO
O
1
7
8
9
1. Selectfluor
cat. 9-fluorenone
Scheme 3. Typical example of enone-directed fluorination
through 300-nm irradiation.
O
O
O
10
11
12
13
MeCN, hv
Ph
F
Ph
HO
SRD involved in ET
2. LiOH, H2O
Experiments: Results from Direct Photolysis. Our initial
intuition for enone-directed fluorination led us to believe
the observed selectivity was due to a Norrish-type pathway.
As such, we proposed that intramolecular HAT may occur
first, and the resulting carbon radical could then be inter-
cepted through fluorination by Selectfluor (SF) before either
cleavage or Yang cyclization could occur (Figure 2, Scenario
I). For example, direct irradiation of compound 1 in acetoni-
trile with 300 nm light in the presence of SF produces the
directed fluorinated product 2 in 70% yield (Scheme 3).4 In
the absence of SF, the above conditions produce a small
quantity of unidentifiable products. Reactions conducted in
the dark and under 400-nm irradiation (using blue LEDs)
only result in recovered starting material. Additionally, the
UV-vis spectra of Selectfluor in MeCN at various concentra-
tions showed no absorption bands above 300 nm (Figure 4).
These experiments potentially implicate a role for the Nor-
rish reaction in initiation (albeit it has an even more ques-
tionable role in chain propagation/fluorination).
Scheme 2. Examples where SRD has been shown to play a
significant role in both hydrogen atom transfer (HAT) and
electron transfer (ET).
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
complex molecule. On the other hand, the tandem C-C bond
cleavage/fluorination of acetals (also involving SRD) must
proceed through some sort of a SET process with no HAT in-
volvement (Scheme 2).9 The flexibility of the SRD/SF pair to
play different roles warned us about jumping to conclu-
sions.10 In the present work, we were once again intrigued
by the possibility that SRD may play an imperative role.
[Ox]
e-
O
H
O
O
O
Figure 3. Mayer's multiple-site concerted proton-electron
transfer (MS-CPET) system.
In regard to PCET, recent work by Knowles and cowork-
ers demonstrates its viability for a number of unique syn-
thetic transformations.11 Additionally, variants of the basic
PCET system should be noted. For example, the subclass
"MS-CPET" (multiple-site concerted proton-electron trans-
fer) has only recently been explored for C-H activation. To
our knowledge, Mayer and coworkers described the first
mechanistic account in 2019, whereby the carboxylate of
fluorenyl-benzoate facilitates rapid cleavage of a benzylic C–
H bond in the presence of a weak one-electron oxidant (Fig-
ure 3).12 For our purposes, this precedent is important as it
parallels, at least superficially, the present findings in sev-
eral ways. In Mayer's case, a carboxylate acts as an internal
base through a 6-membered ring transition state. The oxi-
dant, in analogy to SRD, is an amine radical cation. We bear
in mind what Mayer and coworkers have stated: "MS-CPET
reactivity is increasingly proposed in biological and syn-
thetic contexts, and some reactions typically described as
HAT more resemble MS-CPET. Despite that HAT and MS-
CPET reactions 'look different,' we argue here that these re-
actions lie on a reactivity continuum, and that they are gov-
erned by many of the same key parameters."13 In our case
as well, strict HAT or PCET may represent points on this re-
activity continuum.
Figure 4. UV-vis spectra of Selectfluor in MeCN at various
concentrations.
There are other conceivable pathways for initiation, such
as triplet-triplet energy transfer from substrate to SF, or
electron transfer from the excited substrate to SF that can
generate SRD,14 (although N-F bond activation in SF has
been reported to occur at 400 nm with blue LEDs, as stated
we see no absorption in the UV-vis spectrum in this range).15
These alternatives are not mutually exclusive and may run
parallel to each other; as all the hypotheses involve SRD, the
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 19
1
control
C8H17
Me
2
3
4
5
Me
H
Selectfluor
various fluorinated products
10 mol% benzil
hv (400 nm), MeCN
H
H
H
3
6
7
8
9
competition
AcO
Me
AcO
Me
Me
C8H17
Me
H
Me
Selectfluor
10 mol% benzil
hv (400 nm), MeCN
Me
H
Me
H
Me
H
+
H
H
H
H
H
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
F
H
H
3
AcO
O
AcO
O
4
5
exclusive
Scheme 4. Control and intermolecular competition experiments.
possibility of very complex hybrid mechanisms is possible.
Thus direct photolysis, at least from a mechanistic stand-
point, appears to be anything but direct. In contrast to
enones, the direct photolysis of rigid, optimally configured
ketones produces a generally lower yield of desired fluori-
nated products along with other non-directed fluorinated
species. Decomposition of ketone substrates into a number
of fluorinated products competes with the desired process.
For example, α-cleavage of the O=C—C bond gives rise to ev-
ident acyl fluoride byproducts by 19F NMR.16 Enones do not
seem to be quite as susceptible to these types of cleavage
processes.
hydrogen atom abstractor, oxygen scavenger, organic rea-
gent, and a triplet-triplet energy transfer facilitator).18
OAc
OAc
Me
Me
Selectfluor
sensitizer (10 mol%)
Me
H
Me
H
MeCN, 14 h
hv (400 nm)
H
H
H
H
H
1
F
2
AcO
O
AcO
O
19F NMR yield (%)
entry
sensitizer
-
0
1
2
3
4
5
6
Photoinitiation. The unsuccessful utilization of ketones
(and poor yields of certain products obtained from enones)
prompted us to turn to visible light/near-UV photosensiti-
zation with the goal of increasing yields and improving con-
ditions. Paradoxically, we also sought to clarify (or simplify)
the mechanism through the addition of a photosensitizer.
Table 1 shows a range of photosensitizers that we screened
for both enones and ketones (reactivity was similar with
both; steroid 1 is shown as a model substrate).
43
2-bromo-9-fluorenone
47
55
9-fluorenone
dibenzosuberenone
9,10-phenanthrenequinone
4,4-difluorobenzil
64
67
71
7
8
2,7-dichloro-9-fluorenone
2-chlorothioxanthone
73
A trend stands out: carbonyl-containing sensitizers (such
as benzil) work most effectively in the reaction. Benzil itself
increases the yield considerably (from 70% to 94%) over di-
rect photolysis. One other important fact related to the
mechanism was immediately noted - a photosensitized ap-
proach utilizing 400 nm light (with a narrow 10 nm spectral
dispersion) conclusively rules out a Norrish-type pathway at
play in this case. Due to the ketone/enone absorbance be-
low 350 nm and > 20 kcal/mol triplet energy gap between
benzil and ketone/enone substrates, triplet sensitization is
virtually prohibited.17 In this case, it is not feasible to form
the photoexcited substrate through direct excitation or tri-
plet-triplet energy transfer. Thus, the photosensitizers are
playing a very different initiation role. When a directing car-
bonyl group is not present (control 3) fluorination still re-
sults, although it qualifies as “scattershot,” resulting in a
multitude of fluorinated products, none of which is pro-
duced in useful quantities. A mixture of a known substrate
4 plus hydrocarbon 3 produces product 5 exclusively, even
though 4 is expected to have a higher ionization potential
(Scheme 4; competition). Moreover, hydrocarbon 3 fails to
fluorinate under 300-nm irradiation (Scheme 4; control).
The chemical (and photochemical) behavior of benzil has
been well documented over the past 150 years (e.g. as a
methylbenzoylformate
9
89
benzil
94
10
Table 1. Scope of sensitizers screened under 400-nm irra-
diation.
Our first thought turned to the possibility that photoex-
cited benzil acts through HAT (Scheme 5).19 Photoexcited
carbonyl-containing species are proposed to be competent
hydrogen atom abstractors in a variety of settings.20 In our
case, this concept was unlikely for two reasons: 1) the ener-
getics are not favorable (it would be a fairly endothermic
process for the steroid substrates) and 2) benzil in any event
is not expected to be a selective hydrogen atom abstractor.
The calculated abstraction energy for steroid 1 is uphill by
almost 12 kcal/mol (IEFPCM(CH3CN)UωB97X-D/6-
311++G(2d,2p)); in known cases where benzil engages in
HAT, the hydrogen atom donor is fairly activated, as in cu-
mene and isopropanol.21 Furthermore, if the other photo-
sensitizers of differing shapes, sizes, and electronic proper-
ties operate through HAT, it seems highly unlikely that they
would afford the exact same site-selectivity as well. As can
be seen in Table 1, along with direct photolysis, various
ACS Paragon Plus Environment

Page 5 of 19
1
Journal of the American Chemical Society
other carbonyl photosensitizers afford identical selectivity.
detectable cross products (Scheme 7). The same result was
2
On the other hand, Bunbury and Wang have irradiated ben-
zil in cyclohexane solution, observing in the process both
cleavage products and products resulting from H-abstrac-
tion.18a,b In our case, in a control experiment involving ben-
zil irradiated with steroid 6 at 400 nm, we do not observe the
cleavage byproducts consistent with benzil acting as an ab-
stractor, which should include benzaldehyde.
obtained with 4,4’-difluorobenzil (both of these benzil de-
rivatives work equally well in the reaction). In the presence
of SF, recovered starting materials also showed no evidence
of scrambling, suggesting that SF does not promote the pro-
cess.
3
4
5
6
7
8
intermolecular competition
9
R
OAc
O
Me
3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
R
Me
H
O
+
O
R = F, Me
R
H
H
MeCN, hv (400 nm)
H
O
AcO
6
O
O
with and without
Selectfluor
O
OAc
no crossover detected
Me
OH
Me
H
O
+
H
H
O
AcO
O
Scheme 7. Crossover experiment to probe possible direct
α-cleavage of benzil derivatives in the presence and absence
of Selectfluor.
Scheme 5. Concept of benzil acting through putative HAT.
Appropriately, multifaceted benzil seems to take on yet
another role in the present reaction. For example, we note
the generation of significant quantities of benzoyl fluoride in
all successful fluorination reaction mixtures with SF as rea-
gent. As a control, a mixture of benzil and SF (no substrate)
in MeCN was irradiated by a 400 nm LED lamp and gener-
ated benzoyl fluoride in 44% yield when performed in a
glove box and 34% under normal reaction conditions. In
contrast, benzoyl fluoride is not generated in the dark. As
suggested in Scheme 6, the formation of benzoyl fluoride
may logically be correlated to the generation of SRD. How-
ever, the question remains of how benzoyl fluoride is
formed.
The results are consistent with direct cleavage not occur-
ring to a significant extent, or else recombination of benzoyl
radicals is happening faster than their diffusion. Even very
prolonged irradiations produce no evidence of cross prod-
ucts, which along with precedent suggests that direct α-
cleavage of benzil is simply not occurring under the usual
fluorination conditions (400 nm irradiation). However, we
conducted an intermolecular competitive kinetic isotope ef-
fect experiment and tracked the consumption of the benzil
starting materials (due to the instability of benzoyl fluoride,
we were not able to measure its formation with quantitative
mass spectrometry). We found an enhancement of 13C in the
starting material as the reaction proceeds. This suggests
that cleavage may be occurring through a different pathway
(Scheme 8).
CH2Cl
O
O
N
Selectfluor
2BF4
F
intermolecular competition
N
O
O
O
F
Scheme 6. Formation of benzoyl fluoride under irradiation.
O
Selectfluor
The α-cleavage of benzil is a conceivable pathway in
which the reaction could be initiated. One can imagine the
resultant benzoyl radicals reacting quickly with SF to pro-
duce PhCOF and SRD, thus initiating the reaction, as this pro-
cess is predicted to be highly exothermic. The excitation and
intersystem crossing of benzil to the Tn state may promote
an energetically feasible homolysis – albeit one that has only
been observed in laser pulse studies involving two-photon
excitation.18e,f In fact, evidence for direct α-cleavage (Nor-
rish I) of benzil under more normal chemical conditions is
scant. Nevertheless, assuming that direct cleavage occurs
under our conditions of photoexcitation, a simple synthetic
probe to observe cross products would provide incontro-
vertible proof.
MeCN, hv (400 nm)
O
O
F
O
[
12C-benzil]
13C-benzil]
13C labeled
= 0.96 at 70% conversion
[
Scheme 8. Competitive kinetic isotope effect (KIE) experi-
ment.
Another conceivable route involves excited-state benzil
abstracting a hydrogen atom from acetonitrile and subse-
quently cleaving to generate a benzoyl radical.22 Although
energetically uphill by most measures, it was simple enough
to test this hypothesis. We conducted a KIE experiment
Accordingly, we irradiated a mixture of 4,4’-dimethylben-
zil and benzil in acetonitrile with 400 nm light and found no
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 19
1
2
3
4
5
6
7
8
employing acetonitrile-d3; this experiment showed no
unnecessary for initiation, but that too much retards the in-
change in rate when compared to acetonitrile (Scheme 9).
Additionally, benzaldehyde is once again expected as a by-
product, yet none was detected.
itiation step by quenching the triplet-excited benzil either
chemically or photochemically. Nevertheless, Saltiel’s sem-
inal experiments suggest a plausible analogy. In a more
likely scenario, triplet benzil (drawn in a form emphasizing
its diradical character) is trapped by Selectfluor (instead of
dioxygen) and cleaves into benzoyl fluoride and benzoyl
radical (Figure 5, right). This hypothesis is supported by
DFT calculations (∆E = -59.4 kcal mol−1 for the formation of
products [UωB97XD/6-311+G**(MeCN)]). The calculated
values for fluorination of triplet benzil are considerably
more exothermic than those calculated for the triplet oxy-
gen reaction (∆E = -21.7 kcal mol−1). Also in both cases, the
presumed adducts (with 3O2 and F•) are not stable minima
and dissociate, suggesting a concerted route to products.
Bear in mind that SF is an excellent radical trap and known
to react extraordinarily rapidly with organic free radicals.24
A recent study by Tan and coworkers25 addressed the in-
teraction between photoexcited anthraquinone and Select-
fluor using transient-absorption spectroscopy and DFT cal-
culations. The authors propose that an anthraquinone-Se-
lectfluor exciplex is responsible for initial HAT from their
substrates, and this initiates a chain reaction. While we can-
not rule out the formation of exciplexes in our system, it
seems once again unlikely that HAT from excited state benzil
would be selective in any form.
Selectfluor
O
O
MeCN, hv (400 nm)
F
Selectfluor
O
9
CD3CN, hv (400 nm)
no change
in rate
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 9. KIE experiment probing role of solvent.
One other illuminating piece of data was obtained by the
use of 9-fluorenone as a photoinitiator. The reaction pro-
vides a slightly lower yield of fluorinated ketone/enone in
comparison to the benzil-initiated reaction, but still works
moderately well. In the case of 9-fluorenone, hypothetical
α-cleavage through a Norrish I fragmentation is expected to
be less favorable than that of benzil as the resulting reactive
intermediate consists of an aryl radical (Scheme 10). Nev-
ertheless, purification of the reaction mixture led to the iso-
lation of difluoride 7 (the structure was unambiguously as-
signed via single-crystal X-ray diffraction), evidently result-
ing from difluorination of an aryl-benzoyl radical, which
should also be competent in generating SRD.
O
O
F
Selectfluor (2 equiv.)
3
O
3O2
MeCN, hv (400 nm)
SF
7
F
O
crystal structure
O
O
F
O
O
SRD
+
O
O
not a minimum
not a minimum
Scheme 10. α-Cleavage and difluorination of 9-fluorenone.
Crystal structure of product shown (displacement ellipsoids
at 50% probability with hydrogen atoms refined by a riding
model).
O
O
O
O
An attractive pathway for the production of benzoyl fluo-
ride is suggested by Saltiel’s classic reaction of ground state
triplet dioxygen with photoexcited benzil (Figure 5, left).23
The resulting peroxyl radical intermediate (or perhaps tran-
sition state) cleaves to produce benzoyl radical and the
daughter peroxy radical 8. In order to determine whether
dioxygen was involved in our reaction, we conducted our
original experiments under strict atmospheric regulation,
whereby a reaction mixture containing benzil, substrate 1,
and SF in dioxygen-free acetonitrile (freeze-pump-thaw cy-
cled) was prepared in a glove box and then irradiated. The
resultant crude mixture was found to have fluorinated in a
comparable yield to the typical reaction conditions.
O
8
O
F
+
+
+
SRD
ΔE = -21.7 kcal
ΔE = -59.4 kcal
Figure 5. Analogy between Saltiel's experiments involving
excited benzil and triplet dioxygen (left) and a plausible var-
iant involving Selectfluor (right).
Photoinitiation with Other Promoters. Carbonyl-con-
taining photopromoters that absorb in the region of irradia-
tion are notable for their efficacy in the reaction, which we
attribute to their propensity for a-cleavage in the presence
of SF. Furthermore, a number of “non-carbonyl” photopro-
moters work as well. Anthracene produces a good yield of
In addition, a fluorination reaction conducted under a
pure dioxygen atmosphere failed to fluorinate either the
steroid or benzil - suggesting that dioxygen was not only
ACS Paragon Plus Environment

Page 7 of 19
Journal of the American Chemical Society
F
CH2Cl
N
F
H
1
2
3
4
5
6
7
*
Selectfluor
2BF4
+
MeCN
hv (CFL)
•
N
escape
fluorination
Scheme 11. Effecting the fluorination reaction with non-carbonylic photopromoters.
8
9
product 5 with a CFL bulb, whereas perylene produces no
product. Note that the triplet energies of both anthracene
(ET » 42 kcal/mol) and perylene (ET » 35 kcal/mol) are too
low for sensitization to be viable, and that both anthracene26
and perylene27 absorb within the region of the light source
emission.
During the course of the reaction, anthracene is fluori-
nated (predominately in the 9-position) whereas perylene
is not. One can imagine excited-state anthracene reacting
with SF to liberate SRD and initiating a chain process
(Scheme 11). A general rule of thumb is that any photopro-
moter that works well in the reaction is transformed itself
by fluorination (presumably to produce SRD).
The Triethylborane Method. We established in prior work
that the production of ethyl radicals during the autoxida-
tion29 of triethylborane provides an adequate radical source
to be fluorinated, thus generating SRD.30 Under these
strictly chemical conditions, fluorination of steroidal
enones/ketones is possible (albeit in lower yields) with
identical reactivity patterns to those derived from direct and
benzil-catalyzed photolysis (example in Scheme 13). More-
over, the lower yields observed when using triethylborane
are possibly attributed to maintaining a sufficient quantity
of SRD at any one time, the necessary presence of dioxygen,
and the vagaries of putative chain propagation in general. In
particular, the reaction requires oxygen, but too much also
inhibits the reaction.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Alternative Methods to Generate SRD. To understand the
intimate role that SRD plays within the mechanism, we
sought methods for its generation by purely chemical
means. N,N,N′,N′-Tetramethyl-p-phenylenediamine is an
avid one-electron donor that generates the highly character-
istic dye “Würster’s Blue” in the process.28 It reacts readily
with SF, producing the colored dye immediately. Unfortu-
nately, this reaction cannot be used to initiate a selective
fluorination, as SRD itself is even more highly susceptible to
one electron reduction. The result is almost clean conver-
sion of SF to amine 9 (Scheme 12).
OAc
OAc
Me
Me
Selectfluor (2.2 equiv)
Me
H
Me
H
Et3B (20 mol%)
33%
1.4:1 d.r.
H
H
H
H
MeCN
AcO
H
1
F
2
AcO
O
O
Scheme 13. The triethylborane test in which SRD is gener-
ated under non-photochemical conditions.
Bulk Electrolysis. Using cyclic voltammetry (CV), peak ox-
idation potentials of amine 9 (a direct precursor to SRD) and
substrates 6 and 4 were found to lie between 1.9 and 2.4 V
vs. Ag/Ag+ (Figure 6). Unsurprisingly, the oxidations are ir-
reversible at all scan rates probed, although peak shapes
change. We bore in mind that these outcomes are depend-
ent on the electrochemical solution concentrations, but they
provided a rough guide to voltage tuning for the perfor-
mance of a bulk electrolysis.31 Irreversible peak reduction
of SF was also observed at –0.3 V; the presumably liberated
SRD is quickly reduced itself (dry and de-oxygenated MeCN
with 0.1 M TBAPF6 and a potential sweep rate of 100 mV/s
(vs Fc/Fc+)). We chose to conduct a bulk electrolysis exper-
iment whereby substrate 4 and SF were mixed in MeCN, and
a potential of 1.8 V (approximate anodic potential of amine
NMe2
NMe2
CH2Cl
N
Selectfluor
+
2BF4
N
NMe2
NMe2
+ F
NMe2
NMe2
NMe2
CH2Cl
N
SRD
+
BF4
N
NMe2
9
Scheme 12. Reactions of SF with N,N,N′,N′-tetramethyl-p-
phenylenediamine.
CH2Cl
N
BF4
AcO
AcO
Me
Me
Me
Me
(3.0 equiv.)
N
(no mediator)
Me
Me
trace
Selectfluor (3.0 equiv.)
Selectfluor (3.0 equiv.)
81%
fluorinated
products
H
H
0.1 M TBAPF6
MeCN, 3 h
0.1 M TBAPF6
MeCN, 3 h
H
F
4
5
AcO
O
AcO
O
(+)RVC/(-)Pt
1.8 V vs. Ag/Ag+
divided cell
(+)RVC/(-)Pt
1.8 V vs. Ag/Ag+
divided cell
Scheme 14. Bulk electrolysis experiment supporting involvement of SRD.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 19
1
2
3
4
5
6
7
8
9, and out of range to oxidize compound 4) was applied to
In the absence of SF, unidentifiable products form, but at
the cell for 3 h (Scheme 14, left). Under these conditions,
product yields were very low (< 3%). On the other hand,
electrolysis in the presence of amine 9 resulted in an 81%
yield of product, and once again, the exact same product dis-
tribution was observed as compared to the photolytic ap-
proach (Scheme 14, right). Amine promotor 9 proved abso-
lutely necessary as a mediator – the optimal voltage for the
reaction corresponds very roughly to its oxidation, suggest-
ing that SRD is once again the indispensable actor. Direct
oxidation of the steroid itself produced product (E = 2.3 V),
albeit in only 10% yield, suggesting that a critical threshold
of SRD as a chain carrier was not attained. Use of other
fluorinating agents such as NFSI produced small amounts of
various unselective fluorinated products. This experiment,
besides its innate utility, cleared away a number of mecha-
nistic ambiguities.
a much lower rate than fluorinated byproducts when SF is
present, suggesting that another pathway is at work. Conse-
quently, yields are naturally lower as a bit of the substrate is
sacrificed. The situation is most dramatic for ketones,
whose propensity to fragment seems to be greater than that
for enones.
9
ET/PT. Benzil, photoinitiation, borane initiation and bulk
electrolysis conclusively rule out Norrish II chemistry, as
well as any other chemistry involving photoexcitation of the
enone chromophore with the exception of direct excitation.
If ET/PT were operative, the carbonyl lone pairs must act as
an internal base to deprotonate intramolecularly an opti-
mally poised C-H bond, yielding a protonated carbonyl and
a secondary carbon radical. This step (proton transfer) can
happen sequentially or simultaneously with an electron
transfer; consequently, there exist two reasonable path-
ways: electron transfer followed by proton transfer (ET-PT)
or concerted transfer of the two particles (CPET). One other
pathway (PT-ET) is high energy by any estimation and
should be discounted.32
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CH2Cl
N
BF4
9
N
C8H17
Me
Me
H
6
AcO
O
AcO
Me
Me
Me
OH
OH
H
H
stepwise
concerted
4
O
OH
AcO
O
R
R
R
R
Figure 6. Cyclic voltammograms of compounds 9, 6, and 4
in dry and de-oxygenated MeCN with 0.1 M TBAPF6 and a
potential sweep rate of 100 mV/s (vs Fc/Fc+).
Scheme 16. Gas-Phase McLafferty rearrangement.
Direct Reprise. Reprising briefly the topic of direct photol-
ysis, Scheme 15 provides one clue as to the lower yields ob-
served; the substrate itself likely serves as the initiator. As
mentioned, every successful photopromotor is itself fluori-
nated in order to liberate SRD. This may apply to direct pho-
tolysis as well - the minor acyl fluoride byproducts (highly
characteristic by 19F NMR in the vicinity of +17 ppm) likely
result from the cleavage of putative radical intermediate 10
(Scheme 15).
The Relationship of CPET and ET/PT to the Gas-Phase
McLafferty Rearrangement. The enone activation process
bears a resemblance to the gas-phase McLafferty rearrange-
ment (Scheme 16). This venerable gas-phase reaction in-
volves carbonyl-containing compounds possessing g-hydro-
gens, similar to many of our substrates.33 The (somewhat
limited) mechanistic consensus advocates electron abstrac-
tion, followed by intramolecular HAT and then fragmenta-
tion (the present work could thus be considered a formal
"interrupted" McLafferty reaction). Intramolecular isotope
effects for these rearrangements are documented; these
range over a wide spectrum of values.34 However, intermo-
lecular isotope effects for McLafferty reactions are not
widely known. The McLafferty rearrangement would seem
to be a candidate for PCET, but not much if anything is re-
ported about this option. Djerassi and coworkers35 have ex-
amined potential McLafferty rearrangements in keto ster-
oids and found that only when the interacting carbonyl and
C-H bonds can approach to within 1.5 Å is the rearrange-
ment feasible (Figure 7). In more rigid steroidal ketones it
does not occur, in contrast to the present chemistry. This is
an interesting fact that begs the question whether CPET or
ET is involved in our system at all, as our geometric require-
ments are so different than typical McLafferty substrates.
Granted, McLafferty chemistry is all gas phase, so the lack of
correlation may not be taken as definitive.
direct photolysis
OAc
Me
Me
H
Selectfluor
2 + < 10% acyl
fluoride byproducts
H
H
hv (300 nm), MeCN
AcO
O
OAc
Me
H
OAc
*
Me
Me
H
Me
H
Selectfluor
10
H
H
H
•
AcO
O
AcO
O
F
Scheme 15. Under 300-nm irradiation in the absence of a
sensitizer.
ACS Paragon Plus Environment

Page 9 of 19
Journal of the American Chemical Society
MS-PCET pathway
OAc
Me
1
2
3
4
OAc
Me
CH2Cl
CH2Cl
N
Me
H
N
Me
H
Δ
G = -2.3 kcal/mol
+
+
•
H
H
H
H
[UωB97XD/6-311+G**(MeCN)]
H
N
N
AcO
11
OH
AcO
O
5
•
6
7
8
9
10
11
12
13
14
15
16
17
18
19
comparing radical cation isomers
OAc
OAc
Me
Me
H
Me
H
Me
H
12
2.53 Å
•
H
H
H
2.01 Å
H
H
AcO
O
AcO
O
-4.1 kcal/mol
0.0 kcal/mol
Δ
G
rel [UωB97XD/6-311+G**(MeCN)]
HAT pathway
OAc
Me
OAc
Me
CH2Cl
N
CH2Cl
Me
H
N
Me
H
Δ
G = -7.9 kcal/mol
+
+
•
H
H
H
H
[UωB97XD/6-311+G**(MeCN)]
H
N
N
AcO
O
AcO
O
H
Scheme 17. Comparing calculations of MS-PCET and HAT mechanisms.
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
weaker interaction exists in computed radical cation 12 of
2.37 Å between C and O (Scheme 17). On the other hand, a
straightforward HAT is exothermic by more than 7.9
kcal/mol.
R
Me
Me
Me
Me
H
H
O
H
Me
H
Me
H
Me
H
O
Our efforts to discern between the two viable possibilities
led us to use a mechanistic probe: an isotope effect experi-
ment between 2-(pentadeuteroethyl)cyclo-hexanone and 2-
ethylcyclohexanone). We observed a phenomenological ki-
netic isotope effect in a one-pot intermolecular competition
reaction (KIE = 3.4) and also when comparing initial rates of
reaction of each isotopomer separately (KIE = 3.7). Both re-
sults are large enough to encompass a primary effect (cleav-
age of the C-H bond) along with superimposing secondary
effects (Scheme 18).37 These numbers rule out an initial rate-
limiting ET in ET-PT and argue against pre-equilibrium PT in
PT-ET.
Although we observed a primary KIE, PT-ET (rate-limit-
ing PT) is unreasonable; the rate determining PT would be
highly thermodynamically unfavorable due to the low acid-
ity of the targeted C-H bond and the resulting instability of
the zwitterion intermediate. One important experiment de-
signed to distinguish MS-PCET and HAT from pre-equilib-
rium ET in ET/PT involves the use of SF derivatives pos-
sessing different oxidizing power. SF derivative 13-a has a
more positive anodic peak potential (Ea = 2.14 V) and inflec-
tion-point potential (Ei = 1.88 V) compared to compound 9
(Ea = 1.99 V amd Ei = 1.77 V) (Figure 8), and leads to a faster
reaction (Scheme 19), which could be due to HAT, an en-
hanced electron transfer rate or else fluorination. However,
the observation of a KIE for proton transfer in the reaction
militates against the fluorination step being rate-determin-
ing; thus, this result supports the involvement of HAT or
electron transfer in the rate-determining step. Congruent
with this conclusion is our prior work that established a
very fast rate for the reaction of SF with free radicals.24a
H
H
H
H
H
O
H
no McLafferty
no McLafferty
McLafferty
Figure 7. Examples of steroid cores that were or were not
observed to undergo McLafferty rearrangements in gas-
phase experiments.
intermolecular KIE experiments
O
O
H
H
F
H
CH3
CH3
Selectfluor
KIE = 3.4 (one-pot);
3.7 (separate pots)
+
+
MeCN, benzil
hv (400 nm)
O
O
F
D
D
D
CD3
2H labeled
CD3
Scheme 18. Competitive intermolecular kinetic isotope ef-
fect experiment. The reaction gives rise to minor amounts
of ring-fluorinated byproducts that do not affect the analy-
sis.
One insightful calculation shows that CPET (to be more
accurate: MS-PCET) is modestly exothermic for a typical ste-
roidal substrate (Scheme 17). The precise number, of
course, is to be best viewed as a ballpark figure. More inter-
esting is the comparative energy of radical cation isomers
that favors 11 by 2.3 kcal/mol. The structure of 11 reveals
what could be characterized as a hydrogen bonding contact
between the radical center and the OH proton (2.01 Å).36
A
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 19
1
F
F
2
(6 equiv.)
3
4
5
6
7
8
9
N
AcO
Me
AcO
Me
2BF4
Me
Me
N
yield
0.5 h <1%
yield
time
time
13-b
Me
Me
Selectfluor (6 equiv.)
F
0.5 h
1 h
1.5 h 17%
3%
10%
5
1 h
3%
8%
H
H
benzil (10 mol %)
MeCN, hv (400 nm)
benzil (10 mol %)
MeCN, hv (400 nm)
H
F
1.5 h
4
5
AcO
O
AcO
O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 19. Rate comparative studies for compound 13-b vs. SF; monitored initial rate with excess fluorinating reagent pre-
sent.
40
Ea
at several sites simultaneously. In a simple probe, we ob-
35
serve product distribution in accordance with the “polar ef-
30
25
20
15
10
5
F
F
fect,” a known HAT pathway (Scheme 20).39
CH2Cl
N
N
N
Ei
ε
2.3 : 1.3 : 1.0
γ
β
BF4
BF4
polar effect
9
N
Ea
13-a
O
O
Selectfluor
F
Ei
Me
Me
Me
Me
benzil, MeCN
hv (400 nm)
0
Scheme 20. Distribution of fluorinated products from reac-
tions employing non-rigid carbonyl-containing substrates
are characteristic of the established "polar effect.”
-5
1
1.2
1.4
1.6
1.8
2
2.2
2.4
E / V
In more rigid systems however, if the carbonyl is indeed
templating the approach of SRD to the targeted C-H bond,
then it stands to reason that other HAT agents that do not
possess this ability would afford "scattershot" fluorination if
they were to afford anything at all. A good example would
be the free radical derived from NFSI; it is a known HAT
agent,40 and contains no functional groups with particular
affinity for carbonyl coordination. Scattershot fluorination
is in fact the case; a variety of fluorinated products in low
yield is observed in the reaction of NFSI and a model sub-
strate. A similar result is obtained when N-fluoro-
pyridinium triflate is used – small amounts of unselectively
fluorinated products are observed (Figure 10).
Figure 8. Cyclic voltammograms of compounds 9 and 13-a
in dry and de-oxygenated MeCN with 0.1 M TBAPF6 and a
potential sweep rate of 100 mV/s (vs Fc/Fc+).
Carbonyl-coordinated HAT. Considering the aforemen-
tioned results, we are left with viable pathways in the form
of HAT and MS-PCET. The former, HAT through SRD, would
seem to be disfavored based on calculated C-H bond disso-
ciation energies (a factor of HAT capability).38 DFT calcula-
tions on steroid 6 (ωB97XD/6-311++G** in MeCN) show nu-
merous weaker C-H bonds in the presence of the targeted C-
H bond in substrate 6 (Figure 9). However, this analysis is
naïve as it takes no account of steric and electronic factors.
It is the transition state energies that dictate the selectivity
of HAT, not merely BDEs. Is it possible that SRD interacts
with a proximate carbonyl group in a way to organize a lower
energy transition state?
Ox
CH2Cl
N
selective
OAc
Me
Ox
N
N
Me
H
e-
Ha
Me
Hb
scattershot
H
H
Me
H
Hf
AcO
O
Me
H
Me
Me
O
O
O
H
O
N
He
Hh
S
R
S
R
scattershot
O
O
O
Hc
Hd
Figure 10. Product distribution classified as "selective" or
"scattershot" in reactions involving presumed N-centered
radical intermediates.
Hd
<
Ha
<
Hf
<
Hc
<
He
<
Hb
<
Hh
calculated bond dissociation energies
Figure 9. Relative calculated C-H bond dissociation ener-
gies (weakest to strongest from left to right).
What about an outer-sphere PCET (MS-PCET)? Although
the NFSI-derived free radical may not be sufficiently ener-
getic to engage in PCET, using more reactive analogues of
NFSI (e. g. the DesMarteau reagent)41 produces a similar pat-
tern of scattershot fluorination (Scheme 21). This is notable
as the DesMarteau reagent should be energetically capable
In order to address this issue, we undertook transition
state calculations on a model substrate. In less rigid systems
(for example, entities containing a floppy side chain), the
carbonyl is not adequately locked, and activation may occur
ACS Paragon Plus Environment

Page 11 of 19
1
Journal of the American Chemical Society
of engaging in MS-PCET. Once again, this consideration ap-
Calculations. In the mechanistic study of the carbonyl-di-
2
3
4
5
6
7
8
9
plies to N-fluoropyridinium. On the other hand, in a true MS-
PCET system such as that of Mayer and coworkers, the de-
sired event occurred in the presence of a wide variety of
one-electron oxidants as long as they possessed the correct
potentials. In our case, it is clear that the nature of the reac-
tion is not dependent on oxidation potential, but on chemi-
cal structure. This would seem to be strong evidence for a very
selective and special version of HAT or inner-sphere PCET.
rected reaction, quantum calculations were always destined
to play an important role. If our hypothesis of carbonyl-di-
rected HAT is true, then the corresponding HAT transition
states should be the lowest energy of all reasonable sites in
the molecule. Sampling all chemically distinct hydrogen at-
oms at a sufficient level of theory is a tall order; however, we
strove to be as comprehensive as possible. Additionally,
each carbonyl-activated site is a methylene unit containing
diastereotopic protons; it is quite possible that only one of
whose abstractions would be favored. Take, for example,
model substrate 1 (Figure 11). Abstraction of the α proton
is favored by 2.9 kcal/mol over the β proton. Abstractions
of protons from other logical sites are relatively disfavored
(see SI for structures). The reason is evident from both the
geometry and energetics of the transition states – two key
C–H···O=C hydrogen bonds anchor and stabilize the assem-
bly in the correct orientation. This deduction was supported
by second order perturbation theory analysis of Fock matrix
in natural bond orbital (NBO) basis for alpha and beta man-
ifolds. These hydrogen bonds, involving slightly acidified
protons on SRD, are calculated to be worth 4–5 kcal/mol —
more than enough to torque the system toward directed ab-
straction. The other stabilizing interaction exists between
the transferring hydrogen atom and the carbonyl group.
This is a weak H-bond in its own right and a contributing
factor to the transition state stability as well. In the case of
the lowest energy transition state, all three interactions are
a bit tighter. The transferring hydrogen carries a calculated
partial positive charge of 0.41, which is not unusual for
HAT.42 As for potential inner-sphere PCET, the theoretical
criteria of Mayer and coworkers would seem to disfavor this
possibility.43
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
OAc
OAc
Me
Me
Me
Me
H
H
NFSI, MeCN
hv (300 nm)
F
H
H
H
H
AcO
AcO
O
O
numerous fluorinated products in low yield
R = Ph (NFSI), not energetic enough for
PCET, HAT only
O
O
N
S
R
S
O
O
R = CF3 (DesMarteau reagent), energetic
enough for PCET, but still no selectivity
R
Scheme 21. Reactivity of NFSI derivatives differs from that
of Selectfluor.
Evidence for a Chain Reaction. After the critical HAT step,
the substrate is left with a free radical (Scheme 22, com-
pound 14). The final step of the chain propagation is well
established through prior studies: reaction between the re-
sultant radical and Selectfluor to yield an alkyl fluoride and
to regenerate SRD (Scheme 22, compound 15).29 To verify a
chain process, we calibrated the quantum yield F of the
standard reaction (SF, benzil, 400 nm LED, MeCN) against
the photodecomposition of lime green potassium ferrioxa-
late, a well-established chemical actinometer, and found F =
18.
CH2Cl
C8H17
Me
N
Me
2BF4
H
sensitized pathway:
initation via 400 nm
N
H
H
H
H
AcO
O
*
O
HAT
O
O
C8H17
Me
R
F
CH2Cl
N
SF
SF
Me
H
2BF4
H
H
SRD production
14
N
AcO
O
15
*
C8H17
Me
fluorination
SRD regeneration
Me
H
CH2Cl
N
C8H17
H
H
Me
AcO
O
Me
H
2BF4
direct photolysis pathway:
initiation via 300 nm
N
H
H
F
F
AcO
O
Scheme 22. Chain propagation mechanism at play.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 19
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 12. HAT transition state (top) and intrinsic reaction coordinate (IRC) derived evolution of HAT reaction coordinate
(bottom) with molecular electrostatic potential (MEP) surfaces (isovalue = 0.001, min = 50.0 and max = 110.0) superimposed
on spin orbital densities (isovalue = -0.0045). The transition state optimization and IRC computed at UB3LYP/6-31G(d). The
MEP surfaces and spin orbital densities computed by single point calculations at the UB3LYP-D3/6-311++G(2d,2p).
OAc
negatively charged carbonyl oxygen. Conversely, if this
were a PCET mechanism, it would require the transferring
hydrogen to be a part of a four-electron, three-center array
more consistent with hydrogen bonding.44 In this instance
the carbonyl oxygen would need to have significant radical
cationic character, or at least positive charge build-up rela-
tive to the substrate.
Me
Me
H
1
H
H
H
H
AcO
O
1.40 Å
1.34 Å
N
H
H
N
2.14 Å
2.18 Å
Figure 12 also depicts the evolution of the HAT reaction
coordinate, with images of the spin orbital density with the
electrostatic potential (blue positive, red negative) superim-
posed. Beginning at the bottom right is a precomplex
wherein, there is no orbital density indicative of HAT; in-
stead it looks to be a scenario primed for PCET or at the least
intramolecular hydrogen abstraction by the carbonyl oxy-
gen (NBO analysis shows a very small donation of carbonyl
oxygen lone pair electron density to the C–H bond involved
in hydrogen abstraction of 2.3 kcal/mol). As one moves
along the bottom of the figure toward the left-hand side,
HAT orbital density emerges and at the transition state (top
of Figure 12) HAT type bonding is clearly visible, respec-
tively.
CH2Cl
ΔG‡ = 2.9 kcal/mol
IEFPCM(CH3CN)UωB97X-D/6-311++G(2d,2p)/UB3LYP/6-31G(d)
Figure 11. Comparing transition state energies involving
HAT from each of the diastereotopic protons at C15 of a pro-
totypical enone-containing substrate.
Figure 12 shows an image of the computed spin orbital
density of TS assembly TS-1 with its electrostatic potential
superimposed. This distribution of spin tells an interesting
story - namely, in this transition state, the cationic-radical
nitrogen atom to which the hydrogen is being donated sig-
nificantly lacks electron density — indicative of partial pos-
itive charge — while there is a larger degree of electron den-
sity at the transferring hydrogen atom, thus, suggestive of
negative charge build-up. The developing carbon-centered
radical, on the other hand, is essentially neutral. This spin
orbital density and imbedded charge distribution are indic-
ative of a three-electron, three-center transition state. Rec-
ollect that the lone pair of the distal carbonyl oxygen is di-
rectly pointed at the C–H bond undergoing homolytic cleav-
age; associated with this interaction is favorable Coulombic
attraction between the cationic SRD species and the
ACS Paragon Plus Environment

Page 13 of 19
Journal of the American Chemical Society
1
2
3
4
5
6
7
1) LDA, TMSCl, then Pd(OAc)2
O
OAc
OAc
Me
H
Me
Me
2) D2, Pd/C
3) cat. Na2CO3, H2O
Me
H
Me
H
Me
H
+
4) NaBH4
H
H
H
H
H
D
D
5) Ac2O, pyridine
6) K2Cr2O7, NHS, AcOH
17-β
17-α
TMSO
AcO
O
AcO
O
from dehydroepiandrosterone
1.6/1.0
8
9
10
11
12
13
OAc
OAc
Me
H
Me
Selectfluor
benzil, MeCN
Me
H
Me
H
+
H
F
H
H
F
hv (400 nm)
H
D
18
AcO
O
2
AcO
O
1.6/1.0
Scheme 23. Isotopic-labeling experiment confirming preferential abstraction of the a-hydrogen and deuterium atoms, con-
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
sistent with DFT-predicted HAT hypothesis.
The overall analysis holds for rigid ketones as well (Figure
13). Selective abstraction from ketone 16 is unusual in that
the calculated energies between diastereotopic transition
states is sizeable (3.6 kcal/mol). In the high energy transi-
tion state, the anchoring effect of the two acidic C–H bonds
on SRD is greatly attenuated; one of the contacts is severed
completely.
allylic oxidation with potassium dichromate, N-hydroxy-
succinimide (NHS), and AcOH to produce the product as a
mixture of bD/aD isomers of 1.6/1 (compound 17). The as-
signment of isotopomers was made by 2H NMR, with confir-
mation by 1H NMR (see SI). The use of benzene-d6 as solvent
greatly aids the assignment of diagnostic signals.
Upon standard reaction conditions, remarkably, substrate
17 is converted to a 1.6/1 ratio of labeled isomeric products
(compounds 18 and 2). The result is exactly in line with
what we would expect for preferential abstraction of the a-
hydrogen and deuterium atoms. Preferential b-abstraction,
on the other hand, would have yielded the opposite result.
CH2Cl
2.25 Å
N
1.28 Å
2.29 Å
H
N
H
O
1.45 Å
Conclusion: We have explored the mechanistic possibilities
of our previously reported enone/ketone-directed site-se-
lective sp3 C-H fluorination of terpenoid derivatives. Our
findings suggest intermolecular hydrogen atom transfer
(HAT) chemistry is at play, rather than classical Norrish hy-
drogen atom abstraction as initially conceived. Isotope ef-
fect studies, detailed quantum computations, thermochemi-
cal experiments, and reactions with one-electron oxidants
all point conclusively toward a special type of HAT mecha-
nism in which SRD approaches the targeted C-H bond by co-
ordinating to the proximate carbonyl group. This interest-
ing form of HAT may mimic such venerable reactions as the
Norrish II cleavage and the McLafferty reaction but in actu-
ality is quite different. Finally, this principle of selective, di-
rected HAT may be leveraged in the interaction of SRD and
related radical cations with other functional groups in
works to follow.
Me
Me
H
16
Me
TsO
ΔG‡ = 3.6 kcal/mol
IEFPCM(CH3CN)
U
ω
B97X-D/6-311++G(2d,2p)/UB3LYP/6-31G(d)
Figure 13. Comparing transition state energies involving
HAT from each of the diastereotopic protons of a rigid, non-
steroidal ketone-containing substrate.
Although the magnitude of the KIE in deuterated 2-ethyl-
cyclohexanone indicates a primary isotope effect (Scheme
18), it occurred to us that substrate 1 would provide a way
to affirm our computational predictions based on selective
isotopic labeling. For a more compelling and less ambiguous
case, we sought to make a specifically deuterated steroidal
substrate optimally poised for fluorination (Scheme 23).
For example, the unlabeled variant of 1 is predicted to fluor-
inate through preferential abstraction of the b-C15 hydro-
gen atom. This prediction could be verified by a simple la-
beling experiment. In the event, we began the synthesis
with dehydroepiandrosterone (DHEA). Enolization was fol-
lowed by Pd(II)-catalyzed oxidation to form the enone. Re-
duction with D2, followed by base-catalyzed exchange of the
C16 hydrogen atoms, produced a mixture of isotopomers in
a 1.6/1 ratio. The synthesis is completed by stereospecific
reduction of the ketone carbonyl, acetylation, and standard
ASSOCIATED CONTENT
Supporting Information. The Supporting Information con-
taining experimental procedures, spectra, computational
data, and crystallographic data for compound 7 is available
free of charge on the ACS Publications website at DOI:
AUTHOR INFORMATION
Corresponding Authors
Thomas Lectka - Department of Chemistry, Johns Hopkins
University, 3400 N. Charles St., Baltimore, MD, 21218,
United States; Email: lectka@jhu.edu
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 19
1
2
3
4
5
6
7
8
Travis Dudding - Department of Chemistry, Brock Univer-
sity, 1812 Sir Isaac Brock Way, St. Catharines, ON L2S 3A1,
Canada; Email: tdudding@brocku.ca
Electron Transfer Reactions. Chem. Phys. Chem. 2002, 3, 33-
42.
4 Pitts, C.; Bume, D.; Harry, S.; Siegler M.; Lectka, T. Multiple
Enone-Directed Reactivity Modes Lead to the Selective Pho-
tochemical Fluorination of Polycyclic Terpenoid Deriva-
tives. J. Am. Chem. Soc. 2017, 139, 2208–2211.
Authors’ Notes
C.R.P. current address: Department of Chemistry, Scripps
Research Institute, 10550 North Torrey Pines Road, La Jolla,
California 92037, United States.
5
(a) Bume, D. D.; Pitts, C. R.; Ghorbani, F.; Harry, S. A.;
9
Capilato, J. N.; Siegler, M. A.; Lectka, T. Ketones as Directing
Groups in Photocatalytic sp3 C-H Fluorination. Chem. Sci.
2017, 8, 6918-6923. (b) Bume, D. D.; Harry, S. A.; Pitts, C. R.;
Lectka, T. Sensitized Aliphatic Fluorination Directed by Ter-
penoidal Enones: A 'Visible Light' Approach. J. Org. Chem.
2018, 83, 1565-1575. (c) Bume, D. D.; Harry, S. A.; Lectka, T.;
Pitts, C. R. Catalyzed and Promoted Aliphatic Fluorination. J.
Org. Chem. 2018, 83, 8803-8814. (d) Capilato, J. N.; Pitts, C.
R.; Rowshanpour, R.; Dudding, T.; Lectka, T. Site-Selective
Photochemical Fluorination of Ketals: Unanticipated Out-
comes in Selectivity and Stability. J. Org. Chem. 2020, 85,
2855-2864.
Notes
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The authors declare no competing financial interest.
ACKNOWLEDGMENT
T.L. thanks the National Science Foundation (NSF) (Grant
No. CHE 1800510) for financial support. Mass spectral data
were obtained at University of Delaware’s mass spectrome-
try centers. T.D. acknowledges financial support from the
Natural Science and Engineering Research Council (NSERC)
Discovery Grant (2019-04205). Computations were carried
out using facilities at SHARCNET (Shared Hierarchical Aca-
demic Research Computing Network: www.sharcnet.ca)
and Compute/Calcul Canada. The authors kindly thank Dr.
David Miller (Caltech) for insightful conversations.
6 (a) Norrish, R. G. W.; Bamford, C. H. Photo-decomposition
of Aldehydes and Ketones. Nature 1937, 140, 195-196. (b)
Sauers, R. R.; Edberg, L. A. Modeling of Norrish Type II Reac-
tions by Semiempirical and ab Initio Methodology. J. Org.
Chem. 1994, 59, 7061-7066. (c) Wagner. P. J. Type II Photo-
elimination and Photocyclization of Ketones. Acc. Chem.
Res. 1971, 4, 168-177. (d) Turro, N. J.; Dalton, J. C.; Dawes,
K.; Farrington, G.; Hautala, R.; Morton, D.; Niemczyk, M.;
Schore, N. Molecular Photochemistry of Alkanones in Solu-
tion: alpha-Cleavage, Hydrogen Abstraction, Cycloaddition,
and Sensitization Reactions. Acc. Chem. Res. 1972, 5, 92-101.
(e) Ihmels, H; Scheffer, J. R. The Norrish Type II Reaction in
the Crystalline State: Toward a Better Understanding of the
Geometric Requirements for γ-Hydrogen Atom Abstraction.
Tetrahedron 1999, 55, 885-907.
REFERENCES
(a) Zhang, C; Tang, C.; Jiao, N. Recent Advances in Copper-
Catalyzed Dehydrogenative Functionalization via a Single
Electron Transfer (SET) Process. Chem. Soc. Rev. 2012, 41,
3464-3484. (b) Yoon, C.; Mariano, P. Mechanistic and Syn-
thetic Aspects of Amine-Enone Single Electron Transfer
Photochemistry. Acc. Chem. 1992, 25, 233-240. (c) Suess, A.;
Ertem, M.; Cramer, C.; Stahl, S. Divergence Between Organo-
metallic and Single-Electron-Transfer Mechanisms in Cop-
per (II)-Mediated Aerobic C–H Oxidation. J. Am. Chem. Soc.
2013, 26, 9797-9804. (d) Lewis, F. Proton-Transfer Reac-
tions of Photogenerated Radical Ion Pairs. Acc. Chem. 1986,
19, 401-405. (e) Klein, E.; Lukeš, V. DFT/B3LYP Study of the
Substituent Effect on the Reaction Enthalpies of the Individ-
ual Steps of Single Electron Transfer−Proton Transfer and
Sequential Proton Loss Electron Transfer Mechanisms of
Phenols Antioxidant Action. J. Phys. Chem. A 2006, 110,
12312-12320. (f) Liu, W.; Huang, X.; Cheng, M.-J.; Nielsen, R.
J.; Goddard, W. A., III; Groves, J. T. Science 2012, 337,
1322−1325.
7 Miller, D.; Tarantino, K.; Knowles, R. Proton-Coupled Elec-
tron Transfer in Organic Synthesis: Fundamentals, Applica-
tions, and Opportunities. Top. Curr. Chem. 2016, 30, 374.
8 Bloom, S.; Pitts, C. R.; Miller, D.; Haselton, N.; Holl, M. G.; Ur-
heim, E.; Lectka, T. A Polycomponent Metal-Catalyzed Ali-
phatic, Allylic, and Benzylic Fluorination. Angew. Chem. Int.
Ed. 2012, 51, 10580-10583.
9 Pitts, C. R.; Bloom, M. S.; Bume, D. D.; Zhang, Q. A.; Lectka, T.
Unstrained C–C Bond Activation and Directed Fluorination
Through Photocatalytically-Generated Radical Cations.
Chem. Sci. 2015, 6, 5225-5229. See also: Pitts, C. R.; Ling, B.;
Snyder, J. A.; Bragg, A. E.; Lectka, T. Aminofluorination of Cy-
clopropanes: A Multifold Approach through a Common, Cat-
alytically Generated Intermediate. J. Am. Chem.
Soc. 2016, 138, 6598-6609.
2
(a) Warren, J.; Tronic, T.; Mayer, J. Thermochemistry of
Proton-Coupled Electron Transfer Reagents and its Implica-
tions. Chem. Rev. 2010, 110, 6961-7001. (b) Hammes-
Schiffer, S.; Stuchebrukhov, A. Theory of Coupled Electron
and Proton Transfer Reactions. Chem. Rev. 2010, 110, 6939-
6960.
3 (a) Hammes-Schiffer, S.; Iordanova, N. Theoretical Studies
of Proton-Coupled Electron Transfer Reactions. Biochim. Bi-
ophys. Acta 2004, 1655, 29-36. (b) Hammes-Schiffer, S. Com-
parison of Hydride, Hydrogen Atom, and Proton-Coupled
10 (a) Molnar, I.; Gilmour, R. Catalytic Difluorination of Ole-
fins. J. Am. Chem. Soc. 2016, 138, 5004−5007. (b) Crossley,
S.; Obradors, C.; Martinez, R.; Shenvi, R. Mn-, Fe-, and Co-Cat-
alyzed Radical Hydrofunctionalizations of Olefins. Chem.
ACS Paragon Plus Environment

Page 15 of 19
1
Journal of the American Chemical Society
Rev. 2016, 116, 8912−9000. (c) Nyffeler, P.T.; Durón, S.G.;
Photolysis of Benzil and 4-Hydroxybenzil. Can. J. Chem.
2
Burkart, M.D.; Vincent, S.P.; Wong, C.H. Selectfluor: mecha-
nistic insight and applications. Angew. Chem. Int. Ed. 2005,
44, 192-212. (d) Danahy, K.E.; Cooper, J.C.; Van Humbeck, J.F.
Benzylic Fluorination of Aza-Heterocycles Induced by Sin-
gle-Electron Transfer to Selectfluor. Angew. Chem. Int. Ed.
2018, 57, 5134-5138. (e) Kee, C.W.; Chin, K.F.; Wong, M.W.;
Tan, C.H. Selective fluorination of alkyl C–H bonds via pho-
tocatalysis. Chem. Comm. 2014, 50, 8211-8214. (f) Amaoka,
Y.; Nagatomo, M.; Inoue, M. Metal-free fluorination of C
(sp3)–H bonds using a catalytic N-oxyl radical. Org. Lett.,
2013, 15, 2160-2163. (g) Brunet, V.; O’Hagan, D. Angew.
Chem. Int. Ed. 2008, 47, 1179 – 1182.
1972, 50, 2499-2510. (b) Bunbury, D. L.; Wang, C. T. The
Photolysis of Benzil in Cyclohexane Solution. Can. J. Chem.
1968, 46, 1473-1479. (c) Encinas, M. V.; Scaiano, J. C. Laser
Photolysis Study of the Exciplex between Triplet Benzil and
Triethylamine. J. Am. Chem. Soc. 1979, 101, 7740-7741. (d)
Mohapatra, G. K. D.; Bhattacharya, J.; Bandopadhyay, J.; Bera,
S. C. Flash Photolysis of Benzils. J. Photochem. 1987, 40, 47-
58. (e) McGimpsey, W. G.; Scaiano, J. C. A Two-Photon Study
of the “Reluctant" Norrish Type I Reaction of Benzil. J. Am.
Chem. Soc. 1987, 109, 2179-2181. (f) Scaiano, J. C.; Johnston,
L. J.; McGimpsey, W. G.; Weir, D. Photochemistry of Organic
Reaction Intermediates: Novel Reaction Paths Induced by
Two-Photon Laser Excitation. Acc. Chem. Res. 1988, 21, 22–
29. (g) Mukai, M.; Yamauchi, S.; Hirota, N. Time-Resolved
EPR Study on the Photochemical Reactions of Benzil. J. Phys.
Chem. 1992, 96, 3305-3311. (h) Adam, W.; Oestrich, R. S.
Two-Photon Cleavage of Benzil in the Laser-Jet: Intermolec-
ular Reactions of Transient Benzoyl and tert-Butoxy Radi-
cals in the Photolysis of tert-Butyl Peroxide Mixtures. J. Am.
Chem. Soc. 1993, 115, 3455-3457. (i) Mukai, M.; Yamauchi,
S.; Hirota, N. A Time-Resolved ERR Study of One- and Two-
Photon Processes in the Photochemical Reactions of Benzil.
J. Phys. Chem. 1989, 93, 4411-4413. (j) Kosa, C.; Lukác, I. Pho-
tophysical and Photochemical Properties of Benzil. Chem.
Listy. 1996, 90, 287-294. (k) Furusawa, T.; Kawano, M.; Fu-
jita, M. The Confined Cavity of a Coordination Cage Sup-
presses the Photocleavage of α-Diketones to Give Cycliza-
tion Products through Kinetically Unfavorable Pathways.
Angew. Chem. Int. Ed. 2007, 46, 5717-5719. (l) Volman, D. H.;
Hammond, G. S.; Gollnick, K. A Model for the Influence of Or-
ganized Media on Photochemical Reactions. Advances in
Photochemistry, Volume 16, 2009; Wiley; pp 85-87. (m)
Simburgera, H.; Kern, W.; Hummel, K.; Hagg, C. Photoreac-
tions in Polymers Containing Benzil Units: A Comparative
Study Under Excimer Laser and Hg-Lamp Irradiation. Poly-
mer 2000, 41, 7883-7897 and references therein.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11 (a) Tarantino, K.; Liu, P.; Knowles, R. Catalytic Ketyl-Olefin
Cyclizations Enabled by Proton-Coupled Electron Transfer.
J. Am. Chem. Soc. 2013, 135, 10022-10025. (b) Gentry, E.;
Knowles, R. Synthetic Applications of Proton-Coupled Elec-
tron Transfer. Acc. Chem. Res. 2016, 49, 1546-1556. (c)
Yayla, H.; Wang, H.; Tarantino, K.; Knowles, R. Catalytic Ring-
Opening of Cyclic Alcohols Enabled by PCET Activation of
Strong O–H Bonds. J. Am. Chem. Soc. 2016, 138, 10794-
10797.
12 (a) Markle, T. F.; Darcy, J. W.; Mayer, J. M. A New Strategy
to Efficiently Cleave and Form C–H Bonds Using Proton-Cou-
pled Electron Transfer Sci. Adv. 2018, 4, 5776. (b) Darcy, J.
W.; Kolmar, S. S.; Mayer, J. M. Transition State Asymmetry in
C–H Bond Cleavage by Proton-Coupled Electron Transfer. J.
Am. Chem. Soc. 2019, 141, 10777−10787.
13 Darcy, J. W.; Koronkiewicz B.; Parada G.; Mayer J. A Con-
tinuum of Proton-Coupled Electron Transfer Reactivity. Acc.
Chem. Res. 2018, 51, 2391-2399.
4
Kee, C. W.; Chin, K. F.; Wong, M. W.; Tan, C. H. Selective
Fluorination of Alkyl C–H Bonds via Photocatalysis. Chem.
Comm. 2014, 50, 8211-8214.
15 Niu, L.; Liu, J.; Liang, X. A.; Wang, S.; Lei, A. Visible Light-
Induced Direct α C–H Functionalization of Alcohols. Nature
Comm. 2019, 10, 467.
9 Yang, Y.; Liu, L.; Chen, J.; Han, K. Hydrogen Bonding Tunes
the Early Stage of Hydrogen-Atom Abstracting Reaction.
Phys. Chem. Chem. Phys. 2014, 16, 17828.
6 Meanwell, M.; Lehmann, J.; Eichenberger, M.; Martin, R. E.;
Britton, R. Synthesis of Acyl Fluorides via Photocatalytic
Fluorination of Aldehydic C–H Bonds. Chem. Comm. 2018,
54, 9985-9988.
20 Leigh, W. J.; Lathioor, E. C.; St. Pierre, M. J. Photoinduced
Hydrogen Abstraction from Phenols by Aromatic Ketones. A
New Mechanism for Hydrogen Abstraction by Carbonyl n,π*
and π,π* Triplets. J. Am. Chem. Soc. 1996, 118, 12339-12348.
7 (a) Herkstroeter W. G.; Lamola A. A.; Hammond G. S. Mech-
anisms of Photochemical Reactions in Solution. XXVIII. Val-
ues of Triplet Excitation Energies of Selected Sensitizers. J.
Am. Chem. Soc. 1964, 86, 4537–4540. (b) Schuster, D. I.;
Dunn, D. A.; Heibel, G. E.; Brown, P. B.; Rao, J. M.; Woning, J.;
Bonneau, R. Enone Photochemistry. Dynamic Properties of
Triplet Excited States of Cyclic Conjugated Enones as Re-
vealed by Transient Absorption Spectroscopy. J. Am. Chem.
Soc. 1991, 113, 6245−6255. (c) Evans, T. R.; Leermakers, P.
E. Emission Spectra and Excited-State Geometry of alpha-
Diketones. J. Am. Chem. Soc. 1967, 89, 4380−4382.
21 Zhang, X.; Ma, J.; Li, S.; Li, M. D.; Guan, X.; Lan, X., Phillips,
D. L. Ketyl Radical Formation via Proton-Coupled Electron
Transfer in an Aqueous Solution versus Hydrogen Atom
Transfer in Isopropanol after Photoexcitation of Aromatic
Carbonyl Compounds. J. Org. Chem. 2016, 81, 5330-5336.
22 Huggenberger, C.; Lipscher, J.; Fischer, H. Self-Termination
of Benzoyl Radicals to Ground- and Excited-State Benzil.
Symmetry Control of a Radical Combination. J. Phy. Chem.
1980, 84, 3467-3474.
23 (a) Saltiel, J., and H. C. Curtis. Photooxidation of Benzil. Mol.
Photochem. 1969, 1, 239-243. (b) Lukáč, I.; Kósa, C. The
8
(a) Bunbury D. L.; Chan T. M. The Photolysis of 4-Hy-
droxybenzil and the Effect of Triethylamine on the
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 19
1
2
3
4
5
6
7
8
Formation of Dibenzoyl Peroxide by Photooxidation of Ben-
Reagents and its Implications. Chem. Rev. 2010, 110, 6961–
7001.
zil in a Polymer Film. Macromol. Rapid Commun. 1994, 15,
929-934. (c) Cosa, G.; Scaiano, J. C. Photochemistry of
Diketones: Observation of a Triplet State−Oxygen Adduct. J.
Am. Chem. Soc. 2004, 126, 8636–8637.
33 (a) Williams. H.; Wilson, J. M.; Budzikiewicz H.; Djerassi, C.
Mass Spectrometry in Structural and Stereochemical Prob-
lems. XXIV. A Study of the Hydrogen Transfer Reactions Ac-
companying Fragmentation Processes of 11-Keto Steroids.
Synthesis of Deuterated Androstan-11-ones. J. Am. Chem.
Soc. 1963, 85, 2091-2105. (b) Djerassi, C.; Mutzenbecher; G.
von; Fajkos, J.; Williams, D. H.; Budzikiewicz H. Mass Spec-
trometry in Structural and Stereochemical Problems. LXV.
Synthesis and Fragmentation Behavior of 15-Keto Steroids.
The Importance of Interatomic Distance in the McLafferty
Rearrangement. J. Am. Chem. Soc. 1965, 87, 817.
24 (a) Pitts, C. R.; Bloom, S.; Woltornist, R.; Auvenshine, D. J.;
Ryzhkov, L. R.; Siegler, M. A.; Lectka, T. Direct, Catalytic Mon-
ofluorination of sp3 C–H Bonds: A Radical-Based Mechanism
with Ionic Selectivity. J. Am. Chem. Soc. 2014, 136, 9780-
9791. (b) Rueda-Becerril, M.; Sazepin, C. C.; Leung, J. C. T.;
Okbinoglu, T.; Kennepohl, P.; Paquin, J.-F.; Sammis, G. M. Flu-
orine Transfer to Alkyl Radicals. J. Am. Chem. Soc. 2012, 134,
4026−4029.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
34 (a) Loos, J.; Schröder, D.; Schwarz, H. Diastereoselectivity
in the McLafferty Rearrangement of Photoionized 3-Methyl
Valeramide. J. Chem. Org. 2005, 70, 1073-1076. (b) Loos, J.;
Schröder, D. Schwarz, H.; Thissen, R.; Dutuit, O. Competitive
Reactions and Diastereoselective CH Bond Activation in the
McLafferty Rearrangement of Photoionized 3-Methyl Val-
eramide. Int. J. Mass Spectrom. 2005, 240, 121-137. (c) Loos,
J.; Schröder, D.; Zummack, W.; Schwarz, H.; Thissen, R.; Du-
tuit, O. Dissociation Behavior of Ionized Valeramide: Part I.
Experimental Studies. Int. J. Mass Spectrom. 2002, 214, 105-
128.
25 Kee, J. W.; Shao, H.; Kee, C. W.; Lu, Y.; Soo, H. S.; Tan, C.-H.
Mechanistic Insights for the Photoredox Organocatalytic
Fluorination of Aliphatic Carbons by Anthraquinone Using
Time-Resolved and DFT Studies. Catal. Sci. Technol. 2017, 7,
848.
26 Padhye, M.; McGlynn, S.; Kasha, M. Lowest Triplet State of
Anthracene. J. Chem. Phys. 1956, 24, 588.
27
Parker, C.; Joyce, T. Formation Efficiency and Energy of the
Perylene Triplet. Chem. Comm. 1966, 4, 108.
35 Djerassi, C.; Tökés, L. Mass Spectrometry in Structural and
Stereochemical Problems. XCIII. Further Observations on
the Importance of Interatomic Distance in the McLafferty
Rearrangement. Synthesis and Fragmentation Behavior of
Deuterium-Labeled 12-Keto Steroids. J. Am. Chem. Soc.
1966, 88, 536-544.
36 Hammerum, S. Alkyl Radicals as Hydrogen Bond Acceptors:
Computational Evidence. J. Am. Chem. Soc. 2009, 131, 8627–
8635.
28 Pearson, A. J.; Xiao, W. Fluorescent Photoinduced Electron
Transfer (PET) Sensing Molecules with p-Phenylenediamine
as Electron Donor. J. Org. Chem. 2003, 68, 5361-5368
29 (a) Zhang, Z. C.; Chung, T. M. Reaction Mechanism of Bo-
rane/Oxygen Radical Initiators during the Polymerization of
Fluoromonomers. Macromolecules 2006, 39, 5187−5189.
(b) Mirviss, S. B. Mechanism of the Oxidation of Trialkyl-
boranes. J. Org. Chem. 1967, 32, 1713−171. (c) Szpera, R.;
Moseley, D.; Smith, L.; Sterling, A.; Gouverneur, V. The Fluor-
ination of C-H Bonds: Developments and Perspectives. An-
gew. Chem. Int. Ed. 2019, 58, 14824–14848. (d) Troyano, F.;
37 Simmons, E.; Hartwig, J. On the Interpretation of Deuterium
Kinetic Isotope Effects in C-H Bond Functionalizations by Tran-
sition-Metal Complexes Angew. Chem. Int. Ed. 2012, 51, 3066 –
3072.
Merkens, K.; Gómez-Suárez, A. Selectfluor® Radical Dication
(TEDA2+.) – A Versatile Species in Modern Synthetic Or-
ganic Chemistry. Asian J. Org. Chem. 2020, 9, 1–17.
38
Jing, L.; Nash, J.; Kenttämaa, H. Correlation of Hydrogen-
Atom Abstraction Reaction Efficiencies for Aryl Radicals
with their Vertical Electron Affinities and the Vertical Ioni-
zation Energies of the Hydrogen-Atom Donors. J. Am. Chem.
Soc. 2008, 130, 17697-17709.
30
Pitts, C. R.; Ling, B.; Woltornist, R.; Liu, R.; Lectka T. Tri-
ethylborane-Initiated Radical Chain Fluorination: A Syn-
thetic Method Derived from Mechanistic Insight. J. Org.
Chem. 2014, 79, 8895-8899.
39 (a) Walling, C. Free Radicals in Solution, Wiley, New York,
NY, 1957. (b) Bernardi, R.; Galli, R.; Minisci, F. Polar and Ste-
ric Effects in Hydrogen Abstraction by Dialkylamine Cation
Radicals. J. Chem. Soc. B 1968, 324–325. (c) Minisci, F.; Galli,
R.; Bernardi, R. Polar Effects in Radical Reactions: A New Se-
lective Type of Radical Bromination. Chem. Commun. 1967,
903–904. (d) Zavitsa, A.; Pinto, J. The Meaning of the “Polar
Effect” in Hydrogen Abstractions by Free Radicals. Reac-
tions of the tert-Butoxy Radical. J. Am. Chem. Soc. 1972, 94,
7390–7396. (e) Newhouse, T; Baran, P. If C-H Bonds Could
Talk: Selective C-H Bond Oxidation. Angew. Chem. Int. Ed.
2011, 50, 3362–3374.
31 Takahira, Y.; Chen, M.; Kawamata, Y.; Mykhailiuk, P.; Naka-
mura, H.; Peters, B. K.; Reisberg, S. H.; Li, C.; Chen, L.; Hoshi-
kawa, T.; Shibuguchi, T.; Baran P. S. Electrochemical C(sp3)–
H Fluorination. Synlett 2019, 30, 1178-1782.
32 (a) Weinberg, D. R.; Gagliardi, C. J.; Hull, J. F.; Murphy, C. F.;
Kent, C. A.; Westlake, B. C.; Paul, A.; Ess, D. H.; McCafferty, D.
G.; Meyer, T. Proton-Coupled Electron Transfer. J. Chem. Rev.
2012, 112, 4016–4093. (b) Warren, J.; Tronic, T.; Mayer, J.
Thermochemistry of Proton-Coupled Electron Transfer
ACS Paragon Plus Environment

Page 17 of 19
Journal of the American Chemical Society
1
2
3
4
40
Nodwell, M B.; Bagai, A.; Halperin, S. D.; Martin, R. E.;
Versus HAT: A Direct Theoretical Method for Distinguishing
X-H Bond-Activation Mechanisms. Angew. Chem. Int. Ed.
2018, 57, 11913–11917.
Knust, H.; Britton, R. Direct Photocatalytic Fluorination of
Benzylic C–H Bonds with N-Fluorobenzenesulfonimide.
Chem. Commun. 2015, 51, 11783–11786.
5
41 Singh, S.; DesMarteau, D.; Zuberi, S.; Witz, M.; Huang, H. N-
Fluoroperfluoroalkylsulfonimides. Remarkable New Fluori-
nation Reagents. J. Am. Chem. Soc. 1987, 109, 7194-7196.
6
7
8
9
42 Tishchenko, O.; Truhlar, D. G.; Ceulemans, A.; Nguyen; M. A
Unified Perspective on the Hydrogen Atom Transfer and
Proton-Coupled Electron Transfer Mechanisms in Terms of
Topographic Features of the Ground and Excited Potential
Energy Surfaces as Exemplified by the Reaction between
Phenol and Radicals. J. Am. Chem Soc. 2008, 130, 7000–7010.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
43 Mayer J. M. Simple Marcus-Theory-Type Model for Hydro-
gen-Atom Transfer/Proton-Coupled Electron Transfer. J.
Phys. Chem. Lett. 2011, 2, 1481–1489.
44 (a) Harcourt, R. D. Four-Electron Three-Center Bonding:ꢀ
One-Electron and Concerted Two-Electron Delocalizations
into Bonding and Antibonding Molecular Orbitals. J. Phys.
Chem. A 1999, 103, 4293–4297. (b) Klein, J.; Knizia, G. cPCET
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 19
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
OAc
Me
photo-, electro-, or chemical conditions
site-selective sp3 fluorination
radical chain mechanism
coordinated HAT
Me
H
H
H
H
H
AcO
O
H
R
F
R
N
O
O
H
H
Selectfluor
N
conditions
CH2Cl
ACS Paragon Plus Environment

Page 19 of 19
Journal of the American Chemical Society
OAc
Me
photo-, electro-, or chemical conditions
site-selective sp3 fluorination
radical chain mechanism
coordinated HAT
Me
H
1
2
H
H
H
H
3
AcO
O
4
5
6
7
8
9
H
R
F
R
N
O
O
H
H
Selectfluor
ACS Paragon Plus Environment
N
conditions
CH2Cl