Triethylborane-initiated radical chain fluorination: A synthetic method derived from mechanistic insight

Article abstract of DOI:10.1021/jo501520e

We offer a mild, metal-free sp³ C-H fluorination alternative using Selectfluor and a substoichiometric amount of triethylborane.an established radical initiator in the presence of O₂. This radical-chain-based synthetic method is particularly noteworthy as an offspring of the insight gained from a mechanistic study of copper-promoted aliphatic fluorination, constructively turning O₂ from an enemy to an ally. Furthermore, BEt₃/O₂ is a preferred initiator in industrial processes, as it is economical, is low in toxicity, and lends way to easier workup.

Full text of DOI:10.1021/jo501520e

Note
pubs.acs.org/joc
Triethylborane-Initiated Radical Chain Fluorination: A Synthetic
Method Derived from Mechanistic Insight
Cody Ross Pitts, Bill Ling, Ryan Woltornist, Ran Liu, and Thomas Lectka*
Department of Chemistry, Johns Hopkins University, 3400 North Charles Street, Baltimore, Maryland 21218, United States
S
* Supporting Information
ABSTRACT: We offer a mild, metal-free sp3 C−H fluorination alternative using
Selectfluor and a substoichiometric amount of triethylboranean established
radical initiator in the presence of O2. This radical-chain-based synthetic method is
particularly noteworthy as an offspring of the insight gained from a mechanistic
study of copper-promoted aliphatic fluorination, constructively turning O2 from an
enemy to an ally. Furthermore, BEt3/O2 is a preferred initiator in industrial
processes, as it is economical, is low in toxicity, and lends way to easier workup.
hough the elucidation of a reaction mechanism can be
troublesome, it can provide the insight into developing a
T
new synthetic method that would otherwise remain undis-
covered. For example, our laboratory recently investigated the
mechanism of the copper(I)/Selectfluor sp3 C−H fluorination
system.1 We reported a detailed scheme whereby Selectfluor 1
is used to generate the radical dication species 2 (via inner-
sphere electron transfer from copper(I), accompanied by loss
of fluoride) responsible for H atom abstraction, generating an
alkyl radical. This radical, in turn, reacts homolytically with
Selectfluor to yield the desired fluorinated product and to
regenerate the radical dication, which propagates the chain.
Considering that copper(I) proved to be unnecessary during
the H atom abstraction and fluorination stages of the
mechanism and the radical dication chain propagator is
generated by a homolytic cleavage of the N−F bond in
Selectfluor, we gathered that a catalytic amount of an
established “radical initiator” in the presence of Selectfluor
and a substrate should also effect C−H fluorination in a similar
fashion (Figure 1). If true, this mechanistic hypothesis permits
us to design a new radical chain fluorination method rationally
by choosing an appropriate initiator. Beyond proof of concept,
alternative manners of initiating the same chain propagation
can be envisioned that are advantageous over existing methods.
A number of radical initiators that might be suitable
surrogates come to mind, including (but not limited to)
halogens, AIBN, organic peroxides, and inorganic peroxides;2
however, more punishing conditions such as heat or ultraviolet
light are typically necessary to generate the radicals, which may
also foster issues regarding selectivity. On the other hand,
triethylborane famously undergoes a homolytic substitution
(SH2) reaction with triplet O2 at room temperature (or lower),
from which an ethyl radical is liberated.3 Conceivably, this ethyl
radical will behave like any other alkyl radical in our system to
create the volatile and easily removed fluoroethane upon
reaction with Selectfluor and the desired radical dication 2, thus
initiating the fluorination reaction. Furthermore, in an industrial
Figure 1. Hypothesized alternative initiation to sp3 C−H fluorination
method.
setting, BEt3/O2 is the preferred radical initiator wherever
possible, as the reagent is atom-economical, is low in toxicity,
and lends way to easier workup.4 Accordingly, we explored the
possibility of effecting this reaction by implementing a catalytic
amount of triethylborane, Selectfluor, and a substrate.
We examined a variety of conditions with cyclododecane 3 as
a test substrate (Table 1) and were satisfied to find that stirring
1.0 equiv of cyclododecane with 20 mol % of triethylborane
(administered as a 1.0 M solution in hexanes) and 2.2 equiv of
Selectfluor in anhydrous MeCN (with no measures taken to
remove dissolved O2) at room temperature under N2 will
produce 1-fluorocyclodecane in 50% yield after 4 h.5 The same
Received: July 9, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jo501520e | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
Table 1. Screening for Reaction Initiation Conditions
Scheme 1
entry
initiator
BEt3
F source
yield (%)
1
2
4
5
6
7
8
Selectfluor
Selectfluor
Selectfluor
NFSI
0
50
0
and afforded no fluorinated products. As triethylborane has also
proven effective for the generation of tin and silyl radicals from
trialkyltin hydrides and trialkylsilanes, we examined the reaction
in the presence of each of these species but found significant
depletions in yield. However, evidence of the Si−F bond being
formed can be seen in the crude 19F NMR spectra, which may
indicate that the silyl radicals were at least formed over the
course of each reaction. Finally, diethylzinc was employed as an
alternative initiator to triethylborane;10 the reaction provided 1-
fluorocyclododecane in a low yield and is also a less desirable
alternative, as dialkylzinc species are notably harsher reagents
than trialkylboranes.
B(sec-butyl)3
BEt3
0
BEt3/Ph3SnH
BEt3/R3SiH
ZnEt2
Selectfluor
Selectfluor
Selectfluor
4
8
4
The success of the fluorination reaction appeared to rely
intimately on (1) the purity of the triethylborane reagent and
(2) the amount of O2 present. Regarding the latter, product
yields diminished using acetonitrile that was subjected to
rigorous freeze−pump−thaw degasification, and notably, the
reaction completely shut down in the presence of air or an O2
atmosphere. If the fluorination reaction is, in fact, initiated by
release of ethyl radicals via autoxidation of triethylborane
(Figure 2), this result should be anticipatedO2 reacts with
triethylborane to produce the ethyl radicals responsible for
initiation but also inhibits propagation of the H atom
abstraction/fluorination steps. Whereas O2 played a solely
deleterious role as a quencher in the copper(I)/Selectfluor
system, the BEt3/O2 radical chemistry assigns it a productive
role in reaction initiation. Moreover, the previously observed
induction period due to O2 quenching in the copper system has
virtually disappeared. In turn, overall reaction times have
satisfyingly decreased. Fortunately, we found that the amount
of dissolved O2 in the solvent at 1 atm ([O2] ≈ 8 mM)11 was
Figure 2. Proposed initiation through formation of ethyl radicals.
reaction was attempted using N-fluorobenzenesulfonimide
(NFSI), another precedented source of atomic fluorine by
Sammis and co-workers,6 but no 1-fluorocyclododecane was
observed, indicating that Selectfluor is a necessary player for H
atom abstraction. The Inoue7 and Chen8 laboratories similarly
observed this dependence on Selectfluor in their aliphatic
fluorination systems (using NDHPI and V2O3, respectively), as
have others using photochemical approaches.9 Other trialkyl-
borane reagents such as tri-sec-butylborane were also screened
Figure 3. Energetic landscape of ethyl radical initiated fluorination of cyclodecane at the B3PW91/6-311++G**(MeCN) level.
B
dx.doi.org/10.1021/jo501520e | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
enough to produce satisfactory results in a standard organic
laboratory setting, obviating the need for a more sophisticated
setup controlling [O2]. While we believe product yields and
reaction efficiency could benefit greatly from a detailed kinetic
study on the most optimal [O2] on the basis of our proposed
mechanism, such a study is beyond the scope of this work.
In support of our claim for reaction initiation by ethyl
radicals, we found two particularly convincing artifacts that
signify autoxidation of triethylborane. First, fluoroethane is
unambiguously observed as a triplet of quartets (2J = 47.4 Hz, 3J
= 26.8 Hz) at −212.5 ppm in the crude 19F NMR spectra of all
Table 2. Substrate Scope
fluorination reactions.12 Second, monitoring the reaction by 11
B
NMR in CD3CN revealed a rapid formation of B(OEt)(Et)2 at
+56 ppm, a typical byproduct of the autoxidation reaction.13
Furthermore, considering the BEt3/O2 interplay is utilized in
iodine atom abstraction of alkyl iodides,14 we designed a system
to probe the presence of ethyl radicals under our reaction
conditions using 1-iodoadamantane (4) and NFSI in place of
Selectfluor, as NFSI will act only as a source of atomic fluorine
and will not propagate a chain reaction. We found that 1-
fluoroadamantane was produced in 85% yield based on
triethylborane, feasibly as a result of iodine atom transfer to
the ethyl radical and subsequent fluorination of the adamantyl
radical (Scheme 1). Finally, the reaction of an ethyl radical with
Selectfluor to produce fluoroethane and the radical dication was
calculated at the B3PW91/6-311++G**(MeCN) level to be a
substantial 56 kcal/mol downhill,15 demonstrating a highly
favorable reaction (Figure 3). Thus, this method very likely
operates as a radical chain reaction initiated via a well-
precedented autoxidation mechanism and propagated in an
manner analogous to that for the previously reported
copper(I)/Selectfluor system.
Upon evaluating the scope of the reaction, we found trends
in selectivity very similar to those of the copper(I)/Selectfluor
method. (1) Aliphatic substrates, viz. the cyclic alkanes in Table
2 (5−10), provided monofluorinated adducts in better yields
relative to benzylic substrates (11−15). This may suggest a
minor steric influence of the phenyl ring in the transition state,
characteristic of H atom abstraction by N-radical cations.16 (2)
In almost every instance, strictly monofluorination is observed,
which we have previously attributed to a manifestation of “the
polar effect.”17 A minor amount of difluorination only
materialized in the adamantane-based substrates 9 and 10 in
the methine positions (as previously noted in other Selectfluor-
based systems18). (3) On more complex substrates, i.e. the
androsterone 16 and progesterone 17 derivatives (vide infra), a
good degree of site selectivity is observed inherent to the parent
molecules. Interestingly, fluorination was favored primarily in
the C2 and C3 positions on the androsterone derivative19 and
occurred solely in the benzylic position on the progesterone
derivative. (4) We noted that the reaction conditions generally
endure oxygen-containing substrates well and tend to falter in
fluorinating most nitrogen-containing compounds, likely due to
N oxidation over desired reactivity.20 (5) Finally, note that the
product yields are mostly comparable to those reported for the
copper(I)/Selectfluor system as well as other sp3 C−H
fluorination methods in the literature to date. Beyond mild
reaction conditions and short reaction times, the virtue of using
BEt3 and Selectfluor lies in the minimal contamination from
byproducts upon workup/isolationstarting materials can be
easily recovered and fluorinated products can be obtained in
high purity via chromatography.
These parallels to the copper(I)/Selectfluor system and the
aforementioned experiments maintain the notion that the ethyl
radical liberated by BEt3/O2 acts as an initiator in what we
previously established to be a radical chain fluorination
mechanism propagated by a radical dication. Although the
reaction inherently has a two-edged sensitivity for [O2], under
optimal conditions it provides a mild, cheap, and easy
C
dx.doi.org/10.1021/jo501520e | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
3-Fluoro-1,3-diphenylpropan-1-one (13). 1H NMR (CDCl3):
8.00−7.32 (10 H, m), 6.29−6.08 (1 H, ddd, J = 46.9 Hz, 8.3 Hz,
4.5 Hz), 3.88−3.73 (1 H, ddd, J = 17.1 Hz, 14.8 Hz, 8.2 Hz), 3.42−
3.24 (1 H, ddd, J = 29.6 Hz, 17.0 Hz, 4.1 Hz). 19F NMR (CDCl3):
−172.97 (1 F, ddd, J = 46.4 Hz, 29.9 Hz, 15.5 Hz). Yield: 41 mg
(36%).22
alternative to sp3 C−H fluorination methods requiring
transition metals, ultraviolet light, or “catalysts” that are not
commercially available.
EXPERIMENTAL SECTION
■
Methyl 3-Fluoro-3-phenylpropanoate (14). 1H NMR (CDCl3):
7.41−7.34 (5 H, m), 6.03−5.82 (1 H, ddd, J = 46.7 Hz, 9.0 Hz, 4.1
Hz), 3.74 (3 H, s), 3.11−2.98 (1 H, ddd, J = 16.0 Hz, 13.6 Hz, 9.0
Hz), 2.89−2.71 (1 H, ddd, J = 32.6 Hz, 16.2 Hz, 4.3 Hz). 19F NMR
(CDCl3): −172.92 (1 F, ddd, J = 46.4 Hz, 32.0 Hz, 13.4 Hz). Yield: 28
mg (31%).22
General Considerations. Unless otherwise stated, all reactions
were carried out under strictly anhydrous, air-free conditions under
nitrogen. All solvents were dried and distilled by standard methods. 1H
NMR spectra were acquired on a 400 MHz NMR spectrometer in
CDCl3, 13C spectra were taken on a 300 MHz NMR spectrometer in
CDCl3, and 19F spectra were taken on a 300 MHz NMR spectrometer
in CDCl3 or CD3CN. The 1H, 13C, and 19F chemical shifts are given in
parts per million (δ) with respect to an internal tetramethylsilane
(TMS, δ 0.00 ppm) standard and/or 3-chlorobenzotrifluoride (δ
−64.2 ppm relative to CFCl3).21 NMR data are reported in the
following format: chemical shift (multiplicity (s = singlet, d = doublet,
t = triplet, q = quartet, m = multiplet), integration, coupling constants
(Hz)). IR data were obtained using an ATR-IR instrument. High-
resolution mass spectra (HRMS) were recorded using ESI-TOF
(electrospray ionization time of flight) mass spectrometry. All
measurements were recorded at 25 °C unless otherwise stated.
Characterization data for fluorocycloheptane (5),22 fluorocyclooctane
(6),23 fluorocyclodecane (7),9a fluorocyclododecane (8),7 fluoroada-
mantane (9),24 3-fluoroadamantan-1-ol (10),7 (1-fluoroethyl)benzene
(11),25 4-fluoro-4-phenylbutan-2-one (12),20 3-fluoro-1,3-diphenyl-
propan-1-one (13),22 methyl 3-fluoro-3-phenylpropanoate (14),22 and
2-(fluoro(phenyl)methyl)cyclohexanone (15)26 were consistent with
literature precedent. Compounds 5−7 and 11 are reported as crude
2-(Fluoro(phenyl)methyl)cyclohexanone (15). 1H NMR (CDCl3):
7.67−7.03 (10 H, m), 6.17−5.96 (dd, J = 46.5 Hz, 4.1 Hz), 5.95−5.74
(1 H, dd, J = 45.8 Hz, 7.5 Hz), 3.03−2.80 (1 H, m), 2.77−2.60 (1 H,
m), 2.58−2.22 (4 H, m), 2.18−1.49 (11 H, m), 1.35−1.15 (1 H, m).
19F NMR (CDCl3): −191.84 (1 F, dd, J = 46.4 Hz, 21.7 Hz), −172.64
(1 F, dd, J = 45.4 Hz, 14.4 Hz). Yield: 43 mg (41%).26
3β-Fluoro-5α-androstan-17-one and 2α-Fluoro-5α-androstan-
17-one (Major Products) (16). 19F NMR (CD3CN): −170.7 (1 F,
dm, J = 49.5 Hz); −174.9 (1 F, dm, J = 47.4 Hz). Yield: 47%.19
2-(Fluoro(phenyl)methyl)progesterone (17). 1H NMR (CDCl3): δ
7.41−7.36 (m, 2 H), 7.34−7.27 (m, 3 H), 6.55−6.42 (dd, 1 H, J = 46.2
Hz, 1.8 Hz), 5.84−5.82 (d, 1 H, J = 1.2 Hz), 2.72−2.58 (dddd, 1 H, J
= 30.3 Hz, 13.3 Hz, 5.3 Hz, 2.0 Hz), 2.56−2.50 (t, 1 H, J = 9.2 Hz),
2.42−2.32 (m, 2 H), 2.20−2.15 (m, 1 H), 2.11 (s, 3 H), 2.04−1.99
(dt, 1 H, J = 12.1 Hz, 2.9 Hz), 1.89−1.61 (m, 5 H), 1.53−1.07 (m, 10
H), 1.06−1.03 (s, 3 H), 1.03−1.00 (m, 1 H), 0.61 (s, 3 H). 13C NMR
(CDCl3): δ 209.3 (s), 196.4 (s), 170.9 (s), 139.1 (s), 138.9 (s), 129.9
(s), 128.4 (s), 127.7 (s), 124.7 (s), 124.6 (s), 123.9 (s), 90.2 (d, J =
175.6 Hz), 63.5 (s), 55.9 (s), 53.8 (s), 49.4 (s), 48.1 (s), 47.9 (s), 43.8
(s), 38.7 (s), 38.6 (s), 35.4 (s), 33.7 (s), 33.6 (s), 32.5 (s), 31.7 (s),
31.5 (s), 24.3 (s), 22.8 (s), 20.9 (s), 17.6 (s), 13.3 (s). 19F NMR
(CDCl3): δ −198.57 (dd, 1 F, J = 46.4 Hz, 30.9 Hz). IR (CDCl3)
1701, 1671 cm−1. HRMS-(ESI+): calcd for C28H35FO2Na+ 445.2513,
found 445.2527. Yield: 59 mg (28%).
1
19F and H NMR spectra due to their volatility. Spectral data were
processed with ACD/NMR Processor Academic Edition.27
Sample Procedure. Selectfluor (390.0 mg, 1.1 mmol, 2.2 equiv)
was placed in a 10 mL flame-dried round-bottom flask equipped with a
stir bar under N2. Acetonitrile (6.0 mL) was added to the reaction
flask, and the solution was stirred vigorously at room temperature. 2-
Benzylcyclohexanone (94.0 mg, 0.5 mmol, 1.0 equiv) was added,
followed by 1.0 M triethylborane solution in hexanes (10.0 mg, 0.1
mmol, 0.2 equiv). The reaction mixture was stirred for 4 h. The
product was diluted with Et2O and filtered through Celite. The
solvents were removed by rotary evaporation, and the residue was
subjected to preparative TLC on silica with an ethyl acetate/hexanes
mixture as eluent to afford 2-(fluoro(phenyl)methyl)cyclohexanone as
a clear oil (43 mg, 41%).
ASSOCIATED CONTENT
■
S
* Supporting Information
Figures and tables giving spectral data for all compounds
prepared in the paper and details of the computed structures.
This material is available free of charge via the Internet at
http://pubs.acs.org.
Compound Characterization. Fluorocycloheptane (5). 1H
NMR (CD3CN): 5.26−4.97 (1 H, dm, J = 47.7 Hz), 2.46−1.19 (12
H, m). 19F NMR (CD3CN): −164.55 (1 F, m). Yield: 47%.22
Fluorocyclooctane (6). 1H NMR (CD3CN): 5.25−4.92 (1 H, dm, J
= 46.3), 2.38−1.19 (14 H, m). 19F NMR (CD3CN): −164.51 (1 F,
m). Yield: 41%.23
AUTHOR INFORMATION
■
Corresponding Author
*E-mail for T.L.: lectka@jhu.edu.
Notes
Fluorocyclodecane (7). 1H NMR (CD3CN): 5.36−5.06 (1 H, dm, J
= 46.5 Hz), 2.42−1.16 (18 H, m). 19F NMR (CD3CN): −166.29 (1 F,
m). Yield: 40%.9a
The authors declare no competing financial interest.
1
ACKNOWLEDGMENTS
Fluorocyclododecane (8). H NMR (CDCl3): 4.72 (1 H, dm, J =
■
47.5 Hz), 1.87−1.51 (4 H, m), 1.48−1.26 (18 H, m). 19F NMR
(CDCl3): −176.88 (1 F, m). Yield: 47 mg (50%).7
Fluoroadamantane (9). 1H NMR (CDCl3): 2.27−2.20 (3 H, br s),
1.91−1.86 (6H, m), 1.66−1.60 (6H, m). 19F NMR (CDCl3): −128.5
(1 F, m). Yield: 42%.24
T.L. thanks the NSF (CHE 1152996) for support. C.R.P.
thanks Johns Hopkins for a Gary H. Posner Fellowship.
REFERENCES
■
1
3-Fluoroadamantan-1-ol (10). H NMR (CDCl3): 2.41−2.34 (2
(1) Pitts, C. R.; Bloom, S.; Woltornist, R.; Auvenshine, D.; Ryzhkov,
L. R.; Siegler, M. A.; Lectka, T. J. Am. Chem. Soc. 2014, 136, 9780−
9791.
H, m), 1.92−1.88 (2 H, d, J = 5.7 Hz), 1.84−1.80 (4 H, dd, J = 5.4 Hz,
3.3 Hz), 1.72−1.61 (4 H, m), 1.52−1.47 (3 H, m). 19F NMR
(CDCl3): −132.34 (1 F, m), −138.96 (1 F, m). Yield: 37%.7
(1-Fluoroethyl)benzene (11). 19F NMR (CD3CN): −167.06 (1 F,
dq, J = 47.4 Hz, 23.7 Hz). Yield: 30%.25
(2) For a fairly comprehensive introduction to radical initiators, see:
Denisov, E. T.; Denisova, T. G.; Pokidova, T. S. Handbook of Free
Radical Initiators; Wiley: Hoboken, NJ, 2003.
(3) Ollivier, C.; Renaud, P. Chem. Rev. 2001, 101, 3415−3434 and
references cited therein.
(4) Brotherton, R. J.; Weber, C. J.; Guibert, C. R.; Little, J. L. Boron
Compounds. In Ullman’s Encyclopedia in Industrial Chemistry; Wiley-
VCH: Weinheim, Germany, 2000.
4-Fluoro-4-phenylbutan-2-one (12). 1H NMR (CDCl3): 7.46−
7.32 (5 H, m), 6.08−5.86 (1 H, ddd, J = 46.9 Hz, 8.7 Hz, 3.8 Hz),
3.31−3.16 (1 H, m), 2.94−2.75 (1H, ddd, J = 32.2 Hz, 16.6 Hz, 4.0
Hz), 2.24 (3 H, s). 19F NMR (CDCl3): −173.59 (1 F, ddd, J = 47.4
Hz, 34.0 Hz, 15.5 Hz). Yield: 32 mg (38%).20
D
dx.doi.org/10.1021/jo501520e | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Note
(5) Regarding optimization of our reaction conditions, we found that
heating was unnecessary and promoted ionization over time, which
was deleterious to product yields. Colder temperatures (e.g., 10 °C)
are viable, but there is an increase in reaction times with no increase in
yield. Reactions were monitored at room temperature by TLC and/or
19F NMR; we found that 4 h is a reliable, generalized time period to
accomplish this reaction across all substrates (in some instances, the
reaction may be done in less time). Finally, given the solubility issues
with Selectfluor in most organic solvents and its unique reactivity in
acetonitrile, this is the solvent de choix for aliphatic fluorination.
(6) Rueda-Becerril, M.; Sazepin, C. C.; Leung, J. C. T.; Okbinoglu,
T.; Kennepohl, P.; Paquin, J.-F.; Sammis, G. M. J. Am. Chem. Soc.
2012, 134, 4026−4029.
(7) Amaoka, Y.; Nagatomo, M.; Inoue, M. Org. Lett. 2013, 15, 2160−
2163.
(8) Xia, J.-B.; Ma, Y.; Chen, C. Org. Chem. Front. 2014, 1, 468−472.
(9) However, some of these photochemical approaches may be
operating very differently mechanistically. For some examples, see:
(a) Bloom, S.; Knippel, J. L.; Lectka, T. Chem. Sci. 2014, 5, 1175−
1178. (b) Kee, C. W.; Chin, K. F.; Wong, M. W.; Tan, C.-H. Chem.
Commun. 2014, 50, 8211−8214. (c) Xia, J.-B.; Zhu, C.; Chen, C. J. Am.
Chem. Soc. 2013, 135, 17494−17500.
(10) (a) Bertrand, M. P.; Feray, L.; Nouguier, R.; Perfetti, P. J. Org.
Chem. 1999, 64, 9189−9193. (b) Ryu, I.; Araki, F.; Minakata, S.;
Komatsu, M. Tetrahedron Lett. 1998, 39, 6335−6336.
(11) Achord, J. M.; Hussey, C. L. Anal. Chem. 1980, 52, 601−602.
(12) Tanuma, T.; Ohnishi, K.; Okamoto, H.; Morikawa, S. J. Fluorine
Chem. 1996, 76, 45−48.
(13) Zhang, Z.-C.; Chung, T. C. M. Macromolecules 2006, 39, 5187−
5189.
(14) (a) Byers, J. In Radicals in Organic Synthesis; Renaud, P., Sibi, M.
P., Eds.; Wiley-VCH: Weinheim, Germany, 2001; Vol. 1, p 72.
(b) Curran, D. P.; Chen, M.-H.; Spetzler, E.; Seong, C. M.; Chang, C.-
T. J. Am. Chem. Soc. 1989, 111, 8872−8878.
(15) Frisch, M. J., et al. Gaussian 09, Revision A.1; Gaussian, Inc.,
Wallingford, CT, 2009.
(16) Bernardi, R.; Galli, R.; Minisci, F. J. Chem. Soc. B 1968, 324−
325.
(17) For a more extensive discussion on the ionic-like selectivity of
recent sp C−H fluorination methods, see ref 1 and works cited therein.
(18) For example, see: Xia, J.-B.; Ma, Y.; Chen, C. Org. Chem. Front.
2014, 1, 468−472.
(19) Similar site selectivity was observed on the same substrate by
Groves et al. in a different approach to sp C−H fluorination: Liu, W.;
Huang, X.; Cheng, M.-J.; Nielsen, R. J.; Goddard, W. A., III; Groves, J.
T. Science 2012, 337, 1322−1325.
(20) See discussion and references in: Bloom, S.; Pitts, C. R.;
Woltornist, R.; Griswold, A.; Holl, M. G.; Lectka, T. Org. Lett. 2013,
15, 1722−1724.
(21) Naumann, D.; Kischkewitz, J. J. Fluorine Chem. 1990, 47, 283−
299.
(22) Schneider, H.-J.; Gschwendtner, W.; Heiske, D.; Hoppen, V.;
Thomas, F. Tetrahedron 1977, 33, 1769−1773.
(23) Beaulieu, F.; Beauregard, L.; Courchesne, G.; Couturier, M.;
LaFlamme, F.; L’Heureux, A. Org. Lett. 2009, 11, 5050−5053.
(24) Aoyama, M.; Fukuhara, T.; Hara, S. J. Org. Chem. 2008, 73,
4186−4189.
(25) York, C.; Prakash, G. K. S.; Olah, G. A. Tetrahedron 1996, 52,
9−14.
(26) Bloom, S.; Sharber, S. A.; Holl, M. G.; Knippel, J. L.; Lectka, T.
J. Org. Chem. 2013, 78, 11082−11086.
(27) ACD/NMR Processor Academic ed., version 12.0; Advanced
Chemistry Development, Inc., Toronto, ON, Canada, 2012; www.
acdlabs.com.
E
dx.doi.org/10.1021/jo501520e | J. Org. Chem. XXXX, XXX, XXX−XXX