Selective C-F bond activation: Substitution of unactivated alkyl fluorides using YbI3

Article abstract of DOI:10.1002/anie.201306104

F makes the break: The carbon-fluorine single bond is quite strong, thus making aliphatic C-F bond scission unusually challenging. A new methodology utilizing YbI₃ leads to the conversion of a C-F bond into a C-I bond, and is compatible with various functional groups. The reaction is exceptionally selective towards alkyl fluorides and proceeds under mild conditions. Copyright

Full text of DOI:10.1002/anie.201306104

Angewandte
Chemie
DOI: 10.1002/anie.201306104
Synthetic Methods
À
Selective C F Bond Activation: Substitution of Unactivated Alkyl
Fluorides using YbI3**
Annika M. Trꢀff, Mario Janjetovic, Linda Ta, and Gçran Hilmersson*
À
À
Fluorine is an important functionality which nowadays is
widely used in modern pharmaceuticals, agrochemicals, and
pesticides because of its ability to favorably influence
chemical and physical properties of organic molecules.[1]
group (C F) with a highly reactive group (C I), but also
implement activation of that specific carbon atom towards
general synthetic diversification.
À
During recent studies of reductive cleavage of C F bonds
À
These valuable properties, conferred by C F functionalities,
with Sm(HMDS)2 (HMDS = hexamethyldisilazide), we dis-
covered that traces of alkyl iodides could be detected in the
reaction mixtures.[8] Encouraged by this finding we turned our
attention to trivalent lanthanides. 1-Fluorodecane (1a) was
exposed to SmI3 in THF at room temperature. Surprisingly,
GC analysis revealed a small amount of 1-iododecane within
24 hours (10% yield), thus indicating a novel F/I substitution.
A solvent screen highlighted CH2Cl2 as the most efficient
medium yielding 98% of 1-iododecane within the same time
frame. A correlation between the yield of substitution and the
Lewis acidity of the respective lanthanide triiodide was
examined (Scheme 1).
have led to the generation of numerous methods for the
introduction of fluorine, for example, by electrophilic or
nucleophilic fluorinating reagents, or by radical procedures.[2]
The increasing popularity and ability to introduce fluorine
into organic compounds now provides the opportunity for
À
selective C F bond activation, and the development of novel
methods to synthetically alter these compounds further. Aryl
[3]
À
C F bond activation has been comprehensively elucidated.
Additionally, several procedures for substitution of activated
À
C F bonds, for example, at the benzylic or allylic position or
a to a carbonyl group, have been developed.[4] However,
À
a synthetically valuable method for C F bond activation of
unactivated aliphatic fluorocarbons still remains a challenge.
À
A few attempts to manipulate such C F bonds have been
addressed utilizing strong Lewis acids, which in general are
highly oxophilic, thus making their use impractical in the
presence of other functional groups.[5] Consequently, all of
these Lewis acids only allow substitution of simple, unsub-
stituted alkyl fluorides, thus greatly limiting their synthetic
usefulness.
Trivalent lanthanides are known not only to be excellent
Lewis acids, but also to form strong bonds to fluorides.[6] As
such, it stands to reason that they could potentially be utilized
Scheme 1. Investigating the reactivity of different lanthanide triiodide
reagents. Reaction conditions: LnI3 (0.0484 mmol, 24.2 mm), CH2Cl2
(2.0 mL), 1a (0.0440 mmol, 22.0 mm). n-Dodecane (0.044 mmol,
22 mm) was used as an internal standard. Analyzed by GC-FID.
À
to activate C F bonds. The Lewis acidity of lanthanides
increases from left to right in the periodic table, with
simultaneous decrease in oxophilicity.[7] This conclusive
pattern makes ytterbium an interesting candidate for activa-
The use of LaI3 resulted in only 1% F/I substitution, while
SmI3, DyI3, and YbI3 provided 18%, 50%, and greater than
98% yields, respectively, within 9 hours. Gratifyingly, the
reaction was compatible with standard laboratory quality
(> 99.5%) solvent and could be conducted open to the
atmosphere. However, a small amount of deliberately added
water (1% V/V) was deleterious for the reaction. Thus, the
reaction as well as the reagent (YbI3) is insensitive towards
moisture and air, thereby indicating that these iodides are
stable towards hydrolysis. With a substoichiometric amount of
YbI3 relative to 1-fluorodecane, the reaction rate decreased
significantly without reaching full conversion, thus implying
that only one iodide is readily transferred. Full conversion
was achieved with 1.1 equivalents of YbI3. Employing MgI2
gave a considerably lower yield of 1-iododecane along with
several byproducts according to GC analysis. No formation of
1-iododecane was observed when employing simple iodides
such as KI, LiI, AlI3, or CuI. The reaction was successfully
performed in the dark, thus excluding light induction. The
reaction of YbI3 with 1-fluorodecane was also studied in
À
tion of the aliphatic C F bond. Herein we report our
exploration of YbI3 as a novel, very powerful, iodination
reagent. Selective substitution of alkyl fluorides in the
presence of other common functionalities is expected to
open up new synthetic strategies in organic chemistry. Hence,
it would not only enable the substitution of a highly inert
[*] Dr. A. M. Trꢀff, M. Janjetovic, L. Ta, Prof. Dr. G. Hilmersson
Department of Chemistry and Molecular Biology
University of Gothenburg
Kemivꢀgen 10, SE-412 96 Gothenburg (Sweden)
E-mail: Hilmers@chem.gu.se
[**] The authors gratefully acknowledge the Swedish Research Council
and the University of Gotheburg for financial support. Prof. Per-Ola
Norrby is gratefully acknowledged by his contribution of mecha-
nistic insight.
Supporting information for this article (experimental details) is
available on the WWW under http://dx.doi.org/10.1002/anie.
201306104.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Table 1: Substrate scope of the F/I substitution.
a variety of solvents (see Table S1 in the Supporting
Information). The trend observed was that in nonpolar
solvents such as i-hexane, in which YbI3 was not soluble,
less than 10% yield was obtained within 9 hours. In polar,
coordinating solvents such as THF, EtOH, acetone, Et3N, and
DMF, less than 10% conversion was observed, whereas in
Et2O and MeCN approximately 30% yield of 1-iododecane
was achieved. Almost full conversion and yield was achieved
in polar, noncoordinating solvents such as CHCl3 or CH2Cl2
(> 90% yield). The results above indicate that the reaction is
facilitated by an activation of the fluoride–carbon bond by
ytterbium. A competing interaction of the solvent with
À
ytterbium can prevent the C F bond activation to occur.
Upon addition of an excess of YbI3 (1.5 equiv) to a 1:1:1
mixture of 1-fluoro-, 1-chloro-, and 1-bromodecane, 100%
conversion of 1-fluorodecane into 1-iododecane was observed
within 8 hours, whereas less than 0.5% of 1-chlorodecane and
1-bromodecane were consumed (Scheme 2). No further
substitution of 3 and 4 was observed after leaving the reaction
to run for a longer period of time (24 h), thus demonstrating
exclusive selectivity of the reagent for alkyl fluorides.
Scheme 2. Selectivity study of YbI3 towards different aliphatic carbon–
halogen bonds. Reaction conditions: YbI3 (0.066 mmol, 33 mm),
CH2Cl2 (2.0 mL), 1a (0.044 mmol, 22 mm), 3 (0.044 mmol, 22 mm), 4
(0.044 mmol, 22 mm). n-Dodecane (0.044 mmol, 22 mm) was used as
an internal standard. Analyzed by GC-FID.
The proficiency of any synthetic method depends on
several aspects, one of the most important being high
tolerance towards diverse functional groups. Performing the
substitution of 1-fluorodecane in the presence of various
organic compounds, each containing a common functional
group, gave us some comprehension of the scope of the
method (see Table S2 of the Supporting Information).
Functionalities such as ketone, alcohol, cyanide, trialkyl-
amine, or ether did not interfere with the F/I substitution
reaction. Benzamide and adamantyl carboxylic acid reduced
the rate of the reaction but not the yield, possibly by
coordinating strongly to YbI3. It should be underlined that
the above substrates were recovered in quantitative yield
after the reaction. Thiophenol was not compatible with the
Yields of the isolated iodinated compounds are reported. Note: for 1o
only the alkyl fluoride reacted and not the aryl fluoride. Reaction
conditions: YbI3 (0.484 mmol, 24.2 mm), CH2Cl2 (20 mL), substrate
(0.440 mmol, 22.0 mm).
into the corresponding iodides in excellent yields. The straight
forward isolation of the iodo-substituted compounds involved
a simple filtration and subsequent concentration under
vacuum, thus yielding the product in very high (> 96%)
1
purity as determined by H NMR spectroscopy. The tertiary
iodide resulting from 1c is rather unstable and readily
undergoes elimination to a mixture of alkenes, thus explaining
the moderate yield (82%) upon isolation. No substitution was
observed for substrates having two or three fluorides on the
same carbon atom (1e, 1g), or on sp2-hybridized carbon
atoms (1 f, 1h). These outcomes are consistent with the higher
bond strengths for CF2 and CF3 groups as well as for sp2-
carbon atoms, as compared to monofluorinated sp3-carbon
atoms.[1b] The substitution proceeded in high yield but at
a significantly lower rate with a phenyl group present in the
substrate (1i, 1j). This lower rate is attributed to the likely
interaction of the aromatic ring with YbI3.[7] Substrates
containing an alcohol (1k), ester (1l), and benzyl-protected
reaction conditions because of oxidative formation of
[9]
À
diphenyl disulfide (PhS SPh). Trivalent lanthanides are
also known to mediate transesterification reactions.[10] Never-
theless, when YbI3 was added to a mixture of ethyl acetate
and 4-phenyl-1-propanol in the presence of 1-fluorodecane,
only substitution of the fluoride with iodide was observed.
The rate of the reaction was somewhat lower, thus indicating
a competing coordination of YbI3 with the ester or the
alcohol.
The scope of the reaction was further elaborated by its
application on primary, secondary, and tertiary alkyl fluorides
(1a–d; Table 1), in which all the substrates were converted
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
amine (1m) were all compatible with the reaction conditions,
thus providing iodo-substituted products in excellent yields in
all cases. Notably the tetrahydropuran derivatives (1n, 1o,
and 1p) underwent clean and selective substitution of their
aliphatic fluoride with iodide. The substitution occurs only on
3
2
À
À
the sp C F, without affecting the sp C F/Br (1o, 1p), once
again demonstrating the extraordinary selectivity of the
reaction. Considering that ketones are prone to undergoing
aldol condensation in the presence of trivalent lanthanides,
we were very satisfied to achieve full conversion of the bulky
steroid 1q, which contains both an alkene and a ketone
functionality, within 2 hours.[11] The corresponding iodo com-
pound 2q was isolated in 93% yield. The polyfluorinated
alkyl 1r did not react, possibly because of chelation of Yb3+
Scheme 4. Mechanistic proposal for the F/I substitution.
SN2 mechanism is believed to proceed via a transition state
involving two equivalents of YbI3 (A). The SNi mechanism is
proposed to go through an intimate ion pair (B). Alterna-
with two, or several fluorines. The presence of adjacent
3+
À
fluorines weakens the activation of the C F bond with Yb ,
À
while simultaneously increasing the bond strength to the
carbon center.
tively, simultaneous C F bond activation and F/I substitution
could occur (C). Both intermediates B and C result in
retention of configuration.
Based on simple competition experiments with primary,
secondary, and tertiary alkyl fluorides, valuable mechanistic
insights were gained. Hence, comparing the reaction rates of
1a, 1b, 1c, and 1d revealed a reactivity order of tertiary >
secondary > primary. The same trend was observed when
YbI3 was added to a mixture of 1a, 1b, and 1d. This order of
reactivity may be indicative of an SN1 mechanism. It should
however be emphasized that for 1-fluoroadamantane (1d),
formation of a carbocation is highly unfavorable at the
bridgehead atom.[12] Instead an internal nucleophilic substi-
tution (SNi) mechanism could explain the behavior of 1d.
Further insights into the mechanism were gained through
stereochemical analysis of syn-1n and syn-1o (Scheme 3).
Analysis of the iodo-substituted ether products showed that
anti-iodo ether had been formed in a 10:1 ratio (anti-2n) and
a 5:1 ratio (anti-2o). Likewise, starting from the correspond-
ing anti-alkyl fluorides (anti-1n and anti-1o) resulted in the
respective syn-iodo ether products (14:1 syn-2n and 50:1 syn-
2o). Thus, the reaction displays a high degree of stereospe-
cificity, with inversion at the stereogenic center, and points to
an SN2 reaction pathway.[5c,13,14]
In conclusion an efficient protocol for substitution of alkyl
fluorides with iodide by means of YbI3 under mild reaction
conditions has been developed. To the best of our knowledge,
this is the first report of synthetically useful unactivated
À
aliphatic C F bond substitution. The procedure is exception-
ally selective towards alkyl fluorides, and it is compatible with
a large range of common functional groups. The substitution
is proposed to proceed by an SN2 mechanism with a competing
SNi pathway, with the ratio of products from either pathway
being dependent on the substrate. The YbI3-mediated selec-
tive substitution of the exceptionally strong carbon–fluorine
bond is expected to have considerable impact in modern
organic chemistry as it may initiate novel synthetic routes.
This strategy paves the way for the use of fluorine as a small,
sterically unhindered protecting group which now is easy to
remove. In addition, it is a direct and powerful route for late-
stage incorporation of iodine into complex target molecules.
À
The F/I substitution with YbI3 leads to a highly reactive C I
group, which easily can be converted into practically any
other functional group. Supplementary mechanistic studies
are envisioned to provide a deeper understanding of the
reaction and its selectivity.
Initial rate studies were performed to obtain basic kinetic
information. An approximate rate order of 1.0 with respect to
[1-fluorodecane], and 1.5 with respect to [YbI3] for the
substitution was observed.
Based on the stereochemical outcome and the kinetic
experiments, as well as the competition reaction and rate
order, a mechanistic proposal is outlined in Scheme 4. The
Received: July 13, 2013
Published online: && &&, &&&&
Keywords: chemoselectivity · fluorine · iodine ·
.
synthetic methods · ytterbium
[1] a) S. Purser, P. R. Moore, S. Swallow, V. Gouverneur, Chem. Soc.
Rev. 2008, 37, 320; b) D. OꢀHagan, Chem. Soc. Rev. 2008, 37, 308.
[2] a) N. Al-Maharik, D. OꢀHagan, Aldrichimica Acta 2011, 44, 65;
b) M. Rueda-Becerril, C. C. Sazepin, J. C. T. Leung, T. Okbino-
glu, P. Kennepohl, J.-F. Paquin, G. M. Sammis, J. Am. Chem. Soc.
2012, 134, 4026; c) F. Yin, Z. Wang, Z. Li, C. Li, J. Am. Chem.
Soc. 2012, 134, 10401; d) M. P. Sibi, Y. Landais, Angew. Chem.
2013, 125, 3654; Angew. Chem. Int. Ed. 2013, 52, 3570.
Scheme 3. Stereochemical analysis of tetrahydropyran derivatives 1n
and 1o. Reaction conditions: YbI3 (0.0484 mmol, 2.42 mm), CH2Cl2
(2 mL), substrate (0.044 mmol, 2.2 mm).
[3] H. Amii, K. Uneyama, Chem. Rev. 2009, 109, 2119.
[4] a) L. Zhang, W. Zhang, J. Liu, J. Hu, J. Org. Chem. 2009, 74,
2850; b) G. Blessley, P. Holden, M. Walker, J. M. Brown, V.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
Gouverneur, Org. Lett. 2012, 14, 2754; c) E. Benedetto, M.
Keita, M. Tredwell, C. Hollingworth, J. M. Brown, V. Gouver-
neur, Organometallics 2012, 31, 1408; d) A. Hazari, V. Gouver-
neur, J. M. Brown, Angew. Chem. 2009, 121, 1322; Angew. Chem.
Int. Ed. 2009, 48, 1296; e) K. Mikami, Y. Tomita, Y. Itoh, Angew.
Chem. 2010, 122, 3907; Angew. Chem. Int. Ed. 2010, 49, 3819;
f) T. Iida, R. Hashimoto, K. Aikawa, S. Ito, K. Mikami, Angew.
Chem. 2012, 124, 9673; Angew. Chem. Int. Ed. 2012, 51, 9535.
[5] a) K. Hirano, H. Yorimitsu, K. Oshima, Org. Lett. 2004, 6, 4873;
b) T. Hatakeyama, S. Ito, M. Nakamura, E. Nakamura, J. Am.
Chem. Soc. 2005, 127, 14192; c) J. Terao, S. A. Begum, Y.
Shinohara, M. Tomita, Y. Naitoh, N. Kambe, Chem. Commun.
2007, 855; d) C. Douvris, O. V. Ozerov, Science 2008, 321, 1188;
e) W. Gu, M. R. Haneline, C. Douvris, O. V. Ozerov, J. Am.
Chem. Soc. 2009, 131, 11203; f) J. Choi, D. Y. Wang, S. Kundu, Y.
Choliy, T. J. Emge, K. Krogh-Jespersen, A. S. Goldman, Science
2011, 332, 1545; g) K. Matsubara, T. Ishibashi, Y. Koga, Org.
Lett. 2009, 11, 1765.
[7] R. Anwander, Lanthanides: Chemistry And Use In Organic
Synthesis (Ed.: S. Kobayashi), Springer, Berlin, 1999, pp. 1 – 61.
[8] M. Janjetovic, A. M. Trꢁff, T. Ankner, J. Wettergren, G.
Hilmersson, Chem. Commun. 2013, 49, 1826.
[9] D. Witt, Synthesis 2008, 2491.
[10] S.-I. Fukuzawa, Y. Hongo, Tetrahedron Lett. 1998, 39, 3521.
[11]
J. Collin, N. Giuseppone, P. Van de Weghe, Coord. Chem. Rev.
1998, 178 – 180, 117.
[12] M. B. Smith, J. March, Marchꢁs Advanced Organic Chemistry:
Reactions, Mechanism, and Structure, 6th ed., Wiley, Hoboken,
2007, p. 245.
[13] IUPAC definition of stereospecific. A. D. McNaught, A. Wil-
kinson, Compendium of Chemical Terminology, 2nd ed., Black-
well Scientific Publications, Oxford, 1997; XML on-line cor-
rected version: http://goldbook.iupac.org (2006-) created by M.
Nic, J. Jirat, B. Kosata; Updates compiled by A. Jenkins.
[14] a) A. Nova, R. Mas-Ballestꢂ, G. Ujaque, P. Gonzꢃlez-Duarte,
Chem. Commun. 2008, 3130; b) A. Nova, R. Mas-Ballestꢂ, A.
Lledꢄs, Organometallics 2012, 31, 1245.
[6] a) G. B. Deacon, C. M. Forsyth, P. C. Junk, J. Wang, Chem. Eur.
J. 2009, 15, 3082; b) M. Klahn, U. Rosenthal, Organometallics
2012, 31, 1235.
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Synthetic Methods
A. M. Trꢁff, M. Janjetovic, L. Ta,
G. Hilmersson*
&&&& — &&&&
À
À
Selective C F Bond Activation:
Substitution of Unactivated Alkyl
Fluorides using YbI3
F makes the break: The carbon–fluorine
single bond is quite strong, thus making
bond into a C I bond, and is compatible
with various functional groups. The reac-
tion is exceptionally selective towards
alkyl fluorides and proceeds under mild
conditions.
À
aliphatic C F bond scission unusually
challenging. A new methodology utilizing
À
YbI3 leads to the conversion of a C F
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!