Mild and selective activation and substitution of strong aliphatic C-F bonds

Article abstract of DOI:10.1002/chem.201406097

A procedure for chemoselectively manipulating the strong aliphatic C-F bond with direct transformation into a C-N bond under mild conditions is reported. The activation and subsequent substitution of primary alkyl fluorides is mediated by La[N(SiMe₃)₂]₃, and results in high to excellent yields of tertiary amines. The methodology displays high selectivity towards the C(sp³)-F bond, and a variety of secondary amines are applicable as nucleophiles. Mechanistic investigations reveal a reaction that is first order with respect to [La[N(SiMe₃)₂]₃], [R¹R²NH], and [alkyl fluoride], and a 6-membered cyclic transition state is proposed. In addition, ¹H NMR spectroscopy shows that La[N(SiMe₃)₂]₃ is the active species involved in the substitution and that protonolysis of the amine, yielding La[NR¹R²]₃, lowers the reactivity.

Full text of DOI:10.1002/chem.201406097

DOI: 10.1002/chem.201406097
Full Paper
&
CꢀF Activation
Mild and Selective Activation and Substitution of Strong Aliphatic
CꢀF bonds
Mario Janjetovic, Annika M. Trꢀff, and Gçran Hilmersson*[a]
Abstract: A procedure for chemoselectively manipulating
the strong aliphatic CꢀF bond with direct transformation
into a CꢀN bond under mild conditions is reported. The acti-
vation and subsequent substitution of primary alkyl fluorides
is mediated by La[N(SiMe3)2]3, and results in high to excellent
yields of tertiary amines. The methodology displays high
selectivity towards the C(sp3)ꢀF bond, and a variety of sec-
ondary amines are applicable as nucleophiles. Mechanistic
investigations reveal a reaction that is first order with
respect to [La[N(SiMe3)2]3], [R1R2NH], and [alkyl fluoride], and
a
6-membered cyclic transition state is proposed. In
1
addition, H NMR spectroscopy shows that La[N(SiMe3)2]3 is
the active species involved in the substitution and that
protonolysis of the amine, yielding La[NR1R2]3, lowers the
reactivity.
Introduction
The bond between fluorine and carbon, with a dissociation
energy of 441 kJmolꢀ1, is the strongest covalent single bond
that carbon can form to another heteroatom.[1] As such, effec-
tive CꢀF bond scission is an intellectually appealing task, as
well as one of practical relevance as a way to transform
extremely long-lived and potentially toxic molecules. A fairly
unexplored but potentially promising use of the CꢀF function-
ality is as a chemically inert protecting group, which can be
switched into a chemically labile functionality. However, with
both thermodynamic and kinetic obstacles to transformation,
cleaving the CꢀF bond remains a major challenge in chemistry.
Organofluorines are consequently poor substrates for nucleo-
philic substitution, and are essentially limited to aryl-,[2]
benzyl-,[3] allyl-,[4] and other activated/assisted CꢀF bonds.[3d,5]
In contrast, reports of selective replacement of unactivated
fluorine in C(sp3)ꢀF bonds with other heteroatoms are rare,
but have been achieved utilizing strong Brønsted acids,[6] and
more recently with Lewis acids[7] and transition metal species[8]
(Scheme 1). Among these, Kambe and co-workers reported the
cleavage of C(sp3)ꢀF bonds utilizing organoaluminum re-
agents, including one example of amine substitution.[7e] Where-
as these methods are often limited to simple alkyl fluorides,
highly fluorophilic lanthanide complexes are attractive candi-
dates for activation of more demanding aliphatic organofluor-
ides.[9] Recently, we reported that YbI3 mediates mild and very
fast selective substitution of fluorine with iodide on aliphatic
Scheme 1. Different approaches to CꢀF bond activation and cleavage. Lewis
acid abstraction of fluorine through: a) Direct abstraction forming a
carbocation; b) concerted TS with X representing a nucleophile. Transition
metal-mediated CꢀF activation through: c) Oxidative addition; d) radical
cleavage.
fluorocarbons.[10] The reaction proved highly selective towards
the C(sp3)ꢀF bond and was compatible with several functional-
ized alkyl fluorides. The trivalent lanthanide amide
Ln[N(SiMe3)2]3 is a highly versatile Ln[X]3 complex, which has
been widely used in organic synthesis as a Lewis acid.[11] Many
Lewis acid-promoted CꢀF substitution reactions proceed via
an SN1 mechanism and are therefore limited to tertiary alkyl
fluorides. With the formation of a carbocation intermediate
(Scheme 1a), primary alkyl fluorides usually give unsatisfactory
results with elimination, hydride shift, and rearrangement as
a direct result. Herein we report for the first time a mild and
selective method for the amino substitution of primary alkyl
fluorides mediated by La[N(SiMe3)2]3. We have investigated the
reaction by kinetics and NMR methods to get a deeper under-
standing of what governs CꢀF activation and substitution
mediated by La[N(SiMe3)2]3. In addition we have explored the
utility of C(sp3)ꢀF to C(sp3)ꢀN bond substitution, by including
a wide range of secondary amines and functionalized primary
alkyl fluorides. This methodology opens up new strategies in-
volving stable alkyl fluorides as key intermediates in organic
synthesis.
[a] M. Janjetovic, Dr. A. M. Trꢀff, Prof. Dr. G. Hilmersson
Department of Chemistry and Molecular Biology
University of Gothenburg
Kemivꢀgen 10, 41296, Gothenburg (Sweden)
E-mail: hilmers@chem.gu.se
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201406097.
Chem. Eur. J. 2015, 21, 1 – 7
1
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
tion was found to increase with concentration (Table 1, entry 1
vs. entry 9), and it was even possible to exclude the solvent
entirely (Table 1, entry 10). The reaction was found to be rela-
tively insensitive towards moisture and air, and stoichiometric
amount of H2O could be added, although the rate of the reac-
tion decreased (Table 1, entry 11). With substoichiometric
amounts of La[N(SiMe3)2]3, the conversion of 1a never
exceeded the amount of lanthanide added, showing that one
equivalent of La complex is required to activate one alkyl fluo-
ride towards substitution (Table 1, entry 12). With 1 equivalent
of 2a, the rate of the reaction decreased (Table 1, entry 13).
When utilizing Sm[N(SiMe3)2]3 in CH2Cl2 (Table 1, entry 14), the
substitution reaction took twice as long compared to
La[N(SiMe3)2]3 (entry 1). This decrease in reactivity could be
explained by the Ln3+ ionic radius, where the larger radius
(La>Sm) has a wider metal coordination sphere.[11c] The lan-
thanide salts, YbBr3 (Table 1, entry 15), Yb(OTf)3 (entry 16), and
La(OTf)3 (entry 17), did not facilitate the reaction. Suitable reac-
tion conditions for examining the scope of amine incorpora-
tion were found to be 1.1 equivalents of La[N(SiMe3)2]3 and
3 equivalents of amine in CH2Cl2 (Table 2).
Results and Discussion
Seo and Marks recently reported that La[N(SiMe3)2]3 undergoes
protonolysis with primary and secondary amines to yield the
corresponding lanthanide amido complex and free hexa-
methyldisilazane, HN(SiMe3)2.[11c] Based on this, and our recent
results, we hypothesized that such a complex ought to be an
ideal reagent for nucleophilic substitution of alkyl fluorides,
encompassing both an activator for CꢀF bond cleavage and
a good nucleophile. With these considerations in mind, we
subjected 1-fluorodecane (1a) to a mixture of dibutylamine
(2a) and La[N(SiMe3)2]3 and, within 30 minutes, large amounts
of N,N-dibutyldecan-1-amine (3a) were detected by GC-FID.
Herewith, we have discovered a new way to activate and sub-
stitute a primary alkyl fluoride with a secondary amine. No by-
products, resulting from elimination, rearrangement, or hydride
shift, were detected. Hexamethyldisilazane, which is formed
during the reaction, did not react with the alkyl fluoride. The
reaction was further investigated to find suitable conditions
(Table 1).
Table 1. Optimization of CꢀF substitution reaction between
1-fluorodecane (1a) and dibutylamine (2a).[a]
Table 2. CꢀF substitution of 1-fluorodecane (1a) with various secondary
amines.[a]
Entry Ln[X]3
Solvent
c [m]
t [h] Conv. of 1a [%][b]
1
2
3
4
5
6
7
8
La[N(SiMe3)2]3 CH2Cl2
La[N(SiMe3)2]3 toluene
La[N(SiMe3)2]3 Et2O
La[N(SiMe3)2]3 hexanes
La[N(SiMe3)2]3 THF
La[N(SiMe3)2]3 MeCN
La[N(SiMe3)2]3 EtOAc
La[N(SiMe3)2]3 EtOH
La[N(SiMe3)2]3 CH2Cl2
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.044[c]
–
1
1
1
1
1
1
1
1
1
1
3
>99
97
>99
98
48
8
–
–
49
>99
90
9
10
11
12
13
14
15
16
17
La[N(SiMe3)2]3
–
La[N(SiMe3)2]3 CH2Cl2 +H2O[d] 0.18
La[N(SiMe3)2]3 CH2Cl2
La[N(SiMe3)2]3 CH2Cl2
Sm[N(SiMe3)2]3 CH2Cl2
0.072[e] 1.5
27[f]
73[g]
>99
0.18
0.18
0.18
0.18
0.18
1
2
4
1
1
Yb[Br]3
Yb[OTf]3
La[OTf]3
CH2Cl2
CH2Cl2
CH2Cl2
–
–
–
[a] Reaction conditions: Ln[X]3 (0.044 mmol), 1a (0.040 mmol), 2a
(0.12 mmol), solvent (0.25 mL). [b] Analyzed by GC-FID. n-Dodecane
(0.040 mmol) was used as internal standard. [c] CH2Cl2 (1 mL). [d] H2O
(0.9 mL, 0.050 mmol) was added. [e] La[N(SiMe3)2]3 (0.012 mmol). [f] No
change in conversion was observed with prolonged reaction time (24 h).
[g] 1 equiv of 2a (0.040 mmol). 90% conversion within 5 h.
[a] Reaction conditions: La[N(SiMe3)2]3 (0.176 mmol), 1a (0.16 mmol),
amine (0.48 mmol), CH2Cl2 (1 mL). Yields of the isolated compounds are
reported. [b] La[N(SiMe3)2]3 (0.044 mmol), 1a (0.040 mmol), amine
(0.12 mmol), CH2Cl2 (0.25 mL). Conversion of 1a is reported, measured by
GC-FID, with n-dodecane (0.040 mmol) as internal standard.
Full conversion was reached when the reaction was carried
out in weakly or non-coordinating solvents (Table 1, entries
1–4). Owing to its many TMS groups, the complex was fully
soluble even in nonpolar solvents such as hexanes. THF, which
is commonly known to coordinate strongly to lanthanides,[12]
gave rise to a significant decrease in reactivity (Table 1,
entry 5). Furthermore, no or low reactivities were achieved in
coordinating solvents (Table 1, entry 6–8). The rate of the reac-
A variety of secondary amines readily underwent substitu-
tion, and the corresponding tertiary amines were isolated in
high to excellent yields. The model reaction between 1-fluoro-
decane (1a) and N,N-dibutylamine (2a) reached full conversion
within 1 h and N,N-dibutyldecan-1-amine (3a) was isolated in
94% yield. Several other secondary amines, ranging from
&
&
Chem. Eur. J. 2015, 21, 1 – 7
www.chemeurj.org
2
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
simple to more sterically demanding, were all well-suited for
the substitution reaction, (3b–h). The more sterically challeng-
ing amines required longer reaction times. The weaker nucleo-
philic primary butylamine, did not promote substitution (3d),
nor did any dialkylation of the amine occur. The F to N substi-
tution was favored when utilizing N-allyl-N-cyclopentylamine
(yielding 3e), and we observed no addition of amine to the
allyl group, which has been known to occur in the presence of
similar Lewis acids.[11f,13] The reactivity of the cyclic amines de-
creased according to the following order of ring size 6>7@5
ꢁ4 (3i–l). Due to rapid protonolysis and subsequent precipit-
ation of the thermodynamically favored La(pyrrolidide)3 and
La(azetidide)3, the smaller 5- and 4-membered cyclic amines
afforded no substitution at all (3k and 3l). Even the sterically
hindered tetramethylpiperidine displayed reactivity towards
1-fluorodecane (3m). Mixing La[N(SiMe3)2]3 with morpholine,
N-methylpiperazine, pyrrole, and phthalimide, respectively,
resulted in immediate precipitation, of what we believe to be
the corresponding homoleptic lanthanum amides. Accordingly,
Scheme 3. Functional group compatibility of CꢀF substitution in the
presence of organic molecules containing a common functional group.
Reaction conditions: La[N(SiMe3)2]3 (0.044 mmol), 1a (0.040 mmol), Ar-FG
(0.040 mmol), 2a (0.12 mmol), CH2Cl2 (0.25 mL). Analyzed by GC-FID with
n-dodecane (0.040 mmol) as internal standard.
1
only free HMDS was observed by H NMR, indicating instant-
aneous protonolysis. Only modest to negligible substitution
occurred (see the Supporting Information) with these amines.
The diamine, N,N’-dimethylethylenediamine, provided <10%
conversion, probably due to strong bidentate chelation with
the lanthanide.
The substitution promoted by La[N(SiMe3)2]3 displayed excel-
lent selectivity towards the alkyl fluoride in the presence of
other alkyl halides (Cl, Br, or I; Scheme 2). Within 1 h, full con-
Scheme 2. Selectivity towards CꢀF in the presence of other alkyl halides.
Reaction conditions: La[N(SiMe3)2]3 (0.060 mmol), 1a, 4, 5, 6 (0.040 mmol
each), 2a (0.12 mmol), CH2Cl2 (0.25 mL). Analyzed by GC-FID with
n-dodecane (0.040 mmol) as internal standard.
Scheme 4. CꢀF substitution of more demanding alkyl fluorides. Reaction
conditions: La[N(SiMe3)2]3 (0.176 mmol), alkyl fluoride (0.16 mmol), 2a
(0.48 mmol), CH2Cl2 (1 mL). Yields of the isolated compounds are reported.
version of 1-fluorodecane (1a) was reached, leaving 1-chloro-
decane (4) completely untouched, while less than 2% conver-
sion of 1-bromodecane (5) and 1-iododecane (6) occurred. In
addition, functionalities such as nitrile, alcohol, benzyl ether,
ketone, amide, primary amine, alkyne, benzyl alcohol, nitro,
and heterocyclic ether did not affect the substitution, and full
conversion of 1a into 3a was achieved within 2 h or less
(Scheme 3). Even an aldehyde was compatible with the reac-
tion conditions resulting in >95% conversion of 1a, and only
traces of amide were formed.[11c] Hence, the reaction displays
excellent chemoselectivity. However, the ester functionality
was not compatible and, as described in the literature,[14] full
aminolysis was achieved within 30 min. The scope of the
reaction was further expanded to include more demanding
aliphatic organofluorides (Scheme 4). A secondary fluorine in
the vicinal position to the primary fluorine unfortunately
stopped the reaction (3n). The presence of adjacent fluorines
weakens the activation of the CꢀF bond with La3+, in which
the stabilized gauche conformation of the alkyl fluoride favors
chelation, while making a nucleophilic attack hindered.[1] When
the fluorines were situated further apart the primary fluorine
was fully substituted within 1.5 h, leaving the secondary fluo-
rine unreacted, and the corresponding tertiary amine (3o) was
isolated in 91% yield. The reaction showed exclusive selectivity
towards substitution of the C(sp3)ꢀF bond over a C(sp2)ꢀF
bond, with amine 3p isolated in 92% yield. With an aldehyde
incorporated into the alkyl fluoride 1e, a small temperature in-
crease (358C) was needed to reach full conversion, which
slightly promoted amide formation.[11c] Even so, 3q could be
isolated in 78% yield. Substitution of fluorinated chloroambucil
(1 f) with dibutylamine gave displacement exclusively of the
CꢀF bond while the CꢀCl bonds were untouched, once again
Chem. Eur. J. 2015, 21, 1 – 7
www.chemeurj.org
3
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
than those for a cyclic 4-membered TS (typically 1.1–1.5) using
amines as nucleophiles.[16]
Table 3. Reactivity of different branched alkyl fluorides.[a]
ꢀd½RCH2ꢀFꢃ=dt ¼ k ½La½NðSiMe3Þ2ꢃ3ꢃ ½RCH2ꢀFꢃ ½HNR1R2ꢃ
ð1Þ
a-branched
b-branched
g-branched
According to Seo and Marks, the homoleptic La[N(SiMe3)2]3
undergoes instantaneous protonolysis with secondary amines
to afford the corresponding lanthanide amido complex and
free hexamethyldisilazane in C6D6.[11c] However, careful analysis
1g
no rxn/24 h
1h
no rxn/24 h
1i
1j[b]
93%/1.5 h
1
62%/48 h
of the H NMR spectra acquired directly after mixing 3 equiva-
lents of dibutylamine and 1 equivalent of La[N(SiMe3)2]3 in
CD2Cl2 showed a slightly broad quartet at d=2.57 ppm
(J=7.0 Hz), corresponding to the -CH2-NH-CH2- fragment
(Figure 1A), as well as a broad singlet from the -NH- signal at
d=0.68 ppm. The triplet for the -CH2-NH-CH2- protons is
changed to a quartet due to scalar coupling to the acidic -NH-
[a] Reaction conditions: La[N(SiMe3)2]3 (0.044 mmol), alkyl fluoride
(0.040 mmol), 2a (0.12 mmol), CH2Cl2 (0.25 mL). n-Dodecane (0.040 mmol)
was used as internal standard. Conversion of the alkyl fluoride is reported,
analyzed by GC-FID. [b] La[N(SiMe3)2]3 (0.176 mmol), 1j (0.160 mmol), 2a
(0.480 mmol), CH2Cl2 (1 mL). Yield of the isolated tertiary amine is
reported.
demonstrating the high affinity
towards fluorine. Full conversion
was reached within 2 h and 3r
was isolated in 88% yield.
With increasing steric hin-
drance of the aliphatic organo-
fluorines, the rate of the reaction
decreased (Table 3). No reaction
occurred with a-branched alkyl
fluorides (secondary 1g or terti-
ary 1h), excluding the possibility
of a carbocation pathway (SN1).
For b-branched alkyl fluoride 1i,
62% conversion was reached
within 48 h, whereas g-branched
alkyl fluoride 1j underwent full
conversion within 1.5 h and the
tertiary amine was isolated in
93% yield. We hypothesized that
the F to N substitution is en-
hanced by a strong LaꢀF interac-
tion. To determine the existence
Figure 1. 1H NMR spectroscopic monitoring of the F to N substitution reaction mediated by the lanthanum
amides La[N(SiMe3)2]3 and La[NBu2]3. A) La[N(SiMe3)2]3 (0.080 mmol), 2a (0.24 mmol), CD2Cl2 (0.5 mL), 248C; B) 1a
(0.080 mmol), La[N(SiMe3)2]3 (0.080 mmol), 2a (0.24 mmol), CD2Cl2 (0.5 mL), 248C; C) 1a (0.080 mmol), La(NBu2)3
(ca. 0.080 mmol), CD2Cl2 (0.5 mL), 248C; D) 1a (0.080 mmol), La(NBu2)3 (ca. 0.080 mmol), 2a (0.24 mmol), CD2Cl2
(0.5 mL), 248C.
of such an interaction a titration experiment was conducted,
whereupon 1-fluorodecane (1.25–96.25 mm) was added to a so-
lution of La[N(SiMe3)2]3 (6.25 mm) in CD2Cl2. The 19F NMR chem-
ical shift of 1-fluorodecane was observed to be independent of
the La/F ratio (0.2–15:1), strongly indicating that there is no or
only a very weak coordination to La[N(SiMe3)2]3. To gain further
knowledge about the reaction mechanism, a kinetic study was
performed. From the initial rate kinetics, the substitution reac-
tion was determined to be overall third order, that is, first
order in La[N(SiMe3)2]3, alkyl fluoride, and amine (HNR1R2), and
zero order in HN(SiMe3)2 [Equation (1)]. Furthermore, isotopic
labelling experiments, independently measured to avoid iso-
topic scrambling, revealed a small kinetic isotope effect (KIE) of
k(Bu2NH)/k(Bu2ND)=1.03ꢂ0.09. This indicates that the rate-de-
termining step is not a simple bond-breaking process. A
normal SN2 transition state (TS) would yield an inverse secon-
dary KIE (kH/kD ꢁ0.5–0.9).[15] The observed KIE is slightly lower
proton, and it is only observed for solutions containing
La[N(SiMe3)2]3. The broadening of these peaks indicates a dy-
namic process, in which the amine exchanges between a free
and a La[N(SiMe3)2]3-coordinated form. The shifts are almost in-
dependent of the amine/La ratio and nearly identical to that of
the free amine, which shows that the equilibrium is driven to-
1
wards free amine. In addition, the H NMR spectrum revealed
only a very weak signal from the methyl groups of released
HN(SiMe3)2, altogether showing that protonolysis is not instan-
taneous with Bu2NH in CD2Cl2. Accordingly, when dibutylamine
and La[N(SiMe3)2]3 were allowed to equilibrate for 24 h, a new
signal appeared at d=2.43 ppm (a triplet, J=7.0 Hz; Fig-
ure 1C) in addition to that for HN(SiMe3)2 at d=0.07 ppm.
Clearly, this slow transformation in CD2Cl2 corresponds to the
protonolysis reported by Seo and Marks, which yielded
La(NBu2)3.[11c] We observed no re-formation of La[N(SiMe3)2]3
&
&
Chem. Eur. J. 2015, 21, 1 – 7
www.chemeurj.org
4
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
upon addition of a large excess of HN(SiMe3)2 to this mixture,
suggesting that the protonolysis is irreversible.
Conclusion
Furthermore, we could directly follow the F to N substitution
In summary, we have reported herein a new method for CꢀF
bond activation with subsequent substitution utilizing
La[N(SiMe3)2]3, which transforms a highly inert CꢀF bond into
a CꢀN bond. The La[N(SiMe3)2]3-promoted transformation of
primary alkyl fluorides yields a clean substitution reaction,
without any elimination or rearrangement by-products. The
method tolerates a variety of different secondary amines as
well as primary alkyl fluorides. The substitution shows excellent
selectivity towards the CꢀF bond compared to other aliphatic
carbon–halogen bonds, and various functional groups are
compatible with the reaction conditions. Mechanistic investiga-
tions presented results supporting a transition state involving
1
reaction by H NMR spectroscopy when adding 3 equivalents
of dibutylamine (2a) to a sample containing 1-fluorodecane
(1a, 1 equivalent) and La[N(SiMe3)2]3 (1 equivalent; (Figure 1B).
The broad quartet at d=2.57 ppm was found to decrease over
time, simultaneously with the appearance of a new peak at
d=2.35, assigned to the product N,N-dibutyldecan-1-amine
(3a). However, when utilizing La(NBu2)3 formed in situ, the F to
N substitution was much slower (Figure 1C). Slow substitution
was also observed upon further addition of dibutylamine (2a,
1
3 equivalents) to La(NBu2)3 (Figure 1D). The H NMR spectra of
this mixture similarly displays a broad signal at d=2.58 ppm
from dibutylamine. The 1H NMR experiment conducted with
diisopropylamine gave matching results (see the Supporting
Information).
1
La[N(SiMe3)2]3, R1R2NH, and alkyl fluoride. H NMR spectroscopy
revealed that protonolysis of the amine prior to the F to N sub-
stitution is indeed unfavorable. We envision that this type of
CꢀF bond transformation paves the way for the usage of fluo-
rine as a surrogate protecting group, which can be selectively
converted to a tertiary amine under mild conditions.
Overall, these findings indicate that La[N(SiMe3)2]3 is
a unique and powerful reagent for CꢀF activation, and that
lanthanum amides made from simple dialkylamines
(La[NR1R2]3) are less reactive species. It further demonstrates
that the nucleophilic secondary amine forms
a
pre-
Experimental Section
complex with the La[N(SiMe3)2]3. If the pre-complexed
La[N(SiMe3)2]3···HNR1R2 were to react via a linear TS with the
primary alkyl fluoride, the initial rate would instead be second
order with respect to the amine (HNR1R2), and the measured
KIE would be inverse. Taken together, the results support
a cyclic TS involving [La[N(SiMe3)2]3], [R1R2NH], and [RCH2F],
with the amine being partially deprotonated in the TS
(Scheme 5). A concerted transition state is consistent with the
General procedure for substitution of an alkyl fluoride with
a secondary amine.
The reaction was run under dry and inert conditions. To
La[N(SiMe3)2]3 (372 mg, 0.60 mmol) was added CH2Cl2 (2.5 mL)
followed by 1-fluorodecane (273 mg, 0.44 mmol) and dibutylamine
(200 mL, 1.2 mmol) The reaction was stirred at room temperature
for 1 h, after which it was quenched by adding it directly to a silica
gel column and purified by flash column chromatography, using
4:1 pentane/Et2O as eluent. The product, N,N-dibutyldecan-1-
amine, was isolated as colorless oil (102 mg, 0.38 mmol, 94%
yield). 1H NMR and 13C NMR spectra were in agreement with
previously reported data.[17]
Acknowledgements
The authors gratefully acknowledge the Swedish Research
Council for financial support.
Keywords: amines · chemoselectivity · fluorine · lanthanum ·
synthetic methods
Scheme 5. Proposed mechanism for substitution of an alkyl fluoride with
a secondary amine mediated by La[N(SiMe3)2]3. Pathway I=unproductive
pathway; pathway II=productive pathway. I@II=no substitution or very
slow substitution; I>II=slow substitution; I<II=moderate to fast
substitution; I !II=fast substitution.
[1] D. O’Hagan, Chem. Soc. Rev. 2008, 37, 308–319.
[2] H. Amii, K. Uneyama, Chem. Rev. 2009, 109, 2119–2183.
[3] a) P. A. Champagne, J. Pomarole, M.-E. Therien, Y. Benhassine, S. Beau-
lieu, C. Y. Legault, J.-F. Paquin, Org. Lett. 2013, 15, 2210–2213; b) P. A.
Champagne, A. Saint-Martin, M. Drouin, J.-F. Paquin, Beilstein J. Org.
Chem. 2013, 9, 2451–2456; c) G. Blessley, P. Holden, M. Walker, J. M.
Brown, V. Gouverneur, Org. Lett. 2012, 14, 2754–2757; d) L. J. Zhang, W.
Zhang, J. Liu, J. B. Hu, J. Org. Chem. 2009, 74, 2850–2853.
reaction being concentration dependent. On the basis of the
evidence presented, the rate of the substitution is governed
not only by steric factors, but also by the competing reaction
pathway of protonolysis (Scheme 5, pathway I). The faster the
protonolysis is, the slower the substitution (Scheme 5, path-
way II) proceeds.
[4] a) E. Benedetto, M. Keita, M. Tredwell, C. Hollingworth, J. M. Brown, V.
Gouverneur, Organometallics 2012, 31, 1408–1416; b) X. Pigeon, M. Ber-
geron, F. Barabe, P. Dube, H. N. Frost, J.-F. Paquin, Angew. Chem. Int. Ed.
2010, 49, 1123–1127; Angew. Chem. 2010, 122, 1141–1145; c) A. Hazari,
V. Gouverneur, J. M. Brown, Angew. Chem. Int. Ed. 2009, 48, 1296–1299;
Angew. Chem. 2009, 121, 1322–1325; d) T. Narumi, K. Tomita, E. Inoku-
chi, K. Kobayashi, S. Oishi, H. Ohno, N. Fujii, Tetrahedron 2008, 64,
Chem. Eur. J. 2015, 21, 1 – 7
www.chemeurj.org
5
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
4332–4346; e) T. Narumi, K. Tomita, E. Inokuchi, K. Kobayashi, S. Oishi,
H. Ohno, N. Fujii, Org. Lett. 2007, 9, 3465–3468.
[10] A. M. Trꢀff, M. Janjetovic, L. Ta, G. Hilmersson, Angew. Chem. Int. Ed.
2013, 52, 12073–12076; Angew. Chem. 2013, 125, 12295–12298.
[11] a) L. Hong, Y. Shao, L. Zhang, X. Zhou, Chem. Eur. J. Chem. - Eur. J. 2014,
20, 8551–8555; b) E. Le Roux, Y. Liang, M. P. Storz, R. Anwander, J. Am.
Chem. Soc. 2010, 132, 16368–16371; c) S. Seo, T. J. Marks, Org. Lett.
2008, 10, 317–319; d) Q. Shen, W. Huang, J. Wang, X. Zhou, Organome-
tallics 2008, 27, 301–303; e) X. Yu, S. Seo, T. J. Marks, J. Am. Chem. Soc.
2007, 129, 7244–7245; f) S. Hong, T. J. Marks, Acc. Chem. Res. 2004, 37,
673–686; g) Y. Horino, T. Livinghouse, Organometallics 2004, 23, 12–14;
h) S. Hong, S. Tian, M. V. Metz, T. J. Marks, J. Am. Chem. Soc. 2003, 125,
14768–14783; i) H. Berberich, P. W. Roesky, Angew. Chem. Int. Ed. 1998,
37, 1569–1571; Angew. Chem. 1998, 110, 1618–1620.
[12] R. Anwander in Lanthanides: Chemistry and Use in Organic Synthesis
(Ed.: S. Kobayashi), Springer, Heidelberg, 1999, pp. 1–61.
[13] a) X. Cheng, Y. Xia, H. Wei, B. Xu, C. Zhang, Y. Li, G. Qian, X. Zhang, K. Li,
W. Li, Eur. J. Org. Chem. 2008, 1929–1936; b) J.-S. Ryu, G. Y. Li, T. J.
Marks, J. Am. Chem. Soc. 2003, 125, 12584–12605; c) M. R. Gagne, C. L.
Stern, T. J. Marks, J. Am. Chem. Soc. 1992, 114, 275–294.
[14] H. Morimoto, R. Fujiwara, Y. Shimizu, K. Morisaki, T. Ohshima, Org. Lett.
2014, 16, 2018–2021.
[15] I. Lee, Chem. Soc. Rev. 1995, 24, 223–229.
[16] a) I.-H. Um, A.-R. Bae, T.-I. Um, J. Org. Chem. 2014, 79, 1206–1212;
b) K. S. Jeong, H. K. Oh, Bull. Korean Chem. Soc. 2009, 30, 253–256;
c) M. E. Ul Hoque, S. Dey, A. K. Guha, C. K. Kim, B.-S. Lee, H. W. Lee, J.
Org. Chem. 2007, 72, 5493–5499; d) H. K. Oh, J. S. Ha, D. D. Sung, I. Lee,
J. Org. Chem. 2004, 69, 8219–8223; e) H. J. Koh, S. J. Kang, C. J. Kim,
H. W. Lee, I. Lee, Bull. Korean Chem. Soc. 2003, 24, 925–930.
[5] a) R. Yunoki, A. Yajima, T. Taniguchi, H. Ishibashi, Tetrahedron Lett. 2013,
54, 4102–4105; b) G. Haufe, S. Suzuki, H. Yasui, C. Terada, T. Kitayama,
M. Shiro, N. Shibata, Angew. Chem. Int. Ed. 2012, 51, 12275–12279;
Angew. Chem. 2012, 124, 12441–12445; c) A. Nova, M. Reinhold, R. N.
Perutz, S. A. MacGregor, J. E. McGrady, Organometallics 2010, 29, 1824–
1831.
[6] a) D. Landini, D. Albanese, S. Mottadelli, M. Penso, J. Chem. Soc. Perkin
Trans. 1 1992, 2309–2311; b) M. Namavari, N. Satyamurthy, M. E. Phelps,
J. R. Barrio, Tetrahedron Lett. 1990, 31, 4973–4976.
[7] a) T. Stahl, H. F. T. Klare, M. Oestreich, ACS Catal. 2013, 3, 1578–1587;
b) W. Gu, M. R. Haneline, C. Douvris, O. V. Ozerov, J. Am. Chem. Soc.
2009, 131, 11203–11212; c) C. Douvris, O. V. Ozerov, Science 2008, 321,
1188–1190; d) S. A. Ara Begum, J. Terao, N. Kambe, Chem. Lett. 2007,
36, 196–197; e) J. Terao, S. A. Begum, Y. Shinohara, M. Tomita, Y. Naitoh,
N. Kambe, Chem. Commun. 2007, 855–857; f) V. J. Scott, R. Celenligil-
Cetin, O. V. Ozerov, J. Am. Chem. Soc. 2005, 127, 2852–2853; g) K.
Hirano, H. Yorimitsu, K. Oshima, Org. Lett. 2004, 6, 4873–4875; h) K.
Hirano, K. Fujita, H. Yorimitsu, H. Shinokubo, K. Oshima, Tetrahedron
Lett. 2004, 45, 2555–2557; i) G. Theodoridis, Tetrahedron Lett. 1998, 39,
9365–9368.
[8] a) Z. Mo, Q. Zhang, L. Deng, Organometallics 2012, 31, 6518–6521; b) J.
Choi, D. Y. Wang, S. Kundu, Y. Choliy, T. J. Emge, K. Krogh-Jespersen,
A. S. Goldman, Science 2011, 332, 1545–1548; c) J. Terao, H. Todo, S. A.
Begum, H. Kuniyasu, N. Kambe, Angew. Chem. Int. Ed. 2007, 46, 2086–
2089; Angew. Chem. 2007, 119, 2132–2135; d) J. Terao, N. Kambe, Bull.
Chem. Soc. Jpn. 2006, 79, 663–672; e) J. Terao, H. Watabe, N. Kambe, J.
Am. Chem. Soc. 2005, 127, 3656–3657; f) J. Terao, A. Ikumi, H. Kuniyasu,
N. Kambe, J. Am. Chem. Soc. 2003, 125, 5646–5647.
[17] H. Sajiki, T. Ikawa, K. Hirota, Org. Lett. 2004, 6, 4977–4980.
[9] a) M. Klahn, U. Rosenthal, Organometallics 2012, 31, 1235–1244; b) R.
Dꢂaz-Torres, S. Alvarez, Dalton Trans. 2011, 40, 10742–10750; c) G. B.
Deacon, C. M. Forsyth, P. C. Junk, J. Wang, Chem. Eur. J. 2009, 15, 3082–
3092; d) U. Mazurek, H. Schwarz, Chem. Commun. 2003, 1321–1326.
Received: November 14, 2014
Published online on && &&, 0000
&
&
Chem. Eur. J. 2015, 21, 1 – 7
www.chemeurj.org
6
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
&
CꢀF Activation
M. Janjetovic, A. M. Trꢀff, G. Hilmersson*
&& – &&
Mild and Selective Activation and
Substitution of Strong Aliphatic CꢀF
bonds
F makes the break: La[N(SiMe3)2]3
conditions. The scope and mechanistic
insights of the methodology are
reported. This concept opens up new
strategies involving stable alkyl fluorides
as key intermediates in organic
synthesis.
mediates efficient scission of primary
alkyl fluorides. The strong CꢀF bond of
different functionalized organofluorines
is selectively activated and subsequently
converted to a CꢀN bond, under mild
Chem. Eur. J. 2015, 21, 1 – 7
www.chemeurj.org
7
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ