Aluminum-Catalyzed Cross-Coupling of Silylalkynes with Aliphatic C-F Bonds

Article abstract of DOI:10.1021/acs.orglett.7b00712

We report the generation of aliphatic and benzylic acetylenes via reaction of primary, secondary, and tertiary aliphatic fluorides with various trimethylsilyl acetylides. These reactions are catalyzed by Al and B Lewis acids, most effectively by the extre

Full text of DOI:10.1021/acs.orglett.7b00712

Letter
pubs.acs.org/OrgLett
Aluminum-Catalyzed Cross-Coupling of Silylalkynes with Aliphatic
C−F Bonds
Amit K. Jaiswal, Kelvin K. K. Goh, Simon Sung, and Rowan D. Young*
Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543
S
* Supporting Information
ABSTRACT: We report the generation of aliphatic and benzylic acetylenes via
reaction of primary, secondary, and tertiary aliphatic fluorides with various
trimethylsilyl acetylides. These reactions are catalyzed by Al and B Lewis acids,
most effectively by the extremely fluorophilic tris(pentafluorophenyl)alane,
representing the first example of catalytic incorporation of alkynes into aliphatic
C−F positions. The fluorophilicity of the catalysts gives rise to fluorine selectivity
over other halogens, allowing orthogonal reactivity pathways.
The challenge of activating C−F bonds, coupled with their ever-
increasing abundance in chemical industries, has provided
incentive to chemists in pursuing C−F bond functionalization.12
Recently, we reported the catalytic conversion of C−F to higher
C−X functional groups (X = Cl, Br, I) utilizing halosilanes and Al
catalysts, allowing access to “traditional” transition-metal cross-
coupling chemistry.3 However, the direct coupling of C−F bonds
with organometals for C−C bond formation remains an attractive
but difficult means to late-stage functionalization. Direct C−C
couplings of sp2 C−F bonds are known with Zn, Mg, and B
organometallic reagents catalyzed by transition metals;4 however,
sp3 aliphatic and benzylic substrates remain challenging.5
Given the unique environmental hazard posed by aliphatic
fluorocarbons, their catalytic functionalization through C−C
coupling reactions for reuse in the chemical industry remains an
important chemical goal. Early examples of Lewis acid catalyzed,
aliphatic C−F coupling reactions that proceeded via SE2′
Figure 1. (Top) Previous catalytic systems for C−C coupling of aliphatic
pathways were reported for tertiary fluorides with Si allyl or
fluorides. (Bottom) The installation of acetylenes into C−F bonds using
enolate coupling partners (Figure 1a) and recently extended to
silylalkynes and Al catalysts (this work). [M] = main group complex.
benzylic fluorides.6 Friedel−Crafts alkylations between primary
alkyl fluorides and arene solvent were first reported with boron
halide catalysts by Olah.7 Since then, this has been extended to
In contrast, organosilanes are stable under atmospheric
conditions, commercially available, and/or safe and easy to
synthesize. Organosilanes have been employed as leaving groups
secondary and tertiary alkyl fluorides8 and, more recently, to
benzylic fluoride coupling partners (Figure 1b).6c,9 Ozerov
employed an in situ generated Al catalyst to catalyze the
methyldefluorination of benzylic fluorides with AlMe3 (Figure
1c),10 although this reaction has also been reported to proceed in
the absence of catalyst.11
in allylic and enolic nucleophilic reactions with tertiary fluorides
and as co-reagents in the reaction of benzylic and allylic fluorides
with a range of nucleophilic organyls.6 Given our recent success
utilizing silicon halides as nucleophilic reagents in the
substitution of organofluorides,3 we envisioned the reaction
may be extended to silyl acetylides.
Indeed, although a limited number of metal-catalyzed C−F
bond Grignard couplings are known,12 direct stoichiometric
functionalization of aliphatic C−F bonds has been accomplished
most successfully using highly active organometallic Al reagents,
with reports of alkyl, allyl, aryl, and alkynyl transfer.6a,11 Despite
this success, organoalane reagents suffer greatly from their
extreme reactivity, being pyrophoric and, in some cases,
explosive. Additionally, the incorporation of organogroups into
aluminum’s coordination sphere requires the use of other highly
reactive, dangerous, and/or toxic Sn, Mg, Zn, or Li reagents.
Sonogashira couplings between terminal alkynes and sp2
fluoroarenes can be catalyzed by indium chloride,13 and the
base-induced attack of allylic fluorides by phenyl acetylide has
been reported.5 However, prior to this study, the direct catalytic
coupling of aliphatic C−F bonds with alkyne nucleophiles was
Received: March 20, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b00712
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
undocumented. Herein, we report the development of a method
that utilizes silyl acetylides and Al catalysts in the transfer of
acetylides to primary, secondary, and tertiary aliphatic C−F
positions.
When 1-TMS-2-phenylacetylene and adamantyl fluoride (1)
were stirred for 18 h overnight in the presence of 10 mol % AlMe3,
methyladamantane (18%) and 1-adamantyl-2-phenylacetylene
(1a, 78%) were produced as the dominant products (Table 1,
Table 1. Optimization of Reaction between 1 and 1-TMS-2-
a,b
phenylacetylene
Figure 2. Alkyne reaction scope with 1. Conditions: 1 (1 M), 1 mL
reaction volume, 1.1 equiv alkyne, ACF 3.5 mol %, rt, reaction stopped
after 20 h for analysis, conversion determined by 19F NMR and GC−MS.
(a) Reaction run with 5 mol % ACF and stopped after 8 h.
entry
catalyst
solvent
conv (%)
yield (%)
1
2
3
4
5
6
7
8
A (10 mol %)
A (3 mol %)
B (10 mol %)
C (5 mol %)
C (5 mol %)
D (10 mol %)
E (5 mol %)
PhCl
PhCl
DCE
PhCl
DCE
DCE
DCE
DCE
>99
91
78 (67)
84
and yields (1a−g, ≥88%), and alkynyldefluorination products
were highly favored over Friedel−Crafts alkylation products.
Notably, the yields remained high when the para position
contained electron-donating methoxy (1f, 92%) or electron-
withdrawing chloro (1c, 97%) substituents. Moderate con-
versions were observed for alkenyl (1h) and alkyl (1i,j)
acetylenes, although the corresponding moderate yields showed
the reaction to be relatively clean. The reaction was also selective
for sp3 C−F activation in the presence of sp3 C−Cl as
demonstrated by the absence of C−Cl cleavage in 1j.
79
61
100
100
99
98
98 (96)
88
100
8
>95
6
a
b
General conditions: 1 (1.0 M), 18 h at rt. Yields and conversions are
based on GC−MS, isolated yields in parentheses.
The use of terminal TMS acetylide in the alkynyldefluorination
reaction of 1 resulted in a mixture of three products,
adamantylacetylene (1k), 1-adamantyl-2-TMS-acetylene (1l),
and bis-adamantylacetylene as the dominant product. Increasing
the relative concentration of TMS-acetylene to 5 times that of 1
greatly reduced the yield of bis-adamantylacetylene (<10%) to
give a mixture of 1k and 1l (2.2:1, 70% combined yield) (Figure
3). Conveniently, desilylation of 1l forms 1k (1.0 M TBAF in
THF,15 12 h, 89% conversion), representing an effective route of
acetylide delivery to organofluorides.
entry 1). Concurrently, complete consumption of 1 and
formation of TMSF were observed by 19F NMR spectroscopy.
Alkylation of aliphatic fluorides by Al alkyls has been reported to
be facile, but alkynyldefluorination catalysis continued to proceed
after methylation occurred.6b Thus, the generation of methyl-
adamantane may be indicative of an induction process that
generates an active catalyst. In agreement with this, lower
loadings of AlMe3 led to less methyladamantane byproduct and
higher observed yields of 1a (Table 1, entries 1, 2). Similarly,
when 10 mol % AlCl3 was utilized as a catalyst, chloroadamantane
(16%) was observed as a reaction byproduct (Table 1, entry 3).
However, aluminum halides were less effective catalysts
compared to AlMe3 due to their lower solubility in organic
solvents.
The use of the highly Lewis acidic [Al(C6F5)3]14 (ACF) was
also found to catalyze the reaction effectively (Table 1, entries 4,
5). In this case, very little C6F5 transfer was observed (0.5%), with
5 mol % catalyst loading allowing near-quantitative formation of
1a. Both DCE and PhCl were found to be suitable solvents;
however, the higher boiling point of chlorobenzene and small
quantities (<5%) of Friedel−Crafts alkylation products rendered
DCE the solvent of choice for obtaining analytically pure
products. BF3·OEt2 and [B(C6F5)3](BCF) catalysts also proved
efficient for the alkynyldefluorination of 1 (Table 1, entries 6, 7);
however, ACF proved more effective with challenging fluorides
(discussed below) and thus was chosen as the catalyst of choice
for the remainder of this study. A control experiment in the
absence of Al catalyst resulted in almost no reaction, confirming
the active role of Lewis acid in alkynyldefluorination (Table 1,
entry 8).
Figure 3. (Top) TMS and H compete as leaving groups to produce 1k
and 1l. Only 1k is generated after treatment with TBAF, offering a
convenient route to adamantylacetylene. (Bottom) Without TMS
present, H performed poorly as a leaving group, suggesting that silicon
plays an important role in catalysis turnover.
Attempts to employ terminal acetylenes without TMS groups
failed to produce the reactivity observed in the presence of TMS-
acetylenes. For instance, the reaction of 1 with phenylacetylene
under optimized conditions led to only 40% conversion after 40
h, with a yield of 16% of 1a (Figure 3) (cf. 84% yield for the
synthesis of 1a using 1-TMS-2-phenylacetylene; see Table 1,
Using optimized conditions, the scope of TMS acetylides for
the alkynyldefluorination of adamantyl fluoride was explored
(Figure 2). Generally, arylacetylenes gave excellent conversions
B
DOI: 10.1021/acs.orglett.7b00712
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
entry 2). The ability of silicon to sequester fluoride and promote
defluorination reactions has been previously noted.6c
suggestive of a pathway invoking a carbocation intermediate,
which agrees with previous reports of similar reactions.3
Substrates containing an alkyne functional group generated
moderate yields of diyne product (10a, 11a), signifying that
disubstituted alkynes did not interfere with the reaction.
The activity of other halides was tested to examine if the
reaction was specific to organofluorides.16 When 1-bromoada-
mantane was reacted with 1-TMS-2-phenylacetylene under
optimized conditions, negligible quantities (<5%) of 1a were
observed by GC−MS (Figure 6). The preference of fluorine over
To pursue a more efficient and selective reaction, we explored
the effect of substituents on silicon’s ability to act as a leaving
group. The electronic effects of incorporating either a phenyl or
tert-butyl group into silicon’s coordination sphere were
investigated through the use of bis-silylacetylenes 2a and 2b
(Figure 4). However, the distribution of products suggested that
Figure 4. Competitive reactions between different silyl leaving groups
determined preference based on sterics. The use of reagent 2b offers a
selective desilylation protocol.
steric factors were decisive in product distribution, with TMS
dominating as the leaving group when in competition with either
SiMe2 Bu or SiMe2Ph groups.
Hoping to extend this method to other organofluorides, we
attempted alkynyldefluorination of primary and secondary
fluoride derivatives (Figure 5).
Figure 6. Under optimized reaction conditions, aliphatic bromides were
found to be largely unreactive, allowing selective functionalization of
aliphatic fluoro positions.
t
bromine for this reaction stands in contrast to that of transition-
metal-catalyzed cross-couplings, offering the opportunity to
access divergent reactivity. Indeed, this was demonstrated
through the use of 1-(bromomethyl)-4-(fluoromethyl)benzene
(12). Reaction of 1-n-hexyl-2-TMS-acetylene with 12 resulted in
70% conversion after 18 h (Figure 4). The major product of this
reaction was found to be 12a (40% yield), where alkynylde-
fluorination had occurred. Alkynyldebromination represented
less than 1% of the reaction products. Thus, alkynyldefluorination
offers a new method to access orthogonal reactivity.
To explore the mechanism for alkynyl transfer, in separate
experiments, ACF was mixed solely with stoichiometric
quantities of 1 or (TMS)-2-phenylacetylene. No notable reaction
was observed in the mixture of ACF/TMS-2-phenylacetylene;
however, the reaction between 1 and ACF produced
(pentafluorophenyl)adamantane (Ad-C6F5), as judged by 19F
NMR spectroscopy and later identified by GC−MS. This is in
contrast to producing a carbocation/aluminate ion pair (i.e.,
[Ad][AlF(C6F5)3]), which has been reported in a similar fluoride
abstraction of fluorotriphenylmethane to generate [CPh3][AlF-
(C6F5)3]14b,17 and may be expected to be stable given the high
stability of the adamantium cation.18
Figure 5. Alkynyldefluorination of primary and secondary aliphatic
fluorides. Conditions: 3−11 (1 M), 1 mL reaction volume, 2 equiv
alkyne, ACF 5 mol %, rt, reaction stopped after 20 h for analysis,
conversion determined by 19F NMR and GC−MS. (a) Yield based on
GC−MS. (b) Isomerization products accounted for remainder of
conversion products and could not be separated from 9a using flash
chromatography; 5 equiv of alkyne used.
Indeed, 19F NMR spectroscopy revealed the formation of small
quantities of Ad-C6F5 during catalysis, and Ad-C6F5 (ca. 1%
yield) was routinely identified in post-reaction GC−MS analyses
(see the SI). As described above, transfer products arising from
initial constituents on aluminum were even more prominent
when using AlMe3 or AlCl3 catalysts. The presence of Ad-X (X =
C6F5, Me, Cl) byproducts may indicate an initiation process to
generate an active aluminum fluoride catalyst. Although
undocumented, aluminum fluoride exchange with silyl acetylide
would produce silicon fluoride and aluminum acetylide, which is
known to be capable of undergoing alkynyldefluorination (Figure
7, cycle B).11b
The alkynyldefluorination reaction of benzyl fluorides
tolerated both electron-donating and -withdrawing substituents
in the benzyl fluoride para position, allowing incorporation of
alkyl (4a) and halide (3a, 5a,b) functional groups. α-, β- and γ-
fluoroalkylbenzenes were also found to be stable toward
alkynyldefluorination, albeit inreduced yields (6a−8a). Although
alkynyldefluorination of 1-fluoropentane (9a) proceeded to
complete conversion under our reaction conditions, only 56% of
the desired linear product was observed. Rearrangement
products accounted for the remainder of the conversion,
Given the very high conversion of 1 to products in
alkynyldefluorination with ACF and the lower production of
Ad-C6F5 (cf. Ad-Me and Ad-Cl), X transfer could be a competing
reaction, and the active catalytic pathway could proceed via Lewis
C
DOI: 10.1021/acs.orglett.7b00712
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
REFERENCES
■
́ ́
(1) (a) Wang, J.; Sanchez-Rosello, M.; Acena, J. L.; del Pozo, C.;
̃
Sorochinsky, A. E.; Fustero, S.; Soloshonok, V. A.; Liu, H. Chem. Rev.
2014, 114, 2432. (b) Theodoridis, G. Advances in Fluorine Science. In
Fluorine and the Environment Agrochemicals, Archaeology, Green Chemistry
& Water; Tressaud, A., Ed.; Elsevier, 2006; Vol. 2, pp 121−175.
(c) Sandford, G. Philos. Trans. R. Soc., A 2000, 358, 455. (d) Muller, K.;
Faeh, C.; Diederich, F. Science 2007, 317, 1881. (e) Gillis, E. P.; Eastman,
K. J.; Hill, M. D.; Donnelly, D. J.; Meanwell, N. A. J. Med. Chem. 2015, 58,
8315.
(2) (a) Amii, H.; Uneyama, K. Chem. Rev. 2009, 109, 2119. (b) Stahl,
T.; Klare, H. F. T.; Oestreich, M. ACS Catal. 2013, 3, 1578. (c) Ahrens,
T.; Kohlmann, J.; Ahrens, M.; Braun, T. Chem. Rev. 2015, 115, 931.
(d) Shen, Q.; Huang, Y.-G.; Liu, C.; Xiao, J.-C.; Chen, Q.-Y.; Guo, Y. J.
Fluorine Chem. 2015, 179, 14.
Figure 7. Proposed catalytic cycles for alkynyldefluorination reaction.
See text for mechanism discussion.
(3) Goh, K. K. K.; Sinha, A.; Fraser, C.; Young, R. D. RSC Adv. 2016, 6,
42708.
acid halide abstraction, as proposed by Paquin and Stephan in
related Lewis acid catalyzed C−F arylation reactions.6c,9 Indeed,
fluoride abstraction from Ph3CF to generate [CPh3][AlF-
(C6F5)3] is documented.14b,17 It is plausible that a carbocation
generated in this manner may be susceptible to nucleophilic
attack by silyl acetylide. The catalytic loop in such a cycle would
be closed by silylium recombination with fluoro aluminate to
liberate the catalyst and generate silicon fluoride byproduct,
whose formation provides thermodynamic drive for catalysis
(Figure 7, cycle A).19 It unlikely that in situ generated HF is
responsible for catalysis given that alkynyldefluorination only
proceeded in high yield in the presence of TMS leaving groups,
which are known to sequester HF.8b
Thus, we have developed a transition-metal-free method for
coupling silylalkynes with aliphatic organofluorides. This method
is applicable to primary, secondary, and tertiary fluorides and
works for a range of TMS alkynes. In contrast to many previous
examples of transition-metal-free C−F functionalization, the use
of Si as a leaving group allows regioselectivity control. The
preference of the Al catalysts employed for fluoro over other
halogen positions enables access to orthogonal synthetic
strategies in cooperation with traditional transition-metal
couplings, which typically prefer chloro, bromo, and iodo to
fluoro positions.
(4) Examples: (a) Yoshikai, N.; Mashima, H.; Nakamura, E. J. Am.
Chem. Soc. 2005, 127, 17978. (b) Cargill, M. R.; Sandford, G.; Tadeusiak,
A. J.; Yufit, D. S.; Howard, J. A. K.; Kilickiran, P.; Nelles, G. J. Org. Chem.
2010, 75, 5860. (c) Nakamura, Y.; Yoshikai, N.; Ilies, L.; Nakamura, E.
Org. Lett. 2012, 14, 3316. (d) Tobisu, M.; Xu, T.; Shimasaki, T.; Chatani,
N. J. Am. Chem. Soc. 2011, 133, 19505. (e) Saijo, H.; Sakaguchi, H.;
Ohashi, M.; Ogoshi, S. Organometallics 2014, 33, 3669.
(5) Shibata reported base-catalyzed C−C couplings for fluoro allylic
esters. Such couplings are specific to specialized allylic fluoro substrates
and not applicable to more general aliphatic substrates. See:
(a) Nishimine, T.; Taira, H.; Tokunaga, E.; Shiro, M.; Shibata, N.
Angew. Chem., Int. Ed. 2016, 55, 359. (b) Okusu, S.; Okazaki, H.;
Tokunaga, E.; Soloshonok, V. A.; Shibata, N. Angew. Chem., Int. Ed. 2016,
55, 6744.
(6) (a) Ooi, T.; Uraguchi, D.; Kagashima, N.; Maruoka, K. Tetrahedron
Lett. 1997, 38, 5679. (b) Hirano, K.; Fujita, K.; Yorimitsu, H.;
Shinokubo, H.; Oshima, K. Tetrahedron Lett. 2004, 45, 2555. (c) Zhu,
J.; Per
(7) Olah, G. A.; Kuhn, S.; Olah, J. A. J. Chem. Soc. 1957, 0, 2174.
(8) (a) Luhmann, N.; Panisch, R.; Muller, T. Appl. Organomet. Chem.
́
ez, M.; Stephan, D. W. Angew. Chem., Int. Ed. 2016, 55, 8448.
̈
̈
2010, 24, 533. (b) Dryzhakov, M.; Moran, J. ACS Catal. 2016, 6, 3670.
(9) (a) Zhu, J.; Perez, M.; Stephan, D. W. Angew. Chem., Int. Ed. 2016,
́
55, 8448. (b) Champagne, P. A.; Benhassine, Y.; Desroches, J.; Paquin, J.-
F. Angew. Chem., Int. Ed. 2014, 53, 13835.
(10) Gu, W.; Haneline, M. R.; Douvris, C.; Ozerov, O. V. J. Am. Chem.
Soc. 2009, 131, 11203.
(11) (a) Terao, J.; Nakamura, M.; Kambe, N. Chem. Commun. 2009,
6011. (b) Terao, J.; Begum, S. A.; Shinohara, Y.; Tomita, M.; Naitoh, Y.;
Kambe, N. Chem. Commun. 2007, 855.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.7b00712.
■
S
(12) Metal-catalyzed Grignard reactions with alkyl fluorides:
(a) Iwasaki, T.; Min, X.; Fukuoka, A.; Kuniyasu, H.; Kambe, N. Angew.
Chem., Int. Ed. 2016, 55, 5550. (b) Mo, Z.; Zhang, Q.; Deng, L.
Organometallics 2012, 31, 6518. (c) Terao, J.; Ikumi, A.; Kuniyasu, H.;
Kambe, N. J. Am. Chem. Soc. 2003, 125, 5646. (d) Terao, J.; Todo, H.;
Watanabe, H.; Ikumi, A.; Kambe, N. Angew. Chem., Int. Ed. 2004, 43,
6180. (e) Iwasaki, T.; Takagawa, H.; Singh, S. P.; Kuniyasu, H.; Kambe,
N. J. Am. Chem. Soc. 2013, 135, 9604.
(13) Borah, H. N.; Prajapati, D.; Boruah, R. C. Synlett 2005, 2823.
(14) (a) Hair, G. S.; Cowley, A. H.; Jones, R. A.; McBurnett, B. G.;
Voigt, A. J. Am. Chem. Soc. 1999, 121, 4922. (b) Chen, J.; Chen, E. Y.-X.
Angew. Chem., Int. Ed. 2015, 54, 6842.
(15) Gallagher, W. P.; Maleczka, R. E. J. Org. Chem. 2003, 68, 6775.
(16) Reaction tolerance of aliphatic chloride is evident through the use
of DCE solvent and the stability of product 1j.
(17) (a) Chen, M.-C.; Roberts, J. A. S.; Marks, T. J. J. Am. Chem. Soc.
2004, 126, 4605. (b) Chen, M.-C.; Roberts, J. A. S.; Seyam, A. M.; Li, L.;
Zuccaccia, C.; Stahl, N. G.; Marks, T. J. Organometallics 2006, 25, 2833.
(18) Schleyer, P. R.; Fort, R. C.; Watts, W. E.; Comisarow, M. B.; Olah,
G. A. J. Am. Chem. Soc. 1964, 86, 4195.
Experimental details and spectra of obtained compounds
(PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: rowan.young@nus.edu.sg.
ORCID
Kelvin K. K. Goh: 0000-0001-6740-2176
Simon Sung: 0000-0002-7864-7694
Rowan D. Young: 0000-0001-7437-8944
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the National University of Singapore and the Singapore
Ministry of Education for financial support (WBS R-143-000-
586-112, R-143-000-666-114).
(19) Under the assumption that cycle A is responsible for catalysis, Al
fluoride byproducts arising from a competitive transfer reaction (i.e.,
AlX2F, AlXF2, and AlF3) may also act as Lewis acid catalysts.
D
DOI: 10.1021/acs.orglett.7b00712
Org. Lett. XXXX, XXX, XXX−XXX