Gallium(III)-promoted halocyclizations of 1,6-diynes

Article abstract of DOI:10.1021/acs.orglett.5b00716

Cyclization of 1,6-diynes promoted by stoichiometric Ga(III) halides produces vinyl halides in good to excellent yields. Under acidic conditions, initially formed iodocyclization products undergo in situ Friedel-Crafts cyclizations, giving access to iodoi

Full text of DOI:10.1021/acs.orglett.5b00716

Letter
pubs.acs.org/OrgLett
Gallium(III)-Promoted Halocyclizations of 1,6-Diynes
Kyle R. Strom, Anna C. Impastato, Kenneth J. Moy, Adrian J. Landreth, and John K. Snyder*
Department of Chemistry Boston University, Boston, Massachusetts 02215, United States
S
* Supporting Information
ABSTRACT: Cyclization of 1,6-diynes promoted by stoichio-
metric Ga(III) halides produces vinyl halides in good to
excellent yields. Under acidic conditions, initially formed
iodocyclization products undergo in situ Friedel−Crafts
cyclizations, giving access to iodoindenopyridines. Application
of the vinyl halides in cross-coupling reactions has been
explored, and mechanistic aspects of the cyclization are
discussed.
π-Lewis acid promoted cyclizations have become a highly
productive field of research in the development of new synthetic
methodologies.1 In an earlier investigation, we reported the use
of stoichiometric In(III) as a π-Lewis acid catalyst to promote
cyclization of 1,7-bisalkynyl ethers to tetrahydrofuranyl enones
(Scheme 1, top).2 We noted at the time that decreasing the tether
used as competent ynophiles activating alkynes in hydro-
aminations,7 hydroarylations,8 and ene-yne reactions.9 Of
10
11
particular relevance to this work, InI3 and GaCl3 have
recently been reported as effective catalysts for cationic
polycyclizations initiated by alkynes. Despite these substantial
efforts, yne-yne cyclizations have remained unexplored, and to
the best of our knowledge, this is the first report of a halogenative
cyclization involving GaX3 or InX3.12 Work by Yu reported a
related oxocarbenium ion-initiated bisalkyne halocyclization
triggered by Fe(III) halides.13
Scheme 1. Divergent Cyclization Catalyzed by In(III)
Initial efforts to increase the yield of the In(III)-catalyzed
halogenative cyclization (Scheme 1) were met with limited
success. Attempts at optimization, including changes in
solvent,14 temperature, and concentration proved fruitless in
increasing the yield beyond 30%. The persistently low yields
were found to be primarily attributable to the instability of vinyl
bromide 4a, with decomposition occurring during the relatively
long reaction times required for conversion (>12 h) and
subsequent purification. Given the ability of Ga(III) to promote
ring-opening polymerization reactions of cyclic ethers,15 N-
tosylate (NTs)-tethered diynes were examined as potentially
more stable substrates.
The NTs-linked substrates proved fruitful, and bisalkyne 2b
was converted to vinyl bromide 4b with an 80% isolated yield
(Table 1, entry 1). Screening related π-acids revealed In(III)
chloride and iodide to be suitable catalysts, and the
corresponding vinyl chloride 5b and iodide 6b were isolated in
good to excellent yields (Table 1, entries 2 and 3). The reaction
times increased in the series InI3 < InBr3 < InCl3, with the
chloride requiring refluxing in dichloroethane (DCE) to
complete the reaction in 12 h. Both In(III) triflate with added
bromide and In(I) bromide were ineffective at promoting any
reaction. In an effort to accelerate the reaction, Ga(III) was then
examined as a promoter.
length between the two alkynes by a single methylene to 1,6-
bisalkynyl ether 2a resulted in a modest conversion to
dihydropyranyl vinyl bromide 4a in low isolated yield (Scheme
1, bottom). Given the broad utility of vinyl halides in natural
product synthesis3 and drug development,4 we began developing
this chemistry as a route to new scaffolds for the production of
small-molecule libraries. Herein we report a method to construct
several vinyl halide scaffolds in moderate to excellent yields from
simple starting materials, using readily available Ga(III) halides
as reagents.
In comparison to the considerable effort directed toward
exploiting the π-acidity of the noble metals in the past several
decades,5 the catalytic applications of In(III) and Ga(III) have
only recently begun to garner interest. Perhaps this is due, in part,
to the fact that they do not have an obvious catalytic niche, being
both competent σ- and π-acids. As the former, they have both
been shown to be potent oxophiles activating ketones and
aldehydes, carboxylic acids, amides, epoxides, and ethers to a
range of nucleophilic additions.6 As the latter, they have been
Gallium(III) was found to promote the reaction at a faster rate,
and with GaBr3, vinyl bromide 4b could be isolated nearly
Received: March 11, 2015
Published: April 17, 2015
© 2015 American Chemical Society
2126
DOI: 10.1021/acs.orglett.5b00716
Org. Lett. 2015, 17, 2126−2129

Organic Letters
Letter
Table 1. Halocyclization Catalyst Optimization
Table 2. Iodocyclization Substrate Scope
a
entry
MX3
catalyst (equiv)
time (h)
yield (%)
80
1
2
3
4
5
6
7
8
9
InBr3
InCl3
InI3
1.0
1.0
1.0
1.0
1.0
1.0
1.0
0.2
36
12
3
b
72
93
0
InBr
24
1
GaCl3
GaBr3
GaI3
55
95
88
0.5
0.2
2
c
GaBr3
76
d
GaI3
0.2
5
56
a
b
c
Isolated yield. Run at 84 °C. Run at 50 °C, with 1.0 equiv of
d
TMSBr. Run with 1.0 equiv of TMSI.
quantitatively in 30 min. Conversion to the vinyl chloride 5b and
vinyl iodide 6b also proceeded smoothly, with GaI3 affording
complete conversion to the corresponding vinyl iodide in less
than 10 min (Table 1, entries 7−9). As with In(III) promoters,
the rate of the reactions increased in the series I > Br > Cl.
A variety of halide sources and conditions were screened in an
attempt to render the Ga(III) catalyst substoichiometric. These
included additives such as halide salts (TBAB, NaBr, KI), acid
halides (HBr, HI, pyridinium hydrobromide), and halosilanes
(TMSBr, TMSI), as well as Ga(OTf)3 and In(OTf)3 with added
bromide or other nucleophile traps. Only the halosilanes gave
any catalyst turnover, and vinyl bromide 4b or iodide 6b could be
isolated in 76 and 56% yield, respectively, with 20 mol % catalyst
and stoichiometric halosilane (Table 1, entries 8 and 9). Given
the decrease in yield, more difficult purification, comparable cost
of the gallium salts relative to the halosilanes, and a turnover
number of only 3, stoichiometric Ga(III) halides were used in
subsequent studies.
a
b
c
Isolated yield. Run at 60 °C. Run with 10 mol % of GaCl3.
Diphenyl diyne 2l (Table 2, entry 11) gave the expected
product 6l only in relatively low yield (41%), the major product
being tricyclic scaffold 7, isolated in 46% yield. The use of GaCl3
was effective in increasing the yield of 7 to 82%, requiring only 10
mol % catalyst (Scheme 2).
The range of diynes amenable to this chemistry was then
explored using GaI3 (Table 2). Consistent with previous
observations, etherate 2a (Table 2, entry 1) gave a lower yield
of vinyl idoide 6a in comparison to the N-toslyate 2b.
Accordingly, we focused our efforts on N-tosylates. Terminal
alkynes excluded (Table 2, entry 2), a wide range of arylalkynyl
substrates proved amenable to the reaction, and smooth
conversion to the desired products was observed for most
(Table 2, entries 3−10). Despite excellent conversions (80−
94%), some of the products were unstable to silica gel
chromatography, resulting in significant loss of material (Table
2, entries 2, 8, 9, and 10). Fortunately, basic extractive workup
was effective at purifying the vinyl halides to >91% purity, and
these products could be used in further cross-coupling chemistry
as the near pure material (vida infra). The relative rate of the
cyclizations showed a dependence on the electron-richness of the
arylalkyne. Electron-rich substrates (Table 2, entries 4 and 5)
gave near immediate conversion to product, while electron-
withdrawing groups slowed the reaction considerably (Table 2,
entries 7−10). In the case of benzoate 2i and nitrobenzene 2k,
mild heating was effective in decreasing the reaction time. β-
Trimethylsilyl-substituted alkyne 2m gave only the desilylated
allene 8 on treatment with catalyst (Table 2, entry 12).
Scheme 2. Diphenyldiyne Cycloisomerization
The pathway to products 6 may follow an initial 6-exo-dig
cyclization promoted by Ga(III) activation of diyne A, giving
vinyl carbenium intermediate B (Scheme 3). Trapping of this
vinyl carbenium ion by a halide from the Ga(III) promoter to
vinyl gallate C, followed by protodegallation from adventitious
water or during aqueous quench, delivers vinyl iodides 6.
Products were observed with the halide incorporated adjacent to
the arene as the sole regioisomers. Given that both alkynes are
potential π-bases, it is likely that this regioselectivity is driven by
the stability of the vinyl cation B, favoring halogenation at the
benzylic position. The increase in the reaction time observed
with more electron-deficient arenes is thus rationalized by the
2127
DOI: 10.1021/acs.orglett.5b00716
Org. Lett. 2015, 17, 2126−2129

Organic Letters
Letter
Scheme 3. Proposed Mechanism for the Formation of
Products 6 and 7
Table 3. Tandem Iodocyclization/Friedel−Crafts Scope
a
entry
diyne
A
R1
R2
product (% yield)
b
c
1
2
2a
2b
2d
2e
2f
O
Me
Me
Et
H
H
H
N/A (0 , 0 )
b
NTs
NTs
NTs
NTs
NTs
NTs
NTs
NTs
NTs
9b (91 )
b
c
3
9d (66 , 73 )
b
c
4
Me
Me
Me
Me
Me
Me
Me
p-OMe
3,5-Me
m-OMe
p-Cl
9e (0 , 58 )
b
5
9f (91 )
b
c
6
2g
2h
2i
9g (29 , 52 )
b
7
9h (97 )
b
c
8
p-CO2Me
p-CF3
9i (20 , 33 )
b
c
9
2j
9j (25 , 48 )
b
c
10
2k
p-NO2
N/A (0 , 0 )
a
b
c
Isolated yield. One-pot procedure with TFA. Two-step procedure
with TfOH.
GaI3 via basic workup in a two-step sequence (Table 3, entries 3,
4, 6, 8, and 9). Dipropargyl ether 2a gave only a trace conversion
to the desired indenopyran, decomposing under all conditions.
In addition, 2k failed to react even at high temperatures, stopping
at the vinyl bromide 6k likely due to the low nucleophilicity of the
nitroarene.
The vinyl halides 9 were more stable than their exocyclic vinyl
halide counterparts 6 and could be chromatographed in silica gel
and stored for months without noticeable decomposition.
With access to a range of vinyl iodides 6 and 9, the utility of
these structures in selected cross-couplings was then examined
(Figure 1). Under standard unoptimized conditions, Suzuki and
Sonagashira couplings produced the corresponding arylated and
alkynyl products, respectively, in fair to excellent yields.
Despite the relatively short shelf life of the vinyl halides 6, these
cyclization products proved to be useful in Suzuki cross-
couplings without the need to purify the vinyl halides, as
evidenced by compounds 10c and 10i. The two-step isolated
slower rate of formation of the higher energy vinyl carbenium
intermediate due to increased electron withdrawal on the aryl
ring.
Satisfyingly, this mechanism also anticipates the formation of 7
from 2l as the diphenyl system contains an arene nucleophile
capable of trapping the vinyl cation via an intramolecular
Friedel−Crafts reaction. The increase in the yield of 7 with 10
mol % of GaCl3 is likely a consequence of the slower rate of
chloride trapping of the vinyl carbenium, in comparison to
iodide, allowing greater internal arene addition. Products such as
7 have also been observed with Au(I) catalysis.16 We are
currently further exploring this biscyclization with diaryl
bisalkynes.
During optimization efforts for the formation of 6, several
additives were examined as proton sources to quench the vinyl
gallate intermediate C. While not significant to the formation 6,
the addition of acid (TFA, HBr, and Ga(OTf)3) gave small
amounts of indenopyridine 9b (Table 3, entry 2). This product
was presumed to be the result of a Bronsted acid catalyzed
intramolecular Friedel−Crafts alkylation of the carbenium ion
formed upon protonation of the endoalkene of the vinyl iodide 6.
Attempts to favor this process by the addition of TFA to the
halocyclization products formed in situ were successful. N-Tosyl
diyne 2b could be converted to indenopyridine 9b in a two-step,
one-pot procedure with a 91% overall yield by adding TFA to the
reaction mixture once thin-layer chromatography analysis had
determined that all of the starting diyne 2b had been converted to
6b (Table 3, entry 2). This procedure was effective at further
cyclizing several of the N-tosylate diynes 6 to 9 (Table 2, entries 5
and 7). However, some of the substrates decomposed under
these reaction conditions at rates comparable to the formation of
the desired product. Surprisingly, a two-step procedure where
the vinyl iodide 6 was isolated and then subjected to TFA
without GaI3 present afforded no reaction.17 Presumably, TFA
was acting to liberate HI from GaI3, and the stronger acid was
necessary for protonation. Triflic acid provided a convenient
solution, and the crude vinyl iodides 6 could be converted to 9
with superior yields by treatment with this acid after removal of
Figure 1. Cross-coupling products of vinyl iodides.
2128
DOI: 10.1021/acs.orglett.5b00716
Org. Lett. 2015, 17, 2126−2129

Organic Letters
Letter
(10) Surendra, K.; Corey, E. J. J. Am. Chem. Soc. 2014, 136, 10918−
10920.
(11) (a) Li, H.-J; Guillot, R.; Gandon, V. J. Org. Chem. 2010, 75, 8435−
8449. (b) Michelet, B.; Bour, C.; Gandon, V. Chem.Eur. J. 2014, 20,
14488−14492.
yield of the cross-coupled products prepared from the crude vinyl
iodide far exceeded the isolated yields of the first step.
In conclusion, we have developed a simple method for the
construction of iodotetrahydropyridines and indenopyridines
using GaI3 and have shown that these products are capable
substrates in cross-coupling reactions. These reactions offer
quick access to easily diversifiable scaffolds from simple starting
materials via a previously unreported mechanism wherein GaX3
acts to promote a diyne cyclization and is a halide source. We are
now expanding on the cross-coupling reactions to produce a
library for biological screening.
(12) (a) Murai, M.; Hatano, R.; Kitabata, S.; Ohe, K. Chem. Commun.
2011, 47, 2375−2377. (b) Moriya, T.; Yoneda, S.; Kawana, K.; Ikeda, R.;
Konakahara, T.; Sakai, N. Org. Lett. 2012, 14, 4842−4845.
(13) Xu, T.; Yang, Q.; Ye, W.; Jiang, Q.; Xu, Z.; Chen, J.; Yu, Z.
Chem.Eur. J. 2011, 17, 10547−10551.
(14) Coordinating solvents (ACN, THF, DMF) returned starting
material, whereas noncoordinating solvents (DCE, chlorobenzene,
CHCl3) behaved similarly. DCE was chosen out of convenience.
(15) Dagorne, S.; Normand, M.; Kirillov, E.; Carpentier, J. Coord.
Chem. Rev. 2013, 1869−1886.
ASSOCIATED CONTENT
■
S
* Supporting Information
(16) Lian, J.-J.; Chen, P.-C.; Lin, Y.-P.; Ting, H.-C.; Liu, R.-S. J. Am.
Chem. Soc. 2006, 128, 11372−11373.
(17) Subjecting isolated 6 to the reaction conditions with both GaI3
and TFA did bring about conversion to 9.
General experimental protocols and characterization data for all
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: jsnyder@bu.edu.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the NIGMS CMLD initiative (P50 GM067041) for
partial financial support, and the NSF for supporting the
purchases of the NMR (CHE 0619339) and HRMS
spectrometers (CHE 0443618). We also thank the NSF REU
(CHE 1156666) for summer support for A.J.L., and the Boston
University Undergraduate Research Opportunities Program
(UROP) for support for A.C.I.
REFERENCES
■
(1) (a) Corma, A.; García, H. Chem. Rev. 2003, 103, 4307−4366.
(b) Michelet, V.; Toullec, P. Y.; Genet, J. Angew. Chem., Int. Ed. 2008, 47,
4268−4315.
(2) Gibeau, A. L.; Snyder, J. K. Org. Lett. 2011, 13, 4280−4283.
(3) (a) Dounay, A. B.; Overman, L. E. Chem. Rev. 2003, 103, 2945−
2964. (b) Jones, S. B.; Simmons, B.; Mastraccio, A.; MacMillan, D. W. C.
Nature 2011, 475, 183−188. (c) Ueda, A.; Yamamoto, A.; Kato, D.;
Kishi, Y. J. Am. Chem. Soc. 2014, 136, 5171−5176.
(4) (a) McCubbin, J. A.; Maddess, M. L.; Lautens, M. Org. Lett. 2006, 8,
́
2993−2996. (b) Camacho-Davila, A. A. Synth. Commun. 2008, 38,
3823−3833. (c) Liu, K. K.-C.; Sakya, S. M.; O’Donnell, C. J.; Flick, A. C.;
Ding, H. X. Bioorg. Med. Chem. 2012, 20, 1155−1174.
(5) (a) Shen, H. C. Tetrahedron 2008, 64, 7847−7870. (b) Furstner,
̈
A.; Davies, P. W. Angew. Chem., Int. Ed. 2007, 46, 3410−3449.
(6) (a) Chauhan, K. K.; Frost, C. G. J. Chem. Soc., Perkin Trans. 1 2000,
3015−3019. (b) Gupta, M. K.; O’Sullivan, T. P. RSC Adv. 2013, 3,
25498−25522.
(7) (a) Sarmaa, R.; Prajapati, D. Chem. Commun. 2011, 47, 9525−
9527. (b) Li, L.; Huang, G.; Chen, Z.; Liu, W.; Wang, X.; Chen, Y.; Yang,
L.; Li, W.; Li, Y. Eur. J. Org. Chem. 2012, 5564−5572. (c) Sakai, N.;
Takahashi, N.; Ogiwara, Y. Eur. J. Org. Chem. 2014, 5078−5082.
(8) (a) Inoue, H.; Chatani, N.; Murai, S. J. Org. Chem. 2002, 67, 1414−
1417. (b) Bhaskar, G.; Saikumar, C.; Perumal, P. T. Tetrahedron Lett.
́
2010, 51, 3141−3145. (c) Alonso-Maranon, L.; Martínez, M. M.;
̃
Sarandeses, L. A.; Sestelo, J. P. Org. Biomol. Chem. 2015, 13, 379−387.
(9) (a) Chatani, N.; Inoue, H.; Kotsuma, T.; Murai, S. J. Am. Chem. Soc.
2002, 124, 10294. (b) Miyanohana, Y.; Chatani, N. Org. Lett. 2006, 8,
2155. (c) Hassen, K. B. H.; Gaubert, K.; Vaultier, M.; Pucheault, M.;
Antoniotti, S. Green Chem. Lett. Rev. 2014, 7, 243−249.
2129
DOI: 10.1021/acs.orglett.5b00716
Org. Lett. 2015, 17, 2126−2129