Synthesis and reactivity of dibenzoselenacycloheptynes

Article abstract of DOI:10.1021/ol401225n

Dibenzoselenacycloheptynes were prepared in three steps from commercially available reagents and trapped in situ with benzyl azide to form the corresponding triazoles. Surprisingly, the dibenzoselenacycloheptynes also abstracted hydrogen atoms from solvents such as THF or toluene, forming dibenzoselenacycloheptene products. These alkenyl compounds arise from a hydrogen transfer reaction from solvent to the unisolable intermediate, and we postulate that the reaction proceeds via a radical mechanism originating from the strained alkynyl bond that has unusually high radical character.

Full text of DOI:10.1021/ol401225n

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Synthesis and Reactivity of
Dibenzoselenacycloheptynes
Gabriela de Almeida, Lisa C. Townsend, and Carolyn R. Bertozzi*
Departments of Chemistry and Molecular and Cell Biology and Howard Hughes
Medical Institute, University of California, Berkeley, California 94720, United States
crb@berkeley.edu
Received May 2, 2013
ABSTRACT
Dibenzoselenacycloheptynes were prepared in three steps from commercially available reagents and trapped in situ with benzyl azide to form the
corresponding triazoles. Surprisingly, the dibenzoselenacycloheptynes also abstracted hydrogen atoms from solvents such as THF or toluene, forming
dibenzoselenacycloheptene products. These alkenyl compounds arise from a hydrogen transfer reaction from solvent to the unisolable intermediate,
and we postulate that the reaction proceeds via a radical mechanism originating from the strained alkynyl bond that has unusually high radical character.
Angle-strained cycloalkynes have long been a topic of
interest to chemists due to their unique physical properties
and chemical reactivity.1 In recent years, these attributes
havebeen exploitedfor applications inbiological labeling.2
For example, the ring strain associated with cyclooctynes
(1, Figure 1) enables their facile 1,3-dipolar cycloaddition
with azides, a bioorthogonal reaction often termed “copper-
free click chemistry” that produces a biologically inert triazole
adduct.3,4 Over the past decade, modifications to the
cyclooctyne scaffold have been explored in an effort to
increase the rate of cycloaddition. This has been accom-
plished by modulating the electronics of the cyclooctyne,
for example through the addition of propargylic fluorine
atoms as in 2,5 or by increasing ring strain through the
additionof sp2 centers, asin 3,6 or by fusinga cyclopropane
ring to the cyclooctyne.7
Recently, we explored the utility of cycloheptynes con-
taining an endocyclic heteroatom as reagents for copper-
free click chemistry. We showed that tetramethylthiacy-
cloheptyne (TMTH, 4) is stable under biologically relevant
conditions and that the compound undergoes the desired
cycloaddition with azides faster than any other isolable
cycloalkyne.8 In principle, one might further increase the
ring strain of TMTH through the introduction of two
fused aryl rings. However, dibenzothiacycloheptyne 5 is
known to be unstable, as is its sulfoxide analog and the all-
carbon congoner.9,10
Here we explored the chemistry of derivatives of the
previously unknown dibenzoselenacycloheptyne 6. We
hypothesized that introduction of a selenium atom, which
(1) (a) Krebs, A.; Wilke, J. Top. Curr. Chem. 1983, 109, 189. (b) Hopf,
H. Grunenberg, J. Angle-Strained Cycloalkynes. In Strained Hydro-
carbons: Beyond the van’t Hoff and Le Bel Hypothesis; Dodziuk, H., Ed.;
Wiley-VCH: Weinheim, 2009; pp 375À379. (c) Detert, H., In Targets in
Heterocyclic Systems; Attanasi, O. A., Spinelli, D., Eds.; Royal Society of
Chemistry: Cambridge, 2011; Vol. 15, pp 1À49.
(6) (a) Ning, X. H.; Guo, J.; Wolfert, M. A.; Boons, G.-J. Angew.
Chem., Int. Ed. 2008, 47, 2253. (b) Debets, M. F.; van Berkel, S. S.;
Schoffelen, S.; Rutjes, F. P. J. T.; van Hest, J. C. M.; van Delft, F. L.
Chem. Commun. 2010, 46, 97. (c) Jewett, J. C.; Sletten, E. M.; Bertozzi,
C. R. J. Am. Chem. Soc. 2010, 132, 3688.
(7) Dommerholt, J.; Schmidt, S.; Temming, R.; Hendriks, L. J. A.;
Rutjes, F. P. J. T.; van Hest, J. C. M.; Leeber, D. J.; Friedl, P.; van Delft,
F. L. Angew. Chem., Int. Ed. 2010, 49, 9422.
(2) (a) In zebrafish, see: Laughlin, S. T.; Baskin, J. M.; Amacher,
S. L.; Bertozzi, C. R. Science 2008, 320, 664. (b) In C. elegans, see:
Laughlin, S. T.; Bertozzi, C. R. ACS Chem. Biol. 2009, 4, 1068. (c) In
mice, see: Chang, P. V.; Prescher, J. A.; Sletten, E. M.; Baskin, J. M.;
Miller, I. A.; Agard, N. J.; Bertozzi, C. R. Proc. Natl. Acad. Sci. U.S.A.
2010, 107, 3988. (d) Koo, H.; Lee, S.; Na, H. J.; Kim, S. H.; Hahn, S. K.;
Choi, K.; Kwon, I. C.; Jeong, S. Y.; Kim, K. Angew. Chem., Int. Ed.
2012, 51, 11836–11840.
(3) Agard, N. J.; Prescher, J. A.; Bertozzi, C. R. J. Am. Chem. Soc.
2004, 126, 15046.
(4) Jewett, J. C.; Bertozzi, C. R. Chem. Soc. Rev. 2010, 39, 1272.
(5) Codelli, J. A.; Baskin, J. M.; Agard, N. J.; Bertozzi, C. R. J. Am.
Chem. Soc. 2008, 130, 11486.
(8) de Almeida, G.; Sletten, E. M.; Nakamura, H.; Palaniappan,
K. K.; Bertozzi, C. R. Angew. Chem., Int. Ed. 2012, 51, 2443.
(9) Lorch, M.; Meier, H. Chem. Ber. 1981, 114, 2382.
(10) Tochtermann, W.; Oppenlander, K.; Hoang, M. N.-D. Liebigs
Ann. Chem. 1967, 701, 117.
r
10.1021/ol401225n
XXXX American Chemical Society

setbacks, we were certain that 6 was not too strained to
exist transiently in solution; after all, the sulfur-, oxygen-,
and carbon-cycloheptyne analogs have been indirectly
observed by trapping in situ.9,10 We suspected that the
high temperature and nucleophilic conditions necessary to
cleave the selenadiazole of 9 were simply too forcing even
for transient persistence of 6.
Scheme 2. Synthesis and Trapping of Cycloalkyne 14a
Figure 1. Strained cycloalkynes prepared previously (1À5) and
discussed here (6).
is larger than sulfur, might offset the increased ring strain
resulting from fused aryl rings, thereby leading to a stable
structure.11 Our first approach to the synthesis of 6 in-
volved double elimination of the 1,2 dibrominated derivative
of alkene 7 (8, Scheme 1), an approach that has previously
been used in the preparation of the sulfoxide analog of
compound 5 as well as the dibenzooxacycloheptyne.10 Un-
fortunately, our various attempts to add bromine across the
alkene of 7 led to a complex mixture of products.
Scheme 1. Attempted Synthetic Routes to Alkyne 6 via Alkene 7
a Se(dtc)2 = selenium diethyldithiocarbamate.16
Our next strategy relied on formation of the strained
alkyne by photolysis of a cyclopropenone precursor,14
a
method that has proven to be functional group tolerant
and mild enough to produce strained cyclooctynes in good
yield (Scheme 2).15 To make cyclopropenone 13, we
performed a lithiumÀhalogen exchange on commercially
available aryl bromide 10 and treated the aryl lithium inter-
mediate with selenium diethyldithiocarbamate (Se(dtc)2)16
to yield diaryl selenide 11. Compound 11 underwent a
FriedelÀCrafts reaction with tetrachlorocyclopropene 12
in the presence of AlCl3, and in situ hydrolysis of the
dichlorocyclopropene product resulted in 13. Gratifyingly,
when a solution of 13 in methanol was irradiated in the
presence of benzyl azide, the desired triazole 15 was
obtained in quantitative yield, indicating that the alkyne
photodecarbonylation product 14 had formed. But, we
could not isolate 14 when the same reaction was performed
under ambient conditions without an azide trap.
We next attempted the synthesis of 6 via selenadiazole 9
(Scheme 1, Scheme S1), a route analogous to the one used
to prepare compound 5 and its all-carbon congoner.9
Selenadiazoles can be thermally disproportionated to re-
veal alkynes without the use of harsh reagents,12 and they
can be prepared from ketone precursors that are readily
accessible from alkenes such as 7. However, thermolysis of
selenadiazole 9 proved challenging, and 6 was never ob-
tained. No identifiable derivatives of 6 were isolated in the
presence of traps such as benzyl azide or furan, and
attempts to unmask the alkyne via nucleophilic attack of
the selenadiazole also proved unsuccessful.13 Despite these
In a further attempt to isolate 14, we performed the
photolysis reaction at 0 °C in THF in the absence of a
trap.17 Unexpectedly, alkene 16 was the only product
formed (Scheme S3). When the reaction was repeated in
(14) Poloukhtine, A.; Popik, V. V. J. Phys. Chem. A 2006, 110, 1749.
(15) (a) Friscourt, F.; Ledin, P. A.; Mbua, N. C.; Flanagan-Steet,
H. R.; Wolfert, M. A.; Steet, R.; Boons, G.-J. J. Am. Chem. Soc. 2012,
134, 5381. (b) Fischer, F. R.; Nuckolls, C. Angew. Chem., Int. Ed. 2010,
49, 7257. (c) Poloukhtine, A. A.; Mbua, N. E.; Wolfert, M. A.; Boons,
G.-J.; Popik, V. V. J. Am. Chem. Soc. 2009, 131, 15769.
(11) DFT calculations indicate that the alkynyl bond angles are
approximately 148°, larger (and thus presumably less distorted) than
those shown for the stable compound TMTH1a (Figure S3).
(12) Marynoff, B. E.; Rebarchak, M. C. J. Org. Chem. 1991, 56, 5203.
(13) (a) Arsenyan, P.; Overte, K.; Rubina, K.; Beyakov, S.; Lukevics,
E. Chem. Heterocycl. Compd. 2004, 40, 503. (b) Buhl, H.; Gugel, H.;
Kolshorn, H.; Meier, H. Synthesis 1978, 536.
(16) Panda, A.; Menon, S. C.; Singh, H. B.; Butcher, R. J.
J. Organomet. Chem. 2001, 623, 87.
B
Org. Lett., Vol. XX, No. XX, XXXX

Scheme 3. Photodecarbonylation Products of 13
Scheme 4. Trapping and Reduction of Alkyne 20
(18) (Scheme 3, FigureS1).21,22 Furthermore, DHA should
not initiate radical reactions, so obtaining anthracene as a
reaction product would strongly indicate that the radical
originates from the dibenzoselenacycloheptyne. In accor-
dance with this hypothesis, photodecarbonylation of 13 in
the presence of DHA in toluene did produce anthracene
while a control reaction lacking 13 left DHA unaltered
(Figure S2). From these experiments, we conclude that the
alkynyl bond in 14 has significant diradical character that
can dictate some of its reactivity pathways.
Adding ortho methyl groups to diaryl cycloalkynes has
been shown to greatly decrease their reactivity via steric
blocking of the alkyne.23 We were curious whether such a
modification to 14 might have a differential effect on the
rates of its cycloaddition and hydrogen abstraction reac-
tions. Therefore we prepared 19, the dimethylated analog
of 14, using the same synthetic strategy used in the parent
compound’s synthesis (Scheme S5). Upon irradiation of
the dimethylated cyclopropenone with excess benzyl azide
in THF, triazole 20 was indeed obtained, indicating that 19
had formed in the course of the reaction (Scheme 4).
However, this was the minor product, obtained in only
7% yield, in a mixture with alkene 21 as the major product.
Thus, the bis-orthomethyl groups slowed the rate of the
cycloaddition with benyl azide moreso than the rate of
hydrogen atom transfer. It is important to note that the
mechanism of the cycloaddition with benzyl azide does not
necessarily proceed via the radical; most reactions of this
type are believed to be concerted pericyclic reactions.24
In summary, we developed a short synthetic route to
make dibenzoselenacycloheptynes, a new type of strained
alkyne with unexpected properties. These compounds are
not stable enough to be isolated due to their ability to
abstract hydrogen from solvents such as toluene and THF.
THF with an excess of benzyl azide, a 2:1 mixture of 15 and
16 was obtained (Scheme 3), indicating that the two
products are formed on a similar time scale. Given the
exceedingly short lifetimes of the intermediates in the
photodecarbonylation reaction,14 we presume that the
alkene results from transfer of two hydrogen atoms from
the solvent to the alkyne of 14 or to a photoexcited product
thereof. The decreased yield of compound 16 when the
reaction was performed in the presence of azide indicates
thatit originates from the same intermediate alkyne 14that
reacts with benzyl azide to form 15. To verify that the vinyl
hydrogen atoms were in fact originating from the THF
solvent, we performed the photodecarbonylation reaction
in THF-d8 and obtained only the dideutero alkene analog
of 16 (Scheme S4).18
Highly reactive strained cycloalkynes such as TMTH
and cyclooctyne have been reported to undergo hydrogen
transfer reactions with alcohols, amines, and thiols.1a With
alcohols, the reaction is thought to proceed via a pericyclic
mechanism.19,20 In this case, however, we thought it more
likely that alkyne 14 reacts as a diradical that can abstract
hydrogen atoms from the solvent. In accordance with this
hypothesis, we performed the photodecarbonylation reac-
tion in toluene, a nonprotic and non-nucleophilic solvent,
and again obtained alkene 16 as the major product.
To further explore the possibility that alkyne 14 behaves
as a diradical, we added 9,10-dihydroanthracene (DHA,
17) to the reaction. The weak methylenic CÀH bonds of
DHA make this compound a good donor of two succesive
hydrogen atoms, after which it is converted to anthracene
(22) Kurz, J. L.; Hutton, R.; Westheimer, F. H. J. Am. Chem. Soc.
1961, 83, 584.
(17) We could not perform the reaction at 0 °C in methanol due to the
limited solubility of compound 13.
(18) The analogous reaction in toluene-d8 also yielded the dideutero
alkene. However, in this case the product mixture was much more
complex.
(23) Gordon, C. G.; Mackey, J. L.; Jewett, J. C.; Sletten, E. M.;
Houk, K. N.; Bertozzi, C. R. J. Am. Chem. Soc. 2012, 134, 9199.
(24) Ess, D. H.; Houk, K. N. J. Am. Chem. Soc. 2008, 130, 10187.
(25) Olivella, S.; Pericas, M. A.; Riera, A.; Sole, A. J. Org. Chem.
1987, 52, 4160.
(19) Krebs, A.; Colberg, H. Chem. Ber. 1980, 113, 2007.
(20) The reactions with thiols and amines are less studied, but the
thiol reaction is known not to proceed via a radical mechanism.1a
(21) Kooyman, E. C. Discuss. Faraday Soc. 1951, 10, 163.
(26) Compound 4 is known to react with triplet oxygen ( Turro, N. J.;
Ramamurthy, V.; Liu, K.-C.; Krebs, A.; Kemper, R. J. Am. Chem. Soc.
1976, 98, 6758), but the reaction is 3 orders of magnitude slower than
that with singlet oxygen.
Org. Lett., Vol. XX, No. XX, XXXX
C

Furthermore, this hydrogen abstraction appears to be
initiated by a diradical form of the alkyne, an unusual mode
of reactivity for strained alkynes.25,26 Thus, while these
compounds have limited use as bioorthogonal reagents,
they may find applications in studies of organic radicals.
grant to C.R.B. from the NIH (GM58867). G.d.A. was
supported by a predoctoral Ruth L. Kirschstein National
Research Service Award.
Supporting Information Available. Experimental pro-
cedures and characterization of all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Acknowledgment. We are grateful to Prof. John Jewett
at the Unversity of Arizona for helpful discussions and to
Prof. Felix Fischer at UC Berkeley for access to instrumentation
and helpful discussions. This work was supported by a
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX