A Traceless Tether Strategy for Achieving Formal Intermolecular Hexadehydro-Diels-Alder Reactions

Article abstract of DOI:10.1021/acs.orglett.8b02473

A synthetic strategy formally equivalent to an intermolecular hexadehydro-Diels-Alder (HDDA) reaction is described. Sulfur-based linkers were designed and constructed by joining terminal alkynes or diynes using alkyne thiolate chemistry. The resulting tetraynes and triynes successfully underwent HDDA cyclization and benzyne trapping. Linker removal by reductive desulfurization was uneventful. The strategy was also found suitable for the tetradehydro-Diels-Alder (TDDA) reaction.

Full text of DOI:10.1021/acs.orglett.8b02473

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
A Traceless Tether Strategy for Achieving Formal Intermolecular
Hexadehydro-Diels−Alder Reactions
Merrick Pierson Smela and Thomas R. Hoye*
Department of Chemistry, University of Minnesota, 207 Pleasant Street SE, Minneapolis, Minnesota 55455, United States
S
* Supporting Information
ABSTRACT: A synthetic strategy formally equivalent to an intermolecular
hexadehydro-Diels−Alder (HDDA) reaction is described. Sulfur-based linkers
were designed and constructed by joining terminal alkynes or diynes using
alkyne thiolate chemistry. The resulting tetraynes and triynes successfully
underwent HDDA cyclization and benzyne trapping. Linker removal by
reductive desulfurization was uneventful. The strategy was also found suitable
for the tetradehydro-Diels−Alder (TDDA) reaction.
of a pair such as 4a and 4b to produce 5 are currently
he cyclization of a 1,3-diyne with a diynophile to generate
1
Ta benzyne intermediate in a practical and controlled unknown and would pose considerable difficulties with
kinetics4 and selectivity.
manner, a process now known as the hexadehydro-Diels−
Alder (HDDA) reaction,2 requires that the reacting alkynes be
Our current work was motivated by our interest in
identifying a tether that, once finished with its primary
tethered so as to render the process intramolecular in nature
(cf. 1 to 2, Figure 1). To date, the tethers in the vast majority
purpose of holding the reacting groups proximal, could be
cleaved to give a product identical to that of a bimolecular
HDDA reaction (cf. 3 to 6). One apparent strategy for
accomplishing this is to link the diyne and diynophile through
of reported HDDA substrates have been a three-atom linker
(cf. ABC in 1−3).3 In particular, bimolecular HDDA reactions
a terminal sulfur atom on each (cf. 7), since hydrogenolytic
removal of the sulfide moieties would introduce a pair of ortho-
hydrogen atoms in place of the earlier connecting atoms (cf. 9
to 6). This plan raised, at least, the following questions: could
sufficiently efficient chemistry be found to construct substrates
such as 7, and would the bis-sulfide linkage be compatible with
the HDDA cyclization (7 to 8)? We were optimistic that
benzyne trapping reactions of species such as 8 and final
desulfurization of 9 would proceed well.
One attractive S−L−S tether appeared to be the
methylenedithyl moiety (i.e., SCH2S). This presented the
challenge of how one might prepare a structure such as 15
(Figure 2). To demonstrate proof-of-principle, we decided to
synthesize this symmetrical tetrayne. More specifically toward
that end, alkynyllithium reagents react efficiently with
elemental sulfur (0.125 equiv S8) to generate lithium
alkynethiolates (cf. 11). Moreover, these ambident nucleo-
philes are known to cleanly alkylate at the soft sulfur atom.5
Indeed, we found that the diynes 12 and 13 could be efficiently
prepared by sequential in situ alkylation with either dibromo-
or diiodomethane, followed by alkyne desilylation. These
bisalkynes are known entities, having been previously prepared
by an alternative sequencenamely, sulfenylation of trimethyl-
silylethynyllithium with CH2(SCN)2 (11% after deprotec-
tion).6
Figure 1. Hypothesis driving this study: Temporary tethering of a 1,3-
diyne to a diynophile (cf. 7) to support intramolecular HDDA
cyclization (1 to 3 via 2, or 7 to 9 via 8), and removal of the tether (3
or 9 to 6), allows for an otherwise infeasible intermolecular HDDA
reaction (cf. 4a + 4b to 6 via 5).
Received: August 2, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b02473
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Figure 3. Alternative synthesis of tetrayne substrates and an example
of external benzyne trapping.
substrate 15. When that tetrayne was heated in the presence of
excess furan (47 equiv), the major product formed (i9) was a
furan-trapped adduct analogous to compound 21, and only a
trace of 17 was observed (<1%, 1H NMR analysis of the crude
product mixture). That is, furan reacted sufficiently quickly
with the benzyne to outcompete the intramolecular trapping
by the methoxy group.
Side reactions that we suspect diminish the yield of product
formation are (i) trapping of the benzyne intermediate by a
thioether groups,10 present in both the substrate and product
and/or (ii) methylation of those sulfur atoms by methyl-
oxonium ion intermediates. Consistent with this, we observed
(i) a diminished yield when the reaction of 15 was performed
with a higher initial concentration of tetrayne substrate and (ii)
the yield of the product 21 was significantly higher than that of
17, even though the former reaction was performed at 3-fold
higher initial substrate concentration than the latter. To
possibly protect against these sorts of processes, we explored
modifying the linker with bulky or inductively electron-
withdrawing groups.
With this goal in mind, we screened a series of geminal
dihalides in a reaction with alkynethiolate 11 to find one(s)
suitable for linker construction (Figure 4a). We found that α,α-
dibromotoluene and dichloromethyl methyl ether each readily
gave the corresponding linked diynes 23a and 23b. The
tetraynes 24a and 24b were easily formed by cross-coupling
with 14. Each successfully underwent the HDDA reaction with
trapping by the o-methoxy group to provide the dibenzofurans
25a and 25b.
While searching to identify a linker having a fully substituted
carbon between the bis-sulfide atoms, we found that 2,2-
dichlorobenzodioxole 26, known to engage a simple
arylthiolate,11 reacted with alkynethiolates (Figure 4b).
Although the hydrolytic lability of the resulting dithioorthocar-
bonate precluded use of standard Cadiot−Chodkiewicz
coupling conditions, we adapted the strategy used to prepare
20 (i.e., from diyne 19) to directly produce tetrayne 27. The
HDDA cyclization of this tetrayne in the presence of furan
proceeded uneventfully to produce the expected adduct 28 in
good yield. Similarly to compound 21, 28 underwent clean
reductive desulfurization upon treatment with Raney nickel to
yield 22.
Figure 2. Proof-of-concept: Linker synthesis, Cadiot−Chodkiewicz
coupling, HDDA, and desulfurization.
We were then pleased to see that 13 was an acceptable
donor partner in Cadiot−Chodkiewicz cross-coupling reac-
tions with 1-bromoalkynes. Accordingly, the symmetrical,
SCH2S-linked tetrayne 15 was produced upon reaction of
diyne 13 with the bromoalkyne 14. Moreover, the
alkynylsulfide moieties were compatible with the thermal
conditions required to promote HDDA cyclization. More
specifically, 15, containing this new, unique linker, showed a
1
half-life of ca. 40 min at 85 °C (external bath T, CDCl3, H
NMR monitoring). The benzodithiole derivative 17 was the
major product observed (42% yield after purification). We
have observed participation by o-methoxy groups in other
benzyne trapping reactions via what we proposed as analogs of
the zwitterion 16. The fate of the methyl group was identified
as methanol (from intervention by water), dimethyl ether,
methyl chloride, and 1,1,1-trichloroethane (the last two
following protonation of the zwitterion by chloroform) in
those experiments.7 Finally (and to establish feasibility of the
last phase of this strategy for traceless tethering), the SCH2S
moiety was reductively cleaved upon exposure of 17 to
hydrogen adsorbed on Raney nickel. This reaction also
saturated the alkyne to provide 18, the formal product of a
bimolecular HDDA reaction cascade.
To demonstrate an alternative approach for assembling
other requisite polyyne substrates, we carried out the
experiments summarized in Figure 3. Phenylbutadiyne (19)8
was lithiated, treated with S8, and exposed to CH2I2 to more
directly provide the tetrayne 20. When heated in the presence
of furan, this substrate gave rise to adduct 21 in very good
yield.
Because furan is such an efficient (fast) benzyne trapping
reagent, even though bimolecular, we were curious whether it
would outcompete the internal o-methoxy group present in
Next, we investigated the effect of central linker atom
substitution on the rate of the HDDA cycloisomerization to
the corresponding benzyne intermediate (Figure 4c). By
heating a solution of each tetrayne in deuterated chloroform,
we determined the rates by in situ NMR analysis of substrate
B
DOI: 10.1021/acs.orglett.8b02473
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Figure 5. (a) Phenyl-substituted triyne 30 is diverted to a TDDA
cycloisomerization pathway. (b) When the TDDA reaction is not
possible, triyne 36 undergoes the HDDA reaction in the normal
fashion.
resulting solution of chloromethyl sulfide intermediate (i.e., 29
or 35, the latter demonstrated to be isolable, if desired)13 was
then directly treated with a second alkynethiolate and allowed
to warm to ambient temperature. Using this procedure, we
isolated the asymmetric triynes 30 (62%) and 36 (65%).
30 as well as 36 was heated in the presence of furan, but
quite different courses of reaction occurred. The former
underwent a tetradehydro-Diels−Alder (TDDA)14 reaction to
form the naphthalene derivative 32. When performed in the
absence of furan, the reaction (46 h at 125 °C) gave 32 in high
yield, establishing that the HDDA process conceivable within
32 is not competitive.15 This shows that the traceless tether
idea is applicable to a second type of cycloisomerization, the
TDDA reaction.
On the other hand, when the terminal aryl substituent in 30
was replaced by the trimethylsilyl group in triyne 36 so as to
disallow the TDDA pathway, HDDA reactivity reemerged to
give furan adduct 37. It is also notable that the temperature for
onset of the TDDA reaction of 30 (to give 32) was lower than
that of the HDDA process in substrate 36 (to give 37). While
the HDDA required 135 °C, the TDDA product was formed,
slowly, at 100 °C (see Supporting Information).
Figure 4. Effect of the difference of linker structures on the rates and
efficiencies of the HDDA cycloisomerization and product formation.
consumption relative to an internal standard. In accordance
with our expectations, the reactions followed a first-order rate
law. The relative rates of conversion are normalized to the
slowest reactionthat of the parent tetrayne 20 at 65 °C (t1/2
= 263 min). The related substrate 15, having the same linker,
cyclized at a nearly identical rate. Under these conditions,
substrates 24a, 24b, and 27 reacted at progressively faster rates
(t1/2’s of 87, 50, and 28 min, respectively), reflecting the effect
of greater substitution. Even at ambient temperature, 24b and
27 slowly cyclized (t1/2 approximately 2 days) as observed by
NMR. HDDA reactivity at room temperature is unusual but
not without precedent.12 These experiments show that with
proper substitution a sulfur-based tether can enable an HDDA
cyclization under mild thermal conditions.
To this point, all of the examples involved the synthesis of
symmetric compounds having identical diyne units. We next
modified our approach to allow for the synthesis of the pair of
asymmetrically linked triynes 30 and 36 (Figure 5). Using
chloroiodomethane, we were able to selectively displace the
iodide ion at −78 °C with 1 equiv of an alkynethiolate; the
Not surprisingly, the sulfur−carbon bonds in both of the
product benzodithioles could be cleaved by the action of
Raney nickel (in THF) to give adducts 33 and 38, respectively.
C
DOI: 10.1021/acs.orglett.8b02473
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(3) The lone exception to date is one instance of a triyne joined
through a five-atom linker, leading to a seven-membered ring fused to
the benzyne (and derived benzenoid products).2a
(4) Marell, D. J.; Furan, L. R.; Woods, B. P.; Lei, X.; Bendelsmith, A.
J.; Cramer, C. J.; Hoye, T. R.; Kuwata, K. T. Mechanism of the
Intramolecular Hexadehydro-Diels−Alder Reaction. J. Org. Chem.
2015, 80, 11744−11754.
(5) (a) Brandsma, L.; Verkruijsse, H. D. Synthesis of Acetylenes,
Allenes, and Cumulenes: A Laboratory Manual; Elsevier: Amsterdam,
1981. (b) Sukhai, R. S.; Verboom, W.; Meijer, J.; Schoufs, M. J. M.;
Brandsma, L. A Convenient Synthesis of 1,3-Dithioles, 1,3-
Thiaselenoles and 1,3-Thiatelluroles. Recl. Trav. Chim. Pays-Bas
1981, 100, 10−13.
Additionally, when the hydrogenolysis of 32 was effected in
pyridine,16 the reaction proceeded considerably more slowly
and led to the desulfurized product 34 as the major product,
having its alkyne still intact.
In conclusion, we have (i) put forth the concept of a
traceless tether for the HDDA reaction (Figure 1), (ii)
demonstrated a proof-of-concept example using an SCH2S
linker to support the HDDA reaction, which is then readily
removed by treatment with Ra−Ni (Figure 2), (iii) explored
the effect of substitution on the methylene linker, which
increases the HDDA reactivity (Figure 4), and (iv) extended
the concept to a triyne substrate for each of the TDDA and
HDDA reactions; these triynes were prepared by one-pot
successive alkylation of two different alkynethiolates with
CH2ClI (Figure 5). The HDDA reaction can quickly generate
complex substituted arenes, and traceless tethers enable the
synthesis of a new structural class of products.
(6) Benisch, C.; Bethke, S.; Gleiter, R.; Oeser, T.; Pritzkow, H.;
Rominger, F. Syntheses and Structural Properties of Cyclic
Tetrathiadiynes. Eur. J. Org. Chem. 2000, 2000, 2479−2488.
(7) Xu, F.; Xiao, X.; Hoye, T. R. Photochemical Hexadehydro-
Diels−Alder Reaction. J. Am. Chem. Soc. 2017, 139, 8400−8403.
(8) West, K.; Wang, C.; Batsanov, A. S.; Bryce, M. R. Are Terminal
Aryl Butadiynes Stable? Synthesis and X-Ray Crystal Structures of a
Series of Aryl- and Heteroaryl-Butadiynes (Ar−C⋮C−C⋮C−H). J.
Org. Chem. 2006, 71, 8541−8544.
ASSOCIATED CONTENT
* Supporting Information
■
S
(9) The furan adduct i was isolated as a mixture (approximately 2:1)
1
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b02473.
of two coeluting diastereomeric atropisomers: H NMR (400 MHz,
CDCl3): δ 3.77 (s, 1H), 3.77 (s, 2H), 3.79 (s, 2H), 3.80 (s, 1H), 4.49
(d, J = 9.6 Hz, 0.3H), 4.50 (d, J = 9.6 Hz, 0.7H), 4.58 (d, J = 9.6 Hz,
0.7H), 4.59 (d, J = 9.6 Hz, 0.3H), 5.34 (dd, J = 1.6, 1.0 Hz, 0.7H),
5.43 (dd, J = 1.8, 0.9 Hz, 0.3H), 5.64 (dd, J = 1.7, 1.0 Hz, 0.7H), 5.65
(dd, J = 1.8, 0.9 Hz, 0.3H), 6.84−6.78 (m, 2H), 7.11−6.96 (m, 4.7H),
7.14 (dd, J = 7.5, 1.4 Hz, 0.7H), 7.24−7.18 (m, 1H), 7.41−7.36 (m,
1H), 7.52−7.48 (m, 0.3H), 7.57−7.55 (dd, J = 7.5, 1.6 Hz, 0.7H).
Experimental procedures, data for characterization of
new compounds, and copies of 1H and 13C NMR
spectra (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: hoye@umn.edu.
ORCID
Merrick Pierson Smela: 0000-0001-5816-7098
Thomas R. Hoye: 0000-0001-9318-1477
Notes
(10) Chen, J.; Palani, V.; Hoye, T. R. Reactions of HDDA-Derived
Benzynes with Sulfides: Mechanism, Modes, and Three-Component
Reactions. J. Am. Chem. Soc. 2016, 138, 4318−4321.
The authors declare no competing financial interest.
̈
̈
(11) Gross, H.; Rusche, J.; Bornowski, H. Uber α-Halogenather,
ACKNOWLEDGMENTS
■
̈
XVII. Umsetzung von Brenzcatechin-dichlormethylenather mit O-
und N-funktionellen Verbindungen. Justus Liebigs Ann. Chem. 1964,
675, 142−150.
Support for this work was provided by the Institute of General
Medical Sciences of the U.S. Department of Health and
Human Services (R01 GM65597, then R35 GM127097).
M.P.S. was supported by a Gleysteen−Heisig summer
fellowship and by the University of Minnesota Undergraduate
Research Opportunities Program.
(12) See ref 2a for an account of a serendipitous room-temperature
HDDA reaction.
(13) A small amount of this compound was observed (GC-MS)
following chromatographic purification of product 30. This suggested
that it would likely be feasible to isolate intermediates of this sort, if
desired. However, due to stench and concerns about toxicity we did
not attempt this.
(14) (a) Kawano, T.; Suehiro, M.; Ueda, I. Synthesis and Inclusion
Properties of 6,6′-Bi(Benzo[b]Fluoren-5-ol) Derivative by Cyclo-
aromatization. Chem. Lett. 2006, 35, 58−59. (b) Kimura, H.; Torikai,
K.; Miyawaki, K.; Ueda, I. Scope of the Thermal Cyclization of
Nonconjugated Ene−Yne−Nitrile System: A Facile Synthesis of
Cyanofluorenol Derivatives. Chem. Lett. 2008, 37, 662−663.
(15) For an example of a related substrate in which both HDDA and
TDDA products are formed, see ref 13.
REFERENCES
■
(1) (a) Miyawaki, K.; Suzuki, R.; Kawano, T.; Ueda, I. Cyclo-
aromatization of a Non-Conjugated Polyenyne System. Tetrahedron
Lett. 1997, 38, 3943−3946. (b) Bradley, A. Z.; Johnson, R. P.
Thermolysis of 1,3,8-Nonatriyne: Evidence for Intramolecular [2 + 4]
Cycloaromatization to a Benzyne Intermediate. J. Am. Chem. Soc.
1997, 119, 9917−9918. (c) Tsui, J. A.; Sterenberg, B. T. A Metal-
Templated 4 + 2 Cycloaddition Reaction of an Alkyne and a Diyne
To Form a 1,2-Aryne. Organometallics 2009, 28, 4906−4908.
(d) Karmakar, R.; Mamidipalli, P.; Yun, S. Y.; Lee, D. Alder-Ene
Reactions of Arynes. Org. Lett. 2013, 15, 1938−1941.
(16) Freidlin, L. K.; Polkovnikov, B. D. Effect of Pyridine on the
Rate and Selectivity of the Hydrogenation of Diphenylacetylene and
Cyclopentadiene over Raney Nickel and Platinum Black. Bull. Acad.
Sci. USSR, Div. Chem. Sci. 1956, 5, 1547−1550.
(2) (a) Hoye, T. R.; Baire, B.; Niu, D.; Willoughby, P. H.; Woods, B.
P. The Hexadehydro-Diels-Alder Reaction. Nature 2012, 490, 208−
212. (b) Diamond, O. J.; Marder, T. B. Methodology and
Applications of the Hexadehydro-Diels−Alder (HDDA) Reaction.
Org. Chem. Front. 2017, 4, 891−910.
D
DOI: 10.1021/acs.orglett.8b02473
Org. Lett. XXXX, XXX, XXX−XXX