Reactions of hypervalent iodonium alkynyl triflates with azides: Generation of cyanocarbenes

Article abstract of DOI:10.1002/anie.201203062

HIAT me, baby, one more time: Cyanocarbenes have been formed by the reaction of azides with hypervalent iodonium alkynyl triflates (HIATs). Experimental evidence supports the potential intermediacy of an azide-substituted vinylidene or alkynyl azide, both of which could form a cyanocarbene. Trapping of the vinylidene and cyanocarbene includes O-H insertion, dimethyl sulfoxide coordination, and cyclopropanation reactions. Copyright

Full text of DOI:10.1002/anie.201203062

Angewandte
Chemie
DOI: 10.1002/anie.201203062
Synthetic Methods
Reactions of Hypervalent Iodonium Alkynyl Triflates with Azides:
Generation of Cyanocarbenes**
I. F. Dempsey Hyatt and Mitchell P. Croatt*
The creation of a fundamentally novel process can facilitate
the development of shorter, and often more green, synthe-
ses.[1] Additionally, those novel processes which have the
potential to reveal otherwise unimagined pathways to val-
uable reactive intermediates from simple, readily available
substances warrant closer investigation. Towards these goals,
we report herein a new method to generate cyanocarbene
intermediates (2) from hypervalent iodonium alkynyl triflates
(HIATs; 1) and azides, a reaction that converts an azide and
two carbon atoms of an alkyne into dinitrogen and a cyano-
carbene (Scheme 1). The value of this reactive intermediate,
formed from readily available starting materials, is illustrated
this theoretical work, Banert et al. reported the reactions of
chloroalkynes with sodium azide in dimethyl sulfoxide
(DMSO) which resulted in the trapping of cyanocarbene
products.[4a] The yields reported for products resulting from
cyanocarbenes were in the range of 1–25% with an average
yield of 11%. This proof-of-concept work is very significant,
but the yields and reaction conditions, which required several
days, are undesirable in terms of practicality. In the report by
the Banert group, they showed that the more reactive bromo-
and iodoalkynes did not yield products resulting from alkynyl
azides 5 or cyanocarbenes 2.[4b]
Our current research goal is to obtain cyanocarbene
intermediates and have them react with a variety of substrates
to form complex products in high yields. To accomplish these
goals, HIATs (1, Scheme 1) have been employed as they are
more electrophilic than the previously studied haloalkynes
and also have manageable stabilities. Our proposed mecha-
nism of substitution involves the addition of an azide source
to the b-carbon atom of the alkyne, thereby forming iodoylide
3 which decomposes into iodobenzene and vinylidene 4.[5]
Vinylidene 4 can then undergo a 1,2-rearrangement by
migration of either the azide or the R group to afford alkynyl
azide 5. This alkynyl azide has been shown to extrude
dinitrogen to form cyanocarbene 2 which can react with
a substrate.[3b,4] The synthesis and reactivity of HIATs (1) has
been explored by Stang et al.[6] and variations of them have
recently been synthesized by Waser et al.[7] and others.[8]
Other umpolung alkyne variations[9] were investigated for
the reactions reported herein[10] but the HIATs were chosen
for their ease in synthesis and isolation. Briefly, the HIATs
were formed by first converting terminal alkynes into the
trimethylsilyl or tributylstannyl internal alkynes.[6a,11] The
internal alkynes were then treated with Zefirovꢀs reagent,[12]
the product of the reaction of triflic anhydride and iodoso-
benzene, to form the alkynyl(phenyl)phenyliodonium triflate
(1; R = Ph) or the alkynyl(n-pentyl)phenyliodonium triflate
(1; R = n-pentyl).
À
by three different carbene reactions including O H insertion,
sulfoxide coordination, and cyclopropanation.
Scheme 1. Formation of cyanocarbenes from HIATs.
Cyanocarbenes[2] and alkynyl azides[3] have both been
studied from theoretical and experimental aspects for many
years. In 2007, Prochnow et al. published a report that
described the computationally calculated electronic struc-
tures and theoretical reactions of alkynyl azides, and con-
cluded that they would rapidly decompose to form dinitrogen
and cyanocarbenes rather than form alkynyl nitrenes.[3b] After
[*] Dr. I. F. D. Hyatt, Prof. M. P. Croatt
Department of Chemistry and Biochemistry
University of North Carolina at Greensboro
406 PAS Science Bldg., Greensboro, NC 27403 (USA)
E-mail: mpcroatt@uncg.edu
Based on the hypothesis that HIATs are reactive in the
presence of azides, and by the prior theoretical[2d,13] and
experimental[3d,14] work showing that the resultant cyanocar-
benes would be extremely reactive, the decision was made to
study the reactions using the solvent alone to react with the
cyanocarbenes (Table 1). The first reactions were run in the
presence of water (with dichloromethane as a cosolvent),
methanol, acetone, and DMSO, all of which facilitated the
dissolution of sodium azide.
The product from the reaction of Ph-HIAT (6) with water
(Table 1, entry 1) validated that the initial step of the
mechanism involved the addition of azide to the electrophilic
b-carbon atom. In this case, iodoylide 3 was protonated by
Homepage: http://www.uncg.edu/che/MitchellCroatt.html
[**] Funding for this project from The University of North Carolina at
Greensboro is gratefully acknowledged. The authors thank Dr.
Franklin J. Moy (UNCG), Dr. Brandie Ehrmann (UNCG), and T. N.
Graf and the Oberlies group (UNCG) for assisting with analysis of
NMR spectra, mass spectrometry data, and HPLC usage, respec-
tively. The authors also thank the Joint School of Nanoscience and
Nanoengineering for the use of their NMR facilities to characterize
some compounds.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201203062.
Angew. Chem. Int. Ed. 2012, 51, 1 – 5
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Table 1: Reaction conditions to form and react cyanocarbenes.[a]
temperature yielded higher amounts of 8, the product
resulting from cyanocarbene formation.
When the reaction using tetrabutylammonium azide was
run in the presence of water, the identical product was formed
as with sodium azide (compare entries 1 and 7 in Table 1). Ph-
HIAT (6) was also treated with acetone in an effort to afford
a cyano epoxide (entries 8 and 9), however, the acidity of the
a-hydrogen atom led to protonation of iodoylide 3.[17]
Increasing the temperature formed only trace amounts of 7
and a complex mixture of unknown products. For the reaction
performed with DMSO as the solvent, a known product (9)
which results from the cyanocarbene complexing with DMSO
was observed (entries 10 and 11).[4a] The yield of the isolated
product is low, however, these yields are twice as efficient as
the previously reported formation of sulfoxide adduct 9. The
low yield can be partially attributed to the aqueous extraction
workup and the lack of stability of the compound on silica gel.
Both azide sources reacted similarly and provided approx-
imately the same yield of product 9.
Entry
Solvent
T [8C]
Azide
Product
Yield [%][c]
1
2
3
4
5
6
7
8
CH2Cl2/H2O[b]
MeOH
MeOH
MeOH
MeOH
25
0
25
NaN3
NaN3
NaN3
NBu4N3
NBu4N3
NBu4N3
NBu4N3
NBu4N3
NBu4N3
NaN3
7
39
7 + 30
75
49 + 7
24 + 11
60
22
42
trace
20
27
7 + 8
8
7 + 8
7 + 8
8
7
7
7
9
9
À40
0
MeOH
25
25
CH2Cl2/H2O[b]
Acetone
Acetone
DMSO
À40
9
10
11
25
25
25
DMSO
NBu4N3
[a] Conditions: Ph-HIAT (6; 1 equiv) was dissolved in the indicated
solvent, cooled to the indicated temperature, and the azide added
(1 equiv). [b] 1:1 ratio. [c] Yield of isolated products.
To further explore these reactions, an aliphatic HIAT, n-
pentyl-HIAT (10), was synthesized and used under similar
reaction conditions (Table 2). Interestingly, 10 reacted differ-
water to yield iodonium alkene 7. Though the yield of 7 is low,
it was the only observable product of the reaction after
workup to remove the by-products. The low yield is hypothe-
sized to be due to volatile carbene products[15] and isolation
difficulties such as lack of stability on silica gel and moderate
sensitivity of 7 to light.[16] Iodonium alkene 7 was formed as
a single diastereomer, as confirmed by nOe experiments, and
is likely due to a thermodynamic effect (see below).
Table 2: Reactions of n-pentyl-HIAT (10).[a]
Entry
Solvent
T [8C]
Azide
Product
Yield [%][c]
When HIAT 6 was reacted with sodium azide in methanol
(Table 1, entry 2), two products were formed, iodonium
1
2
3
4
5
6
MeOH
MeOH
MeOH
MeOH
H2O/CH2Cl2
H2O/CH2Cl2
25
0
25
NaN3
NaN3
NBu4N3
NBu4N3
NaN3
11 + 12
11 + 12
11 + 12
11 + 12
11
23 + 4
18 + 4
28 + 14
19 + 11
36
À
alkene 7 and cyano ether 8, which results from a formal O
H insertion. The formation of 8 is assumed to proceed
through nucleophilic addition of the oxygen atom of meth-
anol to electrophilic cyanocarbene 2 with a subsequent
À40
[b]
[b]
25
25
NBu4N3
11
30
À
proton-transfer mechanism instead of a concerted O H
insertion reaction. This process is speculated based on the
[a] Conditions: n-pentyl-HIAT (10; 1 equiv) was dissolved in the
indicated solvent, cooled to the indicated temperature and the azide was
added (1 equiv). [b] 1:1 ratio. [c] Yield of isolated products.
À
large bond dissociation energy for a typical O H bond. By
altering the temperature, the selectivity of the reaction could
be altered (entries 3 and 4) and, gratifyingly, a 75% yield of
cyano ether 8, resulting from a cyanocarbene, could be
isolated. This represents a 15-fold improvement in yield for
ently with MeOH (entries 1–4). The major product is
iodonium alkene 11 in all instances with some production of
À
À
the only other O H insertion reaction of a cyanocarbene
vinyl ether 12, presumably from O H insertion from the
formed from an alkyne and an azide.[4a] It is hypothesized that
protonation of iodoylide 3 happens more readily at lower
temperature because there is not sufficient energy to rapidly
cleave iodobenzene and form vinylidene 4.
vinylidene. This O H insertion is again assumed to occur
À
through nucleophilic addition of methanol to the carbene.
There were also trace amounts of what appears to be the O H
À
insertion product from the cyanocarbene (analogous to
structure 8). It should also be noted that only Z-vinyl ether
12 was formed under all reaction conditions, which was
confirmed by nOe experiments. Similar to the reactions with
6, the major product for the reactions of 10 in the presence of
water was iodonium alkene 11, and it reacted in a completely
diastereoselective fashion (entries 5 and 6).
With reaction conditions for forming cyanocarbenes
obtained, additional substrates and reaction types were
examined (Table 3). The formation of cyanocarbenes was
further validated by the reactions with styrene and allylben-
zene when using 6 (entries 1–4), as cyclopropanation reac-
To further investigate the temperature dependence, and
to make the reaction conditions more amenable to nonpolar
substrates, tetrabutylammonium azide was used in place of
sodium azide. The hygroscopic nature of tetrabutylammo-
nium azide, and the fact that water protonates the iodoylide
(Table 1, entry 1), mandated that all manipulations of this
reagent be performed in a glovebox. The product selectivity
with temperature variance using tetrabutylammonium azide
in MeOH followed the same trend as when using sodium
azide (compare entries 2 and 3 versus entries 4–6). Similar to
the results with sodium azide, reactions run at ambient
2
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Table 3: Cyclopropanation reactions of cyanocarbenes.[a]
Entry
Solvent
Reagents
Product
Yield [%][b]
1
2
CH2Cl2
–
styrene, 6
styrene, 6
40
61
3
4
CH2Cl2
–
allylbenzene, 6
allylbenzene, 6
18
50
5
6
CH2Cl2
–
styrene, 10
styrene, 10
25
42
[a] Conditions: Ph-HIAT (6) or n-pentyl-HIAT (10) was dissolved in
a minimum amount CH2Cl2 and 1 equiv of substrate or 10 equiv of
substrate (solventless reactions) followed by the addition of 1 equiv of
NBu4N3. Reactions were performed at room temperature in the glove
box. [b] Yield of isolated products.
Scheme 2. Mechanistically insightful use of iodonium alkenes.
reported vinyl ether 16.[3g] Vinyl ether 16 is likely formed
because of the presence of the more nucleophilic methoxide
as compared to the previously neutral reaction conditions
used for experiments in Table 1. Thus, it appears that the
phenyl group assists in the formation of the alkynyl azide 5
from vinylidene 4. This assistance could be due to either the
phenyl group migrating more easily than the alkyl group
(traversing a Meisenheimer complex) or the aryl group
assisting in the migration of the azide group by stabilizing
charges at the benzylic position. More information about this
mechanism will be published in due time. Gratifyingly, when
the alkenyl iodonium 7 was treated with potassium tert-
butoxide in the presence of excess styrene, the cyclopropane
13 was isolated in 41% yield (Scheme 2). These results
solidify the proposed hypothesis that deprotonation of the
alkenyl iodonium compounds yield vinylidenes, and subse-
quently cyanocarbenes.
tions occurred in moderate yields. This reactivity is unlike the
results previously reported by Moriarty et al. where iodo-
ylides of 1,3-carbonyl compounds require a metal catalyst to
enable intermolecular cyclopropanations of styrene.[18] To
expand the scope of this reaction, 10 was used in the reaction
with styrene (entries 5 and 6). Importantly, in the absence of
a good nucleophile, such as an alkoxide, or an acidic proton,
the vinylidene rearranged with both HIATs and reacted as
a cyanocarbene. The reactions produced both diastereomers
with no stereoselectivity. The solventless reactions produced
higher yields of cyanocyclopropanes than the reactions which
dissolved the tetrabutylammonium azide and the HIAT in
CH2Cl2 (entries 2, 4, and 6). The lower yield obtained when
using CH2Cl2 can be attributed to an unknown, unstable
CH2Cl2 reaction product which could not be isolated in a pure
form. Attempts were made to isolate the product from the
reaction of the 10 with allylbenzene, however, the presumed
cyclopropanation product was not stable to various purifica-
tion techniques.
To further substantiate the hypothesized mechanism in
Scheme 1, the iodonium alkenyl triflates 7 and 11 were
analyzed in subsequent reactions (Scheme 2). Evidence that
the iodoylide 3 (Scheme 1) can undergo reversible protona-
tion was shown by complete hydrogen–deuterium exchange
when 11 was dissolved in CD3OD and sat for 48 hours (see
1H NMR spectra in the Supporting Information). Since the
iodonium alkenes are likely in equilibrium with iodoylide 3,
the diastereoselective nature of forming 7 and 11 is likely due
to thermodynamic control. Moreover, when 11 was treated
with basic methanol, azide-substituted vinyl ether 12 was
formed in 52% yield (Scheme 2). Presumably, the basic
conditions generate the iodoylide 3 which reenters the
carbene pathway described in Scheme 1.[19] Although forma-
tion of 12 could be explained by an addition/elimination
mechanism, this does not account for the H/D exchange that
was observed in deuterated methanol.
This report describes a method to obtain cyanocarbene
reactive intermediates by reacting electrophilic iodonium
alkynes with azides. The reactions presented herein are the
highest yielding for formation of cyanocarbenes[20] from
alkynes and azides and also cover a breadth of reactivity
À
including O H insertion, sulfoxide complexation, and cyclo-
propanation. Empirical data suggests the formation of the
iodoylide 3 and the azide-substituted vinylidene 4 prior to
formation of the cyanocarbene 2. Further investigation into
the reaction mechanism and potential practical uses of this
reaction process are currently underway.
Received: April 20, 2012
Revised: May 14, 2012
Published online: && &&, &&&&
Keywords: azides · carbenes · cyanides ·
.
hypervalent compounds · synthetic methods
[1] P. A. Wender, V. A. Verma, T. J. Paxton, T. H. Pillow, Acc.
To test if the aromatic group facilitates the elimination of
iodobenzene and subsequent rearrangement to the cyanocar-
bene, 7 was investigated. Unlike the analogous reaction of 11,
the cyano ether 8 was observed along with the previously
Chem. Res. 2008, 41, 40.
[2] a) Z.-X. Zhao, H.-X. Zhang, C.-C. Sun, J. Phys. Chem. A 2008,
112, 12125; b) S. Gronert, J. R. Keeffe, R. A. More OꢀFerrall, J.
Am. Chem. Soc. 2011, 133, 3381; c) G. Maier, H. P. Reisenauer,
Angew. Chem. Int. Ed. 2012, 51, 1 – 5
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
K. Rademacher, Chem. Eur. J. 1998, 4, 1957; d) K. Josef, Chem.
Phys. Lett. 2005, 403, 146.
1809; d) T. Kataoka, S. Watanabe, K. Yamamoto, Tetrahedron
Lett. 1999, 40, 931.
[3] a) S. G. D’yachkova, E. A. Nikitina, N. K. Gusarova, A. I.
Albanov, B. A. Trofimov, Russ. J. Gen. Chem. 2003, 73, 782;
b) E. Prochnow, A. A. Auer, K. Banert, J. Phys. Chem. A 2007,
111, 9945; c) R. Helwig, M. Hanack, Chem. Ber. Recl. 1985, 118,
1008; d) J. H. Boyer, R. Selvaraj, J. Am. Chem. Soc. 1969, 91,
6122; e) R. Tanaka, K. Yamabe, J. Chem. Soc. 1983, 329; f) J. H.
Boyer, C. H. Mack, N. Goebel, L. R. Morgan, J. Org. Chem.
1958, 23, 1051; g) T. Kitamura, P. J. Stang, Tetrahedron Lett.
1988, 29, 1887; h) K. Banert in Organic Azides—Syntheses and
Applications (Eds.: S. Brꢁse, K. Banert), Wiley, Chichester, 2010,
pp. 115 – 166.
[10] M. E. Meza-AviÇa, M. K. Patel, C. B. Lee, T. J. Dietz, M. P.
Croatt, Org. Lett. 2011, 13, 2984.
[11] a) A. Beauchard, V. A. Phillips, M. D. Lloyd, M. D. Threadgill,
Tetrahedron 2009, 65, 8176; b) A. E. Sheshenev, M. S. Baird,
I. G. Bolesov, A. S. Shashkov, Tetrahedron 2009, 65, 10552.
[12] a) T. Kitamura, R. Furuki, H. Taniguchi, P. J. Stang, Mendeleev
Commun. 1991, 1, 148; b) N. S. Zefirov, V. V. Zhdankin, Y. V.
Danꢀkov, A. S. Kozꢀmin, J. Org. Chem. USSR (Engl. Trans.)
1984, 20, 401.
[13] K. S. Kim, H. F. Schaefer, L. Radom, J. A. Pople, J. S. Binkley, J.
Am. Chem. Soc. 1983, 105, 4148.
[4] a) K. Banert, M. Hagedorn, J. Wutke, P. Ecorchard, D.
Schaarschmidt, H. Lang, Chem. Commun. 2010, 46, 4058; b) K.
Banert, R. Arnold, M. Hagedorn, P. Thoss, A. A. Auer, Angew.
Chem. 2012, DOI: 10.1002/ange.201203626; Angew. Chem. Int.
Ed. 2012, DOI: 10.1002/anie.201203626.
[14] a) R. Tanaka, K. Yamabe, J. Chem. Soc. Chem. Commun. 1983,
329; b) H. K. W. Hui, H. Shechter, Tetrahedron Lett. 1982, 23,
5115; c) J. S. Lodaya, G. F. Koser, J. Org. Chem. 1990, 55, 1513;
d) M. Ochiai, M. Kunishima, K. Fuji, Y. Nagao, J. Org. Chem.
1988, 53, 6144; e) A. Koumbis, C. Kyzas, A. Savva, A. Varvoglis,
Molecules 2005, 10, 1340.
[15] Iodobenzene was removed under high vacuum and presumably
other volatile by-products were removed since the mass recovery
was always significantly less than 100% after subjecting the
reaction mixture to the vacuum, even after acounting for the loss
of iodobenzene and dinitrogen. Prior to subjection to high
vacuum, an aldehydic peak was observed by 1H NMR spectros-
copy, thus implicating formation of benzaldehyde. However, this
compound was too volatile to be separated from iodobenzene
and definitively identified.
[16] If left for relatively short periods of time, alkene 7 turns black
and forms iodobenzene along with a mixture of undetermined
oligomeric products.
[17] The authors freshly distilled the acetone prior to the reaction,
however, a small amount of exogenous water as the source for
protons in this reaction cannot be excluded.
[18] R. M. Moriarty, S. Tyagi, M. Kinch, Tetrahedron 2010, 66, 5801.
[19] D. A. Gately, T. A. Luther, J. R. Norton, M. M. Miller, O. P.
Anderson, J. Org. Chem. 1992, 57, 6496.
[20] For a report on forming, characterizing, and reacting alkynyl
azides which was submitted soon after the submission of this
manuscript and published back-to-back, see Ref. [4b].
[5] T. Kitamura, L. Zheng, H. Taniguchi, M. Sakurai, R. Tanaka,
Tetrahedron Lett. 1993, 34, 4055.
[6] a) M. D. Bachi, N. Bar-Ner, C. M. Crittell, P. J. Stang, B. L.
Williamson, J. Org. Chem. 1991, 56, 3912; b) P. J. Stang, Angew.
Chem. 1992, 104, 281; Angew. Chem. Int. Ed. Engl. 1992, 31, 274;
c) P. J. Stang, B. W. Surber, Z. C. Chen, K. A. Roberts, A. G.
Anderson, J. Am. Chem. Soc. 1987, 109, 228; d) P. J. Stang, C. M.
Crittell, Organometallics 1990, 9, 3191; e) V. V. Zhdankin, M. C.
Scheuller, P. J. Stang, Tetrahedron Lett. 1993, 34, 6853; f) B. L.
Williamson, P. J. Stang, A. M. Arif, J. Am. Chem. Soc. 1993, 115,
2590.
[7] a) J. P. Brand, D. F. Gonzalez, S. Nicolai, J. Waser, Chem.
Commun. 2011, 47, 102; b) D. Fernꢂndez Gonzꢂlez, J. P. Brand,
J. Waser, Chem. Eur. J. 2010, 16, 9457.
[8] For reviews of other relevant hypervalent iodonium research,
see: a) T. Kitamura, Y. Fujiwara, Org. Prep. Proced. Int. 1997, 29,
409; b) M. Ochiai, K. Miyamoto, Eur. J. Org. Chem. 2008, 4229;
c) V. V. Zhdankin, P. J. Stang, Chem. Rev. 2008, 108, 5299.
[9] a) M. Ochiai, M. Kunishima, S. Tani, Y. Nagao, J. Am. Chem.
Soc. 1991, 113, 3135; b) S. Watanabe, S. Asaka, T. Kataoka,
Tetrahedron Lett. 2004, 45, 7459; c) T. Kataoka, Y. Banno, S.
Watanabe, T. Iwamura, H. Shimizu, Tetrahedron Lett. 1997, 38,
4
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Synthetic Methods
I. F. D. Hyatt,
M. P. Croatt*
&&&& — &&&&
Reactions of Hypervalent Iodonium
Alkynyl Triflates with Azides: Generation
of Cyanocarbenes
HIAT me, baby, one more time: Cyano-
carbenes have been formed by the reac-
tion of azides with hypervalent iodonium
alkynyl triflates (HIATs). Experimental
evidence supports the potential interme-
diacy of an azide-substituted vinylidene
or alkynyl azide, both of which could form
a cyanocarbene. Trapping of the vinyli-
À
dene and cyanocarbene includes O H
insertion, dimethyl sulfoxide coordina-
tion, and cyclopropanation reactions.
Angew. Chem. Int. Ed. 2012, 51, 1 – 5
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!