Formation and in situ reactions of hypervalent iodonium alkynyl triflates to form cyanocarbenes

Article abstract of DOI:10.1039/c4cc08676g

The conversion of readily available silylalkynes, iodobenzene diacetate, and azide anions was utilized to form and react cyanocarbenes. A copper(ii)-catalyzed reaction was found to react in a different manner. Both of these methods benefit from the formation and in situ reaction of hypervalent iodonium alkynyl triflates in O-H insertion reactions. This journal is

Full text of DOI:10.1039/c4cc08676g

ChemComm
View Article Online
View Journal
COMMUNICATION
Formation and in situ reactions of hypervalent
iodonium alkynyl triflates to form cyanocarbenes†
Cite this:DOI: 10.1039/c4cc08676g
I. F. Dempsey Hyatt, Daniel J. Nasrallah, Michael A. Maxwell,
A. Christina F. Hairston, Manahil M. Abdalhameed and Mitchell P. Croatt*
Received 1st November 2014,
Accepted 17th December 2014
DOI: 10.1039/c4cc08676g
www.rsc.org/chemcomm
The conversion of readily available silylalkynes, iodobenzene diacetate, insertion (via a stepwise process, not a concerted bond insertion),
6
and azide anions was utilized to form and react cyanocarbenes. A sulfoxide complexation, and more recently C–Cl insertion.
copper(II)-catalyzed reaction was found to react in a different manner.
Although this reaction was highly novel, it required the synth-
Both of these methods benefit from the formation and in situ reaction esis of hypervalent iodonium-alkynyl triflates (2, HIATs) to be used
of hypervalent iodonium alkynyl triflates in O–H insertion reactions.
as the source of electrophilic alkynes. The HIATs are specifically
electrophilic at the b-carbon, at which point the azide source adds
A major goal of organic synthesis is the formation of a target (Scheme 1). After a series of intermediates, including an iodoylide
molecule as efficiently as possible, which often involves the shortest (3), vinylidene carbene (5), and alkynyl azide (6), the cyanocarbene
1
sequence of steps. To assist in this effort, the design and develop- (7) is generated. Although this method produced the first high
5a,b
ment of new reactions that convert simple, readily available starting yielding cyanocarbene reaction,
it had two major drawbacks.
materials into products with increased molecular complexity is of First of all, this method required the formation and isolation of the
special priority. Furthermore, new reactions that are fundamentally HIATs which tend to have limited stability, and are even explosive
novel are valuable since they enable different, potentially more in some instances. Secondly, all of the reactive intermediates that
efficient retrosynthetic disconnections. Towards this goal, our are required for generation of the cyanocarbene allow for a variety
group has been interested in the reactions of alkynes and azides of side reactions to compete. The paper herein describes two
to transiently form alkynyl-azides (6) that would extrude dinitro- processes for the in situ formation and reaction of HIATs that
2
gen to form cyanocarbenes (7, Scheme 1). Although the reaction obviates the isolation of the HIAT. The first method uses either
of nucleophilic alkynes and electrophilic azides were found to commercially available alkynes or readily available metallated
3
create triazoles, it was determined that cyanocarbenes could be alkynes, iodobenzene diacetate, and sodium azide to generate
4
formed by using hypervalent iodine to generate electrophilic cyanocarbenes. The second method described utilizes a copper(II)
2
,5
alkynes and react them with nucleophilic azides. Reactions of catalyst to theoretically enable the direct addition of azide anion to
the cyanocarbene explored include cyclopropanation, O–H bond the a-carbon, a process that bypasses most of the previously formed
reactive intermediates that cause side reactions.
Reports on the theoretical and/or experimental evidence of
7
6,8
cyanocarbenes and alkynyl azides laid the ground-work for the
establishment of methods to form these molecules. The utility of
the reaction of HIATs and azides, however, is hampered by the
accessibility of HIATs. Common methods to synthesize HIATs can
be efficient, however, the isolation of the HIATs can be problematic
9
and often require cold, anaerobic conditions. Therefore, it was
decided to create a method whereby the HIATs are formed and
reacted in situ so that the process would be more approachable for
the organic chemist community.
Scheme 1 Previous method for the formation of cyanocarbenes from HIATs.
Department of Chemistry and Biochemistry, University of North Carolina at
Greensboro, Greensboro, NC, 27402, USA. E-mail: mpcroatt@uncg.edu
Investigation into new methods to generate cyanocarbenes
started by naturally expanding our previous methods to an in situ
reaction (Scheme 2) by the use of Kitamura’s previously published
†
Electronic supplementary information (ESI) available: Experimental details,
graphical spectra, and crystallographic data for 19. CCDC 1032122. For ESI and
crystallographic data in CIF or other electronic format see DOI: 10.1039/
c4cc08676g
10
reagent. Since it was determined that the rearrangement of
vinylidene carbene 5 to alkynyl azide 6 proceeds much more
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun.

View Article Online
Communication
ChemComm
containing sodium azide and catalytic amounts of Cu(dppe)(OAc)
2
.
The dppe ligand is important because the Cu(dppe)(N formed in
3 2
)
the reaction is soluble and thus lowers the safety hazards. Although
ligand screening is ongoing, initial results suggest that other biden-
Scheme 2 Method 1 generation of cyanocarbenes and deprotonation of tate ligands (TMEDA) work successfully but tridentate ligands slow
the alkenyl iodonium species (4).
down catalyst activity. As mentioned in Scheme 3, the copper(II)
catalyst facilitates an a-attack on the HIAT which then forms alkynyl
azide directly without the need of a 1,2-shift. The alkynyl azide is then
5
a
efficiently with aryl HIATs compared to an alkyl HIAT, only aryl
groups were investigated in the presented work.
allowed to decompose to the cyanocarbene which is trapped by
methanol. To confirm that the copper is not modifying the azido
alkenyl iodonium intermediates, 25 mol% Cu(dppe)(OAc)2 and
alkenyl iodonium (4, R = Ph) were placed in MeOH but no reaction
occurred even after 24 hours at room temperature. When DBU was
The decision to use Kitamura’s reagent instead of other typical
reagents such as Zefirov’s reagent, cyanophenyliodonium triflate,
11
12
13
14
Koser’s reagent, or other reagents was based on solubility, the
versatility of metals on the alkyne (M = Si, Ge, Sn, or B), chemo-
selectivity, and yield. Since the HIAT does not need to be isolated in
this methodology, unstable HIATs can be generated and used
immediately. Method 1 (Scheme 2) caused the alkenyl iodonium
product (4) to re-enter the cycle by the addition of DBU, however,
vinylidene carbene O–H insertion products still formed. The funda-
mental problem with Method 1 is that b-attack on the alkyne occurs
and the rate of 1,2-shift to form the alkynyl azide competes with
vinylidene carbene reactivity. Thus, an alternative method was
developed where a-attack might be possible.
added to a solution of iodonium 4 in the presence of Cu(dppe)(OAc)
in MeOH, products resulting from both cyanocarbene 7 and vinyli-
2
5a
dene carbene 5 were observed. This further validated that the
reactions utilizing Method 2 (Scheme 4) are having azide delivered to
the a-carbon instead of the b-carbon.
Before comparing the two methods, it was decided to first
explore the optimal catalyst loading for the copper(II) catalyst using
the Ph-HIAT (2, R = Ph; Table 1). It was determined that low copper
catalyst loading (o25%) permitted the formation of products
resulting from vinylidene carbene 5. These by-products presumably
formed via the uncatalyzed background pathway. Therefore,
A publication by Stang et al. showed that copper(I) would
transform a hypervalent iodonium alkynyl tosylate to an alkynyl
25 mol% was used in subsequent reactions.
13b
tosylate. It was theorized that the same process could apply to
our system if the anion of the HIAT went through ligand exchange
to an azide (Scheme 3). It was found that copper(I) catalysts led
to Glaser coupling products but copper(II) catalysts formed the
cyanocarbene products. The method shown in Scheme 3 was
successful in the production of cyanocarbenes but had problems
such as formation of Cu(N ) , which is highly explosive, the use of
The comparison between Method 1 and 2 was performed in
which the metal on the alkyne was altered (Table 2). Although
the yields are not drastically different between the different
metals used, the reactions times were longer for silicon and
germanium than they were for trifluoroborates and tin, as
determined by monitoring the reactions by TLC.
3
2
The next series of experiments were intended to monitor the
scope of the reaction by using different in situ prepared HIATs and
reacting them with MeOH (Table 3). Most of the HIATs in this study
are relatively unstable and tend to decompose when isolation of the
pure HIAT was attempted. The O–H insertion likely involves
nucleophilic addition of the oxygen of methanol to the electrophilic
carbene followed by proton transfer. It should be noted that the
more electron-deficient alkynes required higher temperatures
and longer times to form the HIATs and were also lower yielding
to form the products (entries 1, 2 versus entries 3, 4) with no
observable product with the nitro-group (entries 4–6).
3 2
tin-alkynes, and the insolubility of PhI(N ) at cold temperatures.
Due to these problems, the method was not explored further but it
did give some evidence that a-attack was occurring since no alkenyl
iodonium or vinylidene carbene products were produced.
Building on the method shown in Scheme 3, a new method was
developed. In Method 2 (Scheme 4), an in situ HIAT is formed from
Kitamura’s reagent and is then reacted with a methanol solution
a
Table 1 Optimization of copper catalyst
b
Entry
Cu(dppe)(OAc)
2
(mol%)
Yield (%)
1
2
3
4
5
6
100
50
25
10
5
37
51
45
22
25
19
Scheme 3 Initial result which lead to the development of Method 2.
1
a
Reaction conditions: Ph-HIAT (2, R = Ph, 0.10 g, 0.22 mmol) was dissolved
in 1 mL of DCM and cooled to À40 1C. A separate solution of sodium azide
2
(0.014 g, 0.22 mmol) and Cu(dppe)(OAc) (mol% indicated in Table) in 5 mL
methanol was prepared and transferred to the cold solution of Ph-HIAT (2).
Scheme 4 Method 2 for the generation of cyanocarbenes by a-attack on The solution was allowed to warm to r.t., concentrated, and purified by silica
b
the HIAT.
gel flash chromatography. Isolated yields of 9a (R = Ph).
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2015

View Article Online
ChemComm
Communication
Table 2 Investigation of metal alkynes and method variance
Method 1 and 2 also led to no appreciable amount of the desired
intramolecular cyanocarbene product, however, a white precipi-
tate formed in all the reactions. Analysis by X-ray crystallography
determined that a 1,1-diiodonium alkene (19) had formed. To
validate the reaction and structure, azide was removed as a
reagent and 1,1-diiodonium alkene 19 was again isolated. This
compound is the first structure of this type and was unambi-
guously characterized using X-ray crystallography (see ESI† for
X-ray data and more details).
a
b
c
d
Entry Metal Method Temperature (1C) Time (min) Yield
1
2
3
4
5
6
7
8
BF
BF
TMS
TMS
GeEt
GeEt
3
K
K
1
2
1
2
1
2
1
2
À40
À40
r.t.
r.t.
r.t.
o1
o1
10
10
10
31%
39% (16%)
28%
29%
48%
48% (16%)
39%
41%
e
e
3
3
3
r.t.
10
SnBu
SnBu
3
À40
o1
3
À40
o1
a
The described work shows the development of two methods
Reaction conditions followed either Scheme 2 (Method 1) or Scheme 4
b
(
Method 2) using HIAT 2 (R = Ph). See ESI for specifics. Temperature for the generation of cyanocarbenes from alkynes and azides
c
required for HIAT formation. Time required for HIAT formation.
using commercially available hypervalent iodine sources. Both
methods include an in situ generation of hypervalent iodonium
alkynyl triflate but are mechanistically different in how the
azide attacks the alkyne. Although the yields can be similar
between methods, there is less side-product formation when
using a copper catalyst (Method 2). Other reactions such
as cyclopropanation and intramolecular reactions remain of
current research interest in the group.
We thank the American Chemical Society Petroleum
Research Fund (52488-DNI) and the National Science Founda-
tion (CHE-1351883) for funding of this research and Dr Cynthia
Day (Wake Forest University) and the WFU Chemistry Depart-
ment X-ray Facility for acquisition and solving of crystal struc-
tures. We thank Dr Franklin J. Moy (UNCG) for analysis of NMR
data and Dr Brandie Ehrmann, Vincent Sica, and Lara Fakhouri
for acquisition of HRMS data.
d
e
NMR yields of 9a (R = Ph). Yaield in parenthesis is the vinylidene
5
carbene O–H insertion product.
a
Table 3 Substrate scope
b
Entry
R
Method
Yield (%)
Product
1
2
3
4
5
6
7
8
9
Anisole
Anisole
1
2
1
2
1
2
1
2
1
2
12
50
4
19
0
9b
9b
9c
9c
9d
9d
9e
9e
9f
p-Trifluoromethyl-phenyl
p-Trifluoromethyl-phenyl
p-Nitrophenyl
p-Nitrophenyl
m-Phenoxy-phenyl
m-Phenoxy-phenyl
Naphthyl
0
24
24
16
64
1
0
Naphthyl
9f
Notes and references
1
2
3
P. A. Wender and B. L. Miller, Nature, 2009, 460, 197.
a
I. F. D. Hyatt, M. E. Meza-Avi n˜ a and M. P. Croatt, Synlett, 2012, 2869.
M. E. Meza-Avi n˜ a, M. K. Patel, C. B. Lee, T. J. Dietz and M. P. Croatt,
Org. Lett., 2011, 13, 2984.
Reaction conditions followed either Scheme 2 (Method 1) or Scheme 4
b
(Method 2). See ESI for specifics. Isolated yields.
4
(a) V. V. Zhdankin, Hypervalent Iodine Chemistry - Preparation,
Structure and Synthetic Applications of Polyvalent Iodine Compounds,
John Wiley & Sons, West Sussex, 2014; (b) T. Wirth, Topics in Current
Chemistry, Springer-Verlag, Berlin, 2003, vol. 224, p. 264.
(a) I. F. D. Hyatt and M. P. Croatt, Angew. Chem., Int. Ed., 2012,
To diversify this methodology, an intramolecular substrate
was explored. An inherent difficulty with this approach is that
any nearby nucleophile that can attack the electrophilic cyano-
carbene can also add into the HIAT. An example of this occur-
rence is shown in Scheme 5 where the hydroxyl group attacks
the b-carbon before the azide. Unfortunately, the conditions of
5
5
1, 7511; (b) K. Banert, R. Arnold, M. Hagedorn, P. Thoss and
A. A. Auer, Angew. Chem., Int. Ed., 2012, 51, 7515; (c) T. Kitamura
and P. J. Stang, Tetrahedron Lett., 1988, 29, 1887.
X. Zeng, H. Beckers, J. Seifert and K. Banert, Eur. J. Org. Chem., 2014,
6
7
8
9
4
077.
E. Prochnow, A. A. Auer and K. Banert, J. Phys. Chem. A, 2007,
11, 9945.
1
K. Banert, M. Hagedorn, J. Wutke, P. Ecorchard, D. Schaarschmidt
and H. Lang, Chem. Commun., 2010, 46, 4058.
I. F. D. Hyatt, D. J. Nasrallah and M. P. Croatt, J. Visualized Exp.,
2
013, 79, e50886.
1
1
0 T. Kitamura, M. Kotani and Y. Fujiwara, Synthesis, 1998, 1416.
1 (a) M. D. Bachi, N. Bar-Ner, C. M. Crittell, P. J. Stang and
B. L. Williamson, J. Org. Chem., 1991, 56, 3912; (b) T. Kitamura,
R. Furuki, H. Taniguchi and P. J. Stang, Mendeleev Commun., 1991,
1
, 148.
2 H.-Y. Lee, Y. Jung, Y. Yoon, B. G. Kim and Y. Kim, Org. Lett., 2010,
2, 2672.
3 (a) P. J. Stang, Angew. Chem., Int. Ed. Engl., 1992, 31, 274;
b) P. J. Stang, B. W. Surber, Z. C. Chen, K. A. Roberts and
1
1
1
(
A. G. Anderson, J. Am. Chem. Soc., 1987, 109, 228.
1
4 V. V. Zhdankin, P. J. Persichini III, R. Cui and Y. Jin, Synlett, 2000,
719.
Scheme 5 Novel method for the synthesis of 1,1-diiodonium alkenes.
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun.