Mapping the Interactions of I2, I., I-, and I+ with Alkynes and Their Roles in Iodocyclizations

Article abstract of DOI:10.1002/chem.201500384

A combination of experiment and theory has been used to explore the mechanisms by which molecular iodine (I₂) and iodonium ions (I⁺) activate alkynes towards iodocyclization. Also included in the analysis are the roles of atomic iodi

Full text of DOI:10.1002/chem.201500384

DOI: 10.1002/chem.201500384
Full Paper
&
Iodine
À
+
C
Mapping the Interactions of I2, I , I , and I with Alkynes and
Their Roles in Iodocyclizations
Rohan Volpe,[a] Luigi Aurelio,[a] Murray G. Gillin,[a] Elizabeth H. Krenske,*[b] and
Bernard L. Flynn*[a]
Abstract: A combination of experiment and theory has been
used to explore the mechanisms by which molecular iodine
(I2) and iodonium ions (I+) activate alkynes towards iodocyc-
toward nucleophilic attack, especially for polarized alkynes.
Addition of I2 to alkynes competes with iodocyclization, but
is reversible. This fact, together with the capacity of I2 to ac-
tivate both alkyne carbons towards nucleophilic attack,
makes I2 the reagent of choice (superior to iodonium re-
agents) for iodocyclizations of resistant substrates. The dif-
ferences in the nature of the activated intermediate formed
with I2 versus I+ can also be exploited to accomplish re-
agent-controlled 5-exo/6-endo-divergent iodocyclizations.
lization. Also included in the analysis are the roles of atomic
À
C
iodine (I ) and iodide ion (I ) in mediating the competing ad-
dition of I2 to the alkyne. These studies show that I2 forms
a bridged I2–alkyne complex, in which both alkyne carbons
are activated towards nucleophilic attack, even for quite po-
larized alkynes. By contrast, I+ gives unsymmetrical, open io-
dovinyl cations, in which only one carbon is activated
Introduction
Electrophilic activation of p bonds by halides and halonium
ions is an area of great significance to organic synthesis.[1] Over
the last 15 years, the iodocyclization of alkynes bearing teth-
ered nucleophiles (Scheme 1, 1!2) has emerged as one of the
most important synthetic applications of electrophilic p-activa-
tion.[2,3] Iodocyclizations provide access to an enormous variety
of carbo- and heterocyclic structures (2: R and Nu=carbon
and/or heteroatoms, n=1–3) including benzofurans, indoles,
benzothiophenes, quinolones, isoquinolines, imidazoles, isoxa-
zoles, thiophenes, furans, pyrroles, pyranones, lactones, and
heptenones (see Scheme 1 for examples).[2,3] Iodocyclization
chemistry also enables the construction of more complex poly-
cyclic systems, for example by iodine or iodonium ion-induced
reaction cascades (e.g., 3) or reiterative alkyne coupling/iodo-
cyclization strategies (e.g., 4).[4,5] The versatility and scope of
this chemistry are evidenced by the rapid uptake into drug dis-
covery (e.g., 5 and 6) and materials design efforts.[5–7]
Scheme 1. Iodocyclization of alkynes and applications in chemical synthesis.
Apart from mediating iodocyclizations, electrophilic activa-
tion of alkynes by iodonium reagents has been used to pro-
vide convenient stereo- and regioselective access to tri- and
tetrasubstituted alkenes [Eq (1)].[8] Moreover, the elimination of
I2 from trans-diiodoalkenes has recently been employed to
convert polydiiododiacetylenes into conducting graphenes
[Eq (2)].[9]
[a] R. Volpe, Dr. L. Aurelio, M. G. Gillin, Prof. Dr. B. L. Flynn
Medicinal Chemistry, Monash Institute of Pharmaceutical Sciences
Monash University, 381 Royal Parade, Parkville, VIC 3052 (Australia)
E-mail: bernard.flynn@monash.edu
[b] Dr. E. H. Krenske
School of Chemistry and Molecular Biosciences
The University of Queensland, St Lucia, QLD 4072 (Australia)
E-mail: e.krenske@uq.edu.au
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201500384.
Chem. Eur. J. 2015, 21, 10191 – 10199
10191
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Despite the importance of the reactions of alkynes with I2
and I+ in synthetic chemistry, the mechanistic aspects of these
classes of p-bond activation have received little attention.[10,11]
Furthermore, several classes of alkynes show a resistance to
iodine-mediated 5-endo-dig cyclization, due to the presence of
deactivating substituents and/or competing I2 addition. To
overcome these challenges, we have conducted a detailed ex-
perimental and theoretical investigation of the mechanisms of
alkyne activation by I2 and iodonium reagents. Our results
show that these reactions involve several parallel mechanistic
pathways, the outcomes of which are determined both by the
alkyne substituents and by the reagent (I2 or IX). These findings
provide guidance as to how iodocyclizations of resistant al-
kynes may be accomplished by a suitable choice of electrophil-
ic reagent and conditions. Differences in the mechanisms of
alkyne activation by I2 and IX are shown to provide a means to
control the 5-exo/6-endo regioselectivity.
tions of electron-rich substrates has proven superior to the use
of more powerful iodonium sources, which either add irreversi-
bly to the alkyne (ICl, CF3CO2I, or Pyr2I·BF4) or yield the iodocyc-
lization product in a greatly reduced yield (e.g., N-iodosuccini-
mide (NIS) at 808C gave 12 in 30–40% yield).[5b]
We have now found that making I2 addition reversible also
allows one to achieve iodocyclizations of the second class of
resistant alkynes, in which R’ is an EWG. Our chosen substrates
were the 2-alkynylanisoles 13 (Table 1). Previously, Larock and
Table 1. Optimization of iodocyclization of resistant alkynes 1 (R’=EWG).
Results and Discussion
Entry
13 ([m])
C I2
[m]
t
T
[8C]
13/14/15[a]
The competition between iodocyclization and I2 addition
1
2
3
4
5
6
7
8
9
13a (0.1)
13b (0.1)
13b (0.1)
13b (0.1)
13b (0.4)
13b (0.4)
13c (0.4)
13c (0.4)
13c (0.4)
0.15
0.15
0.15
0.15
1.2
1.2
1.2
1.2
1.2
3 h
18
18
18
80
18
80
18
18
80
0:0:100[b]
90:10:0
77:21:2
57:40:3
3:94:3
The substrate scope of many 5-endo-dig iodocyclizations (1!
2, n=1) is limited by electronic effects (Scheme 1).[2b,3e,j,r–w]
When R in 1 is an electron-donating group (EDG) or R’ is an
electron-withdrawing group (EWG), iodocyclization is often dis-
favored and competing addition of I2 or IX across the triple
bond dominates instead. For alkynes in which R is an EDG,
both Monteiro’s group[3t] and our own laboratory[5b] have dem-
onstrated that the resistance to iodocyclization is most effec-
tively overcome by using I2 as the activating reagent
(Scheme 2). Thus, iodine addition predominates at room tem-
22 h
46 h
22 h
22 h
16 h
70 h
15 d
22 h
0:0:100[c]
7:31:62
6:5:89
0:0:100[d]
[a] Ratios determined by 1H NMR spectroscopy. [b] Reference [3j]: 88%
yield of the isolated product, reaction performed in dichloromethane.
[c] 90% Yield of the isolated product. [d] 85% Yield of the isolated prod-
uct.
co-workers observed[3j] that the treatment of the electronically
unbiased alkyne 13a with I2 in dichloromethane gave the iodo-
cyclization product 15a in less than 3 h at room temperature
(Table 1, entry 1), whereas the nitro-substituted analogue 13b
gave only the iodine adduct 14b. Both reactions were report-
ed to be performed at 0.1m 13 and 0.15m I2. However, we
have found that at such concentrations the addition of iodine
to 13b is very slow. A 10% conversion to 14b was noted after
22 h (Table 1, entry 2); this increased to 21% along with trace
amounts of iodocyclized product after 46 h (entry 3). When we
performed the reaction under conditions previously found to
give complete iodocyclization of electron-rich alkynes 7 and
10 (Scheme 2: i.e., DCE, 808C and 0.15m I2),[3t,5b] we obtained
a mixture containing the starting alkyne 13b (57%), diiodoal-
kene 14b (40%) and a trace amount of iodocyclized product
15b (3%) after 22 h (Table 1, entry 4). Further heating led to
decomposition and complete conversion of 13b to 15b was
not achieved.
Scheme 2. Promoting iodocyclization of resistant alkynes 1 (R=EDG)
through reversible iodine addition/elimination.
perature (7!8 and 10!11) but becomes reversible upon
heating to 808C in 1,2-dichloroethane (DCE). This enables the
diiodoalkene to be irreversibly converted into the iodocycliza-
tion product (9 or 12). The use of I2 for promoting iodocycliza-
A striking difference was observed when the reaction was
performed at higher concentrations (0.4m 13b and 1.2m I2).
Under these conditions, diiodoalkene 14b was formed nearly
quantitatively after 22 h at 188C (Table 1, entry 5). Heating to
Chem. Eur. J. 2015, 21, 10191 – 10199
www.chemeurj.org
10192
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
808C led to complete conversion to 15b (Table 1, entry 6, 90%
yield of the isolated product). A similar pattern was observed
with the carboethoxy derivative 13c (Table 1, entries 7–9), in
which iodocyclization could even be achieved at room temper-
ature when the reaction was performed at high I2 concentra-
tion (89% complete after 15 d; Table 1, entry 8). More conven-
iently, iodocyclization of 13c was complete within 22 h at
808C (Table 1, entry 9, 85% yield of the isolated product). The
rates of iodocyclization and iodine addition were similar in
other solvents (toluene, CH3CN, MeOH) to that observed in
DCE.[12] Attempts to iodocyclize 13b with NIS were not suc-
cessful and only returned the starting material, even after sus-
tained heating (808C, DCE, 40 h).
metrical species (i.e., an iodovinyl cation), even for symmetri-
cally substituted alkynes (R1 =R2).
Heasley and co-workers studied iodine additions to arylethy-
nes.[10c] They proposed that additions to this class of alkynes
C
likely involve a radical chain process. Thus, I (formed by ho-
molysis of I2) attacks the arylethyne to give an open vinyl radi-
cal 19 (R1 =H, R2 =aryl), which attacks I2 to give the diiodoalke-
ne.[10c] They were drawn to this conclusion by the lack of sol-
vent incorporation for reactions performed in methanol; in
contrast, reactions with Br2 gave exclusively the Br/MeO ad-
ducts, presumably by an ionic pathway.
Under each of the above sets of conditions, the trans-diio-
doalkene was formed diastereospecifically.[10,13] Although Heas-
ley did not discuss the stereochemical implications of radical
pathway 19, we would argue that the participation of an open
vinyl radical is not consistent with the exclusive formation of
trans-diiodoalkene. In related work on additions to alkenes, the
groups of Skell and Rack showed that iodine addition proceeds
through a radical mechanism, at least for reactions performed
under ambient light.[14] Anti addition and third-order kinetics
were observed, leading Rack to advance two stereospecific
Proposed mechanisms for I2 addition to alkynes
To understand the variety of conditions required for achieving
iodocyclizations of different resistant substrates (Scheme 2 and
Table 1), we have examined the mechanisms of iodocyclization
and I2 addition with a combination of experimental and theo-
retical techniques. Most prior mechanistic studies concerning
the additions of iodine to alkynes have focused on reactions of
propiolates in aqueous solutions containing varying ratios of I2
and iodide salts (NaI or KI).[10a,b] These studies converged on
three ionic mechanisms (Scheme 3, 16–18). The dominant
C
AdE3 mechanisms: 1) I attacks the alkene to give a bridged
radical intermediate that then attacks I2 (20, also proposed by
Skell), and 2) I attacks the iodine–alkyne complex (21).[14b]
C
Since iodine addition to alkynes generally gives trans-diiodoal-
kenes (anti addition),[13] it seems likely that if it indeed follows
a radical mechanism then it would involve an alkyne equiva-
lent of 20 or 21, rather than open vinyl radical 19 (see also
below).
Kinetic measurements
To explore the possible involvement of radicals in the additions
of iodine to alkynes, we measured the rates of I2 addition to
13c in the light and in the dark (Table 2, entries 1 and 2). Addi-
tion was significantly faster when performed under visible light
(halogen lamp) than when performed in the dark. This result
C
suggests that iodine addition can be mediated by I . Light
Scheme 3. Previously proposed mechanisms for iodine addition to alkynes
(16–19) and alkenes (20 and 21).
versus dark had no effect on the configuration of the diiodoal-
kene (expected to be trans),[13] and did not affect the rate of io-
mechanism depends on the relative concentrations of I2 and IÀ
and on the pH. At high iodide concentrations, the AdE3 mecha-
nisms 16 and 17 are favored. In 16, nucleophilic activation of
the alkyne by IÀ slightly precedes engagement of the electro-
philic I2; this is preferred at low pH (R2 =CO2H). In 17, electro-
philic activation of the alkyne by iodine (I2–alkyne complexa-
tion) promotes nucleophilic attack by IÀ; this pathway is pre-
ferred at high pH (R2 =CO2Na). For reactions involving only I2
and alkyne (no added IÀ), an AdE2 mechanism was proposed
(18).[10b] Activation of the alkyne by an iodonium ion (following
iodine solvolysis) was suggested to give an I+–alkyne complex
that undergoes nucleophilic attack by the counterion IÀ. This
AdE2 pathway is supported by significant solvent (H2O) incor-
poration giving a-iodoketones.[10b] Though 18 is often depicted
as a symmetrically bridged iodonium cation,[2,3,10] recent calcu-
lations[11a] indicate that it is better represented as an unsym-
Table 2. Effects of light and iodide salts on iodocyclization of alkyne 13c.
Entry Conditions
13c/14c/15c
24 h 48 h
10 h
106 h
1
2
3
4
5
1.5 equiv I2
1.5 equiv I2 (light)
1.5 equiv I2, 0.05 equiv TBAI
1.5 equiv I2, 1.5 equiv TBAI
3.0 equiv I2 1.5 equiv TBAI
92:4:4 86:6:8 80:9:11 55:22:23
45:49:6 26:69:5
–
–
–
–
–
–
–
58:16:26
100:0:0 100:0:0
97:2:1 95:2:3
–
–
[a] Reactions performed in the dark unless otherwise specified.
Chem. Eur. J. 2015, 21, 10191 – 10199
www.chemeurj.org
10193
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
docyclization. We also investigated whether iodine addition
might alternatively be catalyzed by IÀ (in the dark). Small
amounts of IÀ (5 mol% tetrabutylammonium iodide (TBAI))
had no effect on the iodine addition rate, whereas larger quan-
tities inhibited reaction (Table 2, entries 3–5). The same effects
were observed with NaI in CH3CN (data not shown). Presuma-
bly, the decrease in iodine addition is attributable to decreases
doalkene 14b or iodocyclized product 15b. The geometries
and energies of the important transition states and intermedi-
ates of these reactions are shown in Figure 1.
Four mechanisms of I2 addition to 13b were identified
(Paths A–D).[18] In Path A, alkyne-mediated heterolysis of I2
(TSA) gives the I+–alkyne adduct A and IÀ, the latter of which
may combine with a second molecule of I2 to give I3À. The I+
–alkyne adduct A is best described as an iodovinyl cation, not
as a bridged iodonium cation. A bridged species could not be
located on the potential energy surface. This is consistent with
previous calculations on I+–alkyne adducts by Yamamoto
et al.,[11a] and mirrors the behavior of Br+ and Cl+, which gen-
erally favor open halovinyl cations relative to cyclic halonium
ions.[11b,d,g] Iodovinyl cation A is formed regioselectively. The
isomeric cation A’ (see lower panel of Figure 1) is disfavored
because the electron-poor nitrophenyl ring provides less stabi-
lization of the positive charge compared with the anisyl ring.
The bond lengths and angles of iodovinyl cations A and A’ in-
dicate that these cations both have considerable allenic
character.
C
in the concentrations of the I2–alkyne complex and of I that
result from conversion of I2 to I3À. The lower electrophilicity of
À
I3 compared with I2 also accounts for the decrease in iodocyc-
lization rate.
We also observed that TBAI efficiently promotes the elimina-
tion of I2 from diiodoalkenes [Eq (3)]. For this reason, iodide
salts may act as a useful alternative to pyrrolidine in the Lewis
base-promoted conversion of polydiiododiacetylenes to con-
ducting graphenes [Eq (2)].
Paths B–D commence with formation of I2–alkyne complex
B. Complex B is computed to be 6.6 kcalmolÀ1 higher in
energy than the isolated reactants (DG). The positive DG stems
primarily from entropic effects; the DH of complexation is
0.3 kcalmolÀ1. Following complex formation, Paths B–D differ
in respect of which species attacks complex B. In Path B, B
reacts with a second molecule of I2 (TSB) to give ion pair [A]I3,
which then gives diiodoalkene 14b plus I2.[19] In Path C, com-
plex B reacts with IÀ (TSC) to give diiodoalkene 14b plus IÀ. In
Theoretical calculations
We examined the mechanisms of iodine addition and iodocyc-
lization onto alkynes by means of density functional theory
(DFT) calculations.[11,15] The calculations were performed at the
M06-2X/6-311G(d,p)//B3LYP/6-31G(d)-LANL2DZ(CH2Cl2) level of
theory, modeling the solvent (dichloromethane) with the SMD
implicit model.[16] This level of theory was chosen after evaluat-
ing the performance of several functionals for reactions of rele-
vance to I2 addition and iodocyclization, namely, addition of I2
to ethyne, five RÀI bond homolyses, and four key steps in the
reactions of alkyne 13b with I2. Full details are provided in the
Supporting Information but the following comments may be
made regarding the expected accuracy of the calculations.
The mechanisms that we consider for I2 additions to alkynes
involve elementary steps representing a wide variety of reac-
tion types. Ionic pathways, radical pathways, and pathways in-
volving closed-shell, polar transition states, all compete. We do
not expect that DFT calculations can estimate the relative rates
and thermodynamics of these reactions with quantitative accu-
racy. Errors arise from several sources, including: 1) The limita-
tions in the ability of a single functional to model chemically
diverse reactivity; 2) errors associated with the computation of
free energies in solution, especially for ionic species; 3) certain
mechanisms of I2 addition having bimolecular rate-determining
transition states, whereas others are termolecular, and 4) the
·
[20]
C
Path D, B reacts with I (TSD) to give 14b plus I . The pre-
À
ferred site of attack differs for IÀ and I ; I attacks b to the ni-
C
C
trophenyl ring, whereas I attacks b to the anisyl ring.
The values of DG° for Paths A–D are 30.2, 24.3, 26.0, and
14.7 kcalmolÀ1, respectively. Qualitatively, these values predict
that the most facile mechanism of diiodoalkene formation is
the radical pathway (Path D). This is consistent with our experi-
mental observations (Table 2) that showed that I2 addition to
13c is accelerated by ambient light. Ionic Path C is considera-
bly higher in energy than Path D. This result, coupled with our
experimental finding that iodide salts either had no effect or
inhibited I2 addition to 13c (Table 2), suggests that Path C is
not mechanistically significant. Under typical experimental con-
ditions, the available concentration of IÀ is low, either because
no iodide salt is added or because any added IÀ combines
À
with I2 (present in excess) to form I3
.
If no IÀ or I were present in the reaction mixture, the possi-
ble mechanisms of I2 formation would be Paths A and B, both
of which involve iodovinyl cation A. The barriers for these two
pathways (30.2 and 24.3 kcalmolÀ1, respectively) are lower
than the barrier for iodocyclization (Path E, 31.1 kcalmolÀ1). Ex-
perimentally, we found that I2 addition to 13b is kinetically fa-
vored over iodocyclization, even in the dark (Table 1, entry 5).
For 13c under similar conditions (Table 2, entry 1), iodocycliza-
tion has about the same rate as I2 addition. In view of these
observations, we cannot rule out the involvement of Path A or
Path B. Path A is first order in I2, whereas Paths B and D are
second order in I2. However, even in the dark, where photoin-
C
À
À
C
concentrations of several intermediates (e.g., I , I , and I3
)
under experimental conditions are not known with certainty.[17]
Therefore, we do not expect the DFT calculations to provide
definitive mechanistic conclusions, but rather we use the calcu-
lations as a qualitative tool to support the inferences derived
from our experimental studies.
Our initial calculations examined the reactions of a represen-
tative electron-poor alkyne (13b) with I2, leading to either diio-
Chem. Eur. J. 2015, 21, 10191 – 10199
www.chemeurj.org
10194
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 1. Mechanisms of iodine addition (Paths A–D) and iodocyclization (Path E) for electron-deficient alkyne 13b in dichloromethane, computed at the
M06-2X/6-311G(d,p)//B3LYP/6-31G(d)-LANL2DZ(CH2Cl2) level of theory. Distances in Š, DG in kcalmolÀ1
.
Chem. Eur. J. 2015, 21, 10191 – 10199
www.chemeurj.org
10195
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
duced homolysis of I2 is suppressed, there is still likely to be
appears likely that iodovinyl cation formation plays an impor-
tant role in the reactions of 10a.
C
a non-negligible concentration of I , due to thermal homolysis
of I2 (or of B).[21] These trace quantities of I would catalyze the
Furthermore, the calculations indicate that the participation
of an iodovinyl cation is consistent with the observed stereose-
lectivity of addition. Experimentally, only a single isomer of
diiodoalkene is obtained, assumed to be trans.[13] Calculations
predict that addition of IÀ to G on the side syn to phenyl (lead-
ing to trans diiodoalkene 11 a) has a 4 kcalmolÀ1 lower barrier
than addition syn to the iodo substituent leading to the cis
C
C
addition of I2 to 13b/c. In the light, the concentration of I
would be much higher. Therefore, we propose that the domi-
nant mechanism of I2 addition to electron-poor alkynes such
C
as 13b and 13c both in the light and in the dark is the I -medi-
ated pathway, Path D.
The predicted mechanism of iodocyclization of 13b is
Path E. This involves intramolecular nucleophilic attack on I2–
alkyne complex B. In principle, an alternative mechanism of io-
docyclization could involve intramolecular nucleophilic attack
in iodovinyl cation A’. However, the very high energy of A’
(40 kcalmolÀ1) makes such a mechanism unlikely.
C
isomer (see the Supporting Information). By contrast, I attack
on I2–alkyne complex H (Path D) is necessarily stereospecific,
giving the trans diiodoalkene. Thus, Paths A, B, and D are all
predicted to lead exclusively to the trans isomer of diiodoal-
kene 11 a, consistent with the experimental result.[22]
As with electron-poor alkyne 13b, iodocyclization of 10a
occurs most readily through Path E. The alternative mechanism
for iodocyclization, involving iodovinyl cation G’, is high in
energy (see TSK and TSL). In agreement with the experiment,
the calculations predict that the kinetic product of the reaction
of 10a with I2 is diiodoalkene 11 a and the thermodynamic
product is iodocyclized compound 12a.
The predicted importance of ionic pathways for I2 addition
to 10a, but not for 13b, is consistent with the experimental
observation that electron-rich alkynes undergo iodine addition
much faster than do electron-deficient alkynes (in the dark).
Electron-donating substituents
The calculations correctly predict the observed thermody-
namic and kinetic products of the reaction of 13b with I2. Ex-
perimentally, diiodoalkene 14b is the kinetic product and iodo-
cyclized product 15b is the thermodynamic product (Table 1).
In the calculations, I2 addition through Path C is the kinetically
favored process, with DG° =14.7 kcalmolÀ1 (provided that ho-
molysis of I2 is not rate-limiting) and an overall DG of À4.0 kcal
molÀ1. Iodocyclization (Path E) has a higher barrier (31.1 kcal
molÀ1) but is thermodynamically favored (À25.6 kcalmolÀ1).
Analogous calculations were performed for I2 addition and
iodocyclization onto electron-rich alkyne 10a (Figure 2). Com-
render I2 addition facile enough
to proceed at room temperature.
For example, the treatment of
10a with 1.0 equivalent of
iodine (0.2m in DCE) gave
>50% conversion to diiodoal-
kene 11 a in <0.5 h at 188C
(data not shown), compared
with 10% conversion after 22 h
at 188C for 13b (Table 1,
entry 2). For 10a, Paths A, B, and
D each become reversible upon
heating, allowing thermally-in-
duced conversion of diiodoal-
kene back to alkyne plus I2 (or
I2–alkyne complex H), followed
by iodocyclization.
Implications for the competi-
tion between I2 addition and
iodocyclization
Figure 2. Mechanisms of iodine addition and iodocyclization for electron-rich alkyne 10a, computed at the M06-
2X/6-311G(d,p)//B3LYP/6-31G(d)-LANL2DZ(CH2Cl2) level of theory. DG in kcalmolÀ1
.
In light of our kinetic measure-
ments and theoretical studies,
pared with the electron-poor alkyne 13b, the main difference
in the mechanistic picture for 10a is that the ionic pathways
involving iodovinyl cation (Paths A and B) are significantly
more facile for 10a. The barriers for these two pathways drop
by 6–13 kcalmolÀ1 on going from 13b to 10a, such that they
now lie only 2–3 kcalmolÀ1 above radical Path D. Taking into
account the expected degree of accuracy of the calculations, it
we propose two general mechanistic scenarios for the compe-
tition between iodine addition and iodocyclization onto al-
kynes (Scheme 4). For electron-deficient alkynes, we propose
that iodine addition proceeds primarily through a radical chain
mechanism (25, 27), resembling the pathways proposed by
Skell and Rack for analogous reactions of alkenes (20 and 21,
C
Scheme 3). Catalytic I may arise from homolysis of I2 and/or
Chem. Eur. J. 2015, 21, 10191 – 10199
www.chemeurj.org
10196 ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
mal barrier to iodocyclization 25!26. A high concentration of
I2 is also important for iodocyclization because it affords an ele-
vated concentration of 25. This is especially true at high tem-
perature, in which complexation is entropically disfavored
(Table 1, compare entries 4 and 6).
For electron-rich alkynes (22, A-ring and/or R=EDG), we
propose that radical reactions still provide the lowest-energy
mechanism of I2 addition, but pathways involving iodovinyl
cations also become important: That is, either alkyne-mediated
heterolysis of I2 or attack by I2 on the I2–alkyne complex, giving
iodovinyl cations 28 and 28’. Interconversion between 28 and
28’ may involve reversion to I2–alkyne complex 25 and/or
alkyne+I2, or, alternatively, a direct 1,2-I+ shift through transi-
tion state TSM. Cation 28 is favored when R is more electron-
donating than the A-ring, and vice versa for 28’. Cation 28
favors cyclization to 23, whereas 28’ favors reaction with IÀ to
give 24. For substrates that favor I2 addition over iodocycliza-
tion (i.e., 22, A-ring=EDG), higher temperatures cause the I2
addition reaction to become reversible, by promoting thermal
heterolysis of 24 to 28’. This affords sustained access to 25
and/or 28 (in equilibrium with 28’), either of which may under-
go iodocyclization. For these alkynes, high I2 concentrations
are not necessary to effect iodocyclization. This can be attribut-
ed to 1) the unimolecular nature of heterolysis of 24 that leads
to re-formation of 28’, and 2) the higher nucleophilicity of the
alkyne, which favors reaction of 22 (A-ring=EDG) with iodine
to give 28 or 25 from which iodocyclization can occur.
endo/exo Selectivity
Further experimental support for the intermediacy of geomet-
rically distinct I2–alkyne (symmetrical) and iodovinyl (unsym-
metrical) intermediates is provided by the exo/endo selectivities
of iodocyclizations involving 1-(2-(methylthio)phenyl)propy-
nones 29 (Scheme 5). We previously reported that the reac-
tions of various analogues of 29 with I2 favor 5-exo products
30,[4a] whereas Larock et al. subsequently reported that ICl
gives exclusively 6-endo products.[3k] We have reinvestigated
Scheme 4. Proposed mechanisms of iodocyclization and iodine addition
onto different classes of alkynes.
homolysis of the I2–alkyne complex 25 to give 27. Notably, the
calculations predict that the IÀI bond in I2–alkyne complex B
(Figure 1) is 6 kcalmolÀ1 weaker with respect to homolysis than
that of I2 itself.[21] Irrespective of the source of I , attack on the
C
I2–alkyne complex 25 gives trans-diiodoalkene 24 stereospecifi-
C
cally. Alternatively, but less favorably, I addition to the alkyne
gives the bridged radical 27, which reacts with I2 through the
C
same transition state as 25+I to give the same product, 24.
The AdE3 nature of both pathways (22!25!24 and 22!
27!24) accounts for the observation that at room tempera-
C
ture in the dark, in which the I concentration is smallest, the
rate of addition is sensitive to I2 concentration. The termolecu-
lar reaction of alkyne with 2”I2 (Path B, Figure 1) may also play
a minor role.
For electron-deficient alkynes (1, R=EWG), the rate of nucle-
C
ophilic attack by XMe is slower than I -catalyzed iodine addi-
tion, even in the dark (compare TSD and TSE, Figure 1). Heat-
C
ing accelerates homolysis of I2 and/or 25 to give I , causing the
Scheme 5. endo/exo Selectivities of iodocyclizations of resistant alkynes 29.
[a] Ratio determined by using H NMR spectroscopy (yield of the isolated
1
iodine addition reaction to become rapidly reversible and pro-
viding sustained access to 25, as well as overcoming the ther-
product in parentheses). [b] Isolated as an E/Z mixture.
Chem. Eur. J. 2015, 21, 10191 – 10199
www.chemeurj.org
10197
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
this reaction by treating 29 with both I2 and ICl in dichlorome-
thane (Scheme 5). As previously reported,[3k,4a] treatment with
I2 favored the exo-isomer 30, whereas ICl gave exclusively the
endo-isomer 31.[23] Differences in the dominant reactive inter-
mediate as a function of reagent explain these results. For I2,
the bridged intermediate 32 is formed, which can cyclize by
either the exo- or endo-pathway and shows modest selectivity
for the exo-isomer 30.[24] On the other hand, ICl acts as an io-
donium source that gives essentially exclusive formation of the
more stabilized iodovinyl cation 33 relative to 34 and thus
strongly favors the endo-product 31. A broader examination of
the scope of this exo/endo divergence is currently underway.
Keywords: alkynes
calculations · iodine · synthetic methods
·
cyclization
·
density functional
[1] For reviews, see: a) M. B. Smith, March’s Advanced Organic Chemistry:
Reactions, Mechanisms, and Structure, 7th ed., Wiley-Interscience, Hobo-
ken, NJ, 2013, ch. 15; b) P. B. de La Mare, R. Bolton, Electrophilic Addi-
tions to Unsaturated Systems, 2nd ed., Elsevier, New York, 1982; c) G. H.
Schmid in The Chemistry of Double-Bonded Functional Groups, Vol. 2,
Suppl. A, Part 1 (Ed.: S. Patai), Wiley, New York, 1989, p. 699; d) M. F.
Ruasse, Adv. Phys. Org. Chem. 1993, 28, 207; e) R. S. Brown, Acc. Chem.
Res. 1997, 30, 131; f) G. Bellucci, R. Bianchini, C. Chiappe in Advances in
Organobromine Chemistry II, (Eds.: J.-R. Desmurs, B. Gerard, M. J. Gold-
stein), Elsevier, Amsterdam, 1995, p. 128; g) G. A. Olah, K. K. Laali, Q.
Wang, G. K. S. Prakash, Onium Ions, Wiley, New York, 1998; ch. 6;
h) G. A. Olah, Halonium Ions, Wiley, New York, 1975.
[2] For reviews, see: a) B. Godoi, R. F. Schumacher, G. Zeni, Chem. Rev. 2011,
111, 2937; b) A. K. Banerjee, M. S. Laya, E. V. Carbrera, Curr. Org. Chem.
2011, 15, 1058–1080; c) A. V. Dubrovskiy, A. Nataliya, N. A. Markina,
R. C. Larock, Comb. Chem. High Throughput Screening 2012, 15, 451;
d) P. T. Parvatkar, P. S. Parameswaran, G. Santosh, S. G. Tilve, Chem. Eur. J.
2012, 18, 5460; e) A. Palisse, S. F. Kirsh, Org. Biomol. Chem. 2012, 10,
8041.
Conclusion
Kinetic and theoretical studies provide new insights into the
mechanistic features responsible for the different outcomes of
iodocyclizations involving I2 and iodonium reagents. Thus, io-
docyclizations mediated by I2 generally involve a relatively
symmetrical I2–alkyne complex, whereas iodocyclizations medi-
ated by iodonium reagents involve (unsymmetrical) iodovinyl
cations and in this respect resemble the reactions of Br2 and
Cl2 more closely. Any unfavorable electronic effects associated
with the alkyne substituents tend to be mitigated in reactions
mediated by I2, because the I2–alkyne complex is relatively
symmetrical and capable of activating both carbons to nucleo-
philic attack. In contrast, in reactions mediated by iodonium re-
agents, undesired reactivity is traced to the formation of the
undesired regioisomer of iodovinyl cation intermediate. These
differences are demonstrated by the superior capacity of I2 to
promote 5-endo-dig iodocyclizations of resistant alkynes 22
(R=EWG or A=EDG) and by the abilities of I2 and I+ to play
complementary roles in the endo/exo-divergent iodocycliza-
[3] See, for example: a) W. M. R. ten Hoedt, G. van Koten, J. G. Noltes, Synth.
Commun. 1977, 7, 61; b) M. G. Banwell, B. L. Flynn, A. C. Wills, E. Hamel,
Aust. J. Chem. 1999, 52, 767; c) R. C. Larock, D. Yue, Tetrahedron Lett.
2001, 42, 6011; d) J. Barluenga, M. Trincado, E. Rubio, J. M. Gonzalez,
Angew. Chem. Int. Ed. 2003, 42, 2406; Angew. Chem. 2003, 115, 2508;
e) A.-Y. Peng, Y.-X. Ding, Org. Lett. 2004, 6, 1119; f) A. Arcadi, S. Cacchi,
G. Fabrizi, F. Marinelli, L. Moro, Synlett 1999, 1432; g) B. L. Flynn, P. Ver-
dier-Pinard, E. Hamel, Org. Lett. 2001, 3, 651; h) D. Yue, R. C. Larock, J.
Org. Chem. 2002, 67, 1905; i) A. Arcadi, G. Bianchi, F. Marinelli, Synthesis
2004, 610; j) D. Yue, T. Yao, R. C. Larock, J. Org. Chem. 2005, 70, 10292;
k) C. Zhou, A. V. Dubrovsky, R. C. Larock, J. Org. Chem. 2006, 71, 1626;
l) D. Yue, N. Della Cµ, R. C. Larock, J. Org. Chem. 2006, 71, 3381; m) C. T.
Bui, B. L. Flynn, J. Comb. Chem. 2006, 8, 163; n) T. Kesharwani, S. A. Wor-
likar, R. C. Larock, J. Org. Chem. 2006, 71, 2307; o) S. A. Worlikar, T. Ke-
sharwani, T. Yao, R. C. Larock, J. Org. Chem. 2007, 72, 1347; p) D. Alves,
C. Luchese, C. W. Nogueira, G. Zeni, J. Org. Chem. 2007, 72, 6726; q) G.
Lamanna, S. Menichetti, Adv. Synth. Catal. 2007, 349, 2188; r) D. Fischer,
H. Tomeba, N. K. Pahadi, N. T. Patil, Y. Yamamoto, Angew. Chem. Int. Ed.
2007, 46, 4764; Angew. Chem. 2007, 119, 4848; s) F. Manarin, J. A.
Roehrs, G. R. Mozzaquatro, R. Brandao, P. H. Menezes, C. W. Nogueira, G.
Zeni, J. Org. Chem. 2009, 74, 2153; t) G. Raffa, S. Belot, G. Balme, N.
Monteiro, Org. Biomol. Chem. 2011, 9, 1474; u) C.-H. Cho, F. Shi, D.-I.
Jung, B. Neuenswander, G. H. Lushington, R. C. Larock, ACS Comb. Sci.
2012, 14, 403; v) T. Okitsu, K. Nakata, K. Nishigaki, N. Michioka, M Karata-
ni, A. Wada, J. Org. Chem. 2014, 79, 5914; w) A. S. Santana, D. B. Carval-
ho, N. S. Cassemiro, L. H. Viana, G. R. Hurtado, M. S. Amaral, N. M.
Kassab, P. G. Guerrero, Jr., S. L. Barbosa, M. J. Dabdoub, A. C. M. Baroni,
Tetrahedron Lett. 2014, 55, 52.
tions of polarized alkynes such as 29. The mechanistic insights
about the differing roles of I2, I , I and IÀ in alkyne activation
+
C
reported herein should facilitate the further development of
reaction classes involving this chemistry, having broad implica-
tions for organic synthesis.
Experimental Section
[4] For cyclization cascades, see: a) K. O. Hessian, B. L. Flynn, Org. Lett.
2003, 5, 4377; b) K. L. Hessian, B. L. Flynn, Org. Lett. 2006, 8, 243; c) R.
Halim, P. J. Scammells, B. L. Flynn, Org. Lett. 2008, 10, 1967; d) R. Halim,
L. Aurelio, P. J. Scammells, B. L. Flynn, J. Org. Chem. 2013, 78, 4708; e) J.
Barluenga, H. Vµzquez-Villa, I. Merino, A. Ballesteros, J. M. Gonzµlez,
Chem. Eur. J. 2006, 12, 5790; f) Y. Liu, S. Zhou, Org. Lett. 2005, 7, 4609.
X. Huang, W. Fu, M. Miao, Tetrahedron Lett. 2008, 49, 2359; g) Y.-X. Xie,
Z.-Y. Yan, B. Qian, W.-Y. Deng, D.-Z. Wang, L.-Y. Wu, X.-Y. Liu, Y.-M. Liang,
Chem. Commun. 2009, 5451; h) J. Barluenga, H. Vµzquez-Villa, A. Balles-
teros, J. M. Gonzµlez, J. Am. Chem. Soc. 2003, 125, 9028.
All experimental and computational details can be found in the
Supporting Information, which includes compound preparation
and characterization, copies of spectra of new compounds, theo-
retical benchmarking for I2 additions and iodocyclizations, and
computed geometries and energies.
Acknowledgements
[5] For iterative coupling-iodocyclization approaches, see: a) S. Mehta, R. C.
Larock, J. Org. Chem. 2010, 75, 1652; b) L. Aurelio, R. Volpe, R. Halim,
P. J. Scammells, B. L. Flynn, Adv. Synth. Catal. 2014, 356, 1974.
[6] For applications in medicinal chemistry, see: a) Y. He, D. Duckett, W.
Chen, Y. Y. Ling, M. D. Cameron, L. Lin, C. H. Ruiz, P. V. LoGrasso, T. M. Ka-
menecka, M. Koenig, Bioorg. Med. Chem. Lett. 2014, 24, 161; b) A. Nakhi,
M. S. Rahman, S. Archana, R. Kishore, G. P. K. Seerapu, K. L. Kumar, D.
Haldar, M. Pal, Bioorg. Med. Chem. Lett. 2013, 23, 4195; c) Y. He, S. Liu, A.
Menon, S. Stanford, E. Oppong, A. M. Gunawan, L. Wu, D. J. Wu, A. M.
We thank the Australian Research Council (DP110100835 to
B.L.F. and FT120100632 to E.H.K.) for financial support. Compu-
tational resources were provided by the Australian Govern-
ment through the National Computational Infrastructure Na-
tional Facility under the National Computational Merit Alloca-
tion Scheme, and by the University of Queensland Research
Computing Centre.
Chem. Eur. J. 2015, 21, 10191 – 10199
www.chemeurj.org
10198
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Barrios, N. Bottini, A. C. B. Cato, Z.-Y. Zhang, J. Med. Chem. 2013, 56,
4990; d) L.-F. Zeng, J. Xu, Y. He, R. He, L. Wu, A. M. Gunawan, Z.-Y.
Zhang, ChemMedChem 2013, 8, 904; e) X. Zhang, Y. He, S, Liu, Z. Yu,
Z.-X. Jiang, Z. Yang, Y. Dong, S. C. Nabinger, L. Wu, A. M. Gunawan, L.
Wang, R. L. Chan, Z.-Y. Zhang, J. Med. Chem. 2010, 53, 2482; f) P. Mahesh
Kumar, K. S. Kumar, C. L. T. Meda, G. R. Reddy, P. K. Mohakhud, K. Muk-
kanti, G. R. Krishna, C. M. Reddy, D. Rambabu, K. S. Kumar, K. K. Priya,
K. S. Chennubhotla, R. K. Banote, P. Kulkarni, K. V. L. Parsa, M. Pal, Med-
ChemComm 2012, 3, 667; g) A. Nakhi, R. Adep, R. Rambabu, R. Kishore,
G. R. Vanaja, A. M. Kalle, M. Pal, Bioorg. Med. Chem. Lett. 2012, 22, 4418–
4427; h) J. H. Kalin, H. Zhang, S. Gaudrel-Grosay, G. Vistoli, A. P. Kozikow-
ski, ChemMedChem 2012, 7, 425; i) T. Vang, Y. Xie, W. H. Liu, D. Viovi, Y.
Liu, S. Wu, D. H. Smith, A. Rinderspacher, C. Chung, G. Gong, T. Mustelin,
D. W. Landry, R. C. Rickert, S. C. Schuerer, A.-X. Deng, L. Tautz, J. Med.
Chem. 2011, 54, 562; j) R. Terazawa, D. R. Garud, N. Hamada, Y. Fujita, T.
Itoh, Y. Nozawa, K. Nakane, T. Deguchi, M. Koketsu, M. Ito, Bioorg. Med.
Chem. 2010, 18, 7001; k) Z. Huang, Y. He, X. Zhang, A. Gunawan, L. Wu,
Z.-Y. Zhang, C. F. Wong, Chem. Biol. Drug Des. 2010, 76, 85; l) B. Li, B.
Zhou, H. Lu, L. Ma, A.-P. Peng, Eur. J. Med. Chem. 2010, 45, 1955; m) M.-
J. R. P. Queiroz, R. C. Calhelha, L. A. Vale-Silva, E. Pinto, M. S. Nascimento,
Eur. J. Med. Chem. 2009, 44, 1893; n) H. Koolman, T. Heinrich, H.
Boettcher, W. Rautneberg, M. Reggelin, Bioorg. Med. Chem. Lett. 2009,
19, 1879; o) E. P. Santín, H. Khanwalkar, J. Voegel, P. Collette, P. Mauvais,
H. Gronemeyer, . R. de Lera, ChemMedChem 2009, 4, 780; p) Z. Wu,
G. S. Minhas, D. Wen, H. Jiang, K. Chen, P. Zimniak, J. Zheng, J. Med.
Chem. 2004, 47, 3282; q) A. M. Palmer, G. Muench, C. Brehm, P. J. Zim-
mermann, W. Buhr, M. P. Feth, W. A. Simon, Bioorg. Med. Chem. 2008,
16, 1511; r) R. Rossi, A. Carpita, F. Bellina, P. Stabilea, L. Manninab, Tetra-
hedron 2003, 59, 2067; s) I. L. Pinto, H. F. Boyd, D. M. B. Hickey, Bioorg.
Med. Chem. Lett. 2000, 10, 2015.
[14] a) P. S. Skell, R. R. Pavlis, J. Am. Chem. Soc. 1964, 86, 2956; b) C. J. Michej-
da, R. L. Ayres, E. P. Rack, J. Am. Chem. Soc. 1971, 93, 1389.
[15] For theoretical studies of iodocyclizations onto alkenes, see: a) M. Tred-
well, J. A. R. Luft, M. Schuler, K. Tenza, K. N. Houk, V. Gouverneur, Angew.
Chem. Int. Ed. 2008, 47, 357; Angew. Chem. 2008, 120, 363; b) T. Hu, K.
Liu, M. Shen, X. Yuan, Y. Tang, C. Li, J. Org. Chem. 2007, 72, 8555; c) R.
Galeazzi, G. Martelli, G. Mobbili, M. Orena, S. Rinaldi, Tetrahedron 2006,
62, 10450; d) J. M. Jordµ-Gregori, M. E. Gonzµlez-Rosende, P. Cava-Mon-
tesinos, J. Sepffllveda-Arques, R. Galeazzi, M. Orena, Tetrahedron: Asym-
metry 2000, 11, 3769; e) J. M. Jordµ-Gregori, M. E. Gonzµlez-Rosende, J.
Sepffllveda-Arques, R. Galeazzi, M. Orena, Tetrahedron: Asymmetry 1999,
10, 1135. For computations on the mechanism of iodine-catalyzed cis/
trans equilibration of alkenes, see: f) S. S. Hepperle, Q. Li, A. L. L. East, J.
Phys. Chem. A 2005, 109, 10975.
[16] B3LYP: a) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785; b) A. D.
Becke, J. Chem. Phys. 1993, 98, 1372; c) A. D. Becke, J. Chem. Phys. 1993,
98, 5648; d) P. J. Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J.
Phys. Chem. 1994, 98, 11623. M06–2X: e) Y. Zhao, D. G. Truhlar, Theor.
Chem. Acc. 2008, 120, 215. SMD: f) A. V. Marenich, C. J. Cramer, D. G.
Truhlar, J. Phys. Chem. B 2009, 113, 6378. Calculations were performed
in Gaussian 09: g) Gaussian 09, Revision D.01, M. J. Frisch, G. W. Trucks,
H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani,
V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakaji-
ma, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr. , J. E.
Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N.
Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C.
Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Och-
terski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman,
J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian, Inc. Wallingford CT, 2009.
[17] For a recent relevant discussion of the limitations of DFT treatments of
multi-step, multi-component reactions of polar molecules, see R. E.
Plata, D. A. Singleton, J. Am. Chem. Soc. 2015, 137, 3811.
[7] M. B. Goldfinger, K. B. Crawford, T. M. Swager, J. Am. Chem. Soc. 1997,
119, 4578.
´
[8] a) J. Barluenga, M. A. Rodriguez, P. J. Campos, J. Org. Chem. 1990, 55,
´
3104; b) J. Barluenga, I. Llorente, L. J. Alvarez-Garciµ, J. M. Gonzalez, P. J.
Campos, M. R. Diµz, S. Garciµ-Granda, J. Am. Chem. Soc. 1997, 119, 6933;
c) N. HØnaff, A. Whiting, J. Chem. Soc. Perkin Trans. 1 2000, 395; d) N.
Okamoto, Y. Miwa, H. Minami, K. Takeda, R. Yanada, J. Org. Chem. 2011,
76, 9133.
[18] With respect to earlier mechanistic proposals (Scheme 3), Path A corre-
sponds to 18, Path C to 17, and Path D to 21. However, the calculations
indicate that 20 is equivalent to 21, insofar as the structure of the tran-
sition state is identical (TSD) and the reactants are the same in both
[9] a) L. Luo. D. Resch, C. Wilhelm, C. N. Young, G. P. Halada, R. J. Gambino,
C. P. Grey, N. S. Goroff, J. Am. Chem. Soc. 2011, 133, 19274; b) A. W. Sun,
J. W. Lauher, N. S. Goroff, Science 2006, 312, 1030; c) L. Luo, C. Wilhelm,
A. Sun, C. P. Grey, J. W. Lauher, N. S. Goroff, J. Am. Chem. Soc. 2008, 130,
7702; d) C. Wilhelm, S. A. Boyd, S. Chawda, F. W. Fowler, N. S. Goroff,
G. P. Halada, C. P. Grey, J. W. Lauher, L. Luo, C. D. Martin, J. B. Parise, C.
Tarabre, J. A. Webb, J. Am. Chem. Soc. 2008, 130, 4415.
[10] For mechanistic studies of iodine addition to alkynes, see: a) M. H.
Wilson, E. Berliner, J. Am. Chem. Soc. 1971, 93, 4126; b) E. Mauger, E.
Berliner, J. Am. Chem. Soc. 1972, 94, 194; c) V. L. Heasley, D. F. Shell-
hamer, L. E. Heasley, D. B. Yaeger, G. E. Heasley, J. Org. Chem. 1980, 45,
4649.
C
cases (alkyne+I2 +I ). We also considered the possibility of direct addi-
À
tion of I3 to electron-poor alkyne 13b; however, no TS for such a pro-
cess could be located.
[19] For a related mechanistic proposal involving alkenes, see: P. W. Robert-
son, J. B. Butchers, R. A. Durham, W. B. Healy, J. K. Heyes, J. K. Johannes-
son, D. A. Tait, J. Chem. Soc. 1950, 2191.
[20] No open vinyl radical (19) could be located: Attempts to optimize this
species led instead to the van der Waals complex F. A similar result was
obtained with B3LYP-D3: Attempted optimization of the vinylic radical
C
PhC(I)=C( )Ph with B3LYP-D3 led only to the van der Waals complex of
[11] For theoretical studies on electrophilic activation of alkenes and alkynes
by halogens and halonium reagents, see: a) Y. Yamamoto, I. D. Gridnev,
N. T. Patil, T. Jin, Chem. Commun. 2009, 5075; b) T. Okazaki, K. K. Laali, J.
Org. Chem. 2006, 71, 9643; c) S. M. Islam, R. A. Poirier, J. Phys. Chem. A
2007, 111, 13218; d) T. Okazaki, K. K. Laali, J. Org. Chem. 2005, 70, 9139;
e) D. Lenoir, C. Chiappe, Chem. Eur. J. 2003, 9, 1036; f) M. V. Frash, A. C.
Hopkinson, D. K. Bohme, J. Phys. Chem. A 1999, 103, 7872; g) R. Bianchi-
ni, C. Chiappe, G. Lo Moro, D. Lenoir, P. Lemmen, N. Goldberg, Chem.
Eur. J. 1999, 5, 1570; h) P. J. Campos, M. A. Rodríguez, J. Org. Chem.
1996, 61, 8664; i) T. P. Hamilton, H. F. Schaefer, III, J. Am. Chem. Soc.
1991, 113, 7147; j) I. G. Csizmadia, V. Lucchini, G. Modena, Theor. Chim.
Acta 1975, 39, 51.
C
PhCCPh and I .
[21] Homolysis of I2 is computed to have DG=21.8 kcalmolÀ1. The comput-
ed DG for homolysis of the IÀI bond in B is 15.6 kcalmolÀ1. For H, the
corresponding value is 10.5 kcalmolÀ1
.
[22] trans-Diiodoalkene 11 a is 2 kcalmolÀ1 more stable than the corre-
sponding cis isomer (see the Supporting Information).
[23] Compound 30 was initially formed as a single isomer (presumably the
E isomer), which isomerized to an E/Z mixture upon work-up and chro-
matography.
[24] Consistent with this idea, computations predict that exo cyclization in
32 (X=I) is favored by 2.8 kcalmolÀ1 (DDG°) over endo cyclization (see
the Supporting Information).
[12] For a complete listing (tabulation) of reaction conditions employed in
the iodocyclization of 13a–c, including those in other solvents, see the
Supporting Information.
Received: January 29, 2015
[13] R. Hollins, M. P. A. Campos, J. Org. Chem. 1979, 44, 3931.
Published online on June 3, 2015
Chem. Eur. J. 2015, 21, 10191 – 10199
www.chemeurj.org
10199
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim