Computationally Assisted Mechanistic Investigation into Hypervalent Iodine Catalysis: Cyclization of N-Allylbenzamide

Article abstract of DOI:10.1021/acs.joc.9b02623

Previous experimental work identified 2-iodoanisole as the best precatalyst for the oxidative cyclization of N-alkenylamides into 2-oxazolines. Herein, we describe our investigation into the effect on the reaction rate based on the structure of the iodoarene precatalyst. We also reveal the mechanism of the cyclization based on DFT modeling and obtain a clear correlation between observed reaction rates and computationally derived activation energies for different iodoarenes. In addition, the rate-limiting step is shown to be the cyclization of the substrate that is zero order in the concentration of the iodoarene precatalyst. The rate of cyclization is found to correlate with the ease of oxidation of the iodoarene; however, the most easily oxidized iodoarenes generate iodine(III) species that decompose readily. Finally, loss of iodoarene from the cyclized intermediate can proceed by either ligand-coupling or S_N2 displacement (reductive elimination), and this is shown to be substrate-dependent.

Full text of DOI:10.1021/acs.joc.9b02623

Subscriber access provided by University of Massachusetts Amherst Libraries
Article
Computationally Assisted Mechanistic Investigation into
Hypervalent Iodine Catalysis: Cyclization of N-Allylbenzamide
Smaher E. Butt, Mirdyul Das, Jean-Marc Sotiropoulos, and Wesley James Moran
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.9b02623 • Publication Date (Web): 01 Nov 2019
Downloaded from pubs.acs.org on November 2, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Computationally Assisted Mechanistic Investigation
into Hypervalent Iodine Catalysis: Cyclization of N-
Allylbenzamide
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Smaher E. Butt,ǁ Mirdyul Das,ǁ Jean-Marc Sotiropoulos§ and Wesley J. Moran*ǁ
ǁ Department of Chemistry, University of Huddersfield, Queensgate, Huddersfield HD1
3DH, U.K.
§ Université de Pau et des Pays de l’Adour, IPREM (CNRS-UMR 5254), Technopole
Hélioparc, 2 Avenue du Président Pierre Angot, 64053 Pau Cedex 09, France
w.j.moran@hud.ac.uk
O
Ar
I
Ph
N
N
H
DFT & kinetic
study of
mechanism
[O]
Ph
OCOCF3
I
O
Ar
OCOCF3
OCOCF3
G of cyclization correlates with ArI oxidation potential
BUT, most easily formed ArI(OCOCF3)2 decompose rapidly
ACS Paragon Plus Environment
1

The Journal of Organic Chemistry
Page 2 of 43
1
2
3
4
5
6
7
8
9
ABSTRACT: Previous experimental work identified 2-iodoanisole as the best precatalyst
for the oxidative cyclization of N-alkenylamides into 2-oxazolines. Herein, we describe
our investigation into the effect on reaction rate based on the structure of the iodoarene
precatalyst. We also reveal the mechanism of the cyclisation based on DFT modelling
and obtain a clear correlation between observed reaction rates and computationally
derived activation energies for different iodoarenes. In addition, the rate-limiting step is
shown to be the cyclization of the substrate which is zero order in the concentration of the
iodoarene precatalyst. The rate of the cyclization is found to correlate with the ease of
oxidation of the iodoarene, however the most easily oxidized iodoarenes generate
iodine(III) species that decompose readily. Finally, loss of iodoarene from the cyclized
intermediate can proceed by either ligand-coupling or SN2 displacement (reductive
elimination), and this is shown to be substrate dependent.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
INTRODUCTION
ACS Paragon Plus Environment
2

Page 3 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Hypervalent iodine chemistry has continued to receive considerable attention from
synthetic practitioners over the past decades with a wide variety of useful reactivities
revealed.1 Hypervalent iodine compounds can possess reactivity profiles comparable to
some heavy metal and transition metal complexes but may not suffer from disadvantages
such as toxicity, cost or scarcity. In particular, the ability to access reactive 3-iodanes in-
situ from iodoarenes and an oxidant has permitted the discovery of a range of reactions
that are catalytic in iodoarene.2 This is especially important as the preparation and
isolation of 3-iodanes is problematic in many cases and often requires considerable
experimental investigation to identify suitable conditions.3 As such, the in-situ generation
and reaction of 3-iodanes can be advantageous as isolation is avoided and, of course,
allows catalytic applications.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fuchigami reported in 1994 that 4-iodoanisole can be used in as little as 5 mol% to
catalyze the gem-difluorination of dithioketals in an electrochemical cell.4 However, the
first examples of iodoarene catalysis are often ascribed to the two groups of Ochiai and
Kita who simultaneously published their studies in 2005. Ochiai described the
iodobenzene-catalyzed -acetoxylation of ketones employing m-chloroperbenzoic acid
ACS Paragon Plus Environment
3

The Journal of Organic Chemistry
Page 4 of 43
1
2
3
4
5
6
7
8
9
as the stoichiometric oxidant.5 Four other iodoarene precatalysts were investigated in this
study but these were all found to give similar or inferior yields of product (67% with p-
methoxy-, 84% with p-methyl-, 81% with p-chloro-, and 58% with p-nitroiodobenzene);
however, there is no mention of the effect on the rate of the reaction. Kita and co-workers
reported the spirocyclization of phenols using as little as 1 mol% of hypervalent iodine
reagent with m-chloroperbenzoic acid as the re-oxidant.6 Later, Li and co-workers
investigated the syn-diacetoxylation of alkenes catalyzed by six different iodoarenes and
discovered that iodomesitylene provided the highest yields and diastereoselectivities
(Scheme 1a).7 Similarly, Zhdankin demonstrated that 5-iodo-m-xylene was superior to
four other iodoarenes in the conversion of oximes into nitrile oxides.8 Murphy described
an iodoarene-catalyzed intramolecular alkene arylation strategy to access polycyclic
aromatic hydrocarbons which was superior with 4-iodotoluene rather than iodobenzene,
4-iodoanisole or 1-iodo-4-nitrobenzene (Scheme 1b).9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
4

Page 5 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Effect of iodoarene substitution on reaction outcomes
(a) Li, 2012
i. 20 mol% ArI, H2O2
OAc
TfOH, AcOH
OAc
ii. Ac2O, pyridine
ArI: PhI (82%, 12.5:1 dr); 4-ClC6H4I (81%, 5.3:1 dr); 4-
MeOC6H4I (85%, 6.7:1 dr); 2,4,6-Me3C6H2I (84%, 14:1 dr)
(b) Murphy, 2018
20 mol% ArI, m-CPBA
.
BF3 OEt2, PhCl, 110 oC
ArI: PhI (63%); 4-NO2C6H4I (48%); 4-MeOC6H4I (70%);
4-MeC6H4I (79%)
(c) Our work, 2015
20 mol% ArI
Ph
O
1.5 equiv TFA
N
O
Ph
N
H
1 equiv Selectfluor
MeCN, rt
OH
1
then 2 M NaOH
2
ArI: PhI (6%), 2-MeOC6H4I (62%); 3-MeOC6H4I (9%);
4-MeOC6H4I (10%); 3,5-Me2C6H3I (13%)
In this regard, we reported that 2-iodoanisole is a superior precatalyst for the oxidative
cyclization of N-alkenylamides such as 1 into 2-oxazolines 2 compared to several other
ACS Paragon Plus Environment
5

The Journal of Organic Chemistry
Page 6 of 43
1
2
3
4
5
6
7
8
9
iodoarenes (Scheme 1c).10 We also discovered that the cyclization of N-propargyl amides
and -amidoketones performed better with 2-iodoanisole as the precatalyst compared to
iodobenzene.11
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We were interested in investigating the mechanism of our cyclization of N-
alkenylamides using kinetic and computational techniques, especially as relatively few
such studies of hypervalent iodine reactions have been reported to date.12,13 Of crucial
importance to our study, was the determination of the reasons behind the different
reaction outcomes with iodoarene precatalyst variation.
Houk, Xue and co-workers recently published a computational study on the mechanism
and origins of chemo- and stereoselectivities in the chiral iodoarene-catalyzed
difluorination of styrenes.14 This report is a tour de force in rationalizing the reasons for
enantioselectivies in the difluorination reaction but does not consider the initial iodoarene
oxidation. Indeed, the authors state that “the overall activation free energy of 18.9
kcal/mol is not consistent with the 60 h reaction times required experimentally.” Studying
the early part of the reaction mechanism is critical for our study as we are interested in
ACS Paragon Plus Environment
6

Page 7 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
understanding the generation and catalytic properties of the hypervalent iodine species
involved.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
Model Reaction, Initial Rate Data and Proposed Catalytic Cycle. We initiated our study
by investigating the rate of cyclization of N-allylbenzamide 1 using different iodoarene
precatalysts under our standard reaction conditions of 1.5 equivalents of Selectfluor as
oxidant, 1.5 equivalents of trifluoroacetic acid (TFA) in acetonitrile at room temperature,
~20 oC (Scheme 2). We saw a clear and substantial difference in initial rates between
different precatalysts as well as marked variation in final yields of products. With
iodobenzene 4 as precatalyst, the initial rate was just 5.16 x 10-6 s-1 meaning that 50%
conversion of starting material took 52 hours. Employing 2-iodotoluene 5 led to a
somewhat faster reaction; however, the presence of the electron-withdrawing p-nitro
group in 6 dramatically reduced the rate. The electron-donating o-methoxy and p-
methoxy groups in 2-iodoanisole 7 and 4-iodoanisole 8 respectively led to significantly
ACS Paragon Plus Environment
7

The Journal of Organic Chemistry
Page 8 of 43
1
2
3
4
5
6
7
8
9
increased rates. The presence of the second o-methoxy substituents in 9 had a
detrimental effect on the rate of cyclization compared to 7, which was somewhat
countered by the ester in the para-position of iodoarene 10. Most important from a
synthetic standpoint is that the use of 2-iodoanisole led to the highest yield of oxazoline
2 after purification, in accord with our previous observations.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Initial rates for conversion of N-allylbenzamide 1 into oxazoline 3 with
iodoarene precatalysts 4-10, determined by 1H NMR analysis
Ph
20 mol% ArI
Ph
1.5 equiv Selectfluor
HN
O
O
O
N
CF3
1.5 equiv TFA
MeCN, rt
O
1
3
I
ArI
Me
I
I
I
Me
OMe
NO2
4
5
6
7
k = 5.16 x 10-6 s-1
k = 7.73 x 10-6 s-1
k = 1.01 x 10-6 s-1
k = 5.78 x 10-5 s-1
I
I
MeO
OMe
I
MeO
OMe
CO2Me
10
k = 5.87 x 10-5 s-1
OMe
8
9
k = 7.00 x 10-5 s-1
k = 5.66 x 10-5 s-1
ACS Paragon Plus Environment
8

Page 9 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
In order to determine the order of the reaction with respect to the iodoarene
concentration, we carried out the Variable Time Normalization Analysis (VTNA), which
involved changing the initial quantity of 2-iodoanisole 7 and monitoring the rate of
reaction by NMR analysis.15 The profiles of reactions performed with different catalyst
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
n
loadings are plotted on a normalized time scale, t[cat]T , against the concentration of the
substrate 1. The best fit of the data led to the conclusion that the reaction is zero order
in 2-iodoanisole 7 (Scheme 3). This unusual finding is in accord with our previous
experimental findings during reaction optimization but it does not help to rationalize the
relatively high catalyst loadings required for acceptable yields of products. A reaction
which is zero order in catalyst is one where the catalyst is fully saturated in substrate(s)
and this is more common in photochemical reactions,16 or heterogeneous catalysis.17 In
our case, this finding can be rationalized by consideration of the NMR spectra obtained
during the kinetic experiments, as only the iodoarenes were observed in their initial
concentrations. That is, no 3-iodanes were observed by NMR analysis during the
course of the reactions. This suggested that the concentration of 3-iodane in the
reaction mixture was very low and upon formation by oxidation of an iodoarene, the 3-
ACS Paragon Plus Environment
9

The Journal of Organic Chemistry
Page 10 of 43
1
2
3
4
5
6
7
8
9
iodane initiated the reaction sequence. In addition, 3-iodanes tend to have low
solubility (especially compared to the precursor iodoarenes), therefore the reaction
could be occurring in a heterogeneous fashion; indeed, precipitation occurs during the
course of these reactions.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 3. Determination of reaction order with respect to concentration of 2-
iodoanisole 7
Ph
7
x mol% 2-iodoanisole
Ph
1.5 equiv Selectfluor
HN
O
O
O
N
CF3
1.5 equiv TFA
MeCN, rt
O
1
3
The mechanism proposed in the original study involved oxidation of the iodoarene
precatalyst to the iodine(III) species, which coordinated to the alkene of N-
ACS Paragon Plus Environment
10

Page 11 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
allylbenzamide 1 and activated it to intramolecular attack by the amide oxygen forming
intermediate 11 (Scheme 4). Loss of a proton and displacement of the iodonium by
trifluoroacetate formed the final product 3. It was found that some of the trifluoroacetate
esters 3 were unstable to chromatographic purification, therefore saponification to the
alcohol 2 was performed during workup. It was anticipated that trifluoroacetate would be
bound to the iodine(III) center during the course of the reaction, but fluoride was also
present in solution. Similar reaction mechanisms have been proposed for other
cyclizations of alkene substrates with hypervalent iodine compounds.18
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4. Proposed mechanism for the iodoarene-catalyzed cyclization of N-
allylbenzamides 1
ACS Paragon Plus Environment
11

The Journal of Organic Chemistry
Page 12 of 43
1
2
3
4
5
L = F or
CF3CO2
O
L
I
R
N
H
Ar
6
7
1
Selectfluor
TFA
L
8
9
R
N
O
L
OMe
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ar
I
H
I
11
TFA
R
R
2 M NaOH
O
O
N
O
N
CF3
OH
O
2
3
, unstable
At this point we decided to study the reaction mechanism computationally to help
rationalize the differences in rates observed with different iodoarenes precatalysts with
the intention to use the knowledge gained to identify superior precatalysts.
Mechanism of Iodoarene-Catalyzed Cyclization of N-allylbenzamide 1. We began our
modelling of the reaction mechanism by ignoring the initial oxidation of the iodoarene,
like other published studies, and started with bis(trifluoroacetoxy)iodobenzene 12, which
should be formed under the reaction conditions (Figure 1). Loss of a trifluoroacetate
ligand from 12 to form 13 followed by formation of activated alkene complex 14, i.e. a
dissociative mechanism, was found to be possible but of high energy. In contrast, an
ACS Paragon Plus Environment
12

Page 13 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
associative mechanism whereby a trifluoroacetate ligand formed a hydrogen-bond with
the incoming amide and underwent a concerted rotation-coordination process was of
lower energy, i.e. formation of intermediate 15a via TS[14-15a]. However, the barrier
was further reduced to 21.3 kcal mol-1 when an extra molecule of trifluoroacetic acid
was involved in the alkene activation i.e., TS[14-15b], leading to the intermediate 15b.
Cyclization of 15b (and 15a) proceeded through TS[15-16], which was higher in energy
by 4.7 kcal mol-1, to form iodonium 16. Loss of a proton formed oxazoline 17 which was
then converted to the final product 3 with loss of iodobenzene 4. Two possible
mechanisms were envisaged here: ligand coupling and SN2 displacement, the latter of
which can also be termed reductive elimination. Modelling revealed that the SN2
process would proceed through TS[17-3]S which is 5 kcal mol-1 lower in energy than the
corresponding ligand coupling transition state TS[17-3]LC. Attempts to model the
reaction mechanism with 3-iodanes bearing fluoride ligands were unsuccessful.19
Attempts to find mechanistic pathways involving cis-addition to the alkene were also not
fruitful.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
13

The Journal of Organic Chemistry
Page 14 of 43
1
2
3
4
5
6
7
8
TS[14-15a]
TS[15-16]
Ph
Ph
CF3
HN
O
O
O
HN
O
H
TFA
Ph
TS[17-3]LC
I
TFA
I
Ph
N
Ph
+26.9
Ph
O
I
O
Ph
14
TFA
N
+25.4
+25.1
1
O
9
Ph
I
TFA
O
CF3
+22.4
TFA
I
Ph
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
+21.3
Ligand Coupling
+21.1
13
TFA
TFA
CF3
15a
+18.4
+20.4
TFA
15b
O
O
TFA
H
N
+13.2
I
Ph
Ph
O
TS[14-15b]
TS[17-3]S
I-Ph
CF3
H
O
O
O
CF3
O
F3C
H
Ph
TFA
Ph
O
O
O
H
N
F3C
I
Ph
N
O
O
TFA
Ph
TFA
H
N
O
I
Ph
TFA
Ph
Ph
HN
O
I
Ph
O
TFA
I
TFA
TFA
SN
2
1
Ph
TFA
O
N
TFA
-0.4
17
+1.1
16
0.0
TFA
I
TFA
TFA
-38.9
Ph
Ph
12
3
N
O
Ph
I
TFA
Figure 1. Calculated potential energy profile for the PhI(O2CCF3)2 12-mediated cyclization
of N-allylbenzamide 1. Free energies are reported in kcal mol-1.
The optimized geometries of all structures in the calculated mechanism conformed to
our expectations with all bond lengths and angles being as anticipated (Figure 2).
Importantly, the whole mechanistic pathway has been connected by IRC (intrinsic
reaction coordinate) calculations (see SI for details).
ACS Paragon Plus Environment
14

Page 15 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
1.9848
3.08
47
2.67
48
34
2.65
41
20
34
41
20
34
41
2.12
3.55
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.10
20
2.12
3.53
TS[14-15a]
15a
TS[14-15b]
O48-I41-O34 = 136.4°
C20-I41-O34 = 153.6°
O48-I41-O34 = 114.5°
C20-I41-O34 = 165.3°
O47-I41-O34 = 137.8°
C20-I41-O34 = 153.5°
47
1.52
3.00
34
41
1.37
2.76
20
2.13
2.23
2.51
2.14
15b
TS[15-16]
17
O47-I41-O34 = 118.8°
C20-I41-O34 = 173.6°
2.54
3.08
2.33
2.68
3.07
3.82
TS[17-3]LC
TS[17-3]S
ACS Paragon Plus Environment
15

The Journal of Organic Chemistry
Page 16 of 43
1
2
3
4
5
6
7
8
9
Figure 2. Calculated geometries of intermediates and transition structures of the
PhI(O2CCF3)2-mediated cyclization of N-allylbenzamide 1. Some hydrogen atoms are
omitted for clarity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Comparing and Contrasting Different Iodoarene Precatalysts. The rate of cyclization of
N-allylbenzamide catalyzed by iodobenzene is slow at room temperature and the
calculated value of 25.1 kcal mol-1 for TS[15-16] is high for a room temperature process.
At this point, we decided to model the transition state of the rate-limiting step with all of
the iodoarenes studied in our initial kinetic work (assuming no change in mechanism or
rate-limiting step).
The modelling of the TS[15-16] transition state for each iodoarene proceeded as
expected, however the presence of a substituent in the 2-position led to two
diastereomeric transition states. For example, with 2-iodoanisole the two diastereomeric
transition states 18a and 18b had an energy difference of 1.3 kcal mol-1 (Figure 3).
ACS Paragon Plus Environment
16

Page 17 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
1.50
9
1.37
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.28
2.39
2.78
2.17
18a
G = +23.4 kcal mol-1
1.51
1.37
2.28
2.79
2.15
2.35
18b
G = +24.7 kcal mol-1
Figure 3. Calculated geometries of two diastereomeric transition states 18a and 18b
and the subsequent cyclized intermediates derived from 2-iodoanisole.
Comparing the experimentally determined initial rates with the computationally derived
free energy values for the rate-limiting step showed a good match for all of the
iodoarenes studied (Table 1). The data suggested that installing an electron-
withdrawing group para to the iodine led to a raising of the activation energy and a
ACS Paragon Plus Environment
17

The Journal of Organic Chemistry
Page 18 of 43
1
2
3
4
5
6
7
8
9
lowering of the rate (entry 3) whereas an electron-donating group in the ortho or para
position lowered the activation energy and increased the rate (entries 4 and 5).
Interestingly, having two ortho-methoxy groups did not lead to an additive effect and the
activation energy was similar to having a single methoxy group in the ortho or para
position (entry 6 versus entries 4 and 5). The only outlier was iodoarene 10 which
performed worse than anticipated from its calculated value (entry 7).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Comparison of experimental rate data with calculated free energies of rate-
limiting step with different iodoarenes
entry ArI
k / s-1
Gǂcalc
/
kcal mol-1
1
2
3
4
iodobenzene 4
5.16 x 10- 25.1
6
2-iodotoluene 5
7.73 x 10- 25.9
6
2-iodo-5-
1.01 x 10- 26.0
6
nitrotoluene 6
2-iodoanisole 7
5.78 x 10- 23.4
5
ACS Paragon Plus Environment
18

Page 19 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
5
6
4-iodoanisole 8
7.00 x 10- 23.2
5
2-iodo-1,3-
5.66 x 10- 23.5
5
dimethoxybenzene
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
7
methyl 4-iodo-3,5- 5.87 x 10- 21.9
5
dimethoxybenzoate
10
Oxidation of Iodoarenes. Computing transition state energies is computationally
demanding, therefore we sought a simple, fast method to predict iodoarene structures
with lower TS[15-16] transition state energies than 2-iodoanisole. The oxidation of the
iodoarene to the 3-iodane had been ignored up to this point in our computations as it
was presumed to be complex with several discrete mechanistic steps. However, our
hypothesis was that more easily oxidized iodoarenes correlated with lower transition state
energies and faster reaction rates. Accordingly, the oxidation potential was computed for
our studied iodoarenes using the method described by Nicewicz and coworkers.20 These
values were compared with the experimental rate values and a clear correlation was
ACS Paragon Plus Environment
19

The Journal of Organic Chemistry
Page 20 of 43
1
2
3
4
5
6
7
8
9
observed, confirming our hypothesis (Figure 4). In addition, the experimental rate values
were shown to be proportional to the corresponding calculated HOMO energies of the
iodoarenes.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
I
Me
I
6
MeO
OMe
NO2
I
CO2Me
5
4
OMe
I
I
Me
10
I
7
8
I
9
MeO
OMe
OMe
9
8
5
4
7
10
6
Figure 4. Correlation between calculated oxidation potentials of iodoarenes and
experimental rate data derived free energies of TS[15-16]
With this correlation confirmed, the identities of iodoarenes with lower oxidation
potentials were sought as these were envisaged to lead to superior catalysts. 1-Iodo-2,4-
dimethoxybenzene 19 and 2-iodo-1,3,5-trimethoxybenzene 20 were quickly identified as
being more easily oxidized and their oxidation potentials were calculated to be 1.40 V and
1.34 V respectively. Unfortunately, kinetic experiments with these precatalysts in our
ACS Paragon Plus Environment
20

Page 21 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
cyclization reaction indicated that the reactions were somewhat slower than when using
2-iodoanisole 7 as the precatalyst and, crucially, the yields of product were only 20-30%.
Furthermore, the iodoarenes 19 and 20 disappeared during the course of the reactions.
In order to probe the oxidation of the iodoarenes, the rate of oxidation by Selectfluor
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
was monitored by H NMR analysis (Scheme 5). Iodobenzene 4, 2-iodotoluene 5 and
ester 10 required about 24 hours to reach full conversion to the 3-iodane (entries 1, 2
and 3). 2-Iodoanisole 7 was fully oxidized in about 14 hours (entry 4) and 4-iodoanisole
8 was slightly quicker (entry 5). However, 1-iodo-2,4-dimethoxybenzene 19 and 2-iodo-
1,3,5-trimethoxybenzene 20 were completely consumed by the oxidant but no products
were evident by NMR analysis (entries 6 and 7). It is known that 1,3,5-trimethoxybenzene
decomposes under electrolysis in aqueous sulfuric acid to generate carboxylic acids,
carbon dioxide and water.21 It appears that iodoarenes 19 and 20 undergo a similar
decomposition process under our reaction conditions on a slightly faster time scale to the
desired cyclization which accounts for the low chemical yields of 2-oxazoline obtained.
Fujita and co-workers discovered that 2-iodo-1,3-dimethoxybenzene 9 can combine with
its oxidized form, i.e. 2-(diacetoxyiodo)-1,3-dimethoxybenzene, under acidic conditions to
ACS Paragon Plus Environment
21

The Journal of Organic Chemistry
Page 22 of 43
1
2
3
4
5
generate a diaryliodonium salt; this side-reaction was proposed to be the cause of the
6
7
8
relatively low yield of product using 9 as a pre-catalyst in an oxylactonization reaction.22
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 5. Monitoring of the oxidation of iodoarenes by 1H NMR
4 equiv Selectfluor
TFA
5 equiv TFA
Ar
I
Ar
I
CD3CN, rt
TFA
E1/2 vs
entry
1
ArI
time/h conversion/%
SCE/V
2.21
I
4
5
24
24
90
I
2
3
2.10
1.88
100
Me
OMe
I
10
24
77
MeO2C
OMe
I
4
5
6
7
8
1.76
1.63
1.40
14
11
24
100
93
OMe
I
MeO
MeO
I
19
decomp
OMe
OMe
I
7
20
1.34
24
decomp
MeO
OMe
Minor modifications to 2-iodoanisole 7 (still our optimal precatalyst) could be
investigated to identify a slightly more efficient catalytic system, however these analogs
would probably not be commercially available and, so, would be of little impact. 1-Iodo-
2,4-dimethoxybenzene 19 could be a highly efficient precatalyst in other hypervalent
ACS Paragon Plus Environment
22

Page 23 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
iodine mediated processes that occur on a faster timescale than its decomposition;
however, not being able to recover the iodoarene is a disadvantage.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Interestingly, the experimentally derived initial rate of cyclization catalyzed by iodoarene
10 matches well with the calculated oxidation potential and HOMO energy but not with
the calculated activation energy value (vide supra). This suggests that there was a
change in the rate determining step with this iodoarene from the cyclization event to the
iodoarene oxidation. Indeed, the rate of oxidation of 10 was comparable with that of
iodobenzene 4.
Displacement of Iodoarene by Ligand Coupling or SN2 Displacement. In our original
report, cyclization of cis-alkene 21 was reported to lead to formation of product 2 as only
one diastereomer; with the relative stereochemistry being assigned by NMR analysis.9
Applying the calculated mechanism for N-allylamide 1 to the cyclization of 21 suggested
that the all-syn diastereomer 23 would be formed rather than 22. It was suspected that a
change in mechanism was responsible for the observed stereochemistry of 22 due to
ligand coupling being favored over SN2 displacement. This SN2 process was more
ACS Paragon Plus Environment
23

The Journal of Organic Chemistry
Page 24 of 43
1
2
3
4
5
6
7
8
9
difficult in this case because of the increased sterics at the secondary carbon atom
bonded to the iodane compared to the primary carbon in 17. In order to confirm this
hypothesis, the cyclization of 24 was calculated using 2-iodoanisole 7 as the pre-
catalyst (Scheme 6). To reduce the computational resource required, the methyl amide
24 was modelled rather than the 4-methoxyphenyl amide 21 as this was not envisaged
to impact on the stereochemistry determining step.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 6. Issues of diastereoselectivity
ACS Paragon Plus Environment
24

Page 25 of 43
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
R
R
20 mol% 2-iodoanisole
1 equiv Selectfluor
1.5 equiv TFA
R
O
O
O
N
N
HN
OH
OH
MeCN, rt
then 2 M NaOH
R = 4-MeOC6H4
O
O
O
O
O
O
21
22
23
observed
not observed
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
TS[27-29]
Me
CF3
O
O
O
N
MeO2C
SN2
I
O
Me
HN
O
O
O
TS[27-28]
CF3
Me
N
O
O
O
+18.9
24
TS[27-29]
O
O
MeO2C
CF3
Me
HN
+15.0
I
TFA
O
TS[27-28]
ligand
coupling
O
I
TFA
O
MeO2C
I-Ar
TFA
24
+1.3
26
TFA
-0.5
27
0.0
25
Me
N
Me
N
TFA
I
TFA
O
O
O
TFA
O
TFA
I
-34.0
28
MeO2C
Me
MeO2C
-36.7
O
N
29
TFA
MeO2C
ACS Paragon Plus Environment
25

The Journal of Organic Chemistry
Page 26 of 43
1
2
3
4
5
6
7
8
9
The free energies of the cyclized structures 26 and 27 resulting from trans-addition to
the alkene compare favorably with the energies of the corresponding structures 16 and
17 for the cyclization of N-allylamide 1. As previously observed, reaction pathways
proceeding through cis-additions could not be found. Conversion of iodane 27 to the final
product could be achieved through either SN2 displacement or ligand coupling, but, in
contrast to N-allylamide, the latter pathway was favored by 3.9 kcal mol-1 resulting in a
highly diastereoselective reaction. This result corroborated our stereochemical
assignment of the product 22 and confirmed our hypothesis. Note that the all-syn
diastereomer 29 is thermodynamically more stable than 28 but the reaction is under
kinetic control. The optimized geometries are shown in Figure 5.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
26

Page 27 of 43
The Journal of Organic Chemistry
1
2
3
4
Ligand Coupling TS
SN2 displacement TS
CF3
5
6
7
8
O
O
O
N
O
N
O
N
O
vs
I
TFA
O
MeO2C
MeO2C
CF3
MeO2C
I
9
O
I
O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
O
O
CF3
TS[27-28]
TS[27-29]
27
48
38
2.44
55
40
13
2.27
15
2.79
40
13
2.53
18
17
13
19
2.55
3.44
15
3.11
16
2.79
16
18
14
19
Angle C38-C55-I13 = 99.78°
Angle O48-C55-I13 = 154.4°
Angle C40-C13-I15 = 107.8°
Dihedral angle C40-C13-I15-C19 = -179.9°
Angle C40-C13-I15 = 110.3°
Dihedral angle C40-C13-I15-C19 = -98.6°
Dihedral angle C38-C55-I13-C17 = 6.93°
Figure 5. Optimized geometries of intermediate 27 and the ligand coupling and SN2
transition states.
CONCLUSION
An in-depth analysis of the iodoarene-catalyzed cyclization of N-allylbenzamide 1 has
been conducted using a combination of computational and kinetic experiments. A
mechanism has been delineated which accounts for the experimental differences in rate
observed with different iodoarene precatalysts and, importantly, the calculated free
energy values correlate with our experimental results. The cyclization event has been
ACS Paragon Plus Environment
27

The Journal of Organic Chemistry
Page 28 of 43
1
2
3
4
5
6
7
8
9
shown to be rate-limiting and more electron-rich iodoarenes lead to lower Gǂ values for
this step. The reaction has been found to be zero order in the precatalyst 2-iodoanisole 7
but the reaction model breaks down with more electron-rich iodoarenes due to
competitive decomposition. The optimal choice of precatalyst for related processes is
likely to be affected by the rate of the desired reaction; i.e. if the rate of reaction is faster
than decomposition then a more electron-rich precatalyst such as 1-iodo-2,4-
dimethoxybenzene 19 could be the best option.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Loss of iodoarene from the reaction intermediate can occur by either SN2 displacement
by a nucleophile (reductive elimination) or by ligand coupling at the iodine(III) center. Both
pathways are possible but SN2 displacement is favored at primary carbons whereas
ligand coupling is preferred with secondary carbons.
EXPERIMENTAL SECTION
1
General. H NMR spectra were recorded in ppm from tetramethylsilane with the solvent
resonance as the internal standard. All purchased reagents were used as received without further
purification. Compounds 1,23 1024 and 2025 were prepared by the literature procedures and the
characterization data matched the reported values.
ACS Paragon Plus Environment
28

Page 29 of 43
The Journal of Organic Chemistry
1
2
3
4
General Procedures for Kinetic Experiments
5
6
7
8
9
Rate Data: Amide 1 (1 equiv) was added to a borosilicate NMR tube along with the iodoarene
precatalyst (0.2 equiv), Selectfluor (1.5 equiv), TFA (1.5 equiv) and CD3CN (1 mL) to generate a
0.3 M solution of 1. The reaction mixture was monitored and the conversion of amide 1 was
recorded by 1H NMR spectroscopy (400 MHz, 298 K) for up to 24 hours at 25 ˚C. Assuming the
reaction is first order in 1, the rate constant was determined by linearization of the decay curve
followed by linear fitting for only the initial part of the reaction. Results are summarized in the
supporting information table S1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Effect of Catalyst Loading: The order of the cyclization reaction with respect to 2-iodoanisole
was determined by the Variable Time Normalization Analysis (VTNA) as described by Burés
(Scheme 3).14 Amide 1 (1 equiv) was added to a borosilicate NMR tube along with 2-iodoanisole
(10 & 30 mol%), Selectfluor (1.5 equiv), TFA (1.5 equiv) and CD3CN (1 mL). The reaction
mixture was monitored by 1H NMR spectroscopy (400 MHz, 298 K) to study the effect of different
catalyst loadings. The concentration of 1 was recorded for a period of 40 hours and plotted against
n
a normalized time scale t[cat]T ; calculated by multiplying each time point by the total
concentration of catalyst used in each experiment raised to an arbitrary power. The power value
was adjusted to be zero as that produced the overlay of the curves (Scheme 3). The numerical data,
along with the concentration of 1 at specific times, are listed in the supporting information.
Computational Methods: Quantum chemical calculations were performed using
Gaussian 09,26 and GaussView was used for molecular modelling.27 The geometry
optimizations were performed in the gas phase under standard conditions using density
ACS Paragon Plus Environment
29