Catalytic and computational studies of N-heterocyclic carbene or phosphine-containing copper(I) complexes for the synthesis of 5-iodo-1,2,3-triazoles

Article abstract of DOI:10.1021/cs500326e

Two complementary catalytic systems are reported for the 1,3-dipolar cycloaddition of azides and iodoalkynes. These are based on two commercially available/readily available copper complexes, [CuCl(IPr)] or [CuI(PPh ₃)₃], which are active at low metal loadings (PPh ₃ system) or in the absence of any other additive (IPr system). These systems were used for the first reported mechanistic studies on this particular reaction. An experimental/computational-DFT approach allowed to establish that (1) some iodoalkynes might be prone to dehalogenation under copper catalysis conditions and, more importantly, (2) two distinct mechanistic pathways are likely to be competitive with these catalysts, either through a copper(III) metallacycle or via direct-activation of the starting iodoalkyne.

Full text of DOI:10.1021/cs500326e

Subscriber access provided by UNIV OF CONNECTICUT
Article
Catalytic and Computational Studies of N-Heterocyclic
Carbene or Phosphine-Containing Copper(I)
Complexes for the Synthesis of 5-Iodo-1,2,3-Triazoles
Steven Lal, Henry S Rzepa, and Silvia Diez-Gonzalez
ACS Catal., Just Accepted Manuscript • Publication Date (Web): 29 May 2014
Downloaded from http://pubs.acs.org on May 30, 2014
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 free 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 accessible to all
readers and 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.
ACS Catalysis 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 46
ACS Catalysis
1
2
3
4
5
6
7
8
9
Catalytic and Computational Studies of N-
Heterocyclic Carbene or Phosphine-Containing
Copper(I) Complexes for the Synthesis of 5-Iodo-
1,2,3-Triazoles
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
Steven Lal, Henry S. Rzepa, and Silvia Díez-González*
Department of Chemistry, Imperial College London, Exhibition Road, South Kensington, SW7
2AZ London (UK)
KEYWORDS: Azide – Iodotriazole – Copper – Cycloaddition – Click Chemistry
Abstract: Two complementary catalytic systems are reported for the 1,3-dipolar cycloaddition
of azides and iodoalkynes. These are based on two commercially available/readily available
copper complexes, [CuCl(IPr)] or [CuI(PPh3)3], which are active at low metal loadings (PPh3
system) or in the absence of any other additive (IPr system). These systems were used for the
first reported mechanistic studies on this particular reaction. An experimental/computational-
DFT approach allowed to establish that 1) some iodoalkynes might be prone to dehalogenation
under copper catalysis conditions and, more importantly, 2) that two distinct mechanistic
pathways are likely to be competitive with these catalysts: either through a copper(III)
metallacycle or via direct π-activation of the starting iodoalkyne.
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 46
1
2
3
4
5
6
Introduction
7
8
9
The versatility and value of heterocycles for a wide range of applications feeds the continuous
synthetic drive for novel and better methodologies aimed at their preparation. Halogenated
heterocycles are of particular interest due to their easier derivatisation, notably via cross-
coupling reactions. 1,3-Dipolar cycloadditions, Huisgen reactions,1 are one of the most general
approaches to five-membered heterocycles, and the preparation of 1,2,3-triazoles from organic
azides and terminal alkynes in particular (CuAAC) has witnessed a true explosion since the
discovery of copper(I) species as outstanding and selective catalysts for this reaction.2
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
Despite the important applications that CuAAC has found in biochemistry, materials or
medicinal science,3 it remains largely limited to terminal alkynes. This is not surprising if we
take into account the well accepted intermediacy of copper-acetylide derivatives in the formation
of 1,2,3-triazoles.4 Hence, besides two isolated examples,5 only halogenated alkynes have been
shown to be active towards cycloaddition reactions under copper catalysis to date. The first
reported example was based on a combination of copper(I) and copper(II) salts for the reaction
of azides and bromoalkynes in THF.6 However, it was not until 2009 that this reaction began
attracting the interest of the chemical community when Fokin and co-workers reported the use of
polytriazole as ligands in the preparation of 5-iodotriazoles.7
In great contrast with the cycloaddition of terminal alkynes, where the ligand-less combination
CuSO4/Na ascorbate is very popular, the presence of coordinating ligands has proved essential
for any iodotriazole to be formed. Hence, simple amines such as triethylamine or DIPEA
(diisopropylethylamine) were screened by the Fokin group, but it was a tris-triazole ligand
(TTTA) and CuI (5 mol % each) that provided the best yields (Scheme 1). An
2
ACS Paragon Plus Environment

Page 3 of 46
ACS Catalysis
1
2
3
4
5
6
7
8
9
iminophosphorane-copper(I) complex led to similar results with a lower copper loading of 2 mol
% in aqueous media.8 Here, besides the ancillary ligands, an excess of lutidine with respect to the
catalyst was required for the reaction to proceed (Scheme 1). Remarkably, in both cases,
filtration or extraction techniques yielded directly analytically pure products.
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
R1
N
N
N
Conditions
R1 N3
+
R2
I
I
R2
CuI mol
(5
TTTA mol
mol
(10
Water, air,
92–98%
%)
%)
RT
[Cu] (2
2,6-lutidine
%)
mol
(5
%)
in
in
RT
THF, air,
73–99%
2
N
=
[Cu]
=
P
N
TTTA
N
N
P
S
(EtO)2
t-Bu
Cu
Cu
N
S
N
P(OEt)2
N
N
N
N
N
P
2
PF6
3
N
n
Scheme 1. Reported catalysts for the cycloaddition of azides and iodoalkynes.
A third catalytic system, phenylethynylcopper(I) in combination with 2 equivalents of NEt3,
has also been reported for two specific examples.9,10 The small number of catalysts known for
the synthesis of iodotriazoles is in stark contrast with the impressive array of those reported for
the cycloaddition reactions of terminal alkynes. This is a clear indication of the bigger catalytic
challenge that the cycloaddition of iodoalkynes represents.
Tris-triazoles, and TTTA in particular, remain the most popular ligands for this reaction,11 and
the only ones that have found application in larger systems, such as the functionalization of
polymers12 and preparation of phthalocyanines.13 Despite its efficiency, important drawbacks
remain for the general use of this family of ligands in this transformation. First of all, on the
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 46
1
2
contrary to copper salts, tris-triazole compounds are extremely expensive.14 Alternatively, the
preparation of TTTA requires tris-propargylamine, and t-butyl azide. The latter compound is
particularly hazardous due to its low molecular weight, which greatly increases the risk of
explosion of organic azides.15 Finally, tris-triazoles are not the best candidates for undertaking
mechanistic studies since Finn and co-workers showed that, due to their relatively weak donor
properties, multiple coordination modes to copper centres are possible depending on the reaction
media and cycloaddition partners.16
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
In consequence, the development of safe and more convenient catalysts is fundamental for
iodotriazoles to have a real impact as it has been the case for the related 5-H-1,2,3-triazoles.17
Capitalizing on our experience on copper-mediated cycloaddition reactions,18 we decided to
explore the activity of well-established copper(I) catalysts bearing either phosphine or N-
heterocyclic carbene ancillary ligands. Herein, we report the catalytic activity of such complexes
in the cycloaddition reaction of organic azides and iodoalkynes, as well as, the mechanistic
studies carried out for this transformation.
Results and Discussion
A) Catalytic Studies
Copper(I) complexes bearing N-heterocyclic carbene ligands: Within the last 20 years, N-
heterocyclic carbenes (NHCs) have gained a prominent place in the chemist’s toolbox thanks to
the outstanding affinities to metal centres in all oxidation states, which have naturally led to their
widespread application in organic synthesis.19 Copper–NHC complexes are not an exception and
they are recognised to be stable species while remaining effective catalysts for a number of
transformations.20
4
ACS Paragon Plus Environment

Page 5 of 46
ACS Catalysis
1
2
Neutral [CuX(NHC)] complexes (X = Cl, Br or I)18b were first tested in the model reaction of
benzyl azide and iodoethynylbenzene. For the readers’ convenience, the NHC ligands used in
this study and their acronyms are shown in Figure 1.
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
N
N
N
N
IPr
SIPr
N
N
N
N
IMes
SIMes
N
N
IAd
Figure 1. N-Heterocyclic carbene ligands
For [CuX(NHC)] complexes bearing an IAd ligand, the trend in conversion was X = I > Br >
Cl (Chart 1). IMes proved to be a better ligand and, contrary to the IAd complexes, a reverse
reactivity trend was observed, with [CuCl(IMes)] slightly outperforming [CuBr(IMes)].
Saturated NHC ligands SIMes and SIPr were also screened with SIMes leading to a higher
conversion of 76% over 24 h. The NHC ligands in Chart 1 are ordered by increasing percent
buried volume,21 and the IPr ligand appears to perfectly meet the steric requirements for this
particular reaction, as both more or less bulky NHC ligands led to lower conversions into 1a.
Overall, the commercially available [CuCl(IPr)] clearly outperformed all of the screened
complexes, including its bromide and iodide analogues (Chart 1). For comparison purposes, the
22
related cationic complex [Cu(IPr)2]PF6 was also tested in the model reaction, but only 19%
conversion was observed under otherwise identical reaction conditions.
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 46
1
2
3
4
5
6
7
8
[CuX(NHC)]
N
5 mol
(
%)
N
N
+
N3
Ph
I
Ph
Ph
THF
(0.25 M)
in
24 h
I
Ph
RT, air,
1a
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
Chart 1. Copper(I)–NHC catalyst screening. 1H NMR conversions are the average of at least
two independent experiments. Bis-IPr = [Cu(IPr)2]PF6
In all cases, the expected 5-iodotriazole 1a was formed exclusively, without any trace of the
corresponding 5-H-triazole. It is important to note that no correlation could be found when
comparing the activity of each of these [CuX(NHC)] complexes in this reaction and in the
related terminal alkyne–azide cycloaddition.18b This might be indicative of significantly
dissimilar mechanistic pathways in both transformations.
The most promising catalyst, [CuCl(IPr)] was then tested in different solvents. No conversion
was observed in DMF, DMSO, 1,4-dioxane, toluene or MeOH and only very sluggish reactions
took place in MeCN (5%), acetone (5%) or water (17%). Remarkably, the reaction proceeded
equally well in THF or in the absence of solvent (83 and 88% conversion, respectively).
Surprisingly, in more concentrated reaction mixtures (0.5 or 1 M in THF, compared to 0.25 M)
only very poor conversions into triazole 1a were observed. On the other hand, decreasing the
concentration to 0.1 M, also inhibited the cycloaddition. Thus, either neat conditions or 0.25 M
6
ACS Paragon Plus Environment

Page 7 of 46
ACS Catalysis
1
2
3
4
5
6
in THF are optimal for the cycloaddition reactions of azides and iodoalkynes in the presence of
[CuCl(IPr)].
7
8
9
Considering the crucial importance that nitrogen-based ligand/additives have had in the
previously reported systems for this reaction (even with pre-formed catalysts9), we next tested a
number of them in an attempt to further increase the conversion in the model reaction (Table 1).
Triethylamine, DIPEA (DIPEA = diisopropylethylamine) and 2,6-lutidine failed to improve the
conversion and similar results were observed in all cases (Table 1, entries 2–4). Interestingly,
lutidine actually showed an inhibitory effect when a lower copper loading was used (Table 1,
entry 5). On the other hand, 1,10-phenanthroline completely inhibited the cycloaddition reaction
(Table 1, entries 6 and 7). It is important to note that phenanthroline has been shown to enhance
the reactivity of NHC–copper complexes in the cycloaddition reaction of organic azides and
terminal alkynes.23
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. Effect of nitrogen-containing additives with [CuCl(IPr)]
[CuCl(IPr)]
5 mol
(
%)
N
Additive mol
(5
%)
N
N
+
N3
Ph
I
Ph
Ph
in air
24 h
Neat,
RT,
I
Ph
1a
Entry
Additive
None
Conv (%)a
1
2
3
4
5
6
88
89
87
81
28b
0
NEt3
DIPEA
2,6-lutidine
2,6-lutidine
1,10-phenanthroline
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 46
1
2
0c
3
4
7
1,10-phenanthroline
5
6
7
b
a 1H NMR conversions are the average of at least two independent experiments. 3 mol %
c
[CuCl(IPr)] used; 61% conversion was observed in the absence of lutidine. Reaction mixture
0.25 M in THF.
8
9
In our case, a bright green colour was observed when 1,10-phenanthroline was added to the
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
reaction mixture, suggesting oxidation to a copper(II) species, and therefore, to a catalytic dead-
end. Complexation of the phenanthroline ligand could also be envisaged, however the reported
[CuCl(SIMes)(phen)] complex is a red compound.24 No green color was observed when THF
was used as solvent, but no triazole product was formed nevertheless.
Next, we attempted to decrease the copper loading, but lower conversions were consistently
observed,25 and a rise in the reaction temperature (from room temperature to 40ºC) led to the
decomposition of the starting iodoalkyne. In consequence, 5 mol % [CuCl(IPr)] was used to
explore the scope of the reaction (Scheme 2). Several iodotriazoles could be prepared in good to
high yields, including examples bearing electron-rich or electron-poor benzylic chains, alcohol
or cyclopropyl groups. On the other hand, bulky substituents and nitrogen-containing functional
groups led to low conversions (Scheme 2, 1j, 1e, and 1f). The later observation correlates well
with the neutral/negative effect of nitrogen-based additives summarized in Table 1. The use of
THF in these sluggish reactions did not improve the reaction conversions. In some cases,
however, a very slight increase in copper loading (6 mol %) had a dramatic effect in the reaction
conversion (Scheme 2, 1l, and 1m). Gratifyingly, in all cases the prepared triazoles were easily
isolated from the reaction mixture after a simple extraction without the need of purification by
column chromatography. Furthermore, no formation of dehalogenated triazoles was observed in
any of these reactions.
8
ACS Paragon Plus Environment

Page 9 of 46
ACS Catalysis
1
2
3
4
5
6
R1
N
mol
[CuCl(IPr)] (5
in
%)
N
N
R1 N3
R2
I
+
24 h
Neat, air, RT,
I
R2
1
7
8
9
N
N
N
N
N
N
N
N
N
N
N
N
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
I
I
Ph
, 75%
I
I
Ph
Ph
MeO
MeO
N
O2
1a
1b
1c
1d
, 84%
, 68%
, 80%
F
N
N
N
N
N
N
N
N
I
N
N
N
N
Ph
NEt2
I
I
I
F
3C
F
MeO
3C
N
, 32%
,
40%a
, 66%
, 32%
a
a
1e
1f
1h
1g
N
N
N
N
N
N
n
-Hex
Ph
BnO
N
N
N
N
N
N
N
N
N
HO
4
O
I
I
I
I
Ph
, 75%
Ph
, 75%
Ph
Ph
I
Ph
, 24%a
a
a
b
b
1i
1l
1m
, 22% ; 85%
1k
, 40% ; 90%
1j
Scheme 2. [CuCl(IPr)]-catalysed synthesis of 5-iodotriazoles. Isolated yields are the average of
at least two independent experiments. a 1H NMR conversions are the average of at least two
independent experiments. b Isolated yield with 6 mol % [CuCl(IPr)]
An important advantage of this NHC-based catalytic system is the use of a robust and
commercially available copper(I) complex,26 active under very simple reaction conditions and in
the absence of any additive. Moreover, the strength of the copper–NHC bond is of great interest
for the study of this cycloaddition reaction form a mechanistic point of view (see below).
Copper(I) complexes bearing phosphine ligands: Even if other phosphorous ligands have
shown to be effective in azide-alkyne cycloaddition reactions,27 triphenylphosphine has arguably
proved to be the most versatile so far.28 Accordingly, we tested our model reaction with different
PPh3-containing copper catalysts (Table 2). Reactions were run neat since no reaction was
observed in either THF, MeCN, toluene or water.
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 46
1
2
3
4
Mono-phosphine catalysts [CuX(PPh3)] led to a good conversion into 1a (Table 2, entries 1
5
6
1
and 2), however, H NMR analysis of the reaction mixtures showed a singlet at δ 5.59 ppm,
7
8
9
which corresponds to the 5-H-triazole analogue of 1a.29 A singlet at δ 3.07 ppm was also
observed, indicating that iodoethynylbenzene had been partially dehalogenated to
phenylacetylene, which would explain the formation of a 5-H-triazole. Gratifyingly, similar
conversions were obtained with the related tris-phosphine complexes (Table 2, entries 2–4),
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
while no 5-H-triazole was generated even if traces of phenylacetylene were observed in the H
NMR spectra of some crude products. Overall, [CuI(PPh3)3] (Table 2, entry 5) was selected as
the catalyst of choice for efficient and clean reactions.
Table 2. PPh3-containing catalysts screening
n
]
[CuX(PPh3)
Ph
N3
N
5 mol
(
%)
N
N
Ph
+
in air
24 h
Neat,
RT,
Ph
I
I
Ph
1a
Entry
Catalyst
Conv (%)a
1
2
3
4
5
[CuCl(PPh3)]
[CuBr(PPh3)]
[CuCl(PPh3)3]
[CuBr(PPh3)3]
[CuI(PPh3)3]
92b
91b
81
81
90
b
a 1H NMR conversions are the average of at least two independent experiments. 5-10% of
dehalogenated triazole observed by 1H NMR.
Next, several nitrogen-containing additives were tested in combination with [CuI(PPh3)3]
(Table 3). The use of 5 mol % triethylamine or DIPEA had little effect in the reaction
10
ACS Paragon Plus Environment

Page 11 of 46
ACS Catalysis
1
2
3
4
5
6
7
8
9
conversions (Table 3, entries 2 and 3). However, 1,10-phenanthroline significantly lowered the
conversion into iodotriazole 1a (Table 3, entry 4). This is in agreement with our previous
observations with [CuCl(IPr)] as catalyst (see Table 1), but in this case no colour change was
observed in the reaction mixture.
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 3. Effect of nitrogen-containing additives with [CuI(PPh3)3]
]
[CuI(PPh3)3
5 mol
(
%)
Ph
N3
N
Additive mol
(5
%)
N
N
+
Ph
in air
24 h
Neat,
RT,
Ph
I
I
Ph
1a
Entry
Additive
None
Conv (%)a
1
2
3
4
5
90
NEt3
84
DIPEA
87
1,10-phenanthroline
2,6-lutidine
18
>95
a 1H NMR conversions are the average of at least two independent experiments.
Pleasantly, the presence of 2,6-lutidine in the reaction mixture resulted in the total conversion
of the starting materials into triazoles 1a (Table 3, entry 5). The catalyst loading was then
optimised using lutidine, but the results obtained in the absence of any additive are also shown
for comparative purposes (Table 4). When the copper loading was decreased down to 1 mol %,
excellent conversions were still obtained as long as lutidine was present in the reaction mixture.
Otherwise, the conversion progressively decreased from 90% to 15% (Table 4, entries 1–5).
Only a moderate conversion into 1a was obtained with 0.5 mol % [CuI(PPh3)3] and 5 mol % of
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 46
1
2
3
4
5
6
7
8
9
lutidine in 24 h, but the reaction still reached completion after 3 days. Increasing the additive
loading while keeping the copper one unchanged only led to poorer results (Table 4, entries 7–9).
On the other hand, 4 mol % of 2,6-lutidine could be used without any change in the reaction
conversion (Table 4, entry 10) but a further decrease led to a dramatic reduction in reactivity
(Table 4, entries 11–12).
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 4. Optimisation of [CuI(PPh3)3] loading in the presence of lutidine.
[
]
CuI(PPh3)3
X mol
(
%)
mol
Ph
N3
N
Lutidine
N
N
(Y
%)
Ph
+
in air
24 h
Neat,
RT,
Ph
I
I
Ph
1a
[lutidine]
(mol %)
Entry [Cu] (mol %)
Conv (%)a
1
5
5
>95 (90)
88 (86)
>95 (84)
>95 (79)
>95 (15)
40 (0)
8
2
4
5
3
3
5
4
2
5
5
1
5
6
0.5
0.5
1
5
7
10
50
100
4
8
71
9
1
41
10
11
12
1
>95
1
3
69
1
1
10
a 1H NMR conversions are the average of at least two independent experiments, conversions
into brackets were obtained in the absence of 2,6-lutidine.
12
ACS Paragon Plus Environment

Page 13 of 46
ACS Catalysis
1
2
3
4
5
6
Since only 4 mol % of 2,6-lutidine is required in order to maximize the catalyst efficiency, we
hypothesized about the possible formation of a copper-lutidine complex under the reaction
7
8
9
1
conditions. The H and 31P NMR spectra of a stoichiometric mixture of [CuI(PPh3)3] and
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
lutidine only showed the signals corresponding to both starting materials and identical results
were obtained using [CuI(PPh3)] as copper source instead. Also, no interaction between the
additive and either the azide or iodoalkyne were evidenced on NMR. It is important to note that,
to the best of our knowledge, despite the numerous CuX–lutidine complexes known, there are no
reported examples of these bearing phosphine ligands.30 Finally, as the formation of
dehalogenated phenylacetylene was observed in some reactions (vide infra), the intermediacy of
an in situ generated N-iodolutidinium salt was also investigated. Thus, benzyl azide and
phenylacetylene were reacted with one equivalent of N-iodolutidinium salt31 in the presence of 5
mol % [CuI(PPh3)3]. However, after 24 h of stirring at room temperature, only traces of
iodotriazole 1a were detected in the reaction mixture. Importantly, no 5-H triazole was produced
neither, which indicates that the copper catalyst is not compatible with lutidinium salts under
these reaction conditions.
The scope of the optimized catalytic system was finally explored (Scheme 3). In general,
similar or better yields were obtained in triazoles 1 when comparing with the [CuCl(IPr)]-based
catalytic system (see Scheme 2), and in some cases the conversions were at least doubled even if
longer reaction times were required (Scheme 3, 1h and 1k). Furthermore, triazoles bearing an
alkene or an amine function (Scheme 3, 1o and 1p) could be prepared under these conditions,
when only sluggish reactions were obtained with the carbene catalyst. On the other hand,
hindered azides and pyridine substituents still led to poor conversions, if any (Scheme 3, 1j and
1q).
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 46
1
2
3
4
5
6
7
8
9
mol
mol
]
(4
[CuI(PPh3)3 (1
2,6-lutidine
%)
%)
R1
N
N
N
R1 N3
R2
I
+
in
Neat, air, RT,
18 h
I
R2
1
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
N
N
N
N
N
N
N
N
N
N
N
Ph
Ph
HO
4
I
Ph
, 86%
I
I
Ph
I
Ph
MeO
,
74%a,b
,
89%c
1a
1b
1h
1i
, 90%
N
N
N
N
N
n
-Hex
Ph
N
N
N
I
N
N
N
N
I
I
Ph
I
n-Bu
Ph
24%a
Ph
N
N
O2
c
a,b
1k
1m
,
, 80%
1n
, 72%
, >95%
1j
N
N
N
N
N
N
N
N
Ph
Ph
Ph
I
N
Ph
I
NEt2
, 85%a,b
I
, 91%
, 0%a
1q
d
1o
1p
Scheme 3. [CuI(PPh3)3]-catalyzed preparation of 5-iodotriazoles. Isolated yields are the average
of at least two independent experiments. a 1H NMR conversions are the average of at least two
independent experiments. b ≈5% of dehalogenated triazole observed by 1H NMR. c Reaction time
= 48 h. d 2 mol % [CuI(PPh3)3]
The remarkable activity of the phosphine-based complex was sometimes hampered by the
1
formation of small quantities (5–10% according to the H NMR spectra) of the corresponding
dehalogenated triazoles. When no 5-H-triazole was formed, the expected iodotriazoles 1 were
easily isolated after hydrolysis and filtration of the reaction mixture. Overall, [CuI(PPh3)3] is an
efficient and convenient catalyst for this cycloaddition reaction. Not only does this catalytic
system employ, to the best of our knowledge, the lowest copper loading reported for this
transformation, but it also relies on a cheap and readily prepared catalyst.32
14
ACS Paragon Plus Environment

Page 15 of 46
ACS Catalysis
1
2
3
4
5
6
B) Mechanistic Studies
7
8
9
With two optimized catalytic systems based on pre-formed catalysts in hand, we next carried
out combined experimental and computational mechanistic studies on the 1,3-dipolar
cycloaddition of azides and iodoalkynes in order to rationalize the observed catalytic effect and
total regioselectivity towards the formation of 5-iodotriazoles.
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
Formation of dehalogenated triazoles as by-products: The minor formation of 5-H-triazoles in
some of our [CuI(PPh3)3]-mediated reactions is undesired but not unprecedented. Buckley et al.
also observed such by-products when using a polymeric alkynylcopper(I) catalyst on water under
microwave irradiation.33 In their case, up to 39% of non-halogenated triazole was obtained. It
was then suggested that the 5-H-triazole was generated by the cycloaddition of an in situ
generated terminal alkyne upon dehalogenation of the starting iodoalkyne. In order to verify this
hypothesis, 1-iodohexyne was subjected to our optimized conditions, and after 18 h of stirring at
room temperature, it was found that 25% of the starting alkyne had been transformed into the
corresponding terminal alkyne, 1-hexyne (Scheme 4).34 Once such dehalogenation starts, it is
probable that an equilibrium is then established due to a copper-mediated H/I exchange, as
previously reported in the presence of CuI and different amines.35 To verify this hypothesis, a
mixture of the iodoethynylbenzene and 1-hexyne was reacted under identical conditions, and
after 18 h stirring at RT, 50% of 1-hexyne had been converted to 1-iodohexyne. On the other
hand, triazole 1n (see Scheme 3) remained unchanged even at higher catalyst loadings, probing
the stability of the reaction products under the catalytic conditions (Scheme 4).
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 46
1
2
3
4
5
6
mol
]
[CuI(PPh3)3 (1
2,6-lutidine
%)
mol
(4
%)
n
-Bu
I
n-Bu
H
in
18 h
7
neat,
air, RT,
8
9
1H NMR
25%
(
conversion)
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
mol
]
[CuI(PPh3)3 (1
2,6-lutidine
mol
%)
Ph
I
Ph
H
(4
in
%)
+
+
18 h
neat,
air, RT,
n
-Bu
H
n
-Bu
I
1H NMR
50%
(
conversion)
mol
]
[CuI(PPh3)3 (5
%)
2,6-lutidine
N
mol
N
N
(20
%)
Ar
NO
REACTION
in
18 h
neat,
Ar
air, RT,
n
I
-Bu
=
2-Ph
NO
p-
1n
Scheme 4. Control reactions for the formation of dehalogenated triazoles.
Computational procedure: In this work, the ωB97XD density-theory functional was used as it
incorporates dispersion interactions to properly model the globular environment, and it has been
shown to reproduce reaction barriers effectively.36 A Def2-SVP basis set was employed,
including a pseudopotential for iodine.37 This results in ~800 basis functions used in the
calculation, our resource-bound approximate upper limit for a general mechanistic exploration
and for calculation of intrinsic reaction coordinates (IRC). All transition states were
characterized by normal coordinate analysis revealing precisely one imaginary mode
corresponding to the intended reaction. In some cases, this was augmented by an IRC
calculation, which also established the identity and synchronicity of the reaction. All of the
located transition states were found to be stable as a singlet state. Full coordinates for all the
stationary points, together with normal mode animations are available via the Web Enhanced
16
ACS Paragon Plus Environment

Page 17 of 46
ACS Catalysis
1
2
3
4
5
6
7
8
9
Tables 1–5 (see the individual references for those tables for access), which also include links to
full details for each calculation found in a digital repository.38 Transition states were relocated
with inclusion of a self-consistent-reaction-cavity continuum SCRF/CPCM solvation model for
dichloromethane, using smoothed reaction cavities for first and second energy derivatives.39 As
the actual reactions were carried out in the absence of solvent, dichloromethane was chosen as a
model solvent to approximately model the dielectric of the medium. Unless stated otherwise, all
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
energy values are free energies ΔG‡
(kcal/mol) for a standard state of 1 atm (0.041 M),
298
normalized to a relative free energy of 0 kcal/mol for the isolated reactants. This includes
thermal energy correction, including entropy. For a bimolecular reaction, a reduction of -
2.7 kcal/mol to the free energy barrier (not applied in the tables) is required to correct to a
standard state of ~4 M, the conditions under which the experiments were conducted.40,41
In all examined reaction pathways, we used our model reaction of benzyl azide and
iodo(ethynylbenzene) and we considered the formation of the corresponding 5-iodo and 4-
iodotriazoles 1a and 2a, even if the latter was never observed experimentally (Scheme 5). The
formation of both regioisomers is strongly exoenergic by around -63 kcal/mol (Table 5).
Ph
N3
N
Az
N
N
N
[Cu]
N
N
Ph
Ph
and/or
+
Ph
I
Ph
I
I
Ph
Alk
1a
2a
Scheme 5. Model reaction for computational studies.
In a first stage, ΔG‡ for the uncatalyzed concerted cycloaddition for both regioisomers were
298
calculated for transition states [3]# and [4]# as 31.4 and 30.2 kcal/mol for the formation of 1a
and 2a, respectively (34.4 and 31.3 when the solvation model is applied, Table 5 and Web
Enhanced Table 1). Since experimentally no reaction was observed at room temperature in the
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 46
1
2
absence of a copper catalyst, pathways with transition states with ΔG‡
higher than ~30
3
4
298
5
6
kcal/mol were not pursued any further.
7
8
9
Table 5. Free energy profile ΔG‡ of the uncatalyzed model cycloaddition reaction.a
298
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
Compound ΔG‡
(kcal/mol) Digital repository
10042/to-10287
10042/to-10298
10042/to-13966
10042/to-10284
10042/24936
298
0.0
Az
0.0
Alk
[3]#
[4]#
1a
31.4
30.2
-62.9
-64.6
10042/24933
2a
a Species labelled with the suffix # correspond to located transition states. See Web Table 1 for
an interactive WEO (Web-enhanced-object) version. Digital repository entries have full details
of each calculation, and can be accessed as e.g. http://doi.org/10042/24933.
Active copper species in the cycloaddition of azides and iodoalkynes: The use of ligands is
crucial in the cycloaddition of azides and iodoalkynes. Examples have been reported where the
catalytically active species were generated in situ upon reaction of the chosen copper source and
ligand precursors,7,11 or alternatively, from a well-defined complex.8 Whereas the in situ strategy
avoids the need to pre-isolate any metallic species, the use of pre-formed catalysts generally
avoids the need for excess of ligand and/or reagents for achieving optimal catalytic performance
and allows for a better control of the nature of the species present in the reaction media.42 In
most cases, however, the complex used is a pre-catalyst, rather than the actual active species.
Hence, we tried to determine the most likely active species generated from either [CuCl(IPr)] or
[CuI(PPh3)3].
In the case of [CuCl(IPr)], the strength of the metal–NHC bond43 and the mild reaction
conditions used were expected to ensure the structural integrity of the catalyst. This was
18
ACS Paragon Plus Environment

Page 19 of 46
ACS Catalysis
1
2
1
3
4
confirmed by the H NMR analysis of the crude reaction mixtures. Accordingly, we used
5
6
[CuCl(IPr)] in our theoretical calculations.
7
8
9
For the phosphine-based reactions, it is important to note that exact nature of CuX/phosphine
systems in solution is very difficult to predict since not only ligand decoordination, but also the
formation of polynuclear species through halogen bonding needs to be accounted for.44 Also, it is
not possible to extrapolate the literature results to our system since the cycloaddition reactions
were run under neat conditions, which will affect the nature of the formed species from the
starting catalytic system. On the other hand, since we could not experimentally clarify the role of
lutidine in the [CuI(PPh3)3]-mediated cycloaddition reactions and the catalytic reactions
proceeded as well in the absence of this additive, lutidine was not included in the following
experiments or in the computational studies. This will also allow comparison of the
computational results obtained for the PPh3 and IPr systems, as lutidine did not improve the
performance of the latter (see Table 1).
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 first ran a series of experiments with different amounts of CuI and free triphenylphosphine.
As expected, no reaction occurred in the absence of any ligand (Table 6, entry 1),45 whereas
among all the in situ generated systems the best conversion was obtained when 10 mol % PPh3
(1:2 copper/ligand ratio) was used (Table 6, entry 3). Higher phosphine loadings had a
deleterious effect on the reaction, which might also be due to inefficient mixing under neat
conditions (Table 6, entries 4 and 5). Overall, the pre-formed complex [CuI(PPh3)3] remained
the best performing catalyst in the series (Table 6, entry 6).
Table 6. In situ generated PPh3-containing catalysts.
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 46
1
2
3
4
5
6
7
8
N
CuI mol
(5
%)
%)
N
N
Ph
Ph
N3
Ph
PPh3 X mol
(
+
in air
24 h
Neat,
RT,
I
Ph
I
1a
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
Entry
PPh3 (mol %)
Conv (%)a
1
0
0
2
5
41
81
70
<5
90
85
3
10
15
20
0
4
5
6b
7b
5
a 1H NMR conversions are the average of at least two independent experiments. b [CuI(PPh3)3]
(5 mol %) used instead of CuI.
Furthermore, the presence of significant amounts of free phosphine in the reactions is expected
to lead to the formation of the corresponding iminophosphorane issue of the Staudinger
reduction of benzyl azide.46 Due to the presence of moisture in the reaction mixture, this
phosphorane should be hydrolysed into benzylamine and triphenylphosphine oxide (Scheme 6).
However, when the model reaction was run in the presence of 5 mol % of CuI and 15 mol % of
PPh3 (see Table 6, entry 4), the appearance of a new multiplet at 4.29 ppm was observed, but no
1
31
traces of benzylamine were detected on the H NMR. Moreover, the P NMR spectrum only
showed two sharp signals at 29.9 and 39.2 ppm, respectively (ratio = 3:2).47 The first one
corresponds to triphenylphosphine oxide, and the second one, which is too downfield for an
iminophosphorane, was attributed to a phosphonium salt based of similar/identical chemical
shifts of benzylaminophosphonium salts reported in the literature (Scheme 6).48 In our case the
most likely counterion for such a salt is iodide resulting from the partial dehalogenation of the
20
ACS Paragon Plus Environment

Page 21 of 46
ACS Catalysis
1
2
3
4
5
6
7
8
9
starting iodoalkyne, as traces of phenylacetylene were also observed in the crude reaction
products. This undesired reaction appeared to be particularly active when using mono-phosphine
complexes as catalysts (see Table 2).
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
+
N3
PPh3
Ph
N2
PPh3
N
Ph
Ph
N PPh3
H
Ph
NH2
+
X
O=PPh3
Scheme 6. Undesired reactivity observed with free PPh3.
When the model reaction was carried out with 5 mol % of [CuI(PPh3)3], the 31P NMR
spectrum of the reaction mixture did not present the signals corresponding to the starting
catalyst, its mono-phosphine analogue or PPh3. Instead, the spectrum showed that
triphenylphosphine oxide was the major phosphorus-containing compound as well as the
formation of two unknown minor products at 9.6 and 6.6 ppm.45 These results seems to suggest
that [CuI(PPh3)3] is not the actual catalyst in the cycloaddition reaction but also, that all PPh3
molecules play an important role in the catalytic cycle, maybe stabilizing more reactive species.
This would avoid their decomposition and a consequent loss of activity, since they are not
available for reacting with benzyl azide. Taking into account these observations, [CuI(PPh3)]
was chosen as catalyst in our computational studies. For comparative purposes, some of the
calculations were also carried out with [CuCl(PPh3)]. Not only was this catalyst shown to be
active in the preliminary screening (see Table 2), it also represents a useful link to assess the
effect of changing either a L or a X ligand from [CuI(PPh3)] to [CuCl(IPr)].
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 46
1
2
3
4
5
6
7
8
9
Plausible catalytic cycles: Two reaction pathways have been previously proposed for the
cycloaddition reaction of azides and iodoalkynes, both heavily based on previous evidence for
the cycloaddition of azides and terminal alkynes (Scheme 7).4,16,49 Path I would initiate with the
formation of a σ-acetylide species. This is a well-established intermediate in the cycloaddition of
azides and terminal alkynes, and therefore from here, the reaction would undergo the broadly
accepted sequence with monosubstituted alkynes. The eventual reaction of a triazolide
intermediate with the starting iodoalkyne would generate the expected product and close the
catalytic cycle.50 In Path II, the starting iodoalkyne would be π-coordinated by the copper
catalyst before interacting with the azide,51 which would lead to the formation of a vinylidene 6-
membered cyclic species that would produce the expected triazoles upon reaction with a
molecule of iodoalkyne (Scheme 7). A very significant difference between both proposals is that
in Path II, the carbon–iodine bond is not severed during the catalytic cycle. Triazolides are
known to be extremely sensitive to the presence of proton sources,52 and therefore unsuitable
intermediates in this reaction. A 5-H-triazole should indeed be the only reaction product under
technical conditions should the reaction proceed via a triazolide intermediate.53
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
R1
R1
R1
R2
N3
N
N3
N
N
I
I
R2
[CuIII]
R2
R2
[Cu]
[Cu]
[Cu]
PATH I
R1
]
N
R1
N
[Cu]
N
N
N
N
I
PATH III
[CuIII
R2
R2
N
R1
[Cu]
N
R1
I
R2
N
N
I
R2
N
N
I
R2
I
R2
I
R2
R1
N
PATH II
R2
N
PATH IV
I
N
N
I
R2
[Cu]
I
R2
I
R2
[Cu]
R2
I
[Cu]
N
R1
N
N
N
[Cu]
R1
N
N3
R1
[Cu]
[Cu]
R1
N3
Scheme 7. Overview of mechanistic pathways for the cycloaddition reaction. [Cu] denotes a
ligated copper(I) center unless indicated otherwise.
22
ACS Paragon Plus Environment

Page 23 of 46
ACS Catalysis
1
2
3
4
5
6
7
8
9
In addition to Path II, we have also investigated two other alternatives: Paths III and IV
(Scheme 7). Path III would involve the oxidative addition of the iodoalkyne onto the copper
catalyst, followed by the cycloaddition step and a reductive elimination. Path IV is based on the
direct activation of the iodoalkyne via π-coordination to the copper center.
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
Path II (CuIII metallacycles): Using our level the theory and [CuI(PPh3)] as catalyst, we were
able to locate not only [5]#, leading to formation of the observed 5-iodotriazole, but also the
regioisomeric [6]# (Scheme 8, Table 7 and Web Enhanced Table 2). Transition state [5]# is 4
kcal/mol lower in ΔG‡
than [6]# (20.6 and 24.3 kcal/mol respectively for gas phase). This
298
difference in energy can account for the complete regioselectivity observed experimentally, even
if it is much smaller than the previously found in calculations with terminal alkynes (i.e. 13
kcal/mol in favor for the formation of 5-H-triazoles).54 The relative configuration of the copper
centre was found to be very important as a transition state [7]# leading to the formation of 1a but
presenting both halogen atoms on the same side of the transition state, was significantly higher in
energy, even higher than regioisomeric [6]#. The transition states leading to the formation of 1a
located with [CuCl(PPh3)] as catalyst, [8]# to [10]#, displayed a very similar free energy to the
corresponding [5]# to [7]# (Scheme 8, Table 7, and Web Enhanced Web Table 2). These results
correlate well with the experimental catalyst screening (see Table 2), where a change of the
halogen on the copper catalyst led only to slightly different performances.
Table 7. Free energy profile ΔG‡ for the formation of iodotriazoles via a CuIII metallacycle.a
298
Compound
ΔG
Digital
(kcal/mol)
repository
[CuI(PPh3)]
0.0
10042/24798
10042/25072
10042/24997
[CuCl(PPh3)] 0.0
[CuCl(IPr)] 0.0
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 46
1
2
3
4
5
6
7
8
9
20.6
24.3
26.5
20.7
23.2
26.6
28.4
34.1
33.1
10042/25071
10042/24601
10042/24668
10042/24488
10042/29498
10042/29027
10042/24666
10042/24831
10042/24843
[5]#
[6]#
[7]#
[8]#
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]#
[10]#
[11]#
[12]#
[13]#
a Species labelled with the suffix # correspond to located transition states. See Web Table 2 for
an interactive WEO (Web-enhanced-object) version.
Similar calculations with [CuCl(IPr)] as catalyst revealed that the barrier in this case slightly
higher ([11]#, 28.4 kcal/mol), but, this transition state is considerably stabilized when compared
to [13]# as well as to its regioisomer leading to the formation of a 4-I-triazole 2a ([12]#, 34.1
kcal/mol) (Scheme 8 and Table 7).
24
ACS Paragon Plus Environment

Page 25 of 46
ACS Catalysis
1
2
3
4
Ph
Ph
N3
Ph
Ph
5
6
7
8
N
I
N
I
+
N
I
N
N
N
N
N
N
Ph
I
I
PPh3
I
Cu
Cu
Cu
Ph
Ph
Ph
+
PPh3
I
PPh3
9
Ph3P
I
Cu
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
[5]#
[7]#
[6]#
0.0
20.6
26.5
24.3
Ph
Ph
Ph
N3
Ph
Ph
N
I
N
+
+
N
N
N
N
N
N
N
I
PPh3
Cl
Cl
Cu
Cu
Cu
Ph
Ph
Ph
PPh3
Cl
PPh3
I
I
Ph3P
Cu Cl
[8]#
20.7
[10]#
26.6
[9]#
23.2
0.0
Ph
Ph
Ph
Ph
N3
N
N
I
N
I
N
N
+
+
N
N
N
N
IPr
Cu
Cl
Ph
IPr
I
Cl
Cu
Cu
Ph
Ph
Ph
Cl
IPr
I
IPr
Cu Cl
[11]#
28.4
[12]#
34.1
[13]#
33.1
0.0
Scheme 8. Located CuIII transition states. Values are for ∆G298 in kcal/mol.
Intrinsic reaction coordinate (IRC) computations were run from geometries [5]# and [8]# to
further validate this mechanistic pathway. These showed two interesting features (Scheme 9,
Web Enhanced Table 2 and Figure 2). First, no π-coordination of the alkyne is involved prior the
formation of [5]# or [8]# as it had been previously proposed,7 and the copper catalyst interacts
with benzyl azide instead.
Both before and after the transition states features appeared in the gradient norm characteristic
of a so-called hidden intermediate in the reaction (Figure 2b, red arrows). A hidden intermediate
(it could also be called a frustrated intermediate) is not expected to have a lifetime longer than a
molecular vibration, but it could potentially be converted into a real intermediate if the reaction
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 46
1
2
conditions or the substitution pattern are changed.55 In our case, at IRC = 2.5 the hidden
intermediates (HI) [5]HI and [8]HI can be found for the formation of the second C–N bond (I–
C∙∙∙N–Bn), which is concomitant with the reductive elimination of the copper to form the triazole
product. Hence, both the cycloaddition and the elimination are part of the same concerted
process, albeit occurring asynchronously. Furthermore, a second hidden intermediate emerged
before [5]#, corresponding to an agostic interaction between the copper center and the alkyne C–
I bond.
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
Ph
N
Ph
N
N
N
N
N
N
N
Ph
1a
N
X
Cu
Ph
I
Ph
X
Ph
Cu
+
PPh3
I
I
PPh3
[CuX(PPh3)]
=
=
=
=
X
X
I
Cl
X
X
I
[5]#
[8]#
[5]HI
Cl
[8]HI
Scheme 9. IRC path from transition state to products.
(a)
(b)
Figure 2. The calculated intrinsic reaction coordinate for [5]#, showing (a) the relative energy
and (b) the computed gradient norm. For the latter, red arrows indicate the occurrence of hidden
26
ACS Paragon Plus Environment

Page 27 of 46
ACS Catalysis
1
2
3
4
5
6
reaction intermediates. The same features are also visible in the IRC computed with a continuum
solvent field.
7
8
9
Path III (Oxidative addition of the iodoalkyne): The oxidative addition of haloalkynes onto
copper(I) centres has already be evoked for transformations such as the Cadiot-Chodkiewicz
coupling of 1-haloalkynes with terminal alkynes,56,57 but not for this type of cycloaddition
reaction. We explored this possibility but we were unable to locate any transition state leading to
the observed 1a.
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
On the other hand, the energy of the found transition states [14]# and [15]# for the formation
of 4-iodotriazole 2a was greater in magnitude than the uncatalysed thermal process by ≈45
kcal/mol (Scheme 10 and Web Enhanced Table 2). In consequence, this sequence was not
studied any further.
Ph
Ph
N3
N
N
N
N
N
Ph
N
N
N
N
Ph
+
+
I
I
I
Ph
I
I
Cu
X
Cu Cl
Cu
L
Ph
Ph
Ph
PPh3
IPr
Catalyst
0.0
Not located
or
[14]#
45.4
[15]#
47.2
=
PPh3
(L
IPr)
Scheme 10. Calculated energy barriers for the cycloaddition reaction after oxidative addition.
Values are for ∆G298 in kcal/mol.
Path IV (Direct activation via π-coordination of the iodoalkyne): Copper(I) alkyne π-
complexes have long been considered essential intermediates in the cycloaddition of azides and
alkynes (either terminal or halogenated).4,7 However, the cycloaddition of terminal alkynes via π-
coordination of the alkyne without its deprotonation has been found to be higher in energy than
the uncatalysed reaction.47 We decided to assess this type of mechanism with iodoalkynes since
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 46
1
2
3
4
5
6
both cycloaddition reactions show significant differences (Schemes 11 and 12; Table 8, and web-
Enhanced Table 3).
7
8
9
From the phosphine catalyst and the model iodoalkyne, two regioisomeric π-complexes 16 and
17 can be formed. Both compounds have similar free energies (-2.1 and 0.1 kcal/mol,
respectively); their formation is exothermic (ΔH ≈15 kcal/mol) but this is negated by loss of
entropy. The reaction of 16 with benzyl azide led to transition state [18]#, a precursor of a 5-
iodotriazole 1a and which is only 19.6 kcal/mol above the starting materials in free energy.
Moreover, its regioisomer [19]# is 5 kcal/mol higher in energy and a transition state [20]#
leading to 1a but issue of the reaction of π-complex 17 displayed an equally high energy
(Scheme 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
PPh3
I
I
Ph3P
Ph
Ph3P
Ph
I
Cu
+
Cu
Cu
Ph
I
I
I
17
16
0.1
-2.1
Ph
N3
N
N
N
N
N
Ph
N
N
Ph
N
N
Ph
Ph
Ph3P
Ph
I
I
I
Ph
I
Cu
Cu
Cu
Ph3P
I
I
PPh3
[18]#
19.6
[19]#
24.5
[20]#
24.4
Scheme 11. Direct π-activation with [CuI(PPh3)]. Values are for ∆G298 in kcal/mol.
Table 8. Free energy profile ΔG‡ for the formation of iodotriazoles via direct π-activation.a
298
Compound ΔG (kcal/mol)
-2.1
Digital repository
10042/25672
16
28
ACS Paragon Plus Environment

Page 29 of 46
ACS Catalysis
1
2
3
4
5
6
7
8
9
0.1
10042/24673
10042/24844
10042/24838
10042/24834
10042/24925
10042/24923
10042/25040
10042/24914
10042/24934
17
19.6
24.5
24.4
3.9
[18]#
[19]#
[20]#
21
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
3.9
22
24.2
30.7
28.0
[23]#
[24]#
[25]#
a Species labelled with the suffix # correspond to located transition states. See Web Table 3 for
an interactive WEO (Web-enhanced-object) version.
Very similar results were obtained with [CuCl(IPr)] as the catalyst, and even if [23]# has a
higher energy of 24.2 kcal/mol, this remains the lowest transition stated located with the NHC
system. Although this value of energy seems slightly high for reactions operating at room
temperature, it should be taken into account that the -2.7 kcal/mol correction for a standard state
of ~4 M (the actual reaction conditions) is not applied in any of the Tables (vide supra).
IPr
I
I
IPr
Ph
IPr
Ph
I
Cu
+
Cu
Cu
Ph
I
I
I
22
3.9
21
3.9
Ph
N3
N
N
N
N
N
Ph
N
N
I
Ph
N
N
Ph
Ph
IPr
Ph
IPr
I
Ph
Cl
I
Cu
Cu
Cu
Cl
Cl
IPr
[23]#
[24]#
30.7
[25]#
28.0
24.2
29
ACS Paragon Plus Environment