Direct Synthesis of Nitriles from Carboxylic Acids Using Indium-Catalyzed Transnitrilation: Mechanistic and Kinetic Study

Article abstract of DOI:10.1021/acscatal.9b02779

Aliphatic and aromatic carboxylic acids can be quantitatively converted to the corresponding nitriles in the presence of catalysts using acetonitrile both as a solvent and reactant at 200 °C. This transformation is based on the acid-nitrile exchange (i.e., transnitrilation) and uses a nontoxic and water resistant catalyst, indium trichloride (InCl₃). The mechanism of the transnitrilation was investigated both experimentally and computationally and compared to the previously proposed mechanism. In contrast to the usually assumed formation of amide as an intermediate, transnitrilation is an equilibrium reaction and proceeds via an equilibrated Mumm reaction with the formation of an imide as an intermediate. A simple and reversible mechanism was proposed for this reaction, which was validated by kinetics measurement and by density functional theory calculations of the reaction intermediates and reaction mechanisms.

Full text of DOI:10.1021/acscatal.9b02779

Subscriber access provided by UNIV OF CAMBRIDGE
Article
Direct Synthesis of Nitriles from Carboxylic Acids Using Indium
Catalyzed Transnitrilation: Mechanistic and Kinetic Study
laurent VANOYE, Ahmad HAMMOUD, Hélène GERARD, Alexandra Barnes,
Régis Philippe, Pascal Fongarland, Alain Favre-Reguillon, and Claude de Bellefon
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.9b02779 • Publication Date (Web): 18 Sep 2019
Downloaded from pubs.acs.org on September 18, 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 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
Direct Synthesis of Nitriles from Carboxylic Acids
Using Indium Catalyzed Transnitrilation:
Mechanistic and Kinetic Study
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
†
‡
†
†
Laurent Vanoye, * Ahmad HAMMOUD, Hélène Gérard, Alexandra Barnes, Régis Philippe,
†
†
†,#
Pascal Fongarland, Claude de Bellefon, Alain Favre-Réguillon *
†
Université Lyon, Laboratoire de Génie des Procédés Catalytiques UMR 5285, CNRS-
CPE Lyon-UCBL, 43 boulevard du 11 novembre 1918, F-69100 Villeurbanne, France
‡
#
Sorbonne Université, CNRS, Laboratoire de Chimie Théorique, F-75252 Paris, France
Conservatoire National des Arts et Métiers, EPN 7, 2 rue Conté, 75003 Paris, France
ACS Paragon Plus Environment
1

ACS Catalysis
Page 2 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
Aliphatic and aromatic carboxylic acids can be quantitatively converted to the corresponding
nitriles in the presence of catalyst using acetonitrile both as solvent and reactant at 200°C. This
transformation is based on the acid-nitrile exchange (i.e. transnitrilation) and uses nontoxic and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
water resistant catalyst, indium trichloride (InCl ). The mechanism of the transnitrilation was
3
investigated both experimentally and computationally and compared to the previously proposed
mechanism. In contrast to the usually assumed formation of amide as intermediate, transnitrilation
is an equilibrium reaction and proceeds via an equilibrated Mumm reaction with the formation of
an imide as an intermediate. A simple and reversible mechanism was proposed for this reaction,
which was validated by kinetics measurement and by DFT calculations of the reaction
intermediates and reaction mechanisms.
KEYWORDS: nitrile, indium, transnitrilation, mechanism, DFT, Mumm rearrangement
ACS Paragon Plus Environment
2

Page 3 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
INTRODUCTION
Nitriles are versatile building blocks and precursors in organic synthesis due to their unique
reactivity and activating ability.1-3 The cyano group itself is ubiquitous in biologically active
compounds, industrial processes and is used as synthons for further synthetic elaboration.4-6
Typical routes to prepare nitriles can be divided into two classes, one involving the introduction
of a cyano group with inorganic or organic cyanating reagents7-8 and the other involving creation
of a cyano group by chemical transformation of other functional groups.9-13
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
14
Alongside these efficient methods, the ready availability of carboxylic acids makes them
extremely promising raw materials for nitriles synthesis. Many protocols for the preparation of
nitriles from carboxylic acids can be found in the literature, most of them consist of two or more
step procedures. Indeed, the biosynthesis of aliphatic nitriles starts from carboxylic acids, which
are converted into their amides and finally dehydrated. 15 The direct nitrilation of fatty acids16,
fatty acid methyl esters (FAME)17 or triglycerides18 by reaction with NH (ammoniation–
3
dehydration) has been studied (Figure 1A). However, due to the high reaction temperature required
(300 to 400°C, unwanted by-products are formed. It should be noticed that procedures for the
dehydration of primary amides under mild conditions have been also proposed.12, 19-20 The
decarboxylative cyanation of carboxylic acids (Figure 1B) was performed using photoredox
21
catalysis using cyanobenziodoxolone. Nevertheless, in this approach, cyanobenziodoxolones had
2
2
to be prepared and one carbon atom is lost. Based on the work of Weinreb et al. using aluminum
amide, a one-step transformation of carboxylic acids into nitriles have been proposed (Figure
1
C).23-24 However, using this methodology, a huge amount of salts are produced.
ACS Paragon Plus Environment
3

ACS Catalysis
Page 4 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A) ammoniation-dehydration
OR
NH3
NH2
R1
R1
R1
C
>
300 °C
> 300 °C
N
O
O
- H O
- H O
2
2
B) decarboxylative cyanation
O
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
I
N
O
OH
N
R1
C
R1
h

O
Ir cat.
C) reduction addition elimination
OH
H
NH X
4
Al
H
R1
R1
R1
C
N
O
O
D) transnitrilation
OH
OH
O
R1
R1
C
+
R2
C
N
+
R2
N
O
Figure 1 Overview of carboxylic acids-to-nitriles conversion routes
Acid-nitrile exchange reaction (transnitrilation, Figure 1D) has been used to accomplish the
direct conversion of adipic acid to adiponitrile using an excess of acetonitrile at 260°C in the
25
presence of phosphoric acid as catalyst. This reaction could be improved by replacing acetonitrile
2
6
27-28
28
with higher boiling nitriles or di-nitriles
, by using a catalytic amount of AlCl , or in the
3
presence of large excess of Brønsted acids at low temperature (100°C).29-31 This method has been
3
2
applied successfully in continuous flow by Kappe et al. under high temperature/high pressure
(350°C and 65bar, i.e. supercritical conditions) without any catalyst.
Although the reaction is straightforward and gives the nitrile product in good yield, the high
temperature/pressure process or the use of excess of Brønsted acids or water sensitive Lewis acids
ACS Paragon Plus Environment
4

Page 5 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
make the protocols less attractive. Therefore, efficient catalytic procedures for the transnitrilation
of carboxylic acid are still in great demand. Herein we report that the nitrile synthesis via the acid-
nitrile exchange reaction may be accomplished in a classical batch reactor, and in high yield using
a catalytic quantity of a readily available, water-tolerant, innocuous Lewis acid, i.e. indium
trichloride (InCl3).33-34 Finally, the theoretical mechanism is revisited (experimentally and
computationally) and a kinetic model capturing the experimental trends is proposed and optimized.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
RESULTS AND DISCUSSION
Catalysts screening. A screening of Lewis and Brønsted acid catalysts was performed using
hexanoic acid 1 (caproic acid) and acetonitrile, this latter both as reagent and solvent. It is
noteworthy that we aimed at developing a versatile chemical transformation without any need for
additional pre-treatment of the solvent (drying, degassing, …). Thus, the reaction was performed
in a non-anhydrous commercial solvent, i.e. with reagent grade commercial acetonitrile and
without any subsequent drying or purification. The initial concentration of 1 was fixed to 0.5 M
and the reaction temperature was set to 200°C. As expected from Antoine’s parameters, an
autogenous pressure around 15 bar was obtained in the batch reactor. The catalyst content was set
at 0.025 M (5 mol%) and the capronitrile 2 yield as the function of time for various catalysts are
shown in Figure 2. Time zero is taken while the reactor reaches 200°C, typically after 15 minutes
heating. At that point, a hexanoic acid 1 conversion below 10% could be noticed but the
capronitrile 2 yield was always below 1%.
ACS Paragon Plus Environment
5

ACS Catalysis
Page 6 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O
N
C
cat. 5 mol%
200°C
OH
1
C
N
2
1
00
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
InCl3
DyCl3
GaCl3
AlCl3
PS-SO3H
uncatalysed
0
50
100
150
200
250
300
350
400
450
500
Time (min)
Figure 2 : Capronitrile 2 yield as the function of time and catalysts. Dots represent experimental
data, dashed lines are guides for the eye and the continuous line represents the model output (vide
infra). Reaction conditions: 200°C, 0.5M of hexanoic acid 1 in reagent-grade acetonitrile, 5 mol%
of catalyst, 15 bar (autogenous pressure).
As seen in Figure 2, the reaction occurs slowly without catalyst at 200°C yielding less than 10%
of capronitrile 2 after 6 h. Moderate yield (i.e. <30% after 8h) was obtained with solid Brønsted
acid, i.e. Purolite CT482. Purolite CT482 is a macroporous sulphonic acid heterogeneous
polymeric catalyst (PS-SO H) designed for use in high-temperature processes (Tmax~200°C).
3
Thus, the moderate yield could not be explained by the thermal degradation of the polymeric resin.
However, one can suspect a deactivation of the catalyst via the hydrolysis of acetonitrile and
3
5
formation of acetamide which is adsorbed on the catalyst surface. Under our conditions (Figure
2
8
2), the similar performance of AlCl compared to literature could be noticed. Despite, the use of
3
ACS Paragon Plus Environment
6

Page 7 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
“
undried” solvent as well as a different nitrile source (acetonitrile versus glutaronitrile) yield for
aliphatic nitrile were comparable (i.e. 45% for capronitrile using 10mol% of AlCl compared to
3
51% for 3-phenylpropanenitrile using 2 mol% of AlCl3).28
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
AlCl has always been considered to be the more attractive catalyst for organic transformations
3
mainly because the Lewis acidity of aluminum compounds are generally higher than those of other
group 13 metals compounds (Al, Ga, In, Tl). It also appears that the cost is a crucial parameter that
limits the development of other group 13 metals-mediated organic transformations. However, with
the exception of thallium which is highly toxic, the unique properties of gallium and indium in
organic transformations make them interesting catalysts for the transnitrilation reaction.34
However, yield were comparable using AlCl or GaCl as a catalyst in reagent-grade acetonitrile.
3
3
Again, the water sensitivity of GaCl could explain the rather low efficiency of the catalyst. Then,
3
InCl was evaluated. InCl appears to be attractive for two reasons: i) it is unaffected by air or
3
3
water33 and can be easily handled without any apparent toxicity; and ii) it exhibits a low
heterophilicity and, therefore, nitrogen- and oxygen-containing functional groups are tolerated in
3
4
the presence of indium compounds. We were pleased to see that, using InCl , the yield of
3
capronitrile 2 higher than 97% was obtained in 7h at 200°C (Figure 2). Evaluation of milder water
tolerant Lewis acid, i.e. DyCl , leads to remarkable results compare to AlCl , but less interesting
3
3
compared to InCl3.
On the basis of these inputs, further experiments were carried out using InCl as a catalyst. The
3
influence of the amount of InCl on the hexanoic acid 1 conversion is reported in Figure 3. A linear
3
relationship between the amount of InCl and initial rate could be observed (Figure 3 insert).
3
Increasing the amount of InCl to 10 mol% resulted in the quantitative conversion of hexanoic acid
3
1 and yield above 98% in less than 5h at 200°C.
ACS Paragon Plus Environment
7

ACS Catalysis
Page 8 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
0 mol%
5 mol%
3 mol%
0 mol%
1
00
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0
0
.5
.4
0.3
0.2
0
0
.1
.0
0
2
4
6
8
10
InCl3 (mol%)
0
50 100 150 200 250 300 350 400 450 500 550 600
Time (min)
Figure 3 : Hexanoic acid 1 conversion as the function of time and InCl loading. Points represent
3
experimental data and lines represent model output (vide infra). Reaction conditions: 200°C, 0.5M
of hexanoic acid 1 in reagent-grade acetonitrile, 0 to 10 mol% of InCl 15 bar (autogenous
3
,
pressure).
ACS Paragon Plus Environment
8

Page 9 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
Determination of the activation energy for the overall transnitrilation reaction was then carried
out between 160 and 270°C (autogenous pressure up to 60 bar). An Arrhenius plot for hexanoic
acid consumption is shown in Figure 4. The least-squares analysis of the data yields an E value
a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-1
-1
of 80  10 kJ mol (19.1  2.5 kcal mol ).
-4.0
-4.5
-5.0
-5.5
-6.0
-6.5
-
7.0
7.5
-
-8.0
0.00024
0.00025
0.00026
0.00027
0.00028
1/RT
Figure 4: Arrhenius plots for the transnitrilation reaction of hexanoic acid. 0.5 M in reagent-grade
-1
acetonitrile, 10mol% InCl , r° in min .
3
Reaction mechanism. To our knowledge, only a few publications dealing with the mechanism
of the transnitrilation reaction have been published.28-32 Moreover, a highly stable intermediate
was suggested in recent publications, i.e. amide.31-32 The first step of the transnitrilation reaction
is shared by all the publications (Figure 5). An acyl iminol A is produced by the reaction of
carboxylic acid B and acetonitrile. Such intermediate has never been observed. It is supposed that
3
6
the rapid rearrangement of A (Mumm rearrangement ) produce the key intermediate imide C
Figure 5).
(
ACS Paragon Plus Environment
9

ACS Catalysis
Page 10 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O
4
O
HO
R
NH2
Mumm
rearrangement
H
O
O
O
amide D
O
O
N
-
(
21.8
-5.9)
+
N
C
R C
N
R
N
H
R
OH
R
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
H20
carboxylic acid B
imide C
nitrile E
acyl iminol A
0.0
4.1
(21.0)
-15.9
(2.2)
0.0
(0.0)
(
0.0)
Figure 5: Proposed mechanism of transnitrilation of carboxylic acid B into nitrile E (adapted from
ref. 31-32) and associated energies (kcal mol ) with respect to the starting material for R = Me.
-1
-1
Gibbs Free Energies (kcal mol ) are given below in parenthesis. Intermediates acyl iminol A and
amide D in square brackets have never been observed.
In the second step of the proposed mechanism, the imide C was then supposed to act as water
3
2
scavenger and was hydrolyzed into acetic acid 4 and amide D. The amide D is then dehydrated
into the corresponding nitrile E, such dehydration is promoted through imide C hydrolysis. The
reverse mechanism can also be invoked to form imide C, by hydration of acetonitrile to yield
ethylamide, which can then undergo condensation with carboxylic acid B to form imide C.
The stability of the postulated intermediates was thus examined computationally for an identity
reaction model (R = Me, Figure 5). The intermediate acyl iminol A is slightly higher in energy
than the reactant and is entropically disfavored due to the associative nature of this step and it
should thus not be isolated. The Mumm rearrangement yielding C and the formation of the amide
-1
D by hydrolysis of C are both found to be exothermic (by 20.0 and 5.9 kcal mol respectively).
As no major entropic effects are expected for these two steps (conservation of number of molecules
between reactant and products), formation of both C and D should be favored over A. In contrast,
ACS Paragon Plus Environment
1
0

Page 11 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
their transformation starting from B (or with respect to the product E, which is identical to the
reactant for R = Me) is entropically disfavored and the value of this entropic contribution highly
depends on the reaction condition (pressure, temperature, concentrations, and also, most probably,
on the nature of R). As a consequence, the stability of both C and D should depend on the reaction
conditions.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
It should be mentioned that the highly stable amide D and/or acetamide have never been
observed.31-32 Even if the dehydration of amide to the corresponding nitrile is a well-documented
process in the literature,37-38 in general dehydration requires the use of highly reactive reagents,
3
7
such as thionyl chloride, phosphorus oxychloride. In the past few years, a number of reagents or
metal-catalyzed methods have emerged for carrying out efficient dehydration of amides to nitriles
under milder conditions, like Pd(II)-catalyzed “water shuffling” process between acetonitrile and
3
9
the corresponding amide. However, the dehydration of amide under the reaction conditions of
transnitrilation is questionable and it should be again emphasized that no amides could be detected
in the reaction mixture.
With the optimized reaction conditions, we explored the mechanism of the transnitrilation
reaction based on analysis of the crude as the function of time. Using GC/MS different
intermediates could be identified in the reaction mixture and their evolution as the function of time
could be easily followed (Figure 6).
ACS Paragon Plus Environment
1
1

ACS Catalysis
Page 12 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O
O
O
a
b
N
0.5
C
OH
N
H
0.4
0.3
0.2
0.1
0.0
hexanoic acid 1
capronitrile 2
N-acetylhexanamide 3
O
O
O
O
O
hexanoic acid 1
capronitrile 2
N-acetylhexanamide 3
Acetic acid 4
N-acetylacetamide 5
N-hexanoylhexanamide 6
OH
acetic acid 4
N
H
N
H
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
N-hexanoylhexanamide 6
N-acetylacetamide 5
O
O
0
50
100
150
200
250
300
350
400
450
O
Time (min)
hexanoic anhydride 7
Figure 6 : a) Composition of the reaction mixture as the function of time. Points represent
experimental data and lines represent model output (vide infra). Reaction conditions: 200°C, 0.5M
of hexanoic acid 1 in reagent-grade acetonitrile, 10 mol% of InCl , 15 bar (autogenous pressure)
3
and b) Structure of molecules identified in the reaction mixture.
Using 10 mol% of InCl , it has already seen that conversion of hexanoic acid 1 into capronitrile
3
2
higher than 97% could be obtained in 5h (Figure 3). N-acetylhexanamide 3 was rapidly produced
within the first 10 min of the reaction and then decreased slowly until the full conversion of
hexanoic acid 1 (7h). N-acetylacetamide 5 was also identified and was slowly produced until a
constant value after 2 h.
Acetic acid 4 was also identified in the reaction mixture. However, as the response factor for
acetic acid 4 is 5 times smaller compared to the reactant with poor peak shape, it was difficult to
quantify it precisely. A very small amount of N-hexanoylhexanamide 6 (<5 mol%) could be
detected in the first hours of the reaction. As expected from the temperature and the presence of
Lewis acid, the reaction mixture contains small amounts of hexanoic anhydride 7 (<3% at 30 min,
not shown in Figure 6a) which slowly disappear. As expected from the literature,31-32 the highly
ACS Paragon Plus Environment
1
2

Page 13 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
stable acetamide 8 or hexanamide 9 have never been observed in the reaction mixture.
3
2
Furthermore, we (See Supplementary Information) and others have shown that the hexanamide
9could not be transformed into capronitrile 2 under our conditions (200°C, 0.5 M in reagent-grade
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
acetonitrile, 5 mol% InCl , 7h).
3
The influence of carboxylic acid concentration was then studied. The concentration of hexanoic
acid 1 in acetonitrile was varied from 0.5 to 5 M while keeping the amount of InCl at 0.05 M
3
(Figure 7).
ACS Paragon Plus Environment
1
3

ACS Catalysis
Page 14 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
0.6
0.5
0
.5 M
0.4
0.3
0.2
0.1
0.0
hexanoic acid 1
capronitrile 2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
N-acetylhexanamide 3
N-hexanoylhexanamide 6
0
0
0
50
50
50
100
100
100
150
150
150
200
250
300
350
400
450
Time (min)
b
2.5
2 M
2.0
1.5
1.0
0.5
0.0
hexanoic acid 1
capronitrile 2
N-acetylhexanamide 3
N-hexanoylhexanamide 6
200
250
300
350
400
450
Time (min)
6
5
4
3
2
1
0
c
5 M
hexanoic acid 1
capronitrile 2
N-acetylhexanamide 3
N-hexanoylhexanamide 6
200
250
300
350
400
450
Time (min)
Figure 7 : Composition of the reaction mixture as the function of time and initial
concentration hexanoic acid 1. Dots represent experimental data and lines represent model
output (vide infra). Reaction conditions: 200°C, hexanoic acid 1 0.5 to 5 M in reagent-grade
acetonitrile, 10 mol% InCl , 15 bar (autogenous pressure).
3
ACS Paragon Plus Environment
1
4

Page 15 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
From Figure 7, we could see the influence of carboxylic acid/nitrile ratio on conversion and
yield. For diluted hexanoic acid 1 (0.5 M) (Figure 7a), total conversion of the carboxylic acid and
yield in capronitrile 2 higher than 97% could be reached in 7h using 0.05 M of InCl . A
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
concentration of imide 3 close to 25 mol% was observed within the first minutes of the experiment
which then decreases slowly and become null after 7 h. For a higher concentration of hexanoic
acid 1 (2 M) (Figure 7b), the total conversion of the carboxylic acid could not be reached. After
7
h, 6 mol% of hexanoic acid 1 is still present in the reaction mixture. Under those conditions, a
lower concentration of N-acetylhexanamide 3 was observed within the first minutes of the
experiment (19 mol%). The concentration of N-acetylhexanamide 3 slowly decreases but after 7 h
at 200°C, 6 mol% of N-acetylhexanamide 3 could be still detected. The yield of capronitrile 2 for
a concentration of 2M was below 85%. By increasing molarity of hexanoic acid 1 to 5 M (Figure
7
c), the yield in capronitrile 2 dropped to 57% and the conversion of hexanoic acid 1 was below
75%. For 5M, the maximum concentration of N-acetylhexanamide 3 was also decreased to 14
mol% after 2 h and was almost constant until the end of the experiment. Under those conditions,
no amide (acetamide 8 or hexanamide 9) could be detected in the reaction mixture and these results
clearly show that the transnitrilation reaction is a fully reversible reaction and the equilibrium can
be shifted by dilution with acetonitrile. The reversibility of the process was further confirmed by
using benzonitrile as a reagent and solvent. 0.5 M of acetic acid in benzonitrile under the standard
conditions (200°C, InCl 10 mol%), benzoic acid and acetonitrile could be detected in the reaction
3
mixture.
The 2 M experiment was used to evaluate the Standard Gibbs free energies for the formation of
-1
N-acetylhexanamide 3 out of acetonitrile and hexanoic acid 1 (namely Δ G°(1) = + 2.6 kcal mol
r
), as well as that for the formation of capronitrile 2 and acetic acid out of N-acetylhexanamide 3
ACS Paragon Plus Environment
1
5

ACS Catalysis
Page 16 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
-1
(
Δ G°(2) = - 2.9 kcal mol ). As expected, their absolute values are close, as the only difference
r
comes from the exchange between methyl and hexyl. These values compare well with the value of
-1
+2.2 kcal mol obtained computationally for the formation of N-acetylhexanamide 3 from the
starting materials (see Computational section for details of the thermochemical evaluation in
Supplementary Information) as the difference clearly falls within the limits of the DFT methods.40
Taking advantage of the observation of the reversibility of the reaction evidenced above, we
proposed an alternative mechanism based on the microreversibility principle applied to the global
process (Figure 8). After formation of the imide C via a Mumm rearrangement of acyl iminol A,
a reverse Mumm rearrangement resulted in the formation of an acetic alkylimidic anhydride F, a
tautomer of A. Elimination of acetic acid 4 from F, via a mechanism reverse of the one used for
the formation of A, leads to the formation of the nitrile E.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Mumm
rearrangement
H
O
O
O
H
O
O
O
O
N
N
O
+
+
N
C
R
N
H
R C
N
HO
R
OH
R
Mumm
rearrangement
R
carboxylic acid B
acyl iminol A
imide C
acetic alkylimidic anhydride F
nitrile E
4
Figure 8: Revisited mechanism of transnitrilation of carboxylic acid B into nitrile E via imide C
and tautomers A and F as intermediates.
In order to confirm the role of the imide C, i.e. N-acetylhexanamide 3 when hexanoic 1 was used
4
1
as substrate, N-acetylhexanamide 3 was synthesized according to a literature procedure and its
reactivity was evaluated under our conditions. The transformation of 3 in acetonitrile in the
presence of 0.04M of InCl at 200°C was studied (Figure 9). As expected, 90 mol% of
3
capronitrile 2 and 5 mol% of hexanoic acid 1 were obtained. The full conversion was attained in
ACS Paragon Plus Environment
1
6

Page 17 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
about 6h. However, a kinetic study as the function of the concentration of N-acetylhexanamide 3
clearly shows that the reaction rate is proportional to its initial concentration (Figure 9). Among
other conclusions, discussed later, this shows that the mechanism is intramolecular.
O
O
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
b
N
H
100
a
3
O
O
80
60
40
20
0
O
H
N
N
H
O
3
3
products
conc.
3
2
1
0
.07 M
0.15 M
.30 M
H
O
0
O
N
O
O
H
N
acetimidic hexanoic anhydride 10
acetic hexanimidic anhydride 11
0
50
100
150
200
250
300
350
400
OH
O
Time (min)
N
+
N
C
C
+
O
HO
1
4
2
Figure 9: a) Rearrangement of N-acetylhexanamide 3 as the function of time and concentration.
Points represent experimental data and lines represent model output (vide infra). Reaction
conditions: 200°C, N-acetylhexanamide 3 0.07 M to 0.3 M in reagent-grade acetonitrile, 0.04 M
of InCl , 15 bar (autogenous pressure) and b) Structure of the molecules identified in the reaction
3
mixture and proposed mechanism applied to the rearrangement of N-acetylhexanamide 3.
Intermediates 10 and 11 in square brackets have never been observed.
The major reaction intermediate, N-acetylhexanamide 3 (i.e. imide C), being assessed, we next
moved to the computational examination of the reaction mechanism on the identity reaction with
R = Me (Figure 8). Applying the microreversibility principle, the only transformation of B to C
needs to be examined (Figure 10). Two uncatalyzed reaction pathways were located. A first one is
a concerted 2+2 cyclic mechanism and directly yields a tautomeric form of C from the starting
ACS Paragon Plus Environment
1
7

ACS Catalysis
Page 18 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
#
reactant. The activation barrier is very high, as expected from a 2+2 pericyclic reaction (E =
-1
5
9.7 kcal mol ). The second pathway takes place in two steps through the formation of A, as
proposed in Figure 8, followed by a four membered-ring Mumm rearrangement. The first step
features the concerted formation of the C-O bond and the transfer of the acidic H from the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-1
carboxylic acid to the forming imine group. The activation energy is about 25 kcal mol and this
step is entropically disfavored (associative step). The second step exhibits similar energy relative
to the starting material and is thus expected to take place in similar conditions than the first one,
without significant accumulation of A. These values are consistent with a reaction that can be
carried out without catalytic activation in our conditions (see Figure 3). It is noteworthy to mention
that the reaction has been observed in supercritical acetonitrile (350°C, 65 bar) in the absence of a
3
2
catalyst. This is confirmed computationally, as in these conditions the Gibbs free energy for TS1
-1
and TS2 (39.4 and 42.4 kcal mol respectively above the starting reactant respectively) allows to
4
predict that the reaction should take place around 10 times faster than in the conditions of Figure
3
(see Supplementary Information for more details).
ACS Paragon Plus Environment
1
8

Page 19 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
O
1
.67
H
O
R
TS1'
9.7
1
.82
1
.79
NH
5
C
1.20
(76.4)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
TS1
TS2
O
24.7
HO
R
22.5
(41.2)
(38.5)
N
6
.7
24.3)
(
4.1
1.4
(
21.0)
-15.9
2.2)
0.0
0.0)
(
18.5)
(
-4.4
(
(4.9)
R
R
O
O
H
O
O
R
R
R
N
B
N
R
O
O
H
O
OH
O
O
H
H
O
O
N
H
N
N
C
N
C
1.66
1.30
A
1.83
C
1.90
1
.18
.31
1.29
1
1.19
-1
Figure 10: Reaction pathways for formation N-acetylamide C. Energies (kcal mol ) are computed
-1
with respect to separated reactant for R = Me (Gibbs Free Energies (kcal mol ) are given below
in parenthesis) and structure of the three localized transition states (distances in Å).
We next examined the catalysis using InCl . First bonding of both acetonitrile and acetic acid to
3
-1
InCl are largely exothermic (by -27.3 and -26.3 kcal mol respectively), contrasting with the
3
-1
coordination of a second molecule of acetonitrile (- 6.8 kcal mol , associated to a positive Gibbs
Free Energy ). Considering these results and the large excess in acetonitrile, it can be considered
that InCl is fully coordinated to MeCN, the obtained complex being the starting material of the
3
reaction. This is fully consistent with the fact that the rate for consumption of the starting
carboxylic acid is proportional to the amount of InCl (Figure 3 insert), which should be totally
3
converted to Me˗CN--InCl3.
ACS Paragon Plus Environment
1
9

ACS Catalysis
Page 20 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Both the stepwise and the pericyclic mechanisms could be localized in a catalytic version. In the
pericyclic mechanism, InCl coordinates and activates the C=O double bond, which becomes more
3
reactive than the C-OH one (compare Figure 10 and Figure 11). The activation barrier for
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-1
formation of the tautomer of C is diminished by nearly 7 kcal mol but remains significantly
higher than the stepwise mechanism. In the latter, the formation of A takes place in through a six-
centered ring as in the non-catalyzed version. Coordination of InCl to the acetonitrile nitrogen
3
slightly activates the CN bonds, so that the energy barrier for this first step is lowered to 17
-1
kcal mol . Evolution of A through an InCl -catalyzed Mumm rearrangement was also located,
3
-1
with slightly lower energy than in the non-catalyzed mechanism (19.0 vs 22.5 kcal mol ), this
…
time exhibiting In O=C binding. C has then obtained through decoordination of InCl thanks to
3
the excess of acetonitrile as the reaction is found to be athermic (see the last step in Figure 11).
This decoordination is confirmed experimentally by the formation of close to 25 mol% of imide
3, which is larger than the amount of InCl introduced (10mol%).
3
Comparison of the theoretical and experimental values for activation energies is carried out using
4
2
the energetic span model. The starting material (solvated InCl + carboxylic acid) is found to be
3
the determining state for intermediates, whereas the Mumm rearrangement is found to be the rate-
determining transition state (See Supplementary Information). This is in line with the first order
fit in carboxylic acid observed for the experimental data (Figure 4 to 8). The energy difference
-1
between these two states (19.0 kcal mol , see Figure 11) is pretty close of the experimentally
measured E value (20.3  2.5 kcal/mol, see Figure 4), considering the intrinsic errors associated
a
with DFT calculations.40
On the other hand, using the reaction pathway of Figure 12 in a reverse manner allows modeling
of the reactivity of N-acetylhexanamide 3 and of the data of Figure 10. The determining
ACS Paragon Plus Environment
2
0

Page 21 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
intermediate is the starting material (solvated InCl + N-acetylamide) and the determining TS for
3
the Mum rearrangement (see Supplementary Information). This is fully consistent with first order
behavior and the independence of the reaction evolution with the concentration of 3.
As a conclusion, the computational study allowed to confirm the two steps mechanism involving
the formation of A followed by a Mumm rearrangement, but also that the intermediate A is not
stable enough to allow its experimental characterization as a reaction intermediate.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
H
1
.42
R
O
In-TS1'
47.8
NH
1.55
1
.97
C
(69.9)
1.20
R
InCl3
R
Cl3In
O
OH
In-TS2
9.0
(41.6)
O
N
In-TS1
7.1
36.0)
1
1
N
(
HO
1.0
(6.7)
-3.7
(18.4)
0.0
(0.0)
-8.7
(16.4)
-16.8
(4.5)
-15.9
(2.1)
R
3
Cl In
O
O
R
O
O
H
R
R
O
R
C
B
N
H
R
O
O
InCl3
MeCN
N
O
OH
InCl3
O
O
N
O
InCl3
N
InCl3
H
+
H
C
N
InCl3
C
N
N
H
C...InCl3
A...InCl3
1.74
.20
1
.01
1
.44
1.95
.28
1
.68
1
1.34
1
Figure 11: Reaction pathways for formation imide C in the case of catalytic activation by InCl3.
-1
Energies (kcal mol ) are computed with respect to separated reactant for R = Me (Gibbs Free
-1
Energies (kcal mol ) are given below in parenthesis) and structure of the three localized transition
states (distances in Å).
ACS Paragon Plus Environment
2
1

ACS Catalysis
Page 22 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Kinetic modeling. The entire reversible mechanism discussed in details before can be
condensed into the reaction steps shown in Scheme 1. This scheme depicts the reaction of hexanoic
acid 1 with acetonitrile ACN to form N-acetylhexanamide 3. The latter can be rearranged into
capronitrile 2 and acetic acid 4 in the second step. By-products, i.e. N-acetylacetamide 5 and
N-hexanoylhexanamide 6 can be formed by the side reaction of acetic acid 4 and acetonitrile ACN
and by the analogous side reaction of hexanoic acid 1 with capronitrile 2.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
OH
O
O
k1
k2
+
C N
O
N
H
ACN
3
1
O
O
O
k3
k4
N
C
+
N
H
HO
3
2
4
OH
O
O
N
k5
C
+
O
N
H
k6
1
6
2
O
O
O
k7
+
C
N
HO
N
5
ACN
k8
H
4
Scheme 1 Proposed reaction steps for the transnitrilation reaction
By considering a constant liquid mixture volume (negligible consumption of acetonitrile ACN
by vaporization to reach the autogenous pressure, as well as a negligible nonideality of the liquid
ACS Paragon Plus Environment
2
2

Page 23 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
mixture) and an ideal batch stirred tank reactor, the following set of ordinary differential equations
(ODE) describing the mass balance is obtained:
d[ ퟏ]
dt
=
=
― k [ ퟏ] × [퐀퐂퐍] + k [ ퟑ] ― k [ ퟏ] × [ퟐ] + k [ ퟔ]
(eq.1)
(eq.2)
(eq.3)
(eq.4)
(eq.5)
(eq.6)
(eq.7)
1
2
5
6
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
d[ ퟑ]
dt
k [ퟏ] × [퐀퐂퐍] ― k [ퟑ] + k [ ퟒ] × [ퟐ]
1
2
4
d[ ퟐ]
dt
= k [ퟑ] ― k [ ퟒ] × [ퟐ] ― k [ퟏ] × [ퟐ] + k [ퟔ]
3
4
5
6
d[ ퟒ]
dt
=
=
=
k [ퟑ] ― k [ퟒ] × [ퟐ] ― k [ ퟒ] × [퐀퐂퐍] + k [ퟓ]
3 4 7 8
d[ ퟓ]
dt
k [ ퟒ] × [퐀퐂퐍] ― k [ퟓ]
7
8
푑[ ퟔ]
푑푡
푘 [ ퟏ] × [ퟐ] ― 푘 [ ퟔ]
5
6
d[ 퐀퐂퐍]
=
― k [ ퟏ] × [ACN] + k [ퟑ] ― k [ퟒ] × [퐀퐂퐍] + k [ퟓ]
1 2 7 8
dt
The mass action law was assumed and applied for these elementary steps. The eight rate
constants were determined by performing a least-squares optimization with the experimental
profiles of the figures 7 and 9 (60 samples and 240 concentration values) using the hybrid “trust-
region” algorithm NL2SOL from Dynafit (Biokin ltd) software.43-44 The best fitting is obtained for
the set of kinetic constants listed in Table 1.
ACS Paragon Plus Environment
2
3

ACS Catalysis
Page 24 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1: Kinetic parameters for InCl catalyzed the transnitrilation reaction
of hexanoic acid at 200°C
3
Entry
k value
std error (%)
-
1
-1
1
2
3
4
5
6
7
8
k1
k2
k3
k4
k5
k6
k7
k8
0.0011 mol .min
0.00571 min-1
0.0318 min-1
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
10
6
-
1
-1
0.00452 mol .min
NA, (fast equilibrium)
NA (fast equilibrium)
0.00217 min-1
9
-
-
12
15
-1
-1
0.00102 mol .min
A very good fit is obtained for the main reagents and products (hexanoic acid 1, capronitrile 2
and N-acetylhexanamide 3) concentration profiles (Figures 2, 3, 6, 7 and 9). A low relative root
means square deviation of 1.2% is obtained. However, the curve fitting is poor for
N-hexylhexanamide 6. This could be explained by the lack of experimental precision of the
concentration of N-hexylhexanamide 6, which is a minor intermediate at the beginning of nearly
all experiments. However, N-hexylhexanamide 6 is nearly at equilibrium during the duration of
the reaction and k /k ratio of 15 could be calculated. A parity plot representing the adequation
6
5
between the experimental data and the model data for species concentrations is given in Figure 12
See also Supplementary Information). The good fitting obtained is in line with the reliability of
the proposed mechanism.
(
ACS Paragon Plus Environment
2
4

Page 25 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
00
5
.0 M hexanoic acid 1
9
8
7
6
5
4
3
2
0
0
0
0
0
0
0
0
2.0 M hexanoic acid 1
0.5 M hexanoic acid 1
0.3 M N-acetylhexanamide 3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
10
0
0
10
20
30
40
50
60
70
80
90
100
Experimental conversion (%)
Figure 12: Parity plot of conversion for hexanoic acid 1 for a concentration of 0.5 to 5.0 M and
conversion for N-acetylhexanamide 3; lines indicate a deviation of 10% from the predicted values
using the kinetic model.
Scope and limitations. Having examined the reaction within the framework of a simple
theoretical model, we wondered about the extension of the scope of the reaction. The
transnitrilation reaction was thus applied to different carboxylic acids (Table 2). Substitution at the
-position of the carboxylic acid (i.e. 2-ethylhexanoic acid 12) does not influence the time to reach
equilibrium Table 2, entry 2) but the amount and the distribution of the intermediates as the
function of time are different. A larger amount of imide is formed starting from hexanoic acid 1
(i.e. a maximum of 22 mol% after 10 min) compare to the amount of imide starting from 2-
ethylhexanoic acid 12 (i.e. 10 mol% after 30 min) but the corresponding nitrile, 2-
ethylhexanenitrile 13 is quantitatively obtained in 7 h. For more crowded cyclohexanecarboxylic
acid 14 (Table 2, entry 3), longer reaction time (14 h) is necessary to reach total conversion
(
> 99%). Nonetheless, a disappointing yield of 86% of cyclohexanecarbonitrile 15 was obtained.
ACS Paragon Plus Environment
2
5

ACS Catalysis
Page 26 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Up to 14 mol% of the corresponding imide, i.e. N-acetylcyclohexanecarboxamide, is still present
in the reaction mixture after 14 h at 200°C in the presence of 10 mol% of InCl . We were pleased
3
to see that oleic acid 16 could be quantitatively transformed into oleonitrile 17 under those
conditions (Table 2, entry 4) and from NMR spectra the double bond remains unaffected by the
reaction conditions. Phenylacetic acid 18 is efficiently transformed into 2-phenylacetonitrile 19
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(62 % yield after 3 h, Table 2, entry 5). However, the thermal stability of 19 seems to be low under
those conditions since by-products appear above 3 h reaction time.
Benzoic acid 20 was less reactive under those conditions. Less than 70% of benzonitrile 21 were
obtained after 7h (Table 2, entry 6). 14h were needed to reach complete conversion and
quantitative yield of benzonitrile 21. It should be noticed that with benzoic acid 20, only trace of
imide, i.e. N-acetylbenzamide, could be detected as the function of time. The reaction conditions
seem to be incompatible with chloro or nitro substituted benzoic acid since several by-products
are obtained (Table 2, entries 7-8). However, 3,4-dihydroxybenzoic acid 24 and
3
,4-dimethoxybenzoic acid 26 are efficiently transformed into the corresponding nitrile Table 2,
entries 9-10). Transnitrilation of furan-2-carboxylic acid 28 results in the destruction of molecular
3
2
structure (Table 2, entry 11), whereas as expected from the literature, furan-3-carboxylic acid 29
could be slowly selectively transformed into the corresponding nitrile 30 (Table 2, entry 12).
ACS Paragon Plus Environment
2
6

Page 27 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
Table 2: Substrate scope
Entry
Carboxylic acid
Nitrile
Conversion (%) a) Yields (%)a)
O
N
C
1
1
2
> 99 %
> 97% b)
OH
O
N
C
2
3
OH 12
> 99 %
> 97% b)
1
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
64 % b)
86% c)
N
C
85%
OH 14
1
5
> 99%
O
N
C
4
5
6
OH
> 99%
> 97% b)
17
16
88%
> 99%
74%
62% d)
O
N
1
9
b), e)
C
OH 18
O
70% b)
N
C
OH
> 97% c)
2
1
> 99%
20
Cl
O
7
8
9
OH
_
> 50%
_b), e)
22
O
_ b), e)
OH
23
_
> 90%
> 99%
O N
2
O
N
HO
C
HO
> 97% d)
OH
25
24
HO
HO
O
73% b)
N
93%
MeO
C
MeO
1
0
OH
> 99%
> 97% c)
2
7
26
MeO
MeO
OH
11
O
_
> 90%
_ b), e)
28
O
O
23% b)
49% c)
N
42%
71%
C
1
2
OH
3
0
29
O
O
Reaction conditions: 200°C, 0.5M of carboxylic acid in reagent-grade acetonitrile, 10 mol% of
a)
InCl , 15 bar (autogenous pressure). Yields determined by GC/FID using tetradecane as an
3
b)
c)
d)
e)
internal standard. Reaction time 7h. Reaction time 14h. Reaction time 3h. Unidentified
by-products are formed resulting from the degradation of the reactant and/or the product.
ACS Paragon Plus Environment
2
7

ACS Catalysis
Page 28 of 33
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
CONCLUSION
In conclusion, we have shown that InCl is an interesting, robust and easily handled catalyst able
3
to effectively transform carboxylic acid into the corresponding nitrile using reagent grade
acetonitrile and relatively low temperature (200°C) in a classical stirred tank reactor (autogenous
pressure of 15 bar). This equilibrated reaction furnished the nitriles in nearly quantitative yield
with a minimum amount of by-products, i.e. acetic acid. A simple and reversible mechanism was
proposed for this reaction, which was validated by kinetics measurement on intermediates and by
DFT calculations of the reaction intermediates and reaction mechanisms.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ASSOCIATED CONTENT
Supporting Information.
General Information
Synthesis of nitriles and N-acetylhexanamide 3
Computational section, general Information
Computation of the equilibrium constants and standard Gibbs free energies
Structures associated with figures 5
Structures associated with figures 10
Structures associated with Figure 11
Evaluation of the reaction conditions on the non-catalyzed reaction
Application of energetic span model to the InCl catalyzed system
3
Parity plot between kinetic model and experimental data n relative concentration
Correlation Matrix
References
ACS Paragon Plus Environment
2
8

Page 29 of 33
ACS Catalysis
1
2
3
4
5
6
7
8
9
AUTHOR INFORMATION
Corresponding Authors
*
*
Email: lva@lgpc.cpe.fr
Email: afr@lgpc.cpe.fr
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACKNOWLEDGMENTS
AH was a master student of Master International “Synthesis, Catalysis and Sustainable
Chemistry”, Université Lyon. Authors would like to thanks program Erasmus Mundus that
supported AB.
ACS Paragon Plus Environment
2
9