Efficient dehydration of primary amides to nitriles catalyzed by phosphorus-chalcogen chelated iron hydrides

Article abstract of DOI:10.1002/aoc.5337

A series of phosphorus-chalcogen chelated hydrido iron (II) complexes 1–7, (o-(R'₂P)-p-R-C₆H₄Y)FeH (PMe₃)₃ (1: R = H, R' = Ph, Y = O; 2: R = Me, R' = Ph, Y = O; 3: R = H, R' = ⁱPr, Y = O; 4: R = Me, R' = ⁱPr, Y = O; 5: R = H, R' = Ph, Y = S; 6: R = Me, R' = Ph, Y = S; 7: R = H, R' = Ph, Y = Se), were synthesized. The catalytic performances of 1–7 for dehydration of amides to nitriles were explored by comparing three factors: (1) different chalcogen coordination atoms Y; (2) R' group of the phosphine moiety; (3) R substituent group at the phenyl ring. It is confirmed that 5 with S as coordination atom has the best catalytic activity and 7 with Se as coordination atom has the poorest catalytic activity among complexes 1, 5 and 7. Electron-rich complex 4 is the best catalyst among the seven complexes and the dehydration reaction was completed by using 2 mol% catalyst loading at 60 °C with 24 hr in the presence of (EtO)₃SiH in THF. Catalyst 4 has good tolerance to many functional groups. Among the seven iron complexes, new complexes 3 and 4 were obtained via the O-H bond activation of the preligands o-ⁱPr₂P(C₆H₄)OH and o-ⁱPr₂P-p-Me-(C₆H₄)OH by Fe(PMe₃)₄. Both 3 and 4 were characterized by spectroscopic methods and X-ray diffraction analysis. The catalytic mechanism was experimentally studied and also proposed.

Full text of DOI:10.1002/aoc.5337

Received: 30 July 2019
DOI: 10.1002/aoc.5337
Revised: 20 September 2019
Accepted: 4 October 2019
F U L L P A P E R
Efficient dehydration of primary amides to nitriles
catalyzed by phosphorus-chalcogen chelated iron hydrides
Kai Li1 | Hongjian Sun1
Shangqing Xie1 | Xiaoyan Li1
| Wenjing Yang1 | Yajie Wang1
| Olaf Fuhr2 | Dieter Fenske2
|
1School of Chemistry and Chemical
A series of phosphorus-chalcogen chelated hydrido iron (II) complexes 1–7,
(o-(R'2P)-p-R-C6H4Y)FeH (PMe3)3 (1: R = H, R' = Ph, Y = O; 2: R = Me,
Engineering, Key Laboratory of Special
Functional Aggregated Materials, Ministry
of Education, Shandong University,
Shanda Nanlu 27, Jinan, 250100, People's
Republic of China
2Institut für Nanotechnologie (INT) und
Karlsruher Nano-Micro-Facility (KNMF),
Karlsruher Institut für Technologie (KIT),
Hermann-von-Helmholtz-Platz 1, 76344,
Eggenstein-Leopoldshafen, Germany
i
i
R' = Ph, Y = O; 3: R = H, R' = Pr, Y = O; 4: R = Me, R' = Pr, Y = O; 5:
R = H, R' = Ph, Y = S; 6: R = Me, R' = Ph, Y = S; 7: R = H, R' = Ph, Y = Se),
were synthesized. The catalytic performances of 1–7 for dehydration of amides
to nitriles were explored by comparing three factors: (1) different chalcogen
coordination atoms Y; (2) R' group of the phosphine moiety; (3) R substituent
group at the phenyl ring. It is confirmed that 5 with S as coordination atom
has the best catalytic activity and 7 with Se as coordination atom has the
poorest catalytic activity among complexes 1, 5 and 7. Electron-rich complex 4
is the best catalyst among the seven complexes and the dehydration reaction
Correspondence
Xiaoyan Li, School of Chemistry and
Chemical Engineering, Key Laboratory of
Special Functional Aggregated Materials,
Ministry of Education, Shandong
University, Shanda Nanlu 27, Jinan
250100, China.
ꢀ
was completed by using 2 mol% catalyst loading at 60 C with 24 hr in the
presence of (EtO)3SiH in THF. Catalyst 4 has good tolerance to many func-
tional groups. Among the seven iron complexes, new complexes 3 and 4 were
obtained via the O-H bond activation of the preligands o-iPr2P(C6H4)OH and
o-iPr2P-p-Me-(C6H4)OH by Fe(PMe3)4. Both 3 and 4 were characterized by
spectroscopic methods and X-ray diffraction analysis. The catalytic mechanism
was experimentally studied and also proposed.
Email: xli63@sdu.edu.cn
Funding information
NSF China, Grant/Award Number:
21572119
K E Y W O R D S
amide, dehydration, iron hydride, nitrile, phosphorus-chalcogen ligand
1 | INTRODUCTION
C-N bonds.[6] The combination of tris (pen-
tafluorophenyl)borane and nitriles lead to the Lewis
Nitriles are important chemical raw materials and are
widely used in daily life such as synthetic fibers,[1] syn-
thetic fragrance,[2] bioactive substances[3] and liquid crys-
tal materials.[4] Nitriles are also important intermediates
in organic chemical transformations because nitrogen
atom can be introduced in production molecules. For
instance, polyamides could be prepared via the reaction
of carbonyl compounds with nitriles.[5] The reactions of
acid/base adduct with a substantial increase in the C ≡ N
bond strength.[7] Tetrazoles and oxadiazolines could also
be synthesized using coordinate nitriles through [2 + 3]
cycloaddition.[8,9]
However, how to prepare nitriles efficiently is still a
challenge for synthetic chemists. Because amides are
cheap and easy to obtain, dehydration of amides to
nitriles has become one of the effective methods in this
field. Traditionally, some harsh dehydrating reagents
such as TiCl4,[10] SOCl2[11] or POCl3[12] were used for this
aziridine
with
the
coordinated
nitriles,
e.g.
[PtCl2(RCN)2], gave rise to (aziridino)amines by forming
Appl Organometal Chem. 2020;e5337.
wileyonlinelibrary.com/journal/aoc
© 2020 John Wiley & Sons, Ltd.
1 of 9
https://doi.org/10.1002/aoc.5337

2 of 9
LI ET AL.
process. However, on the one hand, these substances
themselves are not environmentally friendly; on the other
hand, the reactions with these substances might bring
some harmful byproducts. In addition, the condition of
most of the transformation is not mild. For example, with
chlorosulfonic acid as dehydrating agent the aldoximes
could be obtained from Beckmann rearrangement.[13] The
EtOPOCl2/DBU system was confirmed as a convenient
dehydrating agent for the conversion of primary amides to
nitriles.[14] The amides could also be dehydrated to nitriles
via simultaneous action of triphenylphosphine, carbon tet-
rachlorides and trimethylamine.[15]
In past few years, we found that some hydrido iron
complexes synthesized in our group could catalyze the
dehydration reactions of amides to afford nitriles.[22,23] 15
Amide substrates were catalytically dehydrated to the
corresponding nitriles by hydrido thiophenolato iron (II)
complexes [cis-Fe(H)(SAr)(PMe3)4] using (EtO)3SiH as an
efficient dehydrating agent in the yield up to 93%.[22] It
was confirmed that the hydrido silyl iron complex, [(o-
Ph2PC6H4SiMe2)Fe(H)(PMe3)3], is a good dual catalyst
for both hydrosilylation of carbonyl complexes and dehy-
dration of benzamides.[23] In 2015, we prepared [P, S]-
chelated iron hydrides (5 and 6, SCHEME 1) and found
that they could be used as catalysts for hydrosilylation of
aldehydes and ketones.[24] The [P, Se]-chelated iron
hydride (7, Scheme 1) was also obtained via Se-H bond
activation.[25] However, both [P, S]- or [P, Se]-chelated
iron hydrides for dehydration of amides to nitriles as cat-
alysts have not been studied. The [P, O]-chelated iron
hydrides (1 and 2, Scheme 1) were published by Klein in
2004.[26] The catalytic activity of complexes 1–2 for dehy-
dration of amides to nitriles has also not been explored.
In order to further expand the types of catalysts and com-
pare electronic and steric effect of the hydrido iron (II)
complexes with different chalcogen coordination atoms
on the dehydration of amides to nitriles, in this paper
seven phosphorus-chalcogen chelated hydrido iron (II)
complexes 1–7 were synthesized and the catalytic perfor-
mances of 1–7 for dehydration of amides to nitriles were
explored. Among them iron hydrides (3 and 4, Scheme 1)
were new and were completely characterized.[26]
In recent years, large efforts have been put into the
development of transition-metal catalyzed dehydration of
amides (e.g. Ru, V, Pd, U).[16] In 2008, Nagashima
reported the first example of a transition-metal-catalyzed
Si-H bond activation leading the dehydration of amides
to afford the nitriles.[16a] After two years Kaneda group
designed a monomeric vanadium oxide on hydrotalcite
as an effective heterogeneous catalyst for the highly effi-
cient dehydration of amides to related nitriles.[16b] For a
decade, replacing precious metals with earth-abundant
metals (e.g. Fe, Co, Ni) is getting more and more atten-
tion. Iron, being the most abundant transition metal and
displaying low toxicity, is an effective alternative. In
2009, the first iron-catalyzed dehydration of primary
amides to nitriles with (EtO)2MeSiH as a dehydrating
agents was developed by Beller. Beller considered that
the elimination of silyl ether is the key step in the cata-
lytic mechanism.[17] In 2012, Beller group also found that
[Et3NH][HFe3(CO)11] could catalyze the dehydration of
benzamide to phenylnitrile in 90% yield while benzamide
could be reduced to benzamine with two catalysts
[Et3NH][HFe3(CO)11] /Fe (OAc)2 in 70% yield.
(EtO)2MeSiH was used for both catalytic systems.[18] In
the same year, Enthaler used the simple zinc salt [Zn
(OTf)2] as a Lewis acid catalyst to complete the reaction
of primary amides into the corresponding nitriles with
MSTFA (N-methyl-N-(trimethylsilyl) trifluoroacetamide)
as a dehydrating agent.[19] Before this, his group also
introduced FeCl2Á4H2O/MSTFA system to dehydrate the
primary amides to nitriles.[20] Some carbonyl iron com-
pounds such as [Fe3(CO)12] and [Fe (CO)5] could also be
used to dehydrate amides to amines in the presence of
silanes.[21]
2 | RESULTS AND DISCUSSION
2.1 | Synthesis of [P, Y]-chelated (Y = O,
S and se) hydrido iron (II) complexes1–7
Complexes 1 and 2 were prepared according to the
method by Klein.[26] Complexes 5 and 6 can be obtained
with the method reported in the reference.[24] We dis-
closed the synthesis of complex 7 in 2017.[25] The ligands
ortho-diisopropylphosphinophenols were synthesized
according to the literature.[27] The combination of Fe
(PMe3)4 with ortho-diisopropylphosphinophenols in
diethyl ether afforded complexes 3 and 4 (Scheme 2).
SCHEME 1 Seven catalysts

HOMOGENEOUS CATALYSIS
3 of 9
SCHEME 2 Synthesis of 3 and 4
Complexes 3 and 4 were fully characterized by IR, NMR
(1H, 31P and 13C) and X-ray diffraction analysis.
In the IR spectrum of 3 the typical Fe-H vibration
appeared at 1907 cm−1. This suggests that the O-H bond
was cleavaged. The strong vibration of coordinated PMe3
1
appeared at 935 cm−1. In the H NMR spectrum of 3 the
hydrido resonance was situated at −9.54 ppm as a dddd
peak (2JPdH = 79.3 Hz, 2JPcH = 73.9 Hz, 2JPbH = 52.3 Hz,
2JPaH = 37.5 Hz)[26] caused by the coupling with four P
atoms. In the 31P NMR spectrum of 3 (Figure 1) four sets
of signals were identified at 13.1 (dt), 21.3 (ddd), 39.0 (dt)
and 78.2 (ddd) ppm in the integral ratio of 1: 1: 1: 1. This
implies that the four P atoms are in different chemical
environments. The two dt peaks belong to Pa and Pc
atoms while the two ddd peaks were caused by Pd and Pb
atoms.[26] For complex 4 the similar spectroscopic results
were obtained.
The molecular structures of 3 and 4 were determined
by single crystal X-ray diffraction analysis (Figure 2 and
3). In Figure 2 the Fe atom is centered in a distorted octa-
hedral coordination geometry. Three trimethylphosphine
ligands are in fac-mode. Because the hydrido H has the
strongest trans-influence and O atom has the weakest
trans-influence, Fe1-P4 (2.2403 Å) is the longest and
Fe1-P3 (2.1752 Å) is the shortest Fe-P bond among the
four Fe-P bonds. The bite angle P1-Fe1-O1 is 83.68(5)o.
These structural characteristics are similar with those of
FIGURE 2 Molecular structure of 3. The thermal ellipsoids
are displayed at the 50% probability level, and most of the hydrogen
atoms are omitted for clarity. Selected bond lengths (Å) and angles
[ꢀ]: Fe1-P1 2.2319(7), Fe1-P2 2.2157(7), Fe1-P3 2.1752(7), Fe1-P4
2.2403(7), Fe1-O1 2.0382(16), Fe1-H 1.55(4); P1-Fe1-O1 83.68(5),
O1-Fe1-P2 77.31(5), P2-Fe1-P3 98.51(3), P3-Fe1-P1 98.71(3),
P4-Fe1-H 175(1)
the derivative iron complexes.[26] The Fe-H bond (1.55(4)
Å) is slightly longer than a normal Fe-H bond (~
1.42 Å).[28] Complex 4 has a similar molecular structure
(Figure 3) with 3 but Fe1-H (1.46(3) Å) in 4 is a little bit
shorter than that in 3.
2.2 | Catalytic activities of 1–7 for
dehydration of amides
Complexes 1–7 as catalysts were used in dehydration
of amides to nitriles (Scheme 3). Initially, p-
chlorobenzamide was selected as a template substrate
FIGURE 1 31P NMR spectrum of 3

4 of 9
LI ET AL.
conversion was reached (entry 10, Table 1). Therefore,
it can be concluded that the optimized temperature is
60 ꢀC. When the reaction time was reduced from
24 hr to 22 or 20 hr, the conversion was also reduced
(entries 11 and 12, Table 1). Lower conversion (67%)
was detected when the catalyst loading was reduced to
1 mol% (entry 13, Table 1). In order to explore the
effect of the amount of silanes on the conversion,
(EtO)3SiH in different molar ratios were added to the
catalytic mixture (entries 14–16, Table 1). It can be
seen that the ratio of 1: 3 (substrate: silane) is the best
and the reaction temperature is crucial. We also found
that (EtO)3SiH is the best dehydrating agent for this
catalytic system in comparing with Et3SiH, Ph3SiH,
Ph2SiH2 and PMHS (polymethyl hydrosiloxane) (entries
17–22, Table 1). When the catalytic reaction was con-
ducted in different media (pentane, diethyl ether, tolu-
ene, CH3CN, THF, DMSO, DMF and benzene), it was
found that THF was the best solvent (entries 21–27,
Table 1). Up to now, the optimized catalytic conditions
can be summarized as follows: 2 mol % of catalyst 4
at 60 ꢀC in the presence of 3 equiv of (EtO)3SiH in
THF to react 24 hr (entry 9, Table 1). The control
experiment under the optimized catalytic conditions
showed that there was no conversion without complex
4 (entry 28, Table 1).
Under the optimized catalytic conditions, the scope of
the substrates was expanded (Table 2). The aromatic sub-
strates with special functional groups, such as halogeno-,
amino-, nitro-, phenolic hydroxyl and C=C group, have
good isolated yields (72—93%). Some substrates with
electron-donating group, such as methyl or methoxy
group, could reach good yields (69—84%). Heterocyclic
substances, 2-pyridinecarboxamide and thiophene-2-car-
boxamide, could only be dehydrated in 61 and 57% yields.
In the catalytic process, both nitro- and C=C groups
could not be reduced and remained unchanged. Interest-
ingly, when a group was attached to the ortho-position of
the amides group, such as o-chlorobenzamide and o-
methylbenzamide, the reaction yields are significantly
poorer than those of the substrates with the same group
at the meta- or para-position, such as p-chlorobenzamide
and m-methylbenzamide. We consider that the steric
effect plays an important role for these results due to the
large steric hindrance of the two isopropyl group in 4.
The steric effect hinders the interaction of the hydrido
hydrogen with the amino group of the amides.
FIGURE 3 Molecular structure of 4. The thermal ellipsoids
are displayed at the 50% probability level, and most of the hydrogen
atoms are omitted for clarity. Selected bond lengths (Å) and angles
[ꢀ]: Fe1-P1 2.2392(5), Fe1-P2 2.2249(5), Fe1-P3 2.2251(5), Fe1-P4
2.1710(5), Fe1-O1 2.0209(12), Fe1-H 1.46(3); P1-Fe1-O1 83.68(4),
O1-Fe1-P2 76.10(4), P2-Fe1-P4 97.60(2), P4-Fe1-P1 98.603(19),
P3-Fe1-H 174.5(10)
to screen the catalytic conditions (Table 1). Under the
same catalytic conditions, 2 is a better catalyst than 1
(entries 1 and 2, Table 1). This might be interpreted
with the introduction of para-methyl group in 2. When
the diphenyl groups were replaced by the diisopropyl
groups, significantly better catalytic activities of 3 and
4 were observed in comparison with those of 1 and 2
(entries 3 and 4, Table 1). If the coordination O atom
in 1 or 2 was replaced by S atom (5 or 6), the conver-
sions are not satisfactory comparing with the results of
3 and 4 (entries 5 and 6, Table 1) but better than
those of 1 and 2. Complex 7 with Se-coordination has
the lowest conversion of 56% (entry 7, Table 1). It can
be deduced that the strong coordination ability makes
this complex more stable. In comparison with the
results of entries 1, 5 and 7, it can be concluded
hydrido iron complex 5 with S as coordination atom
has the best catalytic activity while hydrido iron com-
plex 7 with Se as coordination atom has the poorest
catalytic activity among the three complexes 1, 5 and
7. From the results of entry 1 to entry 7 it can be con-
cluded 4 is the best catalyst among the 7 complexes
under the same catalytic conditions. When the reaction
with 4 as catalyst was conducted at different tempera-
tures, it was found that 4 is a better catalyst than 3
(entries 8 and 9, Table 1). When the reaction tempera-
ture was further lowered to 50 ꢀC, only 65%
SCHEME 3 Catalytic dehydration of amides to nitriles

HOMOGENEOUS CATALYSIS
5 of 9
TABLE 1 Optimization of catalytic reaction conditionsa
Entry
1
Cat
1
2
3
4
5
6
7
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
Loading (mol)
silane
solvent
THF
Temp.(ꢀC)
80
Time(h)
24
Conv.b(%)
2%
2%
2%
2%
2%
2%
2%
2%
2%
2%
2%
2%
1%
2%
2%
2%
2%
2%
2%
2%
2%
2%
2%
2%
2%
2%
2%
0
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiHc
(EtO)3SiHd
(EtO)3SiHd
Et3SiH
64
81
95
96
79
84
56
89
94
65
85
69
67
33
95
57
0
2
THF
80
24
3
THF
80
24
4
THF
80
24
5
THF
80
24
6
THF
80
24
7
THF
80
24
8
THF
60
24
9
THF
60
24
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
THF
50
24
THF
60
22
THF
60
20
THF
60
24
THF
60
24
THF
60
24
THF
50
24
THF
60
24
Ph3SiH
THF
60
24
25
7
Ph2SiH2
THF
60
24
PMHS
THF
60
24
51
71
85
90
61
89
57
88
0
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
(EtO)3SiH
Pentane
Et2O
Toluene
CH3CN
DMSO
DMF
Benzene
THF
60
24
60
24
60
24
60
24
60
24
60
24
60
24
60
24
aCatalytic reaction condition: p-chlorbenamide (1.0 mmol), hydrosilane (3.0 mmol) and 1–7 (2% mmol) in 2 ml THF, TꢀC, t h.
bDetermined by GC with dodecane as internal standard.
chydrosilane (2.0 mmol).
dhydrosilane (4.0 mmol).
3 | MECHANISM DISCUSSION
reaction, Path a and Path b. For Path a, 4 reacts with
acetimidic acid, the tautomer of amide, to give rise to
intermediate A with the release of hydrogen gas. Inter-
mediate A interacts with (EtO)3SiH affords intermedi-
ate B with the formation of 4. The “acidic” H of B
reacts with the “basic” H of 4 to produce intermediate
C with the formation of H2. β-Elimination of C
delivers intermediate D with the production of nitrile.
Finally, the reaction of D with (EtO)3SiH gives rise to
disilyl ether with the recovery of 4. Path b is another
alternative. At first, 4 reacts with (EtO)3SiH to afford
intermediate E with the release of H2. E interacts with
amide to delivers B with the formation of 4. The
Beller proposed a mechanism for dehydration of pri-
mary amides involving the formation of disilyl imidate
and the elimination of disiloxane steps.[17] This pro-
posal becomes a presentative mechanism for this kind
catalytic transformations. In 2018, Buchwald reported
an overview of amide-to-nitrile conversion supported
by theoretical calculation and single crystal X-ray dif-
fraction of intermediate.[29] Inspired by these litera-
tures, we proposed a process of tautomerization of
primary amides at the beginning of this reaction
(Scheme 4). There are two possible paths for this

6 of 9
LI ET AL.
TABLE 2 Products of dehydration of primary amides to nitrilesa,b
aCatalytic reaction conditions: amide (1 mmol), (EtO)3SiH (3 mmol), Catalyst 4 (2 mol%) in 2 ml THF, 60 ꢀC, 24 hr.
bIsolated yield.
SCHEME 4 Proposed mechanism

HOMOGENEOUS CATALYSIS
7 of 9
combination of 4 and B produces intermediate C and
H2. The next two steps of Path b are the same as
those in Path a.
KBr disks, were recorded on a Bruker ALPHA FT-IR
instrument. 1H, 13C, 31P spectra were completed by using
Bruker Avance 300 MHz spectrometers. GC was recorded
on a Fuli 9790 instrument. Melting points were measured
in capillaries sealed under N2. Fe (PMe3)4 was synthe-
sized by literature.[30] All the amides and silanes were
purchased and used without further purification.
In order to explore the mechanism of the reaction
experimentally, stoichiometric reaction of complex 4 with
4-methoxybenzamide or (EtO)3SiH were respectively car-
ried out. When 4 reacted with 4-methoxybenzamide, the
vibration of Fe-H bond disappeared by the detection of
infrared spectra (see SIII). When complex 4 reacted with
(EtO)3SiH, the vibration of Fe-H bond still existed. There-
fore, we consider that Path a is possible. In addition, the
release of H2 was confirmed by GC.
5.2 | Synthesis of 3
A solution of 2-diisopropylphosphinophenol (0.79 g,
3.75 mmol) in 30 ml Et2O was slowly added to a solution
of Fe (PMe3)4 (1.35 g, 3.75 mmol) in 40 ml Et2O at
−78 ꢀC, forming an orange mixture. After stirring at
room temperature for 18 h, the turbid orange solution
was evaporated under reduced pressure and the red
powder was extracted with n-pentane and diethyl ether.
Complex 3 (1.15 g, 2.33 mmol) was obtained as red crys-
tals in the yield of 62%. M.p.: 144–145 ꢀC. IR (Nujol mull,
4000–400 cm−1): 3053, 3011 ν (ArH), 1907 ν (Fe–H), 935 ν
4 | CONCLUSIONS
In this work, two new ortho-diisopropylphosphino
phneolato iron (II) hydrides 3 and 4 were prepared and
fully characterized by spectroscopic methods. In order to
compare different chalcogen coordination atoms and the
electronic as well as the steric effects on the catalytic
activities of the iron hydrides, the catalytic performance
of complexes 1–7 for the dehydration of amides was
explored. The catalytic experiments indicate complex 5
with S as coordination atom has the best catalytic activity
and complex 7 with Se as coordination atom has the
poorest catalytic activity among the three complexes 1, 5
and 7. Higher yields could be obtained when diphenyl
groups were replaced by diisopropyl groups in compari-
son the data of 1 and 2 with those of 3 and 4. When the
results of 1 and 2, 3 and 4 as well as 5 and 6 are com-
pared, it could be found that the introduction of methyl
group on the benzene ring can also increase the catalytic
activity of the complex. Among the seven complexes, 7 is
the poorest catalyst while 4 is the best catalyst. We think
that 4 is the most electron-rich complex among them.
With 4 as catalyst this catalytic system has excellent toler-
ance to halogeno-, amino-, nitro-, phenolic hydroxyl,
C=C group. 18 Amide substrates were transformed to the
related nitriles in the isolated yields of 57—93%. A cata-
lytic mechanism was proposed and experimentally
studied.
1
(PMe3). H NMR: (300 MHz, benzene-d6, 300 K/ppm) δ:
2
2
2
−9.54 (dddd, J PdH = 79.3 Hz, J PcH = 73.9 Hz, J
2
PbH = 52.3 Hz, J PdH = 37.5 Hz, FeH, 1H); 1.14 (dd,
J = 12.0, 6.0, PCHCH3, 3H), 1.26 (dd, J = 12.0, 6.0,
PCHCH3, 3H), 1.49 (dd, J = 12.0, 6.0, PCHCH3, 3H), 1.62
(dd, J = 12.0, 6.0, PCHCH3, 3H); 2.38–2.56 (m, PCH
(CH3)2, 2H); 0.98 (d, PMe3, J = 6.0 Hz, 9H), 1.38 (d, PMe3,
J = 6.0 Hz, 18H); 6.60–7.49 (m, ArH, 4H). 31P NMR
(121 MHz, benzene-d6, 300 K/ppm) δ 13.1 (dt, J = 31.5,
24.2 Hz, PaMe3, 1P), 21.3 (ddd, J = 24.2, 40.0, 93.8 Hz,
PbMe3, 1P), 39.0 (dt, J = 40.0, 31.5 Hz, PcMe3, 1P), 78.2
(ddd, J = 24.2, 31.5, 93.8 Hz, Pd iPr2, 1P). 13C NMR
(75 MHz, benzene-d6, 300 K/ppm) δ: 19.8 (t, J = 2.8 Hz,
PCHCH3), 20.9 (d, J = 5.8 Hz, PCHCH3), 21.9 (d,
J = 3.0 Hz, PCHCH3), 22.3 (m, PMe3), 29.7 (dt, J = 30.0,
7.2 Hz, PMe3), 28.5 (s, PMe3), 29.7 (dd, J = 30.0, 7.2 Hz,
PCHCH3), 110.7 (d, J = 4.8 Hz, Ar-C), 118.2 (d,
J = 6.2 Hz, Ar-C), 124.3 (d, J = 7.2 Hz, Ar-C), 124.9 (d,
J = 7.2 Hz, Ar-C), 129.4 (d, J = 2.0 Hz, Ar-C), 177.8 (ddd,
J = 20.8, 7.1, 2.0 Hz, Ar-C). Anal. calcd for C21H46FeOP4
(494.34 g mol−1): C, 51.02; H, 9.38. Found: C, 51.25; H,
9.27.
5 | EXPERIMENTAL
5.3 | Synthesis of 4
5.1 | General procedures and materials
A solution of 2-diisopropylphosphino-5-methylphenol
(0.8 g, 3.61 mmol) in 30 ml Et2O was slowly added to a
solution of Fe (PMe3)4 (1.30 g, 3.61 mmol) in 40 ml Et2O
All air-sensitive reactions were carried out under anhy-
drous and nitrogen atmospheric conditions by a standard
Schlenk technique. Solvents were distilled with sodium
ꢀ
at −78 C, forming an orange mixture. After stirring at
under
nitrogen
atmosphere.
Infrared
spectra
room temperature for 18 hr, the turbid orange solution
was evaporated under reduced pressure and the red
(4000–400 cm−1), as obtained from Nujol mulls between

8 of 9
LI ET AL.
powder was extracted with n-pentane and diethyl ether.
Complex 4 (1.3 g, 2.56 mmol) was obtained as red crystals
anisotropically and all hydrogen atoms except for those
of the disordered solvent molecules were placed using
AFIX instructions. Appropriate restraints or constraints
were applied to the geometry and the atomic displace-
ment parameters of the atoms. CCDC-1897323 (3) and
1865449 (4) contain the supplementary crystallographic
data for this paper. Copies of the data can be obtained
free of charge on application to CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK (fax: (+44)1223–336-033; e-
mail: deposit@ccdc.cam.ac.uk).
ꢀ
in the yield of 71%. M.p.: 124–126 C. IR (Nujol mull,
4000–400 cm−1): 3132 ν (ArH), 1906 ν (Fe–H), 932 ν
1
(PMe3). H NMR: (300 MHz, benzene-d6, 300 K/ppm) δ
−9.53 (dddd, 2J PH = 79.7 Hz, 2J PH = 74.4 Hz, 2J
2
PH = 52.3 Hz, J PH = 37.6 Hz, FeH, 1H), 1.17 (dd,
J = 12.0, 6.0, PCHCH3, 3H), 1.29 (dd, J = 12.0, 6.0,
PCHCH3, 3H), 1.53 (dd, J = 12.0, 6.0, PCHCH3, 3H), 1.65
(dd, J = 12.0, 6.0, PCHCH3, 3H), 2.45 (s, Ar-CH3, 3H),
2.47–2.59 (m, PCHCH3, 2H), 0.99 (d, J = 6.0 Hz, PMe3,
9H), 1.39 (d, J = 6.0 Hz, PMe3, 9H), 1.41 (d, J = 6.0 Hz,
PMe3, 9H), 6.86–7.53 (m, ArH, 3H). 31P NMR (121 MHz,
benzene-d6, 300 K/ppm) δ 13.1 (dt, J = 31.5, 47.2 Hz,
PMe3, 1P), 21.1 (ddd, J = 94.4, 40.0, 24.2 Hz, PMe3, 1P),
38.9 (dt, J = 41.1, 64.1 Hz, PMe3, 1P), 78.2 (ddd, J = 93.2,
ACKNOWLEDGMENTS
We gratefully acknowledge the support by NSF China
No. 21572119.
CONFLICT OF INTEREST
There are no conflicts to declare.
i
33.9, 24.2 Hz, P Pr2, 1P) ppm, 13C NMR (75 MHz,
benzene-d6, 300 K/ppm) δ: 20.9 (s, PCHCH3), 21.9 (d,
J = 2.3 Hz, PCHCH3), 22.3 (m, PMe3), 27.0 (d,
J = 18.2 Hz, PMe3), 28.5 (s, PMe3), 29.6 (dd, J = 30.0,
7.2 Hz, PCHCH3), 117.8 (d, J = 6.8 Hz, Ar-C), 118.3 (d,
J = 4.7 Hz, Ar-C), 123.9 (d, J = 7.2 Hz, Ar-C), 124.4 (d,
J = 7.3 Hz, Ar-C), 129.1 (d, J = 2.4 Hz, Ar-C), 130.6 (s,
Ar-CH3), 176.0 (dd, J = 20.7, 7.1 Hz, Ar-C). Anal. calcd
for C22H48FeOP4 (508.37 g mol−1): C, 51.98; H, 9.52.
Found: C, 51.79; H, 9.61.
FUNDING INFORMATION
National Natural Science Foundation of China,
Grant/Award Number: NSF China No. 21572119.
ORCID
Hongjian Sun https://orcid.org/0000-0003-1237-3771
Xiaoyan Li https://orcid.org/0000-0003-0997-0380
REFERENCES
[1] H. Bogaty, N. R. S. Hollies, J. C. Hintermaier, M. Harrist, Text.
Res. J. 1953, 23, 536.
5.4 | General procedure for the
dehydration of primary amides to nitriles
[2] Q. Z. Yan, D. Y. Kong, M. N. Li, G. H. Hou, G. F. Zi, J. Am.
Chem. Soc. 2015, 137, 10177.
[3] R. Nongrum, G. S. Nongthombam, M. Kharkongor, J. W.
S. Rani, N. Rahman, C. Kathing, B. Myrboh, R. Nongkhlaw,
RSC Adv. 2016, 6, 108384.
[4] M. M. A. Mazza, S. Yamazaki, D. X. Mai, S. Padgaonkar,
S. Peurifoy, A. Goncalves, Y. Wu, Q. Hu, A. M. Scott, Phys.
Chem. Chem. Phys. 2017, 19, 4588.
Under a nitrogen atmosphere, 1 mmol primary amides
was weighed into a 20 ml Schlenk tube containing a
magnetic stirrer. Then, 3 mmol (EtO)3SiH, 1 ml THF
and 2% mmol catalysts were added to the tube. The
ꢀ
tube was stirred under 60 C for 24 hr. The conversion
was determined by GC with n-dodecane as internal
standard. The mixture was extracted with petroleum
ether, dried over anhydrous sodium sulfate and
evaporated under reduced pressure. Column chroma-
tography was used with petroleum ether as developing
solvent.
[5] S. Sharma, M. Tandon, J. W. Lown, J. Org. Chem. 2001, 66,
1030.
[6] R. A. Michelin, M. Mozzon, R. Bertani, F. Benetollo,
G. Bombieri, R. J. Angelici, Inorg. Chim. Acta 1994, 222, 327.
[7] H. Jacobsen, H. Berke, S. Doering, G. Kehr, G. Erker,
R. Froehlich, O. Meyer, Organometallics 1999, 18, 1724.
[8] J. H. Hall, R. Lopez de la Vega, W. L. Purcell, Inorg. Chim.
Acta 1985, 102, 157.
[9] G. Wagner, A. J. L. Pombeiro, V. Kukushkin, J. Am. Chem.
Soc. 2000, 122, 3106.
5.5 | Single crystal X-ray diffraction
[10] W. Lehnert, Tetrahedron Lett. 1971, 12, 1501.
[11] D. Patil, D. Dalal, Syn. Commun. 2013, 43, 118.
[12] B. Rickborn, F. R. Jensen, J. Org. Chem. 1962, 27, 4608.
[13] D. M. Li, F. Shi, S. Guo, Y. Q. Deng, Tetrahedron Lett. 2005,
46, 671.
[14] C. W. Kuo, J. L. Zhu, J. D. Wu, C. M. Chu, C. F. Yao, K.
S. Shia, Chem. Commun. 2007, 301.
[15] R. Appel, R. Kleinstuckück, K. D. Ziehn, Chem. Ber. 1971, 104,
1030.
Bruker Apex II single crystal diffractometer was used
with a Ga Kα radiation (λ = 1.34143) and a CCD area
detector. The structure was solved using the charge-
flipping algorithm, as implemented in the program SUP-
ERFLIP[31] and refined by full-matrix least-squares tech-
niques against F2 (SHELXL)[32] through the OLEX
interface.[33] All non-hydrogen atoms were refined

HOMOGENEOUS CATALYSIS
9 of 9
[16] (a)S. Hanada, Y. Motoyama, H. Nagashima, Eur. J. Org.
Chem. 2008, 2008, 4097. (b)S. Sueoka, T. Mitsudome,
T. Mizugaki, K. Jitsukawa, K. Kaneda, Chem. Commun.
2010, 46, 8243. (c)E. Marzorati, A. Marazzi, S.
I. Maffioli, Org. Lett. 2005, 7, 5237. (d)S. Enthaler,
Chem. A Eur. J. 2011, 17, 9316.
[27] J. Heinicke, R. Kadyrov, M. K. Kindermann, M. Koesling, P.
G. Jones, Chem. Ber. 1996, 129, 1547.
[28] P. Bhattacharya, J. A. Kraus, H. Guan, Organometallics 2011,
30, 4720.
[29] R. Y. Liu, M. Bae, S. L. Buchwald, J. Am. Chem. Soc. 2018, 140,
1627.
[17] S. L. Zhou, D. Addis, S. Das, K. Junge, M. Beller, Chem.
Commun. 2009, 4883.
[18] S. Das, B. Wendt, K. Moeller, K. Junge, M. Beller, Angew.
Chem. Int. Ed. 2012, 51, 1662.
[19] S. Enthaler, S. Inoue, Chem. Asian J. 2012, 7, 169.
[20] S. Enthaler, Eur. J. Org. Chem. 2011, 4760.
[30] H. F. Klein, H. H. Karsch, Chem. Ber. 1977, 110, 2699.
[31] L. Palatinus, G. Chapuis, J. Appl. Cryst. 2007, 40, 786.
[32] G. M. Sheldrick, Acta Crystallogr. 2008, A64, 112.
[33] O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard,
H. Puschmann, J. Appl. Cryst. 2009, 42, 339.
[21] (a)S. L. Zhou, K. Junge, D. Addis, S. Das, M. Beller, Angew.
Chem. Int. Ed. 2009, 48, 9507. (b)Y. Sunada, H. Kawakami,
T. Imaoka, Y. Motoyama, H. Nagashima, Angew. Chem. Int.
Ed. 2009, 48, 9511.
[22] B. Xue, H. Sun, Y. Wang, T. Zheng, X. Li, O. Fuhr, D. Fenske,
Cat. Com. 2016, 86, 148.
SUPPORTING INFORMATION
Additional supporting information may be found online
in the Supporting Information section at the end of this
article.
[23] S. Ren, S. Xie, T. Zheng, Y. Wang, S. Xu, B. Xue, X. Li, H. Sun,
O. Fuhr, D. Fenske, Dalton Trans. 2018, 47, 4352.
[24] B. Xue, H. Sun, X. Li, RSC Adv. 2015, 5, 52000.
[25] H. Zhou, H. Luan, T. Zheng, X. Qi, H. Sun, X. Li, O. Fuhr,
D. Fenske, J. Organomet. Chem. 2017, 853, 107.
How to cite this article: Li K, Sun H, Yang W,
et al. Efficient dehydration of primary amides to
nitriles catalyzed by phosphorus-chalcogen
chelated iron hydrides. Appl Organometal Chem.
2020;e5337. https://doi.org/10.1002/aoc.5337
[26] H. F. Klein, M. Frey, S. Mao, Z. Anorg. Allg. Chem. 2005, 631,
1516.