S. Ren et al.
CatalysisCommunications120(2019)72–75
Fig. 1. Cobalt(III) hydrido complexes 1–3.
synthesis of [CNC]-pincer hydrido cobalt(III) complexes, [(ortho-F4C6-
CH=N-C10H6)Co(III)(H)(PMe3)2] (1), [(2,5-F2C6H2-CH=N-C10H6)Co
(III)(H)(PMe3)2] (2) and [(2,4,5-F3C6H-CH=N-C10H6)Co(III)(H)
(PMe3)2] (3) (Fig. 1), from reactions of polyfluorinated aryl imines with
CoMe(PMe3)4 via selective C–H/C–F bond activation. We explored that
the hydrido cobalt(III) complexes 1–3 as catalysts, (EtO)3SiH as redu-
cing agent for reduction of aldehydes and ketones [28]. Therefore, we
report in this paper the dehydration of primary amides to nitriles cat-
alyzed by [CNC]-pincer hydrido cobalt(III) complexes, (1), (2) and (3)
(Fig. 1) as catalysts using (EtO)3SiH as an efficient reducing agent.
catalyst is necessary for this reaction. However, the conversion was
91% when the reaction time was extended to 36 h (entry 4, Table 1).
The conversion was low when reaction time was 12 h or 18 h (entries
5–6, Table 1). Among complexes 1–3, complex 2 has the highest ac-
reported that the complex [(C6H4eCH]NeC10H6)Co(III)(H)(PMe3)2]
(4) has a high catalytic activity in hydrosilylation of aldehydes and
ketones [30]. When we used complex 4 to catalyze the dehydration of
The introduction of fluorine groups as EWG on the benzene ring makes
the polarity of the Co-H bond in complexes 1–3 greater. Therefore the
nucleophilic ability of the hydrido hydrogen atom of complexes 1–3 is
higher than that of complex 4 and the catalytic activity of complexes
1–3 is stronger than that of complex 4. In comparison with entry 2, the
conversion was only 42% at 40 °C (entry 13, Table 1). However, when
increasing the reaction temperature, the conversion did not increase
(entry 14, Table 1). On the contrary, when the temperature is too high,
stability of the complex 2 in solution by in situ IR and found that
complex 2 is stable at 80 °C in solution. Perhaps under catalytic con-
ditions the catalyst (complex 2) has a deactivation or decomposition in
the presence of other components at high temperature. In order to ex-
plore the effect of the amount of silane on the reaction, (EtO)3SiH in
different molar ratios was added to the reaction mixture (entries 10–12,
Table 1). It was found that 1:3 (substrate:silane) is the best ratio. The
type of solvents is an important factor that affects the amide dehydra-
tion. It is found that THF (entry 2, Table 1) was the best reaction
medium for this catalytic system in comparison with toluene, DMF,
DMSO, dioxane, and acetonitrile (entries 15–19, Table 1).
2. Experimental section
2.1. General procedures and materials
Standard vacuum techniques were used in manipulations of volatile
and air-sensitive materials. Solvents were dried by known procedures
and distilled under nitrogen before use. Infrared spectra
(4000–400 cm−1), as obtained from Nujol mull between KBr disks,
were recorded on a Bruker ALPHA FT-IR instrument. NMR spectra were
recorded using Bruker Avance 300 MHz spectrometer. GC–MS was re-
corded on a TRACE-DSQ instrument and GC was recorded on a Fuli
9790 instrument. Melting points were measured in capillaries sealed
under N2 and were uncorrected. All the amides were purchased and
used without further purification. The purity of the triethoxysilane used
is 95%, the other 5% is tetraethoxysilane. The silanes were purchased
from J&K Scientific. Fe(PMe3)4 [29] and complexes (1–3) [28] were
prepared according to literature procedures and their data of char-
acterization are listed in the Supporting Information.
Caution! (EtO)3 SiH is flammable and highly toxic by inhalation and
may cause skin irritation and blindness. Even if during our studies on
the dehydration of amides, we used it without incident, triethoxysilane
should be used with precaution. Indeed, due to possible silane dis-
proportionation, the formation of an extremely pyrophoric gas (pos-
sibly SiH4) has led to several fires and explosions reported in the lit-
erature.
SI). The conversion was 86% when using Et3SiH as reducing agent in
toluene (entry 2, Table S1 in the SI). All reactions performed in THF or
dioxane or acetonitrile showed low conversions (up to 56%, entries 1, 5
and 6, Table S1 in the SI). The reaction did not occur in DMSO or DMF
(entries 3 and 4, Table S1 in the SI). Ph3SiH as reducing agent per-
formed in low conversions for this system. The 58% conversion was
detected in acetonitrile (entry 6, Table S2 in the SI). The conversion was
0% in THF and 42% in toluene (entries 1 and 2, Table S2 in the SI).
Using Ph2SiH2 as reducing agent, the conversion was 52% in THF or
24% in toluene respectively (entries 1 and 2, Table S3 in the SI). No
conversion was detected in DMF, DMSO, dioxane or acetonitrile (en-
tries 4–5, Table S3 in the SI). PhSiH3 as reducing agent for this reaction
was in moderate conversion. The conversions in THF and toluene were
similar (entries 1 and 2, Table S4 in the SI). The conversion was only
54% using TMDS as reducing agent in toluene and 42% in acetonitrile
(entries 2 and 6, Table S5 in the SI). This reaction did not occur with
TMDS in other solvents (entries 1, 3, 4 and 5, Table S5 in the SI). When
using PMHS as reducing agent the conversion was 71% in toluene and
0% in THF (entries 1 and 2, Table S6 in the SI). The low conversions
were detected in DMF, DMSO, dioxane, acetonitrile respectively (en-
tries 3 and 6, Table S6 in the SI). Compared to these silanes, (EtO)3SiH
as reducing agent in THF has the good conversion of 90% (entry 2,
(See Buchwald safety letter, Chemical & Engineering News (29 Mar
1993) Vol. 71, No. 13, pp. 2.)
2.2. General procedure for the dehydration of amides to nitriles
To a 25 mL Schlenk tube containing a solution of 2 in 2 mL of THF
was added amide (1.0 mmol) and (EtO)3SiH (0.50 g, 3.0 mmol). The
reaction mixture was stirred at 60 °C until there was no amide left
(monitored by TLC and GC–MS). The product was purified according to
literature procedures by Beller [19].
3. Results and discussion
In our initial trials, 4-chlorobenzamide was selected as a model
substrate to optimize the reaction conditions (Table 1). When 2% mol of
(EtO)3SiH reacted in THF at 60 °C. After 24 h, only 73% conversion was
detected (entry 1, Table 1). To our delight, a good conversion (90%)
contrast, when no complex 2 was added to the reaction system, the
reaction could not proceed (entry 3, Table 1). This indicates that the
Based on the above analysis, the optimized conditions of the cata-
lytic reaction are summarized as follows: 2 mol% of hydrido cobalt
complex 2 as a catalyst, 3 equivalents of (EtO)3SiH, in THF at 60 °C for
24 h (entry 2, Table 1). The scope of the substrates was explored under
73