Efficient hydration of nitriles to amides in water, catalyzed by ruthenium hydroxide supported on alumina

Article abstract of DOI:10.1002/anie.200353461

A broad range of amides is accessible in excellent yield (> 99%) by hydration of the corresponding nitriles in water in the presence of the supported ruthenium catalyst Ru(OH)_x/Al₂O₃ (see scheme). For example, the industrially important conversion of acrylonitrile into acrylamide was achieved in quantitative yield and better than 99% selectivity. The catalyst can be reused without loss of catalytic activity and selectivity.

Full text of DOI:10.1002/anie.200353461

Communications
Catalytic Amide Synthesis
Efficient Hydration of Nitriles to Amides in
Water, Catalyzed by Ruthenium Hydroxide
Supported on Alumina**
Kazuya Yamaguchi, Mitsunori Matsushita, and
Noritaka Mizuno*
Hydration of nitriles to the corresponding amides is an
important reaction in academia and industry because of their
usefulness in a wide variety of applications such as inter-
mediates in organic syntheses and as raw materials for
engineering plastics, detergents, and lubricants.[1–4] For exam-
ple, hydration of acrylonitrile produces annually more than
2 ” 105 tons of acrylamide and is the most important technol-
ogy for the production of this chemical.[4–6] Homogeneous
hydration of nitriles is traditionally carried out with acids and
bases such as H2SO4 and NaOH. However, carboxylic acids
are formed as undesirable by-products by hydrolysis of the
starting nitriles and amide products, especially under strongly
basic conditions. In addition, many functional groups, for
[*] Dr. K. Yamaguchi, M. Matsushita, Prof. Dr. N. Mizuno
Department of Applied Chemistry, School of Engineering
The Universityof Tokyo
7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656 (Japan)
Fax: (+81)3-5841-7220
E-mail: tmizuno@mail.ecc.u-tokyo.ac.jp
[**] This work was supported in part bythe Core Research for
Evolutional Science and Technology(CREST) program of Japan
Science and TechnologyAgency(JST) and a Grant-in-Aid for
Scientific Research from the Ministryof Education, Culture, Sports,
Science, and Technologyof Japan.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
1576
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200353461
Angew. Chem. Int. Ed. 2004, 43, 1576 –1580

Angewandte
Chemie
Table 1: Hydration of benzonitrile to benzamide with various catalysts.[a]
example, carbon–carbon double bonds, do not tolerate
such forcing conditions, which results in decreased
selectivity for amides. Only under carefully controlled
conditions with one of equivalent water with respect to
the nitrile could hydration of a nitrile to an amide be
selectively catalyzed by H2SO4 to give the corresponding
amide sulfate salt. Since a stoichiometric amount of
ammonia (two equivalents with respect to amide sulfate)
is required to isolate the free amide, ammonium sulfate is
produced as waste. Therefore, the development of an
efficient, intrinsically non-waste-producing catalytic
hydration system is of great importance.
EntryCataslyt
Conversion of
Selectivityto
Rate”103 [mminÀ1
]
benzonitrile [%] benzamide [%]
1
Ru(OH)x/Al2O3
47
7
7
2
>99
>99
>99
>99
>99
–
–
–
–
–
1.74
0.10
0.24
0.06
0.79
0.00
0.00
0.00
0.00
0.00
0.00
0.00
2[b]
3
(RuCl)2Ca8(PO4)6(OH)2
Ru(OH)3·nH2O
RuO2
RuCl3·nH2O
[Ru(acac)3]
[RuCl2(PPh3)3]
[RuCl2(DMSO)4]
[{RuCl2(p-cymene)}2]
[RuCl2(bpy)2]
4
5
6
7
8
9
10
28
no reaction
no reaction
no reaction
no reaction
no reaction
no reaction
no reaction
11[c] Al2O3
–
–
Many efficient methods that use microorganisms for
enzymatic hydration of nitriles[5,6] or homogeneous com-
plexes of transition metals such as cobalt,[7] copper,[8]
molybdenum,[9] ruthenium,[10] rhodium,[11] palladium,[12,13]
and platinum[14,15] have been reported. However, these
systems have disadvantages: difficulty in catalyst/product
separation and necessity of special handling of micro-
organisms and metal complexes. Despite the advantages
of heterogeneous hydration, for example, easy catalyst/
product separation and recycling, only a few examples
have been reported. Unactivated neutral alumina (stoi-
chiometric hydration without water),[16] Raney copper,[4]
KF/Al2O3,[17] KF/phosphate,[18] MnO2/SiO2,[19,20] Ru-sub-
stituted hydroxyapatite ((RuCl)2Ca8(PO4)6(OH)2),[21] and
NaNO3 on fluoroapatite (NaNO3/FAP)[22] are examples.
However, turnover numbers are still very low (ꢀ 6) or
stoichiometric, catalysts can not be reused, and/or the type of
nitrile is limited. In this context, efficient, widely usable,
reusable catalysts are so far unknown, although heteroge-
neous hydration systems are environmentally and technolog-
ically the most desirable.[23–25] Herein we report that the easily
12[c] Al2O3 treated with
NaOH
13
14
15
NaOH
H2SO4
None
3
>99
–
–
0.05
0.00
0.00
no reaction
no reaction
[a] Reaction conditions: benzonitrile (1 mmol), catalyst (2 mol%), water (3 mL), 403 K,
3 h. Conversion and selectivitywere determined byGC with an internal standard.
[b] Prepared according to the procedure in ref. [20]. [c] 0.2 g. Wilgus et al.[16] reported
that unactivated alumina could act as a stoichiometric reagent (two hydroxy functions
for hydration of one nitrile molecule) for the hydration of nitriles in the absence of
water. If the alumina used can act as a stoichiometric reagent in the same way, the
maximum yield is estimated to be ca. 5%. However, alumina was completely inactive
under these conditions, probablybecause of the presence of water in the present
system.
as RuCl3·nH2O and [RuCl2(PPh3)3] 18-Electron ruthenium
complexes such as [RuCl2(dmso)4], [RuCl2(bpy)2], and
[{RuCl2(p-cymene)}2] were completely inactive.
When the hydration of benzonitrile was carried out at
413 K (see Experimental Section), a quantitative yield of
benzamide was obtained after 6 h, as determined by GC
analysis. The Ru(OH)x/Al2O3 catalyst could be easily sepa-
rated from the reaction mixture by filtration (Figure 1a and
b). After separation of the catalyst, the filtrate was cooled to
273 K, and analytically pure white crystals of benzamide
appeared (Figure 1c). The crystals were isolated from the
water solvent by filtration or decantation in 90% yield.
Recovered Ru(OH)x/Al2O3 could be reused at least twice for
the hydration of benzonitrile without appreciable loss of
catalytic performance (entries 1–3 in Table 2). After the
catalytic hydration of benzonitrile was completed under the
conditions in Table 2, the reaction mixture was filtered to
remove the Ru(OH)x/Al2O3 and product amide. It was
confirmed that no ruthenium was present in the filtrate by
inductively coupled plasma atomic emission spectroscopy
(ICP-AES; detection limit of 7 ppb). Then, benzonitrile
(1 mmol) was again added to the filtrate and the solution
was heated to 413 K. No conversion of benzonitrile was
observed. These results show that any contribution to the
observed catalysis from ruthenium species that leached into
the reaction solution can be ruled out, and the observed
catalysis is truly heterogeneous in nature.[30]
prepared, inexpensive supported ruthenium hydroxide cata-
[26–28]
lyst Ru(OH)x/Al2O3
is effective for the hydration of
various nitriles to amides in water[29] under conditions that are
entirely free of explosive, hazardous, or carcinogenic organic
solvents, even in the workup steps [Eq. (1)].
First, the catalytic activity and selectivity for the hydration
of benzonitrile to benzamide in water at 403 K were
compared with those of a variety of ruthenium catalysts
(Table 1). Hydration did not proceed in the absence of
catalyst. g-Al2O3 and g-Al2O3 treated with NaOH did not
show any catalytic activity under the present reaction
conditions. Among the ruthenium catalysts tested,
Ru(OH)x/Al2O3 had the highest catalytic activity and selec-
tivity for the hydration of benzonitrile to benzamide, and no
benzoic acid could be detected. Under the same conditions, it
was confirmed that Ru(OH)x/Al2O3 did not catalyze the
hydrolysis and dehydration of benzamide. The yield of
benzamide was higher than those of heterogeneous catalysts
such as anhydrous RuO2, Ru(OH)3·nH2O, and (RuCl)2Ca8-
(PO4)6(OH)2, and of homogeneous ruthenium catalysts such
The scope of the present Ru(OH)x/Al2O3 catalyst system
with regard to various kinds of nitriles was examined. The
results are summarized in Table 2. Ru(OH)x/Al2O3 has high
catalytic activity for hydrations of activated, unactivated, and
Angew. Chem. Int. Ed. 2004, 43, 1576 –1580
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1577

Communications
amide (entry 9). Notably, the industrially important hydration
of acrylonitrile proceeded with 2.3 mol% of Ru(OH)x/Al2O3
at 393 K to afford a quantitative yield of acrylamide. Neither
hydration of the carbon–carbon double bond nor polymer-
ization of acrylonitrile and/or acrylamide occurred (entry 10).
Ru(OH)x/Al2O3 successfully catalyzed the hydration of
various nitriles containing heteroatoms such as oxygen,
nitrogen, and sulfur to the corresponding amides in high
yields (entries 6, 8, 11, and 12). The less reactive aliphatic
nitriles could also be hydrated to the corresponding aliphatic
amides (entries 13 and 14).
In the hydration of toluonitriles, the lower reaction rate of
o-tolunitrile (79% yield after 24 h) relative to m- and p-
Table 2: Hydration of various nitriles catalyzed by Ru(OH)x/Al2O3.[a]
EntrySubstrate
Product
t [h]
Conversion [%]
Selectivity[%]
1
6
6
6
99
99
99
>99
>99
>99
2[b]
3[b]
4
6
96
>99
5
6
6
6
>99
>99
99
Figure 1. Hydration of benzonitrile to benzamide.
a) Reaction mixture, b) filtrate after removal of
Ru(OH)x/Al2O3 byfiltration at 363 K, and c) benza-
mide crystals obtained after cooling the filtrate.
>99
7
8
6
6
>99
>99
>99
heterocyclic nitriles in water. Ben-
zonitriles were smoothly hydrated
to give the corresponding benza-
99
mides
in
excellent
yields
(entries 1–8). The rates were not
much influenced by the electronic
effects of the substituents on the
aromatic ring of benzonitriles. A
larger scale experiment (tenfold
scale-up, 0.4 mol% Ru) for benzo-
nitrile at 423 K showed a turnover
frequency (TOF) of 13 hÀ1, and the
turnover number (TON) reached
234. These values are the highest of
those reported for the heterogene-
ous hydration of benzonitrile so
far: KF/Al2O3 (TOF 0.2 hÀ1, TON
0.5 based on KF),[17] KF/phosphate
(0.2 hÀ1, 0.9 based on KF),[18]
MnO2/SiO2 (0.2 hÀ1, 0.7 based on
MnO2),[19] (RuCl)2Ca8(PO4)6(OH)2
(0.2 hÀ1, 6 based on Ru),[21] and
NaNO3 on fluoroapatite (NaNO3/
FAP ; 2 hÀ1, 6 based on NaNO3).[22]
Hydration of the a,b-unsaturated
nitrile cinnamonitrile proceeded
only at the cyano group to afford
the corresponding a,b-unsaturated
9
5
24
6
>99
>99
91
99
>99
98
10[c]
11
12
6
>99
>99
13[d]
14[d]
10
24
92
91
99
97
[a] Reaction conditions: nitrile (1 mmol), Ru(OH)x/Al2O3 (4 mol% Ru), water (3 mL), 413 K. The
conversion and selectivitywere determined byGC with an internal standard. Traces of the corresponding
carboxylic acids, formed by hydrolysis of starting nitriles or amide products, were found as by-products
in some cases. Carbon balance for each reaction was greater than 95%. [b] These experiments used a
recycled catalyst: 1st reuse (entry 2), 2nd reuse (entry 3). The reaction conditions were the same as in
entry 1. The initial rates for the recycle runs were the same as that for the first run with fresh catalyst.
[c] Reaction conditions: acrylonitrile (10 mmol), Ru(OH)x/Al2O3 (2.3 mol% Ru), water (20 mL), 393 K.
The conversion and selectivitywere determined by 1H NMR spectroscopic analysis with an internal
standard. [d] The reaction was carried out at 403 K.
1578
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 1576 –1580

Angewandte
Chemie
toluonitrile indicates a steric effect. This observation strongly
suggests that nitriles coordinate to the ruthenium center on
the surface of Ru(OH)x/Al2O3 and that hydration of nitriles
proceeds by intramolecular attack of a ruthenium hydroxide
species on the coordinated nitrile (steps 1 and 2). When the
hydration of o-, m-, and p-toluonitriles was catalyzed by
NaOH, the reactivity did not vary. In addition, the 1 value for
NaOH-catalyzed hydration in water (external attack of water
on the nitrile carbon atom catalyzed by free OHÀ) should be
positive, in contrast to the negative 1 value for the present
Ru(OH)x/Al2O3-catalyzed hydration (see below). The above
observations show that free OHÀ is not the active species. It
was reported that cobalt(iii) hydroxide undergoes coordina-
tion of a nitrile and that this favors nucleophilic attack of the
metal hydroxide on the proximal nitrile carbon atom rather
than on the distal carbon–carbon double bond, hydration of
which is thus prevented. These results are in agreement with
those for the present system and support coordination of a
nitrile to Run+ (II). Competitive hydration of benzonitrile and
p-substituted benzonitriles was carried out in water at 413 K.
The order of reactivity was p-OCH3 (kX/kH = 1.28) > p-CH3
(1.01) > p-H (1.00) > p-Cl (0.79) > p-COCH3 (0.70). The good
linearity of Hammett plots (lg(kX/kH) versus Brown–Oka-
moto s+, see the Supporting Information) suggests that the
present hydration proceeds with a single mechanism. The
slope of the linear line gave a Hammett 1 value of À0.21. The
negative Hammett 1 value might be due to the formation of a
III or IV together with the formation of the amide product
(step 3).
Kinetic studies revealed a zero-order dependence of the
reaction rate on the concentration of water and a first-order
dependence on both the amount of Ru(OH)x/Al2O3 and the
nitrile concentration (Supporting Information). The Arrhe-
nius plots (observed rate constant kobs versus TÀ1) are also
shown in the Supporting Information (383–423 K). Good
linearity was observed for the Arrhenius plots, which gave the
following activation parameters: Ea = 71.5 kJmolÀ1, DH4°13 K
68.1 kJmolÀ1, DS°413 K = À155.3 JmolÀ1 KÀ1, and DG4°13 K
=
=
132.2 kJmolÀ1. The activation energy Ea is lower than that
of hydration with H2SO4 (Ea = 150.7 kJmolÀ1).[31] The hydra-
tion rate was independent of the content of D2O (Supporting
À
Information). These results show that O H bond dissociation
(step 3) is not included in the rate-limiting step under the
present conditions.[32] The value of the activation entropy
DS°413 K suggests that a bimolecular transition state (step 2) is
included in the rate-limiting step.[32]
In summary, Ru(OH)x/Al2O3 can act as a heterogeneous
catalyst for the hydration of nitriles in water. The hydration of
both activated and unactivated nitriles can be performed with
high conversion and selectivity to give the corresponding
amides. Furthermore, catalyst/product separation is easy, and
Ru(OH)x/Al2O3 is recyclable.
positively charged transition state at the carbon atom Experimental Section
The Ru(OH)x/Al2O3 catalyst was prepared by the procedure reported
adjacent to the phenyl ring.
previously.[26–28] A typical Ru(OH)x/Al2O3-catalyzed hydration was
carried out as follows: Benzonitrile (5 mmol), Ru(OH)x/Al2O3
(4 mol%), and water (15 mL) were placed in a Teflon vessel with a
magnetic stir bar. Next, the Teflon vessel was quickly inserted into an
autoclave, and then the autoclave was heated at 413 K (bath
temperature). The reaction rates were not affected by stirring rates
from 500 to 2000 rpm. After 6 h, benzamide was formed in > 99%
yield, as determined by GC analysis. Ru(OH)x/Al2O3 was removed by
filtration at 363 K. The filtrate was cooled to 273 K, and white crystals
precipitated from the filtrate. The crystalline product was collected by
simple filtration and dried in vacuo to give analytically pure
benzamide in 90% yield. The recovered Ru(OH)x/Al2O3 was further
washed with hot water and an aqueous NaOH solution (pH 13) and
then dried in vacuo before reuse.
On the basis of these results, we propose the possible
mechanism given in Figure 2 for the Ru(OH)x/Al2O3-cata-
Received: December 4, 2003 [Z53461]
Keywords: amides · cyanides · heterogeneous catalysis ·
.
hydration · ruthenium
Figure 2. A proposed reaction mechanism for the Ru(OH)x/Al2O3-cata-
lyzed hydration of nitriles.
lyzed hydration of nitriles. This catalytic hydration can be
divided into steps 1–3: Initially, coordination of a nitrile to the
ruthenium center proceeds to form II (step 1). Then, intra-
molecular nucleophilic attack of hydroxide species in II on
the nitrile carbon atom takes place to afford the ruthenium
iminolate species III or ruthenium h2-amidate species IV
(step 2). Step 2 may include the formation of a carbocation-
type transition state. Regeneration of ruthenium hydroxide
species I proceeds by the ligand exchange between water and
[1] F. Matsuda, Chemtech 1977, 7, 306.
[2] C. E. Mabermann in Encyclopedia of Chemical Technology,
Vol. 1 (Ed.: J. I. Kroschwitz), Wiley, New York, 1991, p. 251 –
266.
[3] D. Lipp in Encyclopedia of Chemical Technology, Vol. 1 (Ed.:
J. I. Kroschwitz), Wiley, New York, 1991, p. 266 – 287.
[4] R. Opsahl in Encyclopedia of Chemical Technology, Vol. 2 (Ed.:
J. I. Kroschwitz), Wiley, New York, 1991, p. 346 – 356.
[5] P. K. Mascharak, Coord. Chem. Rev. 2002, 225, 201, and
references therein.
Angew. Chem. Int. Ed. 2004, 43, 1576 –1580
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1579

Communications
[6] K. Ingvosersen, J. Kamphuis in Enzyme Catalysis in Organic
[21] K. Mori, K. Yamaguchi, T. Mizugaki, K. Ebitani, K. Kaneda,
Chem. Commun. 2001, 461.
[22] A. Solhy, A. Smahi, H. E. Badaoui, B. Elaabar, A. Amoukal, A.
Tikad, S. Sebti, D. J. Macquarrie, Tetrahedron Lett. 2003, 44,
4031.
[23] P. T. Anastas, J. C. Warner, Green Chemistry: Theory and
Practice, Oxford University Press, Oxford, 1998.
[24] J. H. Clark, Green Chem. 1999, 1, 1.
Synthesis, Vol. 1 (Eds.: K. Drauz, H. Waldmann), VCH, Wein-
heim, 1995, p. 365 – 392.
[7] J. H. Kim, J. Britten, J. Chin, J. Am. Chem. Soc. 1993, 115, 3618.
[8] R. Breslow, R. Fairweather, J. Keana, J. Am. Chem. Soc. 1967, 89,
2135.
[9] K. L. Breno, M. D. Pluth, D. R. Tyler, Organometallics 2003, 22,
1203.
[10] S.-I. Murahashi, S. Sasao, E. Saito, T. Naota, J. Org. Chem. 1992,
57, 2521.
[25] Thematic issue on “Green Chemistry”, Acc. Chem. Res. 2002, 35,
685 – 816.
[11] M. C. K.-B. Dojoman, A. N. Ajjou, Tetrahedron Lett. 2000, 41,
4845.
[26] K. Yamaguchi, N. Mizuno, Angew. Chem. 2002, 114, 4720;
Angew. Chem. Int. Ed. 2002, 41, 4538.
[12] G. Villain, A. Gasket, P. H. Kalck, J. Mol. Catal. 1981, 12, 103.
[13] N. V. Kaminskaia, N. M. Kostic, J. Chem. Soc. Dalton Trans.
1996, 3677.
[14] T. Ghaffar, A. W. Parkins, Tetrahedron Lett. 1995, 36, 8657.
[15] T. Ghaffar, A. W. Parkins, J. Mol. Catal. A 2000, 160, 249.
[16] C. P. Wilgus, S. Downing, E. Molitor, S. Bains, R. M. Pagni, G. W.
Kabalka, Tetrahedron Lett. 1995, 36, 3469.
[17] C. G. Rao, Synth. Commun. 1982, 12, 177.
[18] S. Sebti, A. Rhiil, A. Saber, N. Hanafi, Tetrahedron Lett. 1996, 37,
6555.
[27] K. Yamaguchi, N. Mizuno, Angew. Chem. 2003, 115, 1517;
Angew. Chem. Int. Ed. 2003, 42, 1479.
[28] K. Yamaguchi, N. Mizuno, Chem. Eur. J. 2003, 9, 4353.
[29] Thematic issue on “Organic Reaction in Water”, Adv. Synth.
Catal. 2002, 3–4, 219 – 451.
[30] R. A. Sheldon, M. Wallau, I. W. C. E. Arends, U. Schuchardt,
Acc. Chem. Res. 1998, 31, 485.
[31] K. Sugiyama, H. Miura, Y. Nakano, H. Sekiwa, T. Matsuda, Bull.
Chem. Soc. Jpn. 1986, 59, 2983.
[32] K. A. Connors, Chemical Kinetics, The Study of Reaction Rates
in Solution, VCH, New York, 1990.
[19] K.-T. Li, M.-H. Shih, H.-W. Huang, C.-J. Hu, Synthesis 1988, 715.
[20] P. Breuilles, R. Leclerc, D. Uguen, Tetrahedron Lett. 1994, 35,
1401.
1580
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 1576 –1580