A dual site catalyst for mild, selective nitrile reduction

Article abstract of DOI:10.1039/c3cc47384h

We report a novel ruthenium bis(pyrazolyl)borate scaffold that enables cooperative reduction reactivity in which boron and ruthenium centers work in concert to effect selective nitrile reduction. The pre-catalyst compound [κ³-(1-pz)₂HB(N = CHCH₃)]Ru(cymene) ⁺ TfO^- (pz = pyrazolyl) was synthesized using readily-available materials through a straightforward route, thus making it an appealing catalyst for a number of reactions. the Partner Organisations 2014.

Full text of DOI:10.1039/c3cc47384h

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A dual site catalyst for mild, selective nitrile
reduction†
Cite this: Chem. Commun., 2014,
50, 5391
Zhiyao Lu and Travis J. Williams*
Received 27th September 2013,
Accepted 23rd December 2013
DOI: 10.1039/c3cc47384h
www.rsc.org/chemcomm
We report a novel ruthenium bis(pyrazolyl)borate scaffold that enables
cooperative reduction reactivity in which boron and ruthenium centers
work in concert to effect selective nitrile reduction. The pre-catalyst
compound [j3-(1-pz)2HB(N = CHCH3)]Ru(cymene)+ TfOÀ (pz = pyrazolyl)
was synthesized using readily-available materials through a straightforward
route, thus making it an appealing catalyst for a number of reactions.
Fig. 1 Dual site catalysts for hydride manipulation.
As part of our ongoing studies of dual site ruthenium and boron-
containing catalysts for the manipulation of hydride groups, we have
recently reported a series of [di(pyridyl)borate]ruthenium complexes
(1, 2)1 that exhibit remarkable reactivity in a number of applications,2
notably including dehydrogenation of ammonia borane.3 Although
they are successful catalysts, these di(pyridyl)dimethylborate-derived
complexes are cumbersome to prepare, due largely to dependence on
an expensive and reactive BrBMe2 starting material and high water
and oxygen sensitivity of intermediate complexes in their syntheses.
Further, despite their catalytic utility, no direct evidence has been
collected to give a mechanistic account of the cooperative role, if any,
that boron and ruthenium play in the reactive mechanisms of 1 or 2.4
We suspect this is partially due to the robustness of the bridging m-OH
ligand between the boron and ruthenium centers in 2, which inhibits
access to a free borane in catalytic reactions (Fig. 1).
Scheme 1 Synthesis of precursor 3 and complex 6. Molecular structure of 3.
Here we show a conveniently prepared, borate-pendant
ruthenium complex (3) that retains much of the reactivity of
the original di(pyridyl)borate complexes (see ESI†), show that it is an Although 4 can be isolated, it is easily converted in situ to 3 by
efficient and selective catalyst for nitrile reduction, and provide treatment with 1 equiv. of thallium (or silver) triflate in nitrile solution.
evidence for the cooperative role that boron and ruthenium play in The synthesis proceeds in two smooth steps without the need for
the reaction, as a hydride donor and an activating group, respectively. materials that are cost-prohibitive or difficult to manipulate: all
The synthesis of 3 (Scheme 1) proceeds from potassium materials are amenable to handling using standard Schlenk techni-
di(pyrazolyl)borohydride5–7 and commercially-available (cymene)- ques and/or a glove box. 3 can be crystallized using isopropanol and
ruthenium dichloride dimer to give intermediate chloride 4. hexanes; its molecular structure was determined by single crystal X-ray
diffraction (Scheme 1). The crystal structure of 3 shows the borate
Loker Hydrocarbon Research Institute and Department of Chemistry,
University of Southern California, Los Angeles, California 90089-1661, USA.
E-mail: travisw@usc.edu
ligand bonded to boron in a tetrahedral geometry, which has analogy
to the popular tris(pyrazolyl)borohydride (Tp) ligand series.6–8
The synthesis of 3 revealed an important insight into the mecha-
nism of its catalytic reactivity. In the conversion of 4 to 3, a hydride is
transferred from a ligand B–H group to the coordinated nitrile of 5 in
>90% NMR yield (Scheme 1). This reaction is the first example of our
† Electronic supplementary information (ESI) available: Experimental details,
graphical spectra, and kinetics data. CCDC 963291 contains supplementary
crystallographic datafor 3. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c3cc47384h
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 5391--5393 | 5391

View Article Online
Communication
ChemComm
Table 2 Scope of 3-catalyzed nitrile reduction
envisioned cooperative reactivity of ruthenium and boron. Contrary to
the design concept from which we originally prepared 2,2 boron in 3
does not behave as a Lewis acid. The structure of 3 shows that
addition of the B–H group to the nitrile proceeds in a cis fashion, and
NMR evidence reveals that the selectivity for this geometry is exclusive.
Thus, we believe that the mechanism for this reaction involves
intramolecular hydride transfer from boron to carbon, rapidly
followed by (or concerted with) boron–nitrogen coordination.
The stoichiometric reduction of acetonitrile observed in the
synthesis of 3 can be made catalytic by treating 3 with nitrile and
sodium borohydride,9 thus enabling a mild and selective approach to
the synthesis of primary amines from nitriles, which remains a
contemporary topic in bifunctional catalysis.10 Known methods for
nitrile reduction, e.g. use of excess LiAlH4, borohydride reduction of a
nitrilium salt,11 use of metal hydride reagents,12 or high-pressure
hydrogenation13 can be incompatible with important synthetic handles,
such as aryl bromide and nitro groups.14 Milder conditions compatible
with groups like these require a stoichiometric portion of a borane
reagent,15 which in some cases must be independently prepared. The
new catalytic mechanism reported here enables high functional group
tolerance while incorporating an inexpensive reducing agent.
Table 1 shows the discovery and optimization of our conditions for
nitrile reduction. In the presence of 5 mol% 3, 2.0 equiv. of NaBH4
can reduce 4-trifluoromethylbenzonitrile (7a) to the corresponding
benzylamine (8a) in 50% conversion by NMR in 5 hours (entry 1).
When 1.0 equiv. of NaOtBu was added, the analogous reaction
reached >90% yield (>95% conversion) in the same time. With no
catalyst, the reaction was much slower, reaching completion in 7 days
with a 43% overall yield (entry 5). If the Tp-ligated homologue of 3
(6, Scheme 1) is used in this reaction, much of the starting material
decomposes under the reaction conditions. This illustrates that the
m-acetimine ligand in 3 plays an essential role in nitrile reduction.
The substrate scope of nitriles that can be reduced under our
optimized conditions is broad. As shown in Table 2, aromatic,
aliphatic, and heterocyclic nitriles are smoothly reduced to amines
under optimized conditions. Both electron-poor and electron-rich
substrates can be reduced in high yield. For example, in the presence
of 5 mol% 3, 4.0 equiv. of NaBH4 can reduce 4-trifluorobenzonitrile 7a
to the corresponding benzylamine 8a in 82% isolated yield (entry 1).
Electron-rich nitrile 7b can be reduced with a similar facility in 87%
yield (entry 2). Even more oxidative nitroarene 7c can be reduced with
exclusive selectivity for the nitrile over the nitro group (92%, entry 3).
We believe that this startling result is attributable to selective binding
Yielda
(%)
Entry Nitrile
1
Conditions
Product
5 mol% 3
4 eq. NaBH4, 1 eq.
82
87
NaOtBu
MeOH, reflux, 12 h
5 mol% 3
8 eq. NaBH4, 1 eq.
2
3
NaOtBu
MeOH, reflux, 12 h
5 mol% 3
4 eq. NaBH4, 1 eq.
92
NaOtBu
MeOH, reflux, 4 h
5 mol% 3
8 eq. NaBH4, 1 eq.
4
5
6
85
80
84
NaOtBu
MeOH, reflux, 14 h
5 mol% 3
4 eq. NaBH4, 1 eq.
NaOtBu
MeOH, reflux, 8 h
5 mol% 3
4 eq. NaBH4, 1 eq.
NaOtBu
MeOH, reflux, 12 h
5 mol% 3
8 eq. NaBH4, 1 eq.
7
8
9
60
64
56
87
NaOtBu
MeOH, reflux, 12 h
5 mol% 3
4 eq. NaBH4, 1 eq.
NaOtBu
MeOH, reflux, 8 h
5 mol% 3
4 eq. NaBH4, 1 eq.
NaOtBu
MeOH, reflux, 8 h
5 mol% 3
4 eq. NaBH4, 1 eq.
10
NaOtBu
MeOH, reflux, 8 h
a
Reported yields are isolated yields.
Table 1 Optimization of nitrile reduction conditions
and activation of the nitrile over the nitro group by the ruthenium
center of the catalyst. Furthermore, we see this as additional evidence
showing that nitrile binding to ruthenium is important to the catalytic
mechanism.
Despite the highly reducing conditions, an aryl bromide group in
nitrile 7d is not derivatized in the course of amine synthesis. This is an
important result because aryl bromides such as these are high value
substrates for cross-coupling and amination reactions relevant to the
synthesis of medicinally relevant compounds.16 This example further
illustrates high-yielding reduction of an alkyl nitrile. Whereas ketone
groups are known to react with NaBH4, the reduction of 7e (entry 5)
Entry
Catalyst
NaOtBu
5 h conversiona (%)
NMR yieldb
c
1
2
3
4
5
3
3
2
6
None
50
>95
46
41
38
1 equiv.
1 equiv.
1 equiv.
1 equiv.
>90%, 5 h
42%, 245 h
Trace, 245 h
43%, 245 h
No catalyst
a
Starting material consumption by NMR in 5 hours. b Product formed upon
consumption of nitrile and subsequent addition of water. c Not recorded.
5392 | Chem. Commun., 2014, 50, 5391--5393
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
illustrates high-yielding double reduction for the synthesis of amino- selective and high yielding nitrile reduction under mild conditions.
achohol 8e. A similar double reduction is observed in the reaction of Furthermore, this platform has yielded new insight into the
cinnamonitrile (7g, entry 7) to give alkyl amine 8g in 60% yield.
cooperative reactivity of ruthenium and boron by showing a
Reactions of nitriles appended to aromatic heterocycles afforded plausible scenario of how these two centers can work together,
complicated results. For example, pyridine 7h is compatible with the respectively, as an activating group (ruthenium) and a hydride
conditions and resulted in the formation of aminomethylpyridine donor (boron). Ongoing work in our laboratory regards the applica-
8h in 64% yield (entry 7). By remarkable contrast, more electron-rich tion of this system to the reduction of other high-value pi systems
heterocycle systems are not reduced. For example, 2-cyanofuran 7i and the elucidation of the mechanistic details of these reactions.
and 2-cyanothiophene 7j are selectively monohydrated as opposed to
This work was supported by the University of Southern
being reduced. Thus, amides 9i and 9j were isolated as the main California, the Hydrocarbon Research Foundation, and
products (entries 9 and 10). We suspect that the mechanism for the National Science Foundation (CHE-1054910). We thank
these reactions involves the addition of methanol (solvent) to a Ralf Haiges for X-ray crystallography. We thank the NSF (DBI-
ruthenium-coordinated nitrile, and that the amide products are 0821671, CHE-0840366, CHE-1048807) and NIH (S10 RR25432)
formed upon aqueous work-up. We do not currently have a proposal for analytical instrumentation.
to account for the selectivity of hydration versus reduction.
According to the insight gained from the stoichiometric synthesis
of 3 from 4, we propose the following template mechanism for
catalysis (Scheme 2). We suspect that the bridging imine of 3 is
reduced by borohydride to produce the amine product. We do not
Notes and references
1 B. L. Conley and T. J. Williams, J. Am. Chem. Soc., 2010, 132,
1764–1765.
2 B. L. Conley and T. J. Williams, Comments Inorg. Chem., 2011, 32,
have direct evidence for the intermediacy of 10; however, treatment
of 3 with a stoichiometric portion of NaBH4 in methanol-d4 results
in clean desymmetrization of cymene and pyrazole C–H groups of
the catalyst, consistent with the formation of diastereotopic protons,
as expected with the pyramidalization of the bridging nitrogen
ligand (see the ESI† for graphical spectra). Still, the intermediacy
of 10 remains a proposal because we have not established the kinetic
role of this transient material.
195–218.
3 B. L. Conley, D. Guess and T. J. Williams, J. Am. Chem. Soc., 2011,
133, 14212–14215.
4 Z. Lu, B. Malinoski, A. V. Flores, D. Guess, B. L. Conley and
T. J. Williams, Catalysts, 2012, 2, 412–421.
5 Scorpionates: The Coordination of Polypyrazolylborate Ligands, ed.
S. Trofimenko, Imperial College Press, London, 1999.
6 For recent reviews of organometallic chemistry of poly(pyzalolyl)borate
ligated complexes, see: (a) L. M. Caldwell, Adv. Organomet. Chem., 2008,
56, 1–94; (b) M. Lail, K. A. Pittard and T. B. Gunnoe, Adv. Organomet.
Chem., 2008, 56, 95–153; (c) E. Becker, S. Pavli and K. Kirchner, Adv.
Organomet. Chem., 2008, 56, 155–197; (d) I. R. Crossly, Adv. Organomet.
Chem., 2008, 56, 199–321.
7 R. J. Abernethy, A. F. Hill, M. K. Smith and A. C. Willis, Organo-
metallics, 2009, 28, 6152–6159.
8 S. Bhambri and D. A. Tocher, J. Chem. Soc., Dalton Trans., 1997,
3367–3372.
9 A ruthenium-ligated acetonitrile was reduced by NaBH4, however
the system produced unidentified products.13: L. F. Rhodes and
L. M. Venanzi, Inorg. Chem., 1987, 26, 2692–2695.
We propose that the bridging amine ligand is replaced by an
incoming substrate, and the borohydride group of 5 is regenerated
by a hydride from NaBH4, although we do not know the details of
these steps. We observed by 1H-coupled 11B NMR spectroscopy
that treatment of 3 with a stoichiometric portion of NaBH4 in
methanol-d4 results in the formation of (MeO)4BÀ, unreacted BH4
À
and a catalyst doublet, which indicates that a (pz)2BH2 inter-
mediate, if formed, is transient. Thus, we suspect that X in 10 M. Vogt, A. Nerush, M. A. Iron, G. Leitus, Y. Diskin-Posner,
1
L. J. W. Shimon, Y. Ben-David and D. Milstein, J. Am. Chem. Soc.,
2013, 135, 17004–17018.
11 R. F. Borch, Chem. Commun., 1968, 442–443.
Scheme 2 is methoxide. H NMR studies of the working catalyst
reveal that once ligated, the reductions of the nitrile groups and
imine groups are very facile. Thus, the rate-determining step could 12 For example, stoichiometric portions of NiII, CoII, OsIV, IrIII, PtII, are
be amine for nitrile substitution in the conversion of 10. This
mechanism is a subject of ongoing work in our laboratory.
known to mediate nitrile and nitro group reduction in the presence
of excess NaBH4. Y. Suzuki, Y. Miyaji and Z. Imai, Tetrahedron Lett.,
1969, 52, 4555–4558.
In conclusion, we report here a conveniently prepared 13 For selected examples, see: H. Paul, S. Bhaduri and G. K. Lahiri,
Organometallics, 2003, 22, 3019–3021; P. A. Chase, T. Jurca and
D. W. Stephan, Chem. Commun., 2008, 1701–1703; R. Reguillo,
M. Grellier, N. Vautravers, L. Vendier and S. Sabo-Etienne, J. Am.
homologue of our successful di(pyridyl)borate-ligated ruthenium
complexes. The new catalytic scaffold shows comparable reactivity
to the old one in several key reactions and confers new reactivity to
the cooperative ruthenium–boron catalytic motif by enabling
Chem. Soc., 2010, 132, 7854–7855; X. Miao, C. Fischmeister,
C. Bruneau, J.-L. Dubois and J.-L. Couturier, ChemSusChem, 2012,
5, 1410–1414.
14 P. Magnus, N. Sane, B. P. Fauber and V. Lynch, J. Am. Chem. Soc.,
2009, 131, 16045–16047.
15 D. Haddenham, L. Pasumansky, J. DeSoto, S. Eagon and
B. Singaram, J. Org. Chem., 2009, 74, 1964–1970.
16 For selected reviews of utilization of aryl bromides in medicinal synth-
´
esis, see: J. Hassan, M. Sevignon, C. Gozzi, E. Schulz and M. Lemaire,
Chem. Rev., 2002, 102, 1359–1470; D. Alberico, M. E. Scott and
M. Lautens, Chem. Rev., 2007, 107, 174–238; G. Zeni and R. C. Larock,
Chem. Rev., 2006, 106, 4644–4680; J.-P. Corbet and G. Mignani, Chem.
Rev., 2006, 106, 2651–2710; F.-S. Han, Chem. Soc. Rev., 2013, 42,
5270–5298; Y. Aubin, C. Fischmeister, C. M. Thomas and J.-L. Renaud,
Chem. Soc. Rev., 2010, 39, 694–711; G. P. McGlacken and L. M. Bateman,
Chem. Soc. Rev., 2009, 38, 2447–2464; T. R. M. Rauws and B. U. W. Maes,
Chem. Soc. Rev., 2012, 41, 2463–2497.
Scheme 2 Mechanistic template for nitrile reductions.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 5391--5393 | 5393