Heterogeneous dehydrocoupling of amine-borane adducts by skeletal nickel catalysts

Article abstract of DOI:10.1021/ic201809g

Skeletal Ni, produced by the selective leaching of Al from a Ni/Al alloy, has been successfully employed in the catalytic dehydrogenation of various amineborane adducts. The combination of low cost and facile single-step synthesis make this system a potentially attractive alternative to the previously described precious metal and other first-row metal catalysts. The heterogeneous nature of the catalyst facilitates convenient product purification, and this is the first such system to be based on a first-row transition metal. Catalytic dehydrocoupling of Me2NH·BH₃ (1) and Et2NH3 BH3 (5) was demonstrated using 5 mol % skeletal Ni catalyst at 20 °C and produced [Me2N·BH₂]₂ (2) and [Et2N-BH2]2/Et2NdBH2 (6), respectively. The related adduct iPr2NH3 BH ₃ (7) was also dehydrogenated to afford iPr2NdBH2 (8) but with significant catalyst deactivation. Catalytic dehydrocoupling of MeNH₂ 3 BH₃ (9) was found to yield the cyclic triborazane [MeNH-BH ₂]3 (10) as the major product, whereas high molecular weight poly- (methylaminoborane) [MeNH-BH₂]_n (11) (M_w = 78 000 Da, PDI = 1.52) was formed when stoichiometric quantities of Ni were used. Similar reactivity was also observed with NH₃ 3 BH₃ (12), which produced cyclic oligomers and insoluble polymers, [NH₂-BH2] _xx (14), under catalytic and stoichiometric Ni loadings, respectively. Catalyst recycling was hindered by gradual poisoning. A study of possible catalyst poisons suggested that BH3 was the most likely surface poison, in line with previous work on colloidal Rh catalysts. Catalytic borane adducts using skeletal Cu and Fe was also explored. Skeletal Cu was found to be a less active dehydrogenation catalyst for amine-borane adducts but also yielded poly(methylaminoborane) under stoichiometric conditions on reaction with MeNH₂ · BH₃ (9). Skeletal Fe was found to be completely inactive towardamine-borane dehydrogenationr

Full text of DOI:10.1021/ic201809g

ARTICLE
pubs.acs.org/IC
Heterogeneous Dehydrocoupling of AmineꢀBorane Adducts by
Skeletal Nickel Catalysts
Alasdair P. M. Robertson,† Riccardo Suter,†,‡ Laurent Chabanne,† George R. Whittell,† and Ian Manners†,*
†School of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 1TS United Kingdom
‡Department of Chemistry and Applied Biosciences, ETH, H€onggerberg, 8093 Z€urich, Switzerland
S Supporting Information
b
ABSTRACT: Skeletal Ni, produced by the selective leaching of
Al from a Ni/Al alloy, has been successfully employed in the
catalytic dehydrogenation of various amineꢀborane adducts.
The combination of low cost and facile single-step synthesis
make this system a potentially attractive alternative to the
previously described precious metal and other first-row metal
catalysts. The heterogeneous nature of the catalyst facilitates
convenient product purification, and this is the first such system to be based on a first-row transition metal. Catalytic
dehydrocoupling of Me2NH BH3 (1) and Et2NH BH3 (5) was demonstrated using 5 mol % skeletal Ni catalyst at 20 °C and
3
3
produced [Me2NꢀBH2]2 (2) and [Et2NꢀBH2]2/Et2NdBH2 (6), respectively. The related adduct iPr2NH BH3 (7) was also
3
dehydrogenated to afford iPr2NdBH2 (8) but with significant catalyst deactivation. Catalytic dehydrocoupling of MeNH2 BH3 (9)
3
was found to yield the cyclic triborazane [MeNHꢀBH2]3 (10) as the major product, whereas high molecular weight poly-
(methylaminoborane) [MeNHꢀBH2]n (11) (Mw = 78 000 Da, PDI = 1.52) was formed when stoichiometric quantities of Ni were
used. Similar reactivity was also observed with NH3 BH3 (12), which produced cyclic oligomers and insoluble polymers,
3
[NH2ꢀBH2]x (14), under catalytic and stoichiometric Ni loadings, respectively. Catalyst recycling was hindered by gradual
poisoning. A study of possible catalyst poisons suggested that BH3 was the most likely surface poison, in line with previous work on
colloidal Rh catalysts. Catalytic dehydrogenation of amineꢀborane adducts using skeletal Cu and Fe was also explored. Skeletal Cu
was found to be a less active dehydrogenation catalyst for amineꢀborane adducts but also yielded poly(methylaminoborane) under
stoichiometric conditions on reaction with MeNH2 BH3 (9). Skeletal Fe was found to be completely inactive toward
3
amineꢀborane dehydrogenation.
’ INTRODUCTION
interest, due to the low cost of these metals relative to those of the
second and third row.
Interest in amineꢀborane adducts, which are commonly
applied as reducing and hydroborating reagents in organic
synthesis,1,2 has increased dramatically over the past decade.
This increased attention has, in part, resulted from their potential
application in hydrogen storage and transfer3ꢀ6 and also as
precursors to new materials.7ꢀ10 Catalytic dehydrocoupling/
dehydrogenation of amineꢀborane adducts is of key importance
in this respect, and a plethora of catalytic systems have now been
reported, which are based on various transition metals5,11ꢀ25 and
main group elements.26,27 In general, however, the reported
catalyst systems are based on 4d and 5d metals, including Rh, Ir,
Re, and Ru, where the cost is particularly high relative to their 3d
counterparts (Supporting Information, Figure SI 2). With regard
to first-row metal-catalyzed processes, homogeneous systems
based on Ni and Ti have been reported,15,17,28ꢀ30 as have those
containing Cr, Mn and Fe,15,18,19,31,32 although most of the
compounds of the latter required photoirradiation to produce
a catalytic effect. Stoichiometric and catalytic amineꢀborane
dehydrocoupling/dehydrogenation has also been achieved using
frustrated Lewis pair (FLP) chemistry, although this work is
currently in its infancy.33ꢀ35 The development of novel first-row
amineꢀborane dehydrocoupling catalysts is therefore of significant
The analogy between CꢀC bonds and BꢀN bonds is well-
known,36 and amineꢀborane adducts and aminoboranes can be
considered as inorganic analogues of alkanes and olefins, respec-
tively. This relationship has guided catalyst discovery within the
amineꢀborane dehydrogenation field, where the alkane dehy-
drogenation catalyst IrH2POCOP (POCOP = η3-1,3-(OPtBu2)2-
C6H3)37 has been successfully applied to ammoniaꢀborane and
various primary amineꢀborane adducts as substrates to yield a
variety of dehydrogenated products,21,38 including the first
soluble polyaminoboranes (Scheme 1).20 Similarly, Schneider
and Fagnou have also successfully applied Ru catalysts to
amineꢀborane dehydrocoupling,12,14 following previous reports
of their use in alcohol redox processes and ketone hydro-
genation.39ꢀ41 The screening of such catalysts was initially
attempted due to the similarities in both polarity and electronic
structure between ammoniaꢀborane and methanol.
Based upon a combination of the CꢀC/BꢀN bonding analogy
and the previous reports of amineꢀborane dehydrogenation by
Received: August 18, 2011
Published: November 21, 2011
r
2011 American Chemical Society
12680
dx.doi.org/10.1021/ic201809g Inorg. Chem. 2011, 50, 12680–12691
|

Inorganic Chemistry
ARTICLE
Scheme 1. Catalytic AmineꢀBorane Dehydropolymerization by IrH2POCOP20,21,38
Scheme 2. Generalized Aqueous Synthesis of Skeletal Ni
from Ni/Al Alloy
Scheme 3. Catalytic Dehydrogenation of 1 with NiT50 To
Produce the Cyclic Diborazane 2
CdC hydrogenation catalysts, we postulated that skeletal first-
row transition metals may be effective amineꢀborane dehydro-
genation catalysts. Skeletal first-row metals are well-known olefin
hydrogenation catalysts, with their activity first reported in
patents by Raney in 1925 and 1927.42,43 These catalysts are
generally produced by the selective leaching of Al or Si content
from an alloy with the relevant metal,44 and skeletal forms of
Ni,43,45,46 Cu,46ꢀ50 Co,51 and Fe52 have been reported. These
highly reactive, high surface area systems are attractive due partly
to their heterogeneous nature, which facilitates product separa-
tion, and perhaps more importantly due to their relatively low
cost and ease of synthesis.
The potential for skeletal first-row metals to act as cheap,
readily available heterogeneous catalysts provided us with a
strong impetus to investigate their use in the dehydrocoupling
of amineꢀborane adducts. In this paper we report our studies on
the catalytic dehydrogenation of amineꢀborane adducts using
skeletal Ni, Cu, and Fe.
(Scheme 3), which appeared as a triplet with a chemical shift of
4.5 ppm (JBH = 113 Hz).11 The quantitative nature of the
dehydrogenation in this case was confirmed through the use of
an internal NMR standard.59 Analysis of the peak integrals in the
11B NMR spectrum of the resulting mixture confirmed ∼100%
yield of aminoborane 2.
Ni Catalyst System Optimization. The properties of skeletal
Ni catalysts and their resulting hydrogenation activity can be
strongly attenuated by the conditions used during their synthesis
and the catalytic conditions subsequently employed. With re-
spect to tuning the structure of the skeletal metal, the tempera-
ture utilized during the Al leaching process has been shown to
have a significant effect with the catalyst produced at 50 °C
exhibiting the highest activity for hydrogenation reactions.60
Temperatures in excess of 50 °C have been shown to produce
lower surface area materials as a result of particle sintering, which
creates larger pore volumes and reduces catalytic activity.44 At
lower temperatures, the base leaching reaction is markedly
slower, and it has also been suggested that the Ni/Al ratio at
the surface is lower in the resulting catalysts, leading to reduced
hydrogenation activity.46,61
With a view to investigating a range of Ni catalysts for
amineꢀborane dehydrogenation, we therefore prepared two
further batches of skeletal Ni, leaching the Al content at 20 and
100 °C, respectively, to produce a total of three catalysts
henceforth referred to as NiT20, NiT50, and NiT100. Somewhat
conflicting reports exist concerning the sensitivity of the catalyst
synthesis to oxygen, with some preparations not describing the
use of deaerated solvents.53 A sample of NiT50 was therefore also
prepared using aerated solvents (denoted NiT50*) for compara-
tive purposes. The degree of Al leaching was quantified in the
cases of NiT20, NiT50, and NiT100 by atomic absorption spectros-
copy (AAS), which indicated negligible differences between the
three samples with respect to their Ni/Al contents.62
’ RESULTS AND DISCUSSION
Skeletal Ni Catalysts. A representative example of skeletal Ni
was prepared in deaerated aqueous solution via a modification of
literature methods (Scheme 2).53 This approach effectively
involved the base leaching of Al from a commercial 50:50 wt %
alloy of Ni and Al at 50 °C. The solid product was isolated by
decantation of the aqueous solution and then washed succes-
sively with deaerated ethanol and diethyl ether, before the
remaining volatile species were removed under high vacuum.
The catalyst produced under these conditions is henceforth
referred to as NiT50, where the subscript denotes the leaching
temperature.
Although skeletal Ni catalysts are often employed in aqueous
or protic media and indeed are often supplied as aqueous
slurries,54ꢀ56 this would clearly be inappropriate for the non-
hydrolytic amineꢀborane dehydrogenation of interest in this
work. Furthermore, catalytic hydrolysis and alcoholysis of amineꢀ
boranes using various heterogeneous catalysts, including skeletal
Ni, have previously been reported.57,58 In initial experiments,
toluene was therefore utilized as an alternative solvent, and
Studies of the rate of dehydrogenation of 1, using the range of
Ni catalysts with a 5 mol % catalyst loading, indicated clearly that
the catalyst produced at 50 °C under anaerobic conditions
(NiT50) led to the most rapid conversion to 2 (Figure 1). This
result is what would be expected based upon the reported surface
areas of the materials produced at the different temperatures. At
this stage, investigations utilizing NiT20 and NiT100 were discon-
the adduct Me2NH BH3 (1) selected due to its well-defined de-
3
hydrogenation chemistry, which yields the cyclic diborazane
[Me2NꢀBH2]2 (2) as the final product. Treatment of a toluene
solution of 1 with 5 mol % NiT50 produced vigorous bubbling,
with 11B NMR spectroscopy indicating complete consumption
of the adduct within 6 h. The sole product was demonstrated
by 11B NMR spectroscopy to be the expected diborazane 2
tinued, and subsequent reactions were carried out using NiT50
.
The effect of using aerated solvents in the synthesis of skeletal Ni
upon the rate of dehydrogenation of 1 was found to be
significant. NiT50* produced a significantly slower dehydrogena-
tion reaction under the same conditions employed for NiT50
,
12681
dx.doi.org/10.1021/ic201809g |Inorg. Chem. 2011, 50, 12680–12691

Inorganic Chemistry
ARTICLE
Figure 1. Conversion of 1 to 2 with time for 5 mol % loadings of various
skeletal Ni formulations in toluene solution at 20 °C. Conversions based
on integration of 11B NMR spectra.
Figure 3. Conversion of 1 to 2 with time in various solvents at 20 °C
and with 5 mol % NiT50. Conversions based on integration of 11B NMR
spectra.
activity of the skeletal Ni. As previously discussed, skeletal Ni is
often used in aqueous or protic media, with fewer literature
reports documenting its use in common organic solvents.
Kishida and Teranishi, however, investigated the direct hydro-
genation of acetone using skeletal Ni in various organic solvents,
demonstrating decreasing rates of hydrogenation in the order
n-hexane > cyclohexane > methanol and isopropanol.63 We
therefore screened the reaction of 1 with 5 mol % NiT50 in a
variety of common organic solvents and studied the initial rate of
dehydrogenation in each case (Figure 3).
The rate of dehydrogenation was shown to depend strongly
upon the reaction solvent with the least polar, hexane, producing
the most rapid dehydrogenation and reaching complete conver-
sion of 1 to 2 after 1.5 h. As the solvent polarity increased from
toluene to tetrahydrofuran (THF), the activity dropped, which is
presumably due to an increased level of competition in the
binding of solvent and amineꢀborane adduct at the Ni surface.
Although this screening indicated that the use of hexane resulted
in the most rapid reaction, it was considered inappropriate for
further investigations as it was a poor solvent for adduct 1.
Furthermore, other potential substrates, i.e. MeNH2 BH3 and
3
NH3 BH3, are completely insoluble in this solvent, which would
3
make a comparison of reaction rates difficult. Toluene was
therefore accepted as the most effective solvent for further
studies, providing a compromise between catalyst activity and
substrate solubility.
Figure 2. (i) SEM micrograph of NiT50 as prepared. (ii) EDX spectro-
scopic Ni mapping. (iii) EDX spectroscopic Al mapping. (iv) EDX
spectroscopic O mapping.
Investigation of the Heterogeneous Nature of NiT50 in the
Catalytic Dehydrogenation of 1. Although skeletal metals are
accepted heterogeneous catalysts for various processes, it was
conceivable that the metal could be leached into solution,
thereby serving as a precursor to an active homogeneous species.
To confirm the presence of a solely heterogeneous system, the
effects of filtration and Hg poisoning on the catalytic dehydro-
genation of 1 were investigated. Both procedures are accepted
tests for heterogeneous catalysts, with literature precedent for
their use.64ꢀ66 Three parallel reactions of 1 with 5 mol % NiT50
were preformed at 20 °C, and 11B NMR spectra were recorded
for each reaction every 10 min over 2 h. After 55 min, one mixture
was filtered through glass fiber paper, the second was treated with
a large excess of Hg, and the final mixture was preserved as a
control. Upon filtration, all visible bubbling in the solution
stopped and conversion of 1 to 2 was shown to cease immedi-
ately by 11B NMR spectroscopy (Figure 4a). These results are
which was presumably due to partial deactivation of the Ni
surface by oxidation. Deactivation by oxidation was also sug-
gested by the complete catalytic inactivity of a sample of NiT50
toward 1 in toluene solution after exposure of the solid catalyst to
air over 12 h.
Characterization of all catalyst formulations was performed
using scanning electron microscopy (SEM) and energy disper-
sive X-ray (EDX) spectroscopy. All catalysts showed the ex-
pected surface composition, which was dominated primarily by
Ni with lower levels of Al detected. Oxygen was also detected and
was believed to occur as a result of the unavoidable exposure of
the samples to the atmosphere during introduction to the
microscope (Figure 2). (For related data for other catalysts see
Supporting Information, Section 5).
Effect of Solvent on the Rate of Dehydrogenation of 1 with
NiT50. It was important to elucidate the effect of solvent on the
12682
dx.doi.org/10.1021/ic201809g |Inorg. Chem. 2011, 50, 12680–12691

Inorganic Chemistry
ARTICLE
Figure 4. Effects of (a) filtration and (b) Hg poisoning (after 55 min) on the NiT50 catalyzed dehydrogenation of 1 in toluene at 20 °C. Conversions
based on integration of 11B NMR spectra.
Scheme 4. Possible Mechanisms for the Metal-Catalyzed Dehydrogenation of 1
consistent with a catalyst which is insoluble and heterogeneous.
Addition of Hg to the reaction mixture provided almost
identical results, completely halting catalysis (Figure 4b). The
effect of excess Hg pointed further toward a heterogeneous
catalyst system, as Hg is known to form amalgams with
heterogeneous metal catalysts but has little effect on most
homogeneous systems.65,67,68
case of that based on titanocene, “Cp2Ti”.17,28 In the latter
system, experimental and computational evidence for me-
chanism B was elucidated, although a previous computational
study favored mechanism A.78 It has also been shown that in
some systems, most notably in that based on Ru reported by
Schneider and co-workers, that both mechanisms may operate
in parallel.13
During the dehydrogenation of 1 using NiT50 at 20 °C, there
was evidence for the formation of small quantities (<2%) of 3
within the reaction mixture, as evidenced by 11B NMR spectros-
Mechanism of Catalytic Dehydrogenation of 1 with NiT50
.
With respect to the mechanism of dehydrogenation of 1 to
form 2 by NiT50, two likely pathways appear possible. Initial
hydrogen release may occur either via an intra- or intermolecular
process, to form either dimethylaminoborane, Me2NdBH2 (3)
(Scheme 4a, mechanism A), or the linear diborazane, Me2NHꢀ
BH2ꢀNMe2ꢀBH3 (4) (Scheme 4b, mechanism B). These inter-
mediates may then undergo cycloaddition or dehydrogenative
cyclization reactions, respectively, to affordthe cyclic diborazane2.
In the case of mechanism A, H2 release occurs at the metal center
to produce two molecules of 3, which subsequently dimerize in
an off-metal process. In the case of mechanism B, H2 release
occurs in two distinct steps, both of which occur at the metal
center. In an initial step, one equivalent of H2 is lost in an
intermolecular fashion to form 4, with a subsequent intramole-
cular process releasing a further equivalent of H2 and forming the
cyclic dimer 2.
Both mechanisms A and B have been reported experimentally for
metal and FLP-induced amineꢀborane dehydrogenation, with
route A perhaps the most commonly observed,15,16,18,19,33,34,69ꢀ72
with significantsupportfrom computationalstudies.12,73ꢀ75 Route
B is less common but has been postulated as the mechanism for
thermal dehydrocoupling of amineꢀboranes11,76,77 and has been
implicated in various catalytic systems, most convincingly in the
1
copy [δB = 37.0 (t, JBH = 130 Hz)].13,19 The presence of this
species suggested that dehydrogenation via mechanism A was
likely to operate, with 3 known to dimerize in a spontaneous
fashion.69 To probe the mechanism further, a catalytic loading of
50 mol % was employed, in an attempt to increase the quantity of
3 observed in the early stages of reaction. 11B NMR spectroscopic
analysis, however, indicated not only an increase in the magni-
tude of the signal attributed to 3 but also the appearance of a new
peak consistent with the bridging BH2 environment in the linear
diborazane 4 [δB = 1.16 (t, JBH = 106 Hz, BH2)]13,79 (Figure 5).
It is likely that the signal due to the terminal BH3 moiety in the
linear diborazane 4 overlaps with that of unreacted 1, from which
it is indistinguishable in toluene.17
The unexpected appearance of peaks consistent with 4 implied
that dehydrogenation via mechanism B was also occurring to
some extent. To probe the contribution of this mechanism, a
sample of pure, independently prepared 4 was reacted with 5 mol %
NiT50. This led to 20% conversion to 2 over 6 h at 20 °C, as
evidenced by 11B NMR spectroscopy (Scheme 5a), during which
time 1 was found to react to completion under the same con-
ditions (Scheme 5b).
12683
dx.doi.org/10.1021/ic201809g |Inorg. Chem. 2011, 50, 12680–12691

Inorganic Chemistry
ARTICLE
Scheme 6. Proposed Mechanism of Dehydrogenation of 1 by
Skeletal Ni
Scheme 7. Catalytic Dehydrogenation of 5 with 5 mol %
NiT50 in Toluene at 20 °C
Figure 5. 11B NMR spectrum after 5 min of the reaction of 1 with 50
mol % NiT50 in toluene: (a) Me2NdBH2 (3); (b) Me2NHꢀBH2ꢀ
reduced rate of the dehydrogenation reaction relative to that of 1
was initially postulated to be due to the increased steric bulk of
this adduct. The ratio of 6a to 6b, at 1:48 differed from the values
previously published,31,80 where the ratios vary from 40:60 to
10:90. However, the reaction solvents and methods of prepara-
tion vary widely, and this may account for the difference in the
observed product ratios.
NMe2ꢀBH3 (4); (c) [Me2NꢀBH2]2 (2); (d) Me2NH BH3 (1) and
3
Me2NHꢀBH2ꢀNMe2ꢀBH3 (4); and (e) HB(NMe2)2.
Scheme 5. Reactions of (a) Me2NHꢀBH2ꢀNMe2ꢀBH3 (4)
and (b) Me2NH BH3 (1) with NiT50 at 20 °C in Toluene
3
To provide further insight into the origins of both the rate
reduction and the product ratios observed for 5, catalytic
dehydrogenation of the sterically encumbered adduct iPr2NH
3
BH3 (7) was attempted. Upon reaction of 7 with 5 mol % NiT50
at 20 °C, vigorous bubbling was observed which quickly ceased,
with only ∼5% conversion to iPr2NdBH2 (8) [δB 34.6 (t, JBH
=
127 Hz, BH2) ppm] being detected after 16 h. No further
increase in conversion was observed even after 120 h, however,
increasing the catalyst loading to stoichiometric quantities
yielded complete conversion of 7 to the expected aminoborane
8, as evidenced by 11B NMR spectroscopy. The limited extent
of the dehydrogenation of 7 with 5 mol % of NiT50 is not easily
explained solely on steric grounds, as although a reduced rate of
conversion to 8 would be expected, the reaction would also be
anticipated to ultimately reach completion. With a view to
probing the chemistry further, the stoichiometric reaction of 7
with NiT50 was repeated in C6D6, facilitating study of the final
It is likely, based on the reactivity of isolated 4 with NiT50, that
both mechanisms A and B operate in parallel during the reaction
of 1 with NiT50 (Scheme 6). However, the rate of reaction of 1 via
mechanism A is significantly greater than that via mechanism B,
as evidenced by the absence of peaks corresponding to 4 in the
11B NMR spectra under catalytic conditions. It is also concei-
vable, however, that at higher catalyst loadings, an increased
concentration of 3 leads to an appreciable reaction with remain-
ing amineꢀborane adduct 1, in which the metal may or may not
be involved, thereby forming 4 as discussed by Schneider and
Weller, respectively (Scheme 6).13,69 The thermodynamics of the
reaction between 3 and 1 have also been elucidated computa-
tionally by Schneider and co-workers with a reaction enthalpy of
ꢀ2.3 kcal molꢀ1, suggesting a thermodynamically feasible pro-
cess under ambient conditions.13
1
reaction mixture by H and 13C NMR spectroscopies. In this
solvent 11B NMR spectroscopy again indicated 8 as the sole
1
boron-containing product. However, the H NMR spectrum
contained 2 sets of isopropyl resonances in a 7:3 ratio, with the
larger pair being assigned to aminoborane 8 [δH 3.16 (2H,
septet, JHH = 6.79 Hz, CH) and 1.04 (12H, d, JHH = 6.79 Hz,
CH3) ppm]. The second set of isopropyl resonances corre-
sponded to those of iPr2NH in the same solvent [δH 2.81 (2H,
septet, JHH = 6.24 Hz, CH) and 0.95 (12H, d, JHH = 6.24 Hz,
CH3) ppm], suggesting BꢀN bond cleavage had occurred.
Addition of B(OiPr3)3 as an internal NMR standard enabled
quantification of the 11B content of the solution and indicated
the loss of ∼30% of the initial boron present.59 It is highly
likely, therefore, that BꢀN bond cleavage in 7 or 8 leads to
coordination of a borane moiety to the Ni surface, liberating
Reactions of NiT50 with Other Secondary AmineꢀBorane
Adducts. Following the successful catalytic dehydrocoupling of 1
with NiT50, the reactivity of this catalyst with other amineꢀ
borane substrates was also investigated. Reaction with the closely
related Et2NH BH3 (5) with 5 mol % NiT50 led to catalytic
3
dehydrocoupling over 60 h, to yield a 1:48 mixture of Et2NdBH2
(6a) [δB 36.5 (t, JBH = 125 Hz) ppm] and [Et2NꢀBH2]2 (6b)
[δB 1.5 (t, JBH = 112 Hz ppm)],80 respectively (Scheme 7). The
12684
dx.doi.org/10.1021/ic201809g |Inorg. Chem. 2011, 50, 12680–12691

Inorganic Chemistry
ARTICLE
Scheme 8. Ni-Catalyzed Dehydrogenation of 7 in Toluene
Scheme 10. Reaction of 9 with (a) Catalytic and (b) Stoi-
chiometric Quantities of NiT50 in THF at 20 °C
Solution at 20 °Ca
a NiT50BHx is postulated based on stoichiometry of other reagents/
products.
Scheme 9. Dissociation of 7 To Produce Free Amine and
BH3
Scheme 11. Proposed Mechanism of Polymerization of 9 to
Produce 1120
free amine into solution. Such a process may be responsible for
the low catalytic activity of NiT50 toward this substrate, with
BH3 poisoning the catalyst by irreversibly binding to the metal
surface (Scheme 8).
using Rh complexes.11,86,87 In contrast, however, treatment of an
identical solution of 9 in THF with a stoichiometric quantity of
NiT50 at 20 °C for 2 h led to a very broad peak in the 11B NMR
spectrum at ꢀ6 ppm, which was consistent with the formation of
poly(methylaminoborane), [MeNHꢀBH2]n (11).20 The pro-
duct was purified from the catalyst by filtration of the reaction
mixture through glass fiber paper, followed by precipitation into
hexanes at ꢀ78 °C. This procedure furnished a fine white solid in
60% yield (Scheme 10b). Multinuclear NMR spectra of the
product were consistent with those previously reported for
poly(methylaminoborane),20 and GPC analysis confirmed the
macromolecular nature of the material [Mw = 78 000 Da, PDI
(Mw/Mn) = 1.52].
Previous attempts to produce high molecular weight poly-
aminoboranes using heterogeneous Rh catalysts resulted in
relatively low molecular weight polymers. For example, polym-
erization of 9 using Rh/Al2O3 resulted in a material with Mw =
26 000 Da and a PDI = 8.5.20 The formation of high molecular
weight polymers with this heterogeneous Ni system is therefore
particularly significant, due to the facile polymer purification that
is possible as a result of the insoluble nature of the catalyst. When
homogeneous catalysts are employed, unless extensive purifica-
tion is performed, small amounts of the metal complex remain, as
evidenced by the color of the isolated polymers, which results in
the slow degradation of the material. When NiT50 is used, perfectly
white powders can be obtained, due to the absence of residual
metal catalyst. In addition and to the best of our knowledge, this is
only the second report of the synthesis of high molecular weight
polyaminoboranes by a non-precious metal catalyst.19
To probe the source of the free amine, isolated 8 was reacted
with a stoichiometric quantity of NiT50 over 16 h at 20 °C in
C6D6. Over this time period, 1H NMR spectroscopic analysis of
the reaction mixture indicated very minor elimination (∼1%) of
iPr2NH into solution, with the vast majority of 8 remaining
unreacted under these conditions. It is therefore probable that
the source of the free amine in solution is the initial adduct 7 and
that iPr2NH is liberated to solution through the dissociative
equilibrium present in the adduct (Scheme 9).81,82 In this
process, an equal quantity of BH3 must also be released to the
solution, which could then be removed from the equilibrium by
irreversible coordination to Ni.83 The fact that 7 reacts signifi-
cantly in this fashion may be explained to some degree by the
relative bond strengths in the three secondary amineꢀborane
adducts studied. Various studies, both experimental and
computational,84,85 have suggested that increasing steric bulk at
nitrogen leads to a reduction in bond enthalpy for the NꢀB σ-
bond in an amineꢀborane adduct. This is exemplified by the
bond enthalpies that are believed to decrease in the following
series in the following order: Et2NH BH3 > nPr2NH BH3 >
3
3
nBu2NH BH3.84 The weaker BꢀN bond in the case of 7 is likely
3
therefore to favor dissociation, with a greater proportion of free
amine and BH3 in solution than in the case of 5 or 1. The reduced
rate of dehydrogenation of 5 relative to 1 may possibly be
explained by a similar argument, although an increase in steric
encumbrance is also likely to reduce the rate of reaction such that
the order of adduct reactivity is 1 > 5 > 7.
The fact that the products of the reaction of 9 with NiT50 can
be tuned by altering the catalyst loading is significant. This
observation reinforces the previously reported assertion that
metal-catalyzed polymerization of 9 by IrH2POCOP occurs via
an aminoborane intermediate, MeNHdBH2, which is the true
monomer for the polymerization (Scheme 11).20 In this Ni-
based system, where the kinetics of dehydrogenation are mark-
edly slower, catalytic quantities of metal are likely to produce
only low concentrations of this monomer in solution, hence
favoring the formation of cyclic oligomers or short chain poly-
mers. By increasing the catalyst loading to stoichiometric quan-
tities, the rate of dehydrogenation should be significantly
increased. This would be expected to increase the concentration
of monomer in solution, allowing for the formation of longer
polymer chains.
Reactions of NiT50 with MeNH2 BH3 (9) and NH3 BH3 (12).
3
3
The dehydrocoupling of primary amineꢀborane adducts was
also explored, with that of MeNH2 BH3 (9) attempted using 5
3
mol % NiT50 at 20 °C. This reaction was performed in THF due
to the low solubility of 9 in toluene. On catalyst addition, the
reaction mixture was observed to bubble vigorously over 5 min,
and after 16 h ∼15% conversion to two new products was
observed by 11B NMR spectroscopy. Allowing the reaction to stir
for a further 100 h allowed the conversion to products to
reach 70%, with the major peak (with an upfield shoulder) at
ꢀ5.1 ppm now dominating as a triplet in the proton-coupled 11B
NMR spectrum (JBH = 96 Hz). Upon isolation, this product was
identified on the basis of 11B, 1H, and 13C NMR spectroscopies as
the cyclic triborazane [MeNHꢀBH2]3 (10) (Scheme 10a), a
postulated intermediate in the catalytic dehydrocoupling of 9
12685
dx.doi.org/10.1021/ic201809g |Inorg. Chem. 2011, 50, 12680–12691

Inorganic Chemistry
ARTICLE
Scheme 12. Summary of (a) Catalytic and (b) Stoichiometric
Dehydrocoupling of 12 with NiT50 at 20 °C in THF
Scheme 13. Dehydrocoupling Reactivity of Skeletal Cu (a)
Catalytic Dehydrogenation of 1; (b) Stoichiometric Dehy-
drogenation of 1; and (c) Stoichiometric Dehydropolymeri-
zation of 9
Table 1. Effect of BH3 THF on NiT50-Catalyzed (5 mol %)
Dehydrocoupling of 1 To Form 2 at 20 °C in Toluene
3
surface as a metal boride (MꢀB) or metal boryl (MꢀBH2),83
formed by catalytic decomposition of BH3. It is highly likely
therefore that NiT50 is poisoned by BH3 in much the same way, as
implicated in the catalytic dehydrogenation of 7 and by the
significantly reduced catalytic activity observed on addition of
equiv of
conversion
1 h (%)
conversion
6 h (%)
conversion
24 h (%)
BH3 THF
3
0
54
12
0
100
60
0
N/A
100
0
0.1
10
BH3 THF. For completeness, NiT50 was also treated, under
3
identical conditions, with representative examples of the remain-
ing components of a dehydrogenation reaction mixture, specifi-
cally H2 and iPr2NH. These species were found to have only a
minor effect on the rate of dehydrogenation of 1 at 20 °C when
used in 10-fold excess with respect to the catalyst.
The attempted dehydrogenation of the related and intensively
studied adduct, ammoniaꢀborane, NH3 BH3 (12),3,4,88 led
3
to slow catalytic dehydrogenation over 7 days at 20 °C using
5 mol % NiT50 to produce ∼32% conversion to a single product.
The product was assigned on the basis of 11B NMR spectroscopy
as the cyclolinear aminoborane trimer, B-(cyclodiborazanyl)-ami-
noborohydride (13) [δB ꢀ5.5 (d, JBH = 108 Hz, BH), ꢀ11.9
(t, JBH = 98 Hz, BH2),89 ꢀ24.4 (br. s, BH3) (Scheme 12a).19,72 The
use of stoichiometric quantities of NiT50, however, resulted in the
significant precipitation of a white solid over 24 h, with only small
amounts of soluble boron-containing products apparent after 7
days by 11B NMR spectroscopy. The solid, due to its complete
insolubility in common organic solvents, proved to be inseparable
from the catalyst. The solid-state IR spectrum of the product,
however, contained peaks corresponding to NꢀH (3290 and
3246 cmꢀ1) and BꢀH (2323 cmꢀ1) stretching modes, all of which
correlated well with those described for [NH2ꢀBH2]n (14) pre-
pared by an Ir-catalyzed dehydrogenation of 12.7,20 Based on the
similarity in the IR spectrum and the complete insolubility in
common organic solvents, the product was therefore tentatively
assigned as the polyaminoborane 14 (Scheme 12b), analogous to the
Attempts to recycle NiT50 through multiple catalytic dehy-
drogenations using adduct 1 were only partially successful, with a
significant loss in activity apparent after the initial dehydrogenation
cycle. On completion of the dehydrogenation of 1 in toluene at
20 °C using a 5 mol % loading of NiT50, the catalyst was isolated and
dried under high vacuum, and a further equivalent of 1 was added
following resuspension in toluene. No bubbling was observed in
solution upon addition of 1, with almost negligible dehydrocoupling
detected by 11B NMR spectroscopy after 72 h at 20 °C. When the
reaction was repeated without isolation of the catalyst, some catalytic
activity was preserved with 16% conversion to 2possible over 6 h; the
time period over which fresh catalyst completely dehydrogenates 1. It
is therefore clear that the catalytic activity of NiT50 is significantly
reduced after the completion of a single catalytic run, suggesting a
gradual surface poisoning effect either during or after the dehydro-
genation reaction.
Skeletal Catalysts of Other First-Row Transition Metals.
Among the other reported skeletal catalysts of first-row transition
metals, Cu and Fe potentially provide even more environmen-
tally benign and cheaper alternatives to skeletal Ni. Skeletal Cu,
which has received more attention than skeletal Fe, can provide
increased selectivity in catalytic hydrogenation in some cases,50
although the surface area of such catalysts are reported to be up to
an order of magnitude smaller than those of skeletal Ni.46 Complexes
of these metals have only been briefly reported with respect to
catalytic amineꢀborane dehydrogenation chemistry.15,19
soluble [MeNHꢀBH2]n (11) produced from MeNH2 BH3 (9).90
3
Catalyst Poisoning and Recycling Studies. The limited
catalytic effect of NiT50 with respect to the dehydrogenation of
7 (vide supra) suggested the possibility of catalyst poisoning by
BH3. To probe this postulate, the effect of BH3 THF on the
3
catalytic dehydrogenation of 1 was investigated. To stirred
suspensions of NiT50 at 20 °C was added 0.1 and 10 equiv of
BH3 THF, and the mixtures stirred at 20 °C for 16 h before the
3
Treatment of 1 with 5 mol % of skeletal Cu91 in toluene solution
at 20 °C was shown to produce 5% conversion to the expected
aminoborane 2 over 16 h, with no further reaction apparent over
3 days under the same conditions (Scheme 13a). Attempted dehy-
drogenation using a stoichiometric quantity of Cu did however
yield complete conversion to 2 over 4 days at 20 °C, with no loss of
boron content (Scheme 13b).59 Reaction of 7 with 5 mol % Cu
over 16 h produced less than 1% conversion to the corresponding
aminoborane 8, with 10% conversion over the same period when
Cu was employed in stoichiometric quantities and 20% after 4 days.
addition of 20 equiv of 1 (to produce an effective 5 mol % catalyst
loading). In both cases, addition of BH3 THF was found to
3
significantly reduce the activity of NiT50 toward the dehydro-
genation of 1 at 20 °C, with a partial poisoning effect being
observed with 0.1 equiv of BH3 THF and a complete loss of
3
catalytic activity with 10 equiv (Table 1).
The behavior of BH3 THF as a surface poison for hetero-
3
geneous metal catalyst systems has been previously demon-
strated for a related colloidal Rh system,83 wherein treatment
with BH3 THF led to irreversible binding of boron at the metal
3
12686
dx.doi.org/10.1021/ic201809g |Inorg. Chem. 2011, 50, 12680–12691

Inorganic Chemistry
ARTICLE
Reaction of skeletal Cu with 9 in stoichiometric quantities over
16 h furnished 11, which was identified by multinuclear NMR
spectroscopy (Scheme 13c). GPC analysis of this product
confirmed the formation of low molecular weight polymer, with
an estimated Mw of ∼10 000 Da. A more precise molecular weight,
however, could not be determined, as the peak was outside the low
end of the calibrated range for the GPC columns. Nonetheless,
ESI-MS of the product in THF/acetonitrile did confirm formation
of a linear polymer, containing the expected [MeNHꢀBH2]
repeat unit (43.05 Da) and consistent with authentic samples of
the same material (Supporting Information, Scheme SI 7).20 This
technique, however, can also be problematic in providing accurate
molecular weight values for polyaminoboranes20 and was used in
this context solely to confirm the polymeric nature of the product.
Synthesis of the analogous skeletal Fe species was achieved by
base leaching of the Al content from FeAl2 under identical
conditions to those used to produce NiT50 (vide supra). This
catalyst, however, was found to be completely inactive toward
amineꢀborane dehydrocoupling in either catalytic or stoichio-
metric quantities even at 50 °C.
’ EXPERIMENTAL SECTION
All manipulations were carried out under an atmosphere of nitrogen
gas using standard vacuum line and Schlenk techniques or under an
atmosphere of argon within an MBraun glovebox. All solvents were
dried via a Grubbs’ type solvent purification system.92 Deuterated
solvents were purchased from Sigma Aldrich and distilled from potas-
sium (C6D6) or CaH2 (CDCl3). Ni/Al alloy (50 wt %), skeletal Cu
aqueous slurry, and Me2NH BH3 (1) were purchased from Sigma
3
Aldrich, with the latter sublimed prior to use. NaOH was purchased from
Fisher Scientific. MeNH2 BH3 (9), Me2NHꢀBH2ꢀNMe2ꢀBH3 (4),
3
and iPr2NH BH3 (7) were synthesized according to literature proce-
3
dures, with the former two compounds sublimed prior to use.11
iPr2NdBH2 (8) was synthesized via literature method and purified by
distillation prior to use.80 FeAl2 was synthesized via the method of
Johnston and co-workers52 using 99.7% Al powder supplied by Strem
Chemicals, Inc. and 99.99% Fe powder supplied by ABCR-Chemicals.
NMR spectra were recorded using JEOL JNM-ECP300/400 or JNM-
LA300 spectrometers. Chemical shifts were reported relative to residual
solvent peaks (1H and 13C) or to external standards: BF3 OEt2 (11B).
3
Atomic absorption spectroscopy (AAS) was performed on a Unicam
919 AA spectrometer following initial dissolution of the samples in aqua
regia over 5 min. Al analysis was carried out using a nitrous oxide/
acetylene flame and Ni, Fe, and Cu analysis using an air/acetylene flame.
GPC samples were dissolved in the eluent (0.5 mg/mL) and filtered
(Acrodisc, PTFE membrane, 0.45 mm) before analysis. GPC chroma-
tography was performed on a Viscotek VE2001 instrument, using a flow
rate of 1 mL/min of THF containing 0.1 w/w % nBu4NBr, calibrated
using polystyrene standards. The columns used were of grade
GP5000HHR followed by GP2500HHR (Viscotek) at a constant
temperature of 30 °C, and a VE 3580 refractometer was employed. IR
spectra were measured using a Perkin-Elmer FT-IR spectrometer. SEM
was performed on a JEOL JSM 5600LV microscope with an accelerating
voltage of 30 kV and working distance of 15 mm. The instrument was
fitted with an Everhardt-Thornley secondary electron detector and an
Oxford Instruments ISIS 310 system with silicon detector and an
atmospheric thin window (ATW) for energy-dispersive X-ray analysis
(EDX). Samples were prepared in the glovebox, by mounting ∼5 mg of
skeletal metal onto a graphite disk, and were rapidly transferred to the
microscope to minimize the exposure to air. ESI mass spectra were
recorded using a cone potential of +150 V in a THF/acetonitrile mixture
on a Bruker Daltonics Apex IV Fourier transform ion cyclotron mass
spectrometer.
Synthesis of Skeletal NiT50. To a suspension of 50:50 wt % Ni/Al
(2.5 g) in deaerated distilled water at 50 °C was added, over the course of
45 min, a 6 M deaerated aqueous solution of NaOH (23.16 mL, 139.0
mmol), producing vigorous bubbling. Following the addition, the
mixture was stirred at 50 °C for a further 2 h before cooling to 20 °C
and removing the stirrer bar. The dark solid product of the reaction was
isolated by decantation and washed with distilled water under a nitrogen
flow until the washings were neutral. The solid was then washed with
deaerated EtOH (30 mL) and finally deaerated Et2O (30 mL), before
drying under high vacuum for 1 h and transferring to an inert atmo-
sphere glovebox for storage. Yield: 1.2 g, 96%.
’ CONCLUSIONS
Skeletal Ni metal has been demonstrated to be an effective
amineꢀborane dehydrogenation catalyst at 20 °C and is of
immediate interest due to the low cost and ready availability of
this metal. The catalyst was shown to be heterogeneous and
could be simply removed from complete reactions by filtration,
facilitating the ready purification of products. To the best of our
knowledge, this is the first example of a heterogeneous, non-
hydrolytic first-row amineꢀborane dehydrogenation catalyst.
The dehydrogenation of Me2NH BH3 (1) to form [Me2Nꢀ
3
BH2]2 (2) with 5 mol % skeletal Ni reached completion over
6 h at 20 °C and was shown to occur primarily via the monomeric
aminoborane Me2NdBH2 (3) as the major intermediate,
although the linear diborazane Me2NHꢀBH2ꢀNMe2ꢀBH3
(4) was also observed in lower concentrations. Catalytic dehy-
drogenation of Et2NH BH3 (5) was demonstrated at 20 °C,
3
although due to catalyst poisoning, the related adduct iPr2H
3
BH3 (7) was only completely dehydrogenated by employing the
Ni in stoichiometric quantities. Reaction of MeNH2 BH3 (9)
3
with skeletal Ni was shown to produce either the cyclic tribor-
azane [MeNHꢀBH2]3 (10) at 5 mol % catalyst loading or the
high molecular weight poly(methylaminoborane), [MeNHꢀ
BH2]n (11) (Mw = 78 000 Da, PDI = 1.52) when employing
stoichiometric quantities of Ni. This observation provided some
interesting mechanistic insight into the polymerization process,
suggesting that, as previously postulated, the transient amino-
borane, MeNHdBH2, is the true monomer in the polymeriza-
tion. Although polymerizations of 9 with heterogeneous Rh
catalysts have been reported,20 this is the first heterogeneous
system to provide access to high molecular weight polyamino-
boranes with comparatively low PDIs.
Preparation of NiT20 and NiT100 was carried out in an identical
fashion, using leaching temperatures of 20 and 100 °C, respectively. The
leaching of NiT20 was, however, carried out over 72 h due to the reduced
rate of this reaction.
Synthesis of Skeletal Fe. To a suspension of FeAl2 (1 g,
9.10 mmol) in deaerated distilled water (10 mL) was added, over
30 min, a 6 M deaerated aqueous solution of NaOH (10 mL) at 50 °C.
Following the addition, the mixture was stirred at 50 °C for a further 1.5
h before cooling to 20 °C, removing the stirrer bar and isolating the solid
products via decantation. The product was then washed with deaerated
Recycling of the Ni-based catalysts was found to be difficult,
with catalytic activity rapidly lost due to poisoning. An investiga-
tion of various potential poisons suggested that BH3 was the
likely surface poison, as has been observed previously for
heterogeneous Rh systems. Extending the study to include
skeletal Cu and Fe was less successful. Skeletal Cu was found
to exhibit a similar pattern of reactivity to Ni, although with vastly
reduced rates, and skeletal Fe was found to be inactive toward
amineꢀborane dehydrogenation.
12687
dx.doi.org/10.1021/ic201809g |Inorg. Chem. 2011, 50, 12680–12691

Inorganic Chemistry
ARTICLE
distilled H2O until the washings became neutral and subsequently with
deaerated EtOH (30 mL) and Et2O (30 mL), before drying under high
vacuum for 2 h. Yield: 0.47 g, 92%.
Drying of Skeletal Cu. An aqueous slurry of skeletal Cu, purchased
from Sigma Aldrich, was dried under the same regime described for
skeletal Ni and transferred to the glovebox for storage.
Attempted Catalytic Dehydrogenation of iPr2NH BH3 (7)
3
with NiT50. To a solution of 7 (0.19 g, 1.7 mmol) in toluene (4 mL) was
added 5 mol % NiT50 (5.0 mg, 0.09 mmol) at 20 °C, which initially
produced vigorous bubbling. The reaction mixture was then stirred for
16 h at 20 °C before analysis by 11B NMR spectroscopy indicated ∼5%
conversion to iPr2NdBH2 (8), with no further conversion apparent
after 120 h at 20 °C. 11B NMR (96 MHz, toluene): δ 34.6 (t, JBH = 127
Catalytic Dehydrogenation of Me2NH BH3 (1) with NiT50
.
3
Hz, iPr2NdBH2), ꢀ21.2 (q, JBH = 96 Hz, iPr2NH BH3).
To a solution of 1 (100 mg, 1.70 mmol) in toluene (4 mL) was added
5 mol % NiT50 (5 mg, 0.09 mmol), producing immediate and
vigorous bubbling. The solution was then stirred at 20 °C for 6 h,
producing complete conversion to [Me2NꢀBH2]2 (2). The product
was not isolated, but the reaction was shown to be quantitative
through the addition of B(OiPr3) (49 μL, 0.21 mmol) to the filtered
0.5 mL NMR sample. Integration of the signal arising from 2 relative
to the borate (δB +18 ppm) indicated that 100% of the initial boron
content was present as this species. 11B NMR (96 MHz, toluene): δB
4.5 (t, JBH = 113 Hz) ppm. Repeating the reaction in C6D6 enabled
the recording of further NMR data: 1H NMR (300 MHz, C6D6): δ
3.03 (br. q, JBH = 113 Hz, BH2), 2.23 (s, NMe2), 13C NMR (76 MHz,
C6D6): δ 52.0 (s, NMe2).
3
Stoichiometric Dehydrogenation of 7 with NiT50. To a
solution of 7 (0.19 g, 1.7 mmol) in C6D6 (4 mL) was added 100 mol %
NiT50 (0.10 g, 1.7 mmol) at 20 °C, which initially produced vigorous
bubbling. The reaction mixture was then stirred for 6 h at 20 °C before
analysis by 11B NMR spectroscopy indicated 100% conversion of the
soluble boron content to 8. Addition of a stoichiometric quantity of
B(OiPr)3 to the NMR samples however indicated that at least 30% of the
initial boron was not present in solution. 1H and 13C NMR spectroscopies
indicatedthepresenceof iPr2NH(∼30%) insolution, suggestingcleavage
of either the adduct or the aminoborane product. 11B NMR (96 MHz,
1
C6D6): δ 34.5 (t, JBH = 125 Hz, iPr2NdBH2); H NMR (300 MHz,
C6D6): δ 3.16 (2H, septet, JHH = 6.79 Hz, iPr2NdBH2), 2.81 (2H0,
septet, JHH = 6.24, iPr2NH), 1.04 (6H, d, JHH = 6.79 Hz, iPr2NdBH2),
0.95 (6H0, d, JHH = 6.24 Hz, iPr2NH) (The apostrophes used in the above
assignments denote the two different compounds present within the
reaction mixture, for which the resonances were integrated separately);
13C NMR (76 MHz, C6D6): δ 52.8 (s, iPr2NdBH2 (CH)), 45.7 (s,
iPr2NH, (CH)), 25.4 (s, iPr2NdBH2 (CH3)), 24.1 (s, iPr2NH, (CH3)).
Effect of Solvent on the NiT50 Catalyzed Dehydrogenation
of 1. Reaction mixtures were prepared, as for the dehydrogenation
described above, using a 1.70 mmol quantity of 1 in tetrahydrofuran and
hexane. A 5 mol % loading of NiT50 was added at 20 °C, and the mixtures
monitored every 20 min by 11B NMR spectroscopy. In the case of the
hexane reaction, it was necessary to add a stoichiometric quantity of
B(OiPr)3 (29.4 μL) to each 0.3 mL NMR sample to quantify the
conversion, due to the limited solubility of 1 in this solvent.
Tests for Catalyst Heterogeneity. To three separate solutions of
1 (0.1 g, 1.7 mmol) in toluene (4 mL) were added 5 mol % NiT50 (5 mg,
0.09 mmol), the solution stirred rapidly, and ∼0.7 mL of each sample
transferred to a J. Youngs NMR tube. The reactions were then allowed to
proceed as open systems, by opening to an atmosphere of N2 on a
Schlenk line, so that the H2 formed could easily escape. 11B NMR spectra
were recorded every 10 min for 50 min, and conversion to products
calculated based upon integration of the resonances. After 55 min, one of
the reactions was filtered through glass fiber filter paper (Fisher MF-100
grade), before transferring to a fresh J. Youngs NMR tube, while to the
second excess (a drop) mercury was added. The third reaction was left
unchanged. 11B NMR spectra were subsequently recorded every 10 min
for a further 1 h.
Catalytic Dehydrogenation of MeNH2 BH3 (9) with NiT50
.
3
To a solution of 9 (0.20 g, 4.46 mmol) in THF (2 mL) was added 5 mol %
NiT50 (0.013 g, 0.22 mmol) which produced significant bubbling, which
gradually subsided over 5 min. The mixture was stirred at 20 °C for 120 h
before analysis by 11B NMR spectroscopy study indicated 70% conver-
sion of the starting material to [MeNHꢀBH2]3 (10). The volatiles were
then removed under high vacuum to yield an oily white solid, which was
redissolved in CD2Cl2 for further analysis, by 1H and 13C NMR
spectroscopies. 11B NMR (96 MHz, CD2Cl2): δ ꢀ5.7 (t, JBH = 106
Hz, [MeNHꢀBH2]3, e,e,a) ꢀ6.7 (t, JBH = 107 Hz, [MeNHꢀBH2]3, e,e,
e), ꢀ19.6 (q, JBH = 97 Hz, MeNH2 BH3); 1H NMR (300 MHz,
3
CD2Cl2): δ 2.42 (t, JHH = 6.42 Hz, MeNH2 BH3), 2.28 (d, JHH = 6.42
3
Hz, [MeNHꢀBH2]3), 2.12 (d, JHH = 6.06, [MeNHꢀBH2]3); 13C
NMR (76 MHz, CD2Cl2): δ 38.6 (s, [MeNHꢀBH2]3, e,e,a), 36.4 (s,
[MeNHꢀBH2]3, e,e,a), 35.3 (s, [MeNHꢀBH2]3, e,e,e), 34.8 (s,
MeNH2 BH3). The terms e,e,e and e,e,a used in the above assignments
Synthesis of Et2NH BH3 (5). A mixture of [Et2NH2]Cl (7.50 g,
3
3
refer to the orientation of the Me groups on the six-membered ring
which adopts a chair conformation: e,e,e denotes the triply equatorial
isomer, and e,e,a the second isomer where two Me groups are found
equatorial and the third axial.
68.4 mmol) and NaBH4 (2.61 g, 68.9 mmol) was cooled to ꢀ78 °C, and
THF (200 mL), which was also cooled to ꢀ78 °C, was added dropwise
over 15 min. The reaction was then stirred for 30 min before removing
the cooling bath and allowing to mixture to warm to 20 °C. During the
warming process, vigorous gas evolution was observed, and after stirring
for a further 2 h, the reaction mixture was filtered though Celite. Volatiles
were then removed under high vacuum, and the mixture was filtered
through glass fiber filter paper to yield the product as a colorless oil.
Yield: 3.55 g, 60%. 11B NMR (96 MHz, CDCl3): δ ꢀ17.9 (q, JBH = 94
Polymerization of 9 with NiT50. To a solution of 9 (0.50 g, 11.1
mmol) in THF (6 mL) was added NiT50 (0.65 g, 11.1 mmol), which
produced vigorous bubbling. The mixture was stirred at 20 °C for 2 h
before filtering through a glass fiber filter paper and subsequently
precipitating into hexanes (100 mL) at ꢀ78 °C. The solids so formed
were isolated via filter cannula and washed twice with hexanes (50 mL),
before drying under high vacuum for 4 h, to yield the product,
[MeNHꢀBH2]n (11), as a white solid. Yield: 0.29 g, 60%, 11B NMR (96
MHz, CDCl3): δ ꢀ6.5 (br. s, BH2); 1H NMR (300 MHz, CDCl3): δ 2.87
(1H, br. s, NH), 2.26 (3H, br. s, Me), 1.75 (2H, br. s, BH2); 13C NMR (76
MHz, CDCl3): δ 36.0 (s, CH3), GPC: Mw = 78 000 Da, PDI = 1.52.
1
Hz, BH3); H NMR (300 MHz, CDCl3): δ 3.55 (1H, br. s, NH),
2.76 (4H, dq, JHH = 7.34 Hz, JHH = 6.00, CH2), 1.40 (3H, br. q, BH3),
1.19 (6H, t, JHH = 7.34 Hz, CH3); 1H {11B} NMR (300 MHz, CDCl3): δ
3.55 (1H, br. s, NH), 2.76 (4H, dq, JHH = 7.34 Hz, JHH = 6.00, CH2),
1.40 (3H, br. s, BH3), 1.19 (6H, t, JHH = 7.34 Hz, CH3); 13C NMR (76
MHz, CDCl3): δ 48.9 (s, CH3), 11.3 (s, CH2).
Catalytic Dehydrogenation of 5 with NiT50. To a solution of
5 (0.15 g, 1.7 mmol) in toluene (4 mL) was added 5 mol % NiT50
(5.0 mg, 0.09 mmol) at 20 °C, which initially produced vigorous
bubbling. The reaction was then stirred for 60 h at 20 °C before
sampling for 11B NMR spectroscopic analysis. 11B NMR (96 MHz,
toluene): δ 36.5 (t, JBH = 125 Hz, Et2N=BH2), 1.5 (t, JBH = 111 Hz,
[Et2NꢀBH2]2).
Catalytic Dehydrogenation of NH3 BH3 (12) with NiT50.
3
To a solution of 12 (0.2 g, 6.4 mmol) in THF (16 mL) was added
5 mol % NiT50 (0.02 g, 0.32 mmol) which produced significant bubbling,
which gradually subsided over 5 min. The mixture was stirred at
20 °C for 7 days before the clear solution was analyzed by 11B NMR
spectroscopy, which indicated 32% conversion to B-(cyclodibora-
zanyl)aminoborohydride (13). 11B NMR (96 MHz, THF): δ ꢀ5.5
12688
dx.doi.org/10.1021/ic201809g |Inorg. Chem. 2011, 50, 12680–12691

Inorganic Chemistry
ARTICLE
(d, 108 Hz, BH), ꢀ11.9 (t, JBH = 98 Hz, BH2),89 ꢀ22.8 (q, JBH = 97 Hz,
same scale using 5 and 100 mol % skeletal Fe (5 mg, 0.09 mmol, and
100 mg, 1.7 mmol). Under catalytic and stoichiometric Fe loadings, no
evidence of dehydrogenation of the adduct was apparent by 11B NMR
spectroscopy over 4 days at 20 °C or at 50 °C.
NH3 BH3), ꢀ24.4 (br. s, BH3).
3
Polymerization of 12 with NiT50. To a solution of 12 (0.2 g,
6.5 mmol) in THF (16 mL) was added 100 mol % NiT50 (0.38 mg,
6.5 mmol), which produced vigorous initial bubbling. The rapid forma-
tion of a white precipitate was observed, and the mixture was stirred at
20 °C over 7 days. After this time, minimal 12 (or other soluble species)
was apparent in the 11B NMR spectrum of the supernatant liquid. The
liquid was decanted from the reaction mixture, and the solid mixture of
product and catalyst dried under high vacuum over 16 h. The solid
components proved to be inseparable due to its complete insolubility in
all common organic solvents. Solid-state IR of the product/catalyst
mixture however indicated a close correlation with [NH2ꢀBH2]n (14),
previously identified via reaction of 12 with IrH2POCOP. FT-IR: 3290,
’ ASSOCIATED CONTENT
S
Supporting Information. Supporting information con-
b
tains selected spectra, SEM/EDX characterization of the various
catalyst formulations and also details of additional experiments.
This material is available free of charge via the Internet at http://
pubs.acs.org.
3246 (NꢀH), 2323 (BꢀH) cmꢀ1
.
’ AUTHOR INFORMATION
Attempted Recycling of NiT50. To a solution of 1(0.1g, 1.7mmol)
in toluene (4 mL) was added NiT50 (5 mg, 0.09 mmol), and the mixture
was stirred at 20 °C for 16 h. The mixture was then analyzed by 11B NMR
spectroscopy, which indicated complete conversion of the starting material
to yield the cyclic aminoborane 2. The solution was then removed via
syringe, and the solid catalyst was dried under high vacuum over 4 h.
The solid was then resuspended in toluene (4 mL), and further 1 (0.1 g, 1.7
mmol) was added. The mixture was then stirred at 20 °C over72 h, with no
evidence of conversion of 1 by 11B NMR spectroscopy over this period.
An alternative methodology for recycling was then attempted, with
dehydrogenation of 1 (0.1 g, 1.7 mmol) again run to completion over
16 h. In this case, however, the catalyst was not isolated from the mixture,
and solid 1 (0.1 g, 1.7 mmol) was then added. This resulted in 16%
conversion of the second batch of 1 over a period of 6 h.
Corresponding Author
*E-mail: ian.manners@bristol.ac.uk.
’ ACKNOWLEDGMENT
A.P.M.R., L.C. and I.M. acknowledge the EPSRC for funding.
We thank Dr. Joe Gilroy for his useful suggestions with respect to
this project. We thank Florian W€achter at ETH Zurich for the syn-
thesis of FeAl2 and Thomas Davies and Prof. Graham Hutchings at
the Cardiff Institute of Catalysis for useful discussions.
’ REFERENCES
(1) Burkhardt, E. R.; Matos, K. Chem. Rev. 2006, 106, 2617.
(2) Kanth, J. V. B. Aldrichimica Acta 2002, 35, 57.
(3) Stephens, F. H.; Pons, V.; Baker, R. T. Dalton Trans. 2007, 2613.
(4) Staubitz, A.; Robertson, A. P. M.; Manners, I. Chem. Rev. 2010,
110, 4079.
NiT50 Poisoning Studies. To suspensions of NiT50 (5 mg, 0.09
mmol) in toluene (4 mL) was added 0.9 mmol of iPr2NH, BH3 THF
3
(1 M in THF), and the mixtures were stirred for 16 h at 20 °C. Solid 1
(0.1 g, 1.70 mmol) was then added, and each mixture was sampled
after 6 and 24 h for analysis by 11B NMR spectroscopy. An identical
(5) Jiang, Y.; Blacque, O.; Fox, T.; Frech, C. M.; Berke, H.
Organometallics 2009, 28, 5493.
(6) Sloan, M. E.; Staubitz, A.; Lee, K.; Manners, I. Eur. J. Org. Chem.
2011, 672.
(7) Staubitz, A.; Soto, A. P.; Manners, I. Angew. Chem., Int. Ed. 2008,
47, 6212.
experiment using 0.009 mmol of BH3 THF was also carried out under
3
the same conditions. In the case of poisoning with H2, the gas was
bubbled through a toluene suspension of NiT50 over 16 h before
addition of 1.
Attempted Catalytic and Stoichiometric Dehydrogena-
tion of 1 with Skeletal Cu. To a solution of 1 (0.1 g, 1.7 mmol) in
toluene (4 mL) was added 5 or 100 mol % skeletal Cu (5.3 mg,
0.09 mmol or 0.11 g, 1.7 mmol), and the mixture was stirred at 20 °C for
16 h before analysis by 11B NMR spectroscopy. Over this period, the
catalytic reaction was found to yield 5% 2, whereas the stoichiometric
reaction produced 37%. The catalytic reaction produced no further
conversion by 11B NMR spectroscopy over 4 d, with the stoichiometric
reaction reaching complete conversion to 2 over the same period.
Polymerization of 9 with Skeletal Cu. To a solution of 9
(0.20 g, 4.46 mmol) in THF (10 mL) was added skeletal Cu (0.28 g, 4.46
mmol), and the mixture was stirred at 20 °C for 16 h. The mixture was
then analyzed by 11B NMR spectroscopy, indicating 80% conversion of
the starting material (δB = ꢀ19.4 ppm) to a new product (δB = ꢀ5.7
ppm). The mixture was added to rapidly stirred hexanes (100 mL)
at ꢀ78 °C, and the resulting solids were isolated by filter cannula. The
product was then dried under high vacuum for 4 h and subsequently
identified as 11. Yield: 68 mg, 36%, 11B NMR (96 MHz, CDCl3): δ ꢀ5.7
(8) Corso, M.; Auw€arter, W.; Muntwiler, M.; Tamai, A.; Greber, T.;
Osterwalder, J. Science 2004, 303, 217.
(9) Liu, Z.; Song, L.; Zhao, S.; Huang, J.; Ma, L.; Zhang, J.; Lou, J.;
Ajayan, P. M. Nano Lett. 2011, 11, 2032.
(10) Zhong, B.; Song, L.; Huang, X. X.; Wen, G. W.; Xia, L. Mater.
Res. Bull. 2011, 46, 1521.
(11) Jaska, C. A.; Temple, K.; Lough, A. J.; Manners, I. J. Am. Chem.
Soc. 2003, 125, 9424.
(12) Blaquiere, N.; Diallo-Garcia, S.; Gorelsky, S. I.; Black, D. A.;
Fagnou, K. J. Am. Chem. Soc. 2008, 130, 14034.
(13) Friedrich, A.; Drees, M.; Schneider, S. Chem.—Eur. J. 2009,
15, 10339.
(14) K€aβ, M.; Friedrich, A.; Drees, M.; Schneider, S. Angew. Chem.,
Int. Ed. 2009, 48, 905.
(15) Keaton, R. J.; Blacquiere, J. M.; Baker, R. T. J. Am. Chem. Soc.
2007, 129, 1844.
(16) Pun, D.; Lobkovsky, E.; Chirik, P. J. Chem. Commun.
2007, 3297.
(17) Sloan, M. E.; Staubitz, A.; Clark, T. J.; Russell, C. A.; Lloyd-
Jones, G. C.; Manners, I. J. Am. Chem. Soc. 2010, 132, 3831.
(18) Kakizawa, T.; Kawano, Y.; Naganeyama, K.; Shimoi, M. Chem.
Lett. 2011, 40, 171.
(19) Vance, J. R.; Robertson, A. P. M.; Lee, K.; Manners, I. Chem.—Eur.
J. 2011, 17, 4099.
1
(br. s, [MeNHꢀBH2]n), ꢀ19.4 (q, JBH = 96 Hz, MeNH2 BH3); H
3
NMR (300 MHz, CDCl3): δ 2.87 (br. s, NH2), 2.57 (s, MeNH2 BH3),
3
2.25 (br. s, Me), 1.70 (br. s, BH2); 13C NMR (76 MHz, CDCl3):
δ 35.8 (s, CH3), GPC: Mw ∼10 000 Da (see Supporting Information,
Figure SI 6).
Attempted Dehydrogenation of 1 with Skeletal Fe. Using
an identical procedure to that described above for Cu-catalyzed dehy-
drogenation of 1, skeletal Fe-catalyzed reactions were attempted on the
(20) Staubitz, A.; Sloan, M. E.; Robertson, A. P. M.; Friedrich, A.;
Schneider, S.; Gates, P. J.; Schmedt auf der G€unne, J.; Manners, I. J. Am.
Chem. Soc. 2010, 132, 13332.
12689
dx.doi.org/10.1021/ic201809g |Inorg. Chem. 2011, 50, 12680–12691

Inorganic Chemistry
ARTICLE
(21) Dietrich, B. L.; Goldberg, K. I.; Heinekey, D. M.; Autrey, T.;
Linehan, J. C. Inorg. Chem. 2008, 47, 8583.
(56) Aversa, M. C.; Giannetto, P. J. Chem. Soc. Perkin Trans. 2
1984, 81.
(22) Dallanegra, R.; Robertson, A. P. M.; Chaplin, A. B.; Manners, I.;
Weller, A. S. Chem. Commun. 2011, 3763.
(23) Zahmakiran, M.; Ozkar, S. Inorg. Chem. 2009, 48, 8955.
(24) Tang, C. Y.; Thompson, A. L.; Aldridge, S. J. Am. Chem. Soc.
2010, 132, 10578.
(25) Shrestha, R. P.; Diyabalanage, H. V. K.; Semelsberger, T. A.;
Ott, K. C.; Burrell, A. K. Int. J. Hydrogen Energy 2009, 34, 2616.
(26) Hansmann, M. M.; Melen, R. L.; Wright, D. S. Chem. Sci. 2011,
2, 1554.
(57) Couturier, M.; Tucker, J. L.; Andresen, B. M.; Dubꢁe, P.; Negri,
J. T. Org. Lett. 2001, 3, 465.
€
(58) Yao, C. F.; Zhuang, L.; Cao, Y. L.; Ai, X. P.; Yang, H. X. Int. J.
Hydrogen Energ. 2008, 33, 2462.
(59) The peaks within the 11B NMR spectra were quantified by the
addition of a known quantity of B(OiPr)3 to filtered NMR samples (See
ref 19) and integrating versus the resulting peak (δB +18 ppm). The
borate could not be added directly to the reaction mixture as it was found
to interact with skeletal Ni, reducing its dehydrogenation activity.
(60) Freel, J.; Pieters, W. J. M.; Anderson, R. B. J. Catal. 1969, 14,
247.
(27) Cowley, H. J.; Holt, M. S.; Melen, R. L.; Rawson, J. M.; Wright,
D. S. Chem. Commun. 2011, 2682.
(28) Clark, T. J.; Russell, C. A.; Manners, I. J. Am. Chem. Soc. 2006,
128, 9582.
(61) Freel, J.; Robertson, S. D.; Anderson, R. B. J. Catal. 1970, 18,
243.
(29) Vogt, M.; deBruin, B.; Berke, H.; Trincado, M.; Gr€utzmacher, H.
(62) AAS analysis: 50:50 wt % Ni/Al alloy, as purchased (Sigma-
Chem. Sci., 2011, 2, 723.
Aldrich): Ni, 53.48%; Al, 45.64%; NiT20: Ni, 89.34%; Al, 3.62%; NiT50
:
(30) A catalytic system based upon a metalꢀorganic framework
containing active Ni centers has also been reported for ammoniaꢀ
borane dehydrogenation, although the catalysis in this case occurs in the
solid state at 80 °C: Li, Y.; Song, P.; Zheng, J.; Li, X. Chem.—Eur. J. 2010,
16, 10887.
(31) Kawano, Y.; Uruichi, M.; Shimoi, M.; Taki, S.; Kawaguchi, T.;
Kakizawa, T.; Ogino, H. J. Am. Chem. Soc. 2009, 131, 14946.
(32) An FeB nanoalloy has also been employed in the catalytic
dehydrogenation of ammonia-borane in the solid state at elevated
temperatures:He, T.; Wang, J.; Wu, G.; Kim, H.; Proffen, T.; Wu, A.;
Li, W.; Liu, T.; Xiong, Z.; Wu, C.; Chu, H.; Guo, J.; Autrey, T.; Zhang, T.;
Chen, P. Chem.—Eur. J. 2010, 16, 12814.
(33) Miller, A. J. M.; Bercaw, J. E. Chem. Commun. 2010, 1709.
(34) Whittell, G. R.; Balmond, E. I.; Robertson, A. P. M.; Patra, S. K.;
Haddow, M. F.; Manners, I. Eur. J. Inorg. Chem. 2010, 3967.
(35) Chapman, A. M.; Haddow, M. F.; Wass, D. F. J. Am. Chem. Soc.
2011, 133, 8826.
(36) Haaland, A. Angew. Chem., Int. Ed. 1989, 28, 992.
(37) G€ottker-Schnetmann, I.; White, P. S.; Brookhart, M. Organo-
metallics 2004, 23, 1766.
Ni, 87.45%; Al, 7.95%; NiT100: Ni, 85.15%; Al, 5.53%. We assign the
missing mass in each sample to oxygen, likely to be present initially as a
surface oxide following dissolution of the samples in aqua regia in air.
(63) Kishida, S.; Teranishi, S. J. Catal. 1968, 12, 90.
(64) Widegren, J. A.; Finke, R. G. J. Mol. Catal. A: Chem. 2003,
198, 317.
(65) Jaska, C. A.; Manners, I. J. Am. Chem. Soc. 2004, 126, 9776.
(66) Widegren, J. A.; Bennett, M. A.; Finke, R. G. J. Am. Chem. Soc.
2003, 125, 10301.
(67) Anton, D. R.; Crabtree, R. H. Organometallics 1983, 2, 855.
(68) Whitesides, G. M.; Hackett, M.; Brainard, R. L.; Lavalleye, J. P.
P. M.; Sowinski, A. F.; Izumi, A. N.; Moore, S. S.; Brown, D. W.; Staudt,
E. M. Organometallics 1985, 4, 1819.
(69) Stevens, C. J.; Dallanegra, R.; Chaplin, A. B.; Weller, A. S.;
MacGregor, S. A.; Ward, B.; McKay, D.; Alcaraz, G.; Sabo-Etienne, S.
Chem.—Eur. J. 2011, 17, 3011.
(70) Liptrot, D. J.; Hill, M. S.; Mahon, M. F.; MacDougall, D. J.
Chem.—Eur. J. 2010, 16, 8508.
(71) Chen, Y.; Fulton, J. L.; Linehan, J. C.; Autrey, T. J. Am. Chem.
Soc. 2005, 127, 3254.
(38) Denney, M. C.; Pons, V.; Hebden, T. J.; Heinekey, D. M.;
Goldberg, K. I. J. Am. Chem. Soc. 2006, 128, 12048.
(39) Clapham, S. E.; Hadzovic, A.; Morris, R. H. Coord. Chem. Rev.
2004, 248, 2201.
(40) Li, T.; Churlaud, R.; Lough, A. J.; Abdur-Rashid, K.; Morris,
R. H. Organometallics 2004, 23, 6239.
(41) Abdur-Rashid, K.; Clapham, S. E.; Hadzovic, A.; Harvey, J. N.;
Lough, A. J.; Morris, R. H. J. Am. Chem. Soc. 2002, 124, 15104.
(42) Raney, M. Method of Preparing Catalytic Material, U.S.
1563587, 1925.
(72) Pons, V.; Baker, R. T.; Szymczak, N. K.; Heldebrant, D. J.;
Linehan, J. C.; Matus, M. H.; Grant, D. J.; Dixon, D. A. Chem. Commun.
2008, 6597.
(73) Paul, A.; Musgrave, C. B. Angew. Chem., Int. Ed. 2007, 46, 8153.
(74) Yang, X.; Hall, M. B. J. Am. Chem. Soc. 2008, 130, 1798.
(75) Zimmerman, P. M.; Paul, A.; Zhang, Z.; Musgrave, C. B. Angew.
Chem. Int Ed. 2009, 48, 2201.
(76) Ryschkewitsch, G. E.; Wiggins, J. W. Inorg. Chem. 1970, 9, 314.
(77) Douglas, T. M.; Chaplin, A. B.; Weller, A. S.; Yang, X.; Hall,
M. B. J. Am. Chem. Soc. 2009, 131, 15440.
(43) Raney, M. Method of Producing Finely Divided Nickel, U.S.
1628190, 1927.
(44) Smith, A. J.; Trimm, D. L. Annu. Rev. Mater. Res. 2005, 35, 127.
(45) Fouilloux, P. Appl. Catal. 1983, 8, 1.
(46) Ertl, G.; Kn€ozinger, H.; Weitkamp, J. Preparation of Solid
Catalysts; Wiley-VCH: Weinheim, Germany, 1999.
(47) Jones, W. H.; Benning, W. F.; Davis, P.; Mulvey, D. M.; Pollak,
P. I.; Schaeffer, J. C.; Tull, R.; Weinstock, L. M. Ann. N.Y. Acad. Sci. 1969,
158, 471.
(48) Onuoha, N. I.; Wainwright, M. S. Chem. Eng. Commun. 1984, 29, 1.
(49) Tomsett, A. D.; Wainwright, M. S.; Young, D. J. Appl. Catal.
1984, 12, 43.
(78) Luo, Y.; Ohno, K. Organometallics 2007, 26, 3597.
(79) N€oth, H.; Thomas, S. Eur. J. Inorg. Chem. 1999, 1373.
(80) Pasumansky, L.; Haddenham, D.; Clary, J. W.; Fisher, G. B.;
Goralski, C. T.; Singaram, B. J. Org. Chem. 2008, 73, 1898.
(81) Baldwin, R. A.; Washburn, R. M. J. Org. Chem. 1961, 26, 3549.
(82) Brown, H. C.; Schlesinger, H. I.; Cardon, S. Z. J. Am. Chem. Soc.
1942, 64, 325.
(83) Jaska, C. A.; Clark, T. J.; Clendenning, S. B.; Grozea, D.; Turak,
A.; Lu, Z.-H.; Manners, I. J. Am. Chem. Soc. 2005, 127, 5116.
(84) Flores-Segura, H.; Torres, L. A. Struct. Chem. 1997, 8, 227.
(85) Staubitz, A.; Besora, M.; Harvey, J. N.; Manners, I. Inorg. Chem.
2008, 47, 5910.
(50) Ma, L.; Tran, T.; Wainwright, M. S. Top. Catal. 2003, 22, 287.
(51) Reeve, W.; Christian, J. J. Am. Chem. Soc. 1956, 78, 860.
(52) Johnston, W. D.; Heikes, R. R.; Petrolo, J. J. Am. Chem. Soc.
1957, 79, 5388.
(86) Jaska, C. A.; Temple, K.; Lough, A. J.; Manners, I. Chem.
Commun. 2001, 962.
(87) Dallanegra, R.; Chaplin, A. B.; Tsim, J.; Weller, A. S. Chem.
Commun. 2010, 3092.
(53) Pavlic, A. A.; Adkins, H. J. Am. Chem. Soc. 1946, 68, 1471.
(54) Franck-Neumann, M.; Geoffroy, P.; Bissinger, P.; Adelaide, S.
Tetrahedron Lett. 2001, 42, 6401.
(55) Gallezot, P.; Cerino, P. J.; Blanc, B.; Flꢀeche, G.; Fuertes, P.
J. Catal. 1994, 146, 93.
(88) Smythe, N. C.; Gordon, J. C. Eur. J. Inorg. Chem. 2010, 509.
(89) This peak, based on integration versus the remaining peaks of
the trimer, may potentially also contain some contribution from its
isomeric form, [NH2ꢀBH2]3, which has been reported at an almost
identical similar chemical shift: Shaw, W. J.; Linehan, J. C.; Szymczak,
12690
dx.doi.org/10.1021/ic201809g |Inorg. Chem. 2011, 50, 12680–12691

Inorganic Chemistry
ARTICLE
N. K.; Heldebrant, D. J.; Yonker, C.; Camaioni, D. M.; Baker, R. T.;
Autrey, T. Angew. Chem., Int. Ed. 2008, 47, 7493.
(90) Although assignments of the products in the case of reac-
tions with NH3 BH3 are difficult, the close relationship between the
3
behaviour of this adduct and MeNH2 BH3 is described elsewhere (see
3
refs 11 and 20). Therefore, the assignments are also reinforced by the
analogous reactivity of the methylated adduct, which also produces
trimeric products under catalytic conditions and polymers under
stoichiometric conditions.
(91) Purchased from Sigma Aldrich, Ltd. as an aqueous slurry and
dried as described for NiT50
.
(92) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518.
12691
dx.doi.org/10.1021/ic201809g |Inorg. Chem. 2011, 50, 12680–12691