Homogeneous catalytic dehydrogenation/dehydrocoupling of amine-borane adducts by the Rh(I) Wilkinson's complex analogue RhCl(PHCy2) 3 (Cy ) cyclohexyl)

Article abstract of DOI:10.1021/ic801752k

The Rh(l) complex RhCI(PHCy₂)₃ (1) (Cy = cyclohexyl, C₆H₁₁) has been investigated as a catalyst for the dehydrogenation/dehydrocoupling of dimethylamine-borane adduct Me ₂NH·BH₃ (3) at 20 °C to afford the cyclic dimer [Me₂N-BH₂]₂ (4). Unlike previously studied neutral and Cationic Rh(l) precatalysts such as [{Rh(μ-CI)(1,5-cod)} ₂] and [Rh(1,5-cod)₂]OTf (1,5-cod = 1,5-cyclooctadiene, C₈H₁₂, OTf = OSO₂CF₃) with weakly electron-donating ligands at the metal center, which are reduced to catalytically active Rh(0) species, catalytic dehydrogenation of 3 using 1 was found to occur in a homogeneous manner according to nanofiltration experiments, Hg(0) poisoning and kinetic studies. Moreover, the presence of the sterically bulky ligand PHCy₂ in complex 1 has been found to significantly increase the rate of reaction previously reported for Wilkinson's catalyst RhCI(PPh₃)₃. The catalytic activity of 1 toward a range of other amine-borane adducts RR'NH · BH₃ (e.g., RR' = 'Pr ₂, MeBz, MeH) at 20 °C was also investigated. The third row transition metal analogue of 1, the lr(l) complex lrCI(PHCy₂) ₃ (2), was also explored as a catalyst for the dehydrocoupling of 3 and was found to exhibit much reduced catalytic activity compared to 1 but proved significantly more active for sterically encumbered substrates. Addition of the strong Lewis acid B(C₆F₅)₃ as a co-catalyst to both 1 and 2 has been found to significantly increase the rate of the dehydrocoupling reactions in all cases. The Rh(l) complex 1 (but not the lr(l) analogue 2) was also found to be active for the catalytic dehydrocoupling of the phosphine-borane adduct Ph₂PH·BH₃ (14) at 60-90 °C to afford linear dimer Ph₂PH-BH₂-PPh ₂-BH₃ (15).

Full text of DOI:10.1021/ic801752k

Inorg. Chem. 2009, 48, 2429-2435
Homogeneous Catalytic Dehydrogenation/Dehydrocoupling of
Amine-Borane Adducts by the Rh(I) Wilkinson’s Complex Analogue
RhCl(PHCy2)3 (Cy ) cyclohexyl)
Matthew E. Sloan,† Timothy J. Clark,‡ and Ian Manners*,†
School of Chemistry, UniVersity of Bristol, Cantocks Close, Bristol BS8 1TS, U.K., and
Department of Chemistry, UniVersity of Toronto, 80 St. George Street, Ontario M5S 3H6, Canada
Received September 12, 2008
The Rh(I) complex RhCl(PHCy2)3 (1) (Cy ) cyclohexyl, C6H11) has been investigated as a catalyst for the
dehydrogenation/dehydrocoupling of dimethylamine-borane adduct Me2NH·BH3 (3) at 20 °C to afford the cyclic
dimer [Me2N-BH2]2 (4). Unlike previously studied neutral and cationic Rh(I) precatalysts such as [{Rh(µ-Cl)(1,5-
cod)}2] and [Rh(1,5-cod)2]OTf (1,5-cod ) 1,5-cyclooctadiene, C8H12, OTf ) OSO2CF3) with weakly electron-donating
ligands at the metal center, which are reduced to catalytically active Rh(0) species, catalytic dehydrogenation of 3
using 1 was found to occur in a homogeneous manner according to nanofiltration experiments, Hg(0) poisoning
and kinetic studies. Moreover, the presence of the sterically bulky ligand PHCy2 in complex 1 has been found to
significantly increase the rate of reaction previously reported for Wilkinson’s catalyst RhCl(PPh3)3. The catalytic
i
activity of 1 toward a range of other amine-borane adducts RR′NH·BH3 (e.g., RR′ ) Pr2, MeBz, MeH) at 20 °C
was also investigated. The third row transition metal analogue of 1, the Ir(I) complex IrCl(PHCy2)3 (2), was also
explored as a catalyst for the dehydrocoupling of 3 and was found to exhibit much reduced catalytic activity compared
to 1 but proved significantly more active for sterically encumbered substrates. Addition of the strong Lewis acid
B(C6F5)3 as a co-catalyst to both 1 and 2 has been found to significantly increase the rate of the dehydrocoupling
reactions in all cases. The Rh(I) complex 1 (but not the Ir(I) analogue 2) was also found to be active for the
catalytic dehydrocoupling of the phosphine-borane adduct Ph2PH·BH3 (14) at 60-90 °C to afford linear dimer
Ph2PH-BH2-PPh2-BH3 (15).
Introduction
and high molecular weight polymers based on Group 13-
Group 15 skeletons have been accessed via the catalytic
dehydrocoupling of amine-borane and phosphine-borane
adducts.8-10 Furthermore, with the current interest in amine-
borane adducts, such as NH3 ·BH3 (11),11-17 and phosphine-
borane adducts18 as hydrogen-storage and hydrogen-transfer
materials,19,20 this area has attracted intense recent attention,
and many new catalysts based on other metals have been
The use of transition metal catalysts to effect organic
transformations is a thoroughly developed and important area
of research; however, the development of catalytic processes
for inorganic substrates is still in its relative infancy. Catalytic
dehydrogenation/dehydrocoupling of main group species with
E-H (E ) B, N, P, Si etc) bonds is a promising process
and has attracted recent attention as a route to new molecules
and materials.1-7 For example, a variety of rings, chains,
(4) Tilley, T. D. Acc. Chem. Res. 1993, 26, 22–29.
(5) Roering, A. J.; MacMillan, S. N.; Tanski, J. M.; Waterman, R. Inorg.
Chem. 2007, 46, 6855–6857.
(6) Rosenberg, L.; Davis, C. W.; Yao, J. Z. J. Am. Chem. Soc. 2001,
123, 5120–5121.
(7) Greenberg, S.; Stephan, D. W. Chem. Soc. ReV. 2008, 37, 1482–1489.
(8) Dorn, H.; Singh, R. A.; Massey, J. A.; Nelson, J. M.; Jaska, C. A.;
Lough, A. J.; Manners, I. J. Am. Chem. Soc. 2000, 122, 6669–6678.
(9) Jaska, C. A.; Temple, K.; Lough, A. J.; Manners, I. J. Am. Chem.
Soc. 2003, 125, 9424–9434.
(10) Staubitz, A.; Soto, A. P.; Manners, I. Angew. Chem., Int. Ed. 2008,
47, 6212–6215.
* To whom correspondence should be addressed. E-mail: ian.manners@
bristol.ac.uk.
†
University of Bristol.
University of Toronto.
‡
(1) Reichl, J. A.; Berry, D. H. AdV. Organomet. Chem. 1998, 43, 197–
263.
(2) Clark, T. J.; Lee, K.; Manners, I. Chem.sEur. J. 2006, 12, 8634–
8648.
(3) Gauvin, F.; Harrod, J. F.; Woo, H. G. AdV. Organomet. Chem. 1998,
42, 363–405.
10.1021/ic801752k CCC: $40.75  2009 American Chemical Society
Inorganic Chemistry, Vol. 48, No. 6, 2009 2429
Published on Web 02/17/2009

Sloan et al.
reported.14,15,21-23 For example, Heinekey, Goldberg et al.
identified an Ir-pincer complex which efficiently dehydro-
couples 11 in a homogeneous manner and is also active for
primary amine-borane adducts such as MeNH2 ·BH3 at room
temperature.10,15,24 In addition, Chirik et al. recently de-
scribed a highly efficient Group 4 complex with N2 ligands,
[((η5-1,2-(SiMe3)2-C5H3)2Ti)2N2] for the catalytic dehydro-
coupling of Me2NH· BH3 (3), and a mechanism involving
initial B-H oxidative addition was proposed.14 Several other
systems have been calculated computationally providing
useful insight into possible mechanisms.25,26 Although many
catalysts are now available for the dehydrocoupling of Group
13-15 Lewis acid-base adducts, most are relatively sub-
strate specific, and many exhibit only moderate tunability
and activity.
Hg(0), and ligation by substoichiometric amounts of phos-
phines, which all provided support for a heterogeneous
mechanism.27-29
The presence of more strongly electron donating ligands
at the Rh(I) center in the precatalyst, such as phosphines,
would be expected to hinder the reduction process and might
therefore provide access to active homogeneous catalysts.30
However, we have previously found that Wilkinson’s catalyst
RhCl(PPh3)3 shows relatively poor activity for the dehydro-
coupling of amine-borane adducts, with complete conversion
of 3 to 4 requiring about 2 days at 25 °C (with 0.5 mol %
precatalyst in toluene).9 The probable need for ligand loss
at the metal center for efficient catalysis led us to explore
the use of the more sterically demanding ligand PHCy2 in
place of PPh3. In the paper, we report an investigation of
the use of the secondary phosphine analogue of Wilkinson’s
catalyst, RhCl(PHCy2)3 (1), as a precatalyst for the dehy-
drocoupling of amine-borane adducts.31
Scheme 1. General Scheme for the Dehydrocoupling of Primary and
Secondary Amine-Borane Adducts
Results and Discussion
Synthesis and Characterization of the Wilkinson’s
Catalyst Analogue, RhCl(PHCy2)3 (1). The Rh(I) complex
RhCl(PHCy2)3 was prepared as a bright yellow powder by
treatment of [{Rh(µ-Cl)(1,5-cod)}2] with excess PHCy2
according to the literature method (Scheme 2).32 Previous
1
We have previously reported the catalytic dehydrocou-
pling of amine-borane adducts by neutral and cationic
Rh(I) complexes with relatively weak electron-donating
ligands, such as [{Rh(µ-Cl)(1,5-cod)}2] and [Rh(1,5-
cod)2]OTf (Scheme 1).9 In each case, addition of the yellow
Rh(I) precatalyst to the substrate solution was found to result
in the formation of a black reaction mixture after an induction
period. We have provided evidence that this process is
associated with the reduction of Rh(I) to form Rh(0)
nanoclusters and colloids. The dehydrocoupling of 3 using
the precatalyst [{Rh(µ-Cl)(1,5-cod)}2] has been studied in
particular detail. Reduced catalytic activity was observed
upon filtration through nanoporous filters, treatment with
characterization was limited to H NMR spectroscopy and
elemental analysis. We used 31P NMR spectroscopy to
provide additional data. The 1H-coupled 31P NMR spectrum
of 1 exhibited two signals: a doublet of doublets of triplets
at δ ) 50 and a multiplet (overlapped doublet of doublet of
doublets) at δ ) 33. This indicated the presence of two
different phosphorus environments, corresponding to phos-
phine ligands cis and trans to chloride. The signal at δ )
50 split into a doublet due to P-H coupling (1JP-H ) 300
Hz), a second doublet by coupling to Rh (1JP-Rh ) 172 Hz),
and a triplet by coupling to two equivalent phosphorus
centers (2JP-P ) 42 Hz), and on this basis was assigned to
the PHCy2 ligand trans to chloride. By similar reasoning
the multiplet at δ ) 33 that consisted of a doublet (1JP-H
was not resolved due to the second order spectrum), a doublet
(11) Clark, T. J.; Whittell, G. R.; Manners, I. Inorg. Chem. 2007, 46, 7522–
7527.
(12) Marder, T. B. Angew. Chem., Int. Ed. 2007, 46, 8116–8118.
(13) Stephens, F. H.; Pons, V.; Baker, R. T. Dalton Trans. 2007, 2613–
2626.
(14) Pun, D.; Lobkovsky, E.; Chirik, P. J. Chem. Commun. 2007, 3297–
3299.
(15) Denney, M. C.; Pons, V.; Hebden, T. J.; Heinekey, D. M.; Goldberg,
K. I. J. Am. Chem. Soc. 2006, 128, 12048–12049.
(16) Cheng, F. Y.; Ma, H.; Li, Y. M.; Chen, J. Inorg. Chem. 2007, 46,
788–794.
(17) Chandra, M.; Xu, Q. J. Power Sources 2006, 156, 190–194.
(18) Geier, S. J.; Gilbert, T. M.; Stephan, D. W. J. Am. Chem. Soc. 2008,
130, 12632-12633.
(27) Jaska, C. A.; Manners, I. J. Am. Chem. Soc. 2004, 126, 9776–9785.
(28) Chen, Y. S.; Fulton, J. L.; Linehan, J. C.; Autrey, T. J. Am. Chem.
Soc. 2005, 127, 3254–3255.
(29) Fulton, Linehan, Autrey and co-workers have reported independent
studies of this reaction using EXAFS (Extended X-ray Absorption
Fine Structure) analysis. These workers suggested that the major Rh(0)
component formed in their experiments consisted of Rh4-6 nanoclus-
ters. However, their reactions were performed under slightly different
conditions (under H2/He) compared to our experiments (under N2 or
Ar). The induction periods in their experiments were negligible
compared to 45-200 min in our studies. In addition, it is well known
that a true active catalyst, which was indicated to be mainly hete-
rogeneous colloidal Rh aggregates by our studies, often comprises
only an extremely small proportion of the metal present in the reaction
mixture. See refs 23 and 28.
(19) Jiang, Y.; Berke, H. Chem. Commun. 2007, 3571–3573.
(20) Jaska, C. A.; Manners, I. J. Am. Chem. Soc. 2004, 126, 2698–2699.
(21) Yan, J. M.; Zhang, X. B.; Han, S.; Shioyama, H.; Xu, Q. Angew.
Chem., Int. Ed. 2008, 47, 2287–2289.
(22) Clark, T. J.; Russell, C. A.; Manners, I. J. Am. Chem. Soc. 2006, 128,
9582–9583.
(23) Fulton, J. L.; Linehan, J. C.; Autrey, T.; Balasubramanian, M.; Chen,
Y.; Szymczak, N. K. J. Am. Chem. Soc. 2007, 129, 11936–11949.
(24) Dietrich, B. L.; Goldberg, K. I.; Heinekey, D. M.; Autrey, T.; Linehan,
J. C. Inorg. Chem. 2008, 47, 8583–8585.
(30) Clark, T. J.; Jaska, C. A.; Turak, A.; Lough, A. J.; Lu, Z. H.; Manners,
I. Inorg. Chem. 2007, 46, 7394–7402.
(31) Preliminary results were presented in poster form at the AGICHEM
2008 Conference on 09-11 April 2008 at Cardiff, U.K. For a recent
paper reporting independent work on homogeneous Rh dehydrocou-
pling catalysts for amine-borane adducts see Douglas, T. M.; Chaplin,
A. B.; Weller, A. S. J. Am. Chem. Soc. 2008, 130, 14432-14433.
(32) Rigo, P.; Bressan, M. Inorg. Chem. 1976, 15, 220–223.
(25) Yang, X. Z.; Hall, M. B. J. Am. Chem. Soc. 2008, 130, 1798–1799.
(26) Luo, Y.; Ohno, K. Organometallics 2007, 26, 3597–3600.
2430 Inorganic Chemistry, Vol. 48, No. 6, 2009

Dehydrogenation/Dehydrocoupling of Amine-Borane Adducts
by coupling to Rh (1JP-Rh ) 125 Hz), and a doublet by
coupling to another phosphorus center (2JP-P ) 41 Hz), and
on this basis is assigned to the two PHCy2 ligands cis to
chloride (Supporting Information, Figure S-1 and S-2).
of stabilizing ligation, subsequent agglomeration of catalyst
particles via Ostwald ripening leads to a decrease in the
reaction rate. Combined with an induction period, an overall
sigmoidal kinetic profile results, and this is also characteristic
of a heterogeneous catalytic process. If the reaction rate is
not affected by any of the aforementioned techniques, and
no induction period or sigmoidal kinetic profile is observed,
a homogeneous mechanism is indicated.
Scheme 2. Synthesis of 1
We used most of these methods to explore the nature of the
catalytic dehydrocoupling of 3 by precatalyst 1 (1 mol %, 20
°C).34 Filtration of the reaction mixture at about 40% conversion
of 3 through a small pore (e.g. 200 nm) membrane led to no
change in color, and analysis by 11B NMR spectroscopy showed
no suppression of the catalytic activity (Figure 1). Furthermore,
addition of an excess of mercury (70 equiv.) to the reaction
mixture derived from 1 and 3 after about 20% conversion of
the latter also led to no significant reduction in the catalytic
activity (Figure 2). Both results strongly suggest the presence
of a homogeneous catalyst.
In the kinetic profiles in Figures 1 and 2 no induction
period was detected before dehydrocoupling commenced. In
a typical heterogeneous reaction, for example that of 3 with
[{Rh(µ-Cl)(1,5-cod)}2], a sigmoidal plot of reaction progress
versus time is observed. Here, the induction period represents
the reduction of Rh(I) to colloidal Rh(0), and the lowering
in catalytic activity at higher conversion arises from nano-
particle aggregation and a consequential reduction in active
surface area.27 The lack of an induction period and non-
sigmoidal kinetic profile further supports the conclusion that
the catalytic dehydrocoupling of 3 using 1 as a precatalyst
is predominantly homogeneous in nature.
Catalytic Dehydrocoupling Studies with 1. To investi-
gate the activity of 1 toward secondary amine-borane adducts,
a catalytic amount (1 mol %) was added to a toluene solution
of 3 and the reaction was monitored by 11B NMR spectros-
copy. This showed 65% conversion of 3 to 4 (Bδ ) 4.75,
1JB-H ) 110 Hz) in 10 h, with 100% conversion noted after
19 h (Scheme 3). In contrast to the reactions catalyzed by
the Rh(I) complexes [{Rh(µ-Cl)(1,5-cod)}2] and [Rh(1,5-
cod)2]OTf which turn black because of reduction to Rh(0)
species, the reaction mixture remained clear bright yellow.
This suggested that the catalysis using 1 might be homoge-
neous. To determine whether the reaction was homogeneous
or heterogeneous further detailed studies were performed.
Scheme 3. Dehydrocoupling Reactions of Primary and Secondary
Amine-Borane Adducts Investigated
To further probe the catalytic activity of 1, a range of other
amine-borane adducts were also investigated as substrates
(Scheme 3, Table 1). We have previously reported the effect
of the relative steric bulk of ligands at nitrogen on the rate
of reaction with [{Rh(µ-Cl)(1,5-cod)}2] as a precatalyst.9 In
common with these previous results, the rate of catalytic
i
dehydrocoupling of Pr2NH · BH3 (7) was significantly re-
duced relative to that for 3, with no conversion to iPr2N)BH2
(8) after 10 h. Similar treatment of MeBzNH · BH3 (5) (Bz
) CH2C6H5) also led to no conversion after 10 h. Catalyst
1 showed relatively high activity toward the primary amine-
borane adduct MeNH2 · BH3 (9), with 100% conversion to
the borazine [MeN-BH]3 (10) after 10 h.
Intense recent interest in the development of a suitable
portable source of hydrogen has highlighted 11 as a potential
candidate because of its high weight percent content (19 %)
of hydrogen.13 We therefore investigated the activity of 1
toward 11 as a substrate. Addition of a catalytic (1 mol %)
amount of 1 to a diglyme solution of 11 at 45 °C resulted in
the formation of a light yellow solution. However, after 10 h
negligible conversion of 11 had occurred.
To convincingly distinguish between a homogeneous and
a heterogeneous catalytic process it is important to investigate
a reaction using a range of methods. Finke and co-workers
have recently reviewed the most significant of these tech-
niques.33 Of these nanofiltration, Hg(0) poisoning, ligation
studies, and kinetic analysis can provide particularly useful
insight. Thus, strong evidence for a heterogeneous reaction
is provided if the rate of reaction is reduced by (i) filtration
through nanoporous media by removing insoluble catalytic
material, (ii) Hg(0) poisoning through the formation of an
amalgam or coating on the catalyst surface, and (iii) ligation
by <1 equiv. of a donor ligand by blocking of active sites
on the catalyst surface. The presence of an induction period
can also provide evidence for the formation of a heteroge-
neous catalyst by reduction of the precatalyst. In the absence
(34) We did not perform ligation experiments involving the addition of
phosphine as our subsequent studies suggest that the degree of
phosphine dissociation from 1 to form an active catalyst is low. Thus
small amounts of additional phosphine might therefore lead to
significant rate decrease by affecting the dissociation equilibrium rather
than heterogeneous catalyst poisoning confusing the interpretation of
the results.
(33) Widegren, J. A.; Finke, R. G. J. Mol. Cat. A: Chem 2003, 198, 317–
341.
Inorganic Chemistry, Vol. 48, No. 6, 2009 2431

Sloan et al.
Figure 1. Graph showing blank (unfiltered) and filtered (through a 0.2 µm PTFE membrane after 320 min) dehydrocoupling reactions of 3 catalyzed by 1
mol % 1 (in toluene).
Figure 2. Graph showing blank and Hg poisoned (70 equiv added after 180 min) dehydrocoupling reactions of 3 catalyzed by 1 mol % 1 (in toluene).
Table 1. Catalytic Dehydrocoupling Data for Various Group 13/15
Adducts with Rh(I) Complex 1 and Ir(I) Complex 2
Scheme 4. Synthesis of 2
%
%
T/
°C
mol %
catalyst
conversion conversion
substrate
by 1d
by 2d
product
Me2NH·BH3 (3)a
Me2NH·BH3 (3)a
iPr2NH·BH3 (7)a
iPr2NH·BH3 (7)a
20
1
65
100
0
20
0
10
100e
100e
0
0
20
5
15
20
45
25
45
0
4
4
8
8
6
6
10
10
of IrCl(PHCy2)3 (2) was achieved in an analogous manner
to that of 1, with the product also isolated as a bright yellow
powder (Scheme 4). Key characterization was provided by
20 1 + B(C6F5)3
20
20 1 + B(C6F5)3
1
MeBzNH·BH3 (5)a 20
1
1
the H coupled 31P NMR spectrum, which exhibited two
MeBzNH·BH3 (5)a 20 1 + B(C6F5)3
signals: a doublet of doublets at δ ) 26.5 and a doublet of
triplets at δ ) 19.7. As in the case of 1, this indicated the
presence of two different phosphorus environments, corre-
sponding to phosphines cis and trans to a chloride. The signal
at δ ) 26.5 was split into a doublet by P-H coupling (1JP-H
) 218 Hz), a second doublet by coupling to one inequivalent
phosphorus nucleus (2JP-P ) 18 Hz), and on this basis
assigned to two PHCy2 ligands cis to chloride. By similar
reasoning, the signal at δ ) 19.7 was split into a doublet by
P-H coupling (1JP-H ) 327 Hz), and a triplet by coupling
to two equivalent phosphorus centers (2JP-P ) 24 Hz), was
assigned to one PHCy2 ligand trans to chloride (Supporting
Information, Figure S-3 and S-4).
MeNH2 ·BH3 (9)a
MeNH2 ·BH3 (9)a
NH3 ·BH3 (11)b
NH3 ·BH3 (11)b
Ph2PH·BH3 (14)c
Ph2PH·BH3 (14)c
20
20 1 + B(C6F5)3
45
45 1 + B(C6F5)3
90
60
1
1
12, 13
12, 13
15
5
100
80
5
0
0
1
1
15
a Reaction carried out in 2 mL toluene with 1 mmol adduct. b Reaction
carried out in 2 mL diglyme with 1 mmol adduct. c Reaction carried out in
the melt. d % conversion to products calculated by integration of products
against starting material by 11B NMR spectroscopy after 10 h at 20 °C.
e Reaction complete in <10 h.
Catalytic Dehydrocoupling with IrCl(PHCy2)3 (2). To
provide comparative data, we also studied the catalytic
activity of the third row Ir(I) analogue of 1. The synthesis
2432 Inorganic Chemistry, Vol. 48, No. 6, 2009

Dehydrogenation/Dehydrocoupling of Amine-Borane Adducts
Addition of a catalytic amount (1 mol %) of the Ir-complex
2 to a solution of 3 in toluene resulted in the formation of a
clear yellow solution. No significant conversion of 3 to 4
was observed after 10 h by 11B NMR spectroscopy. Surpris-
ingly, in contrast to the case of 1 as precatalyst, the use of
the more sterically encumbered amine-borane adducts 5 and
7 led to conversion of 5% and 20% to 6 and 8, respectively.
The primary amine-borane adduct 9 was found to proceed
to 25% conversion to 10 after the same time period under
the same conditions as 2, whereas no reaction of 11 was
observed by 11B NMR spectroscopy after 10 h.
Enhancement of the Catalytic Activity of 1 and 2
Using B(C6F5)3 as a Co-catalyst. On the basis of the
tentative but reasonable postulate that the catalytic dehy-
drocoupling mechanism with 1 initially proceeds in a manner
similar to that of hydrogenation of alkenes, via the loss of a
phosphine ligand and formation of a 14-electron Rh(I)
species, we attempted to increase the rate of reaction by
addition of a strong Lewis acid to aid dissociation of a Cy2PH
ligand (Scheme 5).35
all cases no evidence for a reaction was apparent by 11B
NMR spectroscopy, indicating that B(C6F5)3 does not
catalyze dehydrocoupling, in contrast to the case of phos-
phine-borane adducts38 and 11 under different conditions
(glyme and tetraglyme at 60 °C).39
Catalytic Dehydrocoupling of Ph2PH · BH3. We also
explored the catalytic dehydrocoupling activity of 1 and 2
toward the secondary phosphine-borane adduct Ph2PH · BH3
(14). In contrast to the case of amine-borane adducts such
as 3, previous studies have indicated that the dehydrocoupling
of 14 by Rh precatalysts such as [{Rh(µ-Cl)(1,5-cod)}2] and
[Rh(1,5-cod)2]OTf are homogeneous processes.27 The adduct
14 has been shown to form the linear dimer, Ph2PH-
BH2-Ph2P-BH3 (15), at 90 °C or cyclic trimer and tetramer,
[Ph2P-BH2]3 (16) and [Ph2P-BH2]4 (17) respectively, at 120
°C when catalyzed by [{Rh(µ-Cl)(1,5-cod)}2] in the melt
(Scheme 6).8
Addition of a catalytic amount of 1 (1 mol %) to solid 14
and heating in the melt at 90 °C resulted in complete
conversion to 15 after 10 h by 11B and 31P NMR spectros-
copy. By comparison, treatment of 14 with 2 resulted in no
detectable reaction by 11B NMR spectroscopy under the same
conditions. To investigate the activity of 1 at a lower
temperature, 14 was heated to 60 °C in the presence of a
catalytic amount of 1 for 10 h. Analysis by 11B and 31P NMR
spectroscopy showed resonances corresponding to 15 (80%)
and starting material 14 (20%). Formation of a small amount
(ca. 1%) of Cy2PH · BH3 was also noted, when 1 and 2 were
used as catalysts, which suggested that some dissociation of
the phosphine-borane adduct 14 and subsequent combination
of the borane with the more basic phosphine PHCy2
dissociated from the precatalyst.
Scheme 5. Proposed Action of B(C6F5)3 on Tris-Phosphine Complexes
In a typical reaction, a toluene solution of B(C6F5)3 and
an equimolar amount of the desired precatalyst were stirred
for 5 min prior to the addition of the amine-borane adduct
in the same solvent. The reaction was then monitored
periodically by 11B NMR spectroscopy. Importantly a
significant increase in the rate of reaction was observed for
all amine-borane/catalyst combinations (Table 1). The most
significant acceleration involved the use of precatalyst 1 with
3, where addition of a catalytic amount (1 mol %) of B(C6F5)3
resulted in 100% conversion to 4 in less than 10 h, compared
to only 65% conversion in the absence of the borane. Over
the course of the catalytic reactions, Cy2PH · B(C6F5)3 was
not detectable in solution. However, on addition of 1 equiv.
of B(C6F5)3 to 1, Cy2PH · B(C6F5)3 was formed, along with
several unidentified species, according to 11B and 31P NMR
spectroscopic analysis (Supporting Information, Figures S-5
and S-6).36
The slower rate of reaction observed in the absence of a
co-catalyst suggests that a low percentage of the precatalyst
is converted into the active species, whereas addition of a
co-catalyst shifts this equilibrium in favor of the active
species (Scheme 5). In an attempt to access improved
catalytic activity without the need for a co-catalyst the bis-
phosphine dimer, [RhCl(PHCy2)2]2 was targeted; however,
synthesis by a variety of methods failed.37 A series of blank
reactions were also performed, where B(C6F5)3 alone was
added to toluene solutions of 3, 5, 7, 9 at 25 °C, or a diglyme
solution of 11 at 45 °C, followed by stirring for 2 days. In
Conclusion
The Rh(I) complex RhCl(PHCy2)3 (1) has been found to
catalyze the dehydrocoupling of amine-borane adduct 3 under
ambient conditions. Moreover, unlike the cases of previously
studied Rh(I) precatalysts with 1,5-cod ligands, the catalytic
reaction of 1 with 3 has been shown to be homogeneous by
a variety of methods. The third row transition metal analogue
of 1, IrCl(PHCy2)3 (2), showed significantly reduced catalytic
dehydrocoupling activity toward 3 but was more active than
1 toward more sterically encumbered substrates. Addition
of the strong Lewis-acid B(C6F5)3 has been found to
significantly increase the rate of the dehydrocoupling reac-
tions catalyzed by 1 and 2, presumably by acting as a
phosphine abstraction reagent.
Complex 1 is a member of the extensive class of neutral
Rh(I) complexes [RhX(PR3)3] (X ) halogen), and this
(37) Synthesis of [RhCl(PHCy2)2]2 was attempted in a manner similar to
the method used for [RhCl(PPh3)2]2, by refluxing in benzene or by
addition of 4 equiv of PHCy2 to [{RhCl(cod)}2]. This resulted in no
reaction in the former case, the formation of RhCl(cod)(PHCy2) and
RhCl(PHCy2)3 in the latter. An attempted synthesis by the addition of
4 equiv of PHCy2 to [{RhCl(coe)2}2] or [{RhCl(CH2CH2)2}2] also
failed with partial conversion to RhCl(PHCy2)3 in both cases.
(38) Denis, J. M.; Forintos, H.; Szelke, H.; Toupet, L.; Pham, T. N.; Madec,
P. J.; Gaumont, A. C. Chem. Commun. 2003, 54–55.
(35) Erker, G. Dalton Trans. 2005, 1883–1890.
(36) Lancaster, S. J.; Mountford, A. J.; Hughes, D. L.; Schormann, M.;
Bochmann, M. J. Organomet. Chem. 2003, 680, 193–205.
(39) Stephens, F. H.; Baker, R. T.; Matus, M. H.; Grant, D. J.; Dixon,
D. A. Angew. Chem., Int. Ed. 2007, 46, 746–749.
Inorganic Chemistry, Vol. 48, No. 6, 2009 2433

Sloan et al.
Scheme 6. Catalytic Dehydrocoupling of 14
PHCy2 (0.793 g, 4.0 mmol). The resulting suspension was stirred
at 70 °C overnight giving a yellow precipitate. The precipitate was
collected, washed with hexanes (3 × 20 mL), and dried. Yield
0.78 g (95%). 1H{31P} NMR (300 MHz, C6D6, 20 °C): δ ) 4.5-3.5
(m, 3H, 3 × PH), 2.79-1.17 (m, 66H, 6 × Cy); 13C NMR (300
MHz, C6D6, 20 °C): δ ) 33.2-30.9 (m), 27.4-26.7 (m); 31P NMR
suggests that fruitful tuning of catalytic activity should be
possible in the future by means of controlling ligand steric
bulk and electronic character.40 We are currently performing
in-depth kinetic experiments and theoretical studies to
provide detailed mechanistic information on the dehydro-
coupling chemistry.
1
2
(300 MHz, C6D6, 20 °C): δ ) 26.5 (dd, JP-H ) 208 Hz, JP-P
)
1
2
18.2 Hz, PHCy2 cis to Cl), 19.7 (dt, JP-H ) 327 Hz, JP-P ) 24
Hz, PHCy2 trans to Cl) (Supporting Information, Figure S-3 and
S-4). EA calculated (%) for IrClP3C36H69: C 52.57, H 8.46; found
C 53.20, H 8.49.
Experimental Section
General Procedures and Materials. All reactions and product
manipulations were performed under an atmosphere of dry nitrogen
using standard Schlenk techniques or in a MBraun glovebox filled
with dry argon. NMR spectra were collected on a Jeol Lambda
300 or Jeol ECP 300. Chemical shifts were referenced to solvent
peaks or internal TMS (1H, 13C), external BF3 ·OEt2 (11B) or H3PO4
(31P). Toluene and hexanes were purified using a Grubbs solvent
General Procedure for the Catalytic Dehydrogenation of
Amine-Borane Adducts. To a solution of amine-borane (1 mmol)
in toluene (2 mL) was added catalyst (1 mol %). The toluene
solution was stirred and an aliquot subsequently removed for
analysis by 11B NMR spectroscopy after 10 h (see Table 1). Product
structures were assigned based on previous literature.9
i
system. PrOH and MeOH were degassed prior to use by three
freeze-pump-thaw cycles. Mercury (99.9995%), [{Rh(µ-Cl)(1,5-
cod)}2], MeNH2 and NH3 ·BH3 were purchased from Aldrich and
used as received. PHCy2, RhCl3 ·3H2O and [{Ir(µ-Cl)(1,5-cod)}2]
were purchased from Strem and used as received. Cyclooctene was
purchased from Acros and used as received. Ethylene was purchased
from BOC and used as received. Me2NH·BH3 (Strem) was sub-
limed twice prior to use, iPr2NH (Aldrich) and BzMeNH (Aldrich)
were distilled from CaH2, BH3 ·THF (Acros) was distilled prior to
use, and B(C6F5)3 (Boulder Scientific) was dried with Me3SiCl and
sublimed prior to use. Filters (Millex-FG, 0.2 µm hydropho-
bic fluoropore PTFE) were purchased from Fischer. Ph2PH,41
RhCl(PHCy2)3,32 iPr2NH · BH3,9 BzMeNH · BH3,9 MeNH2 · BH3,9
Ph2PH·BH3,8 [RhCl(coe)2]2,42 and [RhCl(CH2CH2)2]243 were syn-
thesized by literature methods. All dehydrocoupling experiments
were performed at lease twice to ensure reproducibility.
General Procedure of the B(C6F5)3 Co-catalyzed Dehydro-
genation of Amine-Borane Adducts. A solution of catalyst (1 mol
%) and B(C6F5)3 (1 mol %) in toluene (1 mL) was stirred for 5
min. To this a solution of the desired amine-borane adduct (1 mmol)
in toluene (1 mL) was added. The solution was stirred and an aliquot
subsequently removed for analysis by 11B NMR spectroscopy after
10 h (see Table 1). Product structures were assigned based on
previous literature.9
General Procedure for B(C6F5)3 Blank Reaction. In a typical
reaction, a catalytic amount of B(C6F5)3 was added to a toluene
solution of the desired amine-borane adduct (3, 7, 5, or 9) at 20
°C, or a diglyme solution of 11 at 45 °C, and the solution stirred
for 2 days. An aliquot was then removed for analysis by 11B NMR
spectroscopy, and no reaction was observed in any case.
General Procedure for Filtration Reactions. To a solution of
Me2NH ·BH3 (0.1 g, 1.7 mmol) in toluene (2 mL) was added 1 (11
mg, 0.015 mmol, 1 mol %) with vigorous stirring. The reaction
was monitored by 11B NMR spectroscopy at regular intervals until
40% conversion to products was achieved. The reaction was then
filtered through a 0.2 µm filter into a new vial with a new stir bar
and septum. Monitoring at regular intervals was continued by 11B
NMR spectroscopy. No change in the rate of reaction was observed,
with the reaction going to completion in 19 h.
Synthesis of RhCl(PHCy2)3 (1).32 To a solution of [{Rh(µ-Cl)-
(1,5-cod)}2] (0.246 g, 0.5 mmol) in hexanes (40 mL) PHCy2 (0.793
g, 4.0 mmol) was added. The resulting suspension was stirred
overnight at 70 °C, giving a yellow precipitate. The precipitate was
collected, washed with hexanes (3 × 20 mL) and dried. Yield 0.55 g
1
(75%). H{31P} NMR (300 MHz, C6D6, 20 °C) δ ) 4.3-3.9 (m,
3H, 3 × PH), 2.8-1.1 (m, 66H, 3 × Cy); 31P NMR (300 MHz,
C6D6, 20 °C) δ ) 50 (m (overlapped doublet of doublet of triplets),
1
2
1JP-H ) 300 Hz, JP-Rh ) 172 Hz, JP-P ) 42 Hz, 1 × PHCy2
trans to Cl), δ ) 33 (m (overlapped doublet of doublet of doublets),
1JP-H was not resolved, 1JP-Rh ) 125 Hz, 2JP-P ) 41 Hz, 2 × PHCy2
cis to Cl) (Supporting Information, Figure S-1 and S-2).
General Procedure for Hg Poisoning Reactions. A toluene
solution of catalyst (1 mol %) and 3 was monitored by 11B NMR
spectroscopy, and after 20% conversion to 4, Hg (70 equiv.) was
added with vigorous stirring. 11B NMR monitoring was continued,
and the % conversion compared to a blank reaction. Important:
Good stirring is known to be needed to ensure contact of the catalyst
with Hg(0), a condition required to avoid false negatives in this
experiment.33
Synthesis of IrCl(PHCy2)3 (2). To a solution of [{Ir(µ-Cl)-
(1,5-cod)}2] (0.335 g, 0.5 mmol) in hexanes (40 mL) was added
(40) Tolman, C. A. Chem. Soc. ReV. 1982, 1, 337.
(41) Rohlik, Z.; Holzhauser, P.; Kotek, J.; Rudovsky, J.; Nemec, I.;
Hermann, P.; Lukes, I. J. Organomet. Chem. 2006, 691, 2409–2423.
(42) Vanderent, A.; Onderdelinden, A. L. Inorg. Synth. 1990, 28, 90–92.
(43) Cramer, R. Inorg. Chem. 1962, 1, 722.
Catalytic Dehydrocoupling of NH3 ·BH3(11). To a suspension
of catalyst (0.014 mmol, 1 mol %) in diglyme (2 ml) 11 (0.043 g,
2434 Inorganic Chemistry, Vol. 48, No. 6, 2009

Dehydrogenation/Dehydrocoupling of Amine-Borane Adducts
1.4 mmol) was added and heated to 40 °C, resulting in complete
dissolution. An aliquot was removed for analysis by 11B NMR
spectroscopy after 10 h (see Table 1). Product structures were
assigned based on previous literature.9
B(C6F5)3 Co-catalyzed Dehydrocoupling of NH3BH3(11). To
a suspension of catalyst (0.014 mmol, 1 mol %) and B(C6F5)3 (7.0
mg, 0.014 mmol, 1 mol %) in diglyme (2 ml) 11 (0.043 g, 1.4
mmol) was added and heated to 40 °C, resulting in complete
dissolution. An aliquot was removed for analysis by 11B NMR
spectroscopy after 10 h (see Table 1). Product structures were
assigned based on previous literature.9
Catalytic Dehydrocoupling of Ph2PH·BH3. Ph2PH·BH3 (0.2
g, 1.0 mmol) and the desired catalyst (1 mol %) were heated in the
melt at 60 or 90 °C for 10 h. On cooling, the material was dissolved
and analyzed by 11B NMR spectroscopy (see Table 1). Product
structures were assigned based on previous literature.8
mg, 0.055 mmol), and the resulting mixture stirred for 12 h.
Analysis was achieved by NMR. 11B NMR (300 MHz, d8-toluene)
δ ) -16.1 (JB-P ) 58 Hz), lit. (96.29 MHz, C6D6, 20 °C) δ )
-13.5 (JB-P ) 64 Hz); 31P (300 MHz, d8-toluene) δ ) 7.9 (JB-P
)
56 Hz), lit. (121.5 MHz, C6D6, 20 °C) δ ) 9.29 (JB-P ) 76 Hz)36
(Supporting Information, Figure S-5 and S-6).
Acknowledgment. M.E.S. acknowledges the EPSRC for
funding and Dr. Anne Staubitz for useful discussions. T.J.C.
thanks the Ontario Government for a Graduate Scholarship
(OGS). I.M. thanks the E.U. for a Marie Curie Chair and
the Royal Society for a Wolfson Research Merit Award.
Supporting Information Available: Relevant 31P and 11B NMR
spectra. This material is available free of charge via the Internet at
http://pubs.acs.org.
Stoichiometric Reaction of 1 with B(C6F5)3. To a solution of
1 (0.04 g, 0.055 mmol) in toluene (1 mL) was added B(C6F5)3 (28
IC801752K
Inorganic Chemistry, Vol. 48, No. 6, 2009 2435