Homogeneous catalytic dehydrocoupling/dehydrogenation of amine-borane adducts by early transition metal, group 4 metallocene complexes

Article abstract of DOI:10.1021/ja909535a

The efficient catalytic dehydrocoupling of a range of amine-borane adducts, R'RNH·BH₃ (R' = R = Me 1a; R' = R = Pr 1b; R' = Me, R = CH₂Ph 1c) by a series of group 4 metallocene type precatalysts has been demonstrated. A reduction in catalytic activity was detected upon descending the group and also on substitution of the cyclopentadienyl (Cp) ligands with sterically bulky or electron-donating substituents. Precatalysts Cp₂TiCl₂/2ⁿBuLi and Cp₂Ti(PMe ₃)₂, which are believed to act as precursors to [Cp ₂Ti], were found to promote the transformation of 1a to [Me ₂N-BH₂l₂ (3a) in a homogeneous catalytic process. Mechanistic studies identified the linear dimer Me₂NH- BH₂-NMe₂-BH₃ (2a) as a reaction intermediate, which subsequently undergoes further catalytic dehydrogenation to form cyclic dimer 3a. Synthesis of the ²H-isotopologues of 1a allowed the extraction of phenomenological kinetic isotope effects for 1a → 2a and 2a → 3a from initial rate data, which permitted the proposal of a catalytic cycle with plausible intermediates. Support for the presence of an active Ti(II) catalyst was provided by the lack of reactivity of Ti(III) complexes TiCl ₃ and Cp₂TiCl or Ti(O) in the form of THF soluble colloids or bulk Ti powder toward 1a or 1b. Modeling of the rates of consumption of 1a and formation of 3a during catalysis by Cp₂Ti(PMe₃) ₂ supported this conclusion and allowed the proposal of a two cycle, four step reaction mechanism. The proposed first cycle generates 2a in a two step process. In the second cycle, interaction of 2a with the same catalyst then results in a catalytic dehydrogenative ring closing reaction to form 3a, also in a two step process.

Full text of DOI:10.1021/ja909535a

Published on Web 02/24/2010
Homogeneous Catalytic Dehydrocoupling/Dehydrogenation of
Amine-Borane Adducts by Early Transition Metal, Group 4
Metallocene Complexes
Matthew E. Sloan, Anne Staubitz, Timothy J. Clark, Christopher A. Russell,
Guy C. Lloyd-Jones,* and Ian Manners*
School of Chemistry, UniVersity of Bristol, Bristol, BS8 1TS, United Kingdom
Received November 10, 2009; E-mail: ian.manners@bristol.ac.uk; guy.lloyd-jones@bristol.ac.uk
Abstract: The efficient catalytic dehydrocoupling of a range of amine-borane adducts, R′RNH·BH3 (R′ )
R ) Me 1a; R′ ) R ) iPr 1b; R′ ) Me, R ) CH2Ph 1c) by a series of group 4 metallocene type precatalysts
has been demonstrated. A reduction in catalytic activity was detected upon descending the group and also
on substitution of the cyclopentadienyl (Cp) ligands with sterically bulky or electron-donating substituents.
Precatalysts Cp2TiCl2/2nBuLi and Cp2Ti(PMe3)2, which are believed to act as precursors to [Cp2Ti], were
found to promote the transformation of 1a to [Me2N-BH2]2 (3a) in a homogeneous catalytic process.
Mechanistic studies identified the linear dimer Me2NH-BH2-NMe2-BH3 (2a) as a reaction intermediate,
which subsequently undergoes further catalytic dehydrogenation to form cyclic dimer 3a. Synthesis of the
2H-isotopologues of 1a allowed the extraction of phenomenological kinetic isotope effects for 1a f 2a and
2a f 3a from initial rate data, which permitted the proposal of a catalytic cycle with plausible intermediates.
Support for the presence of an active Ti(II) catalyst was provided by the lack of reactivity of Ti(III) complexes
TiCl3 and Cp2TiCl or Ti(0) in the form of THF soluble colloids or bulk Ti powder toward 1a or 1b. Modeling
of the rates of consumption of 1a and formation of 3a during catalysis by Cp2Ti(PMe3)2 supported this
conclusion and allowed the proposal of a two cycle, four step reaction mechanism. The proposed first
cycle generates 2a in a two step process. In the second cycle, interaction of 2a with the same catalyst
then results in a catalytic dehydrogenative ring closing reaction to form 3a, also in a two step process.
Introduction
We have previously reported the Rh-catalyzed dehydrocou-
pling of phosphine-borane adducts by late transition metal
precatalysts such as [Rh(µ-Cl)(1,5-cod)]2 to yield a range of
The use of transition metal complexes as catalysts to effect
the synthesis and transformations of organic molecules through
C-X (X ) C, N, or O) bond formation and polymerization
processes is a thoroughly developed area of profound current
importance. By comparison, the use of catalytic reactions to
prepare catenated inorganic species is still in its relative infancy,
with the formation of main group element-element bonds
relying almost entirely on salt elimination reactions. Since the
initial reports of transition metal catalyzed formation of B-B
and Si-Si bonds via dehydrogenative coupling by Sneddon1
and Harrod2 in the 1980s, the area has expanded to include a
range of homonuclear and heteronuclear bond forming reactions,
allowing convenient access to main group based rings, chains
and high molecular weight polymers.3 Moreover, with current
intense interest in the potential of a hydrogen economy to
provide a clean source of energy, amine-borane adducts such
as NH3 ·BH3 (1e) are under detailed scrutiny as potential safe
and stable hydrogen storage materials.4-9,16c
(4) (a) Hamilton, C. W.; Baker, R. T.; Staubitz, A.; Manners, I. Chem.
Soc. ReV. 2009, 38, 279–293. (b) Cheng, F.; Ma, H.; Li, Y.; Chen, J.
Inorg. Chem. 2007, 46, 788–794. (c) Chandra, M.; Xu, Q. J. Power
Sources 2006, 156, 190–194. (d) Stephens, F. H.; Baker, R. T.; Matus,
M. H.; Grant, D. J.; Dixon, D. A. Angew. Chem., Int. Ed. 2007, 46,
746–749. (e) Nguyen, M. T.; Nguyen, V. S.; Matus, M. H.; Gopaku-
mar, G.; Dixon, D. A. J. Phys. Chem. A 2007, 111, 679–690. (f) Yoon,
C. W.; Sneddon, L. G. J. Am. Chem. Soc. 2006, 128, 13992–13993.
(g) Xu, Q.; Chandra, M. J. Power Sources 2006, 163, 364–370. (h)
Chandra, M.; Xu, Q. J. Power Sources 2006, 159, 855–860. (i)
Staubitz, A.; Besora, M.; Harvey, J. N.; Manners, I. Inorg. Chem.
2008, 47, 5910–5918. (j) Spielmann, J.; Jansen, G.; Bandmann, H.;
Harder, S. Angew. Chem., Int. Ed. 2008, 47, 6290–6295. (k) Stephens,
F. H.; Pons, V.; Baker, R. T. Dalton Trans. 2007, 2613–2626. (l)
Marder, T. B. Angew. Chem., Int. Ed. 2007, 46, 8116–8118. (m)
Hausdorf, S.; Baitalow, F.; Wolf, G.; Mertens, F. O. R. L. Int. J.
Hydrogen Energy 2008, 33, 608–614. (n) Clark, T. J.; Whittell, G. R.;
Manners, I. Inorg. Chem. 2007, 46, 7522–7527. (o) Alcaraz, G.;
Vendier, L.; Clot, E.; Sabo-Etienne, S. Angew. Chem. Int. Ed. 2010,
49, 918–920. (p) Tang, C. Y.; Thompson, A. L.; Aldridge, S. Angew.
Chem. Int. Ed. 2010, 49, 921–925.
(5) Denney, M. C.; Pons, V.; Hebden, T. J.; Heinekey, D. M.; Goldberg,
K. I. J. Am. Chem. Soc. 2006, 128, 12048–12049.
(1) Corcoran, E. W.; Sneddon, L. G. J. Am. Chem. Soc. 1984, 106, 7793–
7800.
(6) Keaton, R. J.; Blacquiere, J. M.; Baker, R. T. J. Am. Chem. Soc. 2007,
129, 1844–1845.
(2) (a) Aitken, C.; Harrod, J. F.; Samuel, E. J. Organomet. Chem. 1985,
279, C11–C13. (b) Aitken, C. T.; Harrod, J. F.; Samuel, E. J. Am.
Chem. Soc. 1986, 108, 4059–4066.
(7) (a) Ka¨ꢀ, M.; Friedrich, A.; Drees, M.; Schneider, S. Angew. Chem.,
Int. Ed. 2009, 48, 905–907. (b) Friedrich, A.; Drees, M.; Schneider,
S. Chem.sEur. J. 2009, 40, 10339–10342.
(3) Clark, T. J.; Lee, K.; Manners, I. Chem.sEur. J. 2006, 12, 8634–
8648.
(8) Blaquiere, N.; Diallo-Garcia, S.; Gorelsky, S. I.; Black, D. A.; Fagnou,
K. J. Am. Chem. Soc. 2008, 130, 14034–14035.
9
10.1021/ja909535a  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 3831–3841 3831

A R T I C L E S
Sloan et al.
Ir5,19), mid (Re,20 Ru7,8) and early (Cr, Mo, W21) transition metal
catalysts for the dehydrogenation of amine-borane adducts such
as 1a or 1e. Many of these catalytic reactions are substrate
specific and the mechanisms of the dehydrocoupling/dehydro-
genation processes are of key importance. In some cases
mechanistic insight has recently been provided by computational
studies and the isolation of potential intermediates.4a,8,9,22
As part of our efforts to expand the range of transition metal
catalysts for amine-borane dehydrogenation we have explored
the use of group 4 metallocene complexes. Such systems have
previously been shown to be active for the dehydrocoupling/
dehydrogenation of silanes such as PhSiH3.2,23 In 2006, we
briefly reported that [Cp2Ti], generated in situ from Cp2TiCl2/
2nBuLi at -15 °C with subsequent warming to 20 °C, functions
as an efficient homogeneous catalyst for the dehydrocoupling
of secondary amine-borane adducts such as 1a.24 A subsequent
computational investigation of this system by Ohno and Luo
suggested that the mechanism involved an intramolecular,
stepwise process.25 These workers suggested that 1a initially
interacts with the catalytic metal center via a B-H bond,
followed by proton transfer from nitrogen and subsequent
hydride transfer from boron to give the dehydrogenated ami-
noborane Me2NdBH2 (5a) and Cp2TiH2. They proposed that
the former then dimerizes in an uncatalyzed manner while the
resulting titanocene dihydride releases hydrogen to reform
[Cp2Ti] (Scheme 3). In a further development, Chirik and co-
workers reported the dehydrocoupling of 1a and 1e by a series
of other group 4 complexes and observed that increased
substitution of the Cp ligands resulted in reduced catalytic
activity.26 Signficantly, a N2 complex [{(1,3-(SiMe3)2C5H3)2Ti}2-
(η1-N2)] was found to be a highly efficient catalyst for 1a and
a mechanism involving initial B-H oxidative addition was
proposed.
Scheme 1. Catalytic Dehydrocoupling of Primary and Secondary
Phosphine-Borane Adducts by Rh Precatalysts (e.g.,
[Rh(µ-Cl)(1,5-cod)]2)
Scheme 2. Catalytic Dehydrocoupling of Primary and Secondary
Amine-Borane Adducts by Transition Metal Precatalysts (Bn )
CH2Ph)
linear (at 60-90 °C) and cyclic products (at 100-120 °C), in
the case of secondary phosphine-borane adducts, and high
molecular weight polymers in the case of primary phosphine-
borane adducts (Scheme 1). Detailed studies of these reactions
suggested that the active catalyst is homogeneous, an assertion
consistent with nanofiltration, colloidal catalyst poisoning, and
other studies.10-13
We subsequently reported the extension of this catalytic
dehydrocoupling/dehydrogenation chemistry to primary and
secondary amine-borane adducts, and to 1e (Scheme 2). Thus,
in the presence of a variety of late transition metal precatalysts
such as [Rh(µ-Cl)(1,5-cod)]2, a range of cyclic oligomeric or
insoluble polymeric species were formed either at or slightly
above ambient temperature.14,15 Mechanistic studies indicated
that, in the case of [Rh(µ-Cl)(1,5-cod)]2 as a precatalyst, the
dehydrogenation of Me2NH·BH3 (1a) to form the cyclic dimer
[Me2N-BH2]2 (3a) proceeds via reduction to Rh(0) species and
the true catalyst appears to be heterogeneous in nature.11,16
Recent advances reported by a range of research groups have
led to the discovery of new and improved late (Fe, Ni,6 Rh,9,17,18
In this paper, as a follow up to our preliminary communica-
tion,24 we report a detailed study of the catalytic dehydrocou-
pling of amine-borane adducts by group 4 transition metal
(16) The exact nature of the catalytically active Rh(0) species is currently
under debate, see: (a) Chen, Y.; Fulton, J. L.; Linehan, J. C.; Autrey,
T. J. Am. Chem. Soc. 2005, 127, 3254–3255. (b) Fulton, J. L.; Linehan,
J. C.; Autrey, T.; Balasubramanian, M.; Chen, Y.; Szymczak, N. K.
J. Am. Chem. Soc. 2007, 129, 11936–11949. (c) Rousseau, R.;
Schenter, G. K.; Fulton, J. L.; Linehan, J. C.; Engelhard, M. H.; Autrey,
T. J. Am. Chem. Soc. 2009, 131, 10516–1052. (d) Zahmakiran, M.;
Ozkar, S. Inorg. Chem. 2009, 48, 8955–8964.
(17) (a) Dallanegra, R.; Chaplin, A. B.; Weller, A. S. Angew. Chem., Int.
Ed. 2009, 48, 6875–6878. (b) Sloan, M. E.; Clark, T. J.; Manners, I.
Inorg. Chem. 2009, 48, 2429–2435.
(18) Douglas, T. M.; Chaplin, A. B.; Weller, A. S.; Yang, X.; Hall, M. B.
J. Am. Chem. Soc. 2009, 131, 15440–15456.
(19) (a) Hebden, T. J.; Denney, M. C.; Pons, V.; Piccoli, P. M. B.; Koetzle,
T. F.; Schultz, A. J.; Kaminsky, W.; Goldberg, K. I.; Heinekey, D. M.
J. Am. Chem. Soc. 2008, 130, 10812–10820. (b) Dietrich, B. L.;
Goldberg, K. I.; Heinekey, D. M.; Autrey, T.; Linehan, J. C. Inorg.
Chem. 2008, 47, 8583–8585. (c) Staubitz, A.; Soto, A. P.; Manners,
I. Angew. Chem., Int. Ed. 2008, 47, 6212–6215.
(9) Douglas, T. M.; Chaplin, A. B.; Weller, A. S. J. Am. Chem. Soc. 2008,
130, 14432–14433.
(20) (a) Jiang, Y.; Berke, H. Chem. Commun. 2007, 3571–3573. (b) Jiang,
Y.; Blacque, O.; Fox, T.; French, C. M.; Berke, H. Organometallics
2009, 28, 5493–5504.
(10) (a) Dorn, H.; Singh, R. A.; Massey, J. A.; Lough, A. J.; Manners, I.
Angew. Chem., Int. Ed. 1999, 38, 3321–3323. (b) Jaska, C. A.;
Manners, I. J. Am. Chem. Soc. 2004, 126, 1334–1335.
(11) Jaska, C. A.; Manners, I. J. Am. Chem. Soc. 2004, 126, 9776–9785.
(12) 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.
(13) Clark, T. J.; Rodezno, J. M.; Clendenning, S. B.; Aouba, S.; Brodersen,
P. M.; Lough, A. J.; Ruda, H. E.; Manners, I. Chem.sEur. J. 2005,
11, 4526–4534.
(21) Kawano, Y.; Uruichi, M.; Shimoi, M.; Taki, S.; Kawaguchi, T.;
Kakizawa, T.; Ogino, H. J. Am. Chem. Soc. 2009, 131, 14946–14957.
(22) (a) Zimmerman, P. M.; Paul, A.; Zhang, Z. Y.; Musgrave, C. B. Angew.
Chem., Int. Ed. 2009, 48, 2201–2205. (b) Yang, X. Z.; Hall, M. B.
J. Am. Chem. Soc. 2008, 130, 1798–1799.
(23) (a) Gauvin, F.; Harrod, J. F.; Woo, H. G. AdV. Organomet. Chem.
1998, 42, 363–405. (b) Tilley, T. D. Acc. Chem. Res. 1993, 26, 22–
29. (c) Corey, J. Y.; Zhu, X. H.; Bedard, T. C.; Lange, L. D.
Organometallics 1991, 10, 924–930.
(14) Jaska, C. A.; Temple, K.; Lough, A. J.; Manners, I. Chem. Commun.
2001, 962–963.
(24) Clark, T. J.; Russell, C. A.; Manners, I. J. Am. Chem. Soc. 2006, 128,
9582–9583.
(15) Jaska, C. A.; Temple, K.; Lough, A. J.; Manners, I. J. Am. Chem.
Soc. 2003, 125, 9424–9434.
(25) Luo, Y.; Ohno, K. Organometallics 2007, 26, 3597–3600.
9
3832 J. AM. CHEM. SOC. VOL. 132, NO. 11, 2010

Dehydrocoupling/Dehydrogenation of Amine-Borane Adducts
A R T I C L E S
Scheme 3. Intramolecular Stepwise Catalytic Dehydrocoupling of
1a by [Cp2Ti] Proposed by Ohno and Luo25
Figure 1. 11B{1H} NMR (96 MHz, unlocked, toluene) spectrum showing
conversion of 1a (1.3 M in toluene) to 2a and 3a by [Cp2Ti] generated in
situ (2 mol % in toluene), * ) unidentified species. (a) t ) 0 min, (b) t )
15 min, (c) t ) 75 min, (d) t ) 95 min, (e) t ) 135 min, (f) t ) 240 min.
Cp2TinBu2 is generated first, followed by the formation of the
putative Ti(II) product “Cp2Ti” via a ꢀ-hydrogen elimination/
reductive elimination and olefin dissociation pathway.29,30
Presumably the electron-deficient, coordinatively unsaturated Ti
center in the latter facilitates an initial reaction with 1a and
entry into the catalytic cycle. The proposed involvement of Ti(II)
is reinforced by the report of Sneddon and co-workers describing
that hydroboration reactions of decaborane, B10H14, with terminal
olefins can be promoted by the Ti(II) precatalyst Cp2Ti(CO)2.31
During the course of the catalytic dehydrocoupling of 1a, to
generate 3a, the formation of several other boron-containing
species was noted in the 11B{1H} NMR spectrum. In the early
stages, a signal at δ 2 ppm was observed. This signal
corresponds to the internal BH2 moiety in the linear dimer
Me2NH-BH2-NMe2-BH3 (2a), as confirmed by comparison
with an authentic sample of this species. Linear dimer 2a appears
to be an intermediate in the dehydrocoupling of 1a (Figure
Scheme 4. Catalytic Dehydrocoupling of 1a by [Cp2Ti] to Afford 2a
and 3a
1).7b,32-34 This signal split into a triplet when the H coupled
1
11B NMR spectrum was recorded (1JB-H ) 108 Hz), while the
signal for the terminal BH3 unit was obscured by the signal for
starting material 1a (see Figure S1 and S2, Supporting Informa-
tion). The intensity of this signal (δ 2 ppm) increased rapidly
for the first 15 min of the reaction, but subsequently decreased
as 3a is formed. Small signals were also noted in the spectrum
metallocene complexes. Our work includes in-depth kinetic and
mechanistic studies that provide new insight into the pathways
involved for this interesting early transition metal-catalyzed
process.
1
at δ 28 ppm (d, JB-H ) 130 Hz), which was assigned to the
Results
(29) (a) Negishi, E. I.; Cederbaum, F. E.; Takahashi, T. Tetrahedron Lett.
1986, 27, 2829–2832. (b) Negishi, E. I.; Holmes, S. J.; Tour, J. M.;
Miller, J. A.; Cederbaum, F. E.; Swanson, D. R.; Takahashi, T. J. Am.
Chem. Soc. 1989, 111, 3336–3346.
1. Catalytic Dehydrocoupling of Me2NH ·BH3 (1a) with
Cp2TiCl2/2nBuLi. To investigate if catalysts of the type [Cp2Ti]
would be active for the dehydrocoupling of amine-borane
adducts, a toluene solution of Me2NH·BH3 (1a) was treated
with a catalytic amount (2 mol %) of [Cp2Ti] generated in situ.
Extensive H2 release was observed, which was confirmed by
1H NMR spectroscopy (δ(1H) ) 4.46 in C6D6 lit. 4.46)27 along
with full conversion of 1a to yield 3a as the dominant product
after 4 h, as determined by 11B NMR spectroscopy (Scheme
4).28
(30) Hitchcock, P. B.; Kerton, F. M.; Lawless, G. A. J. Am. Chem. Soc.
1998, 120, 10264–10265.
(31) Pender, M. J.; Carroll, P. J.; Sneddon, L. G. J. Am. Chem. Soc. 2001,
123, 12222–12231.
(32) No¨th, H.; Thomas, S. Eur. J. Inorg. Chem. 1999, 1373–1379.
(33) In our preliminary communication (ref 24) we tentatively postulated
a cyclic trimer structure, [Me2N-BH2]3, for the intermediate based on
1
our observation of unresolved H coupled and decoupled 11B NMR
spectra in toluene (see Figure S1, Supporting Information). Our further
studies and private communication with S. Schneider (T. U. Munich)
revealed this assignment to be incorrect. The 11B NMR signals can
be resolved in C6D6 (see Figure S2, Supporting Information). For
further relevant discussions of the same issues see refs 7b and 18. It
should also be noted that the coupling constants reported for linear
dimer Me2NH-BH2-NMe2-BH3 in ref 32 are incorrect. The correct
The exact details of the reaction between Cp2TiCl2 and nBuLi
are not known with certainty but it has been proposed that
(26) Pun, D.; Lobkovsky, E.; Chirik, P. J. Chem. Commun. 2007, 3297–
3299.
1
(27) Bo¨hm, V. P. W.; Brookhart, M. Angew. Chem., Int. Ed. 2001, 40,
4694–4696.
values are 11B NMR (96 MHz, C6D6, 20 °C) δ-13.2 (q, JBH ) 96
1
Hz), 2 (t, JBH ) 108 Hz) (see Supporting Information S1 and S2).
(28) As a control, no dehydrocoupled products were observed when 1a
(34) The catalytic conversion of 2a to 3a has previously been reported
n
was treated with a catalytic amount of BuLi.
using Rh(0) catalysts (see ref 15).
9
J. AM. CHEM. SOC. VOL. 132, NO. 11, 2010 3833

A R T I C L E S
Sloan et al.
Scheme 5. Dehydrocoupling of 1b by Cp2TiCl2/2nBuLi to Give 5b
Scheme 7. Attempted Dehydrocoupling of Primary Amine-Borane
Adducts by [Cp2Ti]
Scheme 6. Dehydrocoupling of 1c by Cp2TiCl2/2nBuLi to Give 3c
limitation in this area. Indeed, treatment of Ph2PH·BH3 with 2
mol % [Cp2Ti] at 20 or 40 °C resulted in no detectable reaction
in either case, according to 31P and 11B NMR spectroscopy. We
have previously described the use of electron-withdrawing
fluorinated aryl substituents at the phosphorus center of
phosphine-borane adducts as a means of lowering catalytic
dehydrocoupling reaction temperatures through activation of the
P-H bond through an inductive effect.13 With this in mind,
the Ti-catalyzed dehydrocoupling of the primary phosphine-
borane adduct (p-CF3C6H4)PH2 ·BH3 was also attempted at 20
and 40 °C; however, again no reaction was detected in either
case.
diaminoborane (Me2N)2BH (4a),35 and at δ 37 ppm, which was
tentatively assigned to monomeric 5a, a possible precursor to
both 3a and 2a.36 The generation of a very small amount of an
unidentified species was also noted at δ(11B) -11.6 ppm.
Interestingly, addition of an independently synthesized sample
of 2a15 to a catalytic amount of [Cp2Ti], generated from
Cp2TiCl2/2nBuLi, resulted in complete consumption of 2a and
conversion to 3a (98%), 4a (∼1%) and 5a (∼1%) after 4 h. A
detailed discussion of the kinetics of the dehydrocoupling of
1a by [Cp2Ti] generated from an alternative precatalyst
Cp2Ti(PMe3)2, is presented later (Section 5).
To further investigate the scope of group 4 catalysts we
moved on to explore related species in an attempt to find systems
with comparative or superior catalytic activity.
2. Catalytic Dehydrocoupling of Other Amine-Borane and
Phosphine-Borane Adducts with Cp2TiCl2/2nBuLi. We also
investigated the catalytic activity of [Cp2Ti] toward more
sterically encumbered amine-borane adducts. Treatment of
iPr2NH·BH3 (1b) and MeBnNH ·BH3 (Bn ) CH2Ph) (1c), with
a catalytic amount of [Cp2Ti] resulted in full conversion to
iPr2NdBH2 (5b) and [MeBnN-BH2]2 (3c) in 1 and 1.5 h
respectively (Schemes 5 and 6). By comparison, we have
previously shown that 1b dehydrogenates much more slowly
in the presence of a Rh(I) precatalyst such as [Rh(µ-Cl)(1,5-
cod)]2 (1 mol %, 25 °C, neat, 92% conversion to 5b after 96 h).15
In a further effort to examine the scope of the Ti-catalyzed
dehydrocoupling reaction toward other related species, the
reactivity of the primary amine-borane adducts MeNH2 ·BH3
(1d) and NH3 ·BH3 (1e) was explored. Previous work has shown
that upon treatment with Rh precatalysts at 45 °C these
amine-borane adducts yield borazines [RN-BH]3 (R ) Me, H),
presumably via the borazanes [RNH-BH2]3, together with
insoluble, presumably cross-linked oligomeric material.15 In
contrast to 1a, species 1d showed negligible reaction by 11B
NMR spectroscopy upon treatment with 2 mol % [Cp2Ti],
yielding less than 1% [MeNH-BH2]3 after 16 h at 20 °C.
Furthermore, 1e showed no conversion at all under the same
conditions in diglyme and also at elevated temperatures (45 °C,
16 h) (Scheme 7).37,38
3. Attempted Dehydrocoupling of 1a and 1b with Other
Group 4 Precatalysts. (a) Attempted Catalytic Dehydrocoupling
with [Cp2Zr] and [Cp2Hf]. Due to the high catalytic activity
exhibited by [Cp2Ti], we also investigated the activity of
analogous group 4 metal complexes. Treatment of a toluene
n
solution of Cp2ZrCl2 with 2 equivalents of BuLi at -15 °C,
and subsequent addition of 1a or 1b resulted in no reaction, as
observed by 11B{1H} NMR spectroscopy. On changing the
reaction solvent to THF, again no reaction was observed for 1a
after 20 h; however, 65% conversion of 1b to 5b was detected
after 20 h by 11B NMR spectroscopy.39,40 By comparison,
treatment of 1a or 1b with [Cp2Hf], generated in a manner
analogous to [Cp2Zr], resulted in no reaction in both cases (Table
1).41 This highlights the decrease in catalytic activity associated
with descending the group, a result also noted by Chirik and
co-workers.26
(b) Attempted Dehydrocoupling with [Me2Si(η5-C5H4)2M] (M
) Ti or Zr). The dimethylsilyl-ansa-dichloro-metallocene Me2Si-
(η5-C5H4)2MCl2 can be isolated as dark red-brown and pale
yellow-green crystals for M ) Ti or Zr, respectively.42 Treat-
ment of a toluene solution of Me2Si(η5-C5H4)2TiCl2 with two
n
equivalents of BuLi at -15 °C, and subsequent warming to
20 °C resulted in the formation of a pale-yellow solution.
Addition of 1a resulted in a color change to brown, with a small
amount of gas evolution. 11B{1H} NMR analysis of an aliquot
In an attempt to further expand the scope of the Ti-catalyzed
reaction, the catalytic dehydrocoupling of the secondary
phosphine-borane adduct Ph2PH·BH3 was also investigated.
Phosphine-borane adducts generally undergo catalytic dehy-
drocoupling with Rh catalysts at more elevated temperatures
(60-120 °C). It therefore appeared likely that the thermal
instability38 of the [Cp2Ti] catalyst might provide a fundamental
(39) By comparison to the stoichiometric addition of 1a or 1b to [Cp2Ti],
which results in fast conversion to 3a and 5b, addition of a
stoichiometric amount of 1a or 1b to [Cp2Zr] resulted in the formation
of several new products in each case. The characterization of these
products is ongoing and will be the subject of a future publication.
No reaction was observed upon addition of Me3N·BH3 to either
[Cp2Ti] or [Cp2Zr].
(40) In contrast to the solvent dependence observed for [Cp2Zr], full
conversion of 1a to 5a by [Cp2Ti] required the same time, independent
of the solvent choice (toluene or THF).
(41) The lack of reactivity of [Cp2Hf] is perhaps unsurprising, as
Cp2Hf(nBu)2 has been shown to be an isolable complex at room
temperature, which is only reactive at high temperatures (100 °C),
see: Burlakov, V. V.; Beweries, T.; Bogdanov, V. S.; Arndt, P.;
Baumann, W.; Petrovskii, P. V.; Spannenberg, A.; Lyssenko, K. A.;
Shur, V. B.; Rosenthal, U. Organometallics 2009, 28, 2864–2870.
(42) Bajgur, C. S.; Tikkanen, W.; Petersen, J. L. Inorg. Chem. 1985, 24,
2539–2546.
(35) (a) No¨th, H.; Wrackmeyer, B. Nuclear Magnetic Resonance of Boron
Compounds; Springer-Verlag: Berlin, 1978. (b) Srivastava, D. K.;
Krannich, L. K.; Watkins, C. L. Inorg. Chem. 1991, 30, 2441–2444.
(36) No¨th, H.; Vahrenkamp, H. Chem. Ber. 1967, 100, 3353–3362.
(37) No stoichiometric reaction was observed for 1d or 1e.
(38) This apparent lack of catalytic activity over time and at temperatures
greater than 20 °C may be attributed to the instability of [Cp2M] (M
) group 4 metal) in solution, which has been well-documented, see:
Negishi, E. I. Dalton Trans. 2005, 827–848.
9
3834 J. AM. CHEM. SOC. VOL. 132, NO. 11, 2010

Dehydrocoupling/Dehydrogenation of Amine-Borane Adducts
A R T I C L E S
Table 1. Summary of Catalyst Screening for the Dehydrocoupling
of 1a (to 3a) and 1b (to 5b) (see Schemes 4 and 5)
Scheme 8. Synthesis of Cp2Ti(PMe3)2
conversion of 1a
time (h)
yield (%)f
conversion of 1b
time (h)
yield (%)f
precatalyst
a,b
Cp2TiCl2
Cp2ZrCl2
4
100
0
0
100
50
0
1
100
65
0
100
100
0
Addition of a catalytic amount (2 mol %) of Cp2Ti(PMe3)2
to a solution of 1a or 1b in toluene resulted in full conversion
to 3a and 5b in 7 and 2 h, respectively (Table 1). As in the
case of the Cp2TiCl2/2nBuLi system, the appearance of a signal
corresponding to 2a early in the reaction, and subsequent
conversion to 3a was observed by 11B{1H} NMR spectroscopy,
as well as the formation of small amounts of 4a and 5a (Figure
2).
c,d
c,d
20
20
7
20
20
20
48
48
8
20
20
2
20
20
20
48
48
96
Cp2HfCl2
a
Cp2Ti(PMe3)2
Me2Si(C5H4)2TiCl2
Me2Si(C5H4)2ZrCl2
a
a
Cps2Tia,i
0
0
0
0
Cp2TiCla or TiCl3
a
Ti (100 mesh)e or Ti colloid
0
0
100g
92h
15
[Rh(µ-Cl)(1,5-cod)]2
(e) Attempted Dehydrocoupling of 1a and 1b with Potential
Ti(III) Precatalysts. While it seems likely that catalytic systems
Cp2TiCl2/2nBuLi and Cp2Ti(PMe3)2 both form an active Ti(II)
species, it was also important to consider the potential role of
Ti(III) complexes. To test this, Cp2TiCl was prepared by the
reaction of Cp2TiCl2 with Zn powder in THF to give a green
solution of the desired complex,45 which upon addition to
solutions of either 1a or 1b resulted in no reaction after 48 h.
No reaction was also detected upon addition of TiCl3 to a
solution of 1a or 1b in toluene. The lack of reactivity showed
that these Ti(III) complexes were inactive as precatalysts for
the dehydrocoupling reaction.
4. Elucidation of the Homogeneous or Heterogeneous Nature
of the Active Ti Catalyst. Due to the single site mode of reactivity
of homogeneous catalysts, the identification of such a process
might allow for the identification of reaction intermediates and
elucidation of the reaction mechanism. It is therefore important
to determine if the Ti(II) catalyst system operates in a
homogeneous or heterogeneous manner. The previously de-
scribed dehydrocoupling of 1a using Rh(I) precatalysts such as
[Rh(µ-Cl)(1,5-cod)]2 showed a sigmoidal-shaped kinetic profile
that can be characteristic of a heterogeneous process, with an
induction period prior to Rh colloid generation and a slow tailing
off of the reaction rate associated with subsequent nanoparticle
aggregation (red squares, Figure 3).11,15 In contrast, the Ti(II)
system [Cp2Ti] showed approximately linear consumption of
1a (black circles, Figure 3). By comparison, reactions with
Cp2Ti(PMe3)2 showed a slight induction period, presumably due
to dissociation of the phosphine ligands from the Ti(II) center,
a Reaction carried out with
4 mmol of adduct and 2 mol %
precatalyst in 3 mL toluene at 20 °C. b Full consumption of 1a and 2a
was observed in the same time period when the reaction was carried out
in THF. c Reaction carried out with 4 mmol of adduct and 5 mol %
precatalyst in 3 mL THF at 20 °C. d No reaction of 1a or 2a was
observed with toluene as the reaction solvent. e Reaction carried out
with 4 mmol of adduct and 10 mol % precatalyst in 3 mL toluene at 20
°C. f Conversion to 3a or 5b by integration of 11B{1H} NMR spectrum,
all reactions were carried out three times with reproducible yields.
g Reaction carried out with 3.3 mmol of adduct and 0.5 mol
%
precatalyst in 2 mL toluene at 25 °C (see ref 14). h Reaction carried out
with 19 mmol of adduct and 1 mol % precatalyst, neat at 25 °C (see ref
i
t
14). Cps ) (η5-(C5Me4(SiMe2 Bu))).
after 20 h indicated 50% conversion to 3a. By comparison
treatment of 1b with a catalytic amount of [Me2Si(η5-C5H4)2Ti]
(2 mol %) resulted in 100% conversion to 5b after 20 h. In
accordance with the decrease in reactivity of Zr catalysts
compared to those of Ti noted earlier,26 no reaction was
observed on addition of 1a or 1b to [Me2Si(η5-C5H4)2Zr],
generated in either toluene or THF.
(c) Attempted Dehydrocoupling of 1a and 1b with Cps2Ti (Cps
) η5-(C5Me4(SiMe2tBu))). The assembly of a catalyst in situ can
lead to problems with reproducibility and handling. It is therefore
prudent to identify a single component catalyst or precatalyst,
which can be added directly to the substrate. One of the few
stable Ti(II) complexes, Cps2Ti (Cps ) (η5-C5Me4(SiMe2 Bu))),
t
was reported by Lawless and co-workers, and can be isolated
as red crystals.30 A report by Chirik and co-workers highlighted
the decrease in catalytic activity of group 4 complexes with
highly substituted cyclopentadienyl ligands toward amine-
borane adducts.26 We also observed that an extremely crowded
Ti(II) center resulted in very inefficient dehydrocoupling. Thus,
treatment of 1a with 2 mol % of Cps2Ti led to no conversion to
3a after 92 h as observed by 11B{1H} NMR spectroscopy.
Analogous treatment of 1b with a catalytic amount (2 mol %)
of Cps2Ti resulted in only 5% conversion to 5b after 92 h.
(d) Catalytic Dehydrocoupling of 1a and 1b with Cp2Ti(PMe3)2.
To access an active and isolable precatalyst, the known Ti(II)
43
complex Cp2Ti(PMe3)2 was explored. A slightly modified
literature procedure was used to prepare this species, with
replacement of the reaction solvent (THF) by diethyl ether
preventing decomposition upon workup, allowing pure (by 1H,
13C and 31P NMR spectroscopy) Cp2Ti(PMe3)2 to be isolated
as a dark red/brown microcrystalline solid without the need for
further purification44 (Scheme 8).
Figure 2. 11B{1H} NMR (96 MHz, unlocked, toluene) spectra showing
conversion of 1a (1.3 M in toluene) to 2a and 3a by Cp2Ti(PMe3)2 as
precatalyst (2 mol %, toluene, 20 °C), * ) unidentified species. (a) t ) 9
min, (b) t ) 15 min, (c) t ) 19 min, (d) t ) 29 min, (e) t ) 46 min, (f) t
) 60 min, (g) t ) 106 min, (h) t ) 162 min, (i) t ) 179 min, (j) t ) 213
min, (k) t ) 245 min, (l) t ) 300 min, (m) t ) 364 min, (n) t )
439 min.
(43) Jegat, C.; Fouassier, M.; Tranquille, M.; Mascetti, J. Inorg. Chem.
1991, 30, 1529–1536.
(44) Attempts to recrystallise the microcrystalline product resulted in
significant decomposition to unidentified species. Analysis of pure
material (as assessed by 1H, 13C and 31P NMR spectroscopy) by
elemental analysis and mass spectrometery failed to provide useful
and reproducible data.
9
J. AM. CHEM. SOC. VOL. 132, NO. 11, 2010 3835

A R T I C L E S
Sloan et al.
Figure 3. Graph of % conversion to 3a vs time for the catalytic
dehydrocoupling of 1a using [Rh(µ-Cl)(1,5-cod)]211 (2 mol % Rh, toluene,
20 °C, red squares), [Cp2Ti] (2 mol % Ti, toluene, black circles) and
Cp2Ti(PMe3)2 (2 mol % Ti, toluene, blue diamonds).
followed by rapid catalytic consumption of 1a to give 3a (blue
diamonds, Figure 3).46
Further insight was provided by experiments involving the
addition of mercury to the catalytically active reaction mixtures.
Mercury is a well-known poison of heterogeneous metal
catalysts, owing either to its adsorption onto the catalyst surface
or formation of an amalgam.47 In contrast to the Rh-catalyzed
dehydrocoupling of 1a, which is completely halted by its
addition,11,15 the Ti-catalyzed reactions were unaffected even
in the presence of a large excess (e.g., 71 equivalents) of the
group 12 metal. A further indication that the latter is a
homogeneous process was provided by nanofiltration experi-
ments. These involved filtering the reaction mixture through a
small pore (200 nm) membrane in order to distinguish between
soluble and insoluble catalysts.48 If the activity is dramatically
lowered upon filtration, then an insoluble heterogeneous catalyst
can be assumed. Significantly, filtration of the catalytically active
[Cp2Ti] and Cp2Ti(PMe3)2 solutions after 80 and 40 min reaction
time respectively resulted in no detectable decrease in the rate
of conversion of 1a to 3a (see Figure S6, Supporting Informa-
tion, for data involving the Cp2Ti(PMe3)2 precatalyst system).
In analyzing these results, an important fact to consider is that
any very small, colloidal Ti particles present in the reaction
mixture would pass through the 200 nm filter and would, if
catalytically active, result in a false positive test for a homo-
geneous catalyst. Notably however, both THF-soluble colloidal
Ti, generated from TiBr4 ·2THF and K[BEt3H], in addition to
bulk Ti powder, were found to be inactive for the catalytic
dehydrocoupling of 1a or 1b. The results involving the kinetic
profile, mercury poisoning and nanofiltration experiments
therefore strongly support the presence of a homogeneous Ti
catalyst.
Figure 4. Graph showing conversion of (a) 1a (1.3 M in toluene, blue
diamonds) to 2a (green squares) and 3a (red triangles) by Cp2Ti(PMe3)2 as
precatalyst (0.027 M, 2 mol % in toluene) and (b) 2a (0.67 M in toluene,
green squares) to 3a (red triangles) by Cp2Ti(PMe3)2 (0.027 M, 4 mol % in
toluene) (small dots ( · · · · · · ) indicate data from kinetic model).
5. Kinetic and Mechanistic Studies for the Cp2Ti(PMe3)2
Precatalyst System. As noted above, when the reaction of 1a
with Cp2Ti(PMe3)2 was monitored by 11B NMR spectroscopy,
significant amounts of 2a were detected at low conversion of
the amine-borane adduct, followed by consumption to form
3a (see Figure 2). It therefore appears likely that formation of
3a occurs in two steps. Initially, the interaction of two
equivalents of 1a with [Cp2Ti] could result in the formation of
a new B-N bond (to give 2a), with the loss of one equivalent
of H2. Interaction of 2a with the same active catalyst could then
result in the loss of another equivalent of H2, and the formation
of another B-N bond to yield 3a, along with trace amounts of
4a and 5a (Scheme 4). Because the results predicted by
computational modeling25 did not reflect the results observed
for the catalytic dehydrocoupling of 1a by [Cp2Ti] or
Cp2Ti(PMe3)2, we decided to carry out a preliminary mechanistic
investigation.
Some insight into the reaction mechanism was obtained by
monitoring reactions at different initial concentrations (1a ) 2
M, Cp2Ti(PMe3)2 ) 0.040 M; 1a ) 0.67 M, Cp2Ti(PMe3)2 )
0.015 M) and different precatalyst loadings (1a ) 1.3 M,
Cp2Ti(PMe3)2 ) 0.040 M; 1a ) 1.3 M, Cp2Ti(PMe3)2 ) 0.013
M). To ensure that the data obtained was of a suitable standard
for a kinetic study, all reactions were repeated several times,
with highly reproducible kinetics observed in every case (see
overlay for three separate experiments in Figures 4 and S9-S12,
Supporting Information). To investigate the conversion of 2a
to 3a a sample of 2a was synthesized by the literature method.15
Addition of a catalytic amount of Cp2Ti(PMe3)2 (0.027 M in
toluene) to a solution of 2a (0.67 M in toluene) resulted in
(45) Rajanbabu, T. V.; Nugent, W. A. J. Am. Chem. Soc. 1994, 116, 986–
997.
(46) As the dehydrocoupling reactions of 1a by both [Cp2Ti] and
Cp2Ti(PMe3)2 occur at very similar rates (after the brief induction
period in the latter case) the presence of phosphine in the reaction
mixture appears to have a negligible impact on the reaction at the
concentrations in our experiments.
(47) (a) 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, 1819–1830. (b) Anton, D. R.;
Crabtree, R. H. Organometallics 1983, 2, 855–859.
(48) Widegren, J. A.; Finke, R. G. J. Mol. Catal. A: Chem. 2003, 198,
317–341.
9
3836 J. AM. CHEM. SOC. VOL. 132, NO. 11, 2010

Dehydrocoupling/Dehydrogenation of Amine-Borane Adducts
A R T I C L E S
Scheme 9. Catalytic Cycle Employed to Model the Kinetics of the [Cp2Ti] Catalyzed Dehydrocoupling of 1a
Scheme 10. Synthesis of Deuterated Dimethylamine-Borane Ad-
ducts 1aND, 1aBD and 1aNDBD
complete conversion to 3a in 5 h (Figures 4b and S8a,
Supporting Information). It is significant to note that the reaction
of preformed 2a with Cp2Ti(PMe3)2 did not result in the
reformation of 1a or significant amounts of 4a (2.5%) and 5a
(0.5%), strongly indicating that 2a is indeed the sole precursor
to 3a.
Although full analytical rate equations can be derived for
catalytic systems, for example using the method of King and
Altman,49 the complex nature of the reaction networks involved
can make it challenging to simplify the rate equation through
the application of limiting conditions. In such cases, computa-
tional modeling can be used to efficiently test various mecha-
nistic hypotheses, through the numerical fitting of simple models
to the data.50 By applying a two-cycle model to our experimental
data, each involving two steps, it was possible to achieve a
reasonable fit (Scheme 9 and Figures 4 and S7-S12, Supporting
Information).
In this simple model, the precatalyst Cp2Ti(PMe3)2 is acti-
vated51 by 1a to form active species [Cp2Ti], which then
interacts with a molecule of 1a to form the intermediate [IntA].
Addition of a further equivalent of 1a to [IntA] results in loss
of both H2 and 2a from the metal center and regeneration of
[Cp2Ti]. The interaction of 2a with the same active catalyst,
[Cp2Ti], then results in the generation of intermediate [IntB],
from which 3a and a second equivalent of H2 are lost, also
regenerating [Cp2Ti]. The resting states of the system are [IntA]
and [IntB], resulting in an approximate pseudo zero-order
consumption of 2a and evolution of 3a. Importantly, the same
model satisfactorily predicted the kinetics of the independent
conversion of 2a to 3a (Figures 4b and S18, Supporting
Information).
by the reaction of Me2NH with BD3 ·THF yielding Me2NH·BD3
(1aBD) (74% yield), exchange of the protic hydrogen of 1a or
1aBD in D2O, yielding Me2ND·BH3 (1aND) (82% yield) and
Me2ND·BD3 (1aNDBD) (84% yield), respectively (Scheme 10).
We performed the catalytic dehydrocoupling of the deuterated
amine-borane adducts 1aND, 1aBD and 1aNDBD with
Cp2Ti(PMe3)2 and observed by 11B NMR spectroscopy the initial
formation of linear dimer, and subsequent conversion to cyclic
dimer, as with the nondeuterated adduct 1a. Reactions reached
100% conversion to cyclic dimer 3a in 32 h, 9 and 33 h for
1aND, 1aBD and 1aNDBD, respectively (compared to 7 h for
1a under similar conditions) (see Figures S7 and S13-S15,
Supporting Information).
Due to oversimplifications necessary in the derivation of the
kinetic model, this data could not be used to extract kinetic
isotope effects pertinent to individual steps. Instead, the pseudo
zero-order phases of consumption of 1a and evolution of 3a,
were employed to extract phenomenological kH/kD values of 3.6
( 0.3 for N-H/D, 1.7 ( 0.3 for B-H/D and 4.5 ( 0.3 for
B-H/D + N-H/D in the conversion of 1a to 2a, and 4.7 (
0.3 for N-H/D, 1.4 ( 0.2 for B-H/D and 5.3 ( 0.3 for B-H/D
+ N-H/D in the conversion of 2a to 3a (Figures S16-S17,
Supporting Information).
The low kH/kD values associated with B-H/D cleavage in
both steps (1a to 2a and 2a to 3a) may arise from small transfer
angles, exceptionally early or late transition states or B-H/D
cleavage outside of the turnover limiting step.52 However, due
to the limitations of the 11B NMR spectroscopic method
employed we did not conduct competition experiments to probe
for the latter.53
The observation of kinetic isotope effects can yield important
information pertaining to reaction mechanisms. We therefore
explored the kinetic isotope effects associated with the dehy-
drocoupling of 1a. The synthesis of deuterated 1a was achieved
(49) King, E. L.; Altman, C. J. Phys. Chem. 1956, 60, 1375–1378.
(50) (a) Stabler, R. N.; Chesick, J. P. Int. J. Kinet. 1978, 10, 461–469. (b)
Weigert, F. J. Comput. Chem. 1987, 11, 273–280.
Discussion
(51) Comparison of initial rates of consumption of 1a and 2a by
Cp2Ti(PMe3)2 indicated a difference in the induction period. Whilst a
dissociative mechanism for the loss of the first PMe3 ligand from the
precatalyst seems reasonable, loss of the second ligand, and thus
generation of the active catalyst, may be influenced differently by 1a
and 1b.
The involvement of linear dimer 2a in the dehydrocoupling
15
of 1a has been previously suggested for [Rh(µ-Cl)(1,5-cod)]2
(52) O’Ferrall, R. A. J. Chem. Soc. B. 1970, 785–790.
9
J. AM. CHEM. SOC. VOL. 132, NO. 11, 2010 3837

A R T I C L E S
Sloan et al.
Scheme 11. Summary of Catalytic Cycle with Proposed Intermediates
as precatalyst and also very recently for cationic Rh-18 and
neutral Ru-based7 systems. However, previous computational
work by Ohno and Luo discounted the formation of 2a during
the catalytic dehydrocoupling of 1a by [Cp2Ti].25 Their calcula-
tions implied that 3a was formed via an intramolecular, stepwise
mechanism (exergonic by 7.4 kcal/mol), where 5a is formed
during the interaction of 1a with [Cp2Ti] and then undergoes
an uncatalyzed dimerization cyclization reaction to form 3a
(Scheme 3). These workers also investigated the possibility of
an intermolecular mechanism (1a f 2a f 3a), but eventually
ruled it out for kinetic and thermodynamic reasons; the transition
state barrier for [Cp2Ti]-catalyzed transformation of 1a to 2a
was calculated to be + 37.2 kcal/mol and the reaction to be
endergonic by 4.1 kcal/mol.25 As a result, they did not calculate
a route to the catalytic ring closing dehydrocoupling of 2a to
yield 3a. It is therefore significant to experimentally observe
that 2a is both generated and catalytically converted to 3a, rather
than 3a being generated by the cycloaddition of 2 molecules of
5a,54 and that in the absence of catalyst, 2a is stable. This
demonstrates that the reaction mechanism is different than that
proposed by Luo and Ohno (Scheme 3).
in proton transfer and formation of intermediate [IntA],
[Cp2Ti(H)NMe2 ·BH3]. The primary KIE of 3.6 ( 0.3 may be
associated with this step; the magnitude of the effect being
similar to that observed in other transition metal mediated N-H
cleavage reactions.55 A reaction mechanism of this type was
proposed by Ohno and Luo as the first step in the intramolecular
stepwise catalytic dehydrocoupling of 1a (Scheme 3). A range
of similar titanium amido-borane complexes have previously
been reported by Lancaster and co-workers,56 while analogous
zirconocene hydrido amido-borane complexes [Cp2Zr(H)NH2 ·
BH3] have recently been reported by Roesler and co-workers.57
The second step would involve the approach of a second
molecule of 1a to [IntA] and formation of a new B-N bond
between the two adducts at the metal center, followed by
dissociation of 2a and loss of H2 from the Ti center and
reformation of the catalytically active species [Cp2Ti].
Conversion of 2a to 3a could then proceed via a similar route
to the formation of 2a, with precoordination of 2a to the Ti-
center through a 3-center-2-electron Ti-H-B interaction at the
BH3 terminus, followed by interaction of Ti with the N-H bond,
forming intermediate [IntB], [Cp2Ti(η2-BH3-NMe2BH2NH-
Me2]. Similar complexes have been reported for the later
transition metals by Weller and co-workers, for example
Regarding the overall mechanism, in Scheme 11 we propose
a two cycle mechanism with plausible proposed structures for
intermediates. We propose that 1a initially coordinates to
[Cp2Ti] via a 3-center-2-electron Ti-H-B interaction to give
[Cp2Ti(η2-BH3 ·NHMe2)]. Complexes of this type have been
characterized for late transition metals. For example, the
involvement of the late transition metal complex [Rh(PiBu3)2-
[Rh(PiBu3)2(η2-H3B·NMe2BH2 ·NHMe2)][BArF ],18 while Shi-
4
moi and co-workers have reported the in situ identification of
a η1-bound early transition metal complex of 2a, [Cr(CO)5(η1-
BH3-NMe2-BH2-NHMe2)].21 Such species could then undergo
a Ti/N-H insertion reaction to yield 3a (Scheme 11).
(η2-H3B·NHMe2)][BArF ] in the catalytic dehydrocoupling of
4
1a has been discussed by Weller and co-workers.9 Interaction
(55) Stubbert, B. D.; Marks, T. J. J. Am. Chem. Soc. 2007, 129, 6149–
6167.
of the protic hydrogen with the Ti(II) center would then result
(56) (a) Fuller, A. M.; Clegg, W.; Harrington, R. W.; Hughes, D. L.;
Lancaster, S. J. Chem. Commun. 2008, 5776–5778. (b) Mountford,
A. J.; Clegg, W.; Coles, S. J.; Harrington, R. W.; Horton, P. N.;
Humphrey, S. M.; Hursthouse, M. B.; Wright, J. A.; Lancaster, S. J.
Chem.sEur. J. 2007, 13, 4535–4547. (c) Mountford, A. J.; Clegg,
W.; Harrington, R. W.; Humphrey, S. M.; Lancaster, S. J. Chem.
Commun. 2005, 2044–2046.
(53) Due to overlap of the signals corresponding to deuterated and non-
deuterated adducts 1a, 1aND, 1aBD and 1aNDBD, analysis of the
corresponding 11B or 1H NMR signals could not provide information
regarding the outcome of a competition reaction.
(54) Analysis of the reaction mixture by 11B NMR spectroscopy identified
small amounts of 5a, which remained constant throughout the reaction,
and were still present after full consumption of 1a. It is therefore clear
that in this reaction 3a is not mainly formed by the cycloaddition of
two molecules of 5a; however, we cannot discount this reaction playing
a role as a minor pathway.
(57) Forster, T. D.; Tuononen, H. M.; Parvez, M.; Roesler, R. J. Am. Chem.
Soc. 2009, 131, 6689–6691.
(58) Franke, R.; Rothe, J.; Pollmann, J.; Hormes, J.; Bo¨nnemann, H.;
Brijoux, W.; Hindenburg, T. J. Am. Chem. Soc. 1996, 118, 12090–
12097.
9
3838 J. AM. CHEM. SOC. VOL. 132, NO. 11, 2010

Dehydrocoupling/Dehydrogenation of Amine-Borane Adducts
A R T I C L E S
Our mechanistic proposals for the [Cp2Ti] system are interest-
ing to compare with those of other groups for related early and
also late transition metal catalysts. As noted above, the
computational results of Luo and Ohno25 for the [Cp2Ti]-
catalyzed dehydrogenation of 1a appear inconsistent with our
results, which point to a key role for the linear dimer 2a. Chirik
and co-workers have also studied the use of group 4 metallocene
complexes for the catalytic dehydrogenation of amine-borane
adducts.26 Based on reactions performed under an atmosphere
of D2 and first order rate constants for 1a as a substrate and the
highly active Ti(II) dinitrogen complex [{(1,3-(SiMe3)2C5H3)2Ti}2-
(η1-N2)] as a precatalyst, these workers suggested that the
reaction proceeds initially via reversible B-H oxidative addition
to form a Ti-boryl complex. Subsequent, reversible ꢀ-hydride
elimination to generate 5a, which then dimerizes to the cyclic
dimer 3a, was then proposed. This mechanism, which also does
not postulate a role for 2a, is similar to that previously proposed
for amineborane dehydrogenation by Ni6 and more recently Re
catalysts.20 In the case of Cr carbonyl21 and Ir pincer
complexes5,19 as catalysts, the formation of linear dimers has
also not been noted and, in these cases, a mechanism that
involves concerted removal of H2 from the substrate has been
proposed.59
of 1a to 3a, and only slow decay of the signal corresponding to
5a (Figure S19 and S20, Supporting Information).62
Summary
We have demonstrated the efficient homogeneous catalytic
dehydrocoupling of the secondary amine-borane adduct 1a by
the putative early transition metal complex [Cp2Ti], generated
in situ by the reaction of Cp2TiCl2 and 2nBuLi or from the
isolable precatalyst Cp2Ti(PMe3)2. Catalytic screening of a series
of group 4 complexes has highlighted the decreased activity on
descending the group and on substitution of the cyclopentadienyl
ligands by bulky or electron donating groups. The lack of
reactivity of the secondary amine-borane adducts 1a and 1b
with either Ti(III) or Ti(0) species supports the presence of a
catalytically active Ti(II) complex. In contrast to the case with
Rh(I) precatalysts, the electron rich and sterically encumbered
amine-borane adduct 1b shows enhanced reactivity, suggesting
that in the case of the early transition metal catalyst the
electronic and/or steric nature of the adduct has an opposite
effect on the reaction rate. Kinetic analysis of the conversion
of 1a to 3a with Cp2Ti(PMe3)2 as a precatalyst indicates a two
stage catalytic reaction mechanism, with initial formation of
linear species 2a, which undergoes a catalytic ring-closing
dehydrocoupling process to afford cyclic species 3a. These
experimental results contrast with theoretical predictions25 that
suggested monomeric species 5a is formed catalytically, and
then undergoes an uncatalysed 2 + 2 cycloaddition reaction to
form 3a.
Recently, other workers have also reported intriguing studies
of the catalytic dehydrogenation of 1a by late transition metal
complexes where the linear dimer 2a has been detected as an
intermediate.60 For example, Weller, Hall and co-workers have
proposed that catalysis by the cationic Rh(I) complexes
[Rh(PR3)2][BArf4] proceeds via sequential B-H/N-H activation
and transfer steps followed by sequential loss of H2 and 5a. In
this case, although linear dimer 2a was detected in reaction
mixtures, the mechanistic role of this species was unclear.
Moreover, conversion of 2a to 3a was not found to occur via a
simple intramolecular route.18 In another study, Schneider and
co-workers have also investigated the dehydrocoupling of 1a
by a Ru amido pincer complex and identified the linear dimer
2a as an important intermediate.7 These workers suggest that
linear 2a and 5a are formed from 1a via competitive pathways.
Interestingly, in this case, conversion of 2a to 3a appears to
take place via a combination of pathways that involve direct
metal-catalyzed dehydrogenative cyclization and B-N bond
scission to afford 5a, which subsequently dimerizes. In our study
with [Cp2Ti] as the catalyst, the former pathway appears to be
dominant for the formation of 3a. Nevertheless, work by
Zimmerman and co-workers61 has highlighted the possible
autocatalytic role of NH2dBH2, which leads to H2 loss and
oligomerization of 1e, without the need for a transition metal
catalyst. In considering an analogous possibility for 1a we were
able to exclude such autocatalytic pathways by observing that
isolating a mixture of products 2a, 3a, 4a, and 5a and starting
material 1a from the catalyst resulted in no further conversion
Experimental Section
General Procedures and Materials. All reactions and manipu-
lations were performed under an atmosphere of dry argon using
standard Schlenk techniques or in a MBraun glovebox filled with
dry argon. NMR experiments were carried out on Jeol ECP 300
MHz (31P, 11B, 1H, 2H), Jeol Lambda 300 MHz (31P, 1H) or Varian
VNMR 400 MHz (1H, 13C) spectrometers. Spectra were internally
2
referenced to residual solvent peaks (1H, H, 13C) or externally
referenced to BF3 ·Et2O (11B) or H3PO4 (31P). Tetrahydrofuran was
distilled from Na/benzophenone. Diethyl ether, toluene and hexanes
were purified using a Grubbs solvent system. Mercury (99.9995%),
Ti powder (100 mesh), TiCl3, TiCl4, Me2NH (2 M solution in THF),
n
MeNH2 (gas), PMe3 and BuLi (1.6 M solution in hexanes) and
Millex Millipore filters (PTFE membrane, 0.2 µm pore size) were
purchased from Aldrich, Cp2TiCl2 (98%) was purchased from
Acros, Zn powder was purchased from BDH laboratories, Na was
purchased from Fisher Scientific, Cp2HfCl2 was purchased from
Strem and Cp2ZrCl2 was purchased from Fluka, all used as received.
BD3 ·THF (1 M THF) was purchased from Cambridge Isotope
Laboratories and was used as received. D2O was purchased from
Fluorochem and was degassed (3 × freeze-pump-thaw cycles)
prior to use. BH3 ·THF (1 M THF) was purchased from Acros and
distilled under a static vacuum prior to use. Me2NH·BH3 (97%)
was purchased from Aldrich and further purified by vacuum
sublimation twice at 20 °C. iPr2NH and MeBnNH were purchased
from Aldrich and were vacuum distilled from CaH2. Cp2TiCl,45
Me2Si(η5-C5H4)2MCl2 where M ) Ti or Zr,42 Cp2Ti(PMe3)2,43 (η5-
30
t
C5Me4SiMe2 Bu)2Ti,
Ph2PH · BH3,12 (p-CF3C6H4)PH2 · BH3,13
i
R′RNH ·BH3 (R′ ) R ) Pr; R′ ) Bn, R ) Me; R′ ) Me, R )
(59) Ankan, P.; Musgrave, C. B. Angew. Chem., Int. Ed. 2007, 46, 8153–
8156.
H),15 Me2NH-BH2-NMe2-BH3 and THF soluble Ti colloids58
15
(60) During the analogous treatment of 1c with [Cp2Ti] 11B NMR signals
assigned to linear dimer MeBnNH-BH2-MeBnN-BH3 were also
detected. The linear dimer was then found to decrease in concentration
as cyclic dimer 3c was formed. In the case of 1b no intermediates
were detected by 11B NMR spectroscopy in the conversion to 5b.
(61) Zimmerman, P. M.; Paul, A.; Zhang, Z.; Musgrave, C. B. Inorg. Chem.
2009, 48, 1069–1081.
were synthesized following literature procedures. All reactions were
repeated at least twice with reproducible yields.
(62) See experimental for details. The compound H2B(µ-NMe2)(µ-H)BH2
was also formed during isolation of the reactant and products from
the catalyst.
9
J. AM. CHEM. SOC. VOL. 132, NO. 11, 2010 3839

A R T I C L E S
Sloan et al.
i
Catalytic Dehydrocoupling of Me2NH ·BH3 (1a), Pr2NH ·BH3
(1b), MeBnNH ·BH3 (1c) or MeNH2 ·BH3 (1d) by Cp2TiCl2/
2nBuLi. To a solution of Cp2TiCl2 (0.020 g, 0.080 mmol) in toluene
(2 mL), nBuLi (0.10 mL, 0.160 mmol) was added at -15 °C giving
an amber solution. After 5 min, the reaction mixture was removed
from the cooling bath and to this solution, 1a (0.24 g, 4.1 mmol),
1b (0.46 g, 4.0 mmol), 1c (0.54 g, 4.0 mmol) or 1d (0.18 g, 4.0
mmol) in toluene (1 mL) was added resulting in a color change to
black and gas evolution in the cases of 1a, 1b and 1c. An aliquot
was then removed and analyzed by 11B NMR spectroscopy,
indicating complete conversion to products 3a 11B NMR δ 4.8 (t,
1JB-H ) 112 Hz), 5b 11B NMR δ 34.7 (t, 1JB-H ) 127 Hz) and 3c
(1 mL) was added, resulting in a color change to dark green. After
stirring at 20 or 40 °C for 16 h, a small amount of solution was
removed from the reaction mixture and the 11B and 31P NMR spectra
were obtained, showing single signals (11B δ -33.6, 31P δ -48.3)
in both cases corresponding to unreacted (p-CF3C6H4)PH2 ·BH3.13
Attempted Dehydrocoupling of 1a and 1b with [Cp2Zr]. To a
solution of Cp2ZrCl2 (0.025 g, 0.085 mmol) in THF (2 mL), nBuLi
(0.10 mL, 0.17 mmol) was added at -15 °C to give a light-green
solution which was stirred for 5 min and then allowed to warm to
20 °C for 5 min. A solution of 1a (0.10 g, 1.7 mmol) or 1b (0.20
g, 1.74 mmol) in THF (1 mL) was added and the solution stirred
for 20 h after an aliquot was removed for analysis by 11B NMR
spectroscopy.
Attempted Dehydrocoupling of 1a and 1b with [Cp2Hf]. To a
solution of Cp2HfCl2 (0.064 g, 0.16 mmol) in THF or toluene (2
mL), nBuLi (0.21 mL, 0.34 mmol) was added at -15 °C to give a
colorless solution which was stirred for 5 min and then allowed to
warm to 20 °C for 5 min. A solution of 1a (0.20 g, 3.5 mmol) or
1b (0.37 g, 3.2 mmol) in THF or toluene (1 mL) was added and
the solution stirred for 20 h, after which analysis of an aliquot by
11B NMR spectroscopy indicated only starting material for 1a and
product 5b when either THF or toluene was used.
11B NMR δ 3.9 (t, JB-H ) 99 Hz) after 4 h, 1 and 1.5 h
1
respectively.
Control Reaction of 1a with nBuLi. To a solution of 1a (0.051
g, 0.87 mmol) in toluene (2 mL), BuLi (0.050 mL, 0.08 mmol)
was added at -10 °C with no observable change. The reaction
mixture was allowed to warm to 20 °C and stirred for 16 h. After
this time, an aliquot of the reaction mixture was removed and the
11B NMR spectrum obtained, showing a single signal corresponding
to unreacted 1a.
n
Catalytic Dehydrocoupling of 1a. Mercury Poisoning Experi-
ment. A typical dehydrocoupling trial with [Cp2Ti] (from systems
Cp2TiCl2/2nBuLi or Cp2Ti(PMe3)2) was allowed to proceed with
monitoring by 11B NMR spectroscopy for 80 min. After this time
the reaction mixture was transferred to a flask containing Hg (71
equiv) and stirred vigorously. IMPORTANT. Good stirring is
necessary to ensure contact of the catalyst with Hg(0), a condition
required to avoid false negatives in this experiment. Periodic
monitoring was performed by 11B NMR spectroscopy, which
showed complete conversion to 3a after 4 and 7 h, for the
precatalyst systems Cp2TiCl2/2nBuLi and Cp2Ti(PMe3)2 respectively.
Catalytic Dehydrocoupling of 1a. Filtration Experiment. A
typical dehydrocoupling trial with [Cp2Ti] (from systems Cp2TiCl2/
2nBuLi or Cp2Ti(PMe3)2) was allowed to proceed with monitoring
by 11B NMR spectroscopy for 80 min. After this time, the mixture
was filtered through a 0.2 µm filter into a new flask with a new stir
bar and the reaction continued with monitoring by 11B NMR
spectroscopy showing full conversion to 3a after 4 and 7 h for the
precatalyst systems Cp2TiCl2/2nBuLi and Cp2Ti(PMe3)2 respectively.
Attempted Reaction of 1a with Colloidal Ti Metal. To a solution
of 1a (0.20 g, 3.4 mmol) in toluene (2 mL), Ti (0.016 g, 0.33 mmol)
in THF (1 mL) was added at 20 °C with no observable change.
The reaction mixture was allowed to stir for 16 h. After this time,
an aliquot of the reaction mixture was removed and the 11B NMR
spectrum showed a single signal corresponding to unreacted 1a.
Attempted Reaction of 1a with Ti Powder. To a solution of 1a
(0.20 g, 3.4 mmol) in toluene (3 mL) Ti powder (16.0 mg, 100
mesh) was added, and the mixture was stirred for 20 h. Analysis
of an aliquot by 11B NMR spectroscopy showed only unreacted
starting material 1a.
Attempted Catalytic Dehydrocoupling of 1e by [Cp2Ti]. To a
solution of [Cp2Ti], generated in the usual manner, a slurry of 1e
(0.13 g, 4.1 mmol) in diglyme (5 mL) was added resulting in a
color change to gray. After stirring at 20 or 45 °C for 16 h, a small
amount of solution was removed from the reaction mixture and
the 11B NMR spectrum obtained which showed a single signal
corresponding to unreacted 1e.
Attempted Catalytic Dehydrocoupling of Ph2PH·BH3 by
[Cp2Ti]. To a solution of [Cp2Ti], generated in the usual manner,
a solution of Ph2PH·BH3 (0.19 g, 0.95 mmol) in toluene (1 mL)
was added resulting in a color change to blue-black. After stirring
at 20 or 40 °C for 16 h, a small amount of solution was removed
from the reaction mixture and the 11B and 31P NMR spectra were
obtained, which showed single signals (11B δ -41.0, 31P δ 1.97)
in each case corresponding to unreacted Ph2PH·BH3.12
Attempted Catalytic Dehydrocoupling of 1a or 1b with
Cp2TiCl. To a solution of Cp2TiCl2 (0.0050 g, 0.020 mmol) in THF
(2 mL) Zn powder (0.0050 g, 0.076 mmol) was added and the
resulting mixture stirred for 30 min, resulting in a bright green
solution. To this a solution of 1a (0.25 g, 4.2 mmol) or 1b (0.36 g,
3.1 mmol) in THF (1 mL) was added. The reaction mixture was
allowed to stir for 20 h. After this time, an aliquot of the reaction
mixture was removed and the 11B NMR spectrum obtained which
showed a single signal corresponding to unreacted 1a or 1b,
respectively.
Attempted Catalytic Dehydrocoupling of 1a or 1b with TiCl3.
To a solution of 1a (0.25 g, 4.2 mmol) or 1b (0.36 g, 3.1 mmol) in
toluene (3 mL), TiCl3 (0.013 g, 0.084 mmol) was added at 20 °C
giving a purple slurry. The reaction mixture was allowed to stir
for 20 h. After this time, an aliquot of the reaction mixture was
removed and the 11B NMR spectrum obtained showing a single
signal corresponding to unreacted 1a and 1b, respectively.
Synthesis of Cp2Ti(PMe3)2. Repeated attempts to synthesize
Cp2Ti(PMe3)2 by established methods failed to achieve pure
material.43 A suspension of Cp2TiCl2 (2.0 g, 8.0 mmol) in a solution
of PMe3 (4.0 mL, 39 mmol) in Et2O (80 mL) was slowly added to
Na/Hg (100 g, 1 wt %), and the resulting mixture stirred for 12 h.
The resulting dark-red/brown solution was filtered through Celite,
and the volatiles removed under a strong flow of argon giving 1.76 g
(66% yield) of the dark-red/black crystalline solid Cp2Ti(PMe3)2.
Attempted recrystallization resulted in partial decomposition of the
1
product and a reduction in purity. H NMR (C6D6, 400 MHz) δ
4.57 (s, 10H), 0.84 (s, 18H), 13C NMR (C6D6, 100 MHz) δ 90.1
(s, 2 × Cp), 22.5 (s, 6 × Me), 31P NMR (C6D6, 121 MHz) δ 40.6
(s).
Synthesis of Me2ND ·BH3 (1aND). Me2NH·BH3 (10.0 g, 0.17
mol) was dissolved in degassed D2O (30 mL, 1.7 mols) and stirred
at 40 °C for 24 h. The solution was then extracted with CH2Cl2 (3
× 100 mL), the organic layer was dried with MgSO4 and filtered,
and the volatiles were removed. The resulting solid was sublimed
twice under dynamic vacuum at 20 °C to give 8.21 g (82% yield)
of 1aND. 1H NMR (CDCl3, 300 MHz) δ 2.5 (s, 6H), 1.5 (q, 1J BH
2
) 92 Hz, 3H), H NMR (CHCl3, 46 MHz) δ 3.8 (br), 13C NMR
(CDCl3, 100 MHz) δ 44.6 (s), 11B NMR (96 MHz, CDCl3) δ -14.1
1
(q, J
) 92 Hz).
BH
Synthesis of Me2NH ·BD3 (1aBD). Me2NH (28 mL, 55.0 mmol,
2 M in THF) was added to a solution of BD3 ·THF (50 mL, 50.0
mmol, 1 M in THF) in THF (100 mL) at -75 °C, and the solution
allowed to warm slowly to 20 °C. Removal of the volatiles gave a
colorless solid which was sublimed under dynamic vacuum twice
at 20 °C to give 2.3 g (74% yield) of 1aBD. 1H NMR (CDCl3, 300
Attempted Catalytic Dehydrocoupling of (p-CF3C6H4)PH2 ·BH3
by [Cp2Ti]. To a solution of [Cp2Ti], generated in the usual manner,
a solution of (p-CF3C6H4)PH2 ·BH3 (0.31 g, 1.6 mmol) in toluene
2
MHz) δ 4.1 (br, 1H), 2.5 (s, 6H), H NMR (CHCl3, 46 MHz) δ
9
3840 J. AM. CHEM. SOC. VOL. 132, NO. 11, 2010

Dehydrocoupling/Dehydrogenation of Amine-Borane Adducts
A R T I C L E S
2.9 (q, br., 1J BD ) 14.2 Hz), 13C NMR (CDCl3, 100 MHz) δ 44.5
(s), 11B NMR (96 MHz, CDCl3) δ -14.1 (septet, 1JBD ) 14.2 Hz).
Synthesis of Me2ND·BD3 (1aNDBD). Me2NH·BD3 (1.1 g, 17.0
mmol) was dissolved in degassed D2O (7 mL, 0.39 mol) at 40 °C
and stirred for 24 h. The solution was then extracted with CH2Cl2
(3 × 100 mL), the organic layer was dried with MgSO4 and filtered,
and the volatiles were removed. The resulting solid was sublimed
twice under dynamic vacuum to give 0.895 g (84% yield) of
change to black and gas evolution. After 50 min reaction time the
reaction mixture was cooled to -78 °C and a distillation bridge
and receiver flask attached. The volatile contents of the reaction
were vacuum transferred to the receiver flask, with heating. Analysis
by 11B NMR, showed a mixture of 1a (∼ 1.4%), 2a (∼22.4%), 3a
(∼66.5%), 4a (∼4.3%), 5a (∼0.7%) and (µ-NMe2)B2H5 (∼4.7%)
(11B NMR (96 MHz, toluene unlocked) δ -18.3 (dt, 1J BHt ) 129
Hz, 1J BHb ) 30 Hz) (lit. δ -17.0 (dt, 1J BHt ) 130 Hz, 1J BHb ) 30
Hz)63) After 24 h reexamination of the mixture by 11B NMR
spectroscopy showed the same product distribution except for the
amount of 5a which had decreased to ∼0.2% (Figure S19 and S20,
Supporting Information).
1
2
1aNDBD. H NMR (CDCl3, 300 MHz) δ 2.5 (s, 6H), H NMR
1
(CHCl3, 46 MHz) δ 4.1 (s br), 2.9 (q, br., J
) 14.1 Hz), 13C
BD
NMR (CDCl3, 100 MHz) δ 44.6 (s), 11B NMR (CDCl3, 96 MHz)
1
δ -14.0 (sept, J
) 14.1 Hz).
BD
General Method for Kinetic Monitoring of Dehydrocoupling
of 1a or 2a by Cp2Ti(PMe3)2. To a solution of Me2N(H/D) ·B(H/
D)3 (8 mmol) or 2a (4 mmol) in toluene (4 mL) Cp2Ti(PMe3)2 (2
mol % or 4 mol % for 1a and 2a respectively) in toluene (2 mL)
was added, and the resulting mixture was monitored by 11B NMR
spectroscopy at regular time intervals until 100% conversion to 3a
was achieved.
Reaction conditions were varied in 1a concentration (0.67, 1.3,
and 2.0 M) at constant precatalyst loading. Precatalyst loading was
varied (1, 2, and 3 mol %) for 1a, and adduct choice (Me2NH ·BH3,
Me2ND·BH3, Me2NH·BD3 or Me2ND·BD3).
Isolation of Reactants and Products from Catalyst. To a solution
of Cp2TiCl2 (0.020 g, 0.080 mmol) in toluene (2 mL), nBuLi (0.10
mL, 0.160 mmol) was added at -15 °C giving an amber solution.
After 5 min, the reaction mixture was removed from the cooling
bath, warmed to room temperature, and to this solution, 1a (0.24
g, 4.1 mmol) in toluene (1 mL) was added resulting in a color
Acknowledgment. This paper is dedicated to Professor Peter
Paetzold (RWTH Aachen) on the occasion of his 75th birthday.
M.E.S. and A.S. acknowledge the EPSRC for funding. T.J.C. thanks
the Ontario Government for a Graduate Scholarship (OGS). G.C.L.-
J. thanks the Royal Society for a Wolfson Research Merit Award.
I.M. thanks the E.U. for a Marie Curie Chair and the Royal Society
for a Wolfson Research Merit Award. We thank Sven Schneider
(Technical University of Munich) for helpful discussion.
Supporting Information Available: Relevant NMR spectra
and kinetic plots. This material is available free of charge via
the Internet at http://pubs.acs.org.
JA909535A
(63) Gaines, D. F.; Schaeffer, R. J. Am. Chem. Soc. 1964, 86, 1505–1507.
9
J. AM. CHEM. SOC. VOL. 132, NO. 11, 2010 3841