Site selectivity and reversibility in the reactions of titanium hydrazides with Si-H, Si-X, C-X and H+ reagents: Ti=Nα 1,2-silane addition, Nβ alkylation, Nα protonation and σ-bond metathesis

Article abstract of DOI:10.1039/c2dt12359b

We report a combined experimental and computational comparative study of the reactions of the homologous titanium dialkyl- and diphenylhydrazido and imido compounds Cp*Ti{MeC(NⁱPr)₂}(NNR₂) (R = Me (1) or Ph (2)) and Cp*Ti{MeC(NⁱPr)₂}(NTol) (3) with silanes, halosilanes, alkyl halides and [Et₃NH][BPh ₄]. Compound 1 underwent reversible Si-H 1,2-addition to TiN _α with RSiH₃ (experimental ΔH ca. -17 kcal mol^-1), and irreversible addition with PhSiH₂X (X = Cl, Br). DFT found that the reaction products and certain intermediates were stabilised by β-NMe₂ coordination to titanium. The Ti-D bond in Cp*Ti{MeC(NⁱPr)₂}(D){N(NMe₂)SiD ₂Ph} underwent σ-bond metathesis with BuSiH₃ and H₂. Compound 1 reacted with RR′SiCl₂ at N _α to transfer both Cl atoms to Ti; 2 underwent a similar reaction. Compound 3 did not react with RSiH₃ or alkyl halides but formed unstable TiN_α 1,2-addition or N_α protonation products with PhSiH₂X or [Et₃NH][BPh ₄]. Compound 1 underwent exclusive alkylation at N_β with RCH₂X (R = H, Me or Ph; X = Br or I) whereas protonation using [Et₃NH][BPh₄] occurred at N_α. DFT studies found that in all cases electrophile addition to N_α (with or without NMe₂ chelation) was thermodynamically favoured compared to addition to N_β. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2dt12359b

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 2277
www.rsc.org/dalton
PAPER
Site selectivity and reversibility in the reactions of titanium hydrazides with
Si–H, Si–X, C–X and H+ reagents: Ti Na 1,2-silane addition, Nb alkylation,
Na protonation and r-bond metathesis†
Pei Jen Tiong,a Ainara Nova,‡b Andrew D. Schwarz,a Jonathan D. Selby,a Eric Clot*b and Philip Mountford*a
Received 6th December 2011, Accepted 9th December 2011
DOI: 10.1039/c2dt12359b
We report a combined experimental and computational comparative study of the reactions of the
homologous titanium dialkyl- and diphenylhydrazido and imido compounds Cp*Ti{MeC(NiPr)2}-
(NNR2) (R = Me (1) or Ph (2)) and Cp*Ti{MeC(NiPr)2}(NTol) (3) with silanes, halosilanes, alkyl
halides and [Et3NH][BPh4]. Compound 1 underwent reversible Si–H 1,2-addition to Ti Na with
RSiH3 (experimental DH ca. -17 kcal mol-1), and irreversible addition with PhSiH2X (X = Cl, Br). DFT
found that the reaction products and certain intermediates were stabilised by b-NMe2 coordination to
titanium. The Ti–D bond in Cp*Ti{MeC(NiPr)2}(D){N(NMe2)SiD2Ph} underwent s-bond metathesis
with BuSiH3 and H2. Compound 1 reacted with RR¢SiCl2 at Na to transfer both Cl atoms to Ti; 2
underwent a similar reaction. Compound 3 did not react with RSiH3 or alkyl halides but formed
unstable Ti Na 1,2-addition or Na protonation products with PhSiH2X or [Et3NH][BPh4]. Compound
1 underwent exclusive alkylation at Nb with RCH2X (R = H, Me or Ph; X = Br or I) whereas protonation
using [Et3NH][BPh4] occurred at Na. DFT studies found that in all cases electrophile addition to Na
(with or without NMe2 chelation) was thermodynamically favoured compared to addition to Nb.
with unsaturated substrates such as alkynes, phosphaalkynes,
allenes, CO2, isocyanates, isothiocyanates, nitriles and azides.6c,7
Introduction
Terminal transition metal hydrazido(2-) (L)M NNR2 (R = H,
alkyl or aryl) have been of ongoing interest for over 30 years.1
Initial studies focused on molybdenum and tungsten derivatives in
thecontextofthebiologicalconversionofdinitrogentoammonia.2
Both the Chatt type3 and Schrock type4 model systems proceed
mechanistically via a series of protonation and electron transfer
events. These involve (L)M NNH2 intermediates immediately
prior to the key Na–Nb bond cleavage step, in agreement with
computational studies.5
We and others have recently been developing group 4 terminal
hydrazido(2-) chemistry which has only recently been developed
in detail following early reports.6 These compounds show a rich
reaction chemistry of both the M Na and Na–Nb bonds. Reac-
tions at the M Na bond include cycloaddition, cycloaddition-
insertion, cycloaddition-elimination and NNR2 group transfer
Na–Nb bond cleavage reactions occur with alkynes, CO and
isonitriles as well as with other substrates.6b,7c,8
Despite this recent activity, the reactions of terminal group 4
M
NNR2 functional groups with Si–X (X = H or halogen) or H–
H bonds remain unexplored, although there is precedent for their
addition to metal–ligand multiple bonds in general.9 Likewise,
the chemistry of metal complexes with silanes and silicon-based
ligands has been a topic of much interest for many years.10 In
this contribution we report a combined experimental and DFT
comparative study of the reactions of titanium hydrazido and
imido compounds with silanes, halosilanes and H2. We also present
alkylation and protonation reactions which have been little studied
for group 4 hydrazides.11 Taken together, these studies provide
a detailed picture of the chemistry of the Ti NNR2 functional
group at the interface with N2 functionalisation chemistry. Part
of this work, which builds on our group’s interest in metal–ligand
multiple bond chemistry in general,12 has been communicated.13
aChemistry Research Laboratory, University of Oxford, Mansfield Road,
Oxford, UK, OX1 3TA. E-mail: philip.mountford@chem.ox.ac.uk
bInstitut Charles Gerhardt Montpellier, CNRS 5253, Universite´ Montpellier
2, cc 1501, Place Euge`ne Bataillon, F-34095, Montpellier Cedex 5, France.
E-mail: eric.clot@univ-montp2.fr
Results and discussion
† Electronic supplementary information (ESI) available: Experimental
details and characterising data; further details of the X-ray structures of 5,
6, 12, 15 and 20-I. CCDC reference numbers 802264–802265 and 849349–
849356. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c2dt12359b
‡ Current address: Institute of Chemical Research of Catalonia (ICIQ),
Ave Pa¨ısos Catalans, 16, 43007 Tarragona, Spain.
In recent reports,7b,7d,12d,12j,14 we showed that the cyclopentadienyl-
amidinate ligand can support a homologous series of tert-
butylimido, arylimido, dimethylhydrazido and diphenylhydrazido
complexes. Differing only in the nature of the Na substituent,
each of these Ti NR or Ti NNR2 linkages can give rise to
This journal is
The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 2277–2288 | 2277
©

View Article Online
different reaction outcomes with unsaturated substrates (cycload-
dition, cycloaddition-insertion, cycloaddition-elimination). In the
current studies we focus on Cp*Ti{MeC(NiPr)2}(NNR2) [R = Me
(1) or Ph (2)]7b and Cp*Ti{MeC(NiPr)2}(NTol) (3).12j The Nb atom
of 1 is strongly pyramidalised, whereas that in 2 is planar due to
conjugation of the Nb lone pair with the phenyl groups. Compound
3 has no NR2 substituent and allows comparison of Ti Na bond
reactivity without b-NR2 group contributions.
Fig.
1
Displacement ellipsoid plots (20% probability) of
Reactions with silanes and dihydrogen
Cp*Ti{MeC(NiPr)2}(H){N(NMe2)SiH2Ph} (4). C-bound
H atoms
We focus first on the reactions of 1–3 with silanes and H2.
Compound 1 reacts with primary aryl silanes and BuSiH3
according to eqn (1) to form the hydride-hydrazide(1-) complexes
Cp*Ti{MeC(NiPr)2}(H){N(NMe2)SiH2R} (R = Ph (4), ArF (5),
ArOMe (6) or Bu (7); ArF = 3,5-C6H3(CF3)2, ArOMe = 4-C6H4OMe)
via net Si–H 1,2-addition to the Ti Na bond of 1. No reaction
occurred between 1 and either H2 or Et3SiH, or between 2 or
3 and any of these primary silanes. Equilibrium mixtures were
formed for 1 and Ph2SiH2, from which no product could be
isolated. Compounds 4–6 were isolated as yellow crystals in 58–
76% yield whereas 7 is a very soluble waxy solid which could
only be obtained in 22% yield. All four tend to lose RSiH3
in vacuo, particularly in the case of 7. When followed on the
NMR tube scale in C6D6 the reactions (1 : 1 stoichiometry) were
essentially quantitative for ArSiH3, whereas with BuSiH3 the
conversion of 1 to 7 was only ca. 85% at room temperature
and an equilibrium mixture was formed. To confirm the spectro-
scopic assignments for Ti–H and Si–H the deuterated compound
Cp*Ti{MeC(NiPr)2}(D){N(NMe2)SiD2Ph} (4-d3) was prepared
from 1 and PhSiD3.
omitted for clarity, and other H atoms drawn as spheres of an arbitrary
radius. Compounds 5 and 6 follows analogous numbering schemes (see
the ESI†).
are given in Table 1. These confirm the 1,2-addition of Si–H to
Ti Na and reveal 5-legged piano stool geometries; the distances
and angles are within the expected general ranges15a and there are
no significant differences between the three structures. The N(1),
N(3), N(4) and H(1) atoms are approximately co-planar. Ti(1) lies
˚
˚
ca. 0.65 A above and N(2) ca. 1.35 A below this plane. The new
2
N(NMe2)SiH2Ar ligands are k -coordinated as is usually found
for titanium hydrazido(1-) complexes,7j,11a,16 except for especially
crowded systems16f or those with b-NPh2 substituents.16h,17 The
Ti(1)–N(1) and N(1)–N(2) distances of ca. 1.95–1.96 and 1.44–
˚
1.46 A are significantly lengthened compared to those of 1.723(3)
and 1.386(2) A, respectively, in 1. These values lie at the longer end
of the ranges previously found for k -coordinated hydrazido(1-)
˚
2
˚
complexes (Ti–Na: 1.832–2.005 (mean ca. 1.90) A; Na–Nb: 1.236–
1.458 (mean ca. 1.41) A; 14 examples15a) and are consistent with
˚
the disruption of the Ti–Na triple bond and a change from sp
hybridisation for N(1) in 1 to sp2 in 4–6. The Ti(1)–N(2) distances
˚
(av. Ti(1)–N(2) = ca. 2.21 A) are only slightly longer than those for
Table 1 Selected bond lengths (A) and angles (◦) for
˚
(1)
Cp*Ti{MeC(NiPr)2}(H){N(NMe2)SiH2R} (R = Ph (4), ArF (5) or
ArOMe (6)). Cpcent refers to the C5Me5 ring carbon centroid
4
5
6
1
The H, 13C and 29Si NMR spectra of 4–7 are consistent with
Ti(1)–Cpcent
2.092
2.092
2.093
Ti(1)–N(1)
Ti(1)–N(2)
Ti(1)–N(3)
Ti(1)–N(4)
Ti(1)–H(1)
N(1)–N(2)
N(1)–Si(1)
Si(1)–H(2)
1.963(3)
2.212(3)
2.197(3)
2.190(3)
1.64(4)
1.460(4)
1.723(3)
1.38(4)
1.41(3)
114.1
152.0
116.0
111.5
96.2
1.963(2)
2.215(2)
2.180(3)
2.178(2)
1.56(4)
1.441(3)
1.722(3)
1.42(4)
1.40(4)
114.8
152.2
115.6
111.9
97.0
1.952(1)
2.201(1)
2.190(1)
2.194(1)
1.63(2)
1.445(1)
1.735(1)
1.40(2)
1.41(2)
114.4
152.3
116.4
111.9
94.3
the structures shown in eqn (1) which have been confirmed by
X-ray crystallography (vide infra). The Ti–H NMR resonances
appear between 5.05 and 5.15 ppm, and those for the inequivalent
Si–H atoms between ca. 4.8 and 5.5 ppm. A single resonance
between -42.1 and -44.3 ppm was observed in the 29Si spectra
of 4–6, and at -36.6 ppm for 7. The n(Si–H) bands (symmetric
and antisymmetric combinations) lie between 2094 and 2195 cm-1
in the IR spectra, with n(Ti–H) appearing between 1558 and
Si(1)–H(3)
Cpcent–Ti(1)–N(1)
Cpcent–Ti(1)–N(2)
Cpcent–Ti(1)–N(3)
Cpcent–Ti(1)–N(4)
Cpcent–Ti(1)–H(1)
N(1)–Ti(1)–N(2)
Ti(1)–N(1)–Si(1)
N(1)–Si(1)–C(3)
1
1565 cm-1. These bands, and the corresponding H NMR reso-
nances, were absent in 4-d3. Terminal titanium hydride compounds
are comparatively uncommon and only a relatively small number
have been structurally authenticated.9c,9d,15
The solid state structure of 4 is shown in Fig. 1, those of 5
and 6 are shown in the ESI,† and selected distances and angles
40.38(10)
155.46(17)
114.02(15)
39.77(10)
152.92(15)
112.40(13)
40.15(4)
155.54(6)
113.71(6)
2278 | Dalton Trans., 2012, 41, 2277–2288
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
˚
the amidinate supporting ligands (av. Ti–N(3,4) = ca. 2.19 A) and
within the usual ranges.
process is fast compared to the s-bond metathesis process since
H appears in the Ti–D and Si–D sites of 4-d3 at the same
apparent rate at room temperature. No Si–H/Si–D exchange
was observed between PhSiD3 and the chloride analogue of
4, namely Cp*Ti{MeC(NiPr)2}(Cl){N(NMe2)SiH2Ph} (8, vide
infra), establishing the pivotal role of Ti–H in the exchange process
as expected. Chirik et al. have reported related processes for a
titanocene imido complex.9l
As mentioned, all four compounds 4–7 have a tendency to
undergo 1,2-elimination of RSiH3 to reform 1. In the cases of 4, 6
and 7 this process is clean and reversible up to 377 K (remarkably)
in toluene-d8 allowing the direct evaluation of the thermodynamic
parameters for eqn (1) via the van’t Hoff analyses summarised in
Fig. 2. In the case of 5 an unknown decomposition process set in
above ca. 313 K and a direct measurement of DH and DS in this
way was not possible.
(2)
These Si–H addition reactions of silanes with 4–7, and
the accompanying s-bond metathesis reactions, are unique in
transition metal dialkylhydrazide chemistry, although they can
be compared with other aspects of titanium hydride/silane
chemistry10c,18 and PhSiH3 addition to the diazoalkane lig-
2
and in Cp*2Ti{k -NNC(H)Tol} giving Cp*2Ti(H){N(NC(H)Tol)-
SiH2Ph}.9c Although titanium amides can undergo Ti–N/Si–H
exchange with silanes,10b,10f ,19 no evidence for hydrazide-to-silicon
transfer chemistry was found in our systems.
Fig. 2 van’t Hoff plot for the reaction Cp*Ti{MeC(NiPr)2}(NNMe2) +
RSiH3 = Cp*Ti{MeC(NiPr)2}(H){N(NMe2)SiH2R} (R = Ph (4), ArOMe
(6) or Bu (7) in toluene-d8). For R = Ph (diamonds): DH = -17.1(2) kcal
mol-1; DS = -43(1) cal mol-1 K-1; DG298 = -4.3(5) kcal mol-1 (R2 = 0.999);
for R = ArOMe (circles): DH = -16.9(1) kcal mol-1; DS = -43(1) cal mol-1
K-1; DG298 = -4.1(3) kcal mol-1 (R2 = 0.999); for R = Bu (squares): DH =
-16.6(2) kcal mol-1; DS = -45(1) cal mol-1 K-1; DG298 = -3.2(5) kcal mol-1
(R2 = 0.999).
Reactions with halosilanes
The reactions of Cp*Ti{MeC(NiPr)2}(NNMe2) (1) with chloro-
and dichlorosilanes are summarised in Scheme 1.
The DH values of -17.1(2) and -16.9(1) kcal mol-1 are identical
within error for 4 and 6. The DH of -16.6(2) kcal mol-1 for 7
was also equivalent within error to that of the two aryl systems.
However, from the measured DG298 values (-4.3(5), -4.1(3) and
-3.2(5) kcal mol-1 for 4, 6 and 7), and qualitative observations
it is clear that 1,2-addition of BuSiH3 is less favourable than
for the aryl silanes. The corresponding DH (-17.2(2) kcal mol-1)
and DG298 (-4.4(5) kcal mol-1) values for the formation of 5 were
determined indirectly via an equilibrium competition experiment
between PhSiH3 and ArFSiH3 for 1, assuming DS for 1,2-addition
is -43(1) cal mol-1 K-1 in each case (cf Fig. 2 caption).
Addition of BuSiH3 (1 equiv.) to a solution of 4-d3 in C6D6
or C6H6 immediately formed an equilibrium mixture of BuSiH3,
4-d3, 7 and PhSiD3 (ratio of 4-d3 : 7 = ca. 3 : 1 in accordance with
1
2
the DG298 values). In addition, after 5 min, the H and H NMR
spectra also showed incorporation of deuterium into the Ti–H
and/or Si–H sites of 7 and BuSiH3. Likewise, hydrogen appeared
in the Ti–D and/or Si–D sites of 4-d3 and PhSiD3. After 4–5 h a
statistical distribution of H and D into all Si–H/D and Ti–H/D
sites was reached.
The implied s-bond metathesis between Ti–H and Si–D (and
Ti–D and Si–H) was also found with H2 and 4-d3 as shown in eqn
(2). Although H2 does not form a stable 1,2-addition product
with 1, after 4 days at room temperature, 30% incorporation
of hydrogen into the initial Ti–D and Si–D sites of 4-d3 was
observed. This reaction occurs via s-bond metathesis between Ti–
D and H–H followed by 1,2-elimination of PhSiHD2. Subsequent
1,2-addition scrambles the Ti–H and Si–D groups. This latter
Scheme 1 Reactions of Cp*Ti{MeC(NiPr)2}(NNMe2) (1) with mono-
and dichlorosilanes.
Reaction with PhSiH2Cl gave Cp*Ti{MeC(NiPr)2}(Cl)-
{N(NMe2)SiH2Ph} (8) as an orange solid. Com-
pound
8
is the chloride analogue of the hydride
Cp*Ti{MeC(NiPr)2}(H){N(NMe2)SiH2Ph} (4) and its 1H
NMR and IR spectra are similar to those of 4 except for
the absence of the signals attributed to Ti–H. The 29Si NMR
resonance for 8 (-45.6 ppm) is very similar to that of 4
(-42.3 ppm). No evidence was found for an alternative product
This journal is
The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 2277–2288 | 2279
©

View Article Online
Table 2 Selected bond lengths (A) and angles (◦) for
Table 3 Selected bond lengths (A) and angles (◦) for
Cp*Ti{N(NMe2)SiH(Ph)N(iPr)CMeNiPr}Cl2 (11) and Cp*Ti{N(NMe2)-
SiMe2N(iPr)CMeNiPr}Cl2 (12). Cpcent refers to the C5Me5 ring carbon
centroid
˚
˚
Cp*Ti{MeC(NiPr)2}(Cl){N(NMe2)SiH2Ph} (8). Cpcent refers to the
C5Me5 ring carbon centroid
Ti(1)–Cpcent
Ti(1)–N(2)
Ti(1)–N(4)
N(1)–N(2)
Cpcent–Ti(1)–N(1)
Cpcent–Ti(1)–N(3)
Cpcent–Ti(1)–Cl(1)
Ti(1)–N(1)–Si(1)
2.136
Ti(1)–N(1)
Ti(1)–N(3)
Ti(1)–Cl(1)
N(1)–Si(1)
Cpcent–Ti(1)–N(2)
Cpcent–Ti(1)–N(4)
N(1)–Ti(1)–N(2)
N(1)–Si(1)–C(3)
1.939(3)
2.145(2)
2.4463(9)
1.737(3)
153.1
106.3
39.72(9)
116.44(13)
2.223(3)
2.234(2)
1.439(3)
114.3
112.9
102.1
11
12
Ti(1)–Cpcent
Ti(1)–N(1)
Ti(1)–N(2)
Ti(1)–Cl(1)
Ti(1)–Cl(2)
N(1)–N(2)
N(1)–Si(1)
N(3)–C(9/5)
2.066
2.083
1.929(2)
2.109(2)
2.3099(8)
2.3693(9)
1.406(3)
1.752(2)
1.392(4)
1.288(4)
115.7
148.8
108.8
108.7
40.44(9)
151.10(14)
116.88(11)
115.4(2)
1.923(3)
2.112(3)
2.3167(12)
2.3579(12)
1.414(4)
1.750(3)
1.409(4)
1.273(5)
118.5
127.5
112.7
108.8
40.68(12)
153.46(18)
114.38(14)
116.2(3)
156.97(15)
Cp*Ti{MeC(NiPr)2}(H){N(NMe2)SiH(Cl)Ph} which would
be the product of Si–H activation. Assuming thermodynamic
control, this indicates that Si–Cl addition to Ti NNMe2 is
preferred to Si–H addition as has been found for group 5 and
6 metal imido compounds.9g–i Unlike the situation for 4, the
1,2-addition of Si–Cl to Ti NNMe2 is irreversible. Compound
8 is reasonably stable in C6D6 solution, but after 24 h at room
temperature, a number of unknown decomposition products
were observed. Gade et al. have observed an apparently related
1,2-addition of Me3SiN3 to a Ti NNMe2 linkage forming a
titanium hydrazide(1-) compound.7j
N(4)–C(9/5)
Cpcent–Ti(1)–N(1)
Cpcent–Ti(1)–N(2)
Cpcent–Ti(1)–Cl(1)
Cpcent–Ti(1)–Cl(2)
N(1)–Ti(1)–N(2)
Ti(1)–N(1)–Si(1)
N(1)–Si(1)–N(3)
N(3)–Si(9/5)–N(4)
gave exclusively 8 and PhSiD3 with no incorporation of deuterium
into the Si–H sites of 8.
The solid state structure of 8 is shown in Fig. 3 and selected
distances and angles are listed in Table 2. Its structure is very
similar to those of 4–6 (Fig. 1) and there are few significant
differences in the key geometrical features. The Ti–Na distance
in 8 is shorter within error compared to those in its hydride
counterparts, but still within the expected ranges;15a this may be
attributed to the higher electronegativity of Cl compared to H.
The Ti(1)–N(2) and Na–Nb distances are the same within error to
those in the hydride analogues. There is a greater asymmetry in
(3)
Reaction of 1 in toluene-d8 with the disubstituted chlorosilanes
Ph(Me)SiHCl or Me2SiHCl also led to 1,2-addition products
(Scheme 1) as judged by H, 13C and 29Si NMR spectroscopy.
˚
the Ti–N(3,4) distances in 8 (D(Ti–N(3,4) = 0.089(3) A) compared
1
to those for 4–6 but the origin of this difference is not clear.
However, neither 9 nor 10 could be isolated on the preparative
scale, unknown mixtures being formed in each instance. Reaction
of 1 with Ph2SiCl2 or Me3SiCl in C6D6 immediately gave a mixture
of unknown products. As an indication of the lower stability of
these more sterically encumbered products we found that reaction
of 9 with PhSiH2Cl in C6D6 cleanly formed 8 and Ph(Me)SiHCl.
However, Ph(Me)SiHCl was nonetheless still able to displace
PhSiH3 from 4 forming 9.
In contrast to the unstable products formed with the
bulkier monochlorosilanes, reaction of 1 with the dichlorosi-
lanes PhSiHCl2 or Me2SiCl2 gave the stable hydrazide(1-)
products Cp*Ti{N(NMe2)SiH(Ph)N(iPr)CMeNiPr}Cl2 (11) and
Cp*Ti{N(NMe2)SiMe2N(iPr)CMeNiPr}Cl2 (12) (Scheme 1) in
which both chlorine atoms of the silane had been transferred
to titanium with concomitant transfer of the amidinate ligand
to silicon. When followed on the NMR tube scale in C6D6 11
and 12 were the only products formed. However, on scale up, the
isolated yields were only 40–50% owing to their high solubility.
Both compounds have been crystallographically characterised.
The molecular structure of 11 is shown in Fig. 4 and that of 12 is
given in the ESI;† selected distances and angles are presented in
Table 3.
Fig.
3 Displacement ellipsoid plot (20% probability) of
Cp*Ti{MeC(NiPr)2}(Cl){N(NMe2)SiH2Ph} (8). C-bound
H atoms
omitted for clarity, and Si-bound H atoms drawn as spheres of an
arbitrary radius.
As mentioned, formation of 8 was Si–Cl bond selective and
irreversible (unlike formation of 4 with PhSiH3), indicating that Si–
H addition to Ti NNMe2 is less favourable than Si–Cl addition.
Consistent with this, treatment of 4 with PhSiH2Cl in C6D6 cleanly
formed 8 and PhSiH3 immediately and quantitatively (eqn (3)). In
principle,20 this reaction could proceed via direct hydride exchange
between 4 and PhSiH2Cl rather than silane elimination from 4
(forming 1) and then 1,2-addition of PhSiH2Cl. To confirm this
latter mechanism, we found that reaction of 4-d3 with PhSiH2Cl
The solid state structures of 11 and 12 are very similar. Each
features a four-legged piano stool geometry, but with Ti(1) and
N(2) lying above and below the plane formed by N(1) and
the two Cl atoms. The Ti(1)–N(1) distances are shorter but
2280 | Dalton Trans., 2012, 41, 2277–2288
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
[Cp*Ti{MeC(NiPr)2}{N(NMe2)H)]+ (23+, vide infra) which has
been isolated as its [BPh4]- salt and structurally characterised. Di-
rect evidence for [13-Br]+ was found using positive ion electrospray
mass spectrometry for a solution of 13 generated in situ in THF.22
As a probe of the energetic preferences of Ti NNMe2
for Si–H, Si–Cl and Si–Br addition further NMR tube
scale competition experiments were carried out. Reac-
tion of Cp*Ti{MeC(NiPr)2}(Cl){N(NMe2)SiH(Me)Ph} (9) with
PhSiH2Br in C6D6 cleanly formed 13 and Ph(Me)SiHCl. Likewise,
reaction of 4 with PhSiH2Br also immediately gave 13 along with
PhSiH3. At first sight, these results parallel those found with
PhSiH2Cl and 4 or 9. However, in contrast to the reaction between
4-d3 and PhSiH2Cl (eqn (3)), H and H NMR monitoring of
that with PhSiH2Br showed not only the formation of PhSiD3
and Cp*Ti{MeC(NiPr)2}(Br){N(NMe2)SiH2Ph} (13) but also
PhSiH2D and Cp*Ti{MeC(NiPr)2}(Br){N(NMe2)SiD2Ph}(13-
d2). The ratio of 13 to 13-d2 was 5 : 3 showing that silane
elimination from 4 followed by PhSiH2Br addition is nonetheless
kinetically favoured compared to direct Ti–D/Si–Br exchange via
nucleophilic substitution.23
Fig.
4 Displacement ellipsoid plot (20% probability) of
Cp*Ti{N(NMe2)SiH(Ph)N(iPr)CMeNiPr}Cl2 (11). C-bound H atoms
omitted for clarity, and H(1) is drawn as a sphere of an arbitrary radius.
Compound 12 follows an analogous numbering scheme (see the ESI†).
comparable with error to that in 8, whereas the Ti–Cl, Ti(1)–N(2)
distances are significantly shorter as expected based on the reduced
coordination number and decreased donor ability of chloride
compared to MeC(NiPr)2. The N(1)–N(2) distances in 11 and
1
2
˚
12 (1.406(3) and 1.414(4) A) are significantly shorter than in 8
˚
˚
(1.439(3) A) and 4–6 (1.441(3)–1.460(4) A) indicative of differing
degrees of reduction of this bond in the different complexes.
Compounds 11 and 12 are fluxional on the NMR timescale
in a process that equivalences the two isopropyl groups of the –
N(iPr)CMeNiPr moiety but not the inequivalent –NMe2 methyl
groups in either case or the SiMe2 methyl groups in 12. The
fluxional process is attributed to rapid 1,3-silicon migration
between the amidinate nitrogen atoms. This type of mechanism
is well-established for silylated amidines.21 Cooling an NMR
solution of 12 to 223 K freezes out this process, whereas for 11,
which possesses a less sterically crowded silicon, broadening of the
isopropyl group resonances was observed but the slow exchange
limit could not be reached.
Cp*Ti{MeC(NiPr)2}(NNMe2) (1) also reacts with PhSiH2Br
in toluene-d8 or THF-d8 to form Cp*Ti{MeC(NiPr)2}(Br)-
{N(NMe2)SiH2Ph} (13) as a single fluxional product in quan-
titative yield (eqn (4)). Compound 13 could not be isolated on
the preparative scale and solutions show extensive decomposition
after only a few hours. At 193 K in toluene-d8 or THF-d8 the 1H,
13C and 29Si NMR data were consistent with the geometry depicted
in eqn (4) (inequivalent, mutually-coupled SiH2 hydrogen atoms
and inequivalent NMe2 methyl groups; a single 29Si resonance at
-41.7 ppm). Above 220 K the SiH2 and NMe2 resonances under-
went pairwise coalescence consistent with the net fluxional process
depicted in eqn (4) (interchange of diastereomers 13 and 13¢).
At first sight this fluxional process could proceed via elim-
ination of PhSiH2Br forming 1 as an intermediate (as found
for 4–7). However, addition of PhSiH2Br (1 equiv.) to a so-
lution of 13 at room temperature gave no broadening of the
free silane resonances whereas those for 13 were indicative
of the fast exchange limit. Thus the process interconverting
13 and 13¢ does not involve release of PhSiH2Br. Instead
we propose a process involving transient Br- ion elimina-
tion via [Cp*Ti{MeC(NiPr)2}{N(NMe2)SiH2Ph}]+ (denoted [13-
Br]+) through which chemical exchange of the Si-H and N-
Me groups can occur. The cation [13-Br]+ is the analogue of
The reactions of the diphenylhydrazido complex
Cp*Ti{MeC(NiPr)2}(NNPh2) (2) are summarised in eqn
(5). Although 2 does not react with RSiH3 (R = aryl, Bu),
it does undergo a relatively slow reaction with 2 equiv. of
the monochlorosilanes PhSiH2Cl or Ph(Me)SiHCl forming
dichloride-hydrazide(1-) complexes Cp*Ti{N(NPh2)SiH2Ph}Cl2
(14) and Cp*Ti{N(NPh2)SiH(Me)Ph}Cl2 (15). The products
are reminiscent of 11 and 12 formed with dichlorosilanes,
but in this case the reactions have involved complete ejection
of the amidinate ligand as the silylated amidines 16 and 17.
Reaction of 2 with 1 equiv. of either monochlorosilane gave
a mixture of unreacted 2 and the products shown in eqn (5).
As expected, 16 and 17 are fluxional on the NMR timescale
and their room temperature spectra show equivalent isopropyl
resonances, consistent with fast 1,3-shifts of the SiH(R)Ph groups
between the nitrogen atoms. Compound 15 was structurally
characterised and the molecular structure and selected distances
and angles are given in the ESI.† The data are analogous to
those for Cp*Ti{N(NMe2)SiH(Ph)N(iPr)CMeNiPr}Cl2 (11) and
Cp*Ti{N(NMe2)SiMe2N(iPr)CMeNiPr}Cl2 (12). The slower
rates of reaction and ready loss of the amidinate ligand with 2 are
attributed to the delocalised nature of the Nb lone pair and the
higher steric demands of the b-NPh2 group (leading to a more
sterically crowded metal centre), respectively, compared to 1.
As an experimental probe of the importance of the
intramolecular NMe2 chelation in the formation of the
products 8–10 and 13 we carried out the reactions of
Cp*Ti{MeC(NiPr)2}(NTol) (3) with PhSiH2Cl and PhSiH2Br
1
as shown in eqn (6). When followed by H NMR in toluene-
d8 quantitative conversion to the silylamido-chloride complexes
Cp*Ti{MeC(NiPr)2}(X){N(Tol)SiH2Ph} (X = Cl (18) or Br (19))
This journal is
The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 2277–2288 | 2281
©

View Article Online
(4)
(5)
2
was observed. The NMR spectra are consistent with the proposed
structures, featuring inequivalent amidinate isopropyl groups (4
different methyl group environments), a pair of mutually coupled
doublets for the diastereotopic SiH2 groups, and a 29Si resonance
at ca. -36 ppm. The reactions of group 5 and 6 imido compounds
with halosilanes to form silylamido products has been studied
in detail previously.9g–i The reverse process, namely formation of
imido compounds by halosilane elimination, is a well-established
synthetic strategy.12a,12c,24
Both 18 and 19 are unstable in solution, fully decomposing to
unknown products after 16 h, and they could not be isolated
on the preparative scale. A DFT model of 18 (vide infra),
CpTi{MeC(NMe)2}(Cl){N(Tol)SiH2Me} (m8Cl), found the ex-
pected four legged piano stool geometry and a sp2 hybridised Na
for the new silylamido ligand. No b-agostic Si–H ◊ ◊ ◊ Ti interactions
were found in this 16 valence electron compound. These have often
been reported for silylamido ligands and the early transition and
rare earth metals.9g–i,20,25 Given the unstable nature of 18 and 19,
no further reactions of 3 with other halosilanes were carried out.
H addition across the Ti–NN bond of Cp*2Ti{k -NNC(H)Tol}
¯
(forming alkylidene hydrazide(1-)-hydride products reminiscent
of 4–7),9c but without experimental or other support.
Nothing is known regarding the mechanism of Si–X (X = H,
halogen) bond addition to transition metal hydrazide complexes.
In particular it is not clear what mechanistic or other role(s) b-
NMe2 chelation plays in the formation of the silane or halo-silane
1,2-addition products described for 1 above. The absence of any
reaction between 3 and PhSiH3, and the low stability of the product
(18) formed with PhSiH2Cl suggest an important thermodynamic
contribution.
Attempts to gain experimental mechanistic insight for the
reaction of 1 with PhSiH3 or PhSiD3 through KIE experi-
ments were unsuccessful, the reactions proceeding to completion
within seconds even at 203 K in toluene-d8. Likewise, the
rapid equilibrium between 4 and free silane (even at 203 K
on the chemical timescale) prevented the indirect measurement
of any KIE by analysis of the 4/4-d3 product distribution
formed in a competition reaction between 1 and PhSiH3 and
PhSiD3. To determine the reaction mechanism(s) and other
factors underpinning this chemistry we carried out DFT(M06)
calculations26 using CpTi{MeC(NMe)2}(NNMe2) (m1, model for
1) and CpTi{MeC(NMe)2}(NPh) (m7, a model for 3) and the
silanes MeSiH3 and MeSiH2Cl. The main results with m1 are
summarised in Fig. 5 where, for each silane, we consider Si–X
(6)
2
addition to either a terminal (“Path 1”) or a k -coordinated (“Path
2”) hydrazide ligand.
DFT study of the mechanism of Si–H and Si–Cl addition
Si–H bond addition. Overall, the calculated thermodynamic
parameters for the reaction m1 + MeSiH3 → m2 (a model
for the real complexes 4–7) are in good agreement with the
experimental results (DFT: DH = -22.5 kcal mol-1; DS = -53
cal mol-1 K-1; experiment: DH = -16.6(2) to -17.2(2) kcal mol-1;
DS = -43(2) to -45(2) cal mol-1 K-1). The calculated values for
the full experimental system 1 + PhSiH3 → 4 showed even better
agreement (DFT: DH = -18.9 kcal mol-1; DS = -49 cal mol-1
K-1; experiment: DH = -17.1(2) kcal mol-1; DS = -43(2) cal
mol-1 K-1). Given the good agreement with the small model m1,
this was generally selected for further studies so as to improve
computational efficiency.
Kinetic or/and computational studies of the mechanism of Si–H
and Si–Cl bond addition to the M E multiple bond of imido,
sulfido, and oxo complexes have been reported by the groups
of Tilley,9a Bergman and Andersen,9d and Chirik,9e as well as by
Nikonov and Mountford.9h Primary KIEs (kinetic isotope effects)
greater than 1 for Si–H addition to titanocene Ti S or Ti
O
bonds suggested a concerted addition.9d,9e For Tilley’s highly bent
tantalum imido system,9a or Nikonov and Mountford’s 18 valence
electron group 5 imides,9h an inverse KIE or other kinetic data
suggested a stepwise process whereby R3SiX adds first to the
Na atom forming an intermediate with a pentacoordinate silicon
centre, followed by Si–H or Si–Cl bond addition to M Na (1,2-
addition) or the metal centre (oxidative addition). Bergman and
Andersen proposed a similar stepwise mechanism for silane Si–
DGsol for the formation of the 2,1-addition product
CpTi{MeC(NMe)2}(SiH2Me){N(NMe2)H}
(m2¢)
was
+6.5 kcal mol-1 (cf -5.3 kcal mol-1 for forming m2) in
2282 | Dalton Trans., 2012, 41, 2277–2288
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
Fig. 5 Energy profiles for the reaction of CpTi{MeC(NMe)2}(NNMe2) (m1) with MeSiH3 (right) and MeSiH2Cl (left). All energies are DGsol in kcal
mol-1 at 298 K. [Ti] = CpTi{MeC(NMe)2}.
agreement with this isomer not being observed experimentally
(note that C–H activation at Ti NR bonds invariably gives
2,1-addition products27). DH for H2 addition to the Ti Na bond
of either m1 or the full experimental system 1 was calculated to
be energetically less favourable relative to Si–H 1,2-addition by
ca. 9 kcal mol-1, whereas the activation barrier was only 20.0 kcal
mol-1 via Path 1 (Fig. 5). This shows that the lack of observation
of H2 addition to 1 is thermodynamic rather than kinetic in
origin.28
m2 with coordination of Nb approximately trans to the Cp ring.
Whereas the new Na–Si bond is more developed in tsm5-6 than
in tsm3-4 (N ◊ ◊ ◊ Si = 2.099 and 2.119 A, respectively), hydrogen
˚
transfer to Ti is not as advanced as illustrated by Si ◊ ◊ ◊ H = 1.594
˚
˚
˚
˚
A (cf 1.491 A in MeSiH3) and Ti ◊ ◊ ◊ H = 1.878 A (cf 1.701 A in
m6). Thus Path 2 resembles slightly more a nucleophilic attack
of Na on silicon, followed by hydrogen transfer to titanium, as
proposed previously by Tilley for Si–H addition to a highly bent
tantalum imido complex.9a Note also that k -hydrazido moieties
2
As shown in Fig. 5, addition of MeSiH3 is predicted to proceed
via Path 1 (i.e. without initial b-NMe2 coordination). The first-
formed adduct m3 features the necessary interactions between Ti
can sometimes be trapped using the strong Lewis acid B(C6F5)3
which coordinates strongly to Na.7c,29 In both tsm3-4 and tsm5-6,
the silicon is 5-coordinate with a trigonal bipyramidal geometry.
Na occupies one of the axial positions and the migrating H atom
and Me group lie in the equatorial plane.
˚
˚
and H (Ti ◊ ◊ ◊ H = 1.990 A), and Si and Na (Si ◊ ◊ ◊ Na = 2.854 A), to
˚
proceed to the 1,2-addition product m4 (Ti–H = 1.707 A; Na–Si =
˚
1.775 A) via the transition state (TS) tsm3-4. This lies 16.3 kcal
The additional stabilisation gained upon coordination of b-
NMe2 (m4 → m2, DG = -13.2 kcal mol-1) moves the system
thermodynamically towards the Si–H activation product. This
stabilising effect is not available to imido complexes which explains
why Cp*Ti{MeC(NiPr)2}(NTol) (3) does not react with silanes.
Calculations on CpTi{MeC(NMe)2}(NPh) (m7) and MeSiH3
show that the energy barrier for the 1,2 addition is accessible
mol-1 above the separated reactants and is thus readily accessible,
consistent with the rapid reaction of 1 with silanes. The concerted
nature of this TS is illustrated by the almost identical values for the
˚
Ti ◊ ◊ ◊ H and Si ◊ ◊ ◊ H distances (1.765 and 1.760 A). Intermediate
m4 is itself not stable, but chelation of the b-NMe2 forms m2 with
an overall reaction DGsol of -5.3 kcal mol-1.
#
Path 2 in Fig. 5 is also accessible but is not kinetically preferred
in comparison with Path 1. Chelation of the b-NMe2 of m1 is facile
and forms m5 which lies only +1.8 kcal mol-1 above m1. Bending
takes place along a direction roughly parallel to the N ◊ ◊ ◊ N vector
of the amidinate ligand and increases the nucleophilicity of Na
as shown by a decrease in natural charge from -0.45 to -0.53.
(DGsol = 19.4 kcal mol-1) and proceeds through a similar TS (tsm7-
˚
˚
8; Ti ◊ ◊ ◊ H = 1.802 A; Si ◊ ◊ ◊ H = 1.699 A) to tsm3-4. However,
although formation of CpTi{MeC(NMe)2}(H){N(Ph)SiH2Me}
(m8) is 4.0 kcal mol-1 more favourable than formation of m4,
the reaction is still endothermic by +3.9 kcal mol-1.
˚
Si–H addition to Ti Na forms m6 (Na–Si = 1.758 A) via the
Si–Cl bond addition. As summarised in Fig. 5, Si–Cl addition
to the Ti NNMe2 bond of m1 can also, in principle, proceed
via either direct addition (Path 1) or via the b-NMe2 chelated
TS tsm5-6 with an activation barrier of 17.8 kcal mol-1 relative
to m5. Isomerisation via m4 finally gives the more stable isomer
This journal is
The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 2277–2288 | 2283
©

View Article Online
isomer m5. The final product m2Cl (model for 8) is 16.5 kcal
mol-1 more stable with respect to the separated reactants than m2
(overall DGsol = -21.8 kcal mol-1), consistent with experimental
observations of the irreversible addition of PhSiH2Cl to 1. The
high barrier to elimination of MeSiH2Cl from m2Cl via tsm1-4Cl
(38.9 kcal mol-1, cf 21.6 kcal mol-1 for elimination of MeSiH3 from
m2) is consistent with the absence of detectable exchange between
8 and free PhSiH2Cl on the NMR timescale.
studied for group 6 compounds in particular in the context of
understanding the key Na–Nb bond cleavage steps in the Chatt
and Schrock cycles. Reaction of terminal hydrazides with alkyl
halides or triflates invariably occurs at Nb and three examples of
the so-formed hydrazidium complexes [(L)M(NNMe3)]+ have been
structurally characterised for titanium11a,11c and molybdenum.30
The situation regarding protonation is less clear-cut. Both Na- and
Nb-protonated complexes have been structurally authenticated for
group 6.31 For molybdenum, Tuczek et al. have shown through
stopped-flow experiments and DFT that protonation at Nb can be
kinetically preferred but that a rapid 1,2-proton shift can occur to
form the Na-protonated counterpart which is thermodynamically
more stable.32 DFT studies by Reiher et al.,33 and by Guha and
Phukan,34 support this view.
The energy of the TS tsm1-4Cl (17.1 kcal mol-1) is similar to
that of tsm3-4 (16.3 kcal mol-1) for Si–H addition. Comparison of
˚
˚
the Si ◊ ◊ ◊ Cl (2.312 A) and Ti ◊ ◊ ◊ Cl (2.586 A) distances in tsm1-
˚
4Cl with those of MeSiH2Cl (Si–Cl = 2.100 A) and the first-
formed product m4Cl (Ti–Cl = 2.373 A), respectively, confirm the
˚
concerted nature of the process. b-NMe2 chelation (m4Cl → m2Cl)
brings a further stabilisation of -11.8 kcal mol-1 (cf -13.2 kcal
mol-1 for m4 → m2). Unlike m4, however, the chloride analogue
m4Cl is predicted to be stable even without b-NMe2 coordination,
in agreement with the observable 1,2-addition product 18 formed
from Cp*Ti{MeC(NiPr)2}(NTol) (3) and PhSiH2Cl (eqn (6)).
Consistent with this, calculations on CpTi{MeC(NMe)2}(NPh)
(m7) and MeSiH2Cl found a thermodynamically stable prod-
A number of charge-neutral group 4 hydrazide(1-) com-
1
2
plexes possessing either a k - or k -N(NRR¢)H ligand have
been reported,11a,16a,16c,16d,16h,29 suggesting (assuming thermody-
namic control) that Na-protonated tautomers are preferred to the
Nb-protonated ones. None of these were prepared by protonation
of a hydrazide(2-) precursor. The only such reaction of a terminal
group 4 hydrazide was reported by Odom to result in Nb-
protonation, albeit without structural characterisation.11b Two
uct CpTi{MeC(NMe)2}(Cl){N(Ph)SiH2Me} (m8Cl, DGsol
=
-12.8 kcal mol-1). The transition state from m7 to m8Cl lies at
complexes containing a k -N(NPh2)B(C6F5)3 ligand have been
2
17.5 kcal mol-1 above the reaction partners (cf 17.1 kcal mol-1 for
tsm1-4Cl) with Si ◊ ◊ ◊ Cl and Ti ◊ ◊ ◊ Cl distances (2.325 and 2.583 A,
reported, prepared by reaction of terminal group 4 hydrazides
with B(C6F5)3 at the Na atom.7c,29
˚
respectively) that are comparable to those of tsm1-4Cl.
To clarify this site-selectivity issue, the thermodynamic pref-
erences involved, and to compare the outcomes of protonation
and alkylation with the Na-silylation found exclusively above we
carried out a series of further reactions with 1–3 and DFT studies
of model compounds.
The reactions of Cp*Ti{MeC(NiPr)2}(NNMe2) (1) with MeI,
EtBr and PhCH2Br are summarised in Scheme 2. In each case, N–
C bond formation occurs exclusively at Nb to form the hydrazidium
cations [Cp*Ti{MeC(NiPr)2}(NNMe2R)]+ (R = Me (20+), Et (21+)
or CH2Ph (22+)) and corresponding non-coordinated anion as
determined by X-ray crystallography (vide infra). No reaction
occurred between 1 and PrCl, and with PhCH2Cl a mixture of
unknown products was formed. There was no reaction of MeI,
The formation of m2Cl via MeSiH2Cl addition to b-NMe2
chelated m5 (Path 2) resembles in general terms that discussed
above for MeSiH3 addition. However, it is now the kinetically
favoured route with tsm5-6Cl lying only 9.6 kcal mol-1 above m5 (cf
DG# = 17.8 kcal mol-1 for tsm5-6) and 5.7 kcal mol-1 below tsm1-
4Cl. The geometry of tsm5-6Cl is also different to those of tsm5-6
and tsm1-4Cl, and is much better described as a pentacoordinate
silicon adduct on Na with little transfer yet of Cl to titanium (Si–
˚
˚
Cl = 2.193 A; Ti ◊ ◊ ◊ Cl = 3.247 A). This difference is attributed to
the electron-withdrawing chloride ligand in MeSiH2Cl compared
to hydride in MeSiH3 and is the origin of the lower energy of
tsm5-6Cl compared to tsm5-6. From tsm5-6Cl the hydrazide(1-)
product m6Cl is formed with all 4 nitrogens and the Cl atom lying
in an approximate plane. Rearrangement proceeds via m4Cl to
place b-NMe2 approximately trans to Cp and below this plane,
forming m2Cl as the final product.
We also considered whether silicon addition to the
Nb of m1 via
a
charge-separated intermediate species
[CpTi{MeC(NMe)2}(NNMe2SiH2Me)]Cl or the neutral analogue
CpTi{MeC(NMe)2}(Cl)(NNMe2SiH2Me) might be feasible (with
subsequent 1,2-migration of SiH2Me from Nb to Na). Both of
these species lie above the energy of transition state tsm5-6Cl and
so this potential pathway was not explored further.
Overall the calculations find a key role for the b-NMe2 group in
the reactions with silanes from both a kinetic and thermodynamic
point of view. Furthermore, the effect on selecting the final reaction
pathway depends also on whether the substrate is a silane or
chlorosilane.
Alkylation and protonation reactions
Scheme 2 Reactions of Cp*Ti{MeC(NiPr)2}(NNMe2) (1) with alkyl
halides and [Et3NH][BPh4]. Anions I- (for 20-I), Br- (for 21-Br and 22-Br)
and [BPh4]- (for 23-BPh4) omitted for clarity.
While silane addition to metal hydrazides has been hitherto unex-
plored, alkylation and protonation reactions have been extensively
2284 | Dalton Trans., 2012, 41, 2277–2288
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
Table 4 Selected bond lengths (A) and angles (◦) for
Table 5 Selected bond lengths (A) and angles (◦) for
[Cp*Ti{MeC(NiPr)2}{N(NMe2)H}][BPh4] (23-BPh4). Cpcent refers
to the C5Me5 ring carbon centroid
˚
˚
[Cp*Ti{MeC(NiPr)2}(NNMe3)]I
(20-I),
[Cp*Ti{MeC(NiPr)2}-
(NNMe2Et)]Br (21-Br) and the corresponding values for the previously
reported Cp*Ti{MeC(NiPr)2}(NNMe2) (1).7b Cpcent refers to the C5Me5
ring carbon centroid
Ti(1)–Cpcent
Ti(1)–N(2)
Ti(1)–N(4)
N(1)–H(1)
Cpcent–Ti(1)–N(1)
Cpcent–Ti(1)–N(3)
N(1)–Ti(1)–N(2)
2.035
Ti(1)–N(1)
Ti(1)–N(3)
N(1)–N(2)
1.889(3)
2.024(3)
1.432(3)
2.176(2)
2.098(3)
1.02(6)
112.8
115.7
40.49(9)
20-I
21-Br
17b
Cpcent–Ti(1)–N(2)
Cpcent–Ti(1)–N(4)
Ti(1)–N(1)–H(1)
140.2
117.3
146(3)
Ti(1)–Cpcent
Ti(1)–N(1)
Ti(1)–N(3)
Ti(1)–N(4)
2.064
2.074
2.063
1.728(3)
2.090(3)
2.078(3)
1.419(4)
125.1
118.6
118.7
1.717(5)
2.083(6)
2.085(5)
1.426(7)
127.5
117.5
117.7
1.723(2)
2.101(1)
2.101(1)
1.386(2)
119.5
120.8
120.8
N(1)–N(2)
Ti–Na distances are consistent with triple bonds as found in 1 and
their imido counterparts.7b,14b
Cpcent –Ti(1)–N(1)
Cpcent–Ti(1)–N(3)
Cpcent–Ti(1)–N(4)
Ti(1)–N(1)–N(2)
Reaction of 1 with [Et3NH][BPh4] occurred immediately in THF
to form 23-BPh4 which contains the base-free, Na-protonated
cation [Cp*Ti{MeC(NiPr)2}{N(NMe2)H}]+. When followed by
1H NMR spectroscopy in THF-d8 23-BPh4 was the only product
observed (along with the Et3N side-product). Compounds 2 and
3 also reacted with [Et3NH][BPh4] in THF-d8 but required an
excess of ammonium salt to drive the first-formed equilibrium
mixtures to completion. The products are tentatively assigned
as the analogues of 23-BPh4 but definitive structural or other
analytical data could not be obtained. Bearing in mind that
certain hydrazido compounds are known to form Na-coordinated
adducts with B(C6F5)3,7c,29 analogous reactions of 1 were carried
out. However, no reaction was observed with either B(C6F5)3 or
Me3N·BH3, whereas with MeH2N·BH3 unknown mixtures were
formed with complex 1H and 11B NMR spectra.
In order to clearly establish the site of protonation in 23-
BPh4 the X-ray structure has been determined. Selected distances
and angles are listed in Table 5 and a view of the cation is
given in Fig. 7 which confirms that Na carries the added H+.
The Ti–Na and Ti–Nb distances to the N(NMe2)H ligand in
23+ are significantly shorter than those for the N(NMe2)SiH2Ar
ligands in 4–6 and 8 which is mainly attributed to the higher
formal charge and lower coordination number in the cation.
The Na–Nb distance is in general equivalent within error to
those in these neutral complexes. While titanium and zirco-
173.5(3)
177.5(5)
159.5(1)
EtBr or PhCH2Br with either Cp*Ti{MeC(NiPr)2}(NNPh2) (2) or
Cp*Ti{MeC(NiPr)2}(NTol) (3). The same reaction outcomes were
found in both benzene and THF in all instances. When followed
in C6D6 or THF-d8 the reactions of 1 with MeI, PhCH2Br or EtBr
were considerably slower than for PhSiH2X (X = H, Cl or Br) which
reacted immediately under the same conditions. The reaction with
MeI was complete only after 2 h, and those with PhCH2Br and
EtBr required up to 2 days. Despite these slow rates, reaction
of PhCH2Br with Cp*Ti{MeC(NiPr)2}(H){N(NMe2)SiH2Ph} (4)
formed exclusively 22-Br and PhSiH3 indicating that Nb-alkylation
is thermodynamically preferred to 1,2-addition of Si–H to Ti Na.
No evidence for direct Ti–H/C–Br exchange was found, in con-
trast to the reaction between 4 with PhSiH2Br. The corresponding
reaction using 4-d3 confirmed the absence of any Ti–D/C–Br
exchange.
The X-ray structures of 20-I and 21-Br have been determined. A
view of the cation 21+ is given in Fig. 6 and that of 20+ is given in the
ESI.† Selected distances are given in Table 4 along with those of
1 for comparison. The structures unambiguously confirm that the
alkyl halides lead to Nb quaterisation as expected.11a,11c In general
terms, 20+ and 21+ have comparable geometries to 1 and also to the
isoelectronic, neutral imido complexes Cp*Ti{RC(NR¢)2}(NXyl)
2
i
(R = Me or Ph; R¢ = Pr or SiMe3; Xyl = 2,6-C6H3Me2).12j,14c The
nium compounds containing the k -N(NMe2)H ligand have been
reported previously, none were prepared by protonation of a
hydrazide(2-) precursor.16a,16h,35 The observation that 1 undergoes
protonation at Na differs from Odom’s report of Nb protonation in
[Ti(dmpa)(bipytBu2)(NNH2Ph)]+ based on spectroscopic data.11b
Ti–Na distances in the cations are comparable within error to
that in 1 whereas the Na–Nb distance is significantly longer in
each case. This is attributed in part to the loss of residual p-
bonding7b between these atoms on quaternarisation. The short
Fig.
6
Displacement ellipsoid plots (20% probability) of
Fig. 7 Displacement ellipsoid plot (20% probability) of the cation
[Cp*Ti{MeC(NiPr)2}{N(NMe2)H}]+ (23+). C-bound H atoms and [BPh4]-
counter-ion omitted for clarity. H(1) is drawn as a sphere of arbitrary
radius.
[Cp*Ti{MeC(NiPr)2}(NNMe2Et)]+ (21+).
H
atoms and bromide
counter-ions omitted for clarity. [Cp*Ti{MeC(NiPr)2}(NNMe3)]+ (20+)
follows an analogous numbering scheme (see the ESI†).
This journal is
The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 2277–2288 | 2285
©

View Article Online
The lack of any reaction between 3 and alkyl halides, and
the observation that 1 reacts with these substrates only at Nb,
shows that Na alkylation is kinetically and/or thermodynamically
disfavoured. The absence of any reaction for 2 compared with 1
is consistent with the less accessible Nb lone pair in the former.
On the other hand, all three react with [Et3NH]+ giving Na–H
bond formation as confirmed by the X-ray structure of 23-[BPh4].
Likewise, all three give only Na–Si bond formation with silanes
and halosilanes. A precise comparison of C–X vs. Si–X reactivity
is the slow reaction of 1 with PhCH2Br to form 22-Br (Nb–C
bond) in contrast to the immediate reaction with PhSiH2Br to
form 13 (Na–Si bond). In addition, DFT found that DGsol for
the reaction of CpTi{MeC(NMe)2}(NNMe2) (m1) with EtBr to
form CpTi{MeC(NMe)2}(Br){N(NMe2)Et} (m9) was favoured by
tendency to form 5-coordinate species makes these types of
pathways kinetically much less accessible. In confirmation of this,
the transition state (tsm5-9Br) linking EtBr + model m5 (cf. Fig.
5) and m9 via a C–Br 1,2-addition process was found at DGsol
=
48.9 kcal mol-1 relative to m1 and EtBr.
Electrophilic attack by alkyl halides at Nb in kinetic preference
to Na can be explained by examination of the HOMO of 1
(modelled by m1) which has ca. 20% Na and. 35% Nb character.7b
In addition, consideration of a space-filling model for 1 shows
that Na is much less sterically accessible than the pyramidalised Nb
atom. Thus both frontier orbital and steric control appear to direct
the outcome of alkylation reactions whose rates are moderated by
the leaving group ability of the halide. Once the new Nb–C bond
has formed it is kinetically impossible to rearrange to the Na-
alkylated isomer. In contrast, as shown experimentally by Tuczek
et al.,32 Nb could be the kinetically preferred site for protonation,
with migration to Na being fast and facile through a 1,2-shift
between the N atoms (mediated in principle by the THF solvent
or liberated Et3N) to form the thermodynamic product.
-23.8 kcal mol-1. This can be compared, for example, with DGsol
=
-21.8 kcal mol-1 for m1 + MeSiH2Cl → m2Cl (Fig. 5) and shows
that C–Br addition across Ti Na is thermodynamically viable.
Conclusions
The different reaction outcomes of Cp*Ti{MeC(NiPr)2}(NNMe2)
(1), Cp*Ti{MeC(NiPr)2}(NNPh2) (2) and Cp*Ti{MeC(NiPr)2}-
(NTol) (3) with silanes, halosilanes, alkyl halides and
[Et3NH][BPh4] represent a unique comparison in transition
metal hydrazide chemistry. Reversible 1,2-addition of Si–H to
Ti NNMe2 allows an experimental determination of the ener-
getics of this process which DFT shows proceeds preferentially via
To test the intrinsic thermodynamic preferences for
Na- and Nb-bond formation with H, alkyl or silyl
groups, DFT calculations were carried out on the isomeric
1
model species [CpTi{MeC(NMe2)2}{k -N(NMe2)R}]+ (m10_R+),
2
[CpTi-{MeC(NMe2)2}{k -N(NMe2)R}]+ (m11_R+) and [CpTi-
{MeC(NMe2)2}(NNMe2R)]+ (m12_R+) for R = H, Me or SiH2Me.
The results are summarised in Fig. 8 in terms of their free energies
relative to that of m12_R+.
1
a non-chelated (k ) hydrazide group. However, b-NMe2 coordi-
nation is essential to stabilise the hydride-hydrazide(1-) product.
According to DFT, H–H addition to Ti Na is less favoured
compared to Si–H by ca. 9 kcal mol-1 and is not observed exper-
imentally. However, net Si–H/Ti–H/H–H exchange does occur
via a series of s-bond metathesis and 1,2-addition/elimination
steps, as established by labelling experiments. Addition of primary
halosilanes to Ti NNMe2 is irreversible and proceeds prefer-
2
entially via a k -coordinated hydrazide ligand and nucleophilic
attack at silicon by Na due to the higher Lewis acidity of silicon
in these substrates. With secondary chlorosilanes the addition
is reversible but still more favourable than for Si–H. Reaction
of dichlorosilanes with 1 results in double Si–Cl activation and
transfer of the amidinate ligand to silicon. The hydrazidium
products (Nb-alkylation in contrast to Na-silylation) formed with
alkyl halides are kinetic rather than thermodynamic products
and arise because the HOMO of 1 is predominantly based on
Nb and carbon is less readily able to adopt the hypercoordinate,
low energy transition states that silicon can access to achieve
1,2-addition to Ti Na. Protonation occurs at Na in terms of
Fig. 8 Model complexes used in the DFT calculations and their relative
energies (DGsol, kcal mol-1 at 298 K). Representative selected distances (A)
˚
and angles (◦): for m10_H, Ti–Na = 1.851, Na–Nb = 1.393, Ti–Na–Nb
=
141.0; for m11_H, Ti–Na = 1.869, Na–Nb = 1.394, Ti–Na–Nb = 79.4; for
m12_H, Ti–Na = 1.740, Na–Nb = 1.400, Ti–Na–Nb = 156.8.
2
In all cases the k -N(NMe2)R isomer m11_R+ is by far the most
thermodynamically stable one. As found in Fig. 5, chelation of the
b-NMe2 group in m10_R+ to form m11_R+ is very favourable.
Fig. 8 also shows that Na-bond formation, even without b-
NMe2 coordination, remains thermodynamically preferred. This
is consistent with previous DFT calculations on group 5 and 6
metals.32–34
These results establish that Si–X and H+ addition reactions
to 1 are under thermodynamic control, whereas those for C–X
are under kinetic control. For both of the low energy pathways
for Si–X 1,2-addition, Fig. 5 shows the importance of accessing
a hypercoordinate silicon centre. Therefore carbon’s reduced
1
2
thermodynamic preference in either the k - or k -coordinated
forms, but mechanistically may proceed via initial attack at Nb
as proposed for other hydrazides, followed by a 1,2-shift to Na.
Acknowledgements
We thank the Malaysian Higher Education Ministry for a
scholarship to P.J.T. and the EPSRC for a Ph.D plus postdoctoral
award to J.D.S., and a postdoctoral fellowship to A. D. S. We
also thank the Spanish MICINN for postdoctoral and Juan la
2286 | Dalton Trans., 2012, 41, 2277–2288
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
Cierva fellowships to A.N., and Dr A. L. Thompson for advice
with regard to the crystallography.
10 (a) H. K. Sharma and K. H. Pannell, Chem. Rev., 1995, 95, 1351; (b) J.
Y. Corey and J. Braddock-Wilking, Chem. Rev., 1999, 99, 175; (c) J. F.
Harrod, Coord. Chem. Rev., 2000, 206–207, 493; (d) G. I. Nikonov, J.
Organomet. Chem., 2001, 635, 24; (e) R. Waterman, P. G. Hayes and T.
D. Tilley, Acc. Chem. Res., 2007, 40, 712; (f) J. Y. Corey, Chem. Rev.,
2011, 111, 863.
Notes and references
11 (a) Y. Li, Y. Shi and A. L. Odom, J. Am. Chem. Soc., 2004, 126, 1794;
(b) S. Patel, Y. Li and A. L. Odom, Inorg. Chem., 2007, 46, 6373; (c) A.
J. Clulow, J. D. Selby, M. G. Cushion, A. D. Schwarz and P. Mountford,
Inorg. Chem., 2008, 47, 12049.
1 (a) B. F. G. Johnson, B. L. Haymore and J. D. Dilworth, in Comprehen-
sive Coordination Chemistry, ed. G. Wilkinson, R. D. Gillard and J. A.
McCleverty, Pergamon Press, Oxford, 1987, p. 100; (b) W. A. Nugent
and J. M. Mayer, Metal–Ligand Multiple Bonds, Wiley-Interscience,
New York, 1988; (c) D. J. Mindiola, Angew. Chem., Int. Ed., 2008, 47,
1557.
2 (a) M. D. Fryzuk, Chem. Rec., 2003, 3, 2; (b) B. A. MacKay and M. D.
Fryzuk, Chem. Rev., 2004, 104, 385; (c) C. M. Kozak and P. Mountford,
Angew. Chem., Int. Ed., 2004, 43, 1186.
3 (a) J. Chatt, A. J. Pearman and R. L. Richards, Nature, 1975, 253, 39;
(b) C. J. Pickett and G. J. Leigh, J. Chem. Soc., Chem. Commun., 1981,
1033; (c) W. Hussain, G. J. Leigh and C. J. Pickett, J. Chem. Soc., Chem.
Commun., 1982, 747; (d) C. J. Pickett and J. Talamarin, Nature, 1985,
317, 652; (e) K. Arashiba, Y. Miyake and Y. Nishibayashi, Nat. Chem.,
2011, 3, 120.
4 (a) D. V. Yandulov and R. R. Schrock, Science, 2003, 76; (b) R. R.
Schrock, Acc. Chem. Res., 2005, 38, 955.
5 (a) G. C. Stephan, C. Sivasankar, F. Studt and F. Tuczek, Chem.–Eur.
J., 2008, 14, 644; (b) F. Studt and F. Tuczek, Angew. Chem., Int. Ed.,
2005, 44, 5639; (c) R. R. Schrock, Angew. Chem., Int. Ed., 2008, 47,
5512.
6 (a) N. Wiberg, H.-W. Haring, G. Huttner and P. Friedrich, Chem. Ber.,
1978, 111, 2708; (b) P. J. Walsh, M. J. Carney and R. G. Bergman,
J. Am. Chem. Soc., 1991, 113, 6343; (c) A. J. Blake, J. M. McInnes,
P. Mountford, G. I. Nikonov, D. Swallow and D. J. Watkin, J. Chem.
Soc., Dalton Trans., 1999, 379; (d) J. L. Thorman and L. K. Woo, Inorg.
Chem., 2000, 39, 1301.
7 (a) S. Banerjee and A. L. Odom, Organometallics, 2006, 25, 3099; (b) P.
J. Tiong, A. Nova, L. R. Groom, A. D. Schwarz, J. D. Selby, A. D.
Schofield, E. Clot and P. Mountford, Organometallics, 2011, 30, 1182;
(c) A. D. Schofield, A. Nova, J. D. Selby, A. D. Schwarz, E. Clot
and P. Mountford, Chem.–Eur. J., 2011, 17, 265; (d) P. J. Tiong, A.
D. Schofield, J. D. Selby, A. Nova, E. Clot and P. Mountford, Chem.
Commun., 2010, 46, 85; (e) A. D. Schofield, A. Nova, J. D. Selby, C.
D. Manley, A. D. Schwarz, E. Clot and P. Mountford, J. Am. Chem.
Soc., 2010, 132, 10484; (f) J. D. Selby, C. Schulten, A. D. Schwarz, A.
Stasch, E. Clot, C. Jones and P. Mountford, Chem. Commun., 2008,
5101; (g) J. D. Selby, C. D. Manley, M. Feliz, A. D. Schwarz, E. Clot
and P. Mountford, Chem. Commun., 2007, 4937; (h) T. Gehrmann, J.
L. Fillol, H. Wadepohl and L. H. Gade, Organometallics, 2010, 29,
28; (i) K. Weitershaus, H. Wadepohl and L. H. Gade, Organometallics,
2009, 28, 3381; (j) K. Weitershaus, J. L. Fillol, H. Wadepohl and L. H.
Gade, Organometallics, 2009, 28, 4747; (k) T. Gehrmann, J. L. Fillol,
H. Wadepohl and L. H. Gade, Angew. Chem., Int. Ed., 2009, 48, 2152.
8 (a) H. Herrmann, J. L. Fillol, H. Wadepohl and L. H. Gade, Angew.
Chem., Int. Ed., 2007, 46, 8426; (b) T. Gehrmann, J. L. Fillol, S. A.
Scholl, H. Wadepohl and L. H. Gade, Angew. Chem., Int. Ed., 2011,
50, 5757.
9 (a) T. I. Gountchev and T. D. Tilley, J. Am. Chem. Soc., 1997, 119,
12831; (b) C. C. Cummins, S. M. Baxter and P. T. Wolczanski, J. Am.
Chem. Soc., 1988, 110, 8731; (c) A. W. Kaplan, J. L. Polse, G. E. Ball, R.
A. Andersen and R. G. Bergman, J. Am. Chem. Soc., 1998, 120, 11649;
(d) Z. K. Sweeny, J. L. Polse, R. G. Bergman and R. A. Andersen,
Organometallics, 1999, 18, 5502; (e) T. E. Hanna, E. Lobkovsky and P.
J. Chirik, Inorg. Chem., 2007, 46, 2359; (f) D. Sieh, J. Schoffel and P.
Burger, Dalton Trans., 2011, 40, 9512; (g) G. I. Nikonov, P. Mountford,
S. K. Ignatov, J. C. Green, P. A. Cooke, M. A. Leech, L. G. Kuzmina,
A. G. Razuvaev, N. H. Rees, A. J. Blake, J. A. K. Howard and D.
A. Lemenovskii, J. Chem. Soc., Dalton Trans., 2001, 2903; (h) S. K.
Ignatov, A. A. Merkoulov, N. H. Rees, S. R. Dubberley, A. G. Razuvaev,
P. Mountford and G. I. Nikonov, Chem.–Eur. J., 2008, 14, 296; (i) S. K.
Ignatov, A. Y. Khalimon, N. H. Rees, A. G. Razuvaev, P. Mountford and
G. I. Nikonov, Inorg. Chem., 2009, 48, 9605; (j) U. Burckhardt, G. L.
Casty, T. D. Tilley, T. K. Woo and U. Rothlisberger, Organometallics,
2000, 19, 3830; (k) J. B. Diminnie and Z. Xue, J. Am. Chem. Soc.,
1997, 119, 12657; (l) T. E. Hanna, I. Keresztes, E. Lobkovsky, W. H.
Bernskoetter and P. J. Chirik, Organometallics, 2004, 23, 3448; (m) J. L.
Polse, R. A. Andersen and R. G. Bergman, J. Am. Chem. Soc., 1998,
120, 13405.
12 (a) P. Mountford, Chem. Commun., 1997, 2127; (b) P. D. Bolton and
P. Mountford, Adv. Synth. Catal., 2005, 347, 355; (c) N. Hazari and
P. Mountford, Acc. Chem. Res., 2005, 38, 839; (d) A. E. Guiducci, A.
R. Cowley, M. E. G. Skinner and P. Mountford, J. Chem. Soc., Dalton
Trans., 2001, 1392; (e) H. R. Bigmore, S. R. Dubberley, M. Kranenburg,
S. C. Lawrence, A. J. Sealey, J. D. Selby, M. Zuideveld, A. R. Cowley and
P. Mountford, Chem. Commun., 2006, 436; (f) A. Bashall, P. E. Collier,
L. H. Gade, M. McPartlin, P. Mountford, S. M. Pugh, S. Radojevic, M.
Schubart, I. J. Scowen and D. J. M. Tro¨sch, Organometallics, 2000, 19,
4784; (g) G. I. Nikonov, A. J. Blake and P. Mountford, Inorg. Chem.,
1997, 36, 1107; (h) S. C. Dunn, P. Mountford and O. V. Shishkin, Inorg.
Chem., 1996, 35, 1006; (i) P. J. Stewart, A. J. Blake and P. Mountford,
Inorg. Chem., 1997, 36, 1982; (j) A. E. Guiducci, C. L. Boyd and P.
Mountford, Organometallics, 2006, 25, 1167.
13 P. J. Tiong, A. Nova, E. Clot and P. Mountford, Chem. Commun., 2011,
47, 3147.
14 (a) C. L. Boyd, A. E. Guiducci, S. R. Dubberley, B. R. Tyrrell and P.
Mountford, J. Chem. Soc., Dalton Trans., 2002, 4175; (b) C. L. Boyd,
E. Clot, A. E. Guiducci and P. Mountford, Organometallics, 2005, 24,
2347; (c) A. E. Guiducci, C. L. Boyd, E. Clot and P. Mountford, Dalton
Trans., 2009, 5960.
15 (a) D. A. Fletcher, R. F. McMeeking and D. Parkin, J. Chem. Inf.
Model., 1996, 36, 746(The UK Chemical Database Service: CSD
version 5.32 updated August 2011); (b) S. K. Ignatov, N. H. Rees,
B. R. Tyrrell, S. R. Dubberley, A. G. Razuvaev, P. Mountford and
G. I. Nikonov, Chem.–Eur. J., 2004, 10, 4991; (c) W. W. Lukens, P.
T. Matsunaga and R. A. Andersen, Organometallics, 1998, 17, 5240;
(d) M. D. Walter, C. D. Sofield and R. A. Andersen, Organometallics,
2008, 27, 2959; (e) J. M. de Wolf, A. Meetsma and J. H. Teuben,
Organometallics, 1995, 14, 5466; (f) Y. You and G. S. Girolami,
Organometallics, 2008, 27, 3172; (g) H. Noth and M. Schmidt,
Organometallics, 1995, 14, 4601.
16 (a) I. A. Latham, G. J. Leigh, G. Huttner and I. Jibril, J. Chem. Soc.,
Dalton Trans., 1986, 385; (b) R. Hemmer, U. Thewalt, D. L. Hughes, G.
J. Leigh and D. G. Walker, J. Organomet. Chem., 1987, 323, C29; (c) D.
L. Hughes, G. J. Leigh and D. G. Walker, J. Chem. Soc., Dalton Trans.,
1989, 1413; (d) D. L. Hughes, M. Jimenez-Tenorio, G. J. Leigh and D.
G. Walker, J. Chem. Soc., Dalton Trans., 1989, 2389; (e) B. Goetze, J.
Knizek, H. Noth and W. Schnick, Eur. J. Inorg. Chem., 2000, 1849;
(f) J.-S. M. Lehn and D. M. Hoffman, Inorg. Chim. Acta, 2003, 345,
327; (g) J. M. Pietryga, J. N. Jones, C. L. B. Macdonald, J. A. Moore
and A. H. Cowley, Polyhedron, 2006, 25, 259; (h) J. D. Selby, M. Feliz,
A. D. Schwarz, E. Clot and P. Mountford, Organometallics, 2011, 30,
2295.
17 T. Janssen, R. Severin, M. Diekmann, M. Friedemann, D. Haase,
W. Saak, S. Doye and R. Beckhaus, Organometallics, 2010, 29,
1806.
18 A. J. Hoskin and D. W. Stephan, Coord. Chem. Rev., 2002, 233–234,
107.
19 X. Liu, Z. i. Wu, H. Cai, Y. Yang, T. Chen, C. E. Vallet, R. A. Zuhr,
D. B. Beach, Z.-H. Peng, Y.-D. Wu, T. E. Concolino, A. L. Rheingold
and Z. Xue, J. Am. Chem. Soc., 2001, 123, 8011.
20 L. J. Procopio, P. J. Carroll and D. H. Berry, J. Am. Chem. Soc., 1994,
116, 177.
21 M. P. Coles, Dalton Trans., 2006, 985.
22 On the available experimental evidence it is not possible to determine
the relative proportions of 13 and [13-Br]+ in solution above the fast
exchange limit.
23 K. Y. Dorogov, A. V. Churakov, L. G. Kuzmina, J. A. K. Howard and
G. I. Nikonov, Eur. J. Inorg. Chem., 2004, 771.
24 (a) A. R. Fout, U. J. Kilgore and D. J. Mindiola, Chem.–Eur. J.,
2007, 13, 9428; (b) D. E. Wigley, Prog. Inorg. Chem., 1994, 42,
239.
25 M. F. Lappert, A. V. Protchenko, P. P. Power and A. Seeber, Metal
Amide Chemistry, Wiley, Chichester, 2009.
This journal is
The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 2277–2288 | 2287
©

View Article Online
26 Y. Zhao and D. G. Truhlar, Theor. Chem. Acc., 2008, 120, 215.
27 (a) J. L. Bennett and P. T. Wolczanski, J. Am. Chem. Soc., 1997, 119,
10696; (b) C. C. Cummins, C. P. Schaller, G. D. Van Duyne, P. T.
Wolczanski, A. W. E. Chan and R. Hoffmann, J. Am. Chem. Soc.,
1991, 113, 2985.
31 (a) T. Takahashi, Y. Mizobe, M. Sato, Y. Uchida and M. Hidai, J.
Am. Chem. Soc., 1979, 101, 3405; (b) J. R. Dilworth, R. A. Henderson,
P. Dahlstrom, T. Nicholson and J. A. Zubieta, J. Chem. Soc., Dalton
Trans., 1987, 529; (c) A. Galindo, A. Hills, D. L. Hughes, R. L. Richards,
M. Hughes and J. Mason, J. Chem. Soc., Dalton Trans., 1990, 283.
32 A. Dreher, K. Mersmann, C. Nather, I. Ivanovic-Burmazovic, R. van
Eldik and F. Tuczek, Inorg. Chem., 2009, 48, 2078.
28 DGsol for H2 addition to 1 was -0.5 kcal mol-1. This slightly
negative value is at the limits of the DFT method, especially
considering the intrinsic inaccuracy of estimating DS using this
method.
33 S. Schenk, B. Le Guennic, B. Kirchner and M. Reiher, Inorg. Chem.,
2008, 47, 3634.
29 H. Herrmann, J. L. Fillol, T. Gehrmann, M. Enders, H. Wadepohl and
L. H. Gade, Chem. Eur. J., 2008, 14, 8131.
30 M. G. Vale and R. R. Schrock, Inorg. Chem., 1993, 32, 2767.
34 A. K. Guha and A. K. Phukan, Inorg. Chem., 2011, 50, 8826.
35 H. Herrmann, H. Wadepohl and L. H. Gade, Dalton Trans., 2008,
2111.
2288 | Dalton Trans., 2012, 41, 2277–2288
This journal is
The Royal Society of Chemistry 2012
©