Two alternative approaches to access mixed hydride-amido zinc complexes: Synthetic, structural and solution implications

Article abstract of DOI:10.1039/c5dt00312a

Using bis(amide) Zn(HMDS)₂ (HMDS = 1,1,1,3,3,3-hexamethyldisilazide) as a precursor, this study explores the synthesis of N-heterocyclic carbene stabilized mixed amido-hydride zinc complexes using two alternative hydride sources, namely dimethy

Full text of DOI:10.1039/c5dt00312a

Dalton
Transactions
View Article Online
View Journal
PAPER
Two alternative approaches to access mixed
hydride-amido zinc complexes: synthetic,
Cite this: DOI: 10.1039/c5dt00312a
structural and solution implications†‡
Andrew J. Roberts,a William Clegg,b Alan R. Kennedy,a Michael R. Probert,b
Stuart D. Robertson*a and Eva Hevia*a
Using bis(amide) Zn(HMDS)2 (HMDS = 1,1,1,3,3,3-hexamethyldisilazide) as a precursor, this study explores
the synthesis of N-heterocyclic carbene stabilized mixed amido-hydride zinc complexes using two
alternative hydride sources, namely dimethylamine borane (DMAB) and phenylsilane PhSiH3. Hydride-rich
zinc cluster Zn4(HMDS)2H6·2IPr (1) (IPr = 1,3-bis(2,6-di-isopropylphenyl)imidazol-2-ylidene), which can
be envisaged as a co-complex of IPr·ZnH2 and (HMDS)ZnH, is obtained when DMAB is employed, with
the concomitant formation of heteroleptic bis(amido)borane [HB(NMe2)(HMDS)] and H2 evolution. NMR
studies in d8-THF show that although the bulky carbene IPr does not bind to the zinc bis(amide), its pres-
ence in the reaction media is required in order to stabilise 1. Reactions using the slightly less sterically
demanding NHC IXy (IXy = 1,3-bis-(2,6-dimethylphenyl)imidazol-2-ylidene) led to the isolation and struc-
tural elucidation of the carbene adduct Zn(HMDS)2·IXy (2). Contrastingly, mixtures of equimolar amounts
of PhSiH3 and the zinc bis(amide) (60 °C, 3 h, hexane) afforded monomeric heteroleptic hydride (HMDS)
ZnH·IPr (3). NMR studies, including DOSY experiments, revealed that while the integrity of 3 is retained in
polar d8-THF solutions, in lower polarity C6D6 it displays a much more complex solution behaviour, being
in equilibrium with the homoleptic species ZnH2·IPr, Zn(HMDS)2 and IPr.
Received 22nd January 2015,
Accepted 27th March 2015
DOI: 10.1039/c5dt00312a
www.rsc.org/dalton
the aggregation state, aiding solubility. Among the most
studied of the early main group metal hydrides are those of
Introduction
Metal hydride species1 are currently at the forefront of metal magnesium and zinc, primarily using either bulky monoanio-
research due to their widespread utility as reagents for chemi- nic ligands such as β-diketiminate (nacnac) and its deriva-
cal transformations such as deprotonation/metallation and tives,4 or neutral N-heterocyclic carbenes5 to sterically protect
reduction2 as well as their potential in other socio-economi- the metal–hydrogen bond (Fig. 1). Bimetallic magnesium/
cally important fields such as energy storage.3 Of particular alkali-metal hydride clusters have also been reported contain-
importance for the latter application are lightweight, environ- ing the secondary amido diisopropylamide6 or 1,1,1,3,3,3-hexa-
mentally benign, high-abundance metals as these are more methyldisilazide [HMDS, N(SiMe3)2] anion.7
economically viable and have better gravimetric hydrogen
Running in parallel to this research has been that of the
storage capacity. Binary metal hydride species, however, have a amine boranes (RNH2·BH3; R2NH·BH3), which contain protic
number of drawbacks such as pyrophoricity and poor solubi- and hydridic hydrogen atoms in close proximity to one
lity (which in turn leads to low reactivity). Consequently, another, making them primed for hydrogen release under the
neutral or anionic co-ligands must be incorporated to lower correct conditions,8 as well as precursors for the synthesis of
B–N oligomers9 and polymers.10 Again, the use of main group
aWestCHEM, Department of Pure and Applied Chemistry, University of Strathclyde,
Glasgow, G1 1XL, UK. E-mail: eva.hevia@strath.ac.uk,
stuart.d.robertson@strath.ac.uk
bSchool of Chemistry, Newcastle University, Newcastle upon Tyne, NE1 7RU, UK
†Dedicated to the memory of Professor Ken Wade, a pioneer and an inspiration
to Inorganic chemists.
‡Electronic supplementary information (ESI) available: NMR spectroscopic data
and X-ray data in crystallographic file (CIF) format for compounds 1, 2 and 3.
CCDC 1042314–1042316. For ESI and crystallographic data in CIF or other elec-
Fig. 1 Nacnac anion (left) and N-heterocyclic carbene (right) typically
tronic format see DOI: 10.1039/c5dt00312a
used as aggregation-lowering co-ligands in metal hydride complexes.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 2 Crystallographically characterized molecular NHC-stabilized
neutral and cationic zinc hydride species. SIMes = 1,3-bis(2,4,6-tri-
Scheme 1
methylphenyl)imidazol-4,5-dihydro-2-ylidene, Cp*
=
pentamethyl-
cyclopentadienyl, IPr = 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene,
IMes
=
1,3-bis(2,4,6-trimethylphenyl)imidazol-2-ylidene, ImMe2iPr2
=
rated N-heterocyclic carbene IPr (IPr
= 1,3-bis(2,6-diiso-
1,3-diisopropyl-4,5-dimethylimidazol-2-ylidene, DMAP
=
4-dimethyl-
aminopyridine, OTf = triflate. References: A, 5g; B, 5b; C, 5b; D, 5f; E, 5c;
F, 5f; G, 5f; H, 5e; I, 5d.
propylphenyl)imidazol-2-ylidene) did not re-dissolve this
precipitate. However, we did discover that adding a molar equi-
valent of IPr to the solution of Zn(HMDS)2 prior to the intro-
duction of the amine borane, followed by removal of THF,
addition of hexane and reintroduction of THF dropwise with
stirring, resulted in a homogenous solution. Storage of this
solution at −30 °C deposited a small crop of colourless
crystals which were revealed by synchrotron single-crystal
X-ray diffraction to be the zinc mixed amido–hydrido cluster
Zn4(HMDS)2H6·2IPr (1) (Scheme 1a and Fig. 3; note that a non-
metals in this area for environmental and economic reasons is
pervasive with prominent contributions arising from the
groups of Harder,11 Hill12 and Wright.13 While magnesium
amidoboranes have been successfully prepared and extensively
studied,14 zinc amidoboranes have never been reported,
although Harder has suggested that these are intermediate
species on the way to the monomeric zinc hydride complex
[CH{C(Me)N(Dipp)}2ZnH] which can be obtained in good
yields by salt metathesis of [CH{C(Me)N(Dipp)}2ZnCl] and a
potassium amidoborane.4c
Our interest in this area was stimulated by the excellent
recent work of Okuda,5b–e and Rivard,5f who have reported
both neutral and cationic NHC-stabilized zinc hydride species
(specifically with a focus on alkyl zinc/zinc halide reagents),
some of which are effective catalysts for hydrosilylation reac-
tions (Fig. 2). Diverging from Harder’s approach, these com-
pounds can be prepared using phenylsilanes as hydride
sources. Given the considerable carbophilicity of zinc, we sur-
mised that zinc amides [specifically Zn(HMDS)2], with their
relatively weaker and kinetically activated Zn–N bonds (com-
pared to Zn–C bonds),15 might provide a novel, more facile
access point to zinc hydride complexes while also considering
that NHCs might stabilize the as yet elusive zinc amido-
boranes better than nacnac anions. We present our findings of
these studies herein.
Results and discussion
Fig. 3 Molecular structure of complex Zn4(HMDS)2H6·2IPr (1) with 50%
We commenced synthetic studies by reacting an equal stoichio-
probability ellipsoids. All non-hydridic hydrogen atoms and THF mole-
metry of dimethylamine borane (DMAB) with Zn(HMDS)2 in cule of solvation have been removed for clarity. Peripheral carbon atoms
have been made more transparent for clarity. Selected bond parameters
(Å and °): Zn1–H12 1.82(3), Zn1–H13 1.81(3), Zn1–H14 1.82(3), Zn2–H12
1.82(3), Zn2–H23 1.83(3), Zn2–H24 1.82(3), Zn3–H13 1.81(3), Zn3–H23
1.82(3), Zn3–H34 1.83(3), Zn4–H14 1.81(3), Zn4–H24 1.82(3), Zn4–H34
THF solution. This mixture immediately precipitated an
insoluble white product despite the presence of a vast
excess of Lewis donating solvent. Addition of tetramethyl-
ethylenediamine (TMEDA), the potentially tetradentate donor
1.83(3), Zn1–C1 2.040(8), Zn2–C28 2.035(9), Zn3–N5 1.944(6), Zn4–N6
tris[2-(dimethylamino)ethyl]amine (Me6TREN), or the unsatu- 1.939(7).
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
interacting molecule of THF is present in the crystal structure). the hydride resonances appear as three broad signals in a
This synthesis could be improved by using less polar cyclo- 1 : 4 : 1 ratio at 2.85, 3.35 and 4.14 ppm respectively, at lower
hexane as the reaction medium, with a larger crop of crystals temperatures (223 K) they appear as three well-resolved multi-
being deposited upon standing at room temperature. Interest- plets (Fig. S2, ESI‡). These chemical shifts are slightly down-
ingly, the crystals grown from the cyclohexane solution (that is, field of those recently reported for related dimeric mixed
in absence of THF) gave essentially the same unit cell para- alkyl–hydride zinc species [{(IMes)ZnMeH}2] (at 2.75 ppm).5c
meters as those with the THF molecule of solvation, The most informative resonance in the 13C{1H} NMR spectrum
suggesting its presence is not influencing the crystal packing.
is that for the carbenic carbon which is found at 181.0 ppm,
Comprising an adamantyl-like {Zn4H6}2+ core, this complex consistent with the retention of the Zn–Ccarbene interaction in
can be envisaged as a co-complex of IPr·ZnH2 and (HMDS) solution.17
ZnH. Interestingly, although a handful of tetranuclear zinc
Considering the heteroleptic constitution of 1, we found
hydride complexes have been reported,16 all of them contain that this compound could reproducibly be obtained in up to
terminal hydride ligands, whereas in 1 each hydride is acting 46% crystalline yield when the reaction was carried out using a
as a bridge between two zinc centres such that the core is a tetra- 3 : 4 : 2 ratio of DMAB, Zn(HMDS)2 and IPr respectively (see
hedral Zn4 cluster with the six hydride ligands slightly dis- Fig. S3 and S4‡ for H and 13C NMR spectra). In order to shed
1
placed from the six edges of the tetrahedron. A similar some light on the formation of 1 and the fate of the original
bridging coordination mode has recently been found in the amine borane, multinuclear (11B and 1H) NMR studies were
lower nuclearity NHC-stabilized cationic cluster {Zn3H4- carried out on the remaining filtrate from this reaction
(IMes)3(THF)}{BPh4}2.5e The hydride ligands in 1 were located mixture. The 11B NMR spectrum (see Fig. S5, ESI‡) revealed
and refined with fixed isotropic displacement parameters, with the formation of a single boron-containing species displaying
the only restraint applied on their positions being that the a doublet at 33.9 ppm (1JBH = 133.8 Hz), consistent with a BH
Zn–H bonds should be approximately equal in length (which functionality.18 Careful inspection of the 1H NMR spectrum
resulted in Zn–H bond lengths of 1.82 0.01 Å; the length was showed two informative singlets at 0.19 and 2.58 ppm along
not specified as part of the refinement input). This value is with a broad 1 : 1 : 1 : 1 quartet between 4 and 5 ppm (which
marginally longer than the Zn–H–Zn distances reported by reverts to a singlet in the boron-decoupled spectrum at
Coles for a {Zn5H6} cluster (mean 1.775 Å)16d and of the brid- 4.54 ppm) with a relative integration ratio of 18 : 6 : 1 (Fig. S6‡),
ging hydride–zinc distances in B (mean 1.712 Å), C (1.759 Å), E which allows the tentative assignment of this boron species as
(mean 1.775 Å), F (mean 1.78 Å) and H [1.65(3)–1.75(4) Å]. Zn– the hetero(bisamido)borane [HB(NMe2)(HMDS)]. Supporting
C(carbene) and Zn–N(HMDS) bond distances are consistent this proposed interpretation, GC-MS analysis of the viscous
with other such crystallographically characterized bonds (vide liquid residue obtained after removal of volatiles in vacuo
infra).
showed a series of peaks centred on m/z 216 with an intensity
We note at this juncture that a magnesium analogue of pattern consistent with that predicted for the molecular ion of
complex 1 has been reported previously by Hill and co-work- this species. Interestingly, a closely related alkyl(amido)borane
ers.5a The metal–hydride, metal–amide and metal–carbene intermediate has been previously isolated from the thermal
bond lengths are understandably shorter in our zinc complex decomposition of alkylstrontium amidoborane [{RSr(NMe2BH3)}2]
compared to the magnesium derivative due to zinc’s smaller (R = CH(SiMe3)2).12c,14b Building on these studies, we believe
size (covalent radius 1.22 Å for Zn vs. 1.41 Å for Mg). This is that, as similarly proposed by Harder for the synthesis of
most clearly seen in the Zn–carbene bonds, which have a [CH{C(Me)N(Dipp)}2ZnH] (vide supra), initially transient zinc
mean value of 2.038 Å, almost 8% shorter than in the Mg (amido)borane intermediate species (I) must be involved in the
complex [2.2063(19) Å]. The Zn–N (mean 1.966 Å) and Zn–H formation of 1 (Scheme 2), which in our case can be obtained
(mean 1.82 Å) are marginally shorter than the corresponding by direct deprotonation of DMAB by the zinc amide. This com-
Mg–N and Mg–H bond distances [2.0049(17) and 1.871(mean) pound must undergo a fast beta-hydride elimination process
Å respectively]. It is of interest that the magnesium complex
was prepared by reaction of Mg(HMDS)2·IPr with phenylsilane
at 60 °C. However, when we attempted the preparation of 1
using this approach, we obtained instead a mononuclear
mixed hydrido–amido complex (vide infra).
1H NMR spectra of 1 in C7D8 solution exhibit single sets of
signals for the IPr and HMDS fragments, along with three reso-
nances that can be assigned to the hydride groups. DOSY
NMR studies showed that all these different ligands present in
1 belong to the same sized species, as the cross points for all
their resonances are aligned in the second dimension
(Fig. S1,‡ average diffusion coefficient 1.19 × 10−9 m2 s−1),
which suggest that the integrity of the cluster is retained in
Scheme 2 Proposed pathway for the formation of 1.
this arene solvent. Interestingly, although at room temperature
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
to form unsaturated and highly reactive H2BvNMe2 along
with a mixed amido(hydride)zinc (II). Assuming that cluster 1
is obtained as a result of the co-complexation of two different
zinc–hydrido species, namely [IPr·ZnH2] and (HMDS)ZnH (vide
supra), at this stage it is possible to envisage two alternative
reaction pathways for the BH2vNMe2 species that, as shown
in Scheme 2, would explain the formation of bis(amido)borane
[HB(NMe2)(HMDS)] (III) as the only boron-containing species
of this reaction. On one hand and similarly to Hill’s investi-
gations on alkylstrontium amidoboranes, this unsaturated
species can undergo insertion in the polarized Zn–N bond of
the amido(hydride)zinc II (pathway a, Scheme 2).19 Sub-
sequent boron hydride elimination may lead to the formation
of bis(amido)borane III along with ZnH2, which in turn can be
trapped by the NHC ligand IPr (as discussed below IPr does
not coordinate to the precursor Zn(HMDS)2) affording homo-
leptic hydride B.5b On the other hand, and under the stoichio-
metric conditions studied, BH2vNMe2 can also react with the
amine HMDS(H) produced in the initial step for the formation
of I, affording III along with H2 elimination (Scheme 2,
Fig. 4 Molecular structure of one of the two crystallographically inde-
pendent molecules of complex Zn(HMDS)2·IXy (2) with 50% probability
ellipsoids. All hydrogen atoms and a disordered THF molecule of crystal-
lisation have been removed for clarity. The second Zn(HMDS)2·IXy mole-
cule has disordered HMDS ligands. Selected bond parameters (Å and °):
Zn1–C13, 2.100(5); Zn1–N1, 1.973(3); Zn1–N2, 1.961(4); N1–Zn1–N2,
123.0(2), N1–Zn1–C13, 119.9(1); N2–Zn1–C13, 117.1(2).
pathway b). Support for this second reaction pathway and the Determined by X-ray diffraction studies, the molecular struc-
concomitant H2 evolution in the formation of 1 was obtained ture of 2 is shown in Fig. 4.
by monitoring the reaction of Zn(HMDS)2/IPr with DMAB in
At 2.100(5) Å, the Zn–carbene distance is at the long end of
deuterated cyclohexane by 1H NMR spectroscopy, which reported three-coordinate zinc atoms coordinated by an NHC23
showed a sharp singlet at 4.54 ppm which can be assigned to as might be expected given our observation that the carbene is
molecular hydrogen.20 It should also be noted that, when this only just sterically unencumbered enough to access the zinc
reaction was carried out under harsher reaction conditions centre. The Zn–N distances (av. 1.967 Å) are consistent with
(3 hours at 60 °C in hexane), the formation of 1 is inhibited, Zn(HMDS)2·ItBu (av. 1.957 Å) and the zinc centre is distorted
affording instead homoleptic dimer [{IPrZnH2}2] (B), which trigonal planar (∑∠ = 360.0°). Compared to free Zn(HMDS)2,24
suggests that under these conditions the insertion reaction of the coordinating carbene understandably forces the N–Zn–N
BH2vNMe2 in the Zn–N bond of II is favoured.21
angle from 175.2° to 123.0(2)°, which is indeed tighter than
As mentioned above, an Mg-analogue of 1 has been pre- that seen in the less bulky ItBu adduct [131.4(2)°].25 Likewise,
viously prepared, using a Mg(HMDS)2·IPr adduct as a precur- the Zn–N bonds are elongated on going from a two-coordinate
sor.5a In the absence of any characterized NHC adducts of zinc (av. 1.833 Å) to a three-coordinate complex (av. 1.967 Å). The
bisamides,22 we also focused on the initial step of the reaction, carbene is considerably rotated to minimize steric clashing of
specifically the reaction of Zn(HMDS)2 with IPr. However, the flanking aromatic groups with the silylamide ligands (the
we were unable to prepare an IPr-stabilized complex of C3N2 ring plane is oriented at 62.4° with respect to the N–Zn–
1
Zn(HMDS)2. A H NMR spectroscopy study in d8-THF showed N plane).
that these two reagents do not interact, as shown in Fig. S7.‡
In C6D6 solution, 2 appears to maintain its structural com-
This suggests that pre-coordination of the carbene to the position as evidenced by the marginal downfield shift of the
metal is not necessary, but rather the free carbene is available HMDS methyl resonance from 0.20 in free Zn(HMDS)2 to
in solution to protect the sterically unencumbered metal 0.22 ppm in 2 as well as the shift of the methyl and backbone
once the reaction to produce the metal–hydride bond has resonances from 2.14 and 6.42 to 2.11 and 5.84 ppm. Further-
occurred. This means that the synthesis of clusters such as 1 more, there is the characteristic movement of the carbenic
or its Mg derivative can potentially be carried out in a more carbon resonance from 218.9 ppm in free IXy to 186.4 ppm
atom-efficient manner, using only half a molar equivalent of when coordinated (see Fig. S8 and S9‡ for full details).
NHC per metal without requiring the preformation of a coordi-
nation adduct between the carbene and the relevant metal DMAB under the same conditions described for the synthesis
bisamide.
of 1, led to the formation of an off-white solid completely in-
Interestingly, the reaction of the NHC complex 2 with
1
IPr would appear to be only slightly too sterically encum- soluble in C6D6, C7D8, or d8-THF; whereas H NMR analysis of
bered to access the Lewis acidic metal centre of Zn(HMDS)2, as the remaining filtrate from this reaction mixture showed the
moving to the moderately less bulky carbene IXy [1,3-bis(2,6- presence of free carbene IXy. These findings reveal that by
dimethylphenyl)imidazol-2-ylidene, which bears methyl rather replacing IPr by moderately less sterically demanding IXy,
than isopropyl arms at the ortho positions of the flanking aro- which coordinates to the Zn precursor to form 2, has a dra-
matic group] resulted in NHC adduct 2 being formed after stir- matic effect in the outcome of this reaction, suggesting that
ring the two together in hexane–THF at room temperature. under these conditions this NHC ligand may not be able to
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
stabilise the in situ formed putative mixed-hydride amido zinc with respect to the latter whose chelating monoanionic β-diketi-
intermediate.
minate ligand is robust and generally non-reactive.
1
We then turned our attention to the reaction of Zn(HMDS)2
The H and 13C NMR spectra of 3 in d8-THF are consistent
with the primary silane PhSiH3 rather than DMAB with the molecular structure. The terminal hydride resonance
(Scheme 1b). Equimolar amounts of the two reagents were appears at 3.20 ppm in d8-THF; cf complex A, 3.73 ppm;
stirred in hexane at 60 °C for 3 hours, again in the presence of complex I, 4.14 ppm (Fig. S10 and S11‡). Interestingly, using
initially non-interacting IPr. Upon cooling, a crop of crystals of the arene solvent C6D6, which a priori could be considered a
(HMDS)ZnH·IPr (3) were obtained in a 54% crystalline yield. 3 much more ‘innocent’ solvent due to its lesser coordinating
was characterized by multinuclear NMR spectroscopy (1H and ability, the solution behaviour of complex 3 is significantly less
13C) and CHN microanalysis. Interestingly, although, under straightforward (see Fig. 6). Focusing on the region around
1
the reaction conditions investigated, it appears that only one 6 ppm in the H NMR spectrum (chosen as the carbene back-
of the amido groups of the zinc precursor is activated towards bone resonance is found here as a clear singlet with no com-
the Si–H/Zn–HMDS exchange reaction, when the reaction is peting resonances nearby), it is clear that there are three
carried out using an excess of the silane (5 molar equivalents) distinct carbenes in solution (two of which appear to be co-
or using the more polar solvent THF, the formation of the ordinated to a metal, corroborated by the 13C NMR spectrum
homoleptic species [{IPrZnH2}2] (B)5b occurs preferentially.
which shows resonances at 191.7 and 180.8 ppm, indicative of
X-ray crystallographic studies established the molecular such coordination). By comparison with the spectra of some
structure of 3 (Fig. 5). Contrasting with that previously known species, we were able to identify two different known
described for tetranuclear cluster 1, this complex exhibits a species, namely Zn(HMDS)2 and free IPr (Fig. 6a). The identity
monomeric arrangement showing that the combined steric of the other two carbene-containing species was not instantly
bulk of HMDS and IPr are sufficient to protect and stabilize clear, although one set of carbene resonances was similar to
1
the three coordinate zinc centre.26
those reported for [ZnH2·IPr]2.5b A H DOSY NMR experiment
Complex 3 is a discrete mononuclear complex with no (Fig. 6b),27 which separates different components in solution
obvious intermolecular interactions in the solid state. The pro- according to their diffusion coefficient (and by extrapolation
nounced steric mismatch of the three ligands around zinc give their molecular weight/volume) in a manner which can be
it a heavily distorted trigonal planar environment. Surprisingly, considered as ‘NMR chromatography’, allowed us to defini-
the largest angle is not that between the two bulkiest ligands,
IPr and HMDS [N–Zn–C angle is 125.9(1)°] but rather that
between HMDS and hydride [126.2(1)°]. The Zn–H bond length
of 1.53(3) Å is similar to other terminal zinc–hydride bond dis-
tances such as those shown in Fig. 2 [A, 1.44(3); B, 1.53(2); C,
1.47(4)/1.56(4); D, 1.54(2); I, 1.61(3) Å] and close to another
mononuclear three-coordinate complex with a terminal Zn–H
bond, namely Harder’s [CH{C(Me)N(Dipp)}2ZnH] [1.46(2) Å].4c
The presence of a bulky, labile neutral donor and a non- chelat-
ing HMDS ligand may well confer greater latent reactivity on 3
Fig. 5 Molecular structure of complex (HMDS)ZnH·IPr (3) with 50%
probability ellipsoids. All hydrogen atoms other than zinc-bound hydride
have been removed for clarity. Selected bond parameters (Å and °): Zn1–
C1, 2.081(2); Zn1–H1A, 1.53(3); Zn1–N3, 1.924(2); C1–Zn1–N3, 125.9(1);
C1–Zn1–H1A, 107.9(1); N3–Zn1–H1A, 126.2(1).
Fig. 6 (a, top) 1H NMR spectrum and (b, bottom) 1H DOSY NMR spec-
trum of crystals of HZn(HMDS)·IPr (3) in C6D6 solution. Values in par-
entheses represented the theoretical molecular weight of the quoted
species.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
tively assign resonances to each of four components in the and stored over 4 Å molecular sieves. DMAB and PhSiH3 were
solution. A set of resonances corresponding to the molecular purchased commercially from Sigma-Aldrich and used as
structure (HMDS)ZnH·IPr was identified as the final IPr-con- received. (HMDS)2Zn,29 IXy and IPr30 were prepared by literature
taining species, leading us to propose that the following equili- methods. NMR spectra were recorded on a Bruker AV 400 MHz
brium (eqn (1)) is occurring in solution.28
spectrometer operating at 400.13 MHz for 1H, 100.62 MHz for
13C and 128.38 MHz for 11B. All 13C spectra were proton
decoupled. 1H and 13C spectra were referenced to the residual
solvent signal.
2ðHMDSÞZnHꢀIPr Ð ZnH2ꢀIPr þ ZnðHMDSÞ2 þ IPr
ð1Þ
This is a simple ligand redistribution from the heteroleptic
zinc compound to give two homoleptic compounds, albeit
with the presence of four components rather than three since
IPr does not coordinate to Zn(HMDS)2 (vide supra). Curiously,
in this particular equilibrium the dihydride species does not
seem to dimerize to give B, perhaps because the dynamic equi-
librium occurs on a faster timescale than the dimerization.
This is supported by the DOSY spectrum which suggests that
ZnH2·IPr (diffusion coefficient = 8.038 × 10−10 m2 s−1) weighs
only marginally more than IPr itself (diffusion coefficient =
8.495 × 10−10 m2s−1), and much less than 3 (diffusion coeffi-
cient = 5.776 × 10−10 m2s−1). Furthermore, only one resonance
at 5.06 ppm is noticed for the hydrides whereas Okuda’s
dimer B displays clear resonances for bridging and terminal
hydride ligands (at 2.23 and 3.57 ppm in C6D6). This reso-
nance is considerably removed from those in dimer B and
perhaps reflects the change from a four-coordinate zinc in the
dimer to a three-coordinate zinc in the monomer. Rivard
noted a broad hydride resonance for complex D at room temp-
erature although this resolved into two distinct resonances
below 0 °C.5f However, in both cases DOSY NMR spectroscopy
showed no hint of monomeric species, suggesting that their
fluxional processes go via retention of Zn–H–Zn bridges. A
monomer-dimer equilibrium was, however, proposed for
complex E on the basis of variable-temperature NMR spectro-
scopy.5c To further support our equilibrium hypothesis we
X-ray crystallography
Crystallographic data were measured on a Rigaku Saturn 724+
(1, using synchrotron radiation at beamline I19 of Diamond
Light Source) or Oxford Diffraction Gemini S laboratory based
diffractometer (2 and 3, using MoKα radiation). Structures
were solved and refined on F2 against all independent reflec-
tions by the full-matrix least-squares method using the
SHELXS and SHELXL programs.31 All non-hydrogen atoms
were refined using anisotropic displacement parameters.
Hydride scattering factors were used for the 6 bridging H
atoms in the core of complex 1. Selected crystallographic
details and refinement details are given in Table 1. CCDC
1042314 to CCDC 1042316 contain the supplementary crystallo-
graphic data for this paper.
Synthesis of [Zn4H6HMDS2·2IPr] 1
A Schlenk tube was charged with Zn(HMDS)2 (193 mg,
0.5 mmol) and IPr (97 mg, 0.25 mmol) to which 5 mL cyclo-
hexane was added. DMAB (22 mg, 0.37 mmol) was added to
this mixture with stirring at room temperature, resulting in a
brown homogeneous mixture within 5 minutes. After stirring
for 30 minutes at room temperature the stirring was stopped.
Crystals were visible within 5 hours (79 mg, 46%). 1H NMR
(400.13 MHz, 300 K, C7D8) δ (ppm) = 0.13 (s, 36H, Si(CH3)3),
1
3
0.98 (d, 24H, CH3, JHH = 6.8 Hz), 1.34 (d, 24H, CH3, 3JHH = 6.8
recorded a H NMR spectrum containing authentic [ZnH2·IPr]2,
3
Zn(HMDS)2 and IPr in C6D6 (that is, we approached the equili-
brium from the other side), which revealed a similar spectrum
to that previously seen for crystals of 3 (see Fig. S12‡ for
details). The presence of Zn(HMDS)2 is paramount to this de-
aggregation as it is not witnessed when only IPr is added to
[ZnH2·IPr]2 (Fig. S12a‡) but rather only occurs once the
Zn(HMDS)2 is added, evidenced by the loss of the two hydride
resonances of [ZnH2·IPr]2 and the concomitant development of
the hydride resonance of ZnH2·IPr (Fig. S12b‡).
Hz), 2.55 (sept, 8H, iPr-CH, JHH = 6.8 Hz), 2.85 (m, broad 1H,
Zn–H), 3.35 (m, broad, 4H, Zn–H), 4.14 (m, broad, 1H, Zn–H),
6.33 (s, 4H, imidazole backbone CH), 7.14 (d, 8H, m-CH, 3JHH
=
7.8 Hz), 7.28 (t, 4H, p-CH, JHH = 7.7 Hz). 13C{1H} NMR
(100.62 MHz, 300 K, C7D8) δ (ppm) = 6.5 (Si(CH3)3), 24.0 (iPr-
CH3), 25.4 (iPr-CH3), 29.0 (iPr-CH), 124.8 (m-CH), 124.9 (NHC-
CH), 131.2 (p-CH), 135.1 (i-C), 144.9 (o-CH), 181.0 (NHC-C).
Elemental analysis (%) for C66H114N6Si4Zn4: calcd: C 58.05, H
8.42, N 6.15; found: C 58.74, H 8.36, N 5.81. IR (nujol): The
ν(Zn–H) absorptions could not be unambiguously assigned.
3
Initial reactivity studies of 3 showed that this compound
reacts at room temperature with one molar equivalent of
DMAB in THF to yield [ZnH2·IPr]2.
Synthesis of Zn(HMDS)2·IXy (2)
To Zn(HMDS)2 (193 mg, 0.5 mmol) in 5 mL of n-hexane was
added IXy (138 mg, 0.5 mmol) and the mixture stirred for
30 minutes at room temperature, yielding a brown mixture.
THF was added dropwise until a solution was obtained
(approx. 0.3 mL). The solution was stirred for 15 minutes
Experimental
General experimental
All reactions and manipulations were performed under a pro- before cooling to −35 °C, giving orange crystals. Removing
tective argon atmosphere using either standard Schlenk tech- solvent in vacuo yielded a brown sticky solid presenting identi-
niques or a glove box. Hexane and THF were dried by heating cal NMR data (245 mg, 74%). 1H NMR (400.13 MHz, 300 K,
to reflux over sodium benzophenone ketyl and then distilled C6D6) δ (ppm) = 0.22 (s, 36H, Si(CH3)3), 2.11 (s, 12H, CH3),
under nitrogen prior to use. Cyclohexane was distilled over CaH2 5.87 (s, 2H, imidazole backbone CH), 6.92 (d, 4H, m-CH, 3JHH
=
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Table 1 Crystallographic data and refinement parameters or complexes 1, 2 and 3
1
2
3
Empirical formula
Mol. mass
Crystal system
Space group
a/Å
b/Å
c/Å
α/°
C
70H122N6OSi4Zn4
C33H60N4O0.5Si4Zn
698.58
Monoclinic
P21/c
21.938(2)
18.8738(14)
21.985(4)
90
C33H55N3Si2Zn
615.35
1437.57
Orthorhombic
Pna21
22.153(7)
16.412(5)
22.139(7)
90
Orthorhombic
P212121
11.9943(5)
13.8275(5)
21.8642(8)
90
β/°
90
90
8049(4)
4
119.304(15)
90
7938.4(14)
8
90
90
3626.2(2)
4
γ/°
V/Å3
Z
λ/Å
T/K
0.6889
100(2)
53 956
15 103
0.0611
12 203
1.053
0.71073
123(2)
69 144
16 583
0.1095
8990
1.033
0.71073
123(2)
13 071
7050
0.0311
5999
1.041
Measured reflections
Unique reflections
Rint
Observed rflns [I > 2σ(I)]
GooF
R [on F, obs rflns only]
wR [on F2, all data]
Largest diff. peak/hole e Å−3
0.0629
0.1767
2.74/−1.22
0.0700
0.1250
0.639/−0.472
0.0399
0.0760
0.387/−0.386
3
7.6 Hz), 7.02 (t, 2H, p-CH, JHH = 7.6 Hz). 13C{1H} NMR methylamino borane (DMAB) and silane PhSiH3. The former
(100.62 MHz, 300 K, C6D6) δ (ppm) = 7.0 (Si(CH3)3), 19.5 (CH3), affords tetranuclear cluster 1 which can be described as a co-
123.3 (imidazole backbone CH), 129.3 (m-CH), 129.9 (p-CH), complex of IPr·ZnH2 and (HMDS)ZnH and whose formation is
135.9 (o-CH), 137.6 (i-C), 186.4 (NHC-C). Elemental analysis likely to occur via a transient Zn amidoborane species, result-
(%) for C33H60N4O0.5Si4Zn: calcd: C 56.74, H 8.66, N 8.02; ing from the deprotonation of DMAB by the zinc amide precur-
found: C 57.14, H 9.28, N 8.16.
sor. 1H NMR monitoring of the reaction revealed the
formation of a single boron species, heteroleptic bis(amido)
borane [HB(NMe2)(HMDS)] along with H2 evolution. Although
the N-heterocyclic ligand IPr does not bind to Zn(HMDS)2, its
presence appears to be crucial for the stabilization of 1. Con-
trastingly, using silane PhSiH3 under harsher reaction con-
ditions, only one of the amido arms of the zinc precursor
undergoes Si–H/Zn–HMDS metathesis to form (HMDS)ZnH·IPr
(3), which exhibits a complex and intriguing behaviour in
benzene solutions. These findings illustrate the structural
diversity of this family of heteroleptic zinc hydride species as
well as their intricate solution chemistry, which can play an
important role when assessing their reactivity.
Synthesis of (HMDS)ZnH·IPr (3)
A
Schlenk tube was charged with Zn(HMDS)2 (193 mg,
0.5 mmol) and IPr (194 mg, 0.5 mmol) to which 15 mL
n-hexane was added. PhSiH3 (0.06 ml, 0.5 mmol) was added to
this mixture with stirring at room temperature. The mixture was
heated to 60 °C for 3 h, leading to a light brown homogeneous
mixture. The reaction was slowly cooled to room temperature,
producing a crop of light brown crystals (116 mg, 54%). 1H
NMR (400.13 MHz, 300 K, d8-THF) δ (ppm) = −0.32 (s, 18H,
3
Si(CH3)3), 1.08 (d, 12H, CH3, JHH = 6.7 Hz), 1.27 (d, 12H, CH3,
3JHH = 6.7 Hz), 2.60 (m, 3H, iPr-CH, JHH = 6.8 Hz), 3.20 (s, 1H,
3
Zn–H), 7.24 (d, 4H, m-CH, 3JHH = 7.6 Hz), 7.34 (t, 2H, p-CH, 3JHH
= 7.9 Hz), 7.35 (s, 2H, imidazole backbone CH). 13C{1H} NMR
(100.62 MHz, 300 K, d8-THF) δ (ppm) = 6.31 (Si(CH3)3), 24.3
(iPr-CH3), 25.3 (iPr-CH3), 29.4 (iPr-CH), 125.2 (m-CH), 126.4
(NHC-CH), 131.2 (p-CH), 135.9 (i-C), 145.6 (o-CH), 180.3 (NHC-
C). Elemental analysis (%) for C33H55N3Si2Zn: calcd: C 64.41,
H 9.01, N 6.83; found: C 64.48, H 9.43, N 6.54. IR (nujol): The
ν(Zn–H) absorption could not be unambiguously assigned.
Acknowledgements
The authors gratefully acknowledge the Royal Society of Edin-
burgh (BP Trust Fellowship to S.D.R.), the European Research
Council (ERC Starting Grant to E.H.), and the EPSRC (grant
award no. EP/L027313/1) for their kind sponsorship of this
research. We thank Diamond Light Source for access to syn-
chrotron facilities (beamline I19).
Conclusions
Access to two novel and distinct heteroleptic amido(hydride)
Zn species has been gained by reacting a mixture of the zinc
bis(amide) Zn(HMDS)2 and the unsaturated N-heterocyclic
carbene IPr with two different hydride sources, namely di-
Notes and references
1 S. Aldridge and A. J. Downs, Chem. Rev., 2001, 101, 3305–
3365.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
2 For an example of the synthesis and utilization of active
M. S. Hill, D. J. Liptrot, D. J. MacDougall and M. F. Mahon,
forms of binary alkali–metal hydrides in metalation and
Chem. Commun., 2011, 47, 9060–9062.
reduction reactions see R. Pi, T. Friedl, P. v. R. Schleyer, 13 (a) H. J. Cowley, M. S. Holt, R. L. Melen, J. M. Rawson and
P. Klusener and L. Brandsma, J. Org. Chem., 1987, 52,
4299–4303.
3 (a) L. Schlapbach and A. Züttel, Nature, 2001, 414, 353–358;
(b) F. Schüth, B. Bogdanovic and M. Felderhoff, Chem.
Commun., 2004, 2249–2258; (c) B. Sakintuna, F. Lamari-
D. S. Wright, Chem. Commun., 2011, 47, 2682–2684;
(b) M. M. Hansmann, R. L. Melen and D. S. Wright, Chem.
Sci., 2011, 2, 1554–1559; (c) R. J. Less, H. R. Simmonds,
S. B. J. Dane and D. S. Wright, Dalton Trans., 2013, 42,
6337–6343.
Darkrim and M. Hirscher, Int. J. Hydrogen Energy, 2007, 32, 14 (a) J. Spielmann, M. Bolte and S. Harder, Chem. Commun.,
1121–1140; (d) I. P. Jain, C. Lal and A. Jain, Int. J. Hydrogen
Energy, 2010, 35, 5133–5144.
2009, 6934–6936; (b) For an excellent overview of the
thermal decomposition of magnesium amido–borane com-
plexes see: J. Spielmann, D. F.-J. Piesik and S. Harder,
Chem. – Eur. J., 2010, 16, 8307–8318; (c) J. Spielmann and
S. Harder, Dalton Trans., 2011, 40, 8314–8319;
(d) S. Harder, J. Spielmann and J. Intemann, Dalton Trans.,
2014, 43, 14284–14290.
4 (a) H. Hao, C. Cui, H. W. Roesky, G. Bai, H.-G. Schmidt and
M. Noltemeyer, Chem. Commun., 2001, 1118–1119;
(b) S. P. Green, C. Jones and A. Stasch, Angew. Chem., Int.
Ed., 2008, 47, 9079–9083; (c) J. Spielmann, D. Piesik,
B. Wittkamp, G. Jansen and S. Harder, Chem. Commun.,
2009, 3455–3456; (d) S. Schulz, T. Eisenmann, 15 D. R. Armstrong, A. M. Drummond, L. Balloch,
D. Schuchmann, M. Bolte, M. Kirchner, R. Boese,
J. Spielmann and S. Harder, Z. Naturforsch., 2009, 64b,
D. V. Graham, E. Hevia and A. R. Kennedy, Organometallics,
2008, 27, 5860–5866.
1397–1400; (e) S. Harder, J. Spielmann, J. Intemann and 16 (a) M. Krieger, B. Neumüller and K. Dehnicke, Z. Anorg.
H. Bandmann, Angew. Chem., Int. Ed., 2011, 50, 4156–4160.
5 (a) M. Arrowsmith, M. S. Hill, D. J. MacDougall and
M. F. Mahon, Angew. Chem., Int. Ed., 2009, 48, 4013–4016;
(b) A. Rit, T. P. Spaniol, L. Maron and J. Okuda, Angew.
Chem., Int. Ed., 2013, 52, 4664–4667; (c) A. Rit,
T. P. Spaniol, L. Maron and J. Okuda, Organometallics,
2014, 33, 2039–2047; (d) A. Rit, T. P. Spaniol and J. Okuda,
Chem. – Asian J., 2014, 9, 612–619; (e) A. Rit, A. Zanardi,
T. P. Spaniol, L. Maron and J. Okuda, Angew. Chem., Int.
Ed., 2014, 53, 13273–13277; (f) P. A. Lummis,
Allg. Chem., 1998, 624, 1563–1564; (b) W. Marciniak,
K. Merz, M. Moreno and M. Driess, Organometallics, 2006,
25, 4931–4933; (c) B. Gutschank, S. Schulz, D. Bläser,
R. Boese and C. Wölper, Organometallics, 2010, 29, 6133–
6136; (d) For a related {Zn5H6} cluster see: M. P. Coles,
S. M. El-Hamruni, J. D. Smith and P. B. Hitchcock, Angew.
Chem., Int. Ed., 2008, 47, 10147–10150; (e) For a related
{Zn5H4} cluster see: M. Kahnes, H. Görls, L. González
and M. Westerhausen, Organometallics, 2010, 29, 3098–
3108.
M. R. Momeni, M. W. Lui, R. McDonald, M. J. Ferguson, 17 (a) P. L. Arnold, I. J. Casely, Z. R. Turner, R. Bellabarba and
M. Miskolzie, A. Brown and E. Rivard, Angew. Chem., Int.
Ed., 2014, 53, 9347–9351; (g) P. Jochmann and
D. W. Stephan, Angew. Chem., Int. Ed., 2013, 52, 9831–9835.
6 P. C. Andrikopolous, D. R. Armstrong, A. R. Kennedy,
R. B. Tooze, Dalton Trans., 2009, 7236–7247;
(b) D. R. Armstrong, S. E. Baillie, V. L. Blair, N. G. Chabloz,
J. Diez, J. Garcia-Alvarez, A. R. Kennedy, S. D. Robertson
and E. Hevia, Chem. Sci., 2013, 4, 4259–4266.
R. E. Mulvey, C. T. O’Hara and R. B. Rowlings, Eur. J. Inorg. 18 This chemical shift is close to that of the related species
Chem., 2003, 3354–3362.
7 D. J. Liptrot, M. S. Hill and M. F. Mahon, Chem. – Eur. J.,
2014, 20, 9871–9874.
(sBu2N)B(H)-HMDS which resonates at 35.2 ppm:
W. R. Nutt and R. L. Wells, Inorg. Chem., 1982, 21, 2473–
2476.
8 (a) A. Staubitz, A. P. M. Robertson, M. E. Sloan and 19 Although the formation of a Zn bis(amidoborane) inter-
I. Manners, Chem. Rev., 2010, 110, 4023–4078;
(b) A. Staubitz, A. P. M. Robertson and I. Manners, Chem.
Rev., 2010, 110, 4079–4124.
9 R. J. Less, R. L. Melen and D. S. Wright, RSC Adv., 2012, 2,
2191–2199.
10 H. C. Johnson, E. M. Leitao, G. R. Whittell, I. Manners,
G. C. Lloyd-Jones and A. S. Weller, J. Am. Chem. Soc., 2014,
136, 9078–9093 and references therein.
11 (a) J. Spielmann, G. Jansen, H. Bandmann and S. Harder,
Angew. Chem., Int. Ed., 2008, 47, 6290–6295; (b) S. Harder,
J. Spielmann and B. Tobey, Chem. – Eur. J., 2012, 18, 1984–
1991.
mediate, as the result of the reaction of Zn(HMDS)2 with 2
equivalents of DMAB, cannot be discounted, considering
the stoichiometry and mild reaction conditions employed
such a scenario sounds unlikely.
20 G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb,
A. Nudelman, B. M. Stoltz, J. E. Bercaw and K. I. Goldberg,
Organometallics, 2010, 29, 2176–2179.
21 11B NMR analysis of crudes and filtrates of the reaction
showed no indication of the formation of cyclic dimeric
species [R2NBH2]2 which have been previously observed as
the ultimate product in some metal-catalysed dehydro-
coupling processes of secondary aminoboranes, see for
examples ref. 12a and b.
12 (a) D. J. Liptrot, M. S. Hill, M. F. Mahon and
D. J. MacDougall, Chem. – Eur. J., 2010, 16, 8508–8515; 22 We note that during the preparation of this manuscript the
(b) M. S. Hill, M. Hodgson, D. J. Liptrot and M. F. Mahon,
Dalton Trans., 2011, 40, 7783–7790; (c) P. Belham,
first NHC adduct of Zn(HMDS)2 was reported, using ItBu
as the carbene. See: A. Baishya, M. K. Barman, T. Peddarao
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
and S. Nembenna, J. Organomet. Chem., 2014, 769, 112– 26 Similarly to the reactivity described for 2 and DMAB, when this
118.
NHC complex was treated with one equivalent of PhSiH3 under
the same conditions employed to prepare 3, an insoluble white
solid was formed which failed to dissolved in C6D6 or d8-THF.
23 A search of the CCDB for 3-coordinate zinc(II) bound to an
NHC yielded 23 Zn–C bond lengths in the range 1.981(6)
[(Ph2CHO)2Zn·IPr; P. Jochmann and D. W. Stephan, Chem. 27 D. Li, I. Keresztes, R. Hopson and P. G. Williard, Acc. Chem.
– Eur. J., 2014, 20, 8370–8378] to 2.118(5) Å [tBu2Zn·IPr; see
ref. 17b] with a mean value of 2.056 Å.
24 G. Margraf, H.-W. Lerner, M. Bolte and M. Wagner, Z.
Anorg. Allg. Chem., 2004, 630, 217–218.
Res., 2009, 42, 270–280.
28 Variable temperature NMR studies of isolated crystals of 3 in
C7D8 showed that the relative ratio of the different components
present in this equilibrium do not vary substantially over the
range of temperatures studied (220–310 K).
25 ItBu is reported as being more sterically demanding
than IPr in a high coordination number complex but 29 H. Bürger, W. Sawodny and U. Wannagat, J. Organomet.
less sterically demanding when the metal has a lower Chem., 1965, 3, 113–120.
coordination number, such as in Zn(HMDS)2·NHC: 30 L. Hintermann, Beilstein J. Org. Chem., 2007, 3, 22.
H. Clavier and S. P. Nolan, Chem. Commun., 2010, 46, 31 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-
841–861.
logr., 2007, 64, 112–122.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.