An unexpected transamination of bis[bis(trimethylsilyl)amido]zinc with dibenzylamine to form bis(dibenzylamido)zinc: Structural studies by NMR spectroscopy, X-ray crystallography and theoretical calculations

Article abstract of DOI:10.1039/b110696a

The transamination of bis[bis(trimethylsilyl)amido]zinc with two molar equivalents of dibenzylamine in benzene solution yields the dimeric, homoleptic, zinc bis(amide) [{(PhCH₂)₂N}₂Zn]₂·C ₆H₆ 1. Characterisation of compound 1 has been performed by single-crystal X-ray diffraction, ¹H/¹³C NMR spectroscopy, IR spectroscopy, melting point and elemental analysis. Variable concentration ¹H NMR spectroscopic studies have shown a dynamic monomer-dimer equilibrium in arene solution. Compound 1 is compared to the previously reported, isostructural magnesium analogue and other known zinc bis(amide) compounds. Theoretical calculations have been carried out at both SCF and DFT levels to probe the energetics involved in the transamination process.

Full text of DOI:10.1039/b110696a

View Article Online / Journal Homepage / Table of Contents for this issue
An unexpected transamination of bis[bis(trimethylsilyl)amido]zinc
with dibenzylamine to form bis(dibenzylamido)zinc: structural
studies by NMR spectroscopy, X-ray crystallography and
theoretical calculations
David R. Armstrong,a Glenn C. Forbes,*a Robert E. Mulvey,*a William Cleggb
and Duncan M. Tookeb
a Department of Pure and Applied Chemistry, University of Strathclyde, Glasgow, UK G1 1XL.
E-mail: r.e.mulvey@strath.ac.uk
b Department of Chemistry, University of Newcastle, Newcastle-Upon-Tyne, UK NE1 7RU
Received 22nd November 2001, Accepted 5th February 2002
First published as an Advance Article on the web 26th March 2002
The transamination of bis[bis(trimethylsilyl)amido]zinc with two molar equivalents of dibenzylamine in benzene
solution yields the dimeric, homoleptic, zinc bis(amide) [{(PhCH2)2N}2Zn]2ؒC6H6 1. Characterisation of compound
1 has been performed by single-crystal X-ray diffraction, 1H/13C NMR spectroscopy, IR spectroscopy, melting point
and elemental analysis. Variable concentration 1H NMR spectroscopic studies have shown a dynamic monomer–
dimer equilibrium in arene solution. Compound 1 is compared to the previously reported, isostructural magnesium
analogue and other known zinc bis(amide) compounds. Theoretical calculations have been carried out at both SCF
and DFT levels to probe the energetics involved in the transamination process.
bis(amides) over the past ten years, including other dimers:
Introduction
16
[(Ph2N)2Zn]2,14 [(Bui2N)2Zn]2,15 [{[Ph(Me3Si)]N}2Zn]2
and
The beginning of organometallic chemistry is usually associated
with the synthesis of diethylzinc1 by Frankland in 1849, the
intention of which was to prepare ethyl radicals from
iodoethane and zinc powder. However, it took a significant
amount of time thereafter until the α-halogenoester-derived
zinc enolates pioneered by Reformatsky2 became the first
synthetically useful organometallic reagents at the turn of the
20th century. Subsequently, these were superseded by the more
reactive and more easily prepared Grignard reagents.3 Organo-
zinc reagents received scant attention until recently with
moderate interest in, for example, Simmons–Smith type cyclo-
propanations4 and Reformatsky reactions.5 However, due to
their low reactivity, organozinc reagents are tolerant to many
more functional groups compared with their lithium and
magnesium counterparts and their synthetic utility is being
re-examined. Among the most useful examples are the cat-
alytic asymmetric addition to aldehydes6 and the nickel- and
palladium-catalysed cross-coupling reactions with organo-
halides.7 Additionally, organozinc compounds of the form
R2Zn, RZn(NRЈ2) or (R2N)2Zn also have potential applications
as p-type dopants in the semiconductor/electronics industry.8
We are interested in the synergic effects brought about by
the incorporation of compounds of the form (R2N)2M, where
M = Zn or Mg, in the same molecular environment as alkali
metal amides, (R2N)M, where M = Li, Na or K. Our group has
observed several examples of unprecedented synergism in these
heterobimetallic systems to date.9 Of most importance are
the regioselective deprotonation of ferrocene10 and of arene
molecules in thermodynamically unfavourable positions.11 Also
of interest is the selective encapsulation of oxide or peroxide.12
In our quest to extend this heterobimetallic work to new ligand
systems, we have focussed on the dibenzylamido ligand, as
previously this has been shown to display an interesting and
versatile structural chemistry.13
[{(Me2SiCH2CH2Me2Si)N}2Zn]2.17 A comparison can also be
made with the known isostructural magnesium derivative
[{(PhCH2)2N}2Mg]2 which has been studied both in the solid
state18 and solution.19 To the best of our knowledge, this allows
the first direct structural comparison between dimeric zinc and
magnesium bis(amides) in the solid state.
Results and discussion
The new homoleptic amide complex 1 was synthesised from
what is seemingly a routine transamination viz. the reaction of
bis[bis(trimethylsilyl)amido]zinc with dibenzylamine in benzene
(eqn. (1)). Crystallisation at ambient temperature gave com-
pound 1 as the only solid found to crystallise from solution.
One benzene solvent molecule per dimer crystallises in the
lattice.
There is no precedent for this specific transamination in
amido chemistry. Conventionally, one would normally expect
Here we report the synthesis and characterisation of a new
homoleptic zinc bis(amide) [{(PhCH2)2N}2Zn]2ؒC6H6 1. This
follows the structural characterisation of a number of zinc
1656
J. Chem. Soc., Dalton Trans., 2002, 1656–1661
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b110696a

View Article Online
bis(trimethylsilyl)amine to displace a dibenzylamido unit from
a metal centre (e.g. zinc), based purely on the relative thermo-
dynamic acidities of the NH protons on each of the parent
amines.20 Here we find the opposite i.e. dibenzylamine displaces
bis(trimethylsilyl)amide. A possible driving force for this
transamination reaction is an increase in the coordination
number at the zinc centre. Due to the inherent steric bulk of the
trimethylsilyl-substituents, the coordinatively unsaturated zinc
bis(amide) [{(Me3Si)2N}2Zn] is monomeric21 in the gas phase,
(SiMe3)2}x], dibenzylamine was added to [{(Me3Si)2N}2Zn] in
a 1 : 1 stoichiometric ratio in benzene solution. However, no
formation of the heteroleptic complex was observed and only
the homoleptic compound 1 was produced. It can be inferred
from this that the target heteroleptic molecule is unstable
with respect to disproportionation. This can be accounted
for by the steric bulk of bis(trimethylsilyl)amido units, which
may preclude dimerisation in this case and, therefore, no
concomitant increase in coordination number is realised.
Compound 1 has a similar melting point to its magnesium
analogue (mp 172–173 ЊC for 1 vs. 175–177 ЊC for [{(PhCH2)2-
N}2Mg]2). This is expected because melting points in com-
pounds of this type are dictated by van der Waals interactions
between the organic groups on the molecular periphery.
1
although its solid state structure is yet to be determined. H
NMR spectroscopic evidence supports a monomeric structure
in solution, since a singlet is observed at 0.20 ppm in C6D6 at
300 K, showing all silyl protons to be equivalent. The closely
related compounds [{[But(Me3Si)]N}2Zn],22 [{(Ph2MeSi)2-
N}2Zn]23 and [(But2N)2Zn]15 are all similarly monomeric (in the
solid state). Conversely, the less sterically demanding nature
of the silylamido-ligand 2,2,5,5-tetramethyl-2,5-disila-1-aza-
cyclopentanide (mean Si–N–Si angle 109Њ cf. 129Њ in [{(Me3Si)2-
The lipophilic benzyl groups in 1 shield the polar core, thus
enforcing appreciable solubility of 1 in arene solvents. This
enabled a detailed NMR spectroscopy study to be carried out.
1
The H NMR spectrum of 1 run in C6D5CD3 at 300 K shows
N}2Zn]), present in [{(Me2SiCH2CH2Me2Si)N}2Zn]2,17 permits
formation of a dimeric structure. The electronics of 2,2,5,5-
tetramethyl-2,5-disila-1-aza-cyclopentanide are not too far
removed from those of bis(trimethylsilyl)amide and therefore
steric effects must play a greater role in determining the
aggregation state of these zinc bis(amides) than electronic
considerations. In compound 1, transamination of the bis-
(trimethylsilyl)amido-ligand by the less sterically demanding
dibenzylamido-functionality leads to a higher aggregate and
higher coordination saturation at the zinc centre.
two sharp signals of equal integral in the benzylic region at 3.00
ppm and 4.07 ppm. The benzylic protons of the parent dibenzyl-
amine appear at 3.55 ppm in the same solvent. An interesting
point is the marked difference between the chemical shifts of
the two resonances (1.07 ppm) in the spectrum of the metal-
lated product 1. A similar situation is observed in the 1H NMR
spectrum (also in C6D5CD3) of the dimeric magnesium ana-
logue [{(PhCH2)2N}2Mg]2.19 These resonances can be assigned
to bridging and terminal dibenzylamido units. Normally, the
higher frequency signal should correspond to the proton with
the greatest electron deficiency viz. the bridging group. Con-
versely, lower frequency resonance should relate to the amido
function with the least electron deficiency i.e. the terminal
group, but this is by no means always the case, as shown by
Westerhausen.24 He notes that in [{(Me3Si)2N}2M]2, where M =
Mg, Ca or Sr, the location of signals corresponding to bridging
and terminal amido groups follows a much more regular
In order to determine if a similar transamination process
would occur on replacement of zinc by magnesium, two molar
equivalents of dibenzylamine were added to a benzene solution
24
containing one molar equivalent of [{(Me3Si)2N}2Mg]2
(eqn. (2)). Since [{(Me3Si)2N}2Mg]2 exists as a dimer–monomer
1
pattern for 13C signals than H signals. In the case of Mg, the
signals for both the bridging Si(CH3)3 resonances (0.45 ppm)
and the bridging Si(CH3)3 resonances (8.12 ppm) appear at a
higher frequency than the respective terminal group resonances
(0.38 ppm and 7.14 ppm, respectively). Conversely, in the
derivatives with M = Ca or Sr, the positions of the 1H bridging
and terminal signals are reversed i.e. when M = Ca, the bridging
resonance is at 0.21 ppm and the terminal resonance at 0.33
ppm; when M = Sr, the bridging and terminal resonances are
0.12 ppm and 0.33 ppm respectively. In both cases the order of
the corresponding 13C signals are the same as those in the Mg
compound.
The 13C NMR spectrum of 1 run in the same solvent at 300 K
shows the presence of two distinct types of benzylic carbon at
53.5 ppm and 58.2 ppm. From a 1H/13C HMQC experiment, the
(larger) peak at 58.2 ppm was shown to couple to both sets
of benzylic protons. The peak at 53.5 ppm is assigned to the
benzylic carbon of the monomer (see later). The majority of the
phenyl-carbons appear as two sets of signals between 126.4
ppm and 130.1 ppm, which are obscured by solvent signals.
However, the two distinct resonances at 141.0 ppm and 144.7
ppm are assigned to the ipso-carbons of the terminal and bridg-
ing dibenzylamido units, respectively. These data are consistent
with the structure determined by X-ray diffraction. In addition,
the 1H NMR spectrum of 1 shows two relatively small but
significant resonances at 3.56 ppm and 3.58 ppm. These signals
are tentatively assigned to the benzyl protons of the two-
coordinate monomeric complex [{(PhCH2)2N}2Zn] 2. The ab
initio geometry optimised structure of the related monomer
[{(PhCH2)2N}2Mg]19 shows that one benzyl group from each
amido-moiety is attracted to the metal centre, thus, rendering
the CH2 groups inequivalent. It is possible that the structure of
2 resembles this and similarly we would expect inequivalence
between these CH2 groups in this case. However, mitigating
against this, it would be expected that the benzyl groups would
equilibrium in arene solution, in the aforementioned reaction
transamination would afford no increase in coordination at the
metal centre (for the dimer). Thus, it would not be expected that
the transamination would take place. After heating the solution
to reflux in benzene for two hours, an aliquot of the reaction
1
solution was taken and the solvent was removed in vacuo. H
NMR spectral evidence of the residue in the benzyl region
lacked the presence of dibenzylamide ligands, but showed a
solitary singlet corresponding to the parent dibenzylamine.
Thus, as expected, the transamination was unsuccessful and the
‘product’ from acidity considerations prevailed. In addition,
during a similar control ‘reaction’ involving equimolar amounts
of [{(Me3Si)2N}2Li]25 and dibenzylamine, no transamination
was observed.
Until now, we have only considered homoleptic amides.
Although alkylzinc(amide) complexes are well known,26 to the
best of our knowledge, there are no heteroleptic zinc bis(amide)
structures available for comparison. In an effort to form the
as yet unknown heteroleptic complex [{(PhCH2)2NZnN-
J. Chem. Soc., Dalton Trans., 2002, 1656–1661
1657

View Article Online
Table 1 Selected bond lengths (Å) and angles (Њ) for [{(PhCH2)2N}2-
Zn]2ؒC6H6 1
be undergoing a fast rotation process at ambient temperature
and therefore they would show a time-averaged equivalency. An
alternative possibility that the geminal benzyl hydrogen atoms
are inequivalent (diastereotopic) should lead to a pair of
doublets. While such is not observed here, this possibility
cannot be completely ruled out, as they may be isochronous due
to accidental equivalence. Cases of accidental equivalence
for prochiral groups are common.27 It is also known that the
outer lines of two mutually coupled doublets can collapse into
the baseline in cases (as here) where the coupling constant (2J)
approaches or is greater than the chemical shift difference (∆ν,
in Hz) between the distinct resonances.
Zn(1)–N(1)
Zn(1)–N(1Ј)
Zn(1)–N(2)
N(1)–C(1)
C(1)–C(2)
N(1)–C(8)
2.0083(13) C(8)–C(9)
2.0620(12) N(2)–C(15)
1.8530(14) N(2)–C(22)
1.4895(19) C(22)–C(23)
1.517(2)
1.462(2)
1.453(2)
1.529(3)
1.520(0)
1.508(2)
C(15)–C(16)
1.4924(19)
N(1)–Zn(1)–N(2)
N(1)–Zn(1)–N(1Ј)
N(1Ј)–Zn(1)–N(2)
Zn(1)–N(1)–Zn(1Ј)
C(1)–N(1)–Zn(1)
C(1)–N(1)–Zn(1Ј)
C(8)–N(1)–Zn(1)
C(8)–N(1)–Zn(1Ј)
141.95(6)
C(22)–N(2)–C(15)
C(1)–N(1)–C(8)
N(1)–C(1)–C(2)
N(1)–C(8)–C(9)
C(22)–N(2)–Zn(1)
N(2)–C(22)–C(23)
C(15)–N(2)–Zn(1)
N(2)–C(15)–C(16)
112.81(13)
109.78(12)
109.38(12)
114.77(13)
124.93(11)
115.94(14)
122.14(11)
114.61(14)
93.12(5)
124.87(6)
86.88(5)
113.83(10)
113.68(9)
118.39(10)
112.70(9)
Variable temperature NMR spectroscopic studies on com-
pound 1 were also carried out (in C6D5CD3), but these revealed
1
little information. During a high temperature H NMR spec-
troscopy study ranging from 300–373 K, the terminal and
bridging benzyl proton resonances failed to coalesce. Since the
corresponding resonances in [{(PhCH2)2N}2Mg]2 coalesce at
368 K, this reflects the greater strength of the Zn–N bond over
the corresponding Mg–N bond in this system (i.e. the former
bonds are more difficult to cleave). A low temperature 13C
NMR spectroscopy study was also undertaken in order to
assess if the signal at 58.2 ppm would split to reveal resonances
for both the terminal and bridging benzylic carbons.
Unfortunately, loss of all benzylic carbon signals at subambient
temperatures was observed, possibly due to precipitation of the
analyte.
Symmetry operations for equivalent atoms Ϫx ϩ 1, Ϫy ϩ 1, Ϫz ϩ 1.
It appears that the dimeric structure of 1 fails to remain
wholly intact in arene solution and a dynamic monomer–dimer
equilibrium exists. This assignment of the peaks to the
monomer 2 was confirmed by a variable concentration study
(over the range 5–20 mg mlϪ1) as can be seen from Fig. 1. This
Fig. 2 The centrosymmetric dimeric molecule of 1 (without solvent or
H atoms). Ring atoms are numbered sequentially.
shown in Table 1. As far as we are aware, this is the first
structural characterisation of a dimeric zinc bis(amide) where a
corresponding dimeric magnesium analogue is available for
comparison. In compound 1, the coordination arrangement is
based on a rigorously planar (ZnN)2 core (crystallographic
inversion symmetry) and, unsurprisingly, the same structural
motif is shared by other zinc bis(amides).14–16 Each zinc bonds
to three (two bridging and one terminal) dibenzylamido ligands
and adopts a pseudo-trigonal planar geometry (mean N–Zn–N
angle 119.98Њ). Likewise, the terminal nitrogen atoms occupy
distorted trigonal planar environments, whilst the bridging
nitrogens are pseudo-tetrahedral. Notably, the steric require-
ments of the dibenzylamido ligand prevent the zinc from
expanding its coordination number from three to four as is
common in zinc bis(dialkylamides)28 and required for a chain
like that observed in a (ZnN)2 polymer. Steric congestion
almost certainly precludes tetra-coordination at zinc to form a
(ZnN)2 polymer chain, which has previously been observed for
magnesium in the lithium amidomagnesiate [{(PhCH2)2N}4-
MgLi2].18 Here, the formally two-coordinate Li centres in the
heterobimetallic complex each draw two of the four bridging
dibenzylamido ligands away from the central magnesium
centre, thus reducing steric hindrance at the magnesium centre.
The Zn–N bridging distances in 1 have a mean value of 2.035
Å and these are not significantly different from the correspond-
Fig. 1 Variable concentration [5 mg mlϪ1(bottom)–20 mg mlϪ1(top)]
1H NMR spectra of 1 in C6D6, clearly showing the dynamic monomer–
dimer equilibrium.
14
15
ing bridging distances in [(Ph2N)2Zn]2 and [(Bui2N)2Zn]2
which are 2.033 Å and 2.028 Å, respectively. However, the
sterically demanding nature of the silylamido ligands in
[{[Ph(Me3Si)]N}2Zn]2 16 results in a marginally larger mean Zn–
N bridging distance of 2.052 Å. The mean Zn–N bridging dis-
tance in 1 is also significantly larger than the terminal bond
distance (1.853 Å), implying that the terminal amido ligands
bind more strongly to the zinc centres. This is entirely expected,
considering the number of metal centres that the amido ligand
was carried out in C6D6 as this afforded greater solubility of 1.
A six-fold increase in the monomer : dimer ratio is observed on
dilution of the solution to 25% of its initial value. A com-
parable dynamic equilibrium is also observed for [{(PhCH2)2-
N}2Mg]2 in C6D5CD3.19
The solid state structure of 1, reveals a centrosymmetric
dimer (Fig. 2). Selected bond angles and bond lengths are
1658
J. Chem. Soc., Dalton Trans., 2002, 1656–1661

View Article Online
spans in each case. The d-electrons in zinc (i.e. Zn2ϩ) do not
favour the formation of π-complexes. In line with this, there
are no short contacts between the metal centres and phenyl-
substituents of the bridging dibenzylamido ligands. The closest
centroid [of ring C(2)–C(7)] to the zinc centre Zn(1) is 3.843 Å
and the nearest ring atom is C(2), at a distance of 3.146 Å.
However, short contacts between the benzyl groups of dibenzyl-
amido ligands and metal centres, in particular lithium, have
been observed previously.13 Comparing 1 to dimeric [Ph2Zn]2,29
it can be seen that much shorter (Ph)C–Zn contacts are present
in the bis(alkyl)zinc compound, whereby the phenyl groups
bridge the two zinc centres asymmetrically, giving two short,
2.006(5) and 2.016(3) Å, and two long, 2.364(5) and 2.442(4) Å,
Zn–C bridging distances.
Table 2 Calculated energies (Hartrees) of the model compounds at
the SCF and DFT levels
Total Energy E (Eh)
Model compound
SCF
DFT
[{(Me3Si)2N}2Zn]
[{(PhCH2)2N}2Zn]
[{(PhCH2)2N}2Zn]2
[{(Me3Si)2N}2Mg]
(PhCH2)2NH
Ϫ3517.725140
Ϫ2963.027849
Ϫ5926.077823
Ϫ1939.844353
Ϫ593.336109
Ϫ870.67181
Ϫ3525.857144
Ϫ2972.480531
Ϫ5944.990648
Ϫ1946.835584
Ϫ597.265016
Ϫ873.947456
(Me3Si)2NH
Table 3 Comparison of bond lengths (Å) and bond angles (Њ) in the
model [{(Me3Si)2N}2Zn] with those of the magnesium congener
[{(Me3Si)2N}2Mg]
Electrostatic (and steric) repulsion between the amido units
(and between the two zinc atoms) causes the endocyclic N–Zn–
N bond angle to mildly contract in comparison to the Zn–N–
Zn bond angle [i.e. 86.88(5)Њ for N–Zn–N cf. 93.12(5)Њ for
Zn–N–Zn]. The following zinc bis(amides) also share these
characteristics: [(Ph2N)2Zn]2,14 [(Bui2N)2Zn]2,15 [{[Ph(Me3Si)]2-
[{(Me3Si)2N}2Zn]
[{(Me3Si)2N}2Mg]
M–N
N–Si
Si–C
1.855
1.731
1.893–1.902
1.914
1.720
1.894–1.908
16
17
N}2Zn]2 and [{(Me2SiCH2CH2Me2Si)N}2Zn]2 with corre-
sponding N–Zn–N and Zn–N–Zn bond angles: [88.14(7)Њ vs.
91.86(7)Њ], [86.63(9)Њ vs. 93.37(9)Њ], [88.74(7)Њ vs. 91.26(7)Њ] and
[88.05(8)Њ vs. 91.92(9)Њ] respectively.
M–N–M
M–N–Si
Si–N–Si
180.0
116.3
127.5
179.9
115.5
129.1
When we compare zinc and magnesium, they are found
to have covalent radii of 1.20 Å and 1.45 Å, respectively.30 In
addition, zinc has a higher electronegativity in comparison to
magnesium: the Pauling values are 1.65 and 1.31, respectively.31
These data suggest that Zn–N bonds are stronger and more
covalent than Mg–N bonds. The Zn–N bridging and terminal
bond distances are found to be in fairly close agreement to the
analogous Mg–N bond distances (i.e. mean distances: Mg–N
bridging 2.088 Å, Mg–N terminal 1.935 Å in [{(PhCH2)2-
N}2Mg]2 cf. Zn–N bridging 2.035 Å, Zn–N terminal 1.853 Å
in 1). These systematic variations are attributed to the higher
electronegativity of zinc over magnesium.
Table 4 Comparison of the dimensions in the model compound
[{(PhCH2)2N}2Zn]2 with those in the crystal structure of 1
Model [{(PhCH2)2N}2Zn]2
Compound 1
Zn–Nbr
Zn–Nt
2.063
1.893
2.035
1.853
Zn–N–Zn
N–Zn–N
86.5
93.5
86.9
93.1
br = bridging, t = terminal.
Other comparisons between zinc and magnesium bis(amides)
are restricted to isomorphous compounds in the monomeric
state. Similar differences in bond distances can be seen in the
vapour phase structures of [{(Me3Si)2N}2Zn]21 and [{(Me3Si)2-
N}2Mg]32 which have M–N bond distances of 1.824 Å and
1.935 Å, respectively and in the solid state structures of
[{(Ph2MeSi)2N}2Zn]23 and [{(Ph2MeSi)2N}2Mg]33 which have
mean M–N bond distances of 1.852 Å and 1.968 Å, respect-
ively. In order to shed light on the energetics involved in the
transamination procedure and to compare the structural data
obtained from the X-ray diffraction study, model compounds
were subjected to geometry optimisation calculations34 carried
out at the SCF level using the 6-31G* basis set.35 These opti-
mised geometries were in turn subjected to single point DFT36
energy calculations using the B3LYP method37 and the 6-31G*
basis set. All of the reaction components in eqn. (3) and (4)
structure is considered (eqn. (4)), the reaction becomes
exothermic (using DFT data), albeit only moderately so. This
clearly suggests that the ability of bis(dibenzylamido)zinc to
dimerise is a key factor in the transamination process (note
that for two monomers to generate a dimer, the energy released
in the formation of new bonds must outweigh the loss of
translational entropy). Table 3 compares the principal dimen-
sions of the model [{(Me3Si)2N}2Zn] monomer with that of its
magnesium congener [{(Me3Si)2N}2Mg]. Most significantly,
this reveals that the Zn–N bond length (1.855 Å) is decidedly
shorter (and, by implication, stronger) than the corresponding
Mg–N bond (length, 1.914 Å).
This size diminution (0.086 Å) which has its origin in the
greater covalency of zinc–organoelement bonds versus their
magnesium counterparts (see above) proves that the steric
demands of the bulky silylamide ligand have a greater influence
on zinc than magnesium. This is consistent with the fact that in
the crystalline state [{(Me3Si)2N}2Mg] is dimeric whereas the
zinc counterpart remains a monomer (in the hypothetical zinc
dimer the amide ligands would approach each other more
closely than they do in the magnesium case). Table 4 lists the
principal dimensions of the model dimer [{(Ph2CH2)2N}2Zn]2.
Observing the data presented in Table 4, we are able to com-
pare how well the model [{(Ph2CH2)2N}2Zn]2 reflects the
structure of experimentally observed compound 1. It can be
seen that the bond lengths and bond angles in both sets of data
are in close agreement. The mean Zn–N bridging and terminal
distances for the model compound vary by less than 2.2% from
the experimentally observed structural data. The endocyclic
Zn–N–Zn and N–Zn–N bond angles observed in the model
compound differ by less than 0.5% from those found in
compound 1.
[{(Me3Si)2N}2Zn] ϩ 2 (PhCH2)2NH
[{(PhCH2)2N}2Zn] ϩ 2 (Me3Si)2NH ∆Ef =
ϩ7.4 (ϩ10.8) kcal molϪ1 (3)
[{(Me3Si)2N}2Zn] ϩ 2 (PhCH2)NH
1/2[{(PhCH2)2N}2Zn]2 ϩ 2 (Me3Si)2NH ∆Ef =
Ϫ1.9 (ϩ4.1) kcal molϪ1 (4)
were independently optimised in this way and their total
energies computed (Table 2) thus allowing the energy of form-
ation (∆Ef in kcal molϪ1) of each model reaction to be
calculated (DFT values are shown along with SCF values).
Eqn. (3), which disregards the aggregation of the bis(dibenzyl-
amido)zinc product and treats it as a monomer, reveals that the
reaction is significantly endothermic at both levels of theory.
However, when the more experimentally realistic dimeric
J. Chem. Soc., Dalton Trans., 2002, 1656–1661
1659

View Article Online
In conclusion, we have shown that [{(PhCH2)2N}2Zn]2ؒC6H6
1 can be synthesised from an unconventional transamination
of [{(Me3Si)2N}2Zn] with dibenzylamine in benzene. Crystalline
1 is dimeric in the solid state as shown by X-ray structural
analysis, but forms a dynamic monomer–dimer equilibrium
α = 67.811(2), β = 73.637(2), γ = 65.482(2)Њ, U = 1298.97(15) Å3,
Z = 1, Dc = 1.271 g cmϪ3, µ = 0.97 mmϪ1 (Mo-Kα, λ = 0.71073
Å), T = 160 K; 10926 measured reflections, 5747 unique (Rint
=
0.017, θ < 28.2Њ, semi-empirical absorption correction); R (F,
F 2 > 2σ) = 0.029, Rw (F 2, all data) = 0.075, goodness of fit = 1.05
for 307 refined parameters, final difference map extremes ϩ0.87
and Ϫ0.26 e ÅϪ3. Hydrogen atoms were constrained. The unit
cell contains one dimeric molecule of the complex and one
molecule of benzene, both lying on inversion centres. There is
no disorder. Programs used were standard Bruker SMART
(data collection), SAINT (integration) and SHELXTL
(structure determination),40 together with local programs.
CCDC reference number 175760.
1
in arene solution, as established by a variable H NMR con-
centration study. Compound 1 has also been shown to be
isostructural to other zinc bis(amides) and its magnesium
derivative [{(PhCH2)2N}2Mg]2.
Experimental
General procedures
See http://www.rsc.org/suppdata/dt/b1/b110696a/ for crystal-
lographic data in CIF or other electronic format.
All reactions were carried out using standard Schlenk tech-
niques38 with products stored in an argon-filled glove box.
Bis[bis(trimethylsilyl)amido]zinc was synthesised according to
Bürger’s method39 and characterised by its 1H NMR spectrum.
Dibenzylamine was distilled from CaH2 and stored over
molecular sieves. Benzene was freshly distilled from Na/
benzophenone, rigorously degassed and stored over molecular
sieves prior to use. Compound 1 has been characterised
by single-crystal X-ray crystallography, NMR and IR spectro-
scopy, melting point and elemental analysis.
Acknowledgements
We gratefully acknowledge the financial support of the EPSRC
(through grant award no. GR/M78113 and for equipment
funding to W.C.). Our thanks are also extended to Dr K. W.
Henderson for helpful discussions.
References
Physical measurements
1 (a) E. Frankland, Ann. Chem Pharm., 1849, 71, 171; (b) for
an excellent review of Frankland’s work, see D. Seyferth,
Organometallics, 2001, 20, 2940.
2 S. N. Reformatsky, Ber. Dtsch. Chem. Ges., 1887, 20, 1210.
3 H. Richey, Jr, Grignard Reagents: New Developments, Wiley,
Chichester, 2000.
4 (a) K.-P. Zeller and H. Gugel, in Houben-Weyl, Methoden der
Organischen Chemie, ed. M. Regitz, G. T. Verlag, Stuttgart, 1989,
vol. E 19b, p. 195; (b) J. Furukawa and N. Kawabata, Adv.
Organomet. Chem., 1974, 12, 83; (c) H. E. Simmons, T. L. Cairns,
S. A. Vladuchick and C. M. Hoines, Org. React., 1973, 201.
5 (a) A. Fürstner, Synthesis, 1989, 571; (b) M. W. Rathke, Org. React.,
1975, 22, 423; (c) M. Gaudemar, Organomet. Chem. Rev., Sect. A,
1972, 8, 183; (d ) P. L. Shriner, Org. React, 1942, 1, 1.
1H and 13C NMR spectral data were recorded on a Bruker DPX
400 MHz spectrometer operating at 400.13MHz for H and
1
100.62 MHz for 13C. 1H and 13C chemical shifts are given
relative to external SiMe4. C, H and N analyses were performed
using a Perkin-Elmer 2400 elemental analyser. The IR spectra
were obtained from Nujol mulls between NaCl plates using a
Nicolet Avatar 360 FT-IR spectrometer. Melting point studies
were carried out in sealed capillaries using an electrothermal
melting point apparatus and are uncorrected.
Preparation of [{(PhCH2)2N}2Zn]2ؒC6H6 1
6 (a) R. Noyori and M. Kitamura, Angew. Chem., Int. Ed. Engl.,
1991, 30, 49; (b) H. Takahashi, T. Kawakita, M. Ohno, M. Yoshioka
and S. Kobayashi, Tetrahedron, 1992, 48, 5691; (c) D. Seebach,
A. K. Beck, B. Schmidt and Y. M. Wang, Tetrahedron, 1994,
50, 4364; (d ) L. Schwink, S. Vettel and P. Knochel, Organometallics,
1993, 14, 5000; (e) L. Pu and H.-B Yu, Chem Rev., 2001,
101, 3.
7 (a) M. Kobayashi and E. Negishi, J. Org. Chem., 1980, 45, 5223; (b)
E. Negishi, V. Bagheri, S. Chatterjee, F.-T Luo, J. A. Miller and
A. T. Stoll, Tetrahedron Lett., 1983, 24, 5181; (c) R. A. Grey, J. Org.
Chem., 1984, 49, 2288; (d ) R. Rossi, F. Bellina, A. Carpita and
R. Gori, Synlett., 1995, 344; (e) E. Negishi, A. O. King and
N. Okukado, J. Org. Chem., 1977, 42, 1821; ( f ) E. Negishi,
H. Matsushitu, M. Kobayashi and C. L. Rand, Tetrahedron Lett.,
1983, 24, 3824.
To
a solution containing bis[bis(trimethylsilyl)amido]zinc
(10 mmol, 3.86 g) and benzene (10 ml) was added dibenzyl-
amine (20 mmol, 3.80 ml) at ambient temperature. The result-
ant clear, deep red solution was stirred for 18 hours, by which
time a colourless solid had precipitated. Recrystallisation of
the solid from benzene yielded cube-shaped crystals of 1 (5.8
mmol, 2.90 g). Yield 58%. mp 172–173 ЊC (Found: C, 73.9; H,
6.3; N, 5.7%. Calculated for C56H56N4Zn2ؒC6H6: C, 74.9; H, 6.4;
N 5.6%). IR (cmϪ1) 2725s, 2674s, 1350m, 1342m, 1319w, 1298w,
1131w, 1053m, 1005w, 953m, 760m, 750m, 741m, 725m, 701m.
1H NMR (δ) in C6D6 (5 mg mlϪ1): 3.04 (4H, s, PhCH2 dimer),
3.56 (1.2H, s, PhCH2 monomer), 3.58 (1.2H, s, PhCH2
monomer), 4.14 (4H, s, PhCH2 dimer), 6.96–7.52 (26H, series
of multiplets, Ph). 1H NMR (δ) in C6D6 (20 mg mlϪ1): 3.04 (4H,
s, PhCH2 dimer), 3.56 (0.2H, s, PhCH2 monomer), 3.58 (0.2H,
s, PhCH2 monomer), 4.14 (4H, s, PhCH2 dimer), 7.01–7.52
(21H, series of multiplets, Ph). 13C NMR (δ) in C6D6 (20 mg
mlϪ1): 58.6 (PhCH2), 126.9–130.4 (series of overlapping signals,
Ph), 141.4 (ipso-C terminal), 145.1 (ipso-C bridging). C6H6
solvent of crystallisation obscured by overlapping phenyl
signals and deuterated solvent. 1H NMR (δ) in C6D5CD3: 3.00
(4H, s, PhCH2 dimer), 3.54 (0.4H, s, PhCH2 monomer), 3.56
(0.4H, s, PhCH2 monomer), 4.07 (4H, s, PhCH2 dimer), 6.98–
7.46 (29H, series of multiplets, Ph). 13C NMR (δ) in C6D5CD3:
53.5 (PhCH2, monomer), 58.2 (PhCH2, dimer), 126.4–130.1
(series of overlapping signals, Ph), 141.0 (ipso-C terminal),
144.7 (ipso-C bridging). C6H6 solvent of crystallisation
obscured by overlapping phenyl signals and deuterated solvent.
8 (a) M. A. Malik, P. O’Brien, M. Montevalli and A. C. Jones, Inorg.
Chem., 1997, 36, 5076; (b) M. A. Malik and P. O’Brien, Polyhedron,
1997, 16, 3593; (c) W. S. Rees, Jr, O. Just, H. Schumann and
R. Weimann, Polyhedron, 1998, 17, 101.
9 R. E. Mulvey, Chem. Commun., 2001, 1049.
10 W. Clegg, K. W. Henderson, A. R. Kennedy, R. E. Mulvey,
C. T. O’Hara, R. B. Rowlings and D. M. Tooke, Angew. Chem., Int.
Ed., 2001, 40, 3902.
11 (a) P. C. Andrews, A. R. Kennedy, R. E. Mulvey, C. L. Raston,
B. A. Roberts and R. B. Rowlings, Angew. Chem., Int. Ed., 2000,
11, 2036; (b) D. R. Armstrong, A. R. Kennedy, R. E. Mulvey and
R. B. Rowlings, Angew. Chem., Int. Ed., 1999, 38, 131.
12 G. C. Forbes, A. R. Kennedy, R. E. Mulvey, R. B. Rowlings,
W. Clegg, S. T. Liddle and C. C. Wilson, Chem. Commun., 2000,
1759.
13 (a) P. C. Andrews, D. R. Armstrong, R. E. Mulvey and D. Reed,
J. Am. Chem. Soc., 1988, 110, 5235; (b) D. R. Armstrong, R. E.
Mulvey, G. T. Walker, D. Barr, R. Snaith, W. Clegg and D. Reed,
J. Chem. Soc., Dalton. Trans., 1988, 617; (c) D. Barr, W. Clegg,
R. E. Mulvey and R. Snaith, J. Chem. Soc., Chem. Commun., 1984,
285, 287.
X-Ray crystallography
Crystal data for 1: C56H56N4Zn2ؒC6H6, M = 993.9, triclinic,
space group P1, a = 10.1530(7), b = 12.1247(8), c = 12.6655(9) Å,
14 M. A. Putzer, A. Dashti-Mommertz, B. Neumuller and
K. Dehnicke, Z. Anorg. Allg. Chem., 1998, 624, 263.
¯
1660 J. Chem. Soc., Dalton Trans., 2002, 1656–1661

View Article Online
15 H. Schumann, J. Gottfriedsen and F. Girgsdies, Z. Anorg. Allg.
Chem., 1997, 623, 1881.
16 H. Schumann, J. Gottfriedsen, S. Dechert and F. Girgsdies,
Z. Anorg. Allg. Chem., 2000, 626, 747.
17 O. Just, D. A. Gaul and W. S. Rees, Jr, Polyhedron, 2001, 20,
815.
18 W. Clegg, K. W. Henderson, R. E. Mulvey and P. A. O’Neil, J. Chem
Soc., Chem. Commun., 1994, 769.
29 P. R. Markies, G. Schat, O. S. Akkerman, F. Bickelhaupt, W. J. J.
Smeets and A. L. Spek, Organometallics, 1990, 9, 2243.
30 (a) J. E. Huheey, E. A. Keiter and R. C. Keiter, Inorganic Chemistry:
Principles of Structure and Reactivity, Harper-Collins, New York,
4th edn., 1993, p. 292; (b) M. C. Ball and A. H. Norbury, Physical
Data for Inorganic Chemists, Longman, London, 1974, pp. 139–144.
31 A. L. Allred, J. Inorg. Nucl. Chem., 1961, 17, 1.
32 M. Westerhausen and W. Schwartz, Z. Anorg. Allg. Chem., 1992,
609, 39.
33 R. A. Bartlett, M. M. Olmstead and P. P. Power, Inorg. Chem., 1994,
33, 4800.
19 W. Clegg, F. J. Craig, K. W. Henderson, A. R. Kennedy,
R. E. Mulvey, P. A. O’Neil and D. Reed, Inorg. Chem., 1997, 36,
6238.
20 (a) R. R. Fraser, T. S. Mansour and S. Savard, J. Org. Chem., 1985,
50, 3232; (b) J. J. Christensen, R. M. Izatt, D. P. Wrathall and
L. D. Hansen, J. Chem. Soc. A, 1969, 1212; (c) J. J. P. Furlong,
E. S. Lewkowitcz and N. S. Nudelman, J. Chem. Soc, Perkin Trans.
2, 1990, 1461 and references therein.
34 Gaussian 98, Revision A. 9, M. J. Frisch, G. W. Trucks, H. B.
Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G.
Zakrzewski, J. A. Montgomery, Jr, R. E. Stratmann, J. C. Burant,
S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C.
Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi,
B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A.
Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D.
Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz,
A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith,
M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe,
P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L. Andres,
C. Gonzalez, M. Head-Gordon, E. S. Replogle, J. A. Pople,
Gaussian, Inc., Pittsburgh, PA, 1998.
21 K. Haaland, K. Hedberg and P. P. Power, Inorg. Chem., 1984, 23,
1972.
22 W. S. Rees, Jr, D. M. Green and W. Hesse, Polyhedron, 1992, 11,
1697.
23 P. P. Power, K. Ruhlandt-Senge and S. C. Shoner, Inorg. Chem.,
1991, 30, 5013.
24 M. Westerhausen, Inorg. Chem., 1991, 30, 96.
25 R. D. Rogers, J. L. Atwood and R. Grüning, J. Organomet. Chem,
1978, 157, 229.
26 (a) M. M. Olmstead, W. J. Grigsby, D. R. Chacon, T. Hascall and
P. P. Power, Inorg. Chim. Acta, 1996, 251, 273; (b) L. M. Engelhardt,
G. E. Jacobsen, W. C. Patalinghug, B. W. Skelton, C. L. Raston and
A. H. White, J. Chem. Soc., Dalton. Trans., 1989, 2859; (c) M. A.
Malik, P. O’Brien, M. Motavalli and A. C. Jones, Inorg. Chem.,
1997, 36, 5076; (d ) N. A. Bell, H. M. M. Shearer and C. B. Spencer,
Acta Crystallogr., Sect. C, 1983, 39, 1182; (e) M. Westerhausen,
T. Bollwein, A. Pfitzner, T. Nilges and H. Deiseroth, Inorg. Chim.
Acta, 2001, 312, 239; ( f ) M. Westerhausen, T. Bollwein,
N. Makropoulos, T. M. Rotter, T. Habereder, M. Suter and H. Nöth,
Eur. J. Inorg. Chem., 2001, 851.
35 (a) K. W. J. Hehre, R. Ditchfield and J. A. Pople, J. Chem. Phys.,
1972, 56, 2257; (b) P. C. Hariharan and J. A. Pople, Theor. Chim.
Acta, 1973, 28, 213; (c) J. D. Dill and J. A. Pople, J. Chem. Phys.,
1975, 62, 292; (d ) V. A. Rassolov, J. A. Pople, M. A. Ratner and
T. L. Windus, J. Chem. Phys., 1998, 109, 1223.
36 W. Kohn and L. J. Sham, Phys. Rev., 1965, 140, A1133.
37 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
38 D. F. Shriver and M. A. Drezdon, Manipulation of Air-Sensitive
Compounds, Wiley, New York, 2nd edn., 1986.
39 H. Bürger, W. Sawodny and U. Wannagat, J. Organomet. Chem.,
1965, 13, 3.
27 W. B. Jennings, Chem. Rev., 1975, 75, 307.
28 M. F. Lappert, A. R. Sanger and R. C. Srivastava, Metal and
Metalloid Amides, Ellis Horwood, Chichester, 1980, pp. 545–566.
40 SMART, SAINT and SADABS, Siemens Analytical X-Ray
Systems, Inc., Madison, WI, 1996; G. M. Sheldrick, SHELXL-97,
University of Göttingen, Germany, 1997.
J. Chem. Soc., Dalton Trans., 2002, 1656–1661
1661