Lithium and aluminium carbamato derivatives of the utility amide 2,2,6,6-tetramethylpiperidide

Article abstract of DOI:10.1039/c0dt00118j

Insertion of CO₂ into the metal-N bond of a series of synthetically important alkali-metal TMP (2,2,6,6-tetramethylpiperidide) complexes has been studied. Determined by X-ray crystallography, the molecular structure of the TMEDA-solvated Li derivative shows a central 8-membered (LiOCO)₂ ring lying in a chair conformation with distorted tetrahedral lithium centres. While trying to obtain crystals of a THF-solvated derivative, a mixed carbonato/carbamato dodecanuclear lithium cluster was formed containing two central (CO₃)^2- fragments and eight O ₂CTMP ligands with four distinct bonding modes. A bisalkylaluminium carbamato complex has also been prepared via two different methods (CO ₂ insertion into a pre-formed Al-N bond and ligand transfer from the corresponding lithium reagent) which adopts a dimeric structure in the solid state.

Full text of DOI:10.1039/c0dt00118j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Lithium and aluminium carbamato derivatives of the utility amide
2,2,6,6-tetramethylpiperidide†‡
Alan R. Kennedy, Robert E. Mulvey,* David E. Oliver and Stuart D. Robertson*
Received 11th March 2010, Accepted 19th May 2010
First published as an Advance Article on the web 2nd June 2010
DOI: 10.1039/c0dt00118j
Insertion of CO2 into the metal–N bond of a series of synthetically important alkali-metal TMP
(2,2,6,6-tetramethylpiperidide) complexes has been studied. Determined by X-ray crystallography, the
molecular structure of the TMEDA-solvated Li derivative shows a central 8-membered (LiOCO)2 ring
lying in a chair conformation with distorted tetrahedral lithium centres. While trying to obtain crystals
of a THF-solvated derivative, a mixed carbonato/carbamato dodecanuclear lithium cluster was formed
containing two central (CO3)2- fragments and eight O2CTMP ligands with four distinct bonding modes.
A bisalkylaluminium carbamato complex has also been prepared via two different methods (CO2
insertion into a pre-formed Al–N bond and ligand transfer from the corresponding lithium reagent)
which adopts a dimeric structure in the solid state.
its low nucleophilicity, which is superior to that of the common
Introduction
alkyllithium reagents nBuLi or tBuLi. These properties of LiTMP
are further enhanced by high solubility in ethereal and hydrocar-
bon solvents, its ready preparation from commercially available
reagents [nBuLi and the parent amine TMP(H)] and relative ease
of handling with respect to alkyllithium reagents.
One of the most important reactions in chemistry is deprotonative
metallation, where a relatively inert C–H bond is replaced with a
more reactive C–M bond (where M is typically an alkali-metal).
Of the reagents capable of effecting such a transformation, alkali-
metal secondary amides (MNR2) are among the most widely uti-
lized. Sterically demanding lithium 2,2,6,6-tetramethylpiperidide
(LiTMP) is one of the most popular such utility amides. First
reported as a base in 1972,1 LiTMP is favoured by synthetic
chemists due to its blend of relatively strong Brønsted basicity,
slightly inferior to that of alkyllithium reagents, coupled with
With reactivity closely correlated to structure, there has
also been widespread interest in the solid-state structures of
alkali-metal utility amides including LiTMP. When pure in the
crystalline state, LiTMP adopts a cyclotetrameric molecular
structure (Scheme 1), which was elucidated by Lappert and
Atwood in 1983.2 In the presence of the popular bidentate
donor N,N,N¢,N¢-tetramethylethylenediamine (TMEDA) this
crystalline form changes to a hemisolvated ‘open dimer’.3 These
alkali-metal amide structures are highly dependent on the identity
of the alkali-metal, since the unsolvated sodium derivative is
a smaller cyclotrimer4 and the TMEDA-solvated sodium and
potassium derivatives5 are more conventional cyclic (MN)2 dimers
with a donor molecule capping each metal. The donor-deficient
WestCHEM, Department of Pure and Applied Chemistry, University of
Strathclyde, Glasgow, G1 1XL, UK. E-mail: r.e.mulvey@strath.ac.uk
† Dedicated to Professor David W. H. Rankin on the occasion of his
retirement for his many outstanding contributions to molecular structural
chemistry.
‡ CCDC reference numbers 769482–769484. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/c0dt00118j
Scheme 1
6190 | Dalton Trans., 2010, 39, 6190–6197
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
open dimer structure is presumably preferred so as to alleviate
the strain which a four-membered Li2N2 ring would experience;
the closed dimer would also bring the bulky diamine and amide
moieties into too close proximity as a consequence of shorter Li–N
bonds.
Despite the picture being clear with regards to the structures of
these compounds in the solid-state, the solution-state structures
are less clear cut. Lucht and Collum have, however, shown through
6Li and 15N NMR spectroscopy in hydrocarbon solvent that the
unsolvated lithium derivative consists of a complicated mixture of
four distinct cyclic tetramers along with a cyclic trimer,6 while the
TMEDA-solvated congener exists as both the open dimer and a
monomer.7
(A – Scheme 2), which contains both bridging and terminal
carbamate groups, was prepared from reaction of AlCl3 with
HNiPr2 in the presence of CO2 25 and was subsequently converted
to the tetranuclear m3-oxo derivative [Al4(m3-O)2(O2CNiPr2)8] by
stoichiometric hydrolysis with H2O.26 No¨th et al. fortuitously ob-
tained the 1 : 2 species [MeAl(O2CTMP)2]2 (B) 27 and the 2 : 3 ionic
species [(tBuAl)2(O2CTMP)3]+[(tBu3Al)2Br]- (C) 28 when trying to
recrystallise the parent bisalkylaluminium amide using dry ice as
a cooling source, while Chang et al. obtained [Me2Al(O2CNR2)]2
i
(R = Et, Pr, D) as one of the products upon bubbling CO2
through the mixed metal cluster [Me2Al(m-NR2)2Mg(m-Me)]x
i
(R = Et, x = 2; R = Pr, x = 4).29 Interestingly, reaction of
(TMP)2AlCl with excess CO2 resulted in selective insertion into
only one Al–N bond to give the alkylamido aluminium carbamate
[(TMP)(O2CTMP)AlCl]2 (E).30
TMP is also an important amide in aluminium chemistry,
particularly in the guise of bisalkylaluminium amido reagents.
Amongst others, Yamamoto et al. have found utility for these
reagents in a variety of organic transformations such as the
regio/stereospecific isomerisation of epoxides to allylic alcohols,8
conversion of epoxy silyl ethers to 1,2-diols,9 addition of an
ester/ketone enolate to an aldehyde,10 diastereoselective acetal
cleavage11 and Fischer indole synthesis.12 These organo-aluminium
compounds have recently enjoyed a renaissance in the flourishing
field of mixed-metalate chemistry,13 where the juxtaposition of
iBu2Al(TMP) with an alkali-metal TMP reagent can effect the
usually difficult a-C deprotonation of TMEDA or PMDETA
(N,N,N¢,N¢¢, N¢¢-pentamethyldiethylenetriamine) resulting in a
new N–CH2–Al bond.14 Indeed, this reagent has even been applied
in the deprotonation of a methyl group of KTMP to give the
erstwhile unknown TMP2- dianion (N–H, CH3 deprotonated).15
Closely related to metal amides are the metal carbamate
family of compounds. Alkali-metal carbamates (MO2CNR2) are
of interest in industrial chemistry as a convenient precursor to
isocyanates16 and are useful intermediates in areas as diverse as
agricultural17 and medicinal chemistry.18 They can be synthesised
in various ways such as by reaction of an alkali-metal tetraphenylb-
orate salt,19 or the alkali-metal itself,20 with a solution of the
desired amine in the presence of CO2. This facile approach of
inserting a CO2 molecule into a metal–nitrogen bond has been well
documented for transition metal carbamates but has seen much
less attention for the alkali-metal analogues.21 Carbon dioxide
is a useful reagent due to its ease of handling and widespread
availability. Indeed, CO2 insertion to produce carbamates is the
current synthetic route of choice given that the original method
involved both the use of highly toxic phosgene and the production
of HCl as a by-product. Katritzky and co-workers have published a
series of papers using CO2 as a reagent for simultaneous protection
of nucleophilic centres and the activation of alternative locations
to nucleophilic attack,22 while Otero et al. have inserted CO2 into
organolithium species to yield lithium carboxylate species.23 Alu-
minium carbamates have seen marginally more interest than their
alkali-metal counterparts, due in part to the ease with which CO2
can be inserted into the Al–N bond. The first example involved
CO2 insertion into Et2AlNEt2, a reaction whose rate was enhanced
by the presence of a donor such as TMEDA.24 Aluminium
carbamates with a 1 : 1, 1 : 2, 1 : 3 or 2 : 3 aluminium : carbamate
ratio have recently been crystallographically characterised, with
the recurring structural motif in these complexes being the
(AlOCO)2 ring with the carbamate functionality acting as a bridge
between metal atoms. The carbamate rich species Al2(O2CNiPr2)6
Scheme 2 Crystallographically characterised aluminium carbamates. *
Fragment is the cationic moiety of an ion-pair with a [(tBu3Al)2Br]-
counter-anion.
Considering the widespread interest in alkali-metal amides it
is surprising to learn that only two solid-state structures are
known for alkali-metal carbamates, neither being derived from
an important utility alkali-metal amide. Snaith et al. reported
the TMEDA-solvated structure of the diphenylamido carbamate
LiO2CNPh2 31 and the THF-solvated structure of an indole-based
32
lithium carbamate as a dimer and tetramer respectively, both
products being obtained by inserting CO2 into a pre-formed Li–
N bond. LiO2C(TMP) has been prepared previously and found
application as a carbamate source in a metathesis reaction to yield
a bisamido gallium carbamate species, however, beyond elemental
analysis and an IR spectrum, no characterisation of this lithium
TMP carbamate was carried out.33 Given the structural diversity of
alkali-metal amides (vide supra) we set our sights on addressing the
paucity of alkali-metal carbamates based on utility amides. Due to
our burgeoning interest in the chemistry of diisobutyl aluminium
TMP we also studied its carbamato derivative, both via insertion
of CO2 into the Al–N bond of iBu2AlTMP and also by using the
lithium TMP carbamate as a ligand transfer agent and report our
findings herein.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6190–6197 | 6191
©

View Article Online
Scheme 3
2-Na did not give suitable NMR spectra in non-donating
Results and discussion
solvents, therefore a d8-THF solution was consequently studied.
While there was still a noticeable amount of undissolved solid
visible in the NMR tube, 2-Na was sufficiently soluble for useful
spectra to be obtained. This suggests that THF has almost
certainly caused de-aggregation of at least some of the original
Following established protocols, the congeneric alkali-metal TMP
salts (1) were prepared by deprotonation of the parent secondary
n
n
amine with BuLi, BuNa and KCH2SiMe3 respectively. These
amides were converted to the corresponding carbamates (2) by
bubbling gaseous CO2 (from a Schlenk flask containing dry ice)
through a hexane solution of 1 via a cannula (Scheme 3). In each
case a white solid precipitated, which was conveniently collected
by Schlenk filtration and washed with hexane.
1
polymer. The H NMR spectrum again shows a minor amount
of TMP(H). Resonances representing the carbamate are slightly
broader than those of the parent amine, with the b and g protons
appearing at a coincidental chemical shift of 1.53 ppm. What is
noticeable here is that in the presence of the Lewis donating THF,
the reversal of resonance chemical shifts as seen in 2-Li is not
witnessed with the methyl group now upfield at 1.45 ppm. Again,
the CO2 fragment is witnessed in the 13C NMR spectrum with a
chemical shift of 166.6 ppm. Unfortunately, NMR spectra of 2-K
could not be obtained due to its exceptionally poor solubility.
Having determined that carbon dioxide had inserted into the
metal–nitrogen bonds to produce some type of polymeric carba-
mate, we attempted to prepare lower aggregates by introducing
the bidentate Lewis donor TMEDA to a hexane suspension of 2.
This only caused dissolution in the case of 2-Li and even then the
solution was partially cloudy. Cooling of this solution to -30 ◦C
yielded a crop of colourless crystals in a 66% isolated yield. A
single crystal was subsequently subjected to an X-ray diffraction
study. Crystallographic data are listed in Table 1.
These unsolvated products proved to be largely insoluble in
non (lone-pair) donating solvents such as benzene and toluene
suggesting that they are polymeric in nature. However, 2-Li did
1
produce weak H and 13C NMR spectra in C7D8 solution. In
addition to anticipated resonances, these spectra showed others
attributable to TMP(H), however, given the relative insolubility
of the carbamate we believe that the TMP(H) is merely a minor
1
impurity rather than a hydrolysis product. Significantly, the H
NMR spectrum shows a reversal in the expected or normal order
of the TMP resonances with the methyl groups now at highest
frequency (1.74 ppm) and the g hydrogen atoms at the lowest
frequency (1.54 ppm). This reversal was witnessed previously in
Linti’s gallium carbamate.33 All the carbon centres are accounted
for in the 13C NMR spectrum, with the most significant being
that in the CO2 group which resonates at 168.2 ppm, a typical
value for a carbamate. For example this functionality in a series
of functionalized pyrrolidine lithium carbamates resonates at
approximately 158 ppm 34 while that of a Mg carbamate tetramer
resonates at 164 ppm.35
The solid-state structure (Fig. 1) shows that 3-Li exists as a
cyclodimer lying about an inversion centre, with a central eight-
membered (LiOCO)2 ring, sharing similar parameters with that
seen previously by Snaith et al. for [LiO2CNPh2·TMEDA]31 and
Table 1 Crystallographic data and refinement details for compounds 3-Li, 4 and 5
3-Li
4
5
Empirical formula
Mol. mass
Crystal system
C32H68Li2N6O4
614.80
Monoclinic
C110H200Li12N8O29
2182.06
Triclinic
¯
P1
12.7355(7)
16.5468(7)
17.0019(9)
73.910(4)
68.784(5)
89.123(4)
3194.7(3)
1
C36H72Al2N2O4
650.92
Monoclinic
Space group
P21/c
P21/n
˚
a [A]
9.9473(7)
8.4476(6)
22.304(2)
90
94.700(6)
90
1867.9(2)
2
1.093
9.2755(3)
15.4153(5)
13.9819(4)
90
99.805(3)
90
1970.0(1)
2
1.097
˚
b [A]
˚
c [A]
a [◦]
b [◦]
g [◦]
V [A ]
Z
3
˚
rcalcd [Mg m-3]
Measured reflections
Unique reflections [I > 2s(I)]
GooF on F2
1.134
30 301
12 862
34 265
4360 [Rint = 0.0405]
12 497 [Rint = 0.0454]
4924 [Rint = 0.0349]
0.887
1.079
1.067
R1, wR2 [I > 2s(I)]
R1, wR2 [all data]
Largest diff. peak/hole [eA ]
0.0445, 0.0912
0.0904, 0.0992
0.208/-0.177
0.0849, 0.2313
0.1184, 0.2490
0.881/-0.422
0.0435, 0.1238
0.0624, 0.1300
0.419/-0.407
-3
˚
6192 | Dalton Trans., 2010, 39, 6190–6197
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Fig. 1 Molecular structure of the lithium TMP carbamate [(TMPCO2) Li·TMEDA]2 (3-Li) with selective atom labelling. Hydrogen atoms are omitted
for clarity and thermal ellipsoids are displayed at 50% probability level. TMP and TMEDA fragments were modelled as disordered over two positions in
a 77 : 23 and 61 : 39 ratio respectively; only the major component is shown. Symmetry transformations used to generate equivalent atoms: 1 - x, 1 - y,
◦
˚
1 - z. Selected bond lengths (A) and angles ( ): Li(1)–O(1A) 1.881(2), Li(1)–O(2) 1.880(2), C(1)–O(1) 1.261(2), C(1)–O(2) 1.255(2), C(1)–N(3) 1.401(2);
O(1)–C(1)–O(2) 124.3(1), O(1)–C(1)–N(3) 118.0(1), O(2)–C(1)–N(3) 117.7(1), C(1)–O(1)–Li(1A) 132.1(1), C(1)–O(2)–Li(1) 127.2(1), O(2)–Li(1)–O(1A)
134.1(1).
also with that of an ether-solvated lithium carboxylate complex.36
As emphasised in Fig. 2, the carbon and oxygen atoms lie in the
same plane, with the two lithium centres positioned above and
non-donating solvents. Consequently its NMR spectra had to be
collected in d8-THF. The 1H NMR spectrum was similar to that of
2-Na in THF, with the b and g protons appearing at a coincidental
chemical shift of 1.52 ppm. This resonance, along with that
representing the methyl groups of the TMP appendage, were
deshielded with respect to those of TMP(H). Two resonances were
witnessed for TMEDA, with the ethylene backbone at a higher
frequency (2.30 ppm) than the methyl hydrogens (2.15 ppm),
consistent with uncoordinated TMEDA37 and suggesting that
THF has displaced TMEDA in the solution structure (indeed 2-Li
in d8-THF displays an identical 1H spectrum). Integration values
showed that TMEDA was present in a 1 : 1 ratio with respect to
the TMP carbamate, consistent with the solid-state structure. Of
most significance in the 13C NMR spectrum was the CO2 peak at
167.2 ppm. All the other carbon atoms displayed resonances at
similar chemical shifts to those in 2-Li and 2-Na.
˚
below this plane at either end [deviation of 0.585(2) A], such that
the ring lies in a chair conformation. This deviation from planarity
is presumably largely a consequence of the lone pairs on oxygen.
Lying in a distorted tetrahedral environment, the lithium atoms
have their coordination sphere completed by the nitrogen atoms
of a chelating TMEDA molecule. The carbamate carbon atoms
are three-coordinate, trigonal planar (R∠C = 360.0◦). As is often
the case, the TMP and TMEDA skeletons are disordered and a
detailed description of their parameters is unwarranted.
We also attempted to grow crystals of the THF-solvated
derivative so as to carry out a true comparison on the effect
of the donor on aggregation state of the lithium carbamate. A
suspension of 2-Li in hexane had THF added until it turned
clear and was then chilled to -30 ◦C. After several weeks, a
few crystals were evident. X-Ray analysis (Fig. 3) showed that
8
rather than a THF solvate, this structure was actually [Li12(m8h -
CO3)2(O2CTMP)8·4THF] (4); a dodecanuclear lithium carbamate
cluster containing a pair of central carbonato CO3 groups. The
organic periphery of this structure was heavily disordered and
therefore a detailed analysis of the structure is unwarranted,
however a number of salient features regarding the central core
were evident.
Fig. 2 Central core of 3-Li.
Presumably this unexpected product was formed as a conse-
quence of adventitious oxidation. The structure is best described
Despite TMEDA causing the unsolvated lithium carbamate to
deaggregate to a dimeric species, 3-Li still proved to be insoluble in
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6190–6197 | 6193
©

View Article Online
8
Fig. 3 Molecular structure of [Li12(m8h -CO3)2(O2CTMP)8·4THF] (4) with selective atom labelling. Hydrogen atoms and carbon atoms of THF and TMP
fragments are omitted for clarity and thermal ellipsoids are disp◦layed at 50% probability level. Symmetry transformations used to generate equivalent
˚
atoms: 1 - x, 1 - y, -z. Selected bond lengths (A) and angles ( ): C(1)–O(1) 1.292(3), C(1)–O(2) 1.269(3), C(1)–O(3) 1.2903(3), Li(1)–O(1) 1.970(5),
Li(2)–O(2) 1.923(5), Li(3)–O(2) 1.908(5), Li(4)–O(3) 1.938(5), Li(5)–O(3) 1.947(5), Li(6)–O(1) 2.080(5), Li(2A)–O(3) 2.085(5), Li(3A)–O(1) 2.196(5);
O(1)–C(1)–O(2) 121.1(2), O(1)–C(1)–O(3) 118.6(2), O(2)–C(1)–O(3) 120.3(2).
as a pair of antiparallel trigonal planar carbonate units (as a
consequence of an inversion centre) each surrounded by six lithium
atoms (R∠C = 360.0◦). Each carbonate oxygen atom coordinates
to two of these six lithium atoms [range of Li–O bond lengths =
Li(1), Li(1A), Li(4) and Li(4A) also have THF completing their
coordination sphere. This results in all the lithium atoms being
in a distorted tetrahedral LiO4 arrangement, with the exception
of Li(5) which is LiO3 distorted trigonal planar. An unusual and
highly interesting feature is that the O2CTMP units display three
of the seven different bridging modes proposed by Chang et al.
˚
1.908(5) to 2.080(5) A] while O(1) and O(3) also show interactions
˚
to a lithium atom of the adjacent plane [Li(3A)–O(1) = 2.196(5) A;
2
3
Li(2A)–O(3) = 2.085(5) A]. The bonding in the CO3Li6 unit is con-
(Fig. 4).39 Specifically, these are m2h between Li(5) and Li(6), m3h
˚
4
sistent with that witnessed by Hyva¨rinen et al. in their carbonate-
centred lithium cluster {[Li4(m3-Cl)(m6-OSiMe2OMe2SiO)(m-
between Li(1), Li(2) and Li(4A) and m4h between Li(3), Li(4),
Li(1A) and Li(6A) and between Li(3), Li(6A), Li(2) and Li(5A).
4
HMPA)(HMPA)3]3(m3-Cl)(m9-CO3)}·2THF [Li–O = 1.92(2) to
The m4h mode of bridging can be further sub-divided in
38
˚
2.08(2) A]. The lithium atoms in 4 are consequently held
complex 4 since one of the bonding modes involves each oxygen
together by a framework of eight bridging O2CTMP units while
atom bonding to two lithium atoms from the same plane while
Fig. 4 Unique bridging modes found in the carbamate anions found in 4.
6194 | Dalton Trans., 2010, 39, 6190–6197 This journal is The Royal Society of Chemistry 2010
©

View Article Online
the other involves each oxygen atom bonding to one lithium atom
from each plane such that the carbamato groups are essentially
perpendicular to one another. This essentially results in four
distinct bonding modes, two more than in the pentanuclear Mg
carbamato cluster reported by Chang et al. The Li–O(carbamato)
distances vary considerably, however a definite trend is evident.
The shortest interactions occur between those oxygen atoms
which bind only to one lithium atom, that is both from the
2
3
˚
m2h carbamate [1.857(6) and 1.820(6) A] and one from the m3h
˚
carbamate [1.891(5) A]. Those bonds between a bridging oxygen
˚
and lithium fall in the range 1.911(6) to 1.974(6) A.
The fact that we only obtained a small amount of this fortuitous
product, coupled with its poor solubility, meant that we were
unable to obtain any further analysis other than a weak 1H NMR
spectrum. Coordinated THF was evident from two appropriate
resonances at chemical shifts of 3.57 and 1.42 ppm. The carbamate
resonances were broad, which is to be expected since four
chemically inequivalent but closely related O2CTMP groups are
present in the crystal structure. The methyl groups resonated at
1.37 ppm with broad signals at 1.32 and 0.92 ppm tentatively
assigned to the b and g hydrogen atoms respectively.
Next, we extended our attention to bisalkylaluminium carba-
mates. Of most interest to us was the diisobutyl TMP derivative
due to its recent utilisation in mixed metallate chemistry (vide
supra). We determined that two different paths could be taken to
the desired species (Scheme 4).
Fig. 5 Molecular structure of the diisobutyl aluminium TMP carbamate
[iBu2Al(O2CTMP)]2 (5) with selective atom labelling. Hydrogen atoms are
omitted for clarity and thermal ellipsoids are displayed at 50% probability
level. C(12) was modelled as disordered over two positions in a 66 : 34
ratio; only the major component is shown. Symmetry transformations
used to generate◦equivalent atoms: 1 - x, -y, 1 - z. Selected bond lengths
˚
(A) and angles ( ): Al(1)–O(1A) 1.827(1), Al(1)–O(2) 1.826(1), C(1)–O(1)
1.284(2), C(1)–O(2) 1.282(2), C(1)–N(1) 1.347(2), Al(1)–C(11) 1.974(2),
Al(1)–C(15) 1.975(2); O(1)–C(1)–O(2) 120.0(1), O(1)–C(1)–N(1) 119.9(1),
O(2)–C(1)–N(1) 120.1(1), C(1)–O(1)–Al(1A) 129.0(1), C(1)–O(2)–Al(1)
132.4(1), O(2)–Al(1)–O(1A) 107.03(5), C(11)–Al(1)–C(15) 120.22(7).
indicate delocalised bonding while the C–N bond of the carbamate
˚
[1.347(2) A] is considerably shorter than a typical C–N single
Scheme 4
bond [CN single bonds of TMP in 5 are 1.526(2) and 1.528(2)
˚
A]. The carbamate carbon atom lies in an almost perfect trigonal
The first path involved insertion of CO2 into the Al–N bond;
a circumstance which has been previously reported.24–25,27–29 The
bisalkylaluminium amide was prepared by a metathesis reaction
using iBu2AlCl and LiTMP, filtered to remove LiCl and then CO2
was passed through in an identical manner to that described for
2. The solution turned cloudy and was chilled to -30 ◦C to afford
a crop of colourless crystals of the new aluminium carbamate 5.
An X-ray diffraction study on a single crystal of 5 established its
molecular structure (Fig. 5).
planar environment with each angle no greater than 0.1◦ from
120◦.
1
The H and 13C NMR spectra of 5, recorded in C6D6, were as
expected. As witnessed earlier for 2-Li, the order of the resonances
for the TMP fragment are reversed with respect to TMP(H), with
the methyl group at 1.47 ppm and the b and g CH2 groups upfield
of this at 1.28 and 1.20 ppm respectively. In the 13C spectrum the
carbamate carbon resonates at 160.8 ppm, which again is a typical
value for such functionality.
This dimeric structure, which lies about an inversion centre,
is similar to that witnessed by Chang et al. in the closely
related dimethyl aluminium diisopropylcarbamate,29 with a central
(AlOCO)2 ring. In 5, this ring has two co-planar OCO fragments
with the metal atoms lying above and below this plane by 0.949 A.
This deviation is considerably greater than that witnessed in
As shown in Scheme 4, a second route to 5 also exists; that
is by using 2-Li as a ligand transfer reagent with the parent
bisalkylaluminium chloride. Such a route has previously been
employed by Feier-Iova and Linti to give a gallium carbamate33 but
to the best of our knowledge has never been used in the synthesis
of an aluminium derivative. The two reagents were allowed to
stir overnight in hexane before precipitated LiCl was removed via
filtration. Concentration and cooling of the sample gave a crop of
˚
3-Li despite the M–O bonds being shorter in 5 and may be due
i
to the lesser steric bulk around Al (2 ¥ Bu group) versus that
1
around Li (1 ¥ TMEDA). Like Chang’s structure, the aluminium
crystals which were shown by H NMR spectroscopy to also be
centres lie in a pseudo-tetrahedral environment, with the greatest
the desired product 5. The crystalline yield (25%) is comparable
to that obtained from the CO2 insertion route (28%). While this at
first seems low, it is actually at the high end of the yields obtained
for other aluminium carbamates mentioned earlier.
i
distortion occurring between the pendant Bu arms [C–Al–C =
120.22(7)◦], probably to keep these branched alkyl groups apart.
˚
The near-identical C–O bond distances [1.282(2) and 1.284(2) A]
This journal is The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6190–6197 | 6195
©

View Article Online
6.0 mmol) was added to the hexane suspension to give a cloudy
solution. This was cooled to -30 ◦C. After 24 h, a crop of X-ray
quality colourless crystals had deposited (0.311 g, 66%). 1H NMR
(300 K, d8-THF): d 2.30 (4H, s, TMEDA CH2), 2.15 (12H, s,
TMEDA Me), 1.52 (6H, br s, TMP b and g CH2), 1.47 (12H, s,
TMP Me). 13C NMR (300 K, d8-THF): d 167.2 (CO2), 58.9
(TMEDA CH2), 55.2 (TMP a), 46.2 (TMEDA Me), 44.1 (TMP
b), 31.1 (TMP Me), 17.7 (TMP g). 7Li NMR (300 K, d8-THF): d
-1.3.
Conclusion
In summary, we have successfully inserted CO2 into the metal–
nitrogen bond of the lithium and sodium derivatives of the util-
ity amide 2,2,6,6-tetramethylpiperidide. The TMEDA-solvated
lithium carbamate is a dimer in the solid state with a central
(LiOCO)2 ring. Attempts to isolate a THF-solvated derivative
resulted instead in the formation of a (CO3)2- centred lithium
carbamato cluster which displays four distinct bonding modes of
the carbamato functionality towards the metal centres. Finally,
the carbamato derivative of the important bisalkylaluminium
i
amide Bu2AlTMP has been prepared via two different routes,
Synthesis of 4
namely salt metathesis via the lithium carbamate with the parent
bisalkylaluminium chloride or by insertion of CO2 into the N–Al
bond of a bisalkylaluminium TMP complex.
This was prepared in the same manner as 3-Li but instead of
TMEDA, THF was added to give a homogenous solution. After
◦
1
several weeks at -30 C, a few crystals had deposited. H NMR
(300 K, C6D6): d 3.57 (m, THF OCH2), 1.42 (m, THF CH2), 1.37
(s, TMP Me), 1.32 (br s, TMP b), 0.92 (br s, TMP g).
Experimental section
Reagents and general procedures
Synthesis of 5
All reactions and manipulations were performed under a protec-
tive argon atmosphere using either standard Schlenk techniques
or a glove box. Hexane was dried by heating to reflux over sodium
benzophenone ketyl and then distilled under nitrogen prior to use.
Method A. A solution of LiTMP (3.2 mmol, prepared in situ from
TMP(H) and nBuLi) in hexane (5 mL) was prepared and iBu2AlCl
(0.61 mL, 3.2 mmol) was added via syringe. After stirring at room
temperature for 1 h, this was filtered to remove LiCl and CO2 was
bubbled through as described above for 2-Li. This was reduced in
volume and cooled to -30 ◦C. After 24 h, a crop of X-ray quality
colourless crystals had deposited (0.295 g, 28%).
˚
TMEDA was distilled over CaH2 and stored over 4 A molecular
sieves prior to use. nButyllithium (1.6 M in hexane) and iBu2AlCl
were purchased from Aldrich and used as received. NMR spectra
were recorded on a Bruker DPX 400 MHz spectrometer operating
1
7
at 400.13 MHz for H, 155.47 MHz for Li and 100.62 MHz for
13C. All 13C spectra were proton decoupled.
Method B. iBu2AlCl (0.38 mL, 2.0 mmol) was added via syringe
to a stirred suspension of 2-Li (0.382 g, 2.0 mmol) in hexane (10
mL). This was allowed to stir overnight and was then filtered,
reduced in volume and cooled to -30 ◦C. After 24 h, a crop of
X-ray quality colourless crystals had deposited (0.162 g, 25%). 1H
NMR (300 K, C6D6): d 2.22 (2H, sept, 3JH,H = 7 Hz, iBu CH), 1.47,
Synthesis of 2-Li
nBuLi (1 mL, 1.6 M, 1.6 mmol) was added slowly to a stirred
solution of TMP(H) (0.27 mL, 1.6 mmol) in hexane (10 mL). After
1 h, CO2 was bubbled through this solution via a cannula from
an adjacent Schlenk flask filled with dry ice. A white precipitate
started to form immediately. After 30 min, the CO2 cannula
was removed and the solid was filtered and washed with more
hexane to give the final product as a white powder (0.238 g, 78%).
Small resonances representing traces of unreacted TMP(H) were
witnessed, only resonances representing the product are reported
3
(12H, s, TMP Me), 1.34 (t, 4H, JH,H = 7 Hz, TMP b), 1.28 (d,
12H, 3JH,H = 7 Hz, iBu CH3), 1.22 (m, 2H, 3JH,H = 7 Hz, TMP g),
0.39 (d, 4H, 3JH,H = 7 Hz, iBu CH2). 13C NMR (300 K, d8-THF):
d 160.8 (CO2), 58.1 (TMP a), 39.3 (TMP b), 29.7 (TMP Me), 28.8
(iBu Me), 26.5 (iBu CH), 22.4 (TMP g), 15.1 (iBu CH2).
Crystal structure determinations. Crystallographic data were
collected at 123 K on an Oxford Diffraction Gemini S Diffrac-
1
here. H NMR (300 K, C7D8): d 1.74, (12H, s, TMP Me), 1.62
(4H, t, TMP b CH2), 1.54 (2H, m, TMP g CH2). 13C NMR (300 K,
C7D8): d 168.2 (CO2), 55.9 (TMP a), 43.1 (TMP b), 31.3 (TMP
Me), 17.1 (TMP g).
˚
tometer with Mo Ka radiation (l = 0.71073 A). Structures were
solved using SHELXS-97,40 and refined to convergence on F2 and
against all independent reflections by the full-matrix least-squares
method using the SHELXL-97 program.40 The quality of structure
4 is limited by disorder and partial occupancy of the solvent THF
molecules and disorder in the ligands. The model adopted is suit-
able for purpose here, that is, identifying chemical identity. CCDC
769482–769484 contain the supplementary crystallographic data
for this paper.‡
Synthesis of 2-Na
This was prepared using the same procedure as described above
for 2-Li except nBuNa was used instead of nBuLi to afford the final
product as a white powder (0.229 g, 69%). As in the previous case,
1
trace amounts of unreacted TMP(H) were detected. H NMR
(300 K, d8-THF): d 1.53 (6H, br s, TMP b and g CH2), 1.45,
(12H, s, TMP Me). 13C NMR (300 K, d8-THF): d 166.6 (CO2),
54.7 (TMP a), 43.6 (TMP b), 31.3 (TMP Me), 17.6 (TMP g).
Acknowledgements
We thank the UK EPSRC (through grant awards: EP/F063733/1
and EP/D076889/1) and the Royal Society (through a Wolfson
merit award to REM) for generously sponsoring this research
programme.
Synthesis of 3-Li
2-Li was prepared as described above. This time, instead of
collecting the solid by filtration, an excess of TMEDA (0.90 mL,
6196 | Dalton Trans., 2010, 39, 6190–6197
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
22 A. R. Katritzky, M. Black and W.-Q. Fan, J. Org. Chem., 1991, 56,
5045–5048 and references therein.
References
1 M. W. Rathke and R. Kow, J. Am. Chem. Soc., 1972, 94, 6854–6856.
2 M. F. Lappert, A. Singh, J. L. Atwood, R. D. Roger and R. Shakir,
J. Am. Chem. Soc., 1983, 105, 302–304.
3 P. G. Williard and Q.-Y. Liu, J. Am. Chem. Soc., 1993, 115, 3380–3381.
4 B. Gehrhus, P. B. Hitchcock, A. R. Kennedy, M. F. Lappert, R. E.
Mulvey and P. J. A. Rodger, J. Organomet. Chem., 1999, 587, 88–92.
5 D. R. Armstrong, D. V. Graham, A. R. Kennedy, R. E. Mulvey and
C. T. O’Hara, Chem.–Eur. J., 2008, 14, 8025–8034.
23 (a) A. Otero, J. Ferna´ndez-Baeza, J. Tejeda, A. Antin˜olo, F. Carrillo-
Hermosilla, E. D´ıez-Barra, A. Lara-Sa´nchez, M. Ferna´ndez-Lo´pez,
M. Lanfranchi and M. A. Pellinghelli, J. Chem. Soc., Dalton Trans.,
1999, 3537–3539; (b) A. Otero, J. Ferna´ndez-Baeza, J. Tejeda, A.
Antin˜olo, F. Carrillo-Hermosilla, E. D´ıez-Barra, A. Lara-Sa´nchez and
M. Ferna´ndez-Lo´pez, J. Chem. Soc., Dalton Trans., 2000, 2367–2374;
(c) A. Otero, J. Ferna´ndez-Baeza, A. Antin˜olo, J. Tejeda, A. Lara-
Sa´nchez, L. Sa´nchez-Barba, M. T. Expo´sito and A. M. Rodr´ıguez,
Dalton Trans., 2003, 1614–1619.
24 S. Inoue and Y. Yokoo, Bull. Chem. Soc. Jpn., 1972, 45, 3651–3653.
25 D. B. Dell’Amico, F. Calderazzo, M. Dell’Innocenti, B. Gu¨ldenpfennig,
S. Ianelli, G. Pelizzi and P. Robino, Gazz. Chim. Ital., 1993, 123, 283–
288.
26 U. Abram, D. B. Dell’Amico, F. Calderazzo, S. Kaskel, L. Labella, F.
Marchetti, R. Rovai and J. Stra¨hle, Chem. Commun., 1997, 1941–1942.
27 K. Knabel, I. Krossing, H. No¨th, H. Schwenk-Kircher, M. Schmidt-
Amelunxen and T. Seifert, Eur. J. Inorg. Chem., 1998, 1095–1114.
28 K. Knabel and H. No¨th, Z. Naturforsch., Teil B, 2005, 60, 1027–1035.
29 C.-C. Chang, B. Srinivas, M.-L. Wu, W.-H. Chiang, M. Y. Chiang and
C.-H. Hsiung, Organometallics, 1995, 14, 5150–5159.
30 T. Habereder, H. No¨th and R. T. Paine, Eur. J. Inorg. Chem., 2007,
4298–4305.
6 B. L. Lucht and D. B. Collum, J. Am. Chem. Soc., 1994, 116, 7949–7950.
7 J. F. Remenar, B. L. Lucht, D. Kruglyak, F. E. Romesberg, J. H. Gilchrist
and D. B. Collum, J. Org. Chem., 1997, 62, 5748–5754.
8 (a) A. Yasuda, S. Tanaka, K. Oshima, H. Yamamoto and H. Nozaki,
J. Am. Chem. Soc., 1974, 96, 6513–6514; (b) A. Yasuda, H. Yamamoto
and H. Nozaki, Bull. Chem. Soc. Jpn., 1979, 52, 1705–1708.
9 S. Tanaka, A. Yasuda, H. Yamamoto and H. Nozaki, J. Am. Chem.
Soc., 1975, 97, 3252–3254.
10 H. Nozaki, K. Oshima, K. Takai and S. Ozawa, Chem. Lett., 1979,
379–380.
11 Y. Naruse and H. Yamamoto, Tetrahedron, 1988, 44, 6021–6029.
12 K. Maruoka, M. Oishi and H. Yamamoto, J. Org. Chem., 1993, 58,
7638–7639.
13 (a) R. E. Mulvey, Organometallics, 2006, 25, 1060–1075; (b) R. E.
Mulvey, F. Mongin, M. Uchiyama and Y. Kondo, Angew. Chem., Int.
Ed., 2007, 46, 3802–3824; (c) R. E. Mulvey, Acc. Chem. Res., 2009, 42,
743–755.
31 S. C. Ball, I. Cragg-Hine, M. G. Davidson, R. P. Davies, A. J. Edwards,
I. Lopez-Solera, P. R. Raithby and R. Snaith, Angew. Chem., Int. Ed.
Engl., 1995, 34, 921–923.
´
14 (a) J. Garcia-Alvarez, D. V. Graham, A. R. Kennedy, R. E. Mulvey and
32 R. P. Davies, P. R. Raithby and R. Snaith, Organometallics, 1996, 15,
S. Weatherstone, Chem. Commun., 2006, 3208–3210; (b) B. Conway,
4355–4356.
´
33 O. Feier-Iova and G. Linti, Z. Anorg. Allg. Chem., 2008, 634, 559–
564.
34 U. Ko¨hn and E. Anders, Tetrahedron Lett., 2002, 43, 9585–9589.
35 Y. Tang, L. N. Zakharov, A. L. Rheingold and R. A. Kemp,
Organometallics, 2004, 23, 4788–4891.
36 J. R. Hagadorn, L. Que Jr and W. B. Tolman, J. Am. Chem. Soc., 1998,
120, 13531–13532.
37 P. C. Andrews, N. D. R. Barnett, R. E. Mulvey, W. Clegg, P. A. O-Neil,
D. Barr, L. Cowton, A. J. Dawson and B. J. Wakefield, J. Organomet.
Chem., 1996, 518, 85–95.
38 K. Hyva¨rinen, M. Klinga and M. Leskela¨, Polyhedron, 1996, 15, 2171–
2177.
39 K.-C. Yang, C.-C. Chang, C.-S. Yeh, G.-H. Lee and S.-M. Peng,
Organometallics, 2001, 20, 126–137.
40 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008,
64, 112.
J. Garcia-Alvarez, E. Hevia, A. R. Kennedy, R. E. Mulvey and S. D.
Robertson, Organometallics, 2009, 28, 6462–6468.
15 B. Conway, A. R. Kennedy, R. E. Mulvey, S. D. Robertson and J.
Garcia-Alvarez, Angew. Chem., Int. Ed., 2010, 49, 3182–3184.
16 (a) D. J. Woodcock, Chem. Commun., 1968, 267–268; (b) D. C. D. Butler
and H. Alper, Chem. Commun., 1998, 2575–2576.
17 T.-T. Wu, J. Huang, N. D. Arrington and G. M. Dill, J. Agric. Food
Chem., 1987, 35, 817–823.
18 I. Vauthey, F. Valot, C. Gozzi, F. Fache and M. Lemaire, Tetrahedron
Lett., 2000, 41, 6347–6350.
19 M. Aresta, A. Dibenedetto and E. Quaranta, J. Chem. Soc., Dalton
Trans., 1995, 3359–3363.
20 A. Belforte and F. Calderazzo, J. Chem. Soc., Dalton Trans., 1989,
1007–1009.
21 D. B. Dell’Amico, F. Calderazzo, L. Labella, F. Marchetti and G.
Pampaloni, Chem. Rev., 2003, 103, 3857–3897.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 6190–6197 | 6197
©