Donor-dictated interlocking co-complexation reactions of LiNHDipp with dimethylzinc: Synthesis and structures of new methyl(amido)zincates

Article abstract of DOI:10.1021/ic900313b

A systematic study of the interlocking eo-complexation reactions between the primary lithium amide LiNHDipp (Dipp = 2,6-diisopropylphenyl) and dimethylzinc in the presence of different donor ligands is presented which concludes that the final outcome of these reactions is largely dictated by the type of structure that is formed when the donor is coordinated to the lithium amide. When chelating diamine TMEDA (N,N,N′,N′- tetramethylethylenediamine) is employed [{Li₂(NHDipp) ₂(TMEDA)}_∞ (1) is obtained, where Li ₂N₂ rings are connected by TMEDA bridges generating a polymeric chain arrangement which does not form a co-complex with Me ₂Zn even in the presence of an excess of TMEDA. The tridentate ligand PMDETA (N,N,N′, N″, N″-pentamethyldiethylenetriamine) when reacted with LiNHDipp forms monomeric [(PMDETA)Li(NHDipp)](4) which successfully forms a mixed-metal co-complex with Me₂Zn affording dialkyl(amido)zincate [(PMDETA)LiZn(NHDipp)(Me)₂] (2). When the co-complexation reaction is carried out in the presence of monodentate tetrahydrofuran (THF), zincate [(THF)₃LiZn(NHDipp)(Me)₂] (3) is obtained which was found to partially decompose in hexane solution after long periods of time at room temperature (2 weeks) to afford the unprecedented "zinc-rich" zincate [(THF)₃LiZn₂(Me) ₃(NHDipp)₂] (5). This compound presents a unique structure in the solid state previously unknown in organozincate chemistry with a trinuclear Li · · · Zn · · · Zn chain arrangement where the metals are connected by only two amido bridges and therefore both zinc centers exhibit trigonal planar geometries. 5 can be prepared in good yields by the rational reaction of LiNHDipp with a 2:1:3 mixture of Me₂Zn, NH₂Dipp and THF. The different solid-state structural motifs of compounds 1,2,4, and 5 have been revealed by X-ray crystallographic studies. Multinuclear NMR (¹H, ¹³C and ⁷Li) spectroscopic data recorded in C₆D₆ solution are also reported for compounds 1-6. Mixed-metal compounds 2 and 5 constitute the first examples of crystallographically characterized alkyl(amido)zineates containing a primary amide.

Full text of DOI:10.1021/ic900313b

5320 Inorg. Chem. 2009, 48, 5320–5327
DOI: 10.1021/ic900313b
Donor-Dictated Interlocking Co-Complexation Reactions of LiNHDipp with
Dimethylzinc: Synthesis and Structures of New Methyl(amido)zincates
William Clegg,‡ David V. Graham,† Emma Herd,† Eva Hevia,*,† Alan R. Kennedy,† Matthew D. McCall,† and
Luca Russo‡
†WestCHEM, Department of Pure and Applied Chemistry, University of Strathclyde, Glasgow G1 1XL, U.K.
and ‡School of Chemistry, Newcastle University, Newcastle upon Tyne NE1 7RU, U.K.
Received February 16, 2009
A systematic study of the interlocking co-complexation reactions between the primary lithium amide LiNHDipp
(Dipp = 2,6-diisopropylphenyl) and dimethylzinc in the presence of different donor ligands is presented which
concludes that the final outcome of these reactions is largely dictated by the type of structure that is formed when the
donor is coordinated to the lithium amide. When chelating diamine TMEDA (N,N,N0,N0-tetramethylethylenediamine) is
employed [{Li2(NHDipp)2(TMEDA)}¥] (1) is obtained, where Li2N2 rings are connected by TMEDA bridges generating
a polymeric chain arrangement which does not form a co-complex with Me2Zn even in the presence of an excess of
TMEDA. The tridentate ligand PMDETA (N,N,N0,N00,N00-pentamethyldiethylenetriamine) when reacted with LiNHDipp
forms monomeric [(PMDETA)Li(NHDipp)](4) which successfully forms a mixed-metal co-complex with Me2Zn
affording dialkyl(amido)zincate [(PMDETA)LiZn(NHDipp)(Me)2] (2). When the co-complexation reaction is carried
out in the presence of monodentate tetrahydrofuran (THF), zincate [(THF)3LiZn(NHDipp)(Me)2] (3) is obtained which
was found to partially decompose in hexane solution after long periods of time at room temperature (2 weeks) to afford
the unprecedented “zinc-rich” zincate [(THF)3LiZn2(Me)3(NHDipp)2] (5). This compound presents a unique structure in
the solid state previously unknown in organozincate chemistry with a trinuclear Li Zn Zn chain arrangement
3 3 3 3 3 3
where the metals are connected by only two amido bridges and therefore both zinc centers exhibit trigonal planar
geometries. 5 can be prepared in good yields by the rational reaction of LiNHDipp with a 2:1:3 mixture of Me2Zn,
NH2Dipp and THF. The different solid-state structural motifs of compounds 1, 2, 4, and 5 have been revealed by X-ray
crystallographic studies. Multinuclear NMR (1H, 13C and 7Li) spectroscopic data recorded in C6D6 solution are also
reported for compounds 1-6. Mixed-metal compounds 2 and 5 constitute the first examples of crystallographically
characterized alkyl(amido)zincates containing a primary amide.
Introduction
etallic reagents, TMP-di-tert-butylzincates (TMP = 2,2,6,
6-tetramethylpiperidide) of lithium2 and sodium3 have
shown a greater metallating (zincating) power as best illu-
strated by the unprecedented 2,6-dideprotonation of
naphthalene4 or the regioselective meta-deprotonation of
dimethylaniline5 by the base [(TMEDA)NaZn(TMP)-
Alkali-metal dialkyl(amido) zincates have recently been
shown to be extremely effective and versatile reagents in
organic synthesis. By combining the high reactivity typically
associated with group 1 organometallic compounds with
the greater selectivity of the neutral organozinc reagents,
these mixed-metal reagents have proved to be highly chemo-
and regioselective reagents in deprotonative metalations,
allowing the direct (one-step) zincation of a wide range of
aromatic substrates with a rich variety of functional groups
that are often incompatible with group 1 organometallic
reagents such as BuLi or LDA.1 Among this family of bim-
(tBu)2] (TMEDA
=
N,N,N0,N0-tetramethylethylenedia-
mine). Amido-dialkylzincates containing other classical uti-
lity secondary amides such as diisopropylamide (DA)6 or
(3) Andrikopoulos, P. C.; Armstrong, D. R.; Barley, H. R. L.; Clegg, W.;
Dale, S. H.; Hevia, E.; Honeyman, G. W.; Kennedy, A. R.; Mulvey, R. E. J.
Am. Chem. Soc. 2005, 127, 6184.
(4) Clegg, W.; Dale, S. H.; Hevia, E.; Hogg, L. M.; Honeyman, G. W.;
Mulvey, R. E.; O’Hara, C. T. Angew. Chem., Int. Ed. 2006, 45, 6548.
(5) Armstrong, D. R.; Clegg, W.; Dale, S. H.; Hevia, E.; Hogg, L. M.;
Honeyman, G. W.; Mulvey, R. E. Angew. Chem., Int. Ed. 2006, 45, 3775.
(6) (a) Armstrong, D. R.; Clegg, W.; Dale, S. H.; Garcia-Alvarez, J.;
Harrington, R. W.; Hevia, E.; Honeyman, G. W.; Kennedy, A. R.; Mulvey,
R. E.; O’Hara, C. T. Chem. Commun. 2008, 187. (b) Clegg, W.; Garcia-
Alvarez, J.; Garcia-Alvarez, P.; Graham, D. V.; Harrington, R. W.; Hevia,
E.; Kennedy, A. R.; Mulvey, R. E.; Russo, L. Organometallics 2008, 27,
2654.
*To whom correspondence should be addressed. E-mail: eva.hevia@
strath.ac.uk.
(1) (a) Mulvey, R. E. Organometallics 2006, 25, 1060. (b) Mulvey, R. E.;
Mongin, F.; Uchiyama, M.; Kondo, Y. Angew. Chem., Int. Ed. 2007, 46, 3802.
(2) (a) Kondo, Y.; Shilai, M.; Uchiyama, M.; Sakamoto, T. J. Am. Chem.
Soc. 1999, 121, 3539. (b) Uchiyama, M.; Matsumoto, Y.; Nobuto, D.;
Furuyama, T.; Yamaguchi, K.; Morokuma, K. J. Am. Chem. Soc. 2006, 128,
8748. (c) Clegg, W.; Dale, S. H.; Hevia, E.; Honeyman, G. W.; Mulvey, R. E.
Angew. Chem., Int. Ed. 2006, 45, 2370.
r
2009 American Chemical Society
pubs.acs.org/IC
Published on Web 05/01/2009

Article
Inorganic Chemistry, Vol. 48, No. 12, 2009 5321
hexamethyldisilazide (HMDS)7 have also been prepared. The
reactivities and the structures of these compounds are inti-
mately related to the nature of the amide ligand employed.
Thus, for example, Kondo and Uchiyama have shown for the
metalation of 3-bromopyridines by lithium di-tert-butyl(ami-
do)zincates that the regioselectivity of the reaction can be
controlled by switching between TMP (deprotonation at the
2-position) and DA (deprotonation at the 4-position).8
Usually the synthetic methodology employed to prepare
these mixed-metal reagents is an interlocking co-complexa-
tion approach (Scheme 1)6b,9 where both monometallic
reagents, the alkali-metal amide and the dialkylzinc com-
pound are combined together in the presence of a Lewis base
(shown as X-X in Scheme 1) such as the chelating TMEDA,
PMDETA (PMDETA = N,N,N0,N00,N00-pentamethylethy-
lenetriamine) or monodentate THF. In non-coordinating
solvents, alkali-metal amides exist in their aggregated form
(for example unsolvated LDA displays a single strand helical
polymeric structure10 whereas NaTMP is a cyclic trimer11).
Thus, the presence of a Lewis donor is required to break these
oligomeric forms into smaller aggregates (preferably mono-
mers) favoring then the co-complexation reaction with the
relevant R2Zn compound.12,13
Scheme 1. Interlocking Co-Complexation Methodology to Prepare
Dimethyl(amido)zincates
As herein reported, these novel heterobimetallic com-
pounds were prepared by using the aforementioned inter-
locking co-complexation approach by combining the lithium
anilide LiNHDipp with dimethylzinc in the presence of Lewis
bases of increasing denticity (monodentate THF, bidentate
TMEDA and tridentate PMDETA). The influence of chan-
ging the Lewis base has also been evaluated. In addition we
report the synthesis and structural characterization of the
unprecedented stoichiometrically unusual “zinc-rich” (1:2,
Li/Zn) trialkyl-bis(amido)zincate [(THF)3LiZn2(Me)3(NH-
Dipp)2] (5) which was obtained unexpectedly as a dispropor-
tionation product of the 1:1, Li/Zn complex [(THF)3LiZn-
(Me)2(NHDipp)] (3).
To date, the synthesis of dialkyl(amido)zincates has been
limited to the use of secondary, sterically demanding amides
and to the best of our knowledge no examples of zincates
involving primary amides have been reported. Herein we
report the preparation and structural elucidation of dimethyl
(amido)zincates derived from the primary anilide NH2Dipp
(Dipp = 2,6-diisopropylphenyl). It can be anticipated that
the participation of this primary amide will have a profound
influence on the structure and in the reactivity of these new
heterobimetallic compounds. The presence of a hydrogen
attached to the nitrogen atom is likely to diminish their
efficiency as metallating reagents (in comparison with those
containing the more basic and sterically demanding second-
ary amides TMP or DA). On the other hand, lithiated
primary amides are known to be versatile and valuable
chemical intermediates, since, for example, they can behave
as either amido (NHR-) or imido (NR2-) transfer agents.14
Therefore these new zincates could be employed as precur-
sors to prepare mixed-metal organoimido (NAr2-) com-
pounds, by the removal of the remaining hydrogen on the
nitrogen.15
Experimental Section
General Information. All reactions were performed under a
protective argon atmosphere using standard Schlenk techni-
ques. Hexane, tetrahydrofuran (THF), and toluene were dried
by heating to reflux over sodium benzophenone ketyl and
distilled under nitrogen prior to use. nBuLi and Me2Zn, were
purchased from Aldrich Chemicals and used as received. NMR
spectra were recorded on a Bruker DPX 400 MHz spectrometer,
operating at 400.13 MHz for 1H, 150.32 MHz for 7Li and 100.62
MHz for 13C. Elemental analyses were carried out on a Perkin-
Elmer 2400 elemental analyzer.
X-ray Crystallography. Crystallographic data were col-
lected on Nonius KappaCCD diffractometers at 123 K for 5
and at 150 K for 1, 2, and4,with Mo KRradiation (λ= 0.71073 A).
The structures were refined by full-matrix least-squares using all
unique F2 values with programs of the SHELX family.16 Specific
crystallographic data and refinement parameters are given in
Table 1. Minor disorder was resolved for THF ligands in 5.
˚
Experimental Section
Synthesis of [{Li2(NHDipp)2(TMEDA)}¥] (1). To a
solution of hexane (5 mL) and 2,6-diisopropylaniline
(0.35 mL, 2 mmol) was added nBuLi (1.6 M in hexane, 1.25 mL,
2 mmol), andtheresultingsuspensionwasstirredfor1h. TMEDA
(0.3 mL, 2 mmol) was then added, followed by toluene (7 mL) and
gentle heating to afford a pale yellow solution. At this stage
Me2Zn (2 mL of a 1 M solution in heptane, 2 mmol) was
introduced. The resulting solution was left to cool overnight in a
Dewar with hot water. Small colorless crystals were isolated (0.309 g,
(7) (a) Westerhausen, M.; Rademarcher, B.; Schwarz, W. Z. Naturforsch.,
B: Chem. Sci. 1994, 49, 199. (b) Armstrong, D. R.; Herd, E.; Graham, D. V.;
Hevia, E.; Kennedy, A. R.; Clegg, W.; Russo, L. Dalton Trans. 2008, 1323.
(8) Imahori, T.; Uchiyama, M.; Sakamoto, T.; Kondo, Y. Chem. Com-
mun. 2001, 2450.
(9) Conway, B.; Graham, D. V.; Hevia, E.; Kennedy, A. R.; Klett, J.;
Mulvey, R. E. Chem. Commun. 2008, 2638.
(10) Barnet, N. D. R.; Mulvey, R. E.; Clegg, W.; O’Neil, P. A. J. Am.
Chem. Soc. 1991, 113, 8187.
(11) Gehrhus, B.; Hitchcok, P. H.; Kennedy, A. R.; Lappert, M. F.;
Mulvey, R. E.; Rodger, P. J. A. J. Organomet. Chem. 1999, 587, 88.
(12) Westerhausen, M.; Rademacher, B.; Schwarz, W. Z. Anorg. Allg.
Chem. 1993, 619, 675.
(13) Recently two examples of unsolvated dialkyl(amido)zincates have
been reported which adopt polymeric structures (a) Kondo, Y.; Morey, J. V.;
Morgan, J. C.; Naka, H.; Nobuto, D.; Raithby, P. R.; Uchiyama, M.;
Wheatley, A. E. H. J. Am. Chem. Soc. 2007, 129, 13360. (b) See reference 7b.
(14) Clegg, W.; Liddle, S. T.; Mulvey, R. E.; Robertson, A. Chem.
Commun. 2000, 223.
(15) Allan, R. E.; Beswick, M. A.; Feeder, N.; Kranz, M.; Mosquera, M.
E. G.; Raithby, P. R.; Wheatley, A. E. H.; Wright, D. S. Inorg. Chem. 1998,
37, 2602.
yield 64%). 1H NMR (400.13 MHz, 298 K, C6D6) δ 7.15 (4H, d,
i
meta), 6.71 (2H, t, Hpara), 3.22 (4H, m, CH, Pr), 2.84 (2H, s,
H
NH), 1.76 (12H, s, CH3, TMEDA), 1.71 (4H, s, CH2, TMEDA),
1.36 (24H, d, CH3, iPr). 7Li NMR (298K, C6D6, reference LiCl
in D2O at 0.00 ppm): δ 1.88. 1H NMR (400.13 MHz, 298 K, d8-
THF) δ 6.68 (4H, d, Hmeta), 6.03 (2H, t, Hpara), 3.22 (4H, m, CH,
iPr), 2.68 (2H, s, NH), 2.32 (4H, s, CH2, TMEDA), 2.16 (12H, s,
CH3, TMEDA), 1.20 (24H, d, CH3, iPr). 13C{1H} NMR
(100.62 MHz, 298 K, d8-THF) δ 158.2 (Cipso), 131.5 (Cortho),
(16) Sheldrick, G. M. Acta Crystallogr., Sect. A. 2008, 64, 112.

5322 Inorganic Chemistry, Vol. 48, No. 12, 2009
Clegg et al.
Table 1. Crystallographic Data
compound
formula
form wt
1
C30H52Li2N4
482.6
triclinic
P1
8.4634(5)
9.1374(10)
10.9932(7)
75.394(9)
87.935(6)
72.640(7)
784.49(11)
1
2
C23H47LiN4Zn 0.5C7H8
498.0
triclinic
P1
9.6061(10)
9.8717(7)
16.5903(17)
91.976(7)
101.641(8)
104.928(7)
1482.6(2)
2
4
C21H41LiN4
356.5
orthorhombic
P212121
10.0117(11)
14.6223(17)
15.682(2)
90
90
90
2295.7(5)
4
1.031
5
C39H69LiN2O3Zn2
751.6
orthorhombic
Pbca
18.1783(1)
18.6099(2)
24.4975(2)
90
90
90
8287.42(12
8
1.205
1.192
123
3
cryst system
space group
˚
a, A
˚
b, A
˚
c, A
R, deg
β, deg
γ, deg
3
˚
V, A
Z
Dcalcd, g cm-3
1.022
0.059
1.116
0.847
μ, mm-1
T, K
0.061
150
150
0.37 ꢀ 0.34 ꢀ 0.29
150
0.61 ꢀ 0.48 ꢀ 0.39
cryst size, mm
reflns measd
unique reflns
Rint
no of params
R (F, F2 > 2σ)
Rw (F2, all data)
goodness of fit
0.82 ꢀ 0.48 ꢀ 0.32
0.32 ꢀ 0.08 ꢀ 0.08
8844
2709
21036
5171
8017
2249
17290
9026
0.029
173
0.043
0.110
1.08
0.20, -0.17
0.033
323
0.028
0.074
1.05
0.35, -0.21
0.030
248
0.047
0.145
0.99
0.44, -0.23
0.024
439
0.045
0.121
1.09
0.99, -0.39
-3
˚
max, min ΔF, e A
122.2 (Cmeta), 109.4 (Cpara), 58.8 (CH2, TMEDA), 46.2 (CH3,
TMEDA), 28.4 (CH, iPr), 23.9 (CH3, iPr). 7Li NMR (298 K, d8-
THF, reference LiCl in D2O at 0.00 ppm): δ -0.62. Elemental
analysis calcd (%) for C30H52Li2N4: C 74.65, H 10.85, N 11.60;
found: C 74.11, H 11.03, N 11.39.
1.01. Satisfactory elemental analysis of this compound could not
be obtained because of its highly air and moisture sensitive
nature.
Synthesis of [(PMDETA)Li(NHDipp)] (4). To a solution
of hexane (5 mL) and 2,6-diisopropylaniline (0.35 mL, 2 mmol)
was added nBuLi (1.6 M in hexane, 1.25 mL, 2 mmol), and the
resulting suspension was stirred for 10 min. PMDETA
(0.42 mL, 2 mmol) was then added affording a pale yellow
solution which deposited colorless crystals when placed in the
freezer (-20 ꢀC) and left overnight (0.564 g, yield 79%).
1H NMR (400.13 MHz, 298 K, C6D6) δ 7.29(2H, d, Hmeta),
Synthesis of [(PMDETA)LiZn(NHDipp)(Me)2] (2). To
a solution of hexane (5 mL) and 2,6-diisopropylaniline (0.35 mL,
2 mmol) was added nBuLi (1.6 M in hexane, 1.25 mL, 2 mmol),
andthe resultingsuspension was stirred for 10 min. Me2Zn (2 mL
of a 1 M solution in heptane, 2 mmol) was then added, and the
reaction mixture stirred for 30 min, followed by the addition of
PMDETA (0.42 mL, 2 mmol). After stirring for 2 h the pre-
cipitate was still present. Addition of toluene (6 mL) and gentle
heatingafforded a yellow solution, which was placed in the fridge
(0 ꢀC) and left overnight. Small colorless crystals were isolated
(0.550 g, yield 61%).1H NMR (400.13 MHz, 298 K, C6D6) δ 7.15
i
6.66 (1H, t, Hpara), 3.34 (2H, m, CH, Pr), 3.09 (1H, s, NH),
2.09 (3H, s, NCH3, PMDETA), 1.89 (12H, bs, CH3 PMDETA),
1.72 (8H, bs, CH2, PMDETA), 1.54 (12H, d, CH3, iPr). 13C{1H}
NMR (100.62 MHz, 298 K, C6D6) δ 159.4 (Cipso), 131.3 (Cortho),
122.3 (Cmeta), 105.9 (Cpara), 57.2, 53.4 (CH2, PMDETA),
45.7 (N(CH3)2, PMDETA), 44.2 (N(CH3), PMDETA),
i
(2H, d, Hmeta), 6.79 (1H, t, Hpara), 3.45 (2H, m, CH, Pr), 2.84
i
i
(1H, s, NH), 2.14 (3H, s, NCH3, PMDETA), 1.72 (20H, bs, CH3
and CH2, PMDETA), 1.38 (12H, d, CH3, iPr), -0.32 (6H, s, Zn-
(CH3)2). 13C{1H} NMR (100.62 MHz, 298 K, C6D6) δ 154.5
(Cipso), 134.1 (Cortho), 122.8 (Cmeta), 114.2 (Cpara), 57.1, 53.2
(CH2, PMDETA), 45.7 (N(CH3)2, PMDETA), 44.7 (N(CH3),
29.0 (CH, Pr), 24.1 (CH3, Pr). 7Li NMR (298 K, C6D6,
reference LiCl in D2O at 0.00 ppm): δ 1.47. Elemental analysis
calcd (%) for C21H41LiN4: C 70.75, H 11.59, N 15.72; found: C
70.21, H 11.64, N 15.10.
Synthesis of [(THF)3LiZn2(Me)3(NHDipp)2] (5). To a
solution of hexane (8 mL) and 2,6-diisopropylaniline (0.35 mL,
2 mmol) was added nBuLi (1.6 M in hexane, 1.25 mL, 2 mmol),
and the resulting suspension was stirred for 10 min. Me2Zn
(4 mL of a 1 M solution in heptane, 4 mmol) was then added, and
the reaction mixture stirred for 20 min. A second equivalent of
2,6-diisopropylaniline (0.35 mL, 2 mmol) was added, and the
suspension stirred for a further 20 min, followed by the addition
of THF (0.49 mL, 6 mmol) to give a white precipitate. The
reaction mixture was then refluxed for 2 h to give a colorless
solution, which on cooling to room temperature gave an oil. The
Schlenk tube was placed in the freezer (-20 ꢀC) and left over-
night. A white microcrystalline solid was isolated (0.856 g, yield
i
i
PMDETA), 27.9 (CH, Pr), 24.5 (CH3, Pr), -8.1 (Zn(CH3)2).
7Li NMR (298 K, C6D6, reference LiCl in D2O at 0.00 ppm): δ
0.77. Elemental analysis calcd (%) for C23H47LiN4Zn: C 61.12,
H 10.48, N 12.40; found: C 61.41, H 10.80, N 12.33.
Synthesis of [(THF)3LiZn(NHDipp)(Me)2] (3). To
a solution of hexane (5 mL) and 2,6-diisopropylaniline
(0.35 mL, 2 mmol) was added nBuLi (1.6 M in hexane,
1.25 mL, 2 mmol), and the resulting suspension was stirred for
10 min. Me2Zn (2 mL of a 1 M solution in heptane, 2 mmol) was
then added, and the reaction mixture stirred for 30 min, followed
by the addition of THF (0.49 mL, 6 mmol), to give an oil, which
remained even after addition of toluene (1 mL) and gentle
heating. The Schlenk tube was placed in the freezer (-20 ꢀC)
and left for several days. A batch of colorless crystals was
1
57%). H NMR (400.13 MHz, 298 K, C6D6) δ 7.12 (4H, d,
H
meta), 6.83 (2H, t, Hpara), 3.31 (4H, m, CH, iPr), 3.29 (12H, m,
1
isolated (0.43 g, yield 43%). H NMR (400.13 MHz, 298 K,
OCH2, THF), 3.08 (2H, bs, NH), 1.35 (24H, d, CH3, iPr), 1.26
(12H, m, CH2, THF), -0.43 (9H, s, Zn(CH3)2). 13C{1H} NMR
(100.62 MHz, 298 K, C6D6) δ 154.5 (Cipso), 134.8 (Cortho), 123.3
C6D6) δ 7.14 (2H, d, Hmeta), 6.75 (1H, t, Hpara), 3.39 (2H, m, CH,
iPr), 3.38 (12H, m, OCH2, THF), 2.86 (1H, s, NH), 1.42 (12H, d,
CH3, iPr), 1.39 (12H, m, CH2, THF), -0.42 (6H, s, Zn(CH3)2).
13C{1H} NMR (100.62 MHz, 298 K, C6D6) δ 154.2 (Cipso), 134.8
(Cortho), 123.3 (Cmeta), 116.2 (Cpara), 68.1 (OCH2, THF), 28.5
(CH, iPr), 25.4 (CH2, THF), 24.2 (CH3, iPr), -9.0 (Zn(CH3)2).
7Li NMR (298 K, C6D6, reference LiCl in D2O at 0.00 ppm): δ
i
(Cmeta), 116.6 (Cpara), 68.1 (OCH2, THF), 28.5 (CH, Pr), 25.4
(CH2, THF), 24.3 (CH3, Pr), -9.1 (Zn(CH3)2). 7Li NMR
i
(298 K, C6D6, reference LiCl in D2O at 0.00 ppm): δ 1.07.
Elemental analysis calcd (%) for C39H69LiN2O3Zn2: C 62.31, H
9.25, N 3.72; found: C 62.28, H 9.60, N 3.91.

Article
Inorganic Chemistry, Vol. 48, No. 12, 2009 5323
Scheme 2
Table 2. Selected Chemical Shifts in the 1H and 7Li NMR Spectra in C6D6 Solution
compound
δ(1H)NH
δ(1H)iPr
δ(1H)Ar
δ(1H)Zn-Me
δ(7Li)
NH2Dipp
3.26
2.84
2.84
2.86
3.09
3.08
2.67, 1.16
3.22, 1.36
3.45, 1.38
3.39, 1.39,
3.34, 1.54
3.31, 1.35
7.03, 6.85
7.15, 6.71
7.16, 6.79
7.14, 6.75
7.29, 6.66
7.12, 6.83
[{Li2(NHDipp)2(TMEDA)}¥] (1)
[(PMDETA)LiZn(NHDipp)(Me)2] (2)
[(THF)3LiZn(NHDipp)(Me)2] (3)
[(PMDETA)Li(NHDipp)] (4)
1.88
0.77
1.01
1.47
1.07
-0.32
-0.42
[(THF)3LiZn2(Me)3(NHDipp)2] (5)
-0.43
Synthesis of [(THF)LiZn(NHDipp)2Me] (6). Isolated
crystals of [(THF)3LiZn2(Me)3(NHDipp)2] (5) (0.751 g,
1 mmol) were dissolved in toluene (10 mL) afforing a colorless
solution which was stirred at room temperature for 10 min.
Volatiles were removed under vacuum affording a colorless
oil which was characterized by multinuclear NMR (1H, 13C
analysis of the crystals formed from this solution in deuter-
ated benzene solution showed the absence of any signal
attributable to the methyl groups from dimethylzinc. Thus,
1
the H NMR spectrum showed the relevant signals for the
amide fragment [a doublet at 7.15 ppm and a triplet at 6.71
ppm for the aromatic protons, a multiplet at 3.22 ppm and a
doublet at 1.35 ppm for the iPr groups and a broad singlet at
2.84 ppm for the NH] and coordinated TMEDA [1.76 for
methyl groups and 1.71 ppm for the CH2 groups] in a ratio of
2:1 (Table 2). The monometallic constitution of 1 was
confirmed when its structure was determined by X-ray
crystallography (Figure 1).
The structure of compound 1 is based on (LiN)2 four-
membered ring dimers linked by monodentate bridging (non-
chelating) TMEDA ligands affording a polymeric chain
arrangement. The lithium atoms display a distorted trigonal
planar geometry (sum of the bond angles around Li: 359.55ꢀ),
bonding to three nitrogens, two from the anilido ligands and
one from the TMEDA. For each (LiN)2 dimer the hydrogen
atoms of the amido ligands (and concomitantly the substi-
tuted aryl substituents) lie transoid to each other relative to
7
1
and Li) spectroscopy. H NMR (400.13 MHz, 298 K, C6D6)
δ 7.06 (4H, d, Hmeta), 6.84 (2H, t, Hpara), 3.35 (4H, bs, CH, iPr),
3.18 (2H, bs, NH), 2.85 (4H, m, OCH2, THF), 1.32 (24H, bs,
i
CH3, Pr), 0.98 (4H, m, CH2, THF), -0.38 (3H, s, Zn(CH3)2).
13C{1H} NMR (100.62 MHz, 298 K, C6D6) δ 1.49.5 (Cipso),
135.4 (Cortho), 123.5 (Cmeta), 118.0 (Cpara), 68.2 (OCH2, THF),
28.9 (CH, iPr), 25.0 (CH3, iPr), 24.1 (CH2, THF), -12.7 (Zn(CH3)2).
7Li NMR (298 K, C6D6, reference LiCl in D2O at 0.00 ppm):
δ 1.40.
Results and Discussion
The synthetic methodology employed in the preparation of
the new lithium alkyl(amido)zincates is depicted in Scheme 2.
The lithium amide was prepared in situ by reaction of
equimolar amounts of nBuLi and freshly distilled amine
NH2Dipp in hexane at 0 ꢀC, forming a white precipitate.
To this suspension a solution of dimethylzinc in hexane was
added and then the stoichiometric amount of the relevant
donor [(i) TMEDA; (ii) PMEDTA, (iii) 3 equiv of THF] was
added to afford pale yellow solutions which on cooling
deposited colorless crystals of the TMEDA-solvated lithium
amide [{Li2(NHDipp)2(TMEDA)}¥] (1), the dimethyl
(anilido)zincates [(PMDETA)LiZn(NHDipp)(Me)2] (2) and
[(THF)3LiZn(NHDipp)(Me)2] (3) respectively in good isolated
yields (56-79%). These new compounds were characterized in
solution using multinuclear NMR spectroscopy (1H, 13C,
and 7Li) and the structures in the solid state for 1 and 2 were
elucidated by X-ray crystallography.
the Li Li vector, probably to minimize the steric repulsions
3 3 3
within the four-membered ring. This polymeric structure
contrasts with the only previous example of a lithium anilide
solvated by TMEDA, namely [Li(NHAr)(TMEDA)] (Ar =
2,4,6-tBu3-C 6H2),17 where the diamine binds to lithium in a
chelating fashion, but it bears close resemblance to the
TMEDA-solvated structure of LDA reported by Williard
made up of cyclic dimers of (LiDA)2 connected by bridging
(non-chelating) TMEDA generating an infinite chain.18
Compound 1 can be obtained as a crystalline solid in
reasonable isolated yields (63-68%). NMR analysis of the
filtrate revealed the co-product of the reaction to be Me2Zn.
Initially the formation of a homogeneous solution on the
addition of TMEDA to the insoluble mixture of LiNHDipp
and dimethylzinc in hexane seemed to indicate the formation
(17) Fjeldberg, T.; Hitchcock, P. B.; Lappert, M. F.; Thorne, A. J. Chem.
Commun. 1984, 822.
(18) Bernstein, M. P.; Romesberg, F. E.; Fuller, D. J.; Harrison, A. T.;
1
of a mixed-metal compound. However, H and 13C NMR
Collum, D. B.; Liu, Q-Y.; Williard, P. G. J. Am. Chem. Soc. 1992, 114, 5100.

5324 Inorganic Chemistry, Vol. 48, No. 12, 2009
Clegg et al.
Figure 2. Molecular structure of 4 with 30% probability displacement
ellipsoids. Hydrogen atoms, except for N-H, are omitted for clarity.
Figure 1. Section of the extended chain structure of 1 with 30% prob-
ability displacement ellipsoids. Hydrogen atoms, except for N-H, are
omitted for clarity. Selected bond distances (A) and bond angles (deg):
Li-N1 1.976(2), Li-N1A 2.000(2), Li-N2 2.080(2), N1-Li-N1A
103.00(10), N1-Li-N2 120.59(12), N1A-Li-N2 135.96(12), Li-
N1-LiA 77.00(10), Li-N2-C15 117.83(10). Symmetry operation
A: 1 - x, -y, 1 - z.
˚
Selected bond distances (A) and bond angles (deg): Li-N1 2.230(5), Li-
˚
N2 2.170(5), Li-N3 2.174(5), Li-N4 1.953(5), N1-Li-N2 84.01(18),
N1-Li-N3 114.1(2), N1-Li-N4 112.4(2), N2-Li-N3 85.16(19), N3-
Li-N4 130.8(3), N2-Li-N4 114.8(2), Li-N4-C10 147.6(3).
established through an X-ray crystallographic study
(Figure 2).
TMEDA (as confirmed by the presence of a singlet at -0.57
ppm in the 1H NMR spectrum in deuterated benzene solu-
tion).19 These results indicate that, although the addition of
TMEDA to unsolvated LiNHDipp breaks its oligomeric/
polymeric ladder structure forming the hexane-soluble TME-
DA-solvated compound 1, the degree of aggregation of the
latter is still too high to favor the formation of the mixed-
metal compound by co-complexation with Me2Zn. Surpris-
ingly, even when an excess of TMEDA (3 equiv) is employed,
compound 1 is still obtained, which shows the high stability
of this polymeric arrangement and its low tendency to cleave
into smaller aggregated forms. In part this reflects the
excellent bridging ability of the sterically little demanding
electron-rich primary amido ligand.
We next endeavored to study the co-complexation of
LiNHDipp and Me2Zn using the tridentate donor PMDE-
TA. From the previous results when TMEDA was employed,
the degree of aggregation of the lithium amide after the
addition of the Lewis base appears to play a crucial role in
determining whether the co-complexation reaction will take
place. In general, the number of reported lithium anilide
structures is relatively small, and there is only one documen-
ted example of a PMDETA complex. Reported by Mulvey,
[(PMDETA)2Li3(NHPh)3]20 exhibits an unusual trinuclear
arrangement combining three-, four- and five-coordinated
lithium centers. Since NH2Dipp is much more sterically
demanding than unsubstituted NH2Ph it can be anticipated
that the relevant lithiated amine solvated by PMDETA will
present an even smaller oligomeric form. Thus, when 1 molar
equiv of PMDETA was added to a suspension of LiNHDipp
in hexane a colorless solution was obtained which deposited
colorless crystals overnight of the desired complex [(PMDE-
TA)Li(NHDipp)](4). NMR (1H, 13C and 7Li) analysis
of 4 revealed the formation of a complex with a PMDETA:
NHDipp ratio of 1:1. The molecular structure of 4 was
Contrasting with the polymeric structure of 1, a simple
monomeric arrangement was revealed for 4, made up by a
distorted tetrahedral lithium center (average angle subtended
at Li, 106.87ꢀ) which binds to four nitrogen atoms, one from
the amide ligand NHDipp and three from the chelating
triamine PMDETA. The Li-Namide bond length [1.953(5)
˚
A] is significantly shorter than those found for the dative Li-
˚
NPMDETA bonds [average Li-N bond length, 2.191 A] and
slightly shorter than that found for the Li-Namide in the
˚
TMEDA-adduct 1 [1.976(5) A]. The only precedent for a
structurally defined monomeric lithium anilide is the afore-
mentioned TMEDA-solvate [Li(NHAr)(TMEDA)] (Ar =
2,4,6-tBu3-C6H2)17 where the lithium center is tricoordinated
in a distorted trigonal planar geometry, forming a stronger
(shorter) bond with the nitrogen atom of the amide ligand
˚
[1.895(8) A] as a consequence of the decrease in the coordina-
tion number of the metal center in comparison to that in 4.
Next we considered the reaction of 4 (prepared in situ) with
1 molar equiv of dimethylzinc. This afforded the heterobi-
metallic complex [(PMDETA)LiZn(NHDipp)(Me)2] (2) as a
crystalline solid in an isolated yield of 65%. Multinuclear
7
NMR (1H, 13C and Li) analysis (see Table 2 and Experi-
mental Section) of these crystals proved unambiguously the
mixed-metal constitution of 2. Thus, for example the 1H NMR
spectrum of 2 in deuterated benzene solution showed a singlet
at -0.32 ppm for the methyl groups which appears at a similar
chemical shift to unsolvated dimethyl zinc in the same
deuterated solvent (-0.52 ppm). In addition, as indicated in
Table 2, the 1H NMR spectrum of 2 also showed a significant
change in the chemical shifts of the relevant signals of the
NHDipp anion (7.16, 6.79, 3.45, 2.84, and 1.38 ppm) in
comparison with those found for the homometallic PMDE-
TA-lithium amide 4 (7.29, 6.66, 3.34, 3.09, and 1.54 ppm).
Furthermore the 7Li NMR spectrum of 2 revealed a singlet at
0.77 ppm, which is considerably different from that observed
for 4 (1.47 ppm). Compound 2 was also subjected to an X-ray
crystallographic study which established its molecular
structure (Figure 3). Unlike the majority of the lithium
dialkylamidozincates previously structurally characterized
which exhibit a closed four-membered [LiNZnC] motif,7b,19
(19) Graham, D. V.; Hevia, E.; Kennedy, A. R.; Mulvey, R. E. Organo-
metallics 2006, 25, 3297.
(20) Barr, D.; Clegg, W.; Cowton, L.; Horsburgh, L.; Mackenzie, F. M.;
Mulvey, R. E. Chem. Commun. 1995, 891.

Article
Inorganic Chemistry, Vol. 48, No. 12, 2009 5325
soluble in hexane at room temperature, and so far attempts to
determine its structure in the solid state by X-ray crystal-
lography have been unsuccessful. That notwithstanding,
its constitution could be unequivocally determined by
a combination of 1H, 13C, and 7Li NMR spectroscopic
studies (see Experimental Section and Table 2). The pre-
sence of lithium was confirmed by a singlet (1.01 ppm) in the
7Li NMR spectrum of 3 in deuterated benzene solution. Its
1H NMR spectrum revealed a singlet at -0.42 ppm for the
methyl groups attached to zinc, the relevant signals for the
amido group (7.14, 6.75, 3.39, 2.86, and 1.39 ppm) at similar
chemical shifts to those found for the related zincate 2 and
sufficiently different from those observed for the
THF-solvated lithium amide (7.16, 6.61, 3.38, 2.79, and
1.40 ppm). In addition, the 1H NMR spectrum showed two
multiplets at 3.28 and 1.30 ppm corresponding to the THF
molecules, and an integration ratio of 3:2:1 with respect to the
THF/Me/NHDipp ligands. When a solution of 3 in hexane
was left at room temperature for 2 weeks it deposited color-
less crystals which displayed a much lower solubility than the
initial crystals of 3 obtained when the solution of hexane was
placed in the freezer (-20 ꢀC). An X-ray crystallographic
determination revealed that these crystals have the unprece-
dented constitution monolithium-dizinc trialkyl-bis(amido)
zincate [(THF)3LiZn2(Me)3(NHDipp)2] (5). This compound
has a unique contacted ion pair structure with a trinuclear
Li Zn Zn chain where the metals are connected exclu-
Figure 3. Molecular structure of 2 with 30% probability displacement
ellipsoids. Hydrogen atoms, except for N-H, are omitted for clarity.
˚
Selected bond distances (A) and bond angles (deg): Zn-C1 1.984(2), Zn-
C2 1.999(2), Zn-N1 2.0869(13), Li-N1 2.055(3), Li-N2 2.220(3), Li-
N3 2.152(3), Li-N4 2.149(3), Li C2 4.230(3), C1-Zn-C2 136.04(9),
3 3 3
C1-Zn-N1 114.55(7), C2-Zn-N1 108.97(7), N1-Li-N2 121.88(13),
N1-Li-N3 114.24(13), N1-Li-N4 118.23(13), N2-Li-N3 84.48(11),
N3-Li-N4 87.72(11), N2-Li-N4 116.94(12), Li-N1-C3 110.96(12).
2 presents an open arrangement where both metals are con-
nected by an amido NHDipp bridge, the span lengths of
˚
˚
which are similar [Li-N1 2.055(3) Zn-N1 2.0869(13) A].
A,
Both methyl groups are bonded solely to zinc, one at a modestly
3 3 3
3 3 3
elongated distance compared with the other [Zn-C2 1.999
sively by the amido ligands while the methyl groups bond
terminally to the zinc centers. The stoichiometry of this “zinc-
rich” compound (1Li:2Zn:5R, R = monoanionic ligand)
contrasts markedly with those found in previously reported
synthetic useful triorganozincates (1Li:1Zn:3R) and lithium-
rich tetraorganozincates (2Li:1Zn:4R).22 Tetraorganozin-
cates are closely related to 5 since they are also trinuclear
compounds.23 However, unlike 5, they usually exhibit near-
linear Li Zn Li arrangements, as for example in
˚
(2), Zn-C1 1.984(2) A]. The lithium center adopts a distorted
tetrahedral geometry (average angle around Li 107.24ꢀ) by
bonding to the tridentate amine PMDETA and the amido
ligand NHDipp, in an almost identical fashion to that in
homometallic 4. While the Li-NPMDETA bond lengths in 2
˚
[average Li-N bond length, 2.173 A] are similar to those
˚
found in 4, the Li-N amide bond in 2 [2.055 A] is remarkably
˚
longer [for 4, Li-Namide: 1.953(5) A]. This can be rationalized
3 3 3
3 3 3
[Li2ZnMe4] [Li Zn Li, 162.48(17)o]23 and [(TMEDA)2-
in terms of the change in the coordination mode of the amido
ligand which is now also bonded to zinc and this also affects
the value of the Li-Namide-C bond angle which becomes
substantially more acute [110.96(12)o in 2, 147.6(3)o in 4]. As
previously mentioned, no interaction is observed between the
3 3 3 3 3 3
Li2Zn(CtCPh)4] [Li Zn Li, 146.70ꢀ],24 and each anio-
3 3 3
3 3 3
nic ligand acts as a bridge between two metals; thus, the zinc
center bonds to four anionic ligands in a distorted tetrahedral
environment. In 5, the arrangement Li Zn Zn is sig-
3 3 3
3 3 3
3 3 3
nificantly distorted from linearity [Li Zn Zn, 115.7(2)o];
lithium center and either of the methyl groups [closest Li
C
3 3 3
3 3 3
21
˚
separation 4.230(3) A]. This distance is too long even to
suggest any type of secondary agostic interaction. This open
structural motif is reminiscent of that previously reported by
us for the related HMDS-lithium zincate [(PMDETA)LiZn-
(HMDS)(Me)2],7b although in this case the amide occupies a
terminal position, solely bonded to zinc, and both metals are
connected by a single methyl bridge. This is probably due to
the dramatic increase in the steric bulk in the amide, since in
HMDS the nitrogen atom is bonded to two bulky trimethyl-
silyl groups.
in addition each zinc center displays a trigonal planar
geometry, bonded to two bridging amido ligands and a
terminal methyl group for Zn1 located in the center of
the chain and bonded to one bridging amido ligand and
two terminal methyl groups for terminal Zn2 (Figure 4).
The trinuclear arrangement is completed by a distorted
tetrahedral lithium atom, which lies almost coplanar with the
N1-Zn1-N2 plane [Li1-N1-Zn1-N2, 175.4 (2)o] and
binds to three THF molecules and the amide ligand bonded
to Zn1, but shows no interaction with Zn2. The Zn1-N bond
We subsequently investigated the co-complexation reac-
tion of Me2Zn with LiNHDipp using the monodentate Lewis
base THF. Thus, when both monometallic reagents were
combined in hexane in the presence of 4 molar equiv of the
cyclic ether a colorless solution was obtained which deposited
colorless crystals of the lithium zincate [(THF)3LiZn-
(NHDipp)(Me)2] (3) at -20 ꢀC. This compound is extremely
˚
˚
lengths [Zn1-N1 2.021(2) A, Zn1-N22.008(3) A] are similar
to those found in the related dimethyl(amido)zincate 3 [Zn-
˚
N, 2.0869(13) A] and in the zinc alkyl amido compound
[{(Me3SiCH2)Zn(NHdipp)}2] (average Zn-N bond length,
(22) For recent applications of tetraorganozincates in organic synthesis
see: Furuyama, T.; Yonehara, M.; Arimoto, S.; Kobayashi, M.; Matsumoto,
Y.; Uchiyama, M. Chem.;Eur. J. 2008, 11, 10348.
(23) Armstrong, D. R.; Dougan, C.; Graham, D. V.; Hevia, E.; Kennedy,
A. R. Organometallics 2008, 27, 6063.
(21) This Li-C separation distance is markedly longer than that found in ,
˚
for example, [(LiMe.THF)4] (average bond length, 2.235 A), see: Ogle, C. A.;
Huckabee, B. K.; Johnson, H. C., IV; Sims, P. F.; Winslow, S. D.; Pinkerton,
A. A. Organometallics, 1993, 12, 1960.
(24) Edwards, A. J.; Fallaize, A.; Raithby, P. R.; Rennie, M. A.; Steiner,
A.; Verhorevoort, K. L.; Wright, D. S. Dalton Trans. 1996, 133.

5326 Inorganic Chemistry, Vol. 48, No. 12, 2009
Clegg et al.
Scheme 3
observetwodifferent signals forboth of these types ofanionic
ligand. Thus, the NMR analysis of 5 in deuterated benzene
suggests that, in solution, there must be a dynamic process
taking place which interconverts the two different types of
methyl and amido groups. Such a possibility would explain
the broad appearance of the signals observed in the 1H and
13C{H} NMR spectra. A possible process would involve the
cleavage of the Zn-N bond of the zinc center that is bonded
to two methyl groups. As previously mentioned, in the
analysis of its molecular structure, in the solid state this
Zn-N bond is considerably elongated (implying weakness)
in comparison with the Zn-N bond lengths of the other zinc
center. If this Zn-N bond was breaking/reforming fast in
solution it should lead to a dynamic equilibrium in solution
between 5 and the methyl(bis-amido) lithium zincate [LiZn-
(NHDipp)2Me] and Me2Zn (Scheme 3).28 This hypothesis is
strongly supported by the fact that the NMR analysis of the
solid residue obtained after dissolving crystals of 5 in
toluene and then removing the solvent under vacuum
proved to be the monolithium-monozinc zincate [(THF)-
LiZn(NHDipp)2Me] (6). Thus, its 1H NMR shows a singlet
at -0.38 ppm for the Zn-Me group, the integration of which
with respect to theamido ligand NHdipp (whichappears now
at 7.06, 6.84, 3.34, 3.18, and 1.32 ppm) has changed to 1:2. In
addition, only one molecule of coordinated THF (at 2.85 and
0.98 ppm) is present in the mixed metal compound and the
7Li NMR spectrum exhibits only a singlet at 1.40 ppm. The
formation of 6 can be rationalized in terms of the high
volatility of dimethylzinc; thus, it will be removed under
vacuum at the same time as the solvent, which will then push
the equilibrium depicted in Scheme 3 toward the right-hand
side and the formation of 6.
Figure 4. Molecular structure of 5 with 30% probability displacement
ellipsoids. Hydrogen atoms, except for N-H, and minor disorder com-
ponents are omitted for clarity. Selected bond distances (A) and bond
˚
angles (deg): Li-N1 2.074(6), Zn1-N1 2.021(2), Zn1-C25 1.969(3),
Zn1-N2 2.008(3), Zn2-N2 2.139(3), Zn2-C38 1.989(3), Zn2-C39
1.983(3), Li-O1 1.961(6), Li-O2 2.007(6), Li-O3 1.967(6), Li C25
3 3 3
4.275(12), Li-N1-Zn1 122.0(2), N1-Zn1-C25 125.79(13), C25-Zn1-
N2 131.66(13), N2-Zn1-N1 102.45(10), Zn1-N2-Zn2 101.31(11),
C39-Zn2-C38 140.31(14), C38-Zn2-N2 109.11(12), C39-Zn2-N2
110.58(12).
2.035 A).25 Remarkably, the interaction of the terminal zinc
˚
(Zn2), which lies almost perpendicular to the N1-Zn-N2
plane [N1-Zn1-N2-Zn2, 78.72(12)o], with the bridging
amido ligand is significantly weaker as shown by the
˚
elongated Zn2-N2 bond [2.139(3) A]. Closer literature
analogs, in terms of stoichiometry, to this novel structural
motif are the cyclopentadienyl sodium zincate [{Na-
(THF)6}+{Zn2Cp5}-]26 which exhibits a solvent-separated
ion pair structure and the bimetallic lithium bis(zinc) diamide
[(Et 2O)Li(μ-L)Zn(μ-L)Zn(LH)] where L is the bidentate 1,2-
benzenebis(neopentylamido) ligand and eachmetalexhibits a
distorted tetrahedral geometry.27
Compound 5 was also characterized in solution using
multinuclear NMR (1H, 13C and 7Li) spectroscopy (Table 2
and Experimental Section). The 1H NMR spectrum in
deuterated benzene solution showed the presence of a single
resonance for the methyl groups (-0.43 ppm) and only one
set of signals for the amido groups (7.12, 6.83, 3.36, 3.08, and
1.35 ppm) whose integration indicated a 3:2, Me/NHDipp
ratio. In addition these resonances were remarkably broader
than those found for the related compounds 2 and 3 that also
Regarding the unexpected formation of 5, a reasonable
assumption would be that compound 3 partially decomposes
when dissolved in hexane solution and stored at room
temperature for long periods of time. A plausible formation
pathway shown in Scheme 4 involves the dimerization of two
molecules of 3. Previously we have demonstrated that by
reducing the steric bulk in secondary amides, alkyl(amido)
zincates tend to adopt structures where the zinc centers are
tetracoordinated.23 Thus, the formation of the putative
intermediate shown in brackets in Scheme 4 is plausible since
NHDipp, which is a primary amide, is much less sterically
demanding than, for example, TMP or DA. This dimeric
1
contain the same set of alkyl and amido ligands. The H
NMR spectrum also displayed two multiplets at 3.27 and
1.26 ppm corresponding to three molecules of coordinated
THF. The 13C{1H} NMR spectrum showed a broad signal at
-9.09 ppm for the methyl groups and, as observed in the 1H
NMR spectrum the resonances related to the amido ligand
were also broad. Furthermore, a single signalat 1.07 ppmwas
observed in the 7Li NMR spectrum. Considering the struc-
ture of 5 in the solid state there are two different amido
groups (one that connects Li and one of the Zn centers, and
another that connects the two zinc atoms) and also two
different types of methyl groups; thus one would expect to
intermediate could then undergo elimination of MeLi 3THF
3
to yield 5.29 NMR analysis of the filtrate of the solution
(28) The presence of a dynamic process in solution was confirmed by a
variable temperature NMR study, which shows that the relevant resonance
for the methyl groups of 5 (at -0.33 ppm at T = 300 K in deuterated toluene)
splits into two singlets (at 0.15 and 0.05 ppm, relative integration 2:1) when
the temperature is decreased to 208 K.
(29) To study if the conversion of 5 into 3 is reversible, isolated crystals of
(25) Olmstead, M. M.; Grigsby, W. J.; Chacon, D. R.; Hascall, T.; Power,
5 and one molar equiv of MeLi 3THF were dissolved in C6D6 solution. 1H
P. P. Inorg. Chim. Acta 1996, 251, 273.
(26) Alvarez, E.; Grirrane, A.; Resa, I.; del Rio, D.; Rodriguez, A.;
3
and 7Li NMR revealed that instead of compound 3, this reaction
ꢀ
formed
a mixture of [(THF)LiZn(NHDipp)2Me](6) and homoleptic
Carmona, E. Angew. Chem., Int. Ed. 2007, 46, 1296.
(27) Hitchcok, P. B.; Lappert, M. F.; Wei, X.-H. Dalton Trans. 2006, 1181.
[(THF)2LiZnMe3].

Article
Inorganic Chemistry, Vol. 48, No. 12, 2009 5327
Scheme 4
now much more reactive and therefore can react successfully
with dimethylzinc to yield to the formation of dimethyl
(amido)zincate [(PMDETA)LiZn(NHDipp)(Me)2] (2) which
displays a rare open motif where both metals are solely
connected by the amido ligand and both methyl groups
coordinate exclusively to zinc. Moving to a monodentate
donor, when THF is employed the reaction yields the zincate
[(THF)3LiZn(NHDipp)(Me)2] (3). This compound is found
to partially decompose in hexane solution after long periods
at room temperature, affording the unprecedented bis(ami-
do)trimethylzincate [(THF)3LiZn2(Me)3(NHDipp)2] (5).
This zinc-rich compound can also be prepared in good yields
by reacting the lithium amide with 2 molar equiv of dimethyl-
zinc, an extra equivalent of the amine NH2Dipp, and 3 equiv
of THF in hexane solution. Compound 5 exhibits a trinuclear
Li Zn Zn chain structure where the metals are con-
3 3 3
3 3 3
nected by the amido ligands and the methyl groups are
exclusively bound to the zinc centers. In the solid state
structure, the outer zinc of the chain is coordinated by the
amido ligand less strongly than the other zinc center present
in the molecule as revealed by the elongation of this Zn-N
bond. In solution in deuterated benzene, the weakness in this
Zn-N bond translates into the existence of a dynamic
equilibrium between 5 and the bis(amido)methylzincate
showed that the major species remaining in solution was
zincate 3 which indicated that the disproportionation process
that affords 5 occurs only to a minor extent. These results
prompted us to investigate a rational synthetic route to
prepare 5. Thus, we found that, when the lithium amide
LiNHDipp was combined with 2 molar equiv of dimethyl-
zinc, 1 equiv of the primary amine NH2Dipp, and 3 equiv of
THF in hexane solution and the mixture was heated under
reflux for 1 h, the reaction afforded a colorless solution. This
solution deposited 5 as a colorless microcrystalline solid in
good isolated yields (58-64%).
[(THF)LiZn(NHDipp)2Me] (6) and Me2Zn 2THF which
3
can be driven toward the formation of 6 by removing volatile
Me2Zn 2THF under vacuum. These results show the com-
3
plicated chemistry that lithium zincates display in solution
and also the great influence that the donor solvent employed
has for the co-complexation reactions to succeed illustrating
that the mere mixing-up of the monometallic components of
a zincate do not always secure the formation of the desired or
expected bimetallic compound.
Conclusions
The final outcome of the interlocking co-complexation
reactions of dimethylzinc with the primary lithium amide
LiNHDipp is dictated by the donor ligand employed and the
type of structure that it forms when coordinated to the
lithium amide. Thus, when the chelating ligand TMEDA is
employed, [{Li2(NHDipp)2(TMEDA)}¥] (1) is obtained,
where Li2N2 rings are connected by TMEDA bridges giving
rise to a polymeric chain. This oligomeric form of the amide is
too stable to react with dimethylzinc and so no mixed-metal
compound is obtained even in the presence of an excess of
TMEDA. In contrast, the tridentate ligand PMDETA is
found to form a monomeric complex when coordinated to
LiNHDipp, namely, [(PMDETA)Li(NHDipp)] (4), which is
Acknowledgment. We thank the Royal Society (Uni-
versity Research Fellowship to E.H.), the Faculty of
Science, University of Strathclyde (starter grant to E.
H.), and EPSRC (DTA award to M.M.) for their gener-
ous sponsorship of this research.
Supporting Information Available: Crystal data in CIF for-
mat. This material is available free of charge via the Internet at
http://pubs.acs.org.