Ligand and coligand effects on ion association in magnesium amides

Article abstract of DOI:10.1021/om101045b

We report here the examination of the effect of ligand bulk and coligand influence on ion association in a family of magnesium amides. Use of the sparsely employed, sterically demanding coligand HMPA (hexamethylphosphoramide, O=P(N(Me)₂)_3<

Full text of DOI:10.1021/om101045b

986 Organometallics 2011, 30, 986–991
DOI: 10.1021/om101045b
Ligand and Coligand Effects on Ion Association in Magnesium Amides
Ana Torvisco and Karin Ruhlandt-Senge*
Department of Chemistry, CST 1-014, Syracuse University, Syracuse, New York 13244, United States
Received November 5, 2010
We report here the examination of the effect of ligand bulk and coligand influence on ion asso-
ciation in a family of magnesium amides. Use of the sparsely employed, sterically demanding
coligand HMPA (hexamethylphosphoramide, OdP(N(Me)2)3, in conjunction with four different
silylamides [N(SiMe3)(R)]- (R = SiMe3, C6H5 (Ph), 2,4,6-MesC6H2 (Mes), 2,6-i-Pr2C6H3 (Dipp))
provides unique insight into factors governing ion association. To this effect, we report three contact
molecules, along with one example of a rare ion-association state with one amido group metal-bound,
whereas the other is unassociated.
Introduction
Structural elucidations of magnesium [N(SiMe3)2]- based
amido compounds date back to the early 1990s, with reports
on the synthesis and characterization of monomeric Mg(N-
(SiMe3)2)(donor)x and coligand-free dimeric [Mg(N(Si-
Me3)2)2]2 species. These compounds have developed into
versatile starting materials and continue to serve as a spring-
board for more complex systems. The steric demand of the
[N(SiMe3)2]- ligand effectively suppresses aggregation, while
the presence of -SiMe3 substituents ensures favorable solubi-
lity in a range of solvents. Extensive studies of these compounds
have examined their utility as reagents, polymerization initia-
tors, MOCVD precursors, and more.4,9-13 Limited studies
have examined the role of ligand size and electronics by
replacing -SiMe3 substituents by variously sized aryl groups;
a selection of these ligands is shown in Figure 1.
Analogous to the [N(SiMe3)2]- ligand, [N(SiMe3)(Mes)]-,
ligand c, led to a four-coordinate monomeric compound,
Mg[N(SiMe3)(Mes)]2(thf)2, with features very similar to
those of Mg[N(SiMe3)]2(thf)2.14,15 However, use of the
slightly bulkier ligand d, [N(SiMe3)(Dipp)]-, resulted in
formation of a rare two-coordinate species, Mg[N(SiMe3)-
(Dipp)]2, available by crystallization from a nonpolar
solvent.16 In the case of ligand d, we see that steric bulk has
a significant impact on kinetic metal stabilization, effectively
suppressing aggregation.
Since the discovery of Grignard reagents in 1900,1,2 the
organometallic chemistry of magnesium has progressed in leaps
and bounds.3 Impressively, new applications for these reagents
are being discovered continuously, making this family of com-
pounds one of the most versatile in the Periodic Table. Organo-
magnesium species have been the most intensively studied;3 less
work has been devoted to the amido derivatives,4 although the
compounds have been established as important, versatile re-
agents. As an example, recent, remarkable reports on hetero-
bimetallic alkali-metal/magnesium amido species illustrate their
unique reactivity, which is markedly different from that of the
homometallic counterparts.5-8
As new applications for organomagnesium reagents con-
tinue to be discovered, the need to tailor ligand systems to
specific applications and to understand structure and function
relationships remain a crucial part of the further development
of magnesium chemistry. Several factors influence the proper-
ties of the reagents, including the type and size of the ligand, the
size, hapticity, and donor strength of the coligands, and
second-order bonding (including Mg-π and agostics), as well
as ion association. Among those factors, ligand characteristics
have been the most extensively examined, but significantly less
is known about the other variables.
*To whom correspondence should be addressed. E-mail: kruhland@syr.
edu. Tel: 315-443-1306. Fax: 315-443-4070.
(1) Elschenbroich, C. Organometallics; Wiley-VCH: Weinheim,
(9) He, X.; Morris, J.; Noll, B.; Brown, S. N.; Henderson, K. W.
J. Am. Chem. Soc. 2006, 128, 13599.
(10) Chamberlain, B. M.; Cheng, M.; Moore, D. R.; Ovitt, T. M.;
Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc. 2001, 123, 3229.
(11) Sarazin, Y.; Schormann, M.; Bochmann, M. Organometallics
2004, 23, 3296.
Germany, 2006.
(2) Hill, A. E. Grignard Reagents New Developments; Wiley and Sons:
Chichester, U.K., 2000.
(3) Alexander, J. S.; Zuniga, M. F.; Guino-o, M. A.; Hahn, R. C.;
Ruhlandt-Senge, K. The Organometallic Chemistry of the Alkaline
Earth Metals. In The Encyclopedia of Inorganic Chemistry; Wiley:
New York, 2006; Vol. 1, p 116.
(4) Westerhausen, M. Coord. Chem. Rev. 1998, 176, 157.
(5) Mulvey, R. E. Chem. Commun. 2001, 1049.
(6) Mulvey, R. E. Organometallics 2006, 25, 1060.
(7) He, X.; Allan, J. F.; Noll, B. C.; Kennedy, A. R.; Henderson,
K. W. J. Am. Chem. Soc. 2005, 127, 6920.
(12) Chisholm, M. H.; Gallucci, J. C.; Phomphrai, K. Inorg. Chem.
2005, 44, 8004.
(13) Winter, C. H.; Sebestl, J. L.; Heeg, M. J. Mater. Res. Soc. Symp.
Proc. 1998, 495, 147.
(14) Gillett-Kunnath, M.; Teng, W.; Vargas, W.; Ruhlandt-Senge, K.
Inorg. Chem. 2005, 44, 4862.
(15) Bradley, D. C.; Hursthouse, M. B.; Ibrahim, A. A.; Abdul,
A. M.; Motevalli, R.; Moseler, R.; Powell, H.; Runnacles, J. D.;
Sullivan, A. C. Polyhedron 1990, 9, 2459.
(8) Krasovskiy, A.; Straub, B. F.; Knochel, P. Angew. Chem., Int. Ed.
2006, 45, 159.
(16) Vargas, W.; Englich, U.; Ruhlandt-Senge, K. Inorg. Chem. 2002,
41, 5602.
r
pubs.acs.org/Organometallics
Published on Web 02/11/2011
2011 American Chemical Society

Article
Organometallics, Vol. 30, No. 5, 2011 987
reactivity when used in organic reactions.27,28 This dynamic
behavior in solution allows for difficulty in isolating stable
species in the solid state.
We report here on a family of magnesium amides based on
the [N(SiMe3)(R)]- ligand system (R = SiMe3, C6H5 (Ph),
2,4,6-MesC6H2 (Mes), 2,6-i-Pr2C6H3 (Dipp)) in conjunction
with HMPA, allowing the investigation of ion association
in the context of steric saturation. Contact molecules are
observed for Mg[N(SiMe3)2]2(hmpa)2 (1), Mg[N(SiMe3)-
(Ph)]2(hmpa)2 (2), and Mg[N(SiMe3)(Mes)]2(hmpa)2 (3), in
addition to the unique, partially separated [Mg{N(SiMe3)-
(Dipp)}(hmpa)3][N(SiMe3)(Dipp)] (4).
Figure 1. Illustration of various silyl and aryl silylamine
ligands.
Scheme 1. Ion-Association Modes for Alkaline-Earth Metals
Experimental Details
General Procedures. All reactions were carried out under strict
inert gas conditions using a Braun Labmaster 100 drybox and/
or modified Schlenk techniques. The solvents, including hex-
anes, toluene, benzene, and tetrahydrofuran (THF), were dried
on a Vacuum Atmospheres Co. Dri-Solv solvent purifier system
and degassed just prior to use. Dibutylmagnesium (statistical
mix of n- and sec-butyl, 1 M solution in heptanes) was obtained
from a commercial source and used as received. HN(SiMe3)2,
HN(SiMe3)(Ph), and HMPA (hexamethylphosphoramide) were
purchased from Aldrich, while HN(SiMe3)(Mes), HN(SiMe3)-
(Dipp) were synthesized according to literature procedures.29 All
ligandsanddonors were driedanddistilledoverCaH2 priortouse.
IR spectra were obtained as Nujol mulls on a Nicolet IR200
spectrometer. 1H NMR and 13C NMR were recorded on a Bruker
DPX-300spectrometerat25ꢀCinC6D6 and referenced toresidual
solvent peaks. Melting points were collected in sealed capillary
tubes and are uncorrected. Due to the pyrophoric nature of these
compounds, satisfactoryelemental analysis could not be obtained.
This is a well-established problem with alkali-metal and alkaline-
earth-metal organometallics.30,31
Single-Crystal X-ray Diffraction Studies. X-ray-quality crystals
for compounds 1-4 were grown as described below. The crystals
were removed from the Schlenk tube under a stream of N2 and
immediately covered with a layer of viscous hydrocarbon oil
(Infineum). A suitable crystal was selected with the aid of a
microscope, attached to a glass fiber, and immediately placed in
the low-temperature nitrogen stream of the diffractometer.32 The
intensity data sets for all compounds were collected using a
Bruker SMART system, complete with a three-circle gonio-
meter and an APEX-CCD detector. Data for compounds 1-4
were collected at 97, 99, 97, and 96 K, respectively, using a custom-
built low-temperature device from Professor H. Hope (UC Davis).
Further data collection, structure solution, and refinement
details have been reported previously.33,34 Crystallographic data
(excluding structure factors) for the structures reported herein
(Table 1) have been deposited in the Cambridge Crystallographic
Data Center (CCDC deposition numbers 794115-794117 for
compounds 2-4). CIF files can be obtained from the CCDC free
of charge via http://www.ccdc.cam.ac.uk/data_request/cif. Dis-
order was typically handled by including split positions for the
A critical but not well-understood factor governing the re-
activity of organomagnesium species is ion association. For
alkaline-earth-metal species, three ion-association modes
may be observed (Scheme 1), contact molecules, separated
ions, and an intermediate state, with one ligand unassociated
and the other bound to a metal.
An important factor in determining ion association is the
metal-ligand bond character, with more covalently bound
ligands less likely to dissociate. As a result, the formation of
separated ions is more pronounced for the heavier s-block
metals. Furthermore, the capacity of ligand charge deloca-
lization plays a major role, as does the ability for the metal’s
steric saturation as achieved through a combination of
factors. With a relatively high charge density for magnesium,
only a few examples for separated magnesium ions have been
reported, including [Mg(15-crown-5)(pyr)2][C5H5]2, [Mg-
(15-crown-5)(thf)2][C5H5]2, [Mg(C5H5)(pmdta)][C5H5], and
[Mg(15-crown-5)(thf)2][SMes*]2 (Mes*=2,4,6-tri-tert-butyl-
phenyl).17,18 Markedly absent from this list, however, are
separated amido species.
We report here on studies utilizing HMPA, a strong Lewis
donor with significant steric bulk. Notably, amido compounds
containing HMPA are much less common, limited to a handful
of magnesium compounds and one heavier calcium silylamido
compound.19-23 The lack of known HMPA early-s-block
derivatives is attributed to the donor’s capacity for inducing
ion dissociation in solution and its dependency on donor con-
centration.24-26 For example, addition of HMPA to organo-
lithium reagents has been shown to typically increase their
(17) Jaenschke, A.; Paap, J.; Behrens, U. Z. Anorg. Allg. Chem. 2008,
634, 461.
(18) Chadwick, S.; Englich, U.; Ruhlandt-Senge, K. Inorg. Chem.
1999, 38, 6289.
(19) Olmstead, M. M.; Power, P. P. Inorg. Chim. Acta 1996, 251, 273.
(20) Clegg, W.; Craig, F. J.; Henderson, K. W.; Kennedy, A. R.;
Mulvey, R. E.; O’Neil, P. A.; Reed, D. Inorg. Chem. 1997, 36, 6238.
(21) Yang, K.-C.; Chiang, M. Y. J. Organomet. Chem. 2002, 648, 176.
(22) Clegg, W.; Mulvey, R. E. Polyhedron 1998, 17, 1923.
(23) Tang, Y.; Kemp, R. A. Inorg. Chim. Acta 2005, 358, 2014.
(24) Reich, H. J.; Green, D. P. J. Am. Chem. Soc. 1989, 111, 8729.
(25) Reich, H. J.; Borst, J. P.; Dykstra, R. R.; Green, D. P. J. Am.
Chem. Soc. 1993, 115, 8728.
(28) Clayden, J.; Yasin, S. A. New J. Chem. 2002, 26 (2), 191.
(29) Chao, Y.-W.; Wexler, P. A.; Wigley, D. E. Inorg. Chem. 1989, 28,
3860.
(30) Andrews, P.; Henderson, K.; Ruhlandt-Senge, K. Comprehen-
sive Organometallic Chemistry III; Elsevier: Oxford, U.K., 2006; Vol. II.
(31) Hays, M. L.; Hanusa, T. P.; Nile, T. A. J. J. Organomet. Chem.
1996, 514, 73.
(32) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112.
(33) Sheldrick, G. M. SHELX-97, Program for crystal structure
€
solution and refinement; University of Gottingen, Gottingen, Germany,
1997.
€
(26) Reich, H. J.; Green, D. P.; Medina, M. A.; Goldenberg, W. S.;
€
Gudmundsson, B. O.; Dykstra, R. R.; Phillips, N. H. J. Am. Chem. Soc.
1998, 120, 7201.
(27) Clayden, J. Organolithiums: Selectivity for Synthesis; Pergamon:
Amsterdam, Boston, 2002.
(34) Sheldrick, G. M. SADABS, Program for scaling and absorption
€
€
correction of area detector data; University of Gottingen, Gottingen,
Germany, 1997.

988 Organometallics, Vol. 30, No. 5, 2011
Table 1. Crystallographic Dataa for Compounds 2-4
Torvisco and Ruhlandt-Senge
2
3
4
formula
fw
MgN8Si2O2P2C30H64
711.32
13.85(8)
16.67(9)
18.84(9)
90
109.20(10)
90
4111.1(4)
4
P21/n
MgN8Si2O2P2C36H76
MgN11Si2O3P3C48H106
795.48
11.69(8)
11.69(8)
34.79(3)
90
90
90
4748(6)
4
1058.84
11.41(2)
21.72(4)
12.80(3)
90
100.03(3)
90
3127.3(6)
2
P21
˚
a (A)
˚
b (A)
˚
c (A)
R (deg)
β (deg)
γ (deg)
V (deg)
Z
space group
dcalcd (Mg/m3)
P4122
1.113
0.193
1.149
0.215
99(2)
1.124
0.188
96(2)
μ (mm-1
T (K)
)
97(2)
2θ range (deg)
no. of indep rflns
no. of params
R1, wR2 (all data)b
R1, wR2 (>2σ)b
1.67-25.00
7242
412
0.0516, 0.1375
0.0434, 0.1308
P
1.74-25.00
4152
240
0.0893, 0.1303
0.0577, 0.1159
P
2.61-28.07
10 956
631
0.0653, 0.1354
0.0570, 0.1311
P
P
Mo KR radiation (λ = 0.710 73 A). b R1 = ||Fo| - |Fc||/ |Fo|; wR2 = [ w(Fo2 - Fc2)2/ w(Fo2)2]1/2
.
a
˚
affected groups and included the refinement of the respective
occupancies. Compound 1 was found to contain a high degree of
positional disorder and refined using split positions for both ligands
(60/40) and HMPA molecules (70/30); however, important struc-
tural features were able to be elucidated.
Typical Experiment. nBu2Mg (1 mL, 1 mmol) was addedslowly
via syringe to a solution of the secondary amine (2 mmol) in 40 mL
hexanes. This was refluxed overnight to afford a clear solution.
The solution was then reduced in volume. A slight excess of
HMPA (2.5 mmol, 0.44 mL) was added via syringe to create a
cloudy suspension. Suitable crystals (colorless blocks) deposited at
room temperature.
(o-CH*) 114.7 (m-CH*), 122.4 (p-CH), 123.7 (o-CH), 144.4
(m-CH), 145.6 (i-C), 146.67 (i-C*). IR (cm-1): ν 2965 s, 2732 m,
2542 s, 1569 m, 1446 s, 1364 s, 1323 s, 1248 m, 1210 w, 1173 w,
922w, 832 w, 738 w, 479 w, 345 w.
Results
Synthetic Aspects. All compounds were synthesized using
alkane elimination, a reaction route utilized to synthesize a
variety of magnesium compounds in both high yield and
purity.14,16,35-38 This route is attractive due to the commer-
cial availability of dibutylmagnesium (n-/sec-Bu2Mg), straight-
forward reaction conditions, and ease of workup (eq 1). These
reactions have been shown to be solvent dependent; perform-
ing these reactions in the presence of polar solvents frequently
led to incomplete reactions.14,16 To avoid these issues, all
reactions were conducted in hexane. Donor adducts were easily
obtained by the addition of a donor to the hexane solutions
after the reaction took place.14,16
Mg[N(SiMe3)2]2(hmpa)2 (1). (2 mmol, 0.32 g, 0.42 mL) Yield:
1
0.99 g, 73%. Mp: 131-135 ꢀC. H NMR (300 MHz, 25 ꢀC,
C6D6): δH 0.549 (s, 36H, -SiCH3), 2.32 (d, 36H, HMPA). 13
C
NMR (300 MHz, 25 ꢀC, C6D6): δC 7.743 (-SiCH3), 37.21 (CH3
HMPA). IR (cm-1): ν 2916 s, 2882 m, 2723 m, 1450 s, 1377 s,
1298 m, 1210 w, 1182 w, 988 w, 902 s, 823 w, 360 w.
Mg[N(SiMe3)(Ph)]2(hmpa)2 (2). (2 mmol, 0.33 g, 0.35 mL)
Yield: 0.96 g, 63%. Mp:160-169 ꢀC. 1H NMR(300 MHz, 25 ꢀC,
C6D6): δH 0.618 (s, 18H, -SiCH3), 2.21 (d, 36H, HMPA), 6.03
(m, 4H, o-CH), 6.76 (m, 4H, p-CH), 7.28 (m, 8H, m-CH). 13C
NMR (300 MHz, 25 ꢀC, C6D6): δC 4.10 (-SiCH3), 36.71 (CH3
HMPA), 118.3 (m-CH), 126.0 (p-CH), 129.9 (o-CH), 162.1 (i-C).
IR (cm-1): ν 2953 s, 2723 m, 1574 m, 1458 s, 1372 s, 1303 s, 1258
m, 1210 w, 1172 w, 919 w, 823 w, 718 w, 489 w.
HMPA f
n-=sec-Bu2Mg þ 2HNðSiMe3ÞðRÞ
hexane; 65 ꢀC
Mg½NðSiMe3ÞðRÞꢀ2ðhmpaÞ2 þ 2BuH
ð1Þ
1 - 3
Mg[N(SiMe3)(Mes)]2(hmpa)2 (3). A white precipitate was
isolated and redissolved in toluene. The solution was filtered,
and crystals (colorless blocks) deposited at room temperature.
(2 mmol, 0.49 g, 0.45 mL) Yield: 0.94 g, 59%. Mp: 167-174 ꢀC.
1H NMR (300 MHz, 25 ꢀC, C6D6): δH 0.42 (s, 18H, -SiCH3),
2.12 (d, 6H, p-CH3), 2.17 (s, 12H, o-CH3), 2.44 (d, 36H, HMPA),
6.84 (s, 4H, m-CH). 13C NMR (300 MHz, 25 ꢀC, C6D6): δC 4.5
(-SiCH3), 36.71 (CH3 HMPA), 20.9 (p-CH3), 21.7 (o-CH3),
125.0 (p-CH), 129.1 (m-CH), 135.0 (o-CH), 153.6 (i-C). IR
(cm-1): ν 2957 s, 2723 m, 2552 s, 1569 m, 1446 s, 1372 s,
1303 s, 1268 m, 1210 w, 1172 w, 919 w, 823 w, 718 w, 489 w, 365 w.
[Mg{N(SiMe3)(Dipp)}(hmpa)3][N(SiMe3)(Dipp)] (4). (2 mmol,
0.49 g, 0.55 mL) Yield: 1.37 g, 65%. Mp: 180-185 ꢀC. 1H
NMR (300 MHz, 25 ꢀC, C6D6): δH 0.19 (s, 9H, -SiCH3*),
0.51 (s, 9H, -SiCH3), 1.08 (d, 12H, -CH(CH3)2*), 1.39 (d, 12H,
-CH(CH3)2), 2.20 (d, 54H, HMPA), 3.99 (m, 2H, -CH(CH3)2),
4.7 (m, 2H, -CH(CH3)2*), 6.77 (m, 4H, p-CH*), 6.99 (m, 4H, p-
CH), 7.09 (m, 8H, m-CH*), 7.12 (m, 8H, m-CH). 13C NMR (300
MHz, 25 ꢀC, C6D6): δC 4.3 (-SiCH3), 4.85 (-SiCH3*), 36.71
(CH3 HMPA), 25.52 (p-CH(CH3)2*), 26.48 (p-CH(CH3)2*),
27.57 (p-CH(CH3)2), 29.46 (p-CH(CH3)2), 106.4 (p-CH*), 111.4
R ¼ SiMe3 ð1Þ; Ph ð2Þ; Mes ð3Þ
Compounds 1-3 have been isolated using the above reaction
route (eq 1). For the bulkier HN(SiMe3)(Dipp), ligand d, addi-
tion of HMPA yielded a rare separated amido species 4 (eq 2).
HMPA f
n-=sec-Bu2Mg þ 2HNðSiMe3ÞðDippÞ
hexane; 65 ꢀC
½MgfNðSiMe3ÞðDippÞgðhmpaÞ3ꢀ½NðSiMe3ÞðDippÞꢀ
4
þ 2BuH
ð2Þ
(35) Wannagat, V. U.; Autzen, H.; Kuckertz, H.; Wismar, H.-J. Z.
Anorg. Allg. Chem. 1972, 394, 254.
(36) Engelhardt, L. M.; Jolly, B. S.; Junk, P. C.; Raston, C. L.;
Skelton, B. W.; White, A. H. Aust. J. Chem. 1986, 39, 1337.
(37) Bartlett, R. A.; Olmstead, M. M.; Power, P. P. Inorg. Chem.
1994, 33, 4800.
(38) Ruhlandt-Senge, K. Inorg. Chem. 1995, 34, 3499.

Article
Organometallics, Vol. 30, No. 5, 2011 989
Table 2. Selected Bond Lengths and Secondary Interactions for Magnesium Amides
av Mg-N av Mg-D
av N-Mg-D av N-Mg-N av D-Mg-D
˚
(A)
˚
(A)
˚
ligand CN
av N-Si (A)
(deg)
(deg)
(deg)
ref
Mg[N(SiMe3)2]2(thf)2
a
a
b
c
c
d
d
4
4
4
4
4
2
4
2.021(5)
2.066(5)
2.049(2)
2.023(1)
2.085(3)
1.919(2)
2.006(3)
2.094(5)
1.994(4)
1.965(1)
2.055(1)
1.989(3)
1.706(5)
1.701(7)
1.708(3)
1.711(1)
108.55(5)
109.24(18)
107.68(7)
105.06(6)
108.33(13)
127.9(2)
120.07(16)
120.89(8)
134.35(6)
122.20(2)
180.00
89.8(2)
98.1(17)
103.89(7)
103.25(5)
98.93(2)
15
b
b
14
b
16
b
Mg[N(SiMe3)2]2(hmpa)2 (1)
Mg[N(SiMe3)(Ph)]2(hmpa)2 (2)
Mg[N(SiMe3)(Mes)]2(thf)2
Mg[N(SiMe3)(Mes)]2(hmpa)2 (3)
Mg[N(SiMe3)(Dipp)]2
1.714(4)
1.707(2)
[Mg{N(SiMe3)(Dipp)}(hmpa)3]
[N(SiMe3)(Dipp)] (4)
1.949(3)
1.708(3), 1.647(4)a
114.39(13)
104.10(12)
a Values for uncoordinated anion. b This work.
Combination of the strong Lewis base HMPA with the
bulkier HN(SiMe3)(Dipp) amine seems to provide the metal
center with enough coordinative saturation for this to occur.
Structural Aspects. Compounds 1-3 crystallize at room
temperature as discrete monomers with distorted-tetra-
hedral geometries around the metal center. Since they are
closely related to the previously synthesized THF-contain-
ing compounds, we chose these for comparison. Com-
pound 4 crystallizes as a rare example of a compound
with one ligand bound to the metal while the other one
remains unassociated. With only a very small number of
early-main-group separated amido species known,39-42
compound 4 represents the first example of such an alka-
line-earth-metal derivative. Pertinent bond lengths and
angles for both families of compounds are given in Table 2.
Throughout compounds 1-4, we consistently see the Mg-O
bonds for HMPA to be shorter than for those of the THF
adducts, attributed to the stronger basicity of HMPA and the
lower coordination number of the donor. Compounds 1-4
exhibit a formal metal coordination number of 4, which is quite
prevalent in magnesium chemistry.
Figure 2. Structure of Mg[N(SiMe3)(Ph)]2(hmpa)2 (2). Hydro-
gen atoms and methyl groups from HMPA are removed for
clarity.
The structural effect of ligand/coligand bulk, as well as
coligand basicity, is nicely demonstrated when comparing
the THF adduct Mg[N(SiMe3)(Mes)]2(thf)214 with the HMPA
analogue Mg[N(SiMe3)(Mes)]2(hmpa)2 (3) (Figure 3). Shorter
Mg[N(SiMe3)2]2(hmpa)2 (1) crystallizes in a distorted-
tetrahedral geometry, similar to the case for the THF adduct
M[N(SiMe3)2]2(thf)2 reported in 1990 by Bradley et al. via
redox transmetalation using a mercuric silylamide.15 As ex-
pected, the stronger Lewis base HMPA allows for a shorter
˚
Mg-D bonds are observed for 3 (1.989(3) A) than for the THF
˚
adduct (2.055(1) A), consistent with the trends discussed above.
The comparison of D-Mg-D angles in Mg[N(SiMe3)(Ph)]2-
(hmpa)2 (2) (103.89(7)ꢀ) and Mg[N(SiMe3)(Mes)]2(hmpa)2 (3)
(98.3(2)ꢀ) showcases the increased ligand bulk. This value
closely resembles the D-Mg-D angle for compound 1 (98.1-
(17)ꢀ). Similarly, the M-N bonds in Mg[N(SiMe3)(Mes)]2-
˚
M-donor bond with a difference of 0.09 A. The bulkier
nature of HMPA is expressed by a D-Mg-D angle of
98.1(17)ꢀ, wider than that for the THF adduct compound,
89.8(2)ꢀ. The wide angle and the shortening of the Mg-D
˚
bond cause slightly longer Mg-N bonds in 1, 2.066(5) A as
˚
˚
(hmpa)2 lengthen to 2.085(3) A, as compared to 2.049(2) A for
Mg[N(SiMe3)(Ph)]2(hmpa)2, again, an effect attributed to li-
gand bulk.
˚
compared to 2.021(5) A in the THF counterpart, and a
narrower N-Mg-N angle of 120.07(16)ꢀ, as compared to
127.9(2)ꢀ in the THF species.
Replacement of the methyl substituents by isopropyl groups
results in a further increase in ligand size, as also demonstrated
by the previous isolation of the two-coordinate Mg[N(SiMe3)-
(Dipp)]2.16 In the presence of 2.5 equiv of HMPA, a rare con-
tact/separated amido species was obtained, [Mg{N(SiMe3)-
(Dipp)}(hmpa)3][N(SiMe3)(Dipp)] (4) (Figure 4). Clearly, the
combination of [N(SiMe3)(Dipp)]- along with three HMPA
coligands provides favorable steric saturation to allow for the
formation of this rare structural motif. Examples of separated
amides in early-main-group chemistry are limited to alkali-metal
In Mg[N(SiMe3)(Ph)]2(hmpa)2 (2), one of the trimethyl-
silyl groups is replaced by a phenyl group (Figure 2).
Possibly due to the two-dimensional nature of the phenyl
substituents as compared to -SiMe3, and the resulting
decrease in steric bulk, the Mg-D bond for 2 is slightly
˚
˚
shorter (1.965(1) A) than for 1 (1.994(4) A); consequently
the D-Mg-D angle widens to 103.89(7)ꢀ. Despite the
smaller ligand size in 2 the N-M-N angle, 120.89(8)ꢀ, is
similar to that of 1, 120.07(16)ꢀ.
species, including [Li(12-crown-4)2][N(SiPh3)2] THF,39,40 [K(18-
3
crown-6)N(Ph)2],41 and {[K(18-crown-6)N(SiMePh2)2]}¥.42 No
examples of separated alkaline-earth-metal amido complexes have
been reported. Furthermore, separated ions involving alkaline-
earth metals are almost exclusively limited to the heavier metals,
where a lower metal charge density and thus weaker metal-ligand
bonding is observed.
(39) Bartlett, R. A.; Power, P. P. J. Am. Chem. Soc. 1987, 109 (12),
6509.
(40) Chen, H.; Barlett, R. A.; Dias, H. V. R.; Olmstead, M. M.;
Power, P. P. J. Am. Chem. Soc. 1989, 111, 4338.
(41) Hitchcock, P. B.; Khvostov, A. V.; Lappert, M. F.; Protchenko,
A. V. J. Organomet. Chem. 2002, 647, 198.
(42) Torvisco, A.; Decker, K.; Uhlig, F.; Ruhlandt-Senge, K. Inorg.
Chem. 2009, 48, 11459.

990 Organometallics, Vol. 30, No. 5, 2011
Torvisco and Ruhlandt-Senge
nolates, thiolates, and stannates.43-51 For magnesium, the
list of separated species is small, but similar strategies have
also proven successful for cyclopentadienides with the iso-
lation of [Mg(15-crown-5)(pyr)2][(C5H5)]2 and [Mg(15-
crown-5)(thf)2][(C5H5)]2, where the equatorial sphere of the
metal center is occupied by the crown, whereas monodentate
Lewis bases are located in the axial positions.17 In the
presence of the tridentate pentamethyldiethylenetriamine
(PMDTA), the contact/separated compound [Mg(C5H5)-
(pmdta)][C5H5] was obtained.17 A rare separated magnesium
thiolate was obtained using a combination of crown ether
and THF, affording [Mg(15-crown-5)(thf)2][SMes*]2.18
Ultimately, the rarity of separated amido complexes has to
be attributed to the ligand itself: namely, the strength and
nature of the metal-ligand bond and whether the deproto-
nated ligand can sustain the negative charge through deloca-
lization or resonance stabilization. Therefore, a closer inspec-
tion of the substitution patterns on the amide ligand and its
effect on the metal-ligand bond is warranted. HN(SiMe3)2 is
sterically demanding enough to allow for the isolation of an
array of compounds with low coordination numbers, but the
limited capacity for charge delocalization results in contact
molecules, a statement also supported by K(18-crown-6)
[N(SiMe3)2]52 and Li(12-crown-4)[N(SiMe3)2],53 which exhi-
bit M-N bonding despite the presence of crown ether. The
substitution of either one or both -SiMe3 groups by phenyl
groups increases the capacity for charge delocalization and, as
such, the likelihood for ion dissociation. Demonstrating
the added component of ligand bulk, only the most steri-
cally demanding ligand [N(SiMe3)(Dipp)]- (d) did initiate ion
separation.
Figure 3. Structure of Mg[N(SiMe3)(Mes)]2(hmpa)2 (3). Hy-
drogen atoms and methyl groups from HMPA are removed
for clarity.
Figure 4. Structure of [Mg{N(SiMe3)(Dipp)}(hmpa)3][N(SiM-
e3)(Dipp)] (4). Hydrogen atoms and methyl groups from HMPA
are removed for clarity.
Upon inspection of the ligands used for the isolation of
group I separated amido species, replacement of both
-SiMe3 with phenyl groups, as seen in [K(18-crown-6)N-
(Ph)2], results in the cleavage of the K-N bond, although not
separated ions.41 Rather, the equatorial positions of the
potassium cation are being filled by crown ether coordina-
tion, while the two axial positions are being occupied by
K-η6-Ph interactions. This coordinative preference may be
understood on the basis of not a single K-π interaction but
the sum of six, as made possible by the large size of the
potassium. In contrast, the smaller magnesium is unable to
engage in such a large number of π nteractions; as such, a
similar coordination mode is not favored. This argument is
further supported by the combination of the HN(Ph)2 ligand
Consistent with the free amido anions for group I metals,
the N-Si distance in 4 for the unassociated anion (1.647(4)
˚
˚
A) is shorter than for the metal-bound amido (1.708(3) A)
ligand. Further, widening of the N-Si-C angle for the
separated ligand (137.41(3)ꢀ) as compared to metal-bound
ligand (119.81ꢀ) also occurs, indicative of part of the negative
charge being distributed into the ligand system. These values
are summarized in Table 3.
Even though it is common for the heavier group II metals
in the presence of HMPA,43-45 a fully dissociated motif is
not observed for magnesium, likely a consequence of the
increased charge density for the lighter metal. This leads to
increased polarization and capacity for covalent interac-
tions. In contrast, an increase in metal size and reduction
of electronegativity for the heavier alkali and alkaline-earth
metals results in decreased covalency and thus facilitates ion
separation.
˚
with the smaller Li (CN=6, Li=0.76, K = 1.38 A), afford-
ing a contact molecule [Li(12-crown-4)(N(Ph)2] despite the
presence of crown ether.54 Furthermore, the increased
capacity for M-π interactions, as a result of larger radii,
provides an attractive avenue for achieving coordinative
saturation.
The combination of 18-crown-6 and HMPA has been
sufficient to create separated ion species for Ca-Ba for
a plethora of ligands, including methanides, silanides, sele-
(48) Ruhlandt-Senge, K.; Englich, U. Chem. Eur. J. 2000, 6, 4063.
(49) Englich, U.; Ruhlandt-Senge, K. Z. Anorg. Allg. Chem. 2001,
627, 851.
(50) Chadwick, S.; Englich, U.; Ruhlandt-Senge, K. Chem. Commun.
1998, 2149.
(51) Englich, U.; Ruhlandt-Senge, K.; Uhlig, F. J. Organomet. Chem.
2000, 613, 139.
(52) Evans, W. J.; Rego, D. B.; Ziller, J. W. Inorg. Chem. 2006, 45,
3437.
(53) Power, P. P.; Xiaojie, X. J. Chem. Soc., Chem. Commun. 1984,
358.
(43) Teng, W.; Englich, U.; Ruhlandt-Senge, K. Angew. Chem., Int.
Ed. 2003, 42, 3661.
(44) Verma, A.; Guino-o, M.; Gillett-Kunnath, M.; Teng, W.; Ruh-
landt-Senge, K. Z. Anorg. Allg. Chem. 2009, 635, 903.
(45) Teng, W.; Ruhlandt-Senge, K. Organometallics 2004, 23, 2694.
(46) Alexander, J. S.; Ruhlandt-Senge, K. Angew. Chem., Int. Ed.
2001, 40, 2658.
(47) Alexander, J. S.; Ruhlandt-Senge, K.; Hope, H. Organometallics
2003, 22, 4933.
(54) Bartlett, R. A.; Dias, H. V. R.; Hope, H.; Murray, B. D.;
Olmstead, M. M.; Power, P. P. J. Am. Chem. Soc. 1986, 108, 6921.

Article
Organometallics, Vol. 30, No. 5, 2011 991
Table 3. Selected Bond Lengths and Angles for Separated Amide Species
˚
˚
av N-Si (A)
av N-Ci (A)
av C-N-C (deg) av Si-N-Si (deg)
av Si-N-Ci (deg)
ref
[K(18-crown-6)N(Ph)2]
{[K(18-crown-6)N(SiMePh2)2]}¥ 1.649(2)
[Li(12-crown-4)2][N(SiPh3)2] THF 1.633(4)
[Mg{N(SiMe3)(Dipp)}(hmpa)3]
[N(SiMe3)(Dipp)] (4)
1.366(5)
120.3(3)
135.56(9)
154.9(3)
41
42
39, 40
b
3
1.708(3), 1.647(4)a 1.410(5), 1.354(5)a
119.81(2), 137.41(3)a
a Values for uncoordinated anion. b This work.
Along the same line, replacement of one or two methyl
groups in each of the -SiMe3 substituents in [N(SiMe3)2]-
demonstrates the effect of charge delocalization and steric
bulk on ion association. If only one of the CH3 groups is
replaced by phenyl, the HN(SiMe2Ph)2 amine ligand in
conjunction with 18-crown-6 affords the contact molecule
K(18-crown-6)[N(SiMe2Ph)2].52 Replacement of a second
-CH3 group by phenyl further increases the capacity for
charge delocalization, affording the unique {[K(18-crown-
6)N(SiMePh2)2]}¥.42 In this species no K-N bond is present,
but the trans positions are filled by K-η1-Ph interactions.
Interestingly, this strategy did not afford a separated ion for
the lithium analogue, again likely a consequence of not only
increased charge density but also a reduced capacity for
M-π interactions due to reduced ionic radius. Only the
replacement of all methyl groups in -SiMe3 by phenyl
afforded separated ions, as seen in [Li(12-crown-4)2][N-
Conclusion
Our results shed light on the delicate balance of metal-
ligand bond strength attributed to charge/size ratios, ligand
bulk, charge delocalization through the ligand system, and
availability of coordinative saturation through either Lewis
base or secondary interactions on ion association.
Acknowledgment. We gratefully acknowledge support
from the National Science Foundation (No. CHE 075-
3807). Purchase of the X-ray diffraction equipment was
made possible with grants from the National Science
Foundation (Nos. CHE-9527858 and CHE-0234912),
Syracuse University, and the W. M. Keck Foundation.
Supporting Information Available: CIF files giving crystal
data for compounds 2-4. This material is available free of
charge via the Internet at http://pubs.acs.org.
(SiPh3)2] THF.39,40
3