Addition of aluminum and gallium species to aromatic and alkyl-substituted 1,4-diaza-1,3-butadiene ligands

Article abstract of DOI:10.1021/ic202583w

In this report, we investigate the interactions of Me_xMCl _3-x (x = 0-3, M = Al, Ga) with various aromatic and alkyl-substituted 1,4-diaza-1,3-butadiene ^RDAB ligands (or α-diimine ligands) to give a variety of structures in solution and in the solid state. In combination with other previously reported structures, certain general trends of reactivity of these species can be deduced, although there are still some unexplained modes of reactivity. The methylated Al species react with aromatic-substituted ^RDAB ligands to provide final products that result from C=N insertion into the Al-CH₃ group followed by rearrangement reactions. The addition of methyl groups onto the backbone of the ^RDAB ligand is insufficient to stop the insertion and rearrangement processes from occurring. In the case of MeAlCl₂ with the bulky ^DiPPDAB ligand, the reaction could be followed spectroscopically from the monoadduct through the inserted/rearranged final product. Methylated Ga species, however, are much less predictable in their behavior with aromatic-substituted ^RDAB ligands. Depending on the exact species and ratios used, coordinated adducts can be formed and identified, or inserted/rearranged products similar to the aluminum reactions can be obtained. Quite interestingly, cation/anion pairs can also be formed in which GaCl₃ or MeGaCl₂ act as a chloride acceptors. This behavior was unique and substantially different from the analogous Al reactions which formed either a dicoordinated adduct or an inserted/rearranged complex. When the stronger-donating alkyl-substituted ^RDAB ligands were used with Me₂GaCl, only cation/anion pairs were obtained. Surprisingly, when the same reactions were performed using Me₂AlCl as a reagent, irreproducible results were obtained.

Full text of DOI:10.1021/ic202583w

Article
pubs.acs.org/IC
Addition of Aluminum and Gallium Species to Aromatic
and Alkyl-Substituted 1,4-Diaza-1,3-butadiene Ligands
Ana M. Felix,† Diane A. Dickie,† Ian S. Horne,† Giang Page,† and Richard A. Kemp*,†,§
†Department of Chemistry and Chemical Biology, University of New Mexico, Albuquerque, New Mexico 87131-2609, United States
§Advanced Materials Laboratory, Sandia National Laboratories, Albuquerque, New Mexico 87106, United States
S
* Supporting Information
ABSTRACT: In this report, we investigate the interactions of MexMCl3−x (x = 0−3, M = Al, Ga) with various aromatic and
alkyl-substituted 1,4-diaza-1,3-butadiene RDAB ligands (or α-diimine ligands) to give a variety of structures in solution and in the
solid state. In combination with other previously reported structures, certain general trends of reactivity of these species can be
deduced, although there are still some unexplained modes of reactivity. The methylated Al species react with aromatic-substituted
RDAB ligands to provide final products that result from CN insertion into the Al−CH3 group followed by rearrangement
R
reactions. The addition of methyl groups onto the backbone of the DAB ligand is insufficient to stop the insertion and
rearrangement processes from occurring. In the case of MeAlCl2 with the bulky DiPPDAB ligand, the reaction could be followed
spectroscopically from the monoadduct through the inserted/rearranged final product. Methylated Ga species, however, are
much less predictable in their behavior with aromatic-substituted RDAB ligands. Depending on the exact species and ratios used,
coordinated adducts can be formed and identified, or inserted/rearranged products similar to the aluminum reactions can be
obtained. Quite interestingly, cation/anion pairs can also be formed in which GaCl3 or MeGaCl2 act as a chloride acceptors. This
behavior was unique and substantially different from the analogous Al reactions which formed either a dicoordinated adduct or
R
an inserted/rearranged complex. When the stronger-donating alkyl-substituted DAB ligands were used with Me2GaCl, only
cation/anion pairs were obtained. Surprisingly, when the same reactions were performed using Me2AlCl as a reagent,
irreproducible results were obtained.
production of CO/olefin copolymers.1d−i,3 However, despite
INTRODUCTION
■
the high level of application within transition-metal chemistry,
Much of the preparative effort put forth by inorganic
R
the investigation of the coordination of DAB ligands toward
coordination chemists lies in the conception and syntheses of
main-group (s and p block) metals has significantly lagged that
new ligands designed to affect the metal center in a rational
of the d and f block elements.
manner. The chemist’s desire to specifically control targeted
properties of the metal of interest has led to an amazingly broad
Our interest in the RDAB family of ligands was initiated over
15 years ago by the desire to prepare olefin polymerization
array of ligand architectures, with significant structural
catalysts that contained main-group elements as the active
variations in electronic and steric features now commonplace.
components rather than transition metals. Albeit unsuccessful
One such family of donors that has seen widespread use is the
family of 1,4-diaza-1,3-butadiene or RDAB ligands (also
in achieving the goal of preparing high-activity polymerization
catalysts with main-group metals, that preliminary study was
commonly referred to as α-diimines). As one might expect,
the genesis of our more recent investigations of the interactions
the vast majority of structural and mechanistic studies using this
of various group 13 metal complexes with ligands containing
ligand family involve transition metals or lanthanide elements,
R
primarily for applications in catalysis or luminescence.1 Several
the DAB backbone. We recently reported a unique, multi-
R
chromic ligand/metal complex prepared from Me2GaCl and
DiPPDAB (DiPP = 2,6-diisopropylphenyl ligand (this compound
will be briefly mentioned later on in this manuscript as
compound 12).4 Quite unusually, this compound could provide
crystallographically identical materials that were either bright
yellow or green in color, or surprisingly, mixed yellow and
reasons for the significant interest in the DAB ligands have
been put forth in a review by van Koten et al.,2 and among the
features of these ligands that lend synthetic versatility are (a)
the inherent flexibility of the ligand backbone, (b) the ability of
the RDAB framework to act as either a σ-donor or a π-acceptor
ligand, and (c) the ability to alter the nature of the RDAB ligand
by varying the pendant R groups. As mentioned, RDAB ligands
have been more recently utilized in a variety of catalytic studies,
for example, in olefin polymerization catalysis as well as in the
Received: November 30, 2011
Published: March 28, 2012
© 2012 American Chemical Society
4650
dx.doi.org/10.1021/ic202583w | Inorg. Chem. 2012, 51, 4650−4662

Inorganic Chemistry
Article
R
Figure 1. Possible structures arising from interactions of ′DAB ligands with, e.g., group 13 R-MX2 species.
added to the deuterated solvents to provide additional drying. Solution
NMR spectral data were collected either on a Bruker AC 250 MHz
spectrometer with a Tecmag MacSpect upgrade or a Bruker Avance III
300 MHz spectrometer at the University of New Mexico NMR
Laboratory. Chemical shifts are reported in parts per million (ppm)
downfield from SiMe4 and were calibrated to the residual protic signal
of the deuterated solvent. Melting points were measured under argon
using a Uni-Melt capillary melting point apparatus and are reported
uncorrected. Elemental analyses were obtained from Columbia
Analytical Services located in Tucson, Arizona. We do note that many
of the isolated compounds reported here will darken and discolor upon
prolonged storage if they contain impurities.
green within a single crystal. The presence of a very small
amount of a highly colored blue byproduct in the yellow crystal
was identified as the likely source for the green color. However,
during our prior catalytic investigations, we became intrigued
not only by the unusual coloration of the crystals but also by
the unusual alkyl transfer and rearrangement reactions that
were observed upon the addition of various methylated group
R
13 species to DAB ligands. We note that the transfer and
rearrangement behavior was not unprecedented, as this
reactivity had first been described by van Koten et al. using
5
R
Me3Al and DAB ligands in 1979. However, there has been
only a limited number of follow-up reports describing this
unusual reactivity using main group metals, certainly when
compared to the considerable number of transition-metal-RDAB
studies that have been reported. These more recent main-group
studies will be noted and described in context when appropriate
within the Results and Discussion section (vide infra), but in
some cases these reports consist only of single crystal X-ray
structures.
Synthesis of (1). DiPPDAB (0.50 g, 1.32 mmol) was added slowly
to a colorless solution of MeAlCl2 (0.15 g, 1.32 mmol) in toluene
(5 mL). The solution turned bright red immediately. The reaction was
stirred overnight at rt, during which time a white precipitate formed.
The product was isolated by filtration, and the supernatant was cooled
to −25 °C to obtain additional product as X-ray-quality single crystals.
Yield: 52% (0.34 g), mp > 200 °C. 1H NMR (300 MHz, δ, in C6D6):
7.19−7.22 (m, 3H, (i-Pr)2C6H3), 6.97−7.08 (m, 3H, (i-Pr)2C6H3),
4.02 (br s, 2H, NC(Me)−CH2-N), 3.78 (sept, 2H, J = 6.9 Hz,
CHMe2), 3.07 (sept, 2H, J = 6.6 Hz, CHMe2), 1.44 (d, 6H, J = 6.6 Hz,
CHMe2), 1.35 (d, 6H, J = 6.6 Hz, CHMe2), 1.31 (d, 6H, J = 6.9 Hz,
CHMe2), 1.19 (s, 3H, NC(Me)−CHH−N), 0.86 (d, 6H, J = 6.9 Hz,
CHMe2). Anal. Calcd. for C27H39N2Cl2Al: C, 66.25; H, 8.03; N, 5.72.
Found: C, 66.85; H, 8.05; N, 5.62.
It was of interest to us to investigate more fully the range of
group 13 compounds that might undergo this transfer reaction,
as well as to expand the substituent types that can be used as
RDAB modifiers. We were also interested in observing whether
the unusual multicolored nature of the crystals described above
was unique and singular, or whether related members of the
RDAB-group 13 family of compounds might also exhibit this
unusual feature. This contribution reports the synthesis and
analytical characterization, including X-ray structural studies, for
the products of Al- and Ga-based complexes with both aryl- and
Synthesis of (2). DiPPDAB (1.00 g, 2.66 mmol) was added slowly
to a colorless solution of Me2AlCl (2.66 mL, 2.66 mmol in hexanes) in
toluene (5 mL); the solution turned dark orange. The reaction was
stirred for 14 h, during which time the reaction turned pale yellow.
Yellow crystals were isolated by slow evaporation. The compound
acquired a green tint of color over time in the drybox. The yield was
essentially quantitative (1.25 g): mp 210 °C decomposed, slight
R
alkyl-substituted DAB ligands. Reaction pathways resulting in
1
either methyl group transfer-rearrangement reactions or simple
Lewis acid−base adduct formation were both observed for
aromatic-substituted RDAB ligands, as well as unusual examples
of cation/anion complexes (Figure 1). However, with alkyl-
substituted RDAB ligands using Me2GaCl, we observed only the
formation of cation/anion pairs. While highly interesting main-
group complexes of cyclic ligands related to RDAB ligands such
as bis(imino)acenaphthenes (BIAN) have been recently
prepared and investigated,6 we have chosen to limit our work
orange color at 230 °C (orange liquid.) H NMR (250 MHz, δ, in
C6D6): 6.98−7.28 (m, 6H, (i-Pr)2C6H3), 4.35 (d, 1H, J = 23 Hz, N
C(Me)−CHH−N), 4.20 (sept, 1H, J = 7 Hz, CHMe2), 3.86 (d, 1H,
J = 23 Hz, NC(Me)−CHH−N), 3.53 (sept, 1H, J = 7 Hz, CHMe2)
3.16 (sept, 1H, J = 6 Hz, CHMe2), 3.00 (sept, 1H, J = 7 Hz, CHMe2),
1.53 (d, 3H, J = 7 Hz, CHMe2), 1.44 (d, 3H, J = 6 Hz, CHMe2), 1.41
(d, 3H, J = 7 Hz, CHMe2), 1.24 (s, 3H, NC(Me)−CHH−N), 1.28
(d, 3H, J = 6 Hz, CHMe2), 1.21 (d, 3H, J = 6 Hz, CHMe2), 1.17 (d,
3H, J = 7 Hz, CHMe2), 0.95 (d, 3H, J = 7 Hz, CHMe2), 0.86 (d, 3H,
J = 7 Hz, CHMe2), −0.33 (s, 3H, AlCH3). Anal. Calcd. for
C28H42N2ClAl: C, 71.69; H, 9.02; N, 5.97. Found: C, 71.42; H,
9.07; N, 5.96.
R
to these more simply substituted DAB ligands.
Synthesis of (3). MesDAB (0.500 g, 1.71 mmol) was added slowly
to a colorless solution of Me2AlCl (1.71 mL, 1.71 mmol) in toluene
(4 mL); the solution turned dark orange. The reaction was stirred for
14 h; during this time the reaction turned pale yellow. Colorless
crystals were isolated by slow evaporation of the solvent. Over time,
compound 3 acquired an orange tint on the crystals. Yield: 73.7%
(0.485 g), mp 183−185 °C. 1H NMR (250 MHz, δ, in CD2Cl2): 7.03
(s, 2H, (Me)3C6H2), 6.92 (s, 2H, (Me)3C6H2), 4.49 (d, 2H, J = 24 Hz,
NC(Me)−CHH−N), 4.26 (d, 2H, J = 24 Hz, NC(Me)−CHH−
N), 2.36 (s, 9H, (2,4,6-(CH3)3C6H2), 2.32 (s, 3H, (p-Me)3C6H2),
2.28 (s, 3H, (o-Me)3C6H2), 2.17 (s, 3H, (o-Me)3C6H2), 1.98 (s, 3H,
NC(Me)−CHH−N), −0.60 (s, 3H, AlCH3). Anal. Calcd. for
C22H30N2ClAl: C, 68.65; H, 7.86; N, 7.28. Found: C, 68.68; H, 8.00;
N, 7.38.
EXPERIMENTAL SECTION
■
General. Standard glovebox and vacuum line techniques were used,
as all reagents and products other than the starting diimines are highly
air- and/or moisture-sensitive.7 The protio-backbone versions of the
substituted RDAB ligands containing R = 2,4,6-trimethylphenyl
(
(
MesDAB),8 2,6-dimethylphenyl (2,6Me2PhDAB),9 2,6-di-i-propylphenyl
DiPPDAB),9 t-butyl (tBuDAB),10 or cyclohexyl (cHexDAB)10 were
prepared as reported previously in the literature, as was the DiPP-
containing dimethylated backbone ligand [DiPPDAB(Me2)].9 MeAlCl2,
MeGaCl2, and Me2GaCl were prepared according to published
procedures.11 Anhydrous solvents were purchased from Aldrich and
were used without further purification. All other reagents were
purchased from commercial sources including Aldrich, Strem, and
Gelest and used without further purification. Molecular sieves were
4651
dx.doi.org/10.1021/ic202583w | Inorg. Chem. 2012, 51, 4650−4662

Inorganic Chemistry
Article
Synthesis of (4). 2,6Me2PhDAB (0.500 g, 1.89 mmol) was added
slowly to a colorless solution of Me2AlCl (1.89 mL, 1.89 mmol) in
toluene (9 mL); the solution turned dark orange. The reaction was
immediately placed in the freezer. Crystals were isolated via slow
cooling at −20 °C. Yield: 59.3% (0.40 g). Crystals were clear with a
yellow tint and appeared to be thermally unstable long-term. 1H NMR
(300 MHz, δ, in C6D6): 6.80−7.06 (m, 6 H, (Me)2C6H3), 4.01 (d, 1H,
J = 25 Hz, NC(Me)−CHH−N), 3.67 (d, 1H, J = 25 Hz, N
C(Me)−CHH−N), 2.52 (s, 6H, (o-Me)2C6H3), 2.21 (s, 3H,
(o-Me)2C6H3), 1.86 (s, 3H, NC(Me)−CHH−N), 0.97 (s, 3H,
(o-Me)2C6H3), −0.29 (s, 3H, AlCH3). Anal. Calcd. for C20H26AlClN2:
C, 67.31; H, 7.34; N, 7.85. Found: C, 65.34; H, 7.76; N, 7.44.
Synthesis of (5). DiPPDAB (1.54 g, 4.09 mmol) was added slowly
to a colorless solution of Me3Al (0.296 g, 4.09 mmol) in toluene
(5 mL); the solution turned red. The reaction was placed in a freezer
at −20 °C. Yellow/orange crystals were isolated. Yield: 67.9% (1.25 g),
solution. Yield: 65.1% (0.23 g), mp −150 °C (dec). 1H NMR
(300 MHz, δ, in C6D6): 10.10 (s, 2H, NCH), 6.86−6.94 (m, 6H,
C6H3), 3.21 (sept, J = 7 Hz, 4H, CHMe2), 1.21 (d, J = 7 Hz, 24H,
CHMe), −0.14 (s, 6H, GaMe2). Anal. Calcd. for C28H42N2Cl4Ga2: C,
48.89; H, 6.15; N, 4.07. Found: C, 49.09; H, 6.23; N, 3.98.
Synthesis of (11). MesDAB (0.500 g, 1.71 mmol) was added slowly
to a colorless solution of Me2GaCl (0.231 g, 1.71 mmol) in toluene
(6 mL), and the solution turned dark orange. The reaction was
immediately placed in the freezer. Orange block crystals were isolated
at −20 °C. Yield: 69.8% (0.510 g), mp 87 °C (dec). 1H NMR
(250 MHz, δ, in CD2Cl2): 8.84 (d, 1H, J = 8.3 Hz, NC(Me)−
CHH−N), 7.61 (d, 1H, J = 8.3 Hz, NC(Me)−CHH−N), 7.09−6.82
(m, 4H, (Me)3C6H2), 2.38−1.97 (m, 18H, (Me)3C6H2), −0.19 (s, 6H,
Ga(CH3)2). Anal. Calcd. for C22H30N2ClGa: C, 61.79; H, 7.07; N,
6.55. Found: C, 61.75; H, 7.24; N, 6.64.
Synthesis of (12). DiPPDAB (1.00 g, 2.66 mmol) was added slowly
to a colorless solution of Me2GaCl (0.360 g, 2.66 mmol) in 10 mL of
toluene. The solution turned deep orange instantaneously. Over 12 h,
the color of the solution became dark green, and rod-shaped yellow
crystals of 1 began to form. The solution was cooled to −30 °C.
Colorless, green, yellow, and orange crystals were isolated and washed
with cold pentane after 12 h. The yield was essentially quantitative
(1.35 g). mp 170 °C with decomposition to an orange solid.
Continued heating to 210 °C leaves a red liquid which will solidify to a
green solid at rt. 1H NMR (250 MHz, δ, in C6D6): 6.98−7.27 (m, 6H,
(i-Pr)2C6H3), 4.57 (d, 1H, J = 23 Hz, NC(Me)−CHH−N), 4.32
(sept, 1H, J = 7 Hz, CHMe2), 3.88 (d, 1H, J = 23 Hz, NC(Me)−
CHH−N), 3.38 (sept, 1H, J = 6 Hz, CHMe2) 3.20 (sept, 1H, J = 6 Hz,
CHMe2), 2.90 (sept, 1H, J = 7 Hz, CHMe2), 1.52 (d, 3H, J = 6 Hz,
CHMe2), 1.48 (d, 3H, J = 6 Hz, CHMe2), 1.45 (d, 3H, J = 7 Hz,
CHMe2), 1.25 (s, 3H, NC(Me)−CHH−N), 1.25 (d, 3H, J = 7 Hz,
CHMe2), 1.24 (d, 3H, J = 6 Hz, CHMe2), 1.16 (d, 3H, J = 7 Hz,
CHMe2), 0.98 (d, 3H, J = 7 Hz, CHMe2), 0.86 (d, 3H, J = 6 Hz,
CHMe2), 0.01 (s, 3H, GaCH3). Anal. Calcd. for C28H42 N2ClGa: C,
65.71; H, 8.27; N, 5.47. Found: C, 65.99; H, 8.46; N, 5.30.
1
mp 180−182 °C. H NMR (250 MHz, δ, in C6D6): 7.10−7.40 (m,
6H, (i-Pr)2C6H3), 4.37 (s, 2H, NC(Me)−CHH−N), 3.63 (sept,
1H, J = 7 Hz, CHMe2), 3.02 (sept, 1H, J = 7 Hz, CHMe2), 1.33
(d, 12H, J = 7 Hz, CHMe2), 1.19 (d, 12H, J = 7 Hz, CHMe2), −0.86
(s, 3H, Al(CH3)2). Anal. Calcd. for C29H45N2Al: C, 77.63; H, 10.11;
N, 6.24. Found: C, 77.53; H, 10.13; N, 6.41.
Synthesis of (6). DiPPDAB(Me2) (1.33 g, 3.29 mmol) was
dissolved in toluene (5 mL). AlMe3 (0.237 g, 3.29 mmol) was
added slowly to give a yellow solution. The reaction was placed in the
freezer at −20 °C. Yellow crystals were isolated. Yield: 65.7% (1.03 g),
1
mp 180−182 °C. H NMR (250 MHz, δ, in CDCl3): 7.35−7.08 (m,
6H, (i-Pr)2C6H3), 3.74 (sept, 2H, J = 7 Hz, CHMe2), 3.03 (sept, 2H,
J = 7 Hz, CHMe2), 2.03 (s, 3H, NC(Me2)-C(Me)-N), 1.34 (s, 6H,
NC(Me2)-C(Me)-N), 1.30 (d, 6H, J = 7 Hz, CHMe2), 1.28 (d, 6H,
J = 7 Hz, CHMe2), 1.15 (d, 6H, J = 7 Hz, CHMe2), 1.08 (d, 6H, J =
7 Hz, CHMe2), −0.88 (s, 6H, Al(CH3)2). Anal. Calcd. for C31H49N2Al:
C, 78.10; H, 10.36; N, 5.88. Found: C, 78.41; H, 10.06; N, 6.01.
Synthesis of (7). GaCl3 (0.468 g, 2.66 mmol) was dissolved in
toluene (5 mL). DiPPDAB (1.00 g, 2.66 mmol) was slowly added. The
reaction mixture turned dark red immediately. The reaction was stirred
for 18 h. Dark orange crystals formed from the dark red solution via
Synthesis of (13). DiPPDAB (1.00 g, 2.66 mmol) was added slowly
to a colorless solution of Me3Ga (0.305 g, 2.66 mmol) in 6 mL of
toluene. The solution turned orange instantaneously. The reaction was
stirred for 18 h. Yellow crystals were isolated via slow evaporation in
toluene. Yield was essentially quantitative (1.33 g). mp 136−138 °C.
1H NMR (250 MHz, δ, in CD2Cl2): 6.95−7.40 (m, 6H, (i-Pr)2C6H3),
4.49 (s, H, NC(Me)−CHH−N), 3.62 (sept, 2H, J = 7 Hz, CHMe2),
2.98 (sept, 2H, J = 7 Hz, CHMe2), 1.93 (s, 3H, NC(Me)−CHH-N),
1.31 (d, 12H, J = 7 Hz, CHMe2), 1.18 (d, 12H, J = 7 Hz, CHMe2),
−0.43 (s, 3H, Ga(CH3)2). Anal. Calcd. for C29H45N2ClGa: C, 70.88;
H, 9.23; N, 5.70. Found: C, 70.93; H, 9.56; N, 5.75.
1
slow evaporation. Yield: 68.1% (1.00 g), mp 118−119 °C. H NMR
(300 MHz, δ, C6D6): 9.13 (d, J = 8.1 Hz, 1H, GaNCH), 7.58 (d, J =
8.1 Hz, 1H, NCH), 6.88−7.11 (overlapping m, C6H3), 3.08 (sept,
J = 6.6 Hz, 2H, CHMe2), 2.70 (sept, 6.9 Hz, 2H, CHMe2), 1.27 (d, J =
6.6 Hz, 6H, CHMe), 0.89 (d, J = 6.9 Hz, 12H, CHMe2), 0.82 (d, J =
6.9 Hz, 6H, CHMe2). Anal. Calcd. for C26H36N2Cl3Ga: C, 56.50; H,
6.57; N, 5.07. Found: C, 56.98; H, 6.62; N, 5.08.
Synthesis of (8). GaCl3 (0.23 g, 1.33 mmol) was dissolved in
toluene (4 mL). DiPPDAB (0.25 g, 0.67 mmol) was slowly added. The
reaction mixture turned dark red immediately, and a red precipitate
formed within minutes. The reaction was stirred overnight at rt, and
then the precipitate was isolated by filtration. Yield: 80.0% (0.39 g),
Synthesis of (14). DiPPDAB(Me2) (0.500 g, 1.24 mmol) was
dissolved in toluene (3 mL). GaMe3 (0.142 g, 1.24 mmol) was added
slowly. The reaction mixture turned yellow. The reaction was stirred
for 3 h. Yellow crystals were isolated by slow evaporation of the
toluene. Yield: 91.4% (0.587 g), mp 155−160 °C. 1H NMR (250 MHz,
δ, in C6D6): 7.30−7.00 (m, 6H, (i-Pr)2C6H3), 3.92 (sept, 2H,
J = 7 Hz, CHMe2), 3.04 (sept, 2H, J = 7 Hz, CHMe2), 1.40 (d, 6H,
J = 7 Hz, CHMe2), 1.37 (s, 3H, NC(Me2)−C(Me)−N), 1.28 (d, 6H,
J = 7 Hz, CHMe2), 1.25 (d, 6H, J = 7 Hz, CHMe2), 1.20 (s, 6H, N
C(Me2)−C(Me)−N), 0.97 (d, 6H, J = 7 Hz, CHMe2), −0.07 (s, 6H,
GaCH3). Anal. Calcd. for C31H49N2Ga: C, 71.68; H, 9.51; N, 5.39.
Found: C, 71.65; H, 9.64; N, 5.32.
1
mp 163−166 °C. H NMR (300 MHz, δ, C6D6): 9.86 (s, 2H, N
CH), 6.87−6.99 (overlapping m, C6H3), 2.75 (sept, J = 6.6 Hz, 4H,
CHMe2), 1.15 (d, J = 6.6 Hz, 12H, CHMe2), 1.10 (d, J = 6.6 Hz, 12H,
CHMe2). Anal. Calcd. for C26H36Cl6N2Ga2: C, 42.85; H, 4.98; N, 3.84.
Found: C, 42.36; H, 4.98; N, 4.04.
Synthesis of (9). DiPPDAB (0.50 g, 1.32 mmol) was added slowly
to a colorless solution of MeGaCl2 (0.21 g, 1.32 mmol) in 5 mL of
toluene. The solution turned dark red instantaneously. The reaction
was stirred for 18 h. Orange crystals were isolated via slow evaporation
1
of the toluene. Yield: 71.8% (0.51 g), mp 124−125 °C. H NMR
Synthesis of (15). tBuDAB (0.500 g, 2.97 mmol) was dissolved in
toluene (3 mL). Me2GaCl (0.402 g, 2.97 mmol) was slowly added.
The reaction was exothermic. White crystals formed immediately from
the dark orange solution. The white crystals redissolved into the
solvent. The reaction was allowed to stir overnight. Orange crystals fell
out of the solution as it became darker in color. The solution was
allowed to slowly evaporate. The crystals were washed with cold
pentane. Over time, the crystals obtained an orange tint. Yield: 48.4%
overall, 96.7% based on Ga (0.430 g). 1H NMR (250 MHz, δ, C6D6):
9.13 (s, 2H, NCH−CHN), 1.08 (s, 18H, C(CH3)3), 0.47 (s, 6H,
(CH3)2Ga), −0.20 (s, 6H, (CH3)2GaCl2). Anal. Calcd. for
(300 MHz, δ, C6D6): 9.32 (d, J = 8 Hz, 1H, GaNCH), 7.60 (d, J =
8 Hz, 1H, NCH), 6.87−7.09 (overlapping m, C6H3), 3.10 (sept,
J = 7 Hz, 2H, CHMe2), 2.72 (sept, J = 7 Hz, 2H, CHMe2), 1.18 (d, J =
7 Hz, 6H, CHMe2), 0.90 (d, J = 7 Hz, 12H, CHMe2), 0.84 (d, J = 7 Hz,
6H, CHMe2), −0.05 (s, 3H, GaMe). Anal. Calcd. for C27H39N2Cl2Ga:
C, 60.93; H, 7.39; N, 5.26. Found: C, 60.89; H, 7.38; N, 5.24.
Synthesis of (10). DiPPDAB (0.19 g, 0.50 mmol) was added slowly
to a solution of MeGaCl2 (0.16 g, 1.0 mmol) in toluene (2 mL). The
solution turned orange immediately, and a precipitate began to form
within a few minutes. After stirring at rt for 30 min, the precipitate was
isolated by filtration. X-ray-quality crystals were grown from the NMR
4652
dx.doi.org/10.1021/ic202583w | Inorg. Chem. 2012, 51, 4650−4662

Inorganic Chemistry
Article
C14H32N2Cl2Ga: C, 38.32; H, 7.35; N, 6.38. Found: C, 38.51; H, 7.50;
N, 6.37.
Table 1. Crystallographic Data and Parameters for Aromatic-
Substituted Al Complexes 1, 2, 3, 4, and 6
Synthesis of (16). cHexDAB (0.500 g, 2.27 mmol) was added
slowly to a colorless solution of Me2GaCl (0.622 g, 4.60 mmol) in
toluene (5 mL) to give a light green colored solution, which was
immediately placed in the freezer. White crystals were isolated at
1
2
3
empirical formula C27H39AlCl2N2
C28H42AlClN2
469.07
C22H30AlClN2
384.91
fw
489.48
172(2)
1
−20 °C. Yield: 79.6% (0.89 g). H NMR (300 MHz, δ, C6D6): 9.19
T, K
223(2)
223(2)
(s, 2H, NCH−CHN), 3.67 (s, 2H, ipso-C6H11), 1.84 (br, 4H,
C6H11), 1.43 (br, 4H, C6H11), 1.32 (br, 2H), 1.00−1.05 (overlapping
m, 10H, C6H11), 0.49 (s, 6H GaMe2), −0.17 (s, 6H, Cl2GaMe2).
Elemental analysis was not obtained due to thermal sensitivity.
Crystallographic Studies. The crystals were coated with oil
(Paratone-N) and were mounted on the nylon fiber of a CryoLoop
(obtained from Hampton Research, U.S.A.) that had been previously
attached to a metallic pin using epoxy. The data were collected at the
temperatures indicated in the tables on a Bruker X8 Apex II dif-
fractometer using monochromated Mo Kα radiation (λ = 0.71073 Å).
Cell parameters were initially retrieved using the APEX2 software12
and refined with the APEX2 software, and the SAINT software was
used for data reduction. The structures were solved using direct
methods and refined with the full-matrix least-squares method on F2
with SHELXTL.13 The non-hydrogen atoms were refined anisotropi-
cally. Hydrogen atoms were included at geometrically idealized
positions and were not refined. The isotropic thermal parameters of
the hydrogen atoms were fixed at 1.5U equiv. of parent atom for
terminal methyl group hydrogens and 1.2U for all others. Positional
disorders and/or symmetry-imposed partial occupancy was identified
and modeled in 1, 2, 10, and 13. Disordered solvent in 8 was
accounted for using SQUEEZE.14 All final crystallographic figures
shown in this document were generated by Diamond, v. 3.2 g.15
Crystallographic data for all compounds are listed in Tables 1, 2, and
3; however, in the interest of journal space, several of the thermal
ellipsoid plots for the compounds are shown only in the Supporting
Information. Structures have been deposited into the Cambridge
Crystallographic database under the following numbers: 1, 870996; 2,
661028; 3, 661029; 4, 661030; 5, 661033 (for information); 6,
661032; 7, 661024; 8, 870997; 9, 661023; 10, 870998; 11, 661025; 12,
661026 (for information); 13, 661027; 14, 661031; 15, 661021; and
16, 61020.
cryst size (mm)
0.31 × 0.24 ×
0.30 × 0.30 ×
0.30 × 0.10 ×
0.15
0.20
0.10
cryst syst
space group
a, Å
orthorhombic
Pnma
orthorhombic
Pnma
triclinic
P1
12.2900(3)
21.4033(6)
10.5018(3)
90.00
12.4112(4)
21.4045(7)
10.5492(4)
90.00
8.2488(3)
10.9973(4)
13.3696(4)
95.589(2)
101.687(2)
107.113(2)
1119.01(7)
2
b, Å
c, Å
α, deg
β, deg
γ, deg
volume, Å3
Z
90.00
90.00
90.00
90.00
2762.46(13)
4
2802.45(17)
4
calcd density,
g/cm3
1.177
1.112
1.142
μ (Mo Kα),
0.284
0.185
0.218
mm−1
a
R1 [I > 2σ(I)]
0.0416
0.0506
0.1243
0.0623
0.1698
6
b
wR2 [I > 2σ(I)] 0.1071
4
empirical formula
C20H26AlClN2
356.86
C31H49AlN2
fw
476.70
T, K
223(2)
223(2)
cryst size (mm)
cryst syst
space group
a, Å
0.50 × 0.20 × 0.20
triclinic
0.46 × 0.41 × 0.25
triclinic
P1
P1
̅
8.3271(5)
10.0610(6)
13.4920(8)
77.792(3)
78.412(4)
66.779(4)
1006.71(10)
2
8.4376(2)
9.7487(3)
19.3503(6)
82.012(2)
80.823(2)
69.652(2)
1467.24(7)
2
b, Å
c, Å
α, deg
β, deg
γ, deg
volume, Å3
RESULTS AND DISCUSSION
■
Aromatic RDAB LigandsAluminum Complexes. In
Z
calcd density, g/cm3
μ (Mo Kα), mm−1
1.177
1.079
1979, an initial report from van Koten et al. reported that the
R
interaction of Me3Al with various DAB ligands (R = aromatic
0.237
0.089
a
or alkyl groups) could produce a range of products that
depended on the nature of the R groups substituted on the
RDAB ligand.5 Briefly summarizing this seminal work, van
Koten et al. showed that (a) large R groups tended to produce
simple 1:1 Lewis acid−base complexes containing dative N→
AlMe3 bonds and (b) aromatic or smaller alkyl R groups tended
to produce inserted products in which a methyl group migrated
from AlMe3 onto the ligand backbone, resulting in formal
insertion of the CN bond of the ligand into an Al−C bond.
R1 [I > 2σ(I)]
0.0569
0.0433
b
wR2 [I > 2σ(I)]
0.1503
0.1223
a
b
2
2
R1 = ∑||Fo| − |Fc||/∑|Fo|. wR2 = {∑[ω(Fo − Fc )2]/
2
∑[w(Fo )2]}1/2
.
stoichiometry, Mair could observe both 1:1 and 2:1
AlCl3−DiPPDAB complexes, and these were structurally
characterized by X-ray crystallography. Quite interestingly,
Mair was unable to prepare either the 1:1 or 2:1 adduct in the
bulk as a pure substance, as mixtures of the two compounds
were produced in all cases. However, single crystals of each
adduct could be physically separated and isolated from the
complex mixture.
In this report, we have investigated the interactions of
aluminum and gallium species (MexMCl3−x, x = 0−3, M = Al,
Ga) with various aromatic and alkyl-substituted DAB ligands,
expanding the range of both main-group starting compounds as
well as the RDAB ligands. In this initial section, the interactions
of methylated aluminum chlorides with aromatic RDAB ligands
will be presented. Shown in Scheme 1 are the overall reactions
for the preparation of these methylated Al complexes, keeping
in mind that Mair had previously reported the interaction of
1
Products were identified and carefully characterized by H and
13C NMR spectroscopy. As well, the authors noted that a
unique rearrangement process was possible in the insertion
reactions in which a ligand backbone hydrogen atom migrates
to an adjacent carbon atom. It is of note that the R =
4-MeO(C6H4) and 4-MeC6H4 RDAB derivatives with AlMe3
were later characterized by this group using single crystal X-ray
diffraction and were definitively shown to have the inserted
structures suggested by the initial NMR studies.16
As further background, Mair and co-workers reported that
when AlCl3 was used in place of methylated Al species with the
DiPPDAB ligand that only Lewis acid−base adducts were formed
and no inserted products were seen.17 Depending on reaction
4653
dx.doi.org/10.1021/ic202583w | Inorg. Chem. 2012, 51, 4650−4662

Inorganic Chemistry
Article
Table 2. Crystallographic Data and Parameters for Aromatic-Substituted Ga Complexes 7, 8, 9, 10, 11, 13, and 14
7
8
9
10
11
13
empirical formula C26H36Cl3GaN2
C26H36Cl6Ga2N2 C27H39Cl2GaN2
c, Å
24.592(2)
90.00
12.9041(19)
103.143(6)
90.911(7)
97.951(6)
1109.1(3)
2
10.5557(3)
fw
552.64
173(2)
728.71
173(2)
532.22
223(2)
α, deg
β, deg
γ, deg
Volume, Å3
Z
90.00
90.00
90.00
2825.03(12)
4
T, K
90.00
cryst size (mm)
0.21 × 0.15 ×
0.14
0.35 × 0.17 ×
0.14
0.44 × 0.33 ×
0.13
90.00
6756.6(10)
8
cryst syst
space group
a, Å
monoclinic
P2(1)/c
25.344(7)
12.688(3)
19.173(15)
90.00
monoclinic
P2(1)/c
10.6473(4)
29.2220(11)
12.8369(5)
90.00
monoclinic
P2(1)/n
12.7846(5)
12.8871(6)
19.0497(8)
90.00
calcd density,
1.3931
1.281
1.155
g/cm3
b, Å
μ (Mo Kα), mm−1 2.009
1.368
0.991
a
c, Å
R1 [I > 2σ(I)]
0.0427
0.1047
0.0528
0.1580
0.0359
0.0854
b
α, deg
β, deg
γ, deg
volume, Å3
Z
wR2 [I > 2σ(I)]
109.00(3)
90.00
114.604(3)
90.00
108.701(2)
90.00
14
empirical formula
C31H49GaN2
519.44
5829(5)
8
3631.4(2)
4
2972.9(2)
4
fw
T, K
223(2)
calcd density,
g/cm3
1.259
1.333
1.189
cryst size (mm)
cryst syst
space group
a, Å
0.62 × 0.46 × 0.30
monoclinic
P2(1)/c
μ (Mo Kα),
1.234
1.942
1.121
mm−1
a
18.6849(12)
10.1700(7)
16.8763(11)
90.00
R1 [I > 2σ(I)]
0.0514
0.1064
10
0.0517
0.1319
11
0.0498
0.1239
13
b
b, Å
wR2 [I > 2σ(I)]
c, Å
α, deg
empirical formula C27H39Cl5Ga2N2
C22H30ClGaN2
427.65
C29H45GaN2
491.39
β, deg
111.818(3)
90.00
fw
708.29
173(2)
γ, deg
volume, Å3
T, K
223(2)
223(2)
2977.2(3)
4
cryst size (mm)
0.39 × 0.19 × 0.15 0.59 × 0.55 ×
0.20 × 0.09 ×
0.07
Z
0.51
calcd density, g/cm3
μ (Mo Kα), mm−1
1.159
cryst syst
space group
a, Å
orthorhombic
Pbca
triclinic
P1
orthorhombic
Pnma
0.944
̅
a
R1 [I > 2σ(I)]
0.0566
16.5170(14)
16.5170(14)
8.5502(12)
12.4887(3)
21.4298(5)
b
wR2 [I > 2σ(I)]
0.1606
b, Å
10.4355(13)
a
b
2
2
R1 = ∑||Fo| − |Fc||/∑|Fo|. wR2 = {∑[ω(Fo − Fc2)2]/∑[w(Fo )2]}1/2
.
Table 3. Crystallographic Data and Parameters for Aliphatic-
Substituted Ga Complexes 15 and 16
Scheme 1. Outline for Products of Methylated Aluminum
Species and DAB Ligands (R = aromatic)
R
15
16
empirical formula
C14H32Cl2Ga2N2
438.76
C25H44Cl2Ga2N2
582.96
fw
T, K
223(2)
223(2)
cryst size (mm)
cryst syst
space group
a, Å
0.40 × 0.31 × 0.17
triclinic
0.51 × 0.18 × 0.16
triclinic
P1
̅
P1
̅
9.8179(4)
10.6882(5)
12.0478(6)
65.440(3)
74.386(3)
69.424(2)
1065.21(8)
2
10.9899(5)
11.3709(6)
12.9317(6)
115.158(3)
92.861(3)
91.203(3)
1459.30(12)
2
b, Å
c, Å
α, deg
β, deg
γ, deg
volume, Å3
Z
calcd density, g/cm3
μ (Mo Kα), mm−1
1.368
1.327
2.774
2.053
R
a
compounds 1−5 utilize the all-protio DAB backbone, while
R1 [I > 2σ(I)]
0.0528
0.0253
R
b
the reaction to prepare 6 uses a dimethylated DAB backbone.
wR2 [I > 2σ(I)]
0.1526
0.0642
a
b
2
2
R1 = ∑||Fo| − |Fc||/∑|Fo|. wR2 = {∑[ω(Fo − Fc )2]/
Fortunately, the syntheses of the various starting ligands
required in our work have been published in the literature,
utilizing the condensation reactions of diketones with primary
amines to generate the substituted RDAB ligands.8−10 In a
2
∑[w(Fo )2]}1/2
.
AlCl3 with aromatic RDAB ligands to give mono- and
diadducts.17 Reactions to produce the inserted/rearranged
R
typical reaction to form a metal complex, the starting DAB
4654
dx.doi.org/10.1021/ic202583w | Inorg. Chem. 2012, 51, 4650−4662

Inorganic Chemistry
Article
Scheme 2. Formation of Final Product 1 Indicating Initial Adduct Formation, Then Insertion Intermediate Compound (Not
Observed), Followed by CH3/H Rearrangement (from van Koten et al.5)
ligand and the specific methylated Al compound were slowly
added together in a 1:1 molar ratio at room temperature under
argon using toluene as a solvent. Depending on the final
thermal sensitivity of each product, the reactions were stirred
either overnight at room temperature or stored at the slightly
lowered temperature of −20 °C. In no case was external
heating required, as good to essentially quantitative yields of the
products were obtained without additional heating. The
products 1−6 were isolated as X-ray quality crystalline solids.
When MeAlCl2 is treated with the bulky DiPPDAB ligand in
toluene solution at room temperature for less than 1 h followed
by solvent removal, an oil was obtained whose 1H NMR
Figure 2. Molecular structure of 1 (50% ellipsoids). For clarity,
spectrum was consistent with a simple adduct; however, upon
redissolution and storage in toluene at −20 °C, this oil
underwent further reaction to deposit analytically pure,
hydrogen atoms are not shown. Due to symmetry, the Al(1)−N(1)
bond is an average length of the expected Al−Ndative and Al−Ncovalent
bonds, and the N(1)−C(13) bond is an average of the expected C−N
colorless, X-ray quality crystals of 1. Investigation of these
and CN bonds. Selected bond lengths (Å) and angles (deg): Al(1)−
1
crystals by H and 13C NMR spectroscopy indicated that the
Cl(1) = 2.1253(10), Al(1)−Cl(2) = 2.1146(10), Al(1)−N(1) =
final geometry in solution was not an adduct but rather a
structure more consistent with an inserted and rearranged
product. In subsequent syntheses of 1, the immediately
formed red solution was allowed to stir overnight, and 1
precipitated out as a white powder during that time. The oil
was not isolated as only the final product was desired. The
reaction sequence given in Scheme 2 is instructive and is based
on the work of van Koten et al.5 This reaction indicates that the
directly inserted product expected from CN insertion into an
Al−C bond after initial adduct formation is not observed.
Rather, a subsequent rearrangement occurs where the H atom
attached to the methylated carbon migrates to the other
backbone carbon. This rearrangement reaction was briefly
described in the Introduction and presented in more detail
previously by van Koten et al. and as such does not require
extensive discussion.5 However, it is important to note that
the two possible inserted-product structures can be easily
1.8717(14), C(13)−N(1) = 1.368(2); Cl(1)−Al(1)−Cl(2) =
108.10(4), N(1)−Al(1)−N(1) = 87.27(9).
different C−N and the CN bonds that are expected. The
Al−N and C−N bond lengths that are observed and reported in
Figure 2 are averages of the two separate values.
Progressing along the MexAlCl3−x (x = 1−3) series we next
investigated the treatment of Me2AlCl with a variety of
substituted aromatic RDAB ligands. The RDAB ligand
substituents chosen for study were the sterically bulky DiPPDAB
ligand used above, the slightly less-bulky MesDAB group, and
the related 2,6Me2PhDAB group. The bulky DiPPDAB ligand
reacted with Me2AlCl in toluene at room temperature to give a
dark orange solution that turned light yellow upon standing for
14 h. Slow evaporation of the solvent afforded 2 as analytically
pure yellow crystals in essentially quantitative yield. Upon
standing at room temperature over time, 2 developed a green
tint and lost crystallinity; however, a satisfactory elemental
analysis for 2 could be obtained. Storage at −20 °C suppressed
formation of the green color. The 1H NMR data for 2
suggested that the structure in solution was similar to its Ga
analogue that we have previously reported (this mixed green/
yellow crystalline compound is also briefly mentioned below as
compound 12)4 and to inserted/rearranged complexes
prepared by van Koten et al.5
In order to confirm the identity of 2, we performed a single
crystal X-ray analysis on the isolated crystals, and the structure
of 2 is shown in Figure 3. Compound 2 is structurally similar to
our previously reported Ga analogue, and the thermal ellipsoid
plot shows that the addition of DiPPDAB to Me2AlCl results in
the rearranged product shown in Scheme 2. While most of the
metrical parameters for 2 are within the values expected and do
not merit extensive discussion, we note that, like 1, compound
2 crystallizes in space group Pnma, and as such there is a crys-
tallographic mirror plane through the molecule. The migrated
1
distinguished in solution by H NMR spectroscopy, as the
directly inserted complex contains a vinyl H proton and an allylic
H proton that couple to each other, while the rearranged product
1
contains two allylic protons and no vinylic protons. The H
NMR spectrum of 1 showed the presence of only allylic protons
on the carbon backbone, characteristic of the rearranged product
seen earlier by both van Koten et al. and us.4,5 However, in order
to confirm the geometry of 1, we performed a single-crystal
X-ray analysis. 1 crystallized in the centrosymmetric space group
Pnma, and the structure is shown in Figure 2. There is a
crystallographic mirror plane through the Al(1) atom that bisects
the five-membered ring. Therefore, the methyl group [C(14)]
that has migrated can occupy either side of the diimine backbone
and crystallographically shows half occupancy on either of the
C(13) carbon atoms on the backbone. In Figure 2, only one
of the possible structures is shown. Due to this crystallographic
symmetry, it is impossible to distinguish between the two
separate Al−N bonds (dative and covalent), as well as the
4655
dx.doi.org/10.1021/ic202583w | Inorg. Chem. 2012, 51, 4650−4662

Inorganic Chemistry
Article
differences in the Al(1)−N(1) [1.817(3) Å] and Al(1)−N(2)
[1.970(3) Å] bond lengths inside the ring clearly show the
distinctive changes between a typical N−Al single bond and an
N→Al dative bond. As well, the two C−N bonds in the ring
also demonstrate substantial differences in lengththe N(2)−
C(20) bond length of 1.286(4) Å is significantly shorter than
the N(1)−C(19) bond length of 1.446(4) Å, thus indicating
the double bond nature of N(2)−C(20) versus the single bond
character of N(1)−C(19). The bond lengths in the ring are
consistent with the rearranged structure shown in Scheme 2.
Multiple reactions of the 2,6Me2PhDAB ligand with Me2AlCl
showed that the product 4 produced in a room temperature
synthesis began to decompose upon storage at 25 °C. The
single crystal X-ray structure of 4 indicated that the structure is
very similar to 3, space group P1, with distinctive bond lengths
within the five-membered ring indicating the presence of
discrete C−N, CN, N−Al, and N→Al bonds within the ring,
consistent with a rearranged product. There are two crystallo-
graphically distinct (but very similar) molecules in the unit cell
of 4. No other bond lengths or angles deserve any extended
discussion. Due to the similarity of the structures of 3 and 4,
the thermal ellipsoid plot of 4 is not shown here. However, a
structural diagram is available in the Supporting Information,
and the structural details are given in Table 1.
Figure 3. Molecular structure of 2 (50% ellipsoids). For clarity,
hydrogen atoms are not shown. Due to symmetry, the Al(1)−N(1)
bond is an average length of the expected Al−Ndative and Al−Ncovalent
bonds, and the N(1)−C(13) bond is an average of the expected C−N
and CN bonds. Selected bond lengths (Å) and angles (deg):
Al(1)−Cl(1) = 2.097(4), Al(1)−C(15) = 2.01(2), Al(1)−N(1) =
1.8890(15); Cl(1)−Al(1)−C(15) = 112.7(7), Al(1)−N(1)−C(1) =
129.10(11), Al(1)−N(1)−C(13) = 112.78(12).
methyl group [C(14)] has 50% occupancy on each of the
C(13) carbon atoms on the backbone, and Cl(1) and C(15)
are disordered in a 65:35 ratio. The distinct differences in the
amido/imido N−Al bonds could not be seen, nor could the
differences in the C−N versus CN bonds. In Figure 3, again
only one of the two possible isomers is shown.
As mentioned in the Introduction, most of the previous
literature in this area had utilized Me3Al as the main-group
metal. RDAB ligands used prior to our work included variations
in which R = 4-MeC6H4 or 4-(MeO)C6H4, both of which had
been structurally characterized by van Koten et al. and shown
to have the “rearranged product” structure.16 To complete the
MexAlCl3−x (x = 1−3) series using the DiPPDAB ligand, we also
examined the interaction of Me3Al with this ligand. The
DiPPDAB ligand and Me3Al were reacted at room temperature in
toluene to form a red solution. Cooling to −20 °C afforded
analytically pure, X-ray quality yellow-orange crystals of 5 in
67.9% isolated yield, and no thermal instability for 5 was
The reactions of Me2AlCl with the slightly less-bulky MesDAB
and 2,6Me2PhDAB ligands proceeded similarly to the reaction of
DiPPDAB with Me2AlCl. A room temperature reaction of
MesDAB and Me2AlCl in toluene led to the eventual formation
of compound 3. Isolation via the gradual removal of solvent
afforded analytically pure, X-ray quality crystals of 3 in 73.7%
1
yield. H NMR analysis of the colorless crystals of 3 in C6D6
indicated a solution geometry consistent with the rearranged
structure seen also in 2. Single crystal X-ray diffraction analysis
confirmed the structure of 3 to be analogous to 2, and the
solid-state structure for 3 is shown in Figure 4. Unlike 2, 3
1
observed. Analysis of the H NMR data indicated that the
solution structure was again the rearranged, five-membered ring
product, consistent with the previous aryl-substituted Al
complexes. We note that we had previously prepared this
compound, characterized it by X-ray crystallography, and
deposited the structure into the Cambridge Structural Database
as a private communication (CCDC 661033). After our
deposition, the structure of 5 was reported as part of a study
by the Gibson group.18 While the exact preparation method
differed as Gibson et al. used refluxing toluene for 12 h, the
overall yields were similar and the structures virtually identical.
As in 2 above, 5 crystallizes in the centrosymmetric space group
Pnma, and as such the migrated methyl group has half-
occupancy on each of the backbone carbon atoms. This plane
of symmetry also enforces that the different Al−N and C−N
bonds cannot be distinguished. As the structure of 5 was
described previously by Gibson et al., it is not necessary to
discuss it more fully. However, for completeness, a thermal
ellipsoid plot of 5 is provided in the Supporting Information.
Figure 4. Molecular structure of 3 (50% ellipsoids). For clarity,
hydrogen atoms are not shown. Selected bond lengths (Å) and angles
(deg): Al(1)−N(1) = 1.817(3), Al(1)−N(2) = 1.970(3), Al(1)−
C(22) = 2.004(3), Al(1)−Cl(1) = 2.1321(17), N(2)−C(20) =
1.286(4), N(1)−C(19) = 1.446(4); Cl(1)−Al(1)−C(22) =
108.45(10), N(1)−Al(1)−N(2) = 85.65(12), Al(1)−N(1)−C(19) =
114.9(2), Al(1)−N(2)−C(20) = 112.7(2).
R
Last, for the aromatic DAB ligand−aluminum complexes,
we were interested in whether we could sterically block the
crystallized in space group P1 and as such did not possess a
crystallographic plane of symmetry bisecting the five-membered
ring. Therefore, individual bond lengths within the ring could
be determined, and the unsymmetrical nature of the
heterocyclic ring can thus be easily observed. There are two
crystallographically distinct (but very similar) molecules in the
unit cell of 3, and as such we will only discuss one of them. The
migration of the methyl group from the Al atom onto the C−C
R
backbone of the DAB ligand. It was clear based on the work
described above, as well as prior van Koten and Vrieze studies,
that in cases of aryl-substituted DAB ligands the migration of a
methyl group to the backbone always occurs, regardless of the
size of the aromatic substituent, while the AlCl3 experiments
4656
dx.doi.org/10.1021/ic202583w | Inorg. Chem. 2012, 51, 4650−4662

Inorganic Chemistry
Article
from the Mair group indicate that chlorides do not migrate. We
attached the sterically largest aromatic groupDiPPonto a
modified DAB ligand that contained a dimethylated rather than
a diprotio backbone (Scheme 1). DiPPDAB(Me2) and Me3Al
were reacted slowly at room temperature in toluene, and the
product was allowed to crystallize at −20 °C. Yellow,
analytically pure crystals were produced in 65.7% yield.
1H NMR analysis indicated the presence of an −AlMe2 moiety,
as well as a singlet integrating on two Me groups on the
diimine backbone and a singlet integrating to a single Me group
on the backbone. Thus, the NMR integration data were
strongly suggestive that product 6 was not the Lewis acid−base
adduct as would be expected if no methyl migration had taken
place. For confirmation, we performed a single crystal X-ray
structural analysis on 6. Compound 6 crystallizes in space
the deep red solution, the monoadduct 7 was isolated in 68.1%
yield as dark orange crystals. The H NMR data indicated that
1
the molecule had an unsymmetrical structure in solution, thus
indicating a monoadduct, but the solution also showed the
presence of free DiPPDAB ligand. The presence of the ligand in
solution was confirmed by an NMR tube experiment in which a
50% excess of ligand was added to the reaction mixture, and the
resonances due to the free ligand increased by 50%. The
observed unsymmetrical structure by NMR was somewhat
unexpected by us based on prior literature. A symmetric
structure would be consistent with either a 1:1 complex of
GaCl3/DiPPDAB ligand in which DiPPDAB is chelated symmetri-
cally to GaCl3 or a situation in which the GaCl3 is rapidly
migrating between the two N atoms, thus equivalencing the
two halves of the molecule on the NMR time scale. van Koten
R
group P1, and the structure is shown in Figure 5. Again, the
et al. had noted that, in the case of AlMe3 and DAB, rapid
̅
exchange occurred even at −90 °C in toluene,5 and so we were
expecting a symmetric structure in the NMR spectrum. As well,
Mair et al. had seen in the AlCl3 case with DiPPDAB the
presence of both mono- and diadducts of AlCl3 with the ligand
in solution.17 We did not observe any peaks in the NMR
spectrum that could be assigned to a symmetrical complex.
Compound 7 crystallized in the space group P2(1)/c, and
the molecular geometry is shown in Figure 6. The solid-state
structure of 7 is a monoadduct in which only the E/E isomer is
found, in contrast to Mair et al. who observed formation of
both the E/E and E/Z isomers for the AlCl3/DiPPDAB
monoadduct.17 The structure resembles that found for the
free ligand, with the obvious addition of the GaCl3 fragment
complexed to one N atom. There are two slightly different
molecules in the unit cell, and only one of them is shown in
Figure 6. There are negligible differences in bond lengths
between these two independent molecules, and only slight
differences are found in the bond angles. If one looks at the
asymmetric unit, one Ga is bound to the “left” nitrogen atom,
and the other is bound to the “right” nitrogen atom. In a
packing diagram, this has the effect of causing each layer of
molecules to have the GaCl3 unit pointing in the opposite
directions. As was seen previously in the AlCl3 case,17 the
conjugation of the diimine backbone is largely unaffected by the
presence of the GaCl3, as the CN bond lengths are identical
in length.
Figure 5. Molecular structure of 6 (50% ellipsoids). For clarity,
hydrogen atoms are not shown. Selected bond lengths (Å) and angles
(deg): Al(1)−N(1) = 1.8471(9), Al(1)−N(2) = 1.9875(10), Al(1)−
C(30) = 1.9680(15), Al(1)−C(31) = 1.9648(15), N(2)−C(25) =
1.2865(14), N(1)−C(27) = 1.4682(14); C(30)−Al(1)−C(31) =
108.48(7), N(1)−Al(1)−N(2) = 84.40(4).
N(1)−Al(1) amido single bond is significantly shorter than the
N(2)→Al(1) dative bond, and the C(25)−N(2) double bond is
much shorter and distinct from the C(27)−N(1) single bond. It is
apparent that the presence of methyl groups on the backbone
prior to reaction is not sufficient to keep the Al−Me groups from
migrating onto the diimine backbone, as the migrated methyl
group can be clearly seen as one of the two methyl groups on
C(27) versus the single methyl group attached to C(25).19
The preparation of the 2:1 GaCl3/DiPPDAB complex 8
followed a similar route to the preparation of 7 except with the
change in molar ratio. The solution again turned dark red
immediately, and a red precipitate was formed within minutes.
After stirring overnight at room temperature, product 8 was
isolated in 80% yield by filtration as an analytically pure
powder. Cooling of the supernatant afforded X-ray quality
R
Aromatic DAB LigandsGallium Complexes. As one
might suspect, analogous reactions utilizing methylated gallium
R
compounds with aromatic DAB ligands are significantly less
studied than their aluminum counterparts and appear to be
limited to our previous report dealing with the multichromic
crystals produced form the interaction of Me2GaCl with the
DiPPDAB ligand.4 We do note, however, that related compounds
of lower-valent GaI−diimine adducts have been well-
documented in particular by the group of Jones,20 but they
do not fall into the scope of our higher-valent GaIII interests.
Again utilizing primarily the DiPPDAB ligand, we investigated
the interactions of aromatic-substituted RDAB ligands with
GaCl3, MeGaCl2, Me2GaCl, and Me3Ga. Shown in Scheme 3
are the reactions that will be discussed in this section, leading to
either adduct products, inserted and rearranged products, or
surprisingly a cation/anion pair.
1
crystals of 8. The H NMR spectral data for the deposited
powder and the isolated crystals were identical and indicated
the presence of a vinyl proton at 9.86 ppm. As we had also seen
with 7, free DiPPDAB ligand was also present in the spectra. This
indicated to us that we had an entirely different solution
geometry with 8 than with the other complexes, and so an
X-ray diffraction analysis was performed on 8.
The structure shown in Figure 7 is quite different from the
previous structures seen by either van Koten et al., Mair et al.,
or us. 8 crystallizes in space group P2(1)/c and consists of a
discrete cation/anion pair. The bond lengths and angles found
in 8 are similar to those found in the cation/anion pair reaction
product of an alkyl-substituted diimine (tBuDAB) with 2 mol of
GaCl3 that was previously reported by Cowley et al., as well as a
Treatment of the DiPPDAB ligand in toluene at room
temperature for 18 h with a 1:1 ratio of GaCl3 afforded the
monoadduct 7 in which GaCl3 has coordinated to one of the
backbone nitrogen atoms. After removal of the toluene from
4657
dx.doi.org/10.1021/ic202583w | Inorg. Chem. 2012, 51, 4650−4662

Inorganic Chemistry
Scheme 3. Outline for Products of Gallium Species and DAB Ligands (R = Aromatic)
Article
R
isolated in 71.8% yield upon the slow removal of solvent. A 1H
NMR spectrum taken of the orange crystals indicated that a
coordination adduct had formed rather than the ultimate
inserted/rearranged product found in the Al analogue 1;
however, recall that an adduct was initially observed during the
1
preparation of 1 as well as an unstable intermediate. The H
NMR of 9 showed a pattern typical of unsymmetrical binding
to the DiPPDAB ligand in which only one N atom is
coordinated. As with 7 and 8, free DiPPDAB ligand was also
1
observed in the H NMR spectrum. The thermal ellipsoid plot
of 9 is shown in the Supporting Information, as the solid-state
structure of 9 is similar to that of the adduct 7 using GaCl3 and
the E/E isomer of the AlCl3 adduct reported by Mair et al.
N(1) has the expected trigonal geometry (∑bondangles N(1) =
359.96°), and the N(1)−C(26) bond length (1.281(4) Å) is
very slightly elongated relative to the N(2)−C(25) bond length
(1.256(4) Å).
Figure 6. Molecular structure of 7 (50% ellipsoids). For clarity,
hydrogen atoms are not shown. Selected bond lengths (Å) and angles
(deg): Ga(1)−N(1) = 2.021(3), N(1)−C(13) = 1.278(4), C(13)−
C(14) = 1.472(5), C(14)−N(2) = 1.268(5), N(2)−C(15) = 1.432(5),
N(1)−C(1) = 1.470(5); C(1)−N(1)−Ga(1) = 118.7(2), C(1)−
N(1)−C(13) = 115.3(3), Ga(1)−N(1)−C(13) = 126.0(3).
As free DiPPDAB ligand was observed in the reaction solution
during preparation of 9, it was of interest to add more MeGaCl2
to attempt to push the reaction toward completion. We next
added a 2:1 ratio of MeGaCl2 relative to DiPPDAB ligand under
similar reaction conditions. The solution turned orange
immediately upon addition, and an orange precipitate began
forming within minutes. After 30 min, the precipitate was
collected by filtration in 65.1% yield (after identification). The
1H NMR spectrum of 10 indicated the presence of vinyl
protons on the DAB backbone, and no allylic proton
resonances, indicating a symmetric structure in solution. As
well, only a single upfield resonance was present in the GaCH3
region that integrated for one GaCH3 per backbone vinyl
BIAN-complex of [GaCl2]+ from the Clyburne group.21 In both
cases, there is found an acute N−Ga−N bite angle of ∼83−84°,
and the bond lengths present in both products are consistent
with those shown in Scheme 4. The structure of 8 is also
consistent with an initial reaction of the DiPPDAB ligand to
coordinate (or possibly chelate) to a molecule of GaCl3,
followed by a reaction with a second molecule of GaCl3 that acts
as a halide abstractor to form the [GaCl4]− anion (Scheme 4).
We have no evidence by either NMR or X-ray data that a
dicoordinated adduct, as reported by Mair et al. for the 2:1 Al
analogue,17 is ever formed. This result may show the subtle
differences in reactivity patterns exhibited by Al versus Ga.
Interaction of the DiPPDAB ligand in a 1:1 ratio with
MeGaCl2 in toluene at room temperature for 18 h gave a deep
red solution from which dark orange crystals of 9 could be
1
proton. On the basis of the H NMR spectrum, one would
likely expect the symmetric structure of 10 to arise from a
dicoordinated adduct of MeGaCl2 with the DiPPDAB ligand.
However, as we saw above in the 2:1 GaCl3 case to produce 8, a
4658
dx.doi.org/10.1021/ic202583w | Inorg. Chem. 2012, 51, 4650−4662

Inorganic Chemistry
Article
Figure 7. Molecular structure of 8 (50% ellipsoids). For clarity, hydrogen atoms are not shown. Selected bond lengths (Å) and angles (deg): Ga(1)−
N(1) = 1.983(3), Ga(1)−N(2) = 2.005(3), N(1)−C(1) = 1.277(5), N(2)−C(2) = 1.283(5), C(1)−C(2) = 1.478(5); N(1)−Ga(1)−N(2) =
83.56(13), Ga(1)−N(1)−C(1) = 111.2(3), Ga(1)−N(2)−C(2) = 109.6(3).
Scheme 4. Reaction of DiPPDAB Ligand and GaCl3 to Yield 8
dicoordinated adduct was not seen, but rather another cation/
anion pair was formed. In the case of 10, however, the expected
cation/anion pair would be [(DiPPDAB)GaMeCl]+ [MeG-
1
aCl3]−, which could give a symmetrical H NMR pattern for
Figure 8. Molecular structure of 10 (50% ellipsoids). For clarity,
hydrogen atoms are not shown. Selected bond lengths (Å) and angles
(deg): Ga(1)−C(28) = 1.931(13), Ga(1)−Cl(5) = 2.023(3), Ga(1)−
N(1) = 2.043(3), Ga(1)−N(2) = 2.037(3); Cl(5)−Ga(1)−C(28) =
116.9(4), N(1)−Ga(1)−N(2) = 80.13(12).
the cation, but not a single −GaCH3 resonance. In order to
confirm the structure of 10, we grew X-ray quality crystals from
the NMR tube. The structure obtained (Figure 8) was a cation/
anion pair. The most reasonable solution to the crystal
structure was obtained with a [GaCl4]− anion and a
[(DiPPDAB)GaMeCl]+ cation in which the Cl and Me were
each disordered over two positions in a roughly 55:45 ratio. We
note that this does not correspond to the reaction
stoichiometry nor the NMR data. The cation/anion pair that
would best fit the NMR spectrum is [(DiPPDAB)-
GaMe2]+[GaCl4]−, which would result from disproportionation
of the MeGaCl2 starting material and is consistent with the
elemental analysis. Also possible would be [(DiPPDAB)-
GaMeCl]+[MeGaCl3]− resulting from simple chloride abstrac-
tion as seen in 8. Attempts to model each of these, however,
were unsatisfactory. Problems with partial occupancy of Me and
Cl in cation/anion pairs were also encountered in the reaction
of Me2AlCl with tBuDAB (vide infra).
temperature the reaction mixture was cooled to −20 °C and
stored. Dark orange, analytically pure crystals of 11 were
isolated in 69.8% yield. 1H NMR data were consistent with the
solution structure of 11 as an adduct, and the X-ray structure of
11 confirmed the NMR assignment (shown in the Supporting
Information and crystallographic data in Table 2). The solid-
state structure is very similar to the other monoadducts
discussed earlier and so no further discussion of the structure is
needed. However, we do note that the only isomer present in
the structure was the E/E isomer with no trace of the E/Z
isomer found.
Interaction of Me2GaCl with DiPPDAB to give 12 has been
reported previously by us,4 although the preparation and
analytical characterization are included here for completeness.
The solid-state structure is the inserted/rearranged product
(vide supra). Quite interestingly, identical crystal structures
were obtained for 12 for yellow, green, or mixed yellow/green
single crystals. The variation in color seen for crystals of 12 was
attributed to the decomposition of a small amount of 12 to
form trace levels of a highly colored blue compound that was
present in the yellow crystals of 12. While intensely colored
solutions and products were prepared in this current study of
methylated Al and Ga species with aromatic DAB ligands, it is
noteworthy that none of the crystalline solids prepared
demonstrated the multichromic behavior shown by 12.
Detailed characterization data for 12 have been published
elsewhere;4 however, for completeness, the thermal ellipsoid
plot of 12 is also shown in the Supporting Information. We
note that we have also previously reported that if 2 equiv of
As MeGaCl2 yields an adduct 9 and the cation/anion pair 10
rather than inserted/rearranged products, these results indicate
that MeGaCl2 is more similar in behavior to AlCl3 and GaCl3
than to the dimethylated species Me2MCl (M = Al, Ga) when
treated with the bulky DiPPDAB ligand. The energetic drive to
insert the CN bond of the aromatic-substituted diimine into
the Ga−CH3 bond does not appear to be sufficient if only one
methyl group is present on the central Ga atom, despite the fact
that the Al analogue 1 does eventually produce an inserted/
rearranged product (Figure 2).
Changing the aromatic RDAB ligand from DiPPDAB to a
smaller MesDAB ligand and changing the gallium species to
Me2GaCl surprisingly did not afford an inserted/rearranged
product with Me2GaCl, but rather again an adduct. Reaction of
MesDAB with Me2GaCl in toluene immediately gave a vivid dark
orange color in solution, and after stirring briefly at room
4659
dx.doi.org/10.1021/ic202583w | Inorg. Chem. 2012, 51, 4650−4662

Inorganic Chemistry
Article
Scheme 5. Outline for Products of Me2GaCl Reactions with AlkylDAB Ligands
Me2GaCl are treated with DiPPDAB at low temperatures (−30 °C),
a diadduct is formed in low yields that was characterized by X-ray
crystallography.4 This product is consistent with the Mair et al.
results17 but different from our results with 8 and 10. This
diadduct converts to 12 upon warming to room temperature in
solution, again inconsistent with the behaviors of 8 and 10, which
are both stable in solution.
a limited number of structural motifs were formed using
aromatic RDAB ligands. As the work described above had in
some cases involved Me2GaCl and had given inserted/
rearranged products instead of simple adducts, we chose to
examine this main group species with both the t-butyl-
substituted tBuDAB ligand and the cyclohexyl-substituted
cHexDAB ligand (Scheme 5) to see if this reactivity trend
continued.
Initial experiments involved the treatment of Me2GaCl with
the tBuDAB ligand. A similar experimental procedure to those
described above was followed, utilizing a 1:1 Ga/tBuDAB ligand
ratio in toluene. In this case, the reaction was somewhat
exothermic. White crystals formed immediately from the dark
orange solution that gradually redissolved into the solvent. The
reaction was allowed to stir overnight. Orange crystals
precipitated from solution as the solution became darker in
color upon gradual removal of the solvent. The isolated crystals
were washed with cold pentane, giving 15 in essentially
quantitative yield (based on Ga) in analytically pure form. The
1H NMR analysis indicated that the structure was a cation/
anion pair similar to what we had seen for 8 and 10, and
this solution structure was confirmed by X-ray diffraction
In a similar preparative route to the other reactions above,
changing the main group species to Me3Ga and using the
DiPPDAB ligand afforded 13 in essentially quantitative yield.
The reaction was allowed to stir at room temperature for 18 h
in toluene, and concentration of the orange solution by gradual
removal of toluene gave 13 as analytically pure, yellow crystals.
1
Analysis of the H NMR spectrum indicated that, as with 12,
the structure in solution was of the inserted/rearranged type. In
this case, the integrated ratio of protons in the −GaMe2 moiety
versus the backbone −CH3 group was important in ration-
alizing the solution structure, along with the absence of vinyl
protons. The solid-state structure was confirmed by single
crystal X-ray diffraction. Briefly, 13 crystallized in the
centrosymmetric space group Pnma and as such had a mirror
plane of symmetry through the GaMe2 group as well as
bisecting the five-membered ring. Thus, individual C−N versus
CN bonds could not be distinguished. As this structure was
very similar to others presented above, the thermal ellipsoid
plot of 13 is presented only in the Supporting Information.
When GaMe3 is treated with the dimethylated backbone
diimine DiPPDAB(Me2) in toluene at room temperature for 3 h,
14 can be isolated in excellent yield (91.4%) as analytically
pure, yellow crystals upon slow removal of the solvent. As with
the Al analogue 6, 1H NMR spectral evidence indicated that the
DiPPDAB ligand had inserted into a Ga−Me bond of the GaMe3
molecule and then undergone rearrangement. As in 6, addition
of the two methyl groups on the backbone of a sterically
(Figure 9). 15 crystallized in space group P1, and the structure
̅
R
crowded DAB ligand is not capable of blocking the insertion
process and forcing an adduct to be formed. The solid-state
structure was determined by X-ray diffraction and showed that
the structures in the solid and solution were consistent. Unlike
13, 14 did not crystallize in a centrosymmetric space group,
thus allowing the individual Ga(1)−N(1), Ga(1)−N(2),
N(1)−C(26), and N(2)−C(25) bond lengths to be distin-
guished. The thermal ellipsoid plot for 14 is shown in the
Supporting Information, while the important crystallographic
data are given in Table 2.
Figure 9. Molecular structure of 15 (50% ellipsoids). For clarity,
hydrogen atoms are not shown. Selected bond lengths (Å) and angles
(deg): Ga(1)−N(1) = 2.044(3), Ga(1)−N(2) = 2.054(3), C(10)−
N(1) = 1.263(5), C(9)−N(2) = 1.258(5); N(1)−Ga(1)−N(2) =
80.80(12).
indicated that the discrete cation and anion pairs were well-
separated and had no interionic contacts. Although there is no
strict plane of symmetry bisecting the five-membered ring,
there is a virtual plane of symmetry as the Ga(1)−N(2) and
Ga(1)−N(1) bond distances are virtually identical (2.054(3)
and 2.040(3) Å, respectively) and the N(1)−C(10) and N(2)−
C(9) bond distances are also identical (1.263(5) and 1.258(5) Å,
respectively). As mentioned previously, Cowley et al. have
reported a structurally characterized [GaCl2]+ cation containing a
tBuDAB ligand that also has two σ-bound N donor atoms, formed
Aliphatic RDAB LigandsGallium and Aluminum
Complexes. In order to serve as comparison molecules, we
were interested in utilizing the more electron-rich, stronger-
donating alkyl-substituted RDAB ligands10 in place of the
various aromatic groups used earlier. Rather than exploring all
possible permutations, it was decided to examine only a small
subset of possible combinations, as we had seen above that only
4660
dx.doi.org/10.1021/ic202583w | Inorg. Chem. 2012, 51, 4650−4662

Inorganic Chemistry
Article
when 2 equiv of GaCl3 were treated with the tBuDAB ligand.21a
We note that the N(6)−Ga(2)−N(7) angle in 15 is slightly
more acute than what is observed in the Cowley et al. compound
or in 8 (80.80(12)° for 15 versus 84.4(4)° for Cowley et al. and
83.56(13) for 8), and the Ga−N bonds found in 15 are slightly
longer as well.
Me2GaCl was treated in a similar fashion with the less-
hindered cHexDAB ligand. A 1:1 ratio of Ga/cHexDAB ligand
produced a light green solution that was immediately placed in
the freezer. White crystals of 16 were isolated at −20 °C. Upon
standing in the freezer, the crystals slowly acquired an orange
color and appeared to be unstable at room temperature. The
isolated yield of 16 was 79.6% when an optimal 2:1 Ga/diimine
ratio was used (after identification of the product). Like the
previous AlkylDAB derivative 15, both the 1H NMR data as well
as the X-ray structure of 16 were consistent with the structural
assignment as a cation/anion pair. 16 crystallized in space
reactions. The addition of methyl groups onto the backbone of
the RDAB ligand is insufficient to stop the insertion and
migration process from occurring. In the case of MeAlCl2 with
DiPPDAB, the reaction could be followed spectroscopically from
the monoadduct through the inserted/rearranged final product.
Methylated gallium species, however, are much less predictable
in their behavior with aromatic-substituted RDAB ligands.
Depending on the exact species used, coordinated adducts can
be formed and identified, or inserted/rearranged products
similar to the aluminum reactions can be obtained. Quite
interesting, cation/anion pairs were formed in which GaCl3 or
MeGaCl2 acted as chloride acceptors, in stark contrast to the
analogous Al reaction which formed either a dicoordinated
adduct or an inserted/rearranged product. When alkyl-
substituted RDAB ligands were used with Me2GaCl, only
cation/anion pairs were obtained. Unfortunately, when the
same reactions were performed using Me2AlCl as a reagent,
only irreproducible results were obtained. The multicolored
nature exhibited by single crystals of 12 was not found in any
other compound, thus indicating that 12 appears to be unique
in its ability to form yellow, green, and mixed yellow−green
single crystals.
group P1, which allowed for all of the various discrete bond
̅
lengths in the heterocycle to be determined (structure shown in
the Supporting Information). Similarly to 15, there was a virtual
plane of symmetry bisecting the five-membered ring due to the
virtually identical bond lengths of Ga(1)−N(1) and Ga(1)−
N(2) and of N(2)−C(1) and N(1)−C(2), respectively. The
bond lengths and angles around the central Ga atom were not
remarkable relative to 15 and as such do not warrant detailed
comments.
ASSOCIATED CONTENT
* Supporting Information
■
S
Crystallographic data in CIF format for compounds 1−4, 6−
11, and 13−16 and thermal ellipsoid plots of compounds 4, 5,
9, 11−14, and 16. This material is available free of charge via
the Internet at http://pubs.acs.org.
Last, we note that we also examined the interactions of
Me2AlCl with the tBuDAB ligand. Quite interestingly, and very
unexpectedly based on our other results, the reactions using
Me2AlCl were quite inconsistent and irreproducible from
experiment to experiment. Reactions performed under
seemingly identical conditions gave completely different
isolated products. We have performed these experiments
using purchased Me2AlCl obtained in hexane solution, as well
as freshly prepared solid Me2AlCl prepared from Me3Al and
AlCl3. We have obtained crystalline material in all cases, but the
various structures obtained do not appear to correlate with any
known preparative variable. It did appear, however, that the
majority of the Me2AlCl/tBuDAB experiments gave cation/
anion pairs. However, the anionic fragment of the molecules,
which would be expected to be the [Me2AlCl2]− anion after
chloride abstraction with Me2AlCl, was found in only some of
the structures. The [MeAlCl3]− anion was also found via
crystallography, thus implying that some type of methyl/halide
exchange was occurring in solution. This dynamic behavior is
similar to the MeGaCl2 case mentioned above that complicates
the crystallographic identification of 10. Aluminum alkyls and
halides are known to redistribute in solution and can be used
deliberately in a preparative sense, and so this complicating
reaction must be occurring at a rate that is proving to be
synthetically confounding. Attempts to understand this ill-
behaved chemistry are currently underway.
AUTHOR INFORMATION
Corresponding Author
*E-mail: rakemp@unm.edu.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was financially supported by the National Science
Foundation (Grant CHE09-11110 to R.A.K.), and the
Laboratory Directed Research and Development Program at
Sandia National Laboratories (LDRD 69940). D.A.D. was
financially supported by the Natural Sciences and Engineering
Research Council (NSERC) of Canada by means of a
Postdoctoral Fellowship. The Bruker X-ray diffractometer was
purchased via a National Science Foundation CRIF:MU award
to the University of New Mexico (CHE04-43580), and the
NMR spectrometers were upgraded via grants from the NSF
(CHE08-40523 and CHE09-46690). Dr. Michael Katz (North-
western University) provided valuable assistance with the X-ray
crystallographic analysis of 10. Sandia National Laboratories is a
multiprogram laboratory managed and operated by Sandia
Corporation, a wholly owned subsidiary of Lockheed Martin
Corporation, for the U.S. Department of Energy’s National
Nuclear Security Administration under contract DE-AC04-
94AL85000.
SUMMARY
■
In this report, we have shown that MexMCl3−x (x = 0−3, M =
Al, Ga) react with various aromatic and alkyl-substituted RDAB
ligands give a variety of structures in solution and in the solid
state. When combined with other related structures previously
published in the literature, general trends of reactivity of these
species can be deduced, although there are still some
unexplained modes of reactivity. All methylated aluminum
species react with aromatic-substituted RDAB ligands to provide
final products that result from insertion and rearrangement
REFERENCES
■
(1) (a) Johnson, L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc.
1995, 117, 6414. (b) Small, B. L.; Brookhart, M.; Bennett, A. M. A.
J. Am. Chem. Soc. 1998, 120, 4049. (c) Svejda, S. A.; Brookhart, M.
Organometallics 1999, 18, 65. (d) Domski, G. J.; Rose, J. M.; Coates,
G. W.; Bolig, A. D.; Brookhart, M. Prog. Polym. Sci. 2007, 32, 30.
(e) Anselment, T. M. J.; Vagin, S. I.; Rieger, B. Dalton Trans. 2008,
4661
dx.doi.org/10.1021/ic202583w | Inorg. Chem. 2012, 51, 4650−4662

Inorganic Chemistry
Article
4537. (f) Gibson, V. C.; O’Reilly, R. K.; Reed, W.; Wass, D. F.; White,
A. J. P.; Williams, D. J. Chem. Commun. 2002, 1850. (g) Gibson, V. C.;
Tomov, A.; Wass, D. F.; White, A. J. P.; Williams, D. J. J. Chem. Soc.,
Dalton Trans. 2002, 2261. (h) O’Reilly, R. K.; Shaver, M. P.; Gibson,
V. C.; White, A. J. P. Macromolecules 2007, 40, 7441. (i) Camacho,
D. H.; Guan, Z. Chem. Commun. 2010, 46, 7879. (j) Guan, Z.;
Popeney, C. S. Top. Organomet. Chem. 2009, 26, 179. (k) Lees, A. J.
Mol. Supramol. Photochem. 2001, 7, 209. (l) Vlcek, A. Jr. Coord. Chem.
Rev. 2002, 230, 225. (m) Hartl, F.; Rosa, P.; Ricard, L.; Le Floch, P.;
Zalis, S. Coord. Chem. Rev. 2007, 251, 557. (n) van Slageren, J.; Klein,
A.; Zalis, S. Coord. Chem. Rev. 2002, 230, 193. (o) Fedushkin, I. L.;
Maslova, O. V.; Lukoyanov, A. N.; Fukin, G. K. C. R. Chim. 2010, 13,
584. (p) Trifonov, A. A. Eur. J. Inorg. Chem. 2007, 3151. (q) Ward,
M. D. Coord. Chem. Rev. 2010, 254, 2634. (r) Imbert, D.; Cantuel, M.;
Buenzli, J.-C. G.; Bernardinelli, G.; Piguet, C. J. Am. Chem. Soc. 2003,
125, 15698.
(2) van Koten, G.; Vrieze, K. Adv. Organomet. Chem. 1982, 21, 151.
(3) Ittel, S. D.; Johnson, L. K.; Brookhart, M. Chem. Rev. 2000, 100,
1169.
(4) Page, G.; Felix, A. M.; Dickie, D. A.; Chen, L.; Kemp, R. A. Main
Group Chem. 2010, 9, 11.
(5) Klerks, J. M.; Stufkens, D. J.; Van Koten, G.; Vrieze, K.
J. Organomet. Chem. 1979, 181, 271.
(6) Hill, N. J.; Vargas-Baca, I.; Cowley, A. H. Dalton Trans. 2009,
240.
(7) Shriver, D. F.; Drezdzon, M. A. The Manipulation of Air-Sensitive
Compounds, 2nd ed.; Wiley-Interscience: New York, 1986; p 336.
(8) tom Dieck, H.; Renk, I. W. Chem. Ber. 1971, 104, 92.
(9) tom Dieck, H.; Svoboda, M.; Greiser, T. Z. Naturforsch., B: Chem.
Sci. 1981, 36B, 823.
(10) Kliegman, J. M.; Barnes, R. K. Tetrahedron 1970, 26, 2555.
(11) Beachley, O. T. Jr; Rosenblum, D. B.; MacRae, D. J.
Organometallics 2001, 20, 945.
(12) APEX2; Bruker AXS, Inc.: Madison, WI, 2007.
(13) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112.
(14) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7.
(15) Brandenburg, K. Diamond, 3.1g; Crystal Impact GbR: Bonn,
Germany, 1999.
(16) Kanters, J. A.; van Mier, G. P. M; Nijs, R. L. L. M.; van der
Steen, F.; van Koten, G. Acta Crystallogr. 1988, C44, 1391.
(17) Hinchliffe, A.; Mair, F. S.; McInnes, E. J. L.; Pritchard, R. G.;
Warren, J. E. Dalton Trans. 2008, 222.
(18) Gibson, V. C.; Redshaw, C.; White, A. J. P.; Williams, D. J.
J. Organomet. Chem. 1998, 550, 453.
(19) After we had deposited this structure into the CSD (661032),
the structure was re-done and published, see: Olson, J. A.; Boyd, R.;
Quail, J. W.; Foley, S. R. Organometallics 2010, 27, 5333.
(20) (a) Baker, R. J.; Farley, R. D.; Jones, C.; Kloth, M.; Murphy,
D. M. J. Chem. Soc., Dalton Trans. 2002, 3844. (b) Bonello, O.; Jones,
C.; Stasch, A.; Woodul, W. D. Organometallics 2010, 29, 4914.
(21) (a) Clyburne, J. A. C.; Culp, R. D.; Kamepalli, S.; Cowley, A. H.;
Decken, A. Inorg. Chem. 1996, 35, 6651. (b) Jenkins, H. A.;
Dumaresque, C. L.; Vidovic, D.; Clyburne, J. A. C. Can. J. Chem.
2002, 80, 1398.
4662
dx.doi.org/10.1021/ic202583w | Inorg. Chem. 2012, 51, 4650−4662