A General Route to Labile Niobium and Tantalum d° Monoimides. Discussion of Metal-Nitrogen Vibrational Modes

Article abstract of DOI:10.1021/ic961360j

Reaction of TaCl₅ with 2 equiv of an amine in the presence of sodium silicate and pyridine affords Ta(NR)Cl₃(py)₂ in good yield. Reaction of NbCl₅ with ZnCl₂ followed by addition of an amine RNH₂ and pyridine affords M(NR)Cl₃(dme) (dme is 1,2-dimethoxyethane). For niobium this reaction proceeds smoothly regardless of the amine but is ineffective with tantalum and alkyl amines. An alternative route involves reaction of TaCl₂ with 3 equiv of RNH₂ to form [RNH₃]₂[Ta(NR)Cl₅], followed by reaction of this salt with ZnCl₂ in the presence of dme. The molecular structure of Nb(N^tBu)Cl₃(dme) (formula C₈H₁₉Cl₃NNbO₂) was determined by X-ray crystallography (monoclinic space group Cc with a = 30.565(4) ?, b = 7.2406(13) ?, c = 13.915(2) ?, = 90.626(7)°, V = 3079.4(8) ?³, Z = 8). The Nb-N bond length is 1.72 ? with a Nb-N-C bond angle of 177° in a distorted octahedral structure. In order to characterize the M-N stretching frequencies in these compounds, IR data for each compound are compared with calculated stretching frequencies using the commercially available Spartan calculation package. These experiments reveal that there is no real M-N stretching frequency in these imidos. Rather, the M-N modes are strongly coupled to N-C and C-H or C-C modes in these imidos. IR active modes are observed at ～ 260 cm^-1 for tantalum alkyl imidos and ～1350 cm^-1 for tantalum aryl imidos. These correspond to a Ta-(N-C) stretch coupled to the CR₃ umbrella deformation.

Full text of DOI:10.1021/ic961360j

Inorg. Chem. 1997, 36, 2647-2655
2647
A General Route to Labile Niobium and Tantalum d0 Monoimides. Discussion of
Metal-Nitrogen Vibrational Modes
Andrey V. Korolev, Arnold L. Rheingold, and Darryl S. Williams*
Department of Chemistry, 145, Wayne State University, Detroit, Michigan 48202, and Department of
Chemistry and Biochemistry, University of Delaware, Newark, Delaware 19716
ReceiVed NoVember 14, 1996X
Reaction of TaCl5 with 2 equiv of an amine in the presence of sodium silicate and pyridine affords Ta(NR)Cl3(py)2
in good yield. Reaction of NbCl5 with ZnCl2 followed by addition of an amine RNH2 and pyridine affords
M(NR)Cl3(dme) (dme is 1,2-dimethoxyethane). For niobium this reaction proceeds smoothly regardless of the
amine but is ineffective with tantalum and alkyl amines. An alternative route involves reaction of TaCl5 with 3
equiv of RNH2 to form [RNH3]2[Ta(NR)Cl5], followed by reaction of this salt with ZnCl2 in the presence of dme.
The molecular structure of Nb(NtBu)Cl3(dme) (formula C8H19Cl3NNbO2) was determined by X-ray crystallography
(monoclinic space group Cc with a ) 30.565(4) Å, b ) 7.2406(13) Å, c ) 13.915(2) Å, â ) 90.626(7)°, V )
3079.4(8) Å3, Z ) 8). The Nb-N bond length is 1.72 Å with a Nb-N-C bond angle of 177° in a distorted
octahedral structure. In order to characterize the M-N stretching frequencies in these compounds, IR data for
each compound are compared with calculated stretching frequencies using the commercially available Spartan
calculation package. These experiments reveal that there is no real M-N stretching frequency in these imidos.
Rather, the M-N modes are strongly coupled to N-C and C-H or C-C modes in these imidos. IR active
modes are observed at ∼1260 cm-1 for tantalum alkyl imidos and ∼1350 cm-1 for tantalum aryl imidos. These
correspond to a Ta-(N-C) stretch coupled to the CR3 umbrella deformation.
Introduction
Transition metal compounds containing imido ligands (RN2-
reaction with silylarylamines in the presence of thf or dme (1,2-
dimethoxyethane), however, affords the etherated compounds.6
In this case, the aniline derivative produced is labile enough to
be replaced with the weakly coordinating dme. These materials
are much more useful for subsequent derivatization.4,6,7
)
have become extremely important in inorganic chemistry.1 These
divalent ligands potentially provide a total of three bonds to a
metal center, one σ and two π, depending on the ability of the
metal to form a total of three bonds to any given imido ligand.1
Early routes to Ta and Nb imido compounds involved reduction
of acetonitrile with MCl4 and zinc2 or reaction of tantalum
neopentylidenes with imines in a Wittig-like reaction.3,4 This
method could be fairly general since a variety of imines could
be used, but the expense of preparing Ta(CHtBu)Cl3(thf)2, the
precursor, is undesirable.
Finally, some clever reactions involving hexamethyldisilazane
with the metal halides afford silylimidos.8 This result is directly
analogous to the above reactions and valuable due to the
instability of Me3SiNH2,9 making it impossible to prepare the
imido compound using the methods described below.
Somewhat later, Nielson reported that silated alkylamines
react with MCl5 (M ) Nb, Ta) to give the dimeric (bridging
halide) compounds shown:5
Our interest in group 5 monoimido compounds stems from
the extremely interesting properties of their photoexcited
states.7,10-12 Examples of luminescence from solutions of these
compounds at room temperature have been reported recently,
with emission quantum yields ranging from 0.25 to 0.0001 and
emission lifetimes from 20 µs to 20 ns.10,11 We have found
that the alkyl substituent has a substantial effect on the
absorption and emission energies in these compounds.10,11,13
The coordinated amine in these compounds can be replaced with
small phosphines, but the resulting species were not isolated
due to their oily, tacky nature. This is particularly problematic
since exchange of the amine with other bases is often difficult,
and its presence obviously detracts from the usefulness of these
materials in the synthesis of alkylated compounds. A similar
(6) Chao, Y.-W.; Wexler, P. A.; Wigley, D. E. Inorg. Chem. 1989, 28,
3860.
(7) Williams, D. S.; Elgammal, R. A.; Heeg, M. J. Manuscript in
preparation.
* To whom correspondence should be addressed at Wayne State
University.
(8) Jones, C. M.; Lerchen, M. E.; Church, C. J.; Schomber, B. M.; Doherty,
N. M. Inorg. Chem. 1990, 29, 1679.
(9) Wiberg, N.; Uhlenbrock, W. Chem. Ber. 1971, 104, 2643.
(10) Korolev, A. V.; Williams, D. S. Manuscript in preparation.
(11) Williams, D. S.; Thompson, D. W.; Korolev, A. V. J. Am. Chem.
Soc. 1996, 118, 6526.
X Abstract published in AdVance ACS Abstracts, May 15, 1997.
(1) Wigley, D. E. Prog. Inorg. Chem 1994, 42, 239.
(2) Finn, P. A.; King, M. S.; Kilty, P. A.; McCarley, R. E. J. Am. Chem.
Soc. 1975, 97, 220.
(3) Rocklage, S. M.; Schrock, R. R. J. Am. Chem. Soc. 1980, 102, 7808.
(4) Rocklage, S. M.; Schrock, R. R. J. Am. Chem. Soc. 1982, 104, 3077.
(5) Bates, P. A.; Nielson, A. J.; Waters, J. M. Polyhedron 1985, 8, 1391.
(12) Heinselman, K. S.; Hopkins, M. D. J. Am. Chem. Soc. 1995, 117,
12340.
(13) Korolev, A. V.; Williams, D. S. Manuscript in preparation.
S0020-1669(96)01360-2 CCC: $14.00 © 1997 American Chemical Society

2648 Inorganic Chemistry, Vol. 36, No. 12, 1997
Korolev et al.
Therefore access to a variety of these imidos is important in
probing electronic structure. In our search for labile, electroni-
cally simple materials, we sought routes other than those above
which lead to monoimido compounds containing simple, labile
ligands, specifically, M(NR)Cl3(dme), directly from RNH2. We
were also interested in modifying the Wigley procedure, eq 3,
in order to avoid generation of the relatively expensive
Me3SiNHAr prior to reaction with the metal halide. Addition-
ally, the usefulness of the above methods will be limited by
our ability to prepare silated amines. We have found a
surprisingly simple reaction sequence that avoids these problems
and is effective for each amine we have investigated.
Finally, we are interested in establishing the metal-nitrogen
stretching frequency in these compounds in order to judge the
variation of the metal-nitrogen interaction; there has been
significant controversy over the location of these absorptions
in IR spectra.1 A number of assignments on group 5 imido
compounds report two absorptions that can be attributed to
M-N modes.1,14 There is reportedly1 a low-frequency mode
in the range 900-1000 cm-1 and a higher energy mode between
1100 and 1350 cm-1. Stra¨hle14 showed that the M-N stretch
moved from 1107 to 963 cm-1 upon changing X from Cl to I
in the series (XN)VCl3.15 The lower frequency mode changed
from 510 to 390 cm-1 in the same comparison. Furthermore,
Stra¨hle predicts higher energy MN modes in alkylimido
compounds.14 Several papers corroborate this conclusion;
however, Osborne makes an opposite assignment in (PhN)VCp2,
based on 15N labeling experiments. The NC mode is assigned
to a 1330 cm-1 absorption, while the 934 cm-1 absorption is
assigned to the MN absorption (Cp2M compounds may be
anomalous). We hope to clarify this argument.
these reactions result in yields of 50% or less. Each of the
tantalum alkylimido bis(pyridine)s, with the exception of
compounds 5a and 6a (R ) CMe2CH2CMe3 and n-Bu), can be
prepared directly from the amine and TaCl5 in reasonable
isolated yields using Na2SiO3. Unfortunately, attempts at
preparation of the dme adducts of alkylimidos using this method
were unsuccessful (<20% yield).
Use of Amines as Proton Acceptors. On the basis of the
related syntheses of groups 616,17 and 718,19 imidos, it seemed
reasonable that pyridine or triethylamine could be used as an
external base in these reactions. Addition of, for example, a
solution of aniline (PhNH2; 1 equiv) and pyridine to TaCl5 in
Et2O/toluene does not result in formation of the imido.
However, pretreatment of the tantalum solution with 2 equiv
of ZnCl2 results in formation of a fine white precipitate which
dissolves upon addition of the amine mixture and a high yield
of Ta(NPh)Cl3(dme) (7) after filtration and recrystallization (eq
7). This method also affords compounds 8-12 (R ) 2,6-
iPr2C6H3 (Ar), 2,6-Me2C6H3 (Ar′), 4-CNC6H4, 2-naphthyl,
2-anthryl) in high yield.
Perhaps the most valuable success of this route is in the
synthesis of compounds 11 and 12. Thus, 2-aminonaphthalene
(NapNH2) and 2-aminoanthracene (AnthNH2) can be readily
converted into the corresponding imidos 11 and 12, respectively
in good yield. On the basis of these observations, this method
should smoothly provide arylimidos of the form Ta(NAryl)-
Cl3(dme) regardless of the amine. Importantly, the lability of
the dme ligand makes these compounds valuable starting
materials for synthesis of other monoimido compounds.4,6,7 For
example, addition of py to Ta(NAr)Cl3(dme) (8)12 affords
Ta(NAr)Cl3(py)2 quantitatively.
Results
The reaction of interest, eq 5, is a simple one, but the
relatively poor solubility these M(NR)Cl3L2 compounds makes
purification difficult, even though conversion is generally high.
We report two general ways of preparing these compounds by
reaction of primary amines with MCl5 (M ) Nb, Ta). One
approach is to use an insoluble inorganic base to remove the
byproduct HCl. The other is to use a combination of a soluble
organic base and zinc chloride for the same task. In each case,
precipitation of the salt byproducts from the reaction mixture
is key to purification of the imido product.
Unfortunately, this method is unsuccessful for synthesis of
the corresponding alkylimidos using dme as the Lewis base,
Ta(NR)Cl3(dme), 1c-6c. In some cases, the desired compound
could be isolated in low yields (<20%). Fortunately, addition
of 3 equiv of tANH2 to a solution of TaCl5 in Et2O/toluene
quantitatively affords the salt [tANH3]+2[Ta(NtA)Cl5]2- (1b),
eq 8. A related transformation has been observed recently for
Synthesis of Tantalum Imides. Generally, results for the
alkylamines were fairly consistent and somewhat different from
the aniline derivatives. As examples, three compounds will be
discussed, Ta(NR)Cl3L2 (R ) CMe2Et (tA), L ) py (1a), L )
1
1/2 dme (1c); R ) C6H5, L ) /2 dme (7)).
Inorganic Bases as Proton Acceptors. Addition of anhy-
drous Na2SiO3 to a solution of TaCl5 in Et2O/toluene followed
by 1 or 2 equiv of tert-amyl amine (tANH2) results in a pale
yellow to yellow-green solution. Pyridine is then added to give
a yellow solution. After the mixture is stirred overnight at
ambient temperature, the product is filtered from the insoluble
salts, and excess amine is readily removed after concentration
in Vacuo with a pentane wash. This procedure yields Ta(NtA)-
Cl3(py)2 (1a) as a pure, slightly yellow powder. Consistently,
use of only 1 equiv of amine results in yields of 40-50%, but
using 2 equiv increases the yield to ∼70%. We tried a number
of inorganic bases, including K2CO3, Na2B4O7, and CaH2, but
arylamines.20 This salt can then be treated with ZnCl2 in the
presence of dme to afford Ta(NtA)Cl3(dme) (1c) in 86% yield,
eq 9. Reactions with other alkylamines also proceed in high
(16) Fox, H. H.; Yap, K. B.; Robbins, J.; Cai, S.; Schrock, R. R. Inorg.
Chem. 1992, 31, 2287.
(17) Schrock, R. R.; DePue, R. T.; Feldman, J.; Yap, K. B.; Yang, D. C.;
Davis, W. M.; Park, L. Y.; DiMare, M.; Schofield, M.; Anhaus, J.;
Walborsky, E.; Evitt, E.; Kru¨ger, C.; Betz, P. Organometallics 1990,
9, 2262.
(18) Toreki, R.; Schrock, R. R.; Davis, W. M. J. Am. Chem. Soc. 1992,
114, 3367.
(19) Williams, D. S.; Schrock, R. R. Organometallics 1993, 12, 1148.
(20) Jayaratne, K. C.; Yap, G. P.; Haggerty, B. S.; Rheingold, A. L.; Winter,
C. H. Inorg. Chem. 1996, 35, 4910.
(14) Dehnicke, K.; Stra¨hle, J. Angew. Chem., Int. Ed. Engl. 1981, 20, 413.
(15) Dehnicke, K.; Stra¨hle, J. Chem. ReV. 1993, 93, 981.

Labile Niobium and Tantalum d0 Monoimides
Inorganic Chemistry, Vol. 36, No. 12, 1997 2649
yield to form the pentachloride salts (2b-6b) and, subsequently,
the corresponding dme adducts (2c-6c).
This final method is highly desirable since the amine
substituent does not seem to affect the course of the reaction
and the product, formed in good yield, is a labile, useful starting
material for further syntheses.
Synthesis of Niobium Imides. Our results for the Nb
compounds differ somewhat. We have been successful in
preparing all of the niobium imidos according to eq 7. Both
the pyridine and dme adducts are thus available directly from
NbCl5, the amine, pyridine or NEt3, and ZnCl2 in high yield.
Reaction of alkylimido pentachloride salts with ZnCl2 and dme
also results in good yields of Nb(NtA)Cl3(dme) (13c) and the
other alkylimides. This method affords the desired compounds
in somewhat better yield. Finally, attempts at using inorganic
bases to prepare the niobium imidos generally fail, particularly
using CaH2 as base. We attribute this to the greater tendency
for Nb to reduce compared to Ta and to the greater lability of
second-row elements compared to their heavier congeners.
Figure 1. Molecular structure of Nb(NCMe3)Cl3(dme) (14).
Spectroscopic Properties. The geometry of these species
is clear from 1H NMR and consistent with previous observations
by Wigley.6 In 1a, for example, the tA geminal methyl groups
are equivalent as are the geminal protons on the methylene
carbon, indicating a plane of symmetry containing the Ta-N-C
axis. The pyridine ligands are inequivalent, however, and one
of the pyridines exchanges rapidly on the NMR time scale with
free pyridine in solution. It seems clear then that the chlorides
are all cis to the imido group and that the pyridines are cis.
Presumably, the pyridine trans to the imido is the one which
exchanges with free pyridine. NMR data for all of the bis-
(pyridine) adducts are similar. Furthermore, data for the dme
adducts are also consistent with the cis,mer-M(NR)Cl3L2
geometry.
Molecular Structure of Nb(NtBu)Cl3(dme). One of the
earliest structural studies on a transition metal imido compound
involved the dimeric diimido NbV compound [Nb2Cl8(NCMe)2-
(η2-NC(Me)C(Me)CN)]2-, which was prepared by reduction of
NbCl4 in acetonitrile.2 This was taken advantage of in a later
paper on synthesis of bridging imidos from imines using NbIV.21
A small number of other structures of group 5 imido compounds
exist formulated as [M(NR)Cl4L]-, where R is an alkyl
substituent. Structures of compounds with weakly coordinating
Lewis bases, such as THF or dme, are rare. Also, we are
interested in structural effects on excited state properties.
Therefore, we crystallographically characterized the molecular
structure of Nb(NtBu)Cl3(dme) (14), shown in Figure 1. There
are two chemically identical molecules (see Experimental
Details), of which molecule A is discussed. Crystallographic
data and relevant bond lengths and angles are listed in Tables
1 and 2, respectively. The structure is intriguing in a number
of ways. The niobium-nitrogen bond length is 1.722(7) Å,
short for an imido ligand, but within ∼0.04 Å of the average
(∼1.76 Å).1 As expected, the Nb-N-C angle is essentially
linear, 176.8(7)°. Also typical of compounds containing
multiply bonded ligands, the equatorial ligands are “pushed
back” from the equatorial plane by the electronic influence of
the imido, resulting in N-Nb-L angles of between 95 and 100°.
The halides are slightly distorted from perfect octahedral
geometry in the plane as well; Cl(1)-Nb-Cl(2) and Cl(3)-
Nb-Cl(2) angles are 94.92(10) and 92.93(11)°, respectively.
This occurs even though the equatorial Nb-O bond is signifi-
cantly shorter (by 0.2 Å) than the Nb-O bond trans to the
Figure 2. Assigned [Ta-N-C] medium frequency vibrational mode
(ν ) 1171 cm-1) calculated for Ta(NCMe3)Cl3. The magnitude of the
vibration is exaggerated for clarity.
Table 1. Crystallographic Data for Nb(NCMe3)Cl3(dme) (14)
formula
fw
C8H19NO2Cl3Nb
360.50
monoclinic
Cc
cryst system
space group
a-c, Å
30.565(4), 7.2406(13), 13.915(2)
â, deg
90.626(7)
3079.4(8)
8
1.555
12.9
0.710 73
249
3.92, 10.25
V, Å3
Z
Dx, g cm-3
µ(Mo KR), cm-1
λ, Å
T, K
R(F), R(wF2),a %
2
a Quantity minimized ) R(F) ) ∑∆/∑(Fo); R(wF2) ) ∑[w(Fo
-
2
Fc2)2]/∑[(wFo )2]1/2; ∆ ) |Fo - Fc|.
imido. The latter may be attributed to a trans influence and
has been observed in the structure of closely related Mo(N)-
Cl3(dme).22
(21) Roskamp, E. J.; Pedersen, S. F. J. Am. Chem. Soc. 1987, 109, 3152.

2650 Inorganic Chemistry, Vol. 36, No. 12, 1997
Korolev et al.
Table 2. Selected Bond Distances (Å) and Angles (deg) for
Nb(NCMe3)Cl3(dme) (14)
of the normal coordinate analysis. Visual inspection of the
animated vibrational modes more clearly shows the atomic
motions involved in each predicted vibrational mode. Modes
in which both Ta and N atoms are significantly distorted from
their equilibrium positions were taken as (Ta-N)-containing
modes. These modes are fairly unambiguous since, in the
1000-1400 cm-1 region of the spectrum, few modes involve
the Ta atom, and only one to three involve both Ta and N to a
significant degree, depending on the molecule.
molecule A
molecule B
Lengths
1.722(7)
2.400(3)
2.366(3)
2.424(3)
2.389(6)
2.187(6)
1.463(1)
Nb-N
1.730(8)
2.385(3)
2.361(3)
2.421(3)
2.375(7)
2.208(6)
1.463(1)
Nb-Cl(1)
Nb-Cl(2)
Nb-Cl(3)
Nb-O(1)
Nb-O(2)
N-C(5)
As the group bonded to nitrogen changes, a noticeable change
in the frequency of the mode which is mainly Ta-N in character
occurs. For the nitride, the Ta-N stretching mode is observed
at 1120 (1008) cm-1, where one would predict on the basis of
previous literature.14 In the protioimide, this stretching fre-
quency decreases to 1083 (975) cm-1, closer to that expected
for an oxo and consistent with what should be a (slight) bond
weakening upon protonation of the nitride. This mode mainly
involves a Ta and N symmetric stretch, coupled with a small
contribution from the N-H stretch.
In the methylimido, however, this mode drastically increases
in energy to ∼1300 (1170) cm-1. This is due to coupling with
modes which are mainly N-C and C-H in character and close
to what is observed experimentally (Vide infra). This mode is
now best visualized as an umbrella deformation of the CH3
group coupled with the Ta-N mode. At higher frequency is
predicted a mainly N-C stretching mode which involves some
tantalum charactersthis can be thought of as an asymmetric
[Ta-N-C] stretching mode, in which significant nitrogen
motion occurs along the Ta-N-C axis. Finally, a symmetric
[Ta-N-C] mode, in which the Ta and C atoms distort along
this axis, is predicted to occur at 661 (595) cm-1. On the basis
of these results, there are three main stretching modes predicted
for an alkyl imido: a low-energy symmetric vibration Ta-N-
C, a medium energy Ta-(N-C) mode, and a higher energy
asymetric Ta-N-C mode, using italics to indicate atomic
motion.
Angles
94.9(1)
162.2(1)
92.9(1)
95.8(3)
81.4(2)
85.4(2)
100.8(3)
91.0(2)
161.7(2)
168.1(3)
97.4(3)
70.9(2)
176.8(7)
Cl(1)-Nb-Cl(2)
Cl(1)-Nb-Cl(3)
Cl(2)-Nb-Cl(3)
Cl(1)-Nb-N
Cl(1)-Nb-O(1)
Cl(1)-Nb-O(2)
Cl(2)-Nb-N
Cl(2)-Nb-O(1)
Cl(2)-Nb-O(2)
N-Nb-O(1)
95.7(1)
162.4(1)
92.1(1)
95.9(3)
81.7(2)
85.7(2)
100.6(3)
90.4(2)
161.6(2)
169.0(3)
97.5(3)
71.7(2)
176.5(7)
N-Nb-O(2)
O(1)-Nb-O(2)
Nb-N-C
Table 3. Calculated Energies in cm-1 for Ta-N-C Modes in
Ta(N)Cl3, Ta(NH)Cl3, Ta(NMe)Cl3, Ta(NCMe3)Cl3, Ta(NPh)Cl3,
and Ta(NCMe3)Cl3(dme)
alkyl
sym deforma- antisym
N-C-R
bending
compound
Ta(N)Cl3
Ta-N stretch
tion
stretch
1120
Ta(NH)Cl3
Ta(NCH3)Cl3
Ta(NPh)Cl3
Ta(NCMe3)Cl3
Ta(NCMe3)Cl3(dme)
1083
1296
1211
1171
1171
3569
1508
661
710
634
635
989 (E)
1475, 1645 993, 1076
1500
1499
1292, 1292
Infrared Data and the Metal-Nitrogen Stretching Fre-
quency. There has been significant controversy over metal-
nitrogen stretching frequencies in transition metal imido com-
pounds.1,23 The main problem is that this mode is strongly
coupled to the N-C stretching frequency in these compounds.14
A secondary problem that we have discovered is significant
coupling to carbon-carbon or carbon-hydrogen modes in the
alkyl group.13 In order to gain insight into this problem, we
have carried out semi-empirical calculations using the Spartan
calculation package24 on a Silicon Graphics Indigo computer
(PM3(tm) parameters) and compared the calculated vibrational
modes with the observed absorptions. A simple series was con-
structed: Ta(N)Cl3, Ta(NH)Cl3, Ta(NCH3)Cl3, Ta(NCMe3)Cl3,
and Ta(NPh)Cl3. Finally, a calculation was carried out for
Ta(NCMe3)Cl3(dme) for comparison with actual spectra ac-
quired in a Nujol mull. The molecular geometries were
optimized, molecular orbitals calculated, and vibrational fre-
quencies (normal modes) estimated with a frequency calculation.
These modes are typically assumed to be overestimated by 10%,
so values quoted here are followed parenthetically with the
calculated frequency scaled by 0.9. We consider these values
as upper and lower estimates of the vibrational frequencies.
The results are tabulated in Table 3. There is a significant
change in the mode which is mainly Ta-N in character as one
proceeds from nitride to the imides. This mode was selected
from the tabulated data (available as Supporting Information)
In the tert-butylimido, the predicted umbrella/Ta-(N-C)
stretching frequency drops again to 1171 (1054) cm-1, some-
what lower than actually observed. This may be due in part to
geometric considerations; the Ta-N bond length is predicted
to be ∼1.77 Å, significantly longer than that observed in 14
and in Ta(NAr)Ph3(THF) (1.71 Å).7 Predicted modes at 634
(571) cm-1 for the symmetric [Ta-N-C] mode and 1500
(1350) cm-1 for the asymmetric [Ta-N-C] mode are again
observed.
In our model of an actual compound, Ta(NCMe3)Cl3(dme),
the Ta-N mode is predicted to be 1171 (1054) cm-1, unchanged
as the coordination number increases from 4 to 6. The geometry
of this compound is predicted to be very similar to that observed
in the structure of 14. In the phenylimido trichloride, the picture
is more complicated. There are a number of modes in the region
1000-1500 cm-1 that have varying degrees of N and Ta
contributions, but there are two main contributors to Ta-N
stretches, one at 1211 (1090) cm-1 and another at 1475 (1327)
cm-1. Results for Ta(NPh)Cl3 are similar, with coupling of aryl
C-C stretches and the Ta-N stretch being the main distortions
in the medium-frequency analog to the umbrella mode in the
alkylimidos.
With these results in mind, infrared data for the imido
compounds were scrutinized in order to find evidence for these
three modes, Table 4. Observation of a strong absorption at
∼1250-1280 cm-1 in otherwise fairly bleak spectra of the salts
[Ta(NR)Cl5]2- (1b-6b) is perhaps the most convincing evi-
dence for the medium frequency Ta-(N-C) mode in these
compounds. This value is close to that predicted for the
(22) Rentschler, E.; Dehnicke, K. Z. Naturforsch. 1993, 48b, 1841.
(23) Nugent, W. A.; Mayer, J. M. Metal-Ligand Multiple Bonds; Wiley:
New York, 1988.
(24) WaVefunction Spartan, 4.0 ed.; Wavefunction, Inc.: 18401 Von
Karman Ave, No. 370, Irvine, CA 92715, 1995.

Labile Niobium and Tantalum d0 Monoimides
Inorganic Chemistry, Vol. 36, No. 12, 1997 2651
amine RNH2 as a Lewis base ligand.5,25 Replacement of RNH2
in these compounds with weakly coordinating ligands such as
pyridine does not occur smoothly.5 Coordinated dme is highly
desirable due to its ready loss in further reaction chemistry, such
as in Grignard reactions.7,16,17 Of more immediate interest to
us is the simplicity of the electronic structure of these
compounds.10-12 The imido ligand is the dominant factor, while
the chlorides and simple Lewis base ligands essentially serve
to fill out the coordination sphere.
Table 4. Observed Infrared Absorptions for “Ta-N-CR3” (in
cm-1) in the Ta and Nb Imido Compoundsa
no.
compound
medium
high
1a
1b
1c
2a
2c
3a
3b
3c
4a
4c
5a
5b
5c
6b
6c
7
Ta(NCMe2Et)Cl3(py)2
[Ta(NCMe2Et)Cl5]2-
Ta(NCMe2Et)Cl3(dme)
Ta(NCMe3)Cl3(py)2
Ta(NCMe3)Cl3(dme)
Ta(NCH(CHMe2)2)Cl3(py)2
[Ta(NCH(CHMe2)2)Cl5]2-
Ta(NCH(CHMe2)2)Cl3(dme)
Ta(NAda)Cl3(py)2
1255
1263
1255
1283
1275
1263
1268
1266
1286
1281
1255
1263
1261
1300
1306
1359
1354
1336
1370
1322
1218
1236
1225
1241
1327
1332
1336
1308
1589
1589
1586
1592
1603
There are a variety of six-coordinate Nb and Ta imidos of
the form M(NR)Cl3L2 where L is RNH2 or PR3.1,5 On the basis
of the previous literature, it appeared difficult to avoid incor-
poration of the amine when more than 1 equiv was added to
the metal halide, so we attempted to use nitrogen bases as HCl
scavengers. None of these preparations met with reasonable
success in the absence of ZnCl2. However, addition of ZnCl2
to MCl5 prior to amine and py results in good to excellent yields
of the arylimidos. For preparation of the alkylimidos, best
results are obtained by reaction of the halide with 3 equiv of
amine and isolation of [RNH3]+2[M(NR)Cl5]2- followed by
treatment with ZnCl2 in the presence of Lewis base. Niobium
alkylimidos may also be prepared in a one-pot reaction between
MCl5, ZnCl2, RNH2, and NEt3 in somewhat lower yield.
We tried a variety of insoluble inorganic bases such as K2CO3,
Na2B4O7, and LiBO2. Use of these bases results in only modest
(< 50%) yields. However, Na2SiO3 works well in the presence
of pyridine, giving 60-80% yields of M(NR)Cl3(py)2 when 2
equiv of amine is used, but only 40% or so in the presence of
1 equiv of amine. Even in the presence of excess base, 1 equiv
of amine is apparently lost as the ammonium salt. It is possible
that some significant concentration of a soluble base is necessary
to deprotonate the initially formed amine (M(NH2R)Cl5) and
amide (M(NHR)Cl5-) complexes, but it is not clear why the
resulting ammonium salt is not efficiently deprotonated by the
insoluble base.
Syntheses of Imidos Using Amines as Starting Materials.
There are now simple, high-yield syntheses for labile adducts
of d0 imido chloride compounds of Nb, Ta, Mo,16 W,17 and
Re18,19 for almost any alkyl or aryl substituent. It is interesting
to compare these syntheses since they are so similar. For Mo,
W, and Re, the most readily available starting materials are
oxides, and ClSiMe3 is used as an oxygen scavenger. The
simplest synthesis involves rhenium bis- and tris(imidos).
Rhenium heptoxide reacts with 6 equiv of an aniline derivative
in the presence of either pyridine or triethylamine to afford
Re(NAr)2Cl3py18 or Re(NAr)3Cl,19 respectively. In reactions
with alkyl amines, the amine serves as base to give Re(NR)3-
(OSiMe3), which can be isolated or reacted in situ with 3 equiv
of HCl to give Re(NR)2Cl3.18 The generality of these reactions
is not known but is probably similar to those described here
(tBu, Ar, and Ar′ were used previously).
Ta(NAda)Cl3(dme)
1585
Ta(NtOct)Cl3(py)2
[Ta(NtOct)Cl5]2-
1589
1599
1585
1580
1583
1606
1608
1593
1591
Ta(NtOct)Cl3(dme)
[Ta(NBu)Cl5]2-
Ta(NBu)Cl3(dme)
Ta(NPh)Cl3(dme)
Ta(NAr)Cl3(dme)
Ta(NAr′)Cl3(dme)
Ta(NPhCN)Cl3(dme)
Ta(NNap)Cl3(dme)
Nb(NCMe2Et)Cl3(py)2
[Nb(NCMe2Et)Cl5]2-
Nb(NCMe2Et)Cl3(dme)
Nb(NCMe3)Cl3(dme)
Nb(NPh)Cl3(dme)
Nb(NAr)Cl3(dme)
Nb(NPhCN)Cl3(dme)
Nb(NNap)Cl3(dme)
8
9
10
11
13a
13b
13c
14
16
17
18
19
1594
1599
1606
1584
1591
1590
t
a In Nujol mulls. Abbreviations: Ada ) 1-adamantyl, Oct )
CMe2CH2CMe3, Ar ) 2,6-iPr2C6H3, Ar′ ) 2,6-Me2C6H3, PhCN )
4-CNC6H4, Nap ) 2-naphthyl, Anth ) 2-anthryl.
methylimido trichloride and somewhat higher than the calculated
value for the tert-butylimido. Previously, Schrock reported that
only one absorption in the IR of Ta(NPh)Cl3(thf)2 (at 1360
cm-1) shifted upon 15N labeling. In agreement with this, the
midenergy Ta-N stretch in 9 is observed at 1359 cm-1, and
the energy of this mode is predicted to be ∼1475 (1327) cm-1
.
A mode is observed in this region of the spectrum for all the
arylimido compounds, while the lowest energy absorption occurs
for 11, the least electron-donating arylimido, at 1322 cm-1
.
Another feature that we are able to establish is a weak absorption
for all of these compounds at approximately 1600 cm-1. The
energy of this absorption appears to be invariant with metal,
and it corresponds to the higher energy (mainly) N-C mode in
these compounds. We were unable to reliably pick out a low
frequency mode predicted by the calculations in the 600-750
cm-1 region, corresponding to a low energy symmetric [Ta-
N-C] stretch. Similar data were obtained for a series of
vanadium, molybdenum, tungsten, and rhenium halo imido
compounds.15
The niobium compounds show similar trends, with the aryl
compounds having higher energy stretching frequencies than
their alkyl analogues. The Nb-N stretching frequencies are
consistently observed ∼30 cm-1 red of their heavier congeners.
In conclusion, these observed stretching frequencies appear to
vary as a function of vibrational mode energies rather than bond
strength, for which we have, at this point, little accurate measure.
Discussion
Our primary interest in this work was to find a high yield
route to alkylimido compounds of the form M(NR)Cl3L2, where
L is a labile Lewis base. These materials are highly desirable
for a number of reasons. The most obvious is their utility as
starting materials for group 5 imido chemistry. Known
examples of M(NR)Cl3Ln (R is alkyl) contain the corresponding
Synthesis of molybdenum bis(imido)s is somewhat more
general.16 Reaction of [NH4]2[Mo2O7] with a variety of aniline
(25) Jones, T. C.; Nielson, A. J.; Rickard, C. E. J. Chem. Soc., Chem.
Commun. 1984, 205.

2652 Inorganic Chemistry, Vol. 36, No. 12, 1997
Korolev et al.
Amylamine, tert-octylamine, tert-butylamine, and 2,4-dimethyl-3-
aminopentane were dried over CaH2 for 24 h. Zinc chloride, TaCl5,
and NbCl5 (Strem), adamantyl amine (Aldrich), 2-aminonaphtalene,
2-aminoanthracene, and 4-aminobenzonitrile (Acros) were purchased
and used as received. All NMR data were recorded in CDCl3 or C6D6
at approximately 22 °C on a Varian Gemini 300 MHz spectrometer.
All chemical shifts were referenced to internal solvent and reported in
ppm downfield from TMS (1H, 13C). CDCl3 was stirred with CaH2
for 24 h, vacuum transferred, and stored in a black bottle over activated
alumina inside the drybox. Activated alumina was prepared by heating
at 100 °C on a high vacuum line for 24 h. Deuterated solvents were
filtered through Celite immediately before use. Microanalyses (C, H,
N) were obtained from Midwest Microlab, 7212 N. Shadeland Ave.,
Indianapolis, IN, 46250. A number of these compounds were
consistently give variable results in combustion analysis. This has been
noted previously for other d0 group 5 compounds, in particular
Ta(CHR)Cl3L2,28 and a sufficient number of examples analyze satis-
factorily.
derivatives in the presence of NEt3 in refluxing dme affords
high to excellent yields (75-99%) of Mo(NAr)2Cl2(dme). For
t-BuNH2, again no additional base is required, and yields are
somewhat lower (∼60%).
Tungsten is perhaps the most complicated. WOCl4 is used
as the starting material with otherwise similar conditions to those
for molybdate.17,26 This reaction is somewhat more variable,
since precise reflux times seem more important for high-yield
reactions.26 Additionally, the noncoordinating base 2,6-lutidine
must be used, since NEt3 does not give good yields.
Preparation of Compounds. Ta(NCMe2Et)Cl3(py)2 (1a). Method
A. TaCl5 (3.00 g, 8.38 mmol) was dissolved in toluene/ether (30 mL,
5:1 by volume). Anhydrous Na2SiO3 (2.05 g, 16.8 mmol) was added
to it followed by the dropwise addition of neat H2NCMe2Et (1.96 mL,
16.8 mmol). The solution turned yellow-green, lightening to pale-green
at the end of the amine addition. Pyridine was added in excess (∼6
mL) producing a yellow solution. The mixture was stirred for 18 h,
filtered through Celite, and evaporated in Vacuo. The resulting solid
was washed with pentane and dried in Vacuo to give 3.18 g of a yellow
powder (72%). Analytically pure samples were recrystallized from
CH2Cl2/pentane.
Method B. TaCl5 (500 mg, 1.40 mmol) was dissolved in toluene/
Et2O (8 mL, 4:1 by volume). Anhydrous Na2SiO3 (341 mg, 2.79 mmol)
was added to the solution followed by the addition of tert-amylamine
(163 mL, 1.40 mmol). The solution became colorless and was stirred
for 3 min. Pyridine was added (∼1 mL), producing a yellow solution.
The mixture was stirred for 30 min, filtered through Celite, and chilled
to -40 °C. A white precipitate was filtered off and the solution
concentrated in Vacuo to give a yellow oil which crystallized after
standing under pentane, giving 0.342 g (46%) of a yellow solid. 1H
NMR (CDCl3) δ 9.01 (d, 2, pyA-Ho), 8.94 (d, 2, pyB-Ho), 7.93 (tt, 1,
pyA-Hp), 7.83 (tt, 1, pyB-Hp), 7.41 (m, 4, py-Hm), 1.66 (q, 2, CMe2CH2-
Me), 1.32 (s, 6, CMe2CH2Me), 1.10 (t, 3, CMe2CH2Me); 13C NMR
(CDCl3) δ 152.9 (pyA-Co), 151.4 (pyB-Co), 140.0 (pyA-Cp), 138.4
(pyBCp), 124.5 (pyA-Cm), 124.3 (pyB-Cm), 69.1 (CMe2CH2 Me), 37.8
(CMe2CH2Me), 29.2 (CMe2CH2Me), 9.2 (CMe2CH2Me). Anal. Calcd
for TaC15H21Cl3N3: C, 33.95; H, 3.99; N, 7.92. Found: C, 33.88; H,
4.23; N, 7.68.
By contrast, for tantalum and niobium we use the halides as
starting materials. We have found that aniline derivatives react
smoothly with niobium or tantalum pentachloride in the presence
of pyridine, ZnCl2, and dme to give good to excellent yields of
M(NAr)Cl3(dme), eq 7. Alkyl amines react similarly with
NbCl5, giving somewhat lower yields. The best way to make
mono(alkylimido)niobium and tantalum compounds (eqs 8 and
9) is to react the pentachloride with 3 equiv of amine, generating
[RNH3]2[M(NR)Cl5], and treating this salt with ZnCl2 in the
presence of dme. This affords the compounds M(NR)Cl3(dme)
in high yields.
Note that, in each of these reactions, an amine, an external
base, and a Lewis acid are used to generate the imido. In the
oxide reactions,16-18 Me3SiCl acts as a Lewis acid to accept
the oxygen, forming (Me3Si)2O. In the halide reactions here,
ZnCl2 is the Lewis acid, reacting with the ammonium chloride
byproduct to form [pyH][ZnCl3(solvent)] or [RNH3][ZnCl3-
(solvent)]. In either situation, the Lewis acid drives the reaction
to completion. This point of view can be extended to the silated
amines reacting with the group 5 halides, in which the
silyl group is the Lewis acid, removing chloride to form
Me3SiCl.5,6,25,27
Finally, we have investigated the infrared spectra of these
simple imido compounds in order to determine the observed
mode which is mainly M-N in character. The wording is
carefully chosen because there is no single mode which is solely
M-N in character. The mode which is mainly M-N in each
of these compounds is not observed at energies similar to that
of a terminal M-O stretch in an oxo compound, around 1000
cm-1. Substantial mixing of C-N and C-C or C-H modes
raises the energy of this mode to ∼1300 cm-1 ((50 cm-1). In
fact, there are at least three modes in imido compounds which
contain a significant degree of metal-nitrogen character.
[H3NCMe2Et]2[Ta(NCMe2Et)Cl5] (1b). TaCl5 (500 mg, 1.40
mmol) was dissolved in toluene (8 mL) by adding Et2O (1 mL).
H2NCMe2Et (489 mL, 4.19 mmol) was added slowly giving a yellow
solution which was stirred for 4 h. This produced a small amount of
white precipitate. The mixture was evaporated in Vacuo leaving a pale-
yellow powder which was washed with pentane and dried in Vacuo to
give 860 mg (99%) of the product: 1H NMR (CDCl3) δ 7.14 (bs, 6,
H3NCMe2Et), 1.74 (q, 4, H3NCMe2CH2CH3), 1.61 (q, 2, NCMe2CH2-
CH3), 1.39 (s, 12, H3NCMe2CH2CH3), 1.31 (s, 6, NCMe CH2CH3),
2
1.10 (t, 3, NCMe2CH2CH3), 0.99 (t, 6, H3NCMe2CH2CH3); 13C NMR
(CDCl3) δ 69.5 (NCMe2CH2CH3), 56.7 (H3NCMe2CH2CH3), 37.5
(NCMe2CH2CH3), 33.3 (H3NCMe2CH2CH3), 29.1 (NCMe2CH2CH3),
25.0 (H3NCMe2CH2CH3), 9.1 (NCMe2CH2CH3), 8.0 (H3NCMe2-
CH2CH3). Anal. Calcd for TaC15H39Cl5N3: C, 29.07; H, 6.34; N, 6.78.
Found: C, 28.68, H, 6.42; N, 7.06.
Experimental Details
General Details. All manipulations were carried out under a dry
dinitrogen atmosphere in a Vacuum/Atmospheres drybox. All solvents
and liquid reagents were distilled under dry argon over sodium/
benzophenone ketyl (THF, ether, pentane), molten sodium (toluene,
pyridine), or CaH2 (dichloromethane, NEt3). Pentane was washed using
5% HNO3/H2SO4 and dried over CaCl2 prior to distillation, with
tetraglyme added to dissolve the ketyl. (2,6-Dimethylphenyl)amine and
aniline were distilled from CaH2 after predrying with solid KOH. tert-
Ta(NCMe2Et)Cl3(dme) (1c). Compound 1b (1.00 g, 1.61 mmol)
was dissolved in toluene (8 mL) by adding 167 mL (1.61 mmol) of
dme. ZnCl2 (660 mg, 4.84 mmol) was added last. The mixture was
stirred for 18 h, during which time the pale-yellow solution became
colorless. The solution was filtered through Celite and evaporated in
Vacuo. This provided a colorless oil which crystallized upon standing
under pentane to give 640 mg (86%) of a white solid: 1H NMR (CDCl3)
δ 4.22 (s, 3, OMeA), 4.17 (m, 2, O(CH2)A), 4.03 (m, 2, O(CH2)B), 3.91
(26) Williams, D. S. Ph. D. Thesis, Massachusetts Institute of Technology,
1993.
(27) Chao, Y.-W.; Wexler, P. A.; Wigley, D. E. Inorg. Chem. 1990, 29,
(28) Rupprecht, G. A.; Messerle, L. W.; Fellman, J. D.; Schrock, R. R. J.
Am. Chem. Soc. 1980, 102, 6236.
4592.

Labile Niobium and Tantalum d0 Monoimides
Inorganic Chemistry, Vol. 36, No. 12, 1997 2653
(s, 3, OMeB), 1.62 (q, 2, CMe2CH2CH3), 1.30 (s, 6, CMe2CH2CH3),
1.11 (CMe2CH2CH3); 13C NMR (CDCl3) δ 76.0 (O(CH2)A), 71.0
(O(CH2)B), 70.6 (OMeA), 69.7 (CMe2CH2CH3), 62.2 (OMeB), 37.9
(CMe2CH2CH3), 29.3 (CMe2CH2CH3), 9.1 (CMe2CH2CH3). Anal.
Calcd for TaC9H21Cl3NO2: C, 23.37; H, 4.58; N, 3.03. Found: C,
23.22; H, 4.58; N, 3.07.
Ta(NCMe3)Cl3(py)2 (2a). Method A. Ta(NCMe3)Cl3(py)2 was
synthesized in 72% yield using the procedure for 1a (method A) with
tert-butylamine.
152.6 (pyA-Co), 151.7 (pyB-Co), 140.0 (pyA-Cp), 138.3 (pyB-Cp), 124.6
(pyA-Cm), 124.2 (pyB-Cm), 82.7 (CH(CHMe2)2), 34.0 (CH(CHMe2)2),
20.9 (CH(CHMe′Me)2), 18.8 (CH(CHMe′Me)2). Anal. Calcd for
TaC17H25Cl3N3: C, 36.55; H, 4.51; N, 7.52. Found: C, 36.38; H, 4.80;
N, 7.68.
[H3NCHiPr2]2[Ta(NCHiPr2)Cl5] (3b). TaCl5 (0.500 g, 1.40 mmol)
was dissolved in toluene (8 mL) by adding Et2O (1 mL). H2NCH-
(CHMe2)2 (0.622 mL, 4.19 mmol) was added slowly giving a yellow
solution which was stirred for 4 h. This produced a small amount of
white precipitate. The mixture was evaporated in Vacuo leaving a pale-
yellow powder which was washed with pentane and dried in Vacuo to
give 0.978 g (99%) of the product: 1H NMR (CDCl3) δ 7.06 (bs, 6,
H3NCH(CHMe2)2), 4.65 (t, 1, NCH(CHMe2)2), 2.80 (t, 2, H3-
NCH(CHMe2)2), 2.05 (m, 4, H3NCH(CHMe2)2), 1.91 (m, 2, NCH-
(CHMe2)2), 1.23 (d, 6, NCH(CHMeMe′)2), 1.06 (m, 30, NCH-
(CHMeMe′)2 and H3NCH(CHMe2)2); 13C NMR (CDCl3) δ 83.4
(NCH(CHMe2)2), 64.2 (H3NCH(CHMe2)2), 34.0 (NCH(CHMe2)2), 28.2
(H3NCH(CHMe2)2), 20.7 (NCH(CHMeMe′)2), 19.8 (H3NCH-
(CHMeMe′)2), 18.7 (NCH(CHMeMe′)2), 17.5 (H3NCH(CHMeMe′)2).
Anal. Calcd for TaC21H51N3Cl5: C, 35.83; H, 7.30; N, 5.97. Found:
C, 35.29, H, 6.94, N, 5.99.
Ta(NCHiPr2)Cl3(dme) (3c). Compound 3b (500 mg, 710 mmol)
was dissolved in toluene (8 mL) by adding 74 mL (710 mmol) of dme.
ZnCl2 (0.290 g, 2.13 mmol) was added last. The mixture was stirred
for 18 h during which time the pale-yellow solution became colorless.
The solution was decanted from an oily, sticky solid and evaporated
in Vacuo. This provided a yellow oil which crystallized upon standing
under pentane to give 291 mg (83%) of a white solid: 1H NMR (CDCl3)
δ 4.32 (t, 1, CH(CHMe2)2), 4.19 (s, 3, OMeA, and m, 2, O(CH2)A),
4.02 (m, 2, O(CH2)B), 3.90 (s, 3, OMeB), 1.92 (m, 2, CH(CHMe2)2),
1.25 (d, 6, CH(CHMe′Me)2), 1.07 (d, 6, CH(CHMe′Me)2); 13C NMR
(CDCl3) δ 83.1 (CH(CHMe2)2), 76.0 (O(CH2)A), 70.4 (O(CH2)B), 70.1
(OMeA), 62.3 (OMeB), 34.1 (CH(CHMe2)2), 20.9 (CH(CHMe′Me)2),
18.6 (CH(CHMe′Me)2). Anal. Calcd for TaC11H25Cl3NO2: C, 26.93;
H, 5.14; N, 2.85. Found: C, 26.89; H, 5.20; N, 2.85.
Method C. TaCl5 (500 mg, 1.40 mmol) was dissolved in of CH2Cl2
(10 mL) by adding ∼0.3 mL of pyridine. To this solution was added
of ZnCl2 (380 mg, 2.79 mmol) producing a white precipitate. The
mixture was cooled to -40 °C, and a solution of H2NtBu (150 mL,
1.40 mmol) and NEt3 (0.389 mL, 2.79 mmol) in CH2Cl2 was added
dropwise, dissolving the precipitate. The reaction mixture was stirred
for 18 h and evaporated in Vacuo leaving yellow oil. This was extracted
with toluene/pentane (4:1 by volume), concentrated in Vacuo, and
crystallized from CH2Cl2/Et2O/pentane to give 344 mg (48%) yellow
crystals: 1H NMR (CDCl3) δ 9.06 (d, 2, pyA-Ho), 8.97 (d, 2, pyB-Ho),
7.95 (tt, 1, pyA-Hp), 7.85 (tt, 1, pyB-Hp), 7.44 (m, 4, py-Hm), 1.38 (s, 9,
CMe3); 13C NMR (CDCl3) δ 152.9 (pyA-Co), 151.5 (pyB-Co), 140.1 (pyA-
Cp), 138.5 (pyB-Cp), 124.6 (pyA-Cm), 124.4 (pyB-Cm), 66.3 (CMe3), 31.6
(CMe 3). Anal. Calcd for TaC14H19Cl3N3: C, 32.55; H, 3.71; N, 8.13.
Found: C, 32.53; H, 3.80; N, 7.97.
[H3NCMe3]2[Ta(NCMe3)Cl5] (2b). Attempts to synthesize this
compound using the procedure for 1b gave the yellow-green powder
which was characterized by NMR as a mixture of two products: Ta-
(Nt-Bu)Cl2(HNt-Bu)(H2Nt-Bu)20 and the above compound. Several
washes of this mixture with pentane/Et2O yielded the white solid. NMR
spectra suggest it to be the mixture of the above compound and
t-BuNH3Cl. Crystallization from CH2Cl2/pentane gave crystals which
were characterized by NMR: 1H NMR (CDCl3) δ 7.30 (bs, 6,
H3NCMe3), 1.48 (s, 18, H3NCMe3), 1.36 (s, 9, NCMe3); 13C NMR
(CDCl3) δ 54.0 (H3NCMe3), 31.4 (NCMe3), 27.9 (H3NCMe3); NCMe3,
n/a.
Ta(NCMe3)Cl3(dme) (2c). TaCl5 (1.00 g, 2.79 mmol) was dissolved
in ∼30 mL of toluene/Et2O mixture (8:1 by volume). tert-Butylamine
(880 ml, 8.38 mmol) was added to this solution. It was allowed to stir
for 15 min, during which time was formed a white precipitate. A 290
mL volume of 1,2-dimethoxyethane was added to this mixture followed
by the addition of ZnCl2 (1.52 g, 11.2 mmol). The reaction mixture
was stirred for 18 h, filtered through Celite, and evaporated in Vacuo
leaving a yellowish oil which was treated with three 4 mL portions of
Et2O. The extracts were combined, evaporated in Vacuo, and crystal-
lized under pentane to give 1.06 g (84%) of the white product.
Analytically pure samples were obtained by the CH2Cl2/pentane re-
crystallization: 1H NMR (CDCl3) δ 4.23 (s, 3, OMeA), 4.17 (m, 2,
O(CH2)A), 4.03 (m, 2, O(CH2)B), 3.91 (s, 3, OMeB), 1.34 (s, 9, CMe3);
13C NMR (CDCl3) δ 76.1 (O(CH2)A), 71.2 (O(CH2)B), 70.7 (OMeA),
66.9 (CMe3), 62.2 (OMeB), 31.6 (CMe3). Anal. Calcd for TaC8H19-
NO2Cl3: C, 21.42; H, 4.27; N, 3.12. Found: C, 21.34; H, 4.07; N,
3.27.
Ta(NCHiPr2)Cl3(py)2 (3a). Method A. This compound was
synthesized in 64% yield using the procedure for 1a (method A), except
H2NCH(CHMe2)2 was used instead of tert-amylamine and the product
was purified using hot ether extraction.
Method B. This compound was synthesized in 40% yield using
the procedure for 1a (method B) with H2NCH(CHMe2)2.
Method D. This compound was synthesized in 44% yield using
the procedure for 1a (method B) with H2NCH(CHMe2)2 and Na2B4O7.
Method E. TaCl5 (500 mg, 1.40 mmol) was dissolved in of CH2Cl2
(8 mL) by adding ∼0.3 mL of pyridine. H2NCH(CHMe2)2 (207 mL,
1.40 mmol) was added slowly to give a yellow solution. CaH2 (58
mg, 1.40 mmol) was added, and the reaction mixture was stirred for 5
h, filtered through Celite, and chilled to -40 °C. A white precipitate
was filtered off, and the mother liquor was concentrated in Vacuo to
give a reddish oil and a white precipitate. The oil was dissolved in
Et2O, layered with pentane, and chilled to -40 °C to give 388 mg
(50%) of red-yellow crystals: 1H NMR (CDCl3) δ 9.01 (d, 2, pyA-Ho),
8.89 (d, 2, pyB-Ho), 7.93 (tt, 1, pyA-Hp), 7.83 (tt, 1, pyB-Hp), 7.41 (m,
4, py-Hm), 4.56 (t, 1, CHi-Pr2), 1.96 (m, 2, CH(CHMe2)2), 1.20 (d, 6,
CH(CHMe′Me)2), 1.03 (d, 6, CH(CHMe′Me)2); 13C NMR (CDCl3) δ
Ta(N-1-adamantyl)Cl3(py)2 (4a). Method A. This compound was
synthesized in 66% yield using the procedure for 1a (method A) with
adamantylamine.
Method F. This compound was synthesized in 30% yield using
the procedure for 1a (method B) with adamantylamine and K2CO3:
1H NMR (CDCl3) δ 9.04 (d, 2, pyA-Ho), 8.97 (d, 2, pyB-Ho), 7.93 (tt,
1, pyA-Hp), 7.86 (tt, 1, pyB-Hp), 7.43 (m, 4, py-Hm), 2.13 (bs, 3,
C(CH2CHCH2)3), 1.96 (d, 6, C(CH2CHCH2)3), 1.60 (m, 6, C(CH2-
CHCH2)3); 13C NMR (CDCl3) δ 153.0 (pyA-Co), 151.4 (pyB-Co), 139.9
(pyA-Cp), 138.6 (pyB-Cp), 124.5 (pyA-Cm), 124.3 (pyB-Cm), 67.1 (C(CH2-
CHCH2)3), 44.7 (C(CH2CHCH2)3), 36.2 (C(CH2CHCH2)3), 29.5
(C(CH2.CH.CH2)3). Anal. Calcd for TaC20H25N3Cl3: C, 40.39; H,
4.24; N, 7.07. Found: C, 40.95; H, 4.38; N, 6.69.
[H3N-1-adamantyl]2[Ta(N-1-adamantyl)Cl5] (4b). This compound
was synthesized in quantitative yield using the procedure for 1b with
1-adamantylamine. We were not able to assign both 1H and 13C NMR
spectra, partially because of low solubility in NMR solvents (CDCl3,
CD2Cl2, C6D6). However, it was successfully used as a starting material
in the further synthesis of Ta(N-1-adamantyl)Cl3(dme).
Ta(N-1-adamantyl)Cl3(dme) (4c). This compound was synthesized
in 78% yield using the procedure for 1c with [H3N-1-adamantyl]2-
[Ta(N-1-adamantyl)Cl5]: 1H NMR (C6D6) δ 3.48 (s, 3, OMeA), 3.43
(s, 3, OMeB), 3.03 (bm, 4, OCH2), 2.00 (d, 6, C(CH2CHCH2)3), 1.94
(bs, 3, C(CH2CHCH2)3), 1.42 (m, 6, C(CH2CHCH2)3); 13C NMR (C6D6)
δ 75.4 (O(C H2)A), 70.7 (O(C H2)B), 70.2 (OMeA), 67.7 (C (CH2-
CHCH2)3), 61.7 (OMeB), 45.4 (C(CH2CHCH2)3), 36.4 (C(C H2-
CHCH2)3), 29.9 (C(CH2CHCH2)3). Anal. Calcd for TaC14H25NO2-
Cl3: C, 31.93; H, 4.78; N, 2.66. Found: C, 31.41; H, 4.82; N, 2.62.
Ta(NCMe2CH2CMe3)Cl3(py)2 (5a). This compound was synthe-
sized by the ligand exchange reaction of 5c with pyridine and
characterized by NMR: 1H NMR (CDCl3) δ 9.00 (tt, 4, py-Ho), 7.93
(tt, 1, pyA-Hp), 7.83 (tt, 1, pyB-Hp), 7.41 (m, 4, py-Hm), 1.69 (s, 2,
CMe2CH2CMe3), 1.47 (s, 6, CMe2CH2CMe3), 1.05 (s, 9, CMe2-
CH2CMe3); 13C NMR (CDCl3) δ 153.0 (pyA-Co), 151.5 (pyB-Co), 139.9
(pyA-Cp), 138.3 (pyB-Cp), 124.5 (pyA-Cm), 124.2 (pyB-Cm), 70.3 (CMe2-
CH2CMe3), 56.1 (CMe2CH2CMe3), 31.7 (CMe2CH2CMe3 and CMe2-

2654 Inorganic Chemistry, Vol. 36, No. 12, 1997
Korolev et al.
CH2CMe3), 31.3 (CMe2CH2CMe3). Anal. Calcd for TaC18H27N3Cl3:
C, 37.75; H, 4.75; N, 7.34. Found: C, 37.65; H, 4.80; N, 7.35.
[H3NCMe2CH2CMe3]2[Ta(NCMe2CH2CMe3)Cl5] (5b). This com-
pound was synthesized in 96% yield using the procedure for 1b with
tert-octylamine: 1H NMR (CDCl3) δ 7.10 (bs, 6, H3NCMe2CH2CMe3),
1.75 (s, 4, H3NCMe2CH2CMe3), 1.59 (s, 2, NCMe2CH2CMe3), 1.52
(s, 12, H3NCMe2CH2CMe3), 1.47 (s, 6, NCMe2CH2CMe3), 1.09 (s, 9,
NCMe2CH2CMe3), 1.05 (s, 18, H3NCMe2CH2CMe3); 13C NMR (CDCl3)
δ 71.1 (NCMe2CH2CMe3), 58.2 (H3NCMe2CH2CMe3), 55.9 (NCMe2CH2-
CMe3), 52.5 (H3NCMeCH2CMe3), 32.1 (NCMe2CH2CMe3), 31.8
(NCMe2CH2CMe3), 31.3 (H3NCMe2CH2CMe3), 27.4 (H3NCMe2-
CH2CMe3); NCMe2CH2CMe3 and H3NCMe2CH2CMe3, n/a.
Ta(NCMe2CH2CMe3)Cl3(dme) (5c). This compound was synthe-
sized in 75% yield using the procedure for 1c with [H3NCMe2CH2-
CMe3]2[Ta(NCMe2CH2CMe3)Cl5]. 1H NMR (CDCl3) δ 4.22 (s, 3,
OMeA), 4.17 (m, 2, O(CH2)A), 4.02 (m, 2, O(CH2)B), 3.90 (s, 3, OMeB),
1.64 (s, 2, CMe2CH2CMe3), 1.45 (s, 6, CMe2CH2CMe3), 1.06 (s, 9,
CMe2CH2CMe3); 13C NMR (CDCl3) δ 76.1 (O(CH2)A), 71.0 (O(CH2)B),
70.9 (CMeCH2CMe3), 70.6 (OMeA), 62.2 (OMeB), 56.2 (CMe2CH2-
CMe3), 31.8 (CMe2CH2CMe3), 31.7 (CMe2CH2CMe3), 31.3 (CMe2-
CH2CMe3).
[H3N(CH2)3CH3]2[Ta(N(CH2)3CH3)Cl5] (6b). This compound was
synthesized in quantitative yield using the procedure for 1b with n-
butylamine: 1H NMR (CDCl3) δ 7.65 (bs, 6, H3N(CH2)3CH3), 5.16 (t,
2, NCH2(CH2)2CH3), 3.04 (m, 4, H3NCH2(CH2)2CMe3), 1.74 (m, 6,
H3NCH2CH2CH2CH3 and NCH2CH2CH2CH3), 1.45 (m, 6, H3NCH2-
CH2CH2CH3 and NCH2CH2CH2CH3), 0.95 (m, 9, H3NCH2CH2CH3 and
NCH2CH2CH3); 13C NMR (CDCl3) δ 61.0 (NCH2(CH2)2CH3), 40.3
(H3NCH2(CH2)2CH3), 34.8 (NCH2CH2CH2CH3), 29.4 (H3NCH2CH2-
CH2CH3), 20.4 (NCH2CH2CH2CH3), 19.8 (H3NCH2CH2CH2CH3), 13.8
(NCH2CH2CH2CH3), 13.4 (H3NCH2CH2CH2CH3).
Ta(N(CH2)3CH3)Cl3(dme) (6c). This compound was synthesized
in 87% yield using the procedure for 1c with [H3N(CH2)3CH3]2-
[Ta(N(CH2)3CH3)Cl5]: 1H NMR (CDCl3) δ 4.88 (t, 2, NCH2(CH2)2-
CH3), 4.18 (s, 3, OMeA), 4.16 (m, 2, O(CH2)A), 4.03 (m, 2, O(CH2)B),
3.91 (s, 3, OMeB), 1.58 (m, 4, NCH2CH2CH2CH3), 0.93 (t, 3, NCH2-
CH2CH3); 13C NMR (CDCl3) δ 75.9 (O(CH2)A), 70.7(O(CH2)B), 70.3
(OMeA), 62.4 (OMeB), 61.3 (NCH2(CH2)2CH3), 34.7 (NCH2CH2CH2-
CH3), 20.3 (NCH2CH2CH2CH3), 13.7 (NCH2CH2CH2CH3).
Ta(NC6H5)Cl3(dme) (7). TaCl5 (500 mg, 1.40 mmol) was dissolved
in dichloromethane (10 mL) by adding ∼0.25 mL of dimethoxyethane.
ZnCl2 (380 mg, 2.79 mmol) was added to the solution, forming a white
precipitate. This was cooled to -40 °C, and a solution of H2NPh (130
mg, 1.40 mmol) and pyridine (226 mL, 2.79 mmol) in CH2Cl2 was
added slowly. This dissolved the precipitate and produced a yellow
solution. The mixture was stirred for 17 h. A precipitate formed and
was filtered off. The mother liquor was concentrated in Vacuo to a
yellow solid. This was recrystallized from CH2Cl2/pentane to give 610
mg (93%) yellow crystals: 1H NMR (CDCl3) δ 7.35 (t, 2, Ph-Hm),
7.07 (d, 2, Ph-Ho), 6.89 (t, 1, Ph-Hp), 4.20 (m, 2, O(CH2)A), 4.13 (s, 3,
OMeA), 4.11 (m, 2, O(CH2)B), 4.02 (s, 3, OMeB); 13C NMR (CDCl3) δ
153.2 (Ph-Ci), 127.9 (Ph-Cm), 126.7 (Ph-Co), 126.0 (Ph-Cp), 76.0
(O(CH2)A), 71.2 (O(CH2)B), 69.9 (OMeA), 63.2 (OMeB). Anal. Calcd
for TaC10H15NCl3O2: C, 25.63; H, 3.23; N, 2.99. Found: C, 25.29;
H, 3.21; N, 3.01.
Ta(N-2,6-i-Pr2C6H3)Cl3(dme) (8). This compound was synthesized
in 99% yield using the procedure for 7 with 2,6-diisopropylphenyl-
amine: 1H NMR (CDCl3) δ 7.18 (d, 2, Ar-Hm), 6.83 (t, 1, Ar-Hp),
4.32 (m, 2, CHMe2), 4.23 (m, 2, O(CH2)A), 4.12 (m, 2, O(CH2)B), 4.10
(s, 3, OMeA), 4.02 (s, 3, OMeB), 1.31 (d, 12, CHMe2); 13C NMR (CDCl3)
δ Ar-Ci was not observed, 148.7 (Ar-Cm), 126.0 (Ar-Co), 121.9 (Ar-
Cp), 76.0 (OCH2)A), 70.8 (O(CH2)B), 69.9 (OMeA), 62.9 (OMeB), 27.4
(CHMe2), 24.7 (CHMe2). This compound has been reported previ-
ously.6
Ta(N-4-CNC6H4)Cl3(dme) (10). This compound was synthesized
in 86% yield using the procedure for 7 with 4-aminobenzonitrile: 1H
NMR (CDCl3) δ 7.64 (d, 2, Ar-Hm), 7.14 (d, 2, Ar-Ho), 4.24 (m, 2,
O(CH2)A), 4.17 (m, 2, O(CH2)B), 4.14 (s, 3, OMeA), 4.09 (s, 3, OMeB);
13C NMR (CDCl3) δ Ci, Cp, CN, n/a, 132.3 (Ar-Cm), 127.4 (Ar-Co),
76.1 (O(C H2)A), 71.3 (O(C H2)B), 69.9 (OMeA), 63.6 (OMeB). Anal.
Calcd for TaC11H14Cl3N2O2: C, 26.77; H, 2.86; N, 5.68. Found: C,
26.25; H, 2.86; N, 5.51.
Ta(N-2-naphthyl)Cl3(dme) (11). This compound was synthesized
in 90% yield using the procedure for 7 with 2-aminonaphthalene. 1H
NMR (CDCl3) δ 7.81 (d, 1, Ar-H4), 7.77 (d, 1, Ar-H5), 7.69 (d, 1,
Ar-H3), 7.46 (m, 2, Ar-H7, H1), 7.31 (m, 2, Ar-H6,H8), 4.22 (m, 2,
O(CH2)A), 4.17 (s, 3, OMeA), 4.15 (m, 2, O(CH2)B), 4.07 (s, 3, OMeB);
13C NMR (CDCl3) δ 150.7 (Ar-C2), 132.5 (Ar-C10), 131.7 (Ar-C9), 128.0
(Ar-C4), 127.9 (Ar-C5), 127.2 (Ar-C3), 126.0 (Ar-C7), 125.9 (Ar-C1),
125.7 (Ar-C6), 124.7 (Ar-C8), 75.9 (O(C H2)A), 71.1 (O(C H2)B), 69.8
(OMeA), 63.2 (OMeB). Anal. Calcd for TaC14H17NO2Cl3: C, 32.42;
H, 3.30; N, 2.70. Found: C, 31.77; H, 3.12; N, 3.17.
Ta(N-2-anthryl)Cl3(dme) (12). This compound was synthesized
in 96% yield using the procedure for 7 with 2-aminoanthracene: 1H
NMR (CDCl3) δ 8.36 (s, 1, Ar-H10), 8.21 (s, 1, Ar-H1), 7.98 (m, 2,
Ar-H4, H5), 7.92 (d, 1, Ar-H3), 7.53 (s, 1, Ar-H9), 7.47 (m, 1, Ar-H7),
7.37 (m, 2, Ar-H6,H8). Anal. Calcd for TaC18H19NO2Cl3: C, 38.02;
H, 3.37; N, 2.46. Found: C, 36.90; H, 3.35; N, 2.45.
Nb(NCMe2Et)Cl3(py)2 (13a). This compound was synthesized in
19% yield using the procedure for 1a (method A) with NbCl5. 1H
NMR (CDCl3) δ 9.02 (dt, 2, pyA-Ho), 8.86 (dt, 2, pyB-Ho), 7.89 (tt, 1,
pyA-Hp), 7.82 (tt, 1, pyB-Hp), 7.39 (m, 4, py-Hm), 1.74 (q, 2, CMe2CH2-
Me), 1.40 (s, 6, CMe2CH2Me), 1.13 (t, 3, CMe2CH2Me); 13C NMR
(CDCl3) δ 152.7 (pyA-Co), 151.2 (pyB-Co), 139.7 (pyA-Cp), 138.2 (pyB-
Cp), 124.3 (pyA-Cm), 124.1 (pyB-Cm), MeCH2Me2C, n/a, 35.9 (CMe2C
H2Me), 26.9 (CMe2CH2Me), 9.2 (CMe2CH2 Me). Anal. Calcd for
NbC15H21N3Cl3: C, 40.70; H, 4.78; N, 9.49. Found: C, 39.55; H, 4.86;
N, 9.13.
[H3NCMe2Et]2[Nb(NCMe2Et)Cl5] (13b). This compound was
synthesized in quantitative yield using the procedure for 1b with
NbCl5: 1H NMR (CDCl3) δ 6.53 (bs, 6, H3NCMe2Et), 1.77 (q, 4, H3-
NCMe2CH2CH3), 1.68 (q, 2, NCMe2CH2CH3), 1.43 (s, 12, H3NCMe2-
CH2CH3), 1.38 (s, 6, NCMe2CH2CH3), 1.11 (t, 3, NCMe2CH2CH3), 1.00
(t, 6, H3NCMe2CH2CH3); 13C NMR (CDCl3) δ NCMe2CH2CH3, n/a,
56.6 (H3NCMe2CH2CH3), 35.6 (NCMe2CH2CH3), 33.3 (H3NCMe2CH2-
CH3), 26.7 (NCMe2CH2CH3), 25.0 (H3NCMe 2CH2CH3), 9.1 (NCMe2-
CH2CH3), 8.1 (H3NCMe2CH2CH3). Anal. Calcd for C15H39N3Cl5Nb:
C, 33.89; H, 7.39; N, 7.90. Found: C, 33.87; H, 7.41; N, 7.89.
Nb(NCMe2Et)Cl3(dme) (13c). NbCl5 (500 mg, 1.85 mmol) was
dissolved in 10 mL of toluene by adding dimethoxyethane (192 mL,
1.85 mmol). To this yellow colored solution was slowly added tert-
amylamine (649 mL, 5.55 mmol). The solution became foggy in 3
min. An excess of ZnCl2 (1.01 g, 7.40 mmol) was added, and the
reaction mixture was stirred for 4 h. It was then chilled to -40 °C for
3 h, filtered through Celite, and concentrated in Vacuo. The resulting
powder was washed with pentane and dried in Vacuo to give 651 mg
(94%) of a yellow solid: 1H NMR (CDCl3) δ 4.08 (m, 2, O(CH2)A),
4.05 (s, 3, OMeA), 4.02 (m, 2, O(CH2)B), 3.87 (s, 3, OMeB), 1.69 (q, 2,
CMe2CH2CH3), 1.37 (s, 6, CMe2CH2CH3), 1.14 (CMe2CH2CH3); 13C
NMR (CDCl3) δ 75.5 (O(CH2)A), 70.4 (O(CH2)B), 69.6 (OMeA), 61.9
(OMeB), CMe2CH2CH3 not observed, 36.0 (C(Me)2CH2CH3), 27.1
(CMe2CH2CH3), 9.1 (CMe2CH2CH3). Anal. Calcd for NbC9H21NO2-
Cl3: C, 28.86; H, 5.65; N, 3.74. Found: C, 26.83; H, 5.54; N, 4.04.
Nb(NCMe3)Cl3(dme) (14). NbCl5 (1.00 g, 3.70 mmol) was
dissolved in dichloromethane (30 mL) by adding dimethoxyethane (384
mL, 3.70 mmol). ZnCl2 (1.01 g, 7.40 mmol) was added to the solution,
forming a white precipitate. This was cooled to -40 °C, and a solution
of tert-butylamine (389 mL, 3.70 mmol) and triethylamine (1.03 mL,
7.40 mmol) in CH2Cl2 (10 mL) was slowly added. The precipitate
dissolved, and the yellow solution was stirred for 18 h, during which
time it darkened. It was concentrated in Vacuo to a dark green oil,
which was extracted with toluene, filtered through Celite, and concen-
trated in Vacuo. The resulting yellow oil crystallized after standing
with pentane to give 933 mg (70%) of a yellow solid. 1H NMR (CDCl3)
δ 4.08 (m, 2, O(CH2)A), 4.06 (s, 3, OMeA), 4.02 (m, 2, O(CH2)B), 3.89
Ta(N-2,6-Me2C6H3)Cl3(dme) (9). This compound was synthesized
in 68% yield using the procedure for 7 with 2,6-dimethylaniline: 1H
NMR (CDCl3) δ 7.03 (d, 2, Ar-Hm), 6.63 (t, 1, Ar-Hp), 4.22 (m, 2,
O(CH2)A), 4.12 (s, 3, OMeA), 4.11 (m, 2, O(CH2)B), 4.01 (s, 3, OMeB),
2.82 (s, 6, Ar-Me); 13C NMR (CDCl3) δ Ci n/a, 138.5 (Ar-Co), 126.8
(Ar-Cm), 125.4 (Ar-Cp), 76.0 (O(CH2)A), 70.9 (O(CH2)B), 70.2 (OMeA),
62.9 (OMeB), 19.1 (Ar-CH3). Anal. Calcd for TaC12H19Cl3NO2: C,
29.02; H, 3.86; N, 2.82. Found: C, 28.83; H, 3.81; N, 2.96.

Labile Niobium and Tantalum d0 Monoimides
Inorganic Chemistry, Vol. 36, No. 12, 1997 2655
(s, 3, OMeB), 1.42 (CMe3); 13C NMR (CDCl3) δ 75.4 (OCH2)A), 70.5
(OCH2)B), 69.7 (OMeA), 61.9 (OMeB) CMe3, n/a, 29.4 (CMe3).
Nb(N-1-adamantyl)Cl3(py)2 (15a). Compound 15c (250 mg, 57.0
mmol) was dissolved in CH2Cl2 (5 mL). Pyridine was added (∼0.3
mL) and the solution stirred for 3 h. It was concentrated in Vacuo to
give a yellow oil which crystallized upon standing under pentane to a
yellow solid quantitative yield: 1H NMR (CDCl3) δ 9.04 (d, 2, pyA-
Ho), 8.88 (d, 2, pyB-Ho), 7.87 (m, 2, py-Hp), 7.41 (m, 4, py-Hm), 2.13
(s, 9, C(CH2CHCH2)3 and C(CH2CHCH2)3), 1.63 (m, 6, C(CH2-
CHCH2)3); 13C NMR (CDCl3) δ 152.8 (pyA-Co), 151.1 (pyB-Co), 139.7
(pyA-Cp), 138.3 (pyB-Cp), 124.3 (pyA-Cm), 124.2 (pyB-Cm), C(CH2-
CHCH2)3 not observed, 42.2 (C(CH2CHCH2)3), 36.0 (C(CH2CHCH2)3),
29.1 (C(CH2CHCH2)3). Anal. Calcd for NbC20H25N3Cl3: C, 47.39;
H, 4.98; N, 8.29. Found: C, 46.85; H, 5.18; N, 8.12. This compound
consistently analyzed low in carbon.
Nb(N-2-naphthyl)Cl3(dme) (19). This compound was synthesized
in quantitative yield using the procedure for 11 with NbCl5: 1H NMR
(CDCl3) δ 7.78 (m, 4, Ar-H4,H5,H3, H1), 7.54 (dd, 1, Ar-H8), 7.49 (m,
1, Ar-H7), 7.40 (m, 1, Ar-H6), 4.13 (m, 4, OCH2), 4.03 (s, 3, OMeA),
4.01 (s, 3, OMeB); 13C NMR (CDCl3) δ Ar-C2, n/a, 132.6 (Ar-C10),
132.2 (Ar-C9), 128.5 (Ar-C4), 128.2 (Ar-C5), 127.6 (Ar-C3), 126.73
(Ar-C7), 126.66 (Ar-C1), 124.0 (Ar-C6), 123.2 (Ar-C8), 75.3 (O(CH2)A),
70.9 (O(CH2)B), 68.6 (OMeA), 62.8 (OMeB). Anal. Calcd for
NbC14H17NO2Cl3: C, 39.05; H, 3.98; N, 3.25. Found: C, 38.87; H,
3.94; N, 3.21.
Nb(N-2-anthryl)Cl3(dme) (20). This compound was synthesized
in quantitative yield using the procedure for 12 with NbCl5: 1H NMR
(CDCl3) δ 8.36 (s, 1, Ar-H10), 8.33 (s, 1, Ar-H1), 7.96 (m, 4,
Ar-H4,H5,H3,H9), 7.47 (m, 1, Ar-H7), 7.37 (m, 2, Ar-H6,H8), 4.16 (s, 4,
OCH2), 4.06 (s, 6, OMe).
Nb(N-1-adamantyl)Cl3(dme) (15c). NbCl5 (500 mg, 1.85 mmol)
was dissolved in CH2Cl2 (10 mL) by adding ∼0.3 mL of dimethoxy-
ethane. ZnCl2 (504 mg, 3.70 mmol) was added, and a white precipitate
formed. This was cooled to -40 °C, and a solution of adamantylamine
(280 mg, 1.85 mmol) and triethylamine (516 mL, 3.70 mmol) in CH2-
Cl2 (5 mL) was slowly added. The reaction mixture was stirred for 16
h, during which time the precipitate dissolved and a yellow-brown
solution formed. This was evaporated in Vacuo to give a yellow solid
and a black oil which was treated two times with 5 mL of toluene/
CH2Cl2 (4:1 by volume) in order to extract the product. The extracts
were combined, concentrated in Vacuo, and recrystallized from CH2-
Cl2/pentane to give 578 mg (71%) of a yellow solid: 1H NMR (CDCl3)
δ 4.09 (s, 3, OMeA), 4.07 (m, 2, O(CH2)A), 4.01 (m, 2, O(CH2)B), 3.88
(s, 3, OMeB), 2.09 (bs, 3, C(CH2CHCH2)3), 2.06 (s, 6, C(CH2CHCH2)3),
1.61 (m, 6, C(CH2CHCH2)3); 13C NMR (CDCl3) δ 75.4 (O(CH2)A),
70.4 (O(CH2)B), 70.1 (OMeA), 61.8 (OMeB), C(CH2CHCH2)3 not
observed, 42.5 (C(CH2CHCH2)3), 35.8 (C(CH2CHCH2)3), 29.1 (C(CH2-
CHCH2)3). Anal. Calcd for NbC14H25NCl3O2: C, 38.33; H, 5.75; N,
3.19. Found: C, 37.34; H, 5.66; N, 3.19. This compound consistently
analyzed low in carbon.
Crystallographic Structure Determination. Crystallographic data
are collected in Table 3. Yellow-green rod-shaped crystals were
photographically characterized and determined to belong to the mono-
clinic crystal system. Systematic absences in the diffraction data
indicated that space group was either C2/c or Cc. The noncentrosym-
metric alternative was ultimately chosen on the basis of the absence of
either inversional or 2-fold rotational relationships between molecules
in the unit cell. The asymmetric unit consists of two uncorrelated
chemically identical molecules. An empirical correction for absorption
was applied to the diffraction data (Tmax/min ) 1.11). The structure was
solved by direct methods, completed from difference Fourier maps,
and refined with anisotropic thermal parameters for all non-hydrogen
atoms except for carbon to conserve data. Hydrogen atoms were
idealized. All computations used SHELXTL 4.2 software (G. Sheldrick,
Siemens XRD, Madison, WI).
Spartan Calculations. The molecules were constructed in the
builder application of Spartan 4.0 on a Silicon Graphics Indigo
workstation. The molecules were then geometry-optimized using the
PM3(TM) parameter set. An MO calculation was then performed (not
necessary), followed by a frequency calculation to obtain the predicted
normal vibrational modes. These were tabulated and examined with
the animated modes using the Spartan graphical interface.
Nb(NC6H5)Cl3(dme) (16). This compound was synthesized in
quantitative yield using the procedure for 7 with NbCl5. 1H NMR
(CDCl3) δ 7.35 (m, 4, Ph-Ho,Hm), 7.11 (tt, 1, Ph-Hp), 4.12 (m, 4, OCH2),
4.02 (s, 3, OMeA), 4.00 (s, 3, OMeB); 13C NMR (CDCl3) δ Ph-Ci not
observed, 128.4 (Ph-Co), 127.4 (Ph-Cp), 125.2 (Ph-Cm), 75.3 (O(C H2)A),
70.9 (O(C H2)B), 68.6 (OMeA), 62.8 (OMeB).
Acknowledgment. Support for this work was generously
provided by the National Science Foundation under Grant No.
CHE-9632543, the donors of the Petroleum Research Fund
(Grant No. 31361-G3), administered by the American Chemical
Society, and Wayne State University. D.S.W. thanks Professor
B. Schlegel for allowing our use of Spartan and for helpful
discussions.
Nb(N-2,6-i-Pr2C6H3)Cl3(dme) (17). This compound was synthe-
sized in quantitative yield using the procedure for 8 with NbCl5: 1H
NMR (CDCl3) δ 7.11 (d, 2, Ar-Hm), 7.00 (t, 1, Ar-Hp), 4.40 (m, 2, CH
Me2), 4.16 (m, 2, O(CH2)A), 4.10 (m, 2, O(CH2)B), 3.97 (s, 3, OMeA),
3.96 (s, 3, OMeB), 1.32 (d, 12, CHMe2); 13C NMR (CDCl3) δ Ar-Ci
not observed, 148.6 (Ar-Cm), 127.3 (Ar-Co), 122.7 (Ar-Cp), 75.6
(OCH2)A), 70.6 (O(CH2)B), 68.7 (OMeA), 62.5 (OMeB), 27.9 (CHMe2),
24.7 (CHMe2). This compound has been reported previously.12
Nb(N-4-CNC6H4)Cl3(dme) (18). This compound was synthesized
in 75% yield using the procedure for 10 with NbCl5: 1H NMR (CDCl3)
δ 7.63 (d, 2, Ar-Hm), 7.43 (d, 2, Ar-Ho), 4.15 (bs, 4, OCH2), 4.04 (s,
3, OMeA), 3.97 (s, 3, OMeB); 13C NMR (CDCl3) δ 155.9 (Ar-Ci), 132.7
(Ar-Cm), 125.8 (Ar-Co), 118.2 (Ar-C N), 110.0 (Ar-Cp), 75.4 (O(CH2)A),
71.1 (O(CH2)B), 68.7 (OMeA), 63.2 (OMeB). Anal. Calcd for
NbC11H14N2O2Cl3: C, 32.58; H, 3.48; N, 6.91. Found: C, 31.66; H,
3.33; N, 6.82.
Supporting Information Available: Tables of output files for the
frequency calculations of Ta(N)Cl3, Ta(NH)Cl3, Ta(NMe)Cl3,
Ta(NCMe3)Cl3, Ta(NPh)Cl3, and Ta(NCMe3)Cl3(dme) (40 pages). An
X-ray crystallographic file in CIF format for complex 14 is available
on the Internet only. Ordering and access information is given on any
current masthead page.
IC961360J