Investigation of transition metal-imido bonding in M(NBut) 2(dpma)

Article abstract of DOI:10.1021/ic049644o

A complete series down group 6 of the formula M(NBu^t) ₂(dpma) has been synthesized, where dpma is N,N-di-(pyrrolyl-α -methyl)-N-methylamine. A fourth complex, Mo(NAr)₂(dpma) (4), was also prepared, where Ar is 2,6-diisopropylphenyl. All four of these complexes display geometries in the solid state best described as square pyramidal with one imido ligand occupying the axial position and the other an equatorial site. In all cases, the axial imido ligand has a significantly smaller M-N(imido)-C bond angle with respect to the equatorial multiple-bond substituent. From the ¹H, ¹³C, and ¹⁴N NMR spectra, the axial (bent) imido appears to be more electron-rich than the equatorial and linear imido, with the differences becoming less pronounced down the column. The angular deformation energies for the axial imido ligands were studied by DFT in order to discern if and to what extent imido bond angles were important energetically. The electronic energies associated with straightening the axial imido ligand, while holding the remainder of the molecule at the ground-state geometry, for the Cr, Mo, and W derivatives were calculated as 4.5, 2.7, and 2.0 kcal/mol, respectively. A straight-line plot is found for deformation energies versus estimated electronegativity of the group 6 metals in the +6 oxidation state. The study suggests that the electronic differences between metal imido ligands of different angles are quite small; however, the effects may be more pronounced for metal centers with higher electronegativity, e.g. Cr(VI) with electron-withdrawing ligands.

Full text of DOI:10.1021/ic049644o

Inorg. Chem. 2004, 43, 3605−3617
Investigation of Transition Metal−Imido Bonding in M(NBut)2(dpma)
James T. Ciszewski, James F. Harrison, and Aaron L. Odom*
Department of Chemistry, Michigan State UniVersity, East Lansing, Michigan 48824
Received March 17, 2004
t
A complete series down group 6 of the formula M(NBu)2(dpma) has been synthesized, where dpma is N,N-di-
(pyrrolyl-R-methyl)-N-methylamine. A fourth complex, Mo(NAr)2(dpma) (4), was also prepared, where Ar is 2,6-
diisopropylphenyl. All four of these complexes display geometries in the solid state best described as square
pyramidal with one imido ligand occupying the axial position and the other an equatorial site. In all cases, the axial
imido ligand has a significantly smaller M−N(imido)−C bond angle with respect to the equatorial multiple-bond
substituent. From the 1H, 13C, and 14N NMR spectra, the axial (bent) imido appears to be more electron-rich than
the equatorial and linear imido, with the differences becoming less pronounced down the column. The angular
deformation energies for the axial imido ligands were studied by DFT in order to discern if and to what extent imido
bond angles were important energetically. The electronic energies associated with straightening the axial imido
ligand, while holding the remainder of the molecule at the ground-state geometry, for the Cr, Mo, and W derivatives
were calculated as 4.5, 2.7, and 2.0 kcal/mol, respectively. A straight-line plot is found for deformation energies
versus estimated electronegativity of the group 6 metals in the +6 oxidation state. The study suggests that the
electronic differences between metal imido ligands of different angles are quite small; however, the effects may be
more pronounced for metal centers with higher electronegativity, e.g. Cr(VI) with electron-withdrawing ligands.
Introduction
and offer a rich chemistry.5 In the solid state, many group 6
bis(imido) complexes possess nonequivalent imido substit-
uents where one imido is significantly bent with respect to
the other (vide infra). The classic example of a bent imido
is found in Mo(NPh)2(S2CNEt2)2,6 which has two different
In recent years, a resurgence of interest in transition
metal-imido complexes has been sparked by new catalytic
reactions where NAr is present as a supporting ligand,1 highly
reactive early metal imido compounds that participate in
C-H activation,2 catalytic formation of C-N bonds with
titanium imido intermediates,3 and other processes.4 Bis-
(imido) complexes of group 6 have been extensively explored
(3) For examples, see: (a) Cao, C.; Ciszewski, J. T.; Odom, A.
L.Organometallics 2001, 20, 5011-5013; Organometallics 2002, 21,
5148. (b) Shi, Y.; Ciszewski, J. T.; Odom, A. L. Organometallics
2001, 20, 3967-3969; Organometallics 2002, 21, 5148. (c) Hill, J.
E.; Profilet, R. D.; Fanwick, P. E.; Rothwell, I. P. Angew. Chem., Int.
Ed. Engl. 1990, 29, 664. (d) Tillack, A.; Castro, I. G.; Hartung, C.
G.; Beller, M. Angew. Chem., Int. Ed. 2002, 41, 2541. (e) Pohlki, F.;
Heutling, A.; Bytschkov, I.; Hotopp, T.; Doye, S. Synlett 2002, 799.
(f) Haak, E.; Bytschkov, I.; Doye, S. Angew. Chem., Int. Ed. 1999,
38, 3389. (g) Johnson, J. S.; Bergman, R. G. J. Am. Chem. Soc. 2001,
123, 2923. (h) Ackermann, L.; Bergman, R. G. Org. Lett. 2002, 4,
1475. (i) Ong, T.-G.; Yap, G. P. A.; Richeson, D. S. J. Am. Chem.
Soc. 2003, 125, 8100-8101. (j) Ackermann, L.; Bergman, R. G.; Loy,
R. N. J. Am. Chem. Soc. 2003, 125, 11956-11963. (k) Cao, C.; Shi,
Y.; Odom, A. L. J. Am. Chem. Soc. 2003, 125, 2880-2881.
(4) For a few examples, see: (a) Bruno, J. W.; Li, X. J. Organometallics
2000, 19, 4672-4674. (b) Zuckerman, R. L.; Bergman, R. G.
Organometallics 2000, 19, 4795-4809. (c) Bashall, A.; Collier, P.
E.; Gade L. H.; McPartlin, M.; Mountford P.; Pugh, S. M.; Radojevic,
S.; Schubart, M.; Scowen, I. J.; Trosch, D. J. M. Organometallics
2000, 19, 4784-4794. (d) McInnes, J. M.; Mountford, P. J. Chem.
Soc., Chem. Commun. 1998, 1669-1670. (e) Cantrell, G. K.; Meyer,
T. Y. J. Am. Chem. Soc. 1998, 120, 8035-8042.
* Author to whom correspondence should be addressed. E-mail: odom@
cem.msu.edu.
(1) (a) Buchmeiser, M. R. Chem. ReV. 2000, 100, 1565-1604. (b) Phillips,
A. J.; Abell, A. D. Aldrichimica Acta 1999, 32, 75-89. (c) Schrock,
R. R. Tetrahedron 1999, 55, 8141-8153. (d) Schrock, R. R. Olefin
Metathesis by Well-Defined Complexes of Molybdenum and Tungsten.
In Topics in Organometallic Chemistry; Fu¨rstner, A., Ed.; Springer:
Berlin, 1998; Vol. 1. (e) Jamieson, J. Y.; Schrock, R. R.; Davis, W.
M.; Bonitatebus, P. J.; Zhu, S. S.; Hoveyda, A. H. Organometallics
2000, 19, 925-930. (f) Adams, J. A.; Ford, J. G.; Stamatos, P. J.;
Hoveyda, A. H. J. Org. Chem. 1999, 64, 9690-9696. (g) Nugent, W.
A.; Feldman, J.; Calabrese, J. C. J. Am. Chem. Soc. 1995, 117, 8992-
8998. (h) La, D. S.; Alexander, J. B.; Cefalo, D. R.; Graf, D. D.;
Hoveyda, A. H.; Schrock, R. R. J. Am. Chem. Soc. 1998, 120, 9720-
9721. (i) Alexander, J. B.; La, D. S.; Cefalo, D. R.; Hoveyda, A. H.;
Schrock, R. R. J. Am. Chem. Soc. 1998, 120, 4041-4042. (j) Fujimura,
O.; Grubbs, R. H. J. Org. Chem. 1998, 63, 824-832. (k) Schrock, R.
R. Acc. Chem. Res. 1990, 23, 158-165.
(5) Wigley, D. E. Organoimido Complexes of the Transition Metals. In
Progress in Inorganic Chemistry; Karlin, K. D., Ed.; John Wiley &
Sons: New York, 1994; Vol. 42, p 239.
(2) Bennett, J. L.; Wolczanski, P. T. J. Am. Chem. Soc. 1997, 119, 10696-
10719.
10.1021/ic049644o CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/12/2004
Inorganic Chemistry, Vol. 43, No. 12, 2004 3605

Ciszewski et al.
Chart 1. A Few of the Pertinent Valence Bond Structures for a
Transition Metal Bound Imido Ligand
Figure 1. Statistics on the bis(imido) complexes of group 6 from the
Cambridge Structural Database. Each entry is the sum of the two imido
bond angles in a complex.
mandated by the electronic structure of most systems is not
well established. In addition, imido bending may be of more
electronic importance in some systems than in others. In this
study, we sought to determine if potential electronic conse-
quences of imido bending were intimately related to elec-
tronegativity changes at the metal center and bond polarity.
One commonly discussed bonding model for imido ligands
uses three valence bond structures to interpret metal-imido
interactions (Chart 1).5,10 Structure I is a bent imido where
an sp2-hybridized nitrogen acts as a 2-electron donor to the
metal (using the neutral method for counting electrons).
Structure II is a linear imido where an sp-hybridized nitrogen
acts as a 2-electron donor, and the lone pair of electrons
occupies an orbital that is principally N(2p) in character.
Structure III is a linear imido where an sp-hybridized
nitrogen acts as a 4-electron donor with the lone pair on
nitrogen donating into a π-acceptor orbital of the metal.
Combinations of these three bonding forms have been used
to qualitatively categorize metal imido interactions. Among
these three canonical forms, structure I can be unambiguously
identified in the solid state where it is found, but distinction
between structures II and III can be difficult based on metric
considerations. Even so, the differences in bonding between
structures II and III clearly have important consequences
with respect to the electronics of the system. Although II is
geometrically similar to III, its electron donation more
closely resembles mode I.
Mo-N-Ph angles in the solid state of 169.4(4)° and 139.4-
(4)°. Although complexes may have very different imido
geometries in the solid state, it is rare to find bis(imido)
groups that do not equilibrate rapidly in solution, character-
ized by exchange averaged NMR resonances for the imido
groups.7,8 Alternatively, the distortion existent in the solid
state may not persist in fluid solution, which is also consistent
with imido group equivalence.
A statistical analysis of the bis(imido) complexes of the
group 6 transition elements in the Cambridge Structural
Database is shown in Figure 1. In order to avoid problems
with determining if a particular imido in a complex was bent
relative to another when there may be a partial third bond
to both imido ligands, the imido bond angles in each complex
were summed. The complex with the lowest sum of bond
angles is the aforementioned Mo(NPh)2(S2CNEt2)2 at 309°.6
The average sum of imido bond angles was 328° with a
standard deviation of 8°. In this set, it was no more likely
for 6-coordinate complexes, where the maximum bond order
to the imido ligands is 2.5, to have smaller imido angles
than 4- and 5-coordinate complexes. However, even though
4- and 5-coordinate complexes may have two full triple
bonds to the imido ligands, π-bonding to a non-imido ligand
may compete with imido π-bonding.9 Of the bis(imido)
complexes examined, 30% had at least one imido with a bond
angle of 155° or less. Consequently, imido ligands of group
6 bis(imido) complexes often have fairly small M-N(imi-
do)-C bond angles. A statistical analysis does not suggest,
however, anything about whether the imido ligands are bent
for electronic reasons nor the energy associated with the
bending.
Orbital symmetry arguments have been used to discrimi-
nate between II and III. For example, II has been used to
describe imido bonding in the complex Ta(Cp*)2(NPh)H by
Bercaw and co-workers.11 For a similar niobium phenylimido
complex Nb(Me3SiCp)2(NPh)Cl, the Nb-N-C(ipso) angle
is somewhat bent at 165°.12 Calculations suggest that the
potential energy surface describing imido bending is es-
While imido bending is not uncommon in bis(imido)
complexes, the extent to which angular deformation is
(6) Haymore, B. L.; Maatta, E. A.; Wentworth, R. A. D. J. Am. Chem.
Soc. 1979, 101, 2063.
(7) (a) Barrie, P.; Coffey, T. A.; Forster, G. D.; Hogarth, G. J. Chem.
Soc., Dalton Trans. 1999, 4519-4528. (b) Strong, J. B.; Yap, G. P.
A.; Ostrander, R.; Liable-Sands, L. M.; Rheingold, A. L.; Thouvenot,
R.; Gouzerh, P.; Maatta, E. A. J. Am. Chem. Soc. 2000, 122, 639-
649.
(10) For the sake of simplicity, we give a few of the canonical valence
bond forms. A more complete discussion involving all of the limiting
structures as applied to imido bonding is found in the following:
Cundari, T. R. Chem. ReV. 2000, 100, 807-818 and references therein.
(11) Parkin, G.; van Asselt, A.; Leahy, D. J.; Whinnery, L.; Huya, N. G.;
Quan, R. W.; Henling, L. M.; Schaefer, W. P.; Santarsiero, B. D.;
Bercaw, J. E. Inorg. Chem. 1992, 31, 82.
(8) Bradley, D. C.; Hodge, S. R.; Runnacles, J. D.; Hughes, M.; Mason,
J.; Richards, R. L. J. Chem. Soc., Dalton Trans. 1992, 1663-1668.
(9) Lin, Z.; Hall, M. B. Coord. Chem. ReV. 1993, 123, 149.
(12) Antin˜olo, A.; Espinosa, P.; Fajardo, M.; Go´mez-Sal, P.; Lo´pez-
Mardomingo, C.; Martin-Alonso, A.; Otero, A. J. Chem. Soc., Dalton
Trans. 1995, 1007-1013.
3606 Inorganic Chemistry, Vol. 43, No. 12, 2004

Transition Metal-Imido Bonding in M(NBut)2(dpma)
sentially flat to a bond angle of 140°. Consistent with earlier
theoretical findings by Jo¨rgenson,13 electron density in both
complexes can be accepted by the phenyl ring reducing imido
π-bonding and flattening the potential energy surfaces
describing imido angle deformation. For Nb(Cp)2(NBut)Cl,
a relatively short Nb-N(imido) bond and a linear Nb-N-
C(quaternary) angle suggests a structure more closely ap-
proximated by III.14
di(pyrrolyl-R-methyl)-N-methylamine (dpma), we have syn-
thesized three bis(tert-butyl)imido complexes spanning group
6, M(NBut)2(dpma). These complexes represent new mem-
bers of a rare class of bis(imido) complexes where the two
imido groups are nonequivalent in fluid solution on the NMR
time scale. The three bis(imido) dpma complexes are very
closely related structurally and contain two nonequivalent
imido groups in the solid state. Consequently, this class is
well suited for the study of metal-imido bonding interac-
tions. Because all the members of the triad have structures
that are closely related, small differences in the spectroscopy
largely are attributed to electronic consequences of proceed-
ing from one metal in group 6 to another. We also have
prepared Mo[N(2,6-Pri2C6H3)]2(dpma) for comparisons be-
tween alkyl and aryl substituents, which will be briefly
discussed.
In order to discover what may be learned from this series,
the compounds were scrutinized using 1H NMR, 13C NMR,
CP-MAS 13C NMR, 14N NMR, and X-ray diffraction. The
spectroscopic data were analyzed in a variety of ways to
identify the techniques for probing imido electronic structure
that are conceptually useful for the system. While the
spectroscopic data provide information about imido ligands
in differing environments on the same metal, the experi-
mental data cannot distinguish between imido differences
due to the dissimilar overall environments and imido
geometry, e.g. bond angle. Consequently, density functional
theory was used to explore the electronic energy associated
with imido bending.
Calculations suggest that energy differences between linear
and bent imido structures are quite small for many com-
plexes,12,13 a behavior not mirrored by more covalent nitrogen
double bonds such as those found in organic imines.15 Indeed
structure II is highly reminiscent of the transition state in
the “lateral shift” mechanism for imine isomerization, the
favored pathway of cis-trans interconversion for most
organic imines.16 A mechanism less likely in more covalent
systems involves polarization of the C-N double bond to
give a zwitterionic transition state. In transition metal imidos,
where the M-N double bond is more highly polarized, the
zwitterionic structures should be much larger participants.
Consequently, canonical limiting polar forms exemplified by
IV and V (Chart 1) should be considered in the description
of transition metal-imido compounds.10 Interconversion
between the two zwitterionic limiting structures causes the
bond angle to change dramatically, but electron density at
nitrogen and the metal remain relatively constant. As this
paper is largely concerned with imido bonding alterations
concomitant with electronegativity change at a metal center,
some polar forms are included for discussion purposes.
Polarity of metal nitrogen bonds increases when descending
a triad (vide infra).17 Consequently, the descriptions I, II,
and III should be accompanied by larger contributions from
the more polar structures IV and V for the heavier congeners.
Wilkinson and co-workers prepared Cr(NBut)2{[ButNC-
(O)]2NBut},18 which exhibits two nonequivalent imido sub-
stituents on the NMR time scale in solution.19 Empirically,
it appears that the combination of a tridentate, dianionic
ligand and two monodentate substituents on a transition metal
often leads to complexes with nonequivalent monodentate
groups that do not rapidly exchange, at least on the NMR
time scale.20,21 Utilizing the tridentate, dianionic ligand N,N-
Results and Discussion
Syntheses and Structures. The starting material H2dpma
is readily synthesized via Mannich reaction between pyrrole,
formaldehyde, and methylamine hydrochloride. Treatment
of H2dpma with a slight excess of LiBun gives, after workup,
Li2dpma.20
Et2O
CrBr2(py)(NBut)2 + Li2dpma
8
Cr(NBu )2(dpma)
t
-2LiBr
-pyridine
1
34%
(1)
Et2O
MoCl2(NBut)2(dme) + Li2dpma
W(NHBut)2(NBut)2 + H2dpma
8
t
Mo(NBu )2(dpma)
(13) Jørgensen, K. A. Inorg. Chem. 1993, 32, 1521-1522.
(14) Chernega, A. N.; Green, M. L. H.; Sua´rez, A. G. J. Chem. Soc., Dalton
Trans. 1993, 3031.
(15) McCarty, C. G. syn-anti Isomerization and rearrangements. In The
Chemistry of the Carbon-Nitrogen Double Bond; Patai, S., Ed.; The
Chemistry of Functional Groups; Interscience Publishers: London,
1970; p 365-464.
(16) (a) Cyclic Organonitrogen Stereodynamics; Lambert, J. B., Takeuchi,
Y., Eds.; VCH Publishers: New York, 1992. (b) Guerra, A.; Lunazzi,
L. J. Org. Chem. 1995, 60, 7959-7965. (c) Kontoyianni, M.; Hoffman,
A. J.; Bowen, J. P. J. Comput. Chem. 1992, 13, 57-65.
(17) For an example of recent calculations on oxo systems, see: Gisdakis,
P.; Rosch, N. J. Am. Chem. Soc. 2001, 123, 697-701.
(18) Lam, H.-W.; Wilkinson, G.; Hussain-Bates, B.; Hursthouse, M. B. J.
Chem. Soc., Dalton Trans. 1993, 781-788.
(19) For additional systems involving nonexchanging bent and linear imidos
on the NMR time scale, see: (a) Danopoulos, A. A.; Wilkinson, G.;
Hussain, B.; Hursthouse, M. B. J. Chem. Soc., Chem. Commun. 1989,
896-897. (b) Danopoulos, A. A.; Leung, W.-H.; Wilkinson, G.;
Hussain-Bates, B.; Hursthouse, M. B. Polyhedron 1990, 9, 2625-
2634.
-2LiCl
38%
2
(2)
Et2O
t
8
W(NBu )2(dpma)
-2H2NBut
50%
3
(3)
Et2O
Mo(Ndip) (dpma)
MoCl2(Ndip)2(dme) + Li2dpma -2LiCl8
2
4
35%
(4)
Reaction of Cr(NBut)2(Br)2(py)22 and Li2dpma in ethereal
solvent yields dark red Cr(NBut)2(dpma) (1) (eq 1). The
(21) Schrock, R. R.; Seidel, S. W.; Schrodi, Y.; Davis, W. M. Organo-
metallics 1999, 18, 428-437.
(22) Meijboom, N.; Schaverien, C. J.; Orpen, A. G. Organometallics 1990,
9, 774-782.
(20) Li, Y.; Turnas, A.; Ciszewski, J. T.; Odom, A. L. Inorg. Chem. 2002,
41, 6298.
Inorganic Chemistry, Vol. 43, No. 12, 2004 3607

Ciszewski et al.
Table 1. Selected Bond Distances and Angles from X-ray Diffraction
on Complexes 1-4a
Mo[N(2,6-
distances (Å) and
angles (deg)
Cr(NBut)2- Mo(NBut)2- W(NBut)2- Pri2C6H3)]2-
(dpma) (1) (dpma) (2) (dpma) (3) (dpma) (4)
M-N(4)
1.6256(18)
1.6409(18)
1.996(2)
1.731(6)
1.732(6)
2.068(6)
2.339(7)
1.748(3)
1.757(3)
2.067(3)
2.321(3)
1.751(5)
1.728(4)
2.073(5)
2.290(5)
M-N(5)
M-N(pyrrolyl) av
M-N(3)
2.1717(18)
M-N(4)-C
M-N(5)-C
N(4)-M-N(5)
N(pyrrolyl)-M-N(4) av 99.35(8)
N(pyrrolyl)-M-N(5) av 96.64(8)
N(3)-M-N(4)
N(3)-M-N(5)
N(pyrrolyl)-M-N(3) av 76.05(8)
151.10(16)
175.28(16)
112.48(9)
156.5(6)
171.6(5)
110.5(3)
104.2(3)
99.3(3)
98.5(3)
150.8(3)
72.6(3)
158.2(3)
170.4(3)
111.40(14) 111.2(2)
103.57(13) 103.4(2)
99.32(12)
98.89(12)
149.51(12) 141.8(2)
72.57(11) 72.79(18)
155.8(5)
175.2(5)
98.1(2)
106.9(2)
113.64(8)
133.88(8)
Figure 2. ORTEP representation of the solid-state structure of Cr(NBut)2-
(dpma) (1) from X-ray diffraction. Ellipsoids are at the 50% probability
level.
a For the numbering scheme, see Figure 2.
and second largest angles around the metal center, respec-
tively.25 The value of τ can vary from 0 (ideal square
pyramid) to 1 (ideal trigonal bipyramid). Consistent with
square pyramidal as the closest descriptor for 1-3, τ is
computed as 0.28, 0.22, and 0.19, respectively.
solid-state structure exhibits bond angles around chromium
inconsistent with either common limiting form for a penta-
coordinate complex, but it is best described (vide infra) as a
pseudo square pyramid with the imido having the smaller
Cr-N(imido)-C bond angle occupying the axial position
(Figure 2). Having two different imido substituents, one bent
and one linear, is common for bis(imido) complexes in the
solid state. Almost invariably, however, the two imido
substituents equilibrate in solution, giving rise to a single
NMR resonance for the substituents on the imido groups.7,8
In contrast, there are two sharp, distinct tert-butyl resonances
Arylimido Mo[N(2,6-Pri2C6H3)]2(dpma) (4) was prepared
from Mo[N(2,6-Pri2C6H3)]2Cl2(dme)23 and Li2dpma in 35%
yield (eq 4). X-ray diffraction on 4 yields a structure
remarkably similar to the tert-butylimido derivative 2.
Arylimido 4 has bent and linear imido angles of 155.8(5)°
and 175.2(5)°. The Mo-N(imido) bond distances do vary
with statistical significance at 1.728(4) and 1.751(5) Å. The
largest difference in the bis(arylimido) structure from those
observed for the bis(tert-butylimido) complexes is found in
the angles to the dpma ligand, which are altered to accom-
modate the isopropyl substituents. Bis(arylimido) 4 has a
structure that closely resembles a square pyramid as judged
by the τ-parameter of 0.01 for the solid-state structure.
The M(imido)2 core is remarkably well retained across
the entire series as exemplified by the N(4)-M-N(5) bond
angles, which are between 110.5° and 112.5° in all four
complexes.
1
observed in the solution H NMR of 1 from -80 °C to 80
°C. The chelating dpma ligand apparently prohibits equili-
bration of the two imido substituents in solution on the NMR
time scale.
By reactions outlined in eqs 2 and 3, molybdenum and
tungsten analogues of 1, Mo(NBut)2(dpma) (2) and W(NBut)2-
(dpma) (3), have been synthesized. Molybdenum bis(imido)
2 is readily prepared from Mo(NBut)2Cl2(dme)23 and Li2-
dpma. The tungsten analogue 3 was prepared by transami-
24
nation on W(NBut)2(NHBut)2 with H2dpma. These com-
plexes also display two different resonances due to the tert-
1
1H and 13C NMR Spectroscopy. The H and 13C NMR
1
butylimido groups that do not equilibrate on the H NMR
spectra of M(NBut)2(dpma), where M ) Cr, Mo, and W,
display nonequivalent tert-butyl resonances. Utilizing NOE
experiments, we were able to confirm that the overall
geometry in the solid state was retained in solution.
Moreover, we were able to determine which resonances in
time scale even at 80 °C.
The solid-state structures for 1-3 are quite similar (Table
1) with the largest variations found in chromium-containing
1 due to its smaller atomic radius. The differences in the
imido bond angles for 1-3 are slight. The two imido angles
in 1 measure 151.1(2)° and 175.3(2)°. The Cr-N(imido)
bond distances differ with statistical significance in the
chromium derivative at 1.626(2) and 1.641(2) Å for linear
and bent, respectively. In the tungsten structure, W-N(imido)
distances do not differ with statistical significance at 1.748-
(3) and 1.757(3) Å. The molybdenum bis(imido) 2 is
structurally identical to the tungsten complex 3 within
statistical errors for bond distances and angles.
1
the H NMR correspond to the axial (bent) and equatorial
(linear) tert-butylimido protons for each compound. Once
all of the resonances were assigned in the 1H NMR, (1H,13C)-
HMQC NMR spectroscopy26 was used to relate the assign-
ments to the 13C NMR. The full assignments for all of the
tert-butylimido complexes can be found in the Experimental
Section. The NOE experiments confirm what might have
been expected; namely, the nuclei near or adjacent to the
axial (bent) imido nitrogen are shielded with respect to
The accuracy of describing complexes 1-3 as square
pyramids may be quantified using the continuous symmetry
parameter τ ) (R - â)/60, where R and â are the largest
(25) (a) Alvarez, S.; Llunell, M. Dalton Trans. 2000, 3288-3303. (b)
Addison, A. W.; Rao, T. N.; Reedijk, J.; van Rijn, J.; Verschoor, G.
C. Dalton Trans. 1984, 1349-1356. (c) Zabrodsky, H.; Peleg, S.;
Avnir, D. J. Am. Chem. Soc. 1992, 114, 7843-7851.
(26) Minoretti, A.; Aue, W. P.; Reinhold: M.; Ernst, R. R. J. Magn. Reson.
1980, 40, 175-190.
(23) Fox, H. H.; Yap, K. B.; Robbins, J.; Cai, S.; Schrock, R. R. Inorg.
Chem. 1992, 31, 2287-2289.
(24) Nugent, W. A.; Harlow, R. L. Inorg. Chem. 1980, 19, 777-779.
3608 Inorganic Chemistry, Vol. 43, No. 12, 2004

Transition Metal-Imido Bonding in M(NBut)2(dpma)
comparable nuclei on the equatorial (linear) imido, which is
consistent with a more electron-rich nitrogen associated with
the axial imido.
Derivatives containing the tert-butylimido group enable
relatively direct study of M-N bonding by comparison of
13C NMR chemical shifts.27 This is often accomplished by
taking the difference between quaternary and primary carbon
resonances in the 13C NMR for the tert-butylimido group,
which is a quantity referred to here as ∆δRâ. For tert-
butylimido complexes, the larger values of ∆δRâ are often
attributed to the most covalent metal-nitrogen interactions
with the highest amount of triple bond character (structure
III in Chart 1). Smaller values of ∆δRâ are believed to be
the result of higher electron density on the imido nitrogen,
which should be due to greater participation by structures I,
II, IV, or V. For analogous complexes down a transition
metal triad, the polarity increases result in smaller ∆δRâ
values for the heavier congeners, which is consistent with
the greater participation of modes IV and V for isostructural
complexes. In our system, polarity increases also result in a
smaller ∆δRâ as one proceeds from chromium to molybde-
num to tungsten with values of 47.17, 38.54, and 34.17 ppm,
respectively.28
While the method of using ∆δRâ values within an
isostructural series to evaluate imido electronics seems to
have merit, comparisons between the various ligand sets are
tenuous. Disparate ligand geometries naturally lead to
different interactions with the imido-metal core and may
not accurately describe the σ + π donor ability of all ligand
sets. The utility of the method is as a qualitative estimate of
ligand donor ability when used outside of an isostructural
series. However, the results with various ligand sets are
intriguing and at least consistent with what might be expected
intuitively for the donor properties of some common ligand
sets.
Figure 3. Plot of bis(tert-butylimido) complexes of group 6 ∆δRâ values
versus Sanderson electronegativity values for the metals in the +6 oxidation
state. 0: M(Tp*)Cl(NBut)2, y ) 22.785 + 8.8668x, R ) 0.9974. 4:
M(dpma)(NBut)2, y ) 21.613 + 7.6024x, R ) 0.99963. ×: M(OSiR3)2-
(NBut)2, y ) 20.194 + 7.6545x, R ) 0.99999. ]: M(Cp)(Me)(NBut)2, y
) 24.696 + 5.3457x, R ) 0.99852.
nitrogen bonds. Plots of ∆δRâ versus estimated electro-
negativity of the metal center for four different bis(tert-
butylimido) systems are found in Figure 3. The four different
systems are M(Tp*)Cl(NBut)2,32 M(OSiR3)2(NBut)2,27,33
M(Cp)Me(NBut)2,22,34 and M(dpma)(NBut)2, where M ) Cr,
Mo, and W. The linear fits for all four lines are quite
respectable, suggesting that metal electronegativity is a large
contributor to imido nitrogen electron density. The linear fits
also suggest that ∆δRâ values are a reasonable measure of
imido nitrogen electron density, at least in isostructural
systems. The good linear fits do not necessarily indicate that
Sanderson’s electronegatiVities for M(VI) are accurate but
suggest only that the effectiVe electronegatiVity differences
are accurately portrayed within each set of complexes. In
fact, the effective electronegativity of the metal centers in
each of the series of complexes should be different. The
Sanderson electronegativities were calculated using ligands
generally more electronegative and poorer π-donors than
most of those represented by this set. Consequently, the
Sanderson numbers will overestimate the electronegativity
of the metal center. In other words, we have no method for
experimentally referencing our plots to give the “true”
effective electronegativities on the Pauling scale. However,
the more electron-deficient ligand sets would be expected
to cause the greatest change in the metal’s effective elec-
tronegativity and will have the largest slope. The most
electron-rich ligand sets should have the smallest slope. The
slopes of the lines are consistent with the intuitive ordering
of donor ability of the ligand sets. The least electron-donating
might be expected to be M(NBut)2(Tp*)Cl, which has an
8.87 slope value, with a zwitterionic structure placing a
formal positive charge on the metal and an electronegative
Many studies demonstrating linear correlations29 between
electronegativities and chemical shifts have appeared in the
literature.30 As may be expected, many of the values for 13
C
NMR chemical shift differences correlate with estimated
electronegativities of the group 6 metals in the +6 oxidation
state31 and thus with expected polarity changes in the metal-
(27) (a) Nugent, W. A.; McKinney, R. J.; Kasowski, R. V.; Van-Catledge,
F. A. Inorg. Chim. Acta 1982, 65, L91-L93. (b) Ashcroft, B. R.;
Clark, G. R.; Nielson, A. J.; Rickard, C. E. F. Polyhedron 1986, 5,
2081-2091. (c) Nugent, W. A.; Mayer, J. M. Metal-Ligand Multiple
Bonds; Interscience: New York, 1988; pp 133-135.
(28) The values for the dpma complexes reported were obtained by taking
the average of the CR resonances minus the average of the Câ
resonances in each of the spectra. This was done for comparison
purposes with other bis(imido) complexes where the imidos are quickly
exchanging.
(29) For a discussion and an extensive listing of applications in organic
chemistry of NMR as it relates to charge densities, see: Emsley, J.
W.; Feeney, J.; Sutcliffe, L. H. High-Resolution Nuclear Magnetic
Resonance Spectroscopy; Pergamon Press: Oxford, 1966. A brief
discussion on paramagnetic contributions to the chemical shift and
relevant references are found in the discussion on 14N NMR of this
book.
(30) (a) Gasteiger, J.; Marsili, M. Org. Magn. Reson. 1981, 15, 353-360.
(b) Spiesecke, H.; Schneider, W. G. J. Chem. Phys. 1961, 35, 722-
730. (c) Cavanaugh, J. R.; Dailey, B. P. J. Chem. Phys. 1961, 14,
1099-1107. (d) Dailey, B. P.; Shoolery, J. N. J. Am. Chem. Soc. 1955,
77, 3977-3981.
(31) For a discussion of Pauling electronegativities versus transition metal
oxidation states, see: Sanderson, R. T. Inorg. Chem. 1986, 25, 3518-
3522. The same procedure using optical electronegativities provides
plots of equally good linearity. For a discussion of optical electro-
negativity, see: Jørgensen, C. K. Electron Transfer Spectra. Prog.
Inorg. Chem. 1971, 12, 101. For a similar correlation between optical
electronegativity and ∆δ values, see ref 27c.
(32) Sundermeyer, J.; Putterlik, J.; Foth, M.; Field, J. S.; Ramesar, N. Chem.
Ber. 1994, 127, 1201. Tp* is tris(3,5-dimethyl-1-pyrazolyl)hydroborate.
(33) For chromium and molybdenum, R is But. For tungsten, R is Ph.
(34) Radius, U.; Sundermeyer, J. Chem. Ber. 1992, 125, 2183-2186.
Inorganic Chemistry, Vol. 43, No. 12, 2004 3609

Ciszewski et al.
nitrogens comparable in magnitude to changing the metal
center from chromium to tungsten. The effects can be used
to estimate the donor ability of the dpma ligand, which is
given to be similar to dichloride and bis(trialkylsiloxy).37
Differences in solid-state 13C NMR resonances for the
ipso-carbon of aryl substituents on imido ligands have been
successfully correlated to imido bending within Mo(NR)2-
(S2CNEt2)2 complexes.7a For our system in solution, the
difference in chemical shift, ∆δl-b, for the quaternary carbons
on the axial (linear) tert-butyl groups minus the equatorial
(bent), i.e., ∆δl-b ) C(R)linear - C(R)bent, was found to change
as Cr > W > Mo with ∆δl-b equal to 2.1, 1.3, and 0.8 ppm,
respectively.38 We do not currently have a satisfactory
explanation for the observed trend. Despite this, the overall
magnitude of ∆δl-b is descriptive. Even though the difference
in M-N(imido)-C bond angle is >25° for all the complexes
and the two imido substituents reside in distinct coordination
positions, the value of ∆δl-b is less than 2.2 ppm, signifi-
cantly smaller than the effect on ∆δRâ of changing the metal
center or other ligands on the metal.
Figure 4. Estimation of electron-withdrawing ability of various ligand
sets on M(NBut)2. Values are in arbitrary units with bis(tert-butyl)diazene
as reference of 0. Metal centers estimated to be more electron-withdrawing
than the reference are positive.
Consequent to the solution 13C NMR studies, we can
discern that imido nitrogen electron density is greatly affected
by changes in metal center and other ligands on the metal.
The value of ∆δRâ tracks with the estimated electron-
withdrawing ability of the metal center. Imido environment
appears to make a measurable but small contribution to the
chlorine atom. The next most electron-deficient ligand set
by this measure is M(NBut)2(OSiR3)2 with a 7.65 slope value,
which is very similar to M(NBut)2(dpma) with a 7.60 slope
value. The most electron-rich ligand set in the series is
M(NBut)2(Cp)Me with a slope of 5.35, which has quite
electron-donating Cp and Me groups.
electron density on nitrogen as measured by ∆δl-b
.
Solid-State 13C NMR Spectroscopy. In order to bridge
the gap from the solid-state structural data to the solution-
based NMR data, it was desirable to show that the complex
in the solid state gives comparable data to solution measure-
ments by some technique. The CP-MAS 13C NMR spectra
were obtained, which allowed direct comparison with our
solution data. The data correlate amazingly well as can be
seen in Table 2. However, some differences between the
solution and solid-state data are apparent.
A method for referencing the ∆δRâ plots to the Pauling
scale is lacking. However, the plots can be referenced to an
arbitrary value of ∆δRâ. A natural choice for an arbitrary
reference for ButNdM complexes is the covalent system bis-
(tert-butyl)diazene, ButNdNBut (∆δRâ ) 39.0 ppm), where
the imido substituent is bound to itself, resulting in a
minimum of bond polarity. The results are provided in Figure
4. This reinterpretation of the data in Figure 3 allows
estimation of relatiVe electron-withdrawing abilities of metal
centers in different complexes versus a covalent, double-
bond reference. Complexes with positive values presumably
would be withdrawing more electron density from the imido
nitrogen relative to the reference system, which may be
accomplished through triple bonding as in resonance form
III in Chart 1. Negative values would be found for systems
having more participation of polar resonance forms IV and
V. According to this technique, chromium with these ligand
sets is always more electron-withdrawing than NBut and
tungsten less. Molybdenum varied from more to less electron-
withdrawing versus the reference depending on the ligand
set.
The most interesting feature of the solid-state data involves
the ∆δl-b values. The ordering in the solid state was Cr >
Mo > W with values of 4.0, 0.9, and 0.8 ppm, respectively.
In the solid state, chromium complex 1 has a significantly
larger ∆δl-b whereas molybdenum- and tungsten-containing
2 and 3 are similar. In other words, the ∆δl-b values change
as Cr > Mo ∼ W using this technique, which is similar to
the results found with 14N NMR (vide infra).
14N NMR Spectroscopy. Another tool for investigation
of imido bonding is provided by 14N NMR.39 For example,
(36) For a similar discussion on bis(1-adamantylimido)chromium(VI)
complexes, see: Coles, M. P.; Gibson, V. C.; Clegg, W.; Elsegood,
M. R. J. Polyhedron 1998, 2483-2489.
(37) Similar results are found when series of molybdenum complexes are
examined. Cole, S. C.; Coles, M. P.; Hitchcock, P. B. Dalton Trans.
2002, 4168.
A larger set of ligands, X, in X2Cr(NBut)2 can be examined
to estimate the σ + π donation. For X2 ) (neophyl)2,22
(38) An alternative measure of linear versus bent comparison would be to
use ∆δRâ(linear) - ∆δRâ(bent). The solution numbers using this method
would still go in the order Cr > W > Mo with values of 1.2, 0.7, and
0.1, respectively. However, the result is similar in that the ∆δ values
due to changing the imido environment are far smaller (<1.2 ppm)
than changing the metal or ligands on the metal.
22
(neopentyl)2,22 (benzyl)2,35 (OSiMe3)2,22,27 dpma, and Cl2
the electronic changes result in ∆δRâ values that range over
14.8 ppm.36 In other words, altering the other ligands on the
group 6 metal can make electronic changes at the imido
(39) 14N NMR spectra are referenced to the ammonia scale with NH3 as
0.00 ppm. On this scale, neat nitromethane has a resonance at 380.4
ppm. Another common scale uses ammonium as the reference, which
appears at ∼21 ppm on the ammonia scale.
(35) Coles, M. P.; Dalby, C. I.; Gibson, V. C.; Clegg, W.; Elsegood, M.
R. J. Chem. Commun. 1995, 1709-1711.
3610 Inorganic Chemistry, Vol. 43, No. 12, 2004

Transition Metal-Imido Bonding in M(NBut)2(dpma)
Table 2. Comparison of Solution and Solid-State 13C NMR Data
Cr(dpma)-
Mo(dpma)-
W(dpma)-
(NBut)2 (1)
(NBut)2 (2)
(NBut)2 (3)
C(CH3)3 bent
solution
solid
30.2
30.0
31.0
31.0
32.5
32.2
C(CH3)3 linear
solution
solid
31.1
30.6
31.7
32.3
33.1
33.6
CMe3 bent
solution
solid
76.8
75.8
69.5
69.2
66.8
66.8
CMe3 linear
solution
solid
78.9
79.8
70.3
70.1
68.1
67.6
CH2
solution
solid
NCH3
solution
solid
62.9
62.3
60.7
60.3
60.9
59.3
Figure 5. Resonances due to the imido nitrogens in the 14N NMR of Cr-
(NBut)2(dpma) (1) [top], Mo(NBut)2(dpma) (2) [middle], and W(NBut)2-
(dpma) (3) [bottom].
46.0
44.6
44.6
45.2
44.7
42.3
1H NMR, and 13C NMR, the imido nitrogens manifest a very
different result when interrogated by 14N NMR (Figure 5).
The only complex that exhibits two clearly distinct imido
resonances is Cr(NBut)2(dpma) (1). The spectrum of 1 has
one relatively sharp peak at 588 ppm (∆ν1/2 ) 382 Hz)
attributable to the axial (bent) imido nitrogen7b and a broader
resonance at 560 ppm (∆ν1/2 ) 688 Hz) for the equatorial
(linear) imido nitrogen. The molybdenum derivative,
Mo(NBut)2(dpma) (2), was also interrogated by 14N NMR,
and evinces two unresolved resonances at 472 (∆ν1/2 ) 248
Hz) and 458 ppm (∆ν1/2 ) 1100 Hz). The heaviest congener
of the triad, W(NBut)2(dpma) (3), reveals a single apparent
imido resonance at ∼413 ppm. However, simulation of the
spectrum requires two resonances at 415 (∆ν1/2 ) 808 Hz)
and 413 ppm (∆ν1/2 ) 190 Hz) to model the shape of the
peak.
pyrrole-4-C
solution
solid
pyrrole-3-C
solution
solid
pyrrole-2-C
solution
solid
pyrrole-1-C
solution
solid
103.0
103.2
105.0
103.7
105.7
106.8
109.3
108.3
110.0
109.3
111.0
108.9
132.6
132.0
133.4
132.4
134.3
134.1
138.9
137.1
139.0
139.3
139.8
140.3
Osborn and Le Ny40 reported the 14N NMR of several
tungsten mono(imido) complexes. The complexes were
axially symmetric with trigonal bipyramidal geometries. A
consequence of the highly symmetric environment, reso-
nances in the 14N NMR were relatively narrow (∆ν1/2 <100
Hz). The lower symmetry of the dpma complexes in our
system leads to broader lines; however, all of the resonances
corresponding to the nitrogen atoms of the complexes are
readily observed.
Shielding in nitrogen NMR is largely ascribed to the
paramagnetic term (σp) as described by the Ramsey equa-
tion.41 Imido bonding as probed by 14N NMR has been
thoroughly discussed by Bradley and co-workers.8 Imido
resonances in nitrogen NMR are known to shield signifi-
cantly as one proceeds down a column. The shielding is
reasonably attributed to higher negative charges on nitrogen
as M-N(imido) bond polarity increases and to increases in
∆E due to larger ligand field splittings on proceeding down
a column.8 Nitrogen NMR is expected to be reasonably
sensitive to the imido M-N-C angle. For a linear versus
bent imido nitrogen, the bent imido will be expected to show
decreased ∆E due at least in part to participation of lower
energy n f π* circulations.8 Consequently, the less shielded
imido nitrogen resonances are attributed to the axial (bent)
imido nitrogens.
As expected, the imido nitrogen resonances do signifi-
cantly increase in shielding from 1-3. The shielding increase
is dramatic at >150 ppm down the triad. Similar trends are
seen in the nitrogen NMR of nitrido42 complexes.43
Comparison of the two molybdenum bis(imido) complexes
2 and 4 provides a crude gauge of imido substituent effects
on the 14N shift. While alkyl imido 2 displays two resonances
for the imido groups, aryl imido 4 has a single peak in the
14N NMR for the imido nitrogen atoms at 429 ppm (∆ν1/2
) 648 Hz). Modeling of the imido resonance reveals that
the peak shape is Lorentzian. Consequently, it is assumed
that the two resonances corresponding to the differentiated
imido nitrogens are not separated enough for deconvolution.
One consequence of increased M-N(imido) polarity31 may
be decreased sensitivity of chemical shifts to imido nitrogen
environment. This is substantiated by the sensitivity of the
14N resonance shifts to nitrogen environment in the more
covalent chromium complex 1, and lower sensitivity of 14
N
(42) (a) Mason, J. Chem. ReV. 1981, 81, 205-227. (b) Donovan-Mtunzi,
S.; Richards, R. L.; Mason, J. J. Chem. Soc., Dalton Trans. 1984,
1329. (c) Dilworth, J. R.; Donovan-Mtunzi, S.; Kan C. T.; Richards,
R. L.; Mason, J. Inorg. Chim. Acta 1981, 53, L161-L162. (d) Odom,
A. L.; Cummins, C. C. Organometallics 1996, 15, 898-900.
While all four compounds discussed above display two
clearly distinct imido substituents by X-ray crystallography,
(40) Le Ny, J. P.; Osborn, J. A. Organometallics 1991, 10, 1546-1550.
(41) (a) Webb, G. A.; Witanowski, M. Theoretical Background to Nitrogen
NMR. In Nitrogen NMR; Witanowski, M., Webb, G. A., Eds.; Plenum
Press: London, 1973. (b) Mason, J. Multinuclear NMR; Plenum
Press: New York, 1987.
(43) For discussion of similar effects in 17O NMR metal-oxo shifts, see:
(a) Kidd, R. G. Can. J. of Chem. 1967, 45, 605-608. (b) Klemperer,
W. G. Angew. Chem., Int. Ed. Engl. 1978, 17, 246-254. (c) Miller,
K. F.; Wentworth, R. A. D. Inorg. Chem. 1979, 18, 984-988.
Inorganic Chemistry, Vol. 43, No. 12, 2004 3611

Ciszewski et al.
Figure 6. Comparison of bond distances (Å) and angles (deg) from X-ray diffraction (top) on 1-3 and the optimized structures from DFT calculations
(bottom).
NMR resonance shifts for the more polar tungsten-containing
3. Indeed, the two imido nitrogens become increasingly
difficult to distinguish on proceeding from chromium to
molybdenum to tungsten. If one assumes that one of the
imido substituents is slightly more electron-rich than the
other, which is consistent with the 13C NMR data, the two
imido substituents would react slightly differently to decreas-
ing the metal’s electronegativity on proceeding down the
column. Rising polarity in moving to the heavier congeners
effectively decreases the differences in nitrogen shielding
in the different environments. A similar argument can be
put forth to explain differences between alkyl imido 2 and
aryl imido 4. While two imido peaks are noticeable for 2,
the aryl substituents on the imido nitrogens of 4 result in
more electronegative imido nitrogens and higher Mo-
N(imido) bond polarity. Consequently, one resonance is
observed for the two different imido nitrogens of 4. In
addition, the aryl group will resonance stabilize the imido
“lone pair” leading to an increase in ∆E for n f π*
paramagnetic circulations, which leads to the ∼30 ppm
shielding of the arylimido nitrogen relative to the tert-butyl
derivative.
An alternative explanation would rely on the fact that the
major factor leading to differences in 14N NMR chemical
shifts is the imido bond angle differences. The other NMR
nuclei would need to be more sensitive to other environ-
mental issues. The more polar imido bonds may have a lower
energy bent to linear conversion barrier, which is leading to
fluxional processes equilibrating the imido bond angles on
the 14N NMR time scale. Judging from the solid-state
structures and density functional theory calculations (vide
infra), the minima of the potential energy surfaces associated
with imido bond angle deformation are similar, but the
energies associated with a specific angle change are different.
One possible experimental probe would be temperature
dependence in nitrogen NMR of the complexes. The
complexes should all reach a low temperature limit with two
nonequivalent imido resonances similar to the chromium
structure. Unfortunately, the broadening of the peaks at low
temperature quickly becomes prohibitive. On raising the
temperature, solutions of 2 gave sharper, clearer resonances
with some shifting. However, the shift difference between
axial and equatorial imido nitrogen resonances did not change
significantly over the temperature range we were able to
access. Consequently, there is no fluxionality indicated by
the 14N NMR spectra, which is consistent with the spectros-
copy of the other nuclei.
From examining 14N NMR of complexes 1-3, a few
conclusions may be made. First, the imido nitrogen reso-
nances shield significantly as one proceeds down the triad.
Second, lowering the metal’s electronegativity has the
apparent result of reducing the effective difference between
the two imido nitrogens in the two different positions of the
square pyramid. Like the ∆δl-b values discussed earlier, the
differences between axial (bent) versus equatorial (linear)
imido ligands followed the trend Cr > Mo ∼ W. However,
consistent with the previous spectroscopic results, the shifts
due to changing the metal down the triad were substantially
larger than due to electronic differences on the two imido
ligands.
The 14N NMR resonances corresponding to the amine
donors of the dpma region are all in the typical range for
amine complexes. Modeling of the tertiary amine peaks gave
resonance maximums for compounds 1-4 at 84 (∆ν1/2
)
1024 Hz), 74 (∆ν1/2 ) 1594 Hz), 79 ppm (∆ν1/2 ) 1750
Hz), and 42 (∆ν1/2 ) 3132 Hz), respectively.
Correlations between 14N NMR chemical shifts of the
pyrrolyl nitrogen and π-accepting ability of the transition
3612 Inorganic Chemistry, Vol. 43, No. 12, 2004

Transition Metal-Imido Bonding in M(NBut)2(dpma)
metal center would be a potentially useful contribution to
metal-pyrrolyl chemistry. Evidence in the literature suggests
an intimate link between pyrrole nitrogen NMR shifts and
π-electron density in the ring.44 For free azoles, 14N NMR
chemical shifts were found to vary linearly with calculated
π-charge densities (SCF-PPP-MO method) of the aromatic
ring. Therefore, as applied to our system, significant per-
turbations in the π-system due to metal center competition
for the pyrrolyl nitrogen lone pair could induce dramatic
changes in chemical shifts of pyrrolyl nitrogen resonances.
For Cr(NBut)2(dpma) (1), the 14N NMR resonance for the
pyrrolyl nitrogens is found at 195 ppm (∆ν1/2 ) 125 Hz).
The molybdenum and tungsten pyrrolyl nitrogen shifts are
found at 198 (∆ν1/2 ) 372 Hz) and 201 ppm (∆ν1/2 ) 421
Hz), respectively. The arylimido-containing 4 has a pyrrolyl
nitrogen resonance at 206 ppm (∆ν1/2 ) 895 Hz). The shifts
display a slight deshielding as one proceeds down the triad,
which is consistent with slightly more π-bonding between
the heavier congeners and the pyrrolyl ring. In addition, it
appears that the less electron-donating arylimido groups on
4 lead to additional pyrrolyl nitrogen electron-density dona-
tion. However, the differences in chemical shifts are rela-
tively small and suggest only little or no significant bonding
differences for this particular series. The relatively small 14
N
NMR chemical shifts on proceeding down the column may
be related to competition for the π-acceptor orbital between
the axial imido and the pyrrolyl nitrogen lone pairs, which
would strongly attenuate the pyrrolyl nitrogen π-donation
relative to complexes where a low energy metal orbital is
available to accept the nitrogen electron density.45
Examination of M(NBut)2(dpma) Using DFT. Examina-
tion of the structures of M(NBut)2(dpma) complexes by
density functional theory provided optimized atomic positions
in remarkable agreement with the single-crystal X-ray
diffraction experiments. The metal-ligand distances and
M-N-C angles for complexes 1-3 are shown in Figure 6.
Like the distances from X-ray diffraction, the bond distances
from chromium to the axial and equatorial imido nitrogens
are essentially the same. Correlations between bond distance
and bond order seem to rarely be valid. The correlations often
break down unexpectedly, and bond distances are the product
(44) Witanowski, M.; Stefaniak, L.; Januszewski, H.; Grabowski, Z.
Tetrahedron 1972, 28, 637-653.
(45) Pyrrolyl 14N NMR chemical shifts for the isostructural series
M(dpma)2, where M is Ti, Zr, and Hf, were found to vary over a 47
ppm range. The larger range in these molecules is attributable to
stronger π-bond to the metal complexes of this group 4 series with
available orbitals, without competing π-donation, to accept π-electron
density from the pyrroles. See ref 20.
(46) (a) Steffey, B. D.; Fanwick, P. E., Rothwell, I. P. Polyhedron 1990,
9, 963-968. For similar examples of early metal alkoxides with little
or no correlation between bond distance and angle, see: (b) Howard,
W. A.; Trnka, T. M.; Parkin, G. Inorg. Chem. 1995, 34, 5900. (c)
Wolczanski, P. T. Polyhedron 1995, 14, 3335. For systems where
alkoxide bond angle does seem to correlate with bond distance, see:
(d) Tomaszewski, R.; Arif, A. M.; Ernst, R. D. J. Chem. Soc., Dalton
Trans. 1999, 1883-90. (e) Huffman, J. C.; Moloy, K. G.; Marsella,
J. A.; Caulton, K. G. J. Am. Chem. Soc. 1980, 102, 3009. For a recent
computational study supporting Ti-OR triple bonding, see: (f)
Dobado, J. A.; Molina, J. M.; Uggla, R.; Sundberg, M. R. Inorg. Chem.
2000, 39, 2831. For an excellent experimental study comparing
π-bonding in a variety of monoanionic ligands to a common metal
fragment, see: (g) Lukens, W. W., Jr.; Smith, M. R., III; Andersen,
R. A. J. Am. Chem. Soc. 1996, 118, 1719-1728.
Figure 7. Plots of calculated energies versus angle for the axial imido of
compounds 1-3.
of many different factors.9 For example, Rothwell and co-
workers were able to show experimentally that metal
alkoxide bond angle has no apparent correlation to alkoxide
bond distance.46 Examination of the imido complexes
M(NBut)2(dpma) provides a similar conclusion. The two
M-N(imido) bonds in each molecule seem quite similar
despite their different environments.
Inorganic Chemistry, Vol. 43, No. 12, 2004 3613

Ciszewski et al.
were purely σ-only donors, there would potentially be no
electronic reason for the axial imido M-N(imido)-C angle
to deviate from linearity. However, any π-bonding from the
pyrrolyl ligands could compete with axial imido π-bonding.
Several molecular orbitals were found with contributions
from the pyrrolyl π-systems and the axial imido consistent
with this assertion. In addition, the 14N NMR resonances for
the pyrrolyl nitrogen are suggestive of attenuated π-bonding
to those substituents due to competition with the imido (vide
supra).
Conclusions
In this study, we sought to quantify the relationship
between electronegativity and energy associated with imido
angle deformation. While none of the techniques described
above were convincing on their own, the self-consistent
picture created by the host of techniques employed does seem
to imply several conclusions concerning the bis(imido)
systems described.
Figure 8. Plot of calculated energy difference from minimum to 175° in
axial imido angle versus estimated electronegativities on the Pauling scale.
Using the spectroscopic techniques above, it is not possible
to distinguish between different factors leading to resonance
differences for the two imido substituents. It was desirable
to probe the effect of changing the bond angle on the axial
(bent) imido to determine if there is an energetic barrier of
some kind associated with the imido bond angle. The energy
associated with changing the angle of the axial imido
substituent M-N(imido)-C(quaternary But) to 175°, which
is approximately the angle of the equatorial imido, was
explored using DFT. The calculations were done by changing
the axial imido angle and holding the remainder of the
molecule static at the ground-state geometry; this may
provide an upper limit on the energy associated with the
geometry change. The electronic energy associated with
straightening the axial imido for chromium-containing 1 was
calculated as 4.5 kcal/mol (Figure 7). A similar calculation
by Legzdins and co-workers for imido bending in Cr(O)-
(NMe)(Me)(NPri2) provided a similar energy associated with
Cr-N-C angle deformation of 4.4 kcal/mol.47 Examination
of 2 and 3 using the same technique provided energies for
bent imido straightening of 2.7 and 2.0 kcal/mol, respectively.
With energies for imido angle deformation of this mag-
nitude, M-N(imido)-C bond angles should be changing
rapidly on the NMR time scale in room temperature
solutions. The electronic differences measured in fluid
solutions probably represent differences in equilibrium
angular values for imido substituents and differences in
overall coordination environment.
While our studies suggest that the axial imido is bent
(solution and solid) and that it is more electron-rich relative
to the equatorial imido, they do not necessarily establish
causality. In other words, we can show that the axial imido
is more electron-rich and that it has a smaller imido angle,
but we cannot discern definitively that the axial imido is
more electron-rich because it is bent rather than having this
effect as a product of its environment (axial versus equato-
rial). However, the assertion that the axial imido is bent due
to a larger amount of electron-density (or vice-versa) is
reasonable if the canonical forms in Chart 1 are a reasonable
description of the bonding and assuming that imido ligands
have a propensity to form triple bonds (form III, Chart 1)
when possible. We will assume that imido bond angle and
nitrogen electron-density are related. With this caveat, we
offer the following conclusions.
1
The spectroscopic evidence, which includes H, 13C, and
14N NMR, suggests that M(NBut)2(dpma) complexes do not
exchange imido substituents in fluid solution on those time
scales. Consequently, it is possible to study the differences
in the two imido nitrogens due to their differing environments
spectroscopically. For example, it was found that the more
1
shielded tert-butyl resonances in the H NMR were associ-
ated with the, apparently more electron-rich, axial (bent)
imido substituents in complexes 1-3. However, it is likely
from the calculated electronic energies associated with imido
bending that the imido substituents sweep through large
angles in solution and observed differences in the two imido
substituents are due to variances in equilibrium positions and
different environments, e.g. axial vs equatorial, for the two
multiple-bond ligands.
The fact that ∆δl-b is nonzero for each complex suggests
that there are indeed different electron densities on imido
nitrogens in the axial and equatorial orientations. The value
of ∆δl-b in solution and in the solid state changed as Cr >
Mo ∼ W, similar to differences between imido nitrogen
chemical shifts in the 14N NMR.48 However, the effects of
changing the donor ability of the other ligands or the metal
center, as judged by ∆δRâ values, can be significantly larger
Interestingly, taking the calculated values for the axial
imido angle deformation and plotting versus estimated
electronegativity in the +6 oxidation state31 provided a
straight line. This suggests that the angular deformation
energy is closely associated with the M-N(imido) bond
polarity (Figure 8). If factors other than bond polarity were
greatly affecting the electronic energy associated with angle
deformation, one might expect the plot in Figure 8 to deviate
strongly from linearity.
One question we hoped to address using the calculations
was why the axial imido angle deviates from linearity and
has a barrier to straightening. If the pyrrolyl substitutents
(47) Jandciu, E. W.; Legzdins, P.; McNeil, W. S.; Patrick, B. O.; Smith,
K. M. Chem. Commun. 2000, 1809.
3614 Inorganic Chemistry, Vol. 43, No. 12, 2004

Transition Metal-Imido Bonding in M(NBut)2(dpma)
than the effects of differing imido environments. Therefore,
imido environment on the metal center is likely less important
to imido bonding than these metal-nitrogen polarity issues.49
With the above points in mind, a relatively simple
description of imido bonding for this series is made appar-
ent.50 Polar forms such as IV and V will contribute to the
bonding. Because these zwitterionic contributors both have
a M-N(imido) bond order of one, rehybridization involving
exchange of IV for V or vice versa does not greatly affect
the bonding or nitrogen electron density. In this model, the
barrier to straightening or bending an imido bond will be
determined by the energy required to exchange bent imido
covalent form I with linear imido forms II and III, the extent
to which the polar forms IV and V participate, and steric
effects. Consequently, one expects that as the M-N(imido)
bond becomes more covalent for metal centers with relatively
high electronegativity, e.g. chromium(VI) with very electron-
withdrawing groups, the barrier to nitrogen rehybridization
(bending or straightening) should increase. Consistent with
this assertion, density functional theory on the complexes
in this study found that the chromium imido complex 1
displays a higher barrier to imido angle deformation than
the more polar, isostructural tungsten complex. The linear
relationship between calculated barriers and estimated elec-
tronegativity suggests that bond polarity is a major factor in
determining imido deformation energies.
In highly covalent systems, the barrier to isomerization
could potentially approach those in organic imines, which
typically range from 15 to >23 kcal/mol.15,16 Consequently,
the conclusions on this system are consistent with the results
of other studies that have demonstrated calculable47 and
observable7 differences in imido bonding associated with
imido bending, which are usually small. In addition, study
of this system suggests that larger barriers to imido angle
deformation due to electronic structure may be found for
chromium(VI) or other transition metals where the effective
electronegativity more closely approximates that of nitrogen
and where the bending is mandated9 by the electronics of
the system, rather than sterics51 or some other effect.
The slightly higher electron density on the bent imido
nitrogens of some complexes in combination with greater
nitrogen kinetic accessibility potentially could lead to reaction
rate and thermodynamic enhancements relative to linear
imido ligands on the same complex. The reactivity differ-
ences relative to a linear imido are likely to be more
important as M-N(imido) bonding becomes more covalent.
However, for many systems, M-N(imido)-C bending may
be a slight attenuation acting on the larger effect of metal-
nitrogen bond polarity in determining nitrogen electron
density,52 which is largely established by the metal, the
metal’s formal oxidation state, the other ligands, and the
substituent on the imido nitrogen.
Experimental Section
General Considerations. All manipulations of air-sensitive
materials were carried out in an MBraun glovebox under an
atmosphere of purified nitrogen. Ethereal solvents and pentane were
purchased from Aldrich Chemical Co. and distilled from purple
sodium benzophenone ketyl. Toluene was purchased from Aldrich
Chemical Co., refluxed over molten sodium for at least 2 days,
and distilled. Dichloromethane was purchased from Spectrum
Chemical Co., refluxed with calcium hydride for at least 2 days,
and distilled. NMR solvents were purchased from Cambridge
Isotopes Laboratories, Inc. Deuterated benzene was distilled from
purple sodium benzophenone ketyl. Deuterated toluene was de-
gassed and dried with neutral activated alumina. NMR solvents
were stored in sealed containers equipped with a Teflon stopcock
in the drybox prior to use. Spectra were taken on Varian instruments
located in the Max T. Rogers Instrumentation Facility. Routine
coupling constants are not reported. Alumina, silica, and Celite were
dried at >200 °C under dynamic vacuum for at least 12 h and then
stored under an inert atmosphere. Cr(NBut)2(Br)2(py),22 Mo[N(2,6-
Pri2C6H3)]2Cl2(dme),23 W(NBut)2(NHBut)2,24 and Mo(NBut)2Cl2(1,2-
dimethoxyethane)23 were prepared by literature methods. H2dpma
was prepared by the literature procedure.20 Bis(tert-butyl)diazene
was purchased from Aldrich Chemical Co., and spectroscopic
1
examinations were carried out in d7-toluene by H [δ ) 1.18 (s,
But)] and 13C NMR [δ ) 65.95 C(CH3)3, 26.90 C(CH3)3].
Combustion analyses were performed by Oneida Research Services
in Whitesboro, NY.
Procedure for Density Functional Theory Calculations. The
calculated molecular structures were obtained by geometry opti-
mizations at the SCF level using the LANL2dz effective core
potential and the associated 3s3p3d basis set53 for the transition
metals Cr, Mo, and W while the all electron 3-21g basis54 was
used for the main group elements (H, C, O and N). The potential
energy curves for the axial imido angle were calculated at the DFT
level using the B3LYP functional.55 As the angle was varied, we
kept the remainder of the molecule at the optimal SCF geometry,
a procedure that may overestimate the barrier for imido angle
deformation. All calculations used the Gaussian 98-program56 as
implemented on the Chemistry Departments Silicon Graphics Origin
3400 computer.
(52) For an interesting reactivity study on an imido with an extremely small
angle, see: Gountchev, T. I.; Tilley, T. D. J. Am. Chem. Soc. 1997,
119, 12831-12841.
(53) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 284.
(54) Gordon, M. S.; Binkley, J. S.; Pople, J. A.; Pietro, W. J.; Hehre, W.
J. J. Am. Chem. Soc. 1982, 104, 2797.
(55) Becke, A. D. J. Chem. Phys. 1993, 98, 1372.
(56) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels,
A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone,
V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;
Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Salvador, P.; Dannenberg, J. J.; Malick, D. K.; Rabuck,
A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J.
V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.;
Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.;
Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.;
Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian
98, revision A.11.1; Gaussian, Inc.: Pittsburgh, PA, 2001.
(48) Imido bond angle has recently been correlated with 95Mo NMR
shielding as well. Minelli, M.; Hoang, M. L.; Kraus, M.; Kucera, G.;
Loertscher, J.; Reynolds, M.; Timm, N.; Chiang, M. Y.; Powell, D.
Inorg. Chem. 2002, 41, 5954.
(49) For a reference demonstrating through SCF-XR-SW analysis the high
electron density on even linear imido nitrogens, see: Schofield, M.
H.; Kee, T. P.; Anhaus, J. T.; Schrock, R. R.; Johnson, K. H.; Davis,
W. M. Inorg. Chem. 1991, 30, 3595-3604.
(50) Similar arguments relating bond order and bond polarity are put forth
in ref 27c, Chapter 2.
(51) Coffey, T. A.; Forster, G. D.; Hogarth, G.; Sella, A. Polyhedron 1993,
12, 2741-2743.
Inorganic Chemistry, Vol. 43, No. 12, 2004 3615

Ciszewski et al.
Preparation of Li2dpma. A solution of H2dpma (5.00 g, 26.4
mmol) in approximately 100 mL of toluene was cooled to near
frozen in a liquid nitrogen cold well within the drybox. To the
stirring cold solution was slowly added 1.6 M n-butyllithium (35
mL, 56 mmol, 2.12 equiv) by pipet. The product precipitated as a
white solid during addition. When addition was complete, the
solution was stirred for 5 min, and then 100 mL of pentane was
added. The solid was collected on a frit and was washed repeatedly
with pentane. Drying in vacuo yielded Li2dpma (5.30 g, 26.4 mmol,
∼100%).
resulting solution was allowed to warm to box temperature and
stirred for 2 h. Volatiles were removed in vacuo to yield a brown
oil. To the oil was added 2 mL of pentane. The solution was stirred
for several minutes. Volatiles were again removed, which resulted
in a yellow solid. Recrystallization from pentane gave 3 as a tan
1
solid in 50% yield (0.255 g, 0.499 mmol). H NMR (toluene-d8):
δ ) 7.16 (s, 2H, pyrrole-5-H), 6.38 (m, 2H, pyrrole-4-H), 6.20
(m, 2H, pyrrole-4-H), 4.48 (d, J ) 12.6 Hz, 2H, N-CHH-pyrrole,
anti to methyl), 3.46 (d, J ) 12.6 Hz, 2H, N-CHH-pyrrole, syn to
methyl), 1.94 (s, 3H, NCH3), 1.45 (s, 9H, NC(CH3)3, linear), 1.12
(s, 9H, NC(CH3)3, bent). 13C NMR (toluene-d8): δ ) 139.76
(pyrrole-1-C), 134.35 (pyrrole-2-C), 111.00 (pyrrole-3-C), 105.73
(pyrrole-4-C), 68.06 (NCMe3, linear), 66.81 (NCMe3, bent), 60.87
(CH2), 44.658 (NCH3), 33.08 (NC(CH3)3, linear), 32.51 (NC(CH3)3,
bent). 14N NMR (benzene-d6): δ ) 415.21 (∆ν1/2 ) 808 Hz),
412.70 (∆ν1/2 ) 190 Hz), 201.44 (∆ν1/2 ) 421 Hz), 78.71 (∆ν1/2
) 1750 Hz). Anal. Calcd for C19H31N5W: C, 44.46; H, 6.09; N,
13.64. Found: C, 43.99; H, 5.86; N, 13.69.
Preparation of Cr(NBut)2(dpma) (1). A solution of Li2dpma
(0.1006 g, 0.500 mmol) in 5 mL of ether was cooled to near frozen
in a liquid nitrogen cold well. This was added to a cold solution of
Br2Cr(NBut)2(py) (0.2163 g, 0.499 mmol) in 5 mL of ether. The
resulting solution was allowed to warm to box temperature and
stirred for 2 h. The solids formed were filtered off, and ether was
removed in vacuo. The residue was recrystallized from ether/
pentane, yielding 1 as a dark red powder in 34% yield (0.0655 g,
1
Preparation of Mo[N(2,6-Pri2C6H3)]2(dpma) (4). A solution
of Li2dpma (1.001 g, 4.97 mmol) in 25 mL of ether was cooled to
near frozen. This cold solution was added to Cl2Mo(Ndip)2(dme)
(3.038 g, 5.00 mmol) in 25 mL of ether. The resulting solution
was allowed to warm to box temperature and stirred overnight.
Lithium chloride was filtered off, and ether was removed in vacuo.
The resulting brown solid was recrystallized from ether/pentane,
0.172 mmol). H NMR (toluene-d8): δ ) 6.97 (s, 2H, pyrrole-5-
H), 6.60 (m, 2H, pyrrole-4-H), 6.37 (m Hz, 2H, pyrrole-3-H), 4.37
(d, J ) 12.7 Hz, 2H, N-CHH-pyrrole, anti to methyl), 3.57 (d, J )
12.7 Hz, 2H, N-CHH-pyrrole, syn to methyl), 2.41 (s, 3H, NCH3),
1.47 (s, 9H, NC(CH3)3, linear), 1.14 (s, 9H, NC(CH3)3, bent). 13C
NMR (toluene-d8): δ ) 138.95 (pyrrole-1-C), 132.56 (pyrrole-2-
C), 109.33 (pyrrole-3-C), 103.00 (pyrrole-4-C), 78.89 (NCMe3,
linear), 76.78 (NCMe3, bent), 62.88 (methine CH2), 46.03 (NCH3),
31.12 (NC(CH3)3, linear), 30.22 (NC(CH3)3, bent). 14N NMR
1
which gave 4 as a tan solid in 35% yield (1.10 g, 1.74 mmol). H
NMR (benzene-d6): δ ) 6.93-7.02 (m, 6H), 6.91 (m, 2H), 6.40
(m, 2H), 6.26 (m, 2H), 4.43 (d, J ) 13.0 Hz, 2H), 3.90 (h, J ) 6.7
Hz, 2H), 3.46 (d, J ) 13.0 Hz, 2H), 3.18 (h, J ) 6.7 Hz, 2H), 2.23
(s, 3H), 1.05 (d, J ) 6.1 Hz, 9H), 1.04 (d, J ) 6.1 Hz, 9H). 13C
NMR (benzene-d6): δ ) 153.22, 151.90, 147.91, 143.43, 139.20,
131.26, 129.55, 128.29, 128.29, 128.19, 127.81, 126.65, 122.95,
122.35, 111.06, 110.99, 105.84, 60.56, 45.46, 28.88, 28.80, 23.58,
23.42. 14N NMR (benzene-d6): δ ) 429.32 (∆ν1/2 ) 648 Hz),
205.85 (∆ν1/2 ) 895 Hz), 42.49 (∆ν1/2 ) 3132 Hz). Anal. Calcd
for C35H47N5Mo: C, 66.34; H, 7.48; N, 11.05. Found: C, 66.39;
H, 7.42; N, 11.03.
(benzene-d6): δ ) 587.90 (∆ν1/2 ) 382.27 Hz), 559.80 (∆ν1/2
)
688 Hz), 194.84 (∆ν1/2 ) 327 Hz), 83.66 (∆ν1/2 ) 1024 Hz). Anal.
Calcd for C19H31N5Cr: C, 59.82; H, 8.19; N, 18.36. Found: C,
60.20; H, 8.22; N, 17.98.
Preparation of Mo(NBut)2(dpma) (2). A solution of Li2dpma
(0.2014 g, 1.00 mmol) in a mixture of 5 mL of toluene and 1 mL
of ether was cooled in a liquid nitrogen temperature cold well to
near frozen. This solution was added to a cold, stirring solution of
Cl2Mo(NBut)2(dme) (0.4007 g, 1.00 mmol) in 5 mL of toluene.
The reaction solution was allowed to warm to box temperature and
stirred for 2 h. The solution was filtered, and volatiles were removed
in vacuo. The resulting brown solid was recrystallized from pentane,
giving 2 as a yellow solid in 38% yield (0.161 g, 0.378 mmol). 1H
NMR (toluene-d8): δ ) 7.09 (s, 2H, pyrrole-5-H), 6.51 (m, 2H,
pyrrole-4-H), 6.31 (m, 2H, pyrrole-3-H), 4.28 (d, J ) 12.7 Hz,
2H, N-CHH-pyrrole, anti to methyl), 3.46 (d, J ) 12.7 Hz, 2H,
N-CHH-pyrrole, syn to methyl), 2.17 (s, 3H, NCH3), 1.49 (s, 9H,
NC(CH3)3, linear), 1.21 (s, 9H, NC(CH3)3, bent). 13C NMR (toluene-
d8): δ ) 139.00 (pyrrole-1-C), 133.37 (pyrrole-2-C), 110.04
(pyrrole-3-C), 104.96 (pyrrole-4-C), 70.29 (NCMe3, linear), 69.51
(NCMe3, bent), 60.67 (CH2), 44.61 (NCH3), 31.69 (NC(CH3)3,
linear), 31.04 (NC(CH3)3, bent). 14N NMR (23 °C, benzene-d6): δ
) 472 (∆ν1/2 ) 248 Hz), 458 ppm (∆ν1/2 ) 1100 Hz), 198.30 (∆ν1/2
) 372 Hz), 73.75 (∆ν1/2 ) 1594 Hz). 14N NMR (59 °C, toluene-
d8): δ ) 427.03 (∆ν1/2 ) 273 Hz), 412.98 (∆ν1/2 ) 792 Hz), 206.76
(∆ν1/2 ) 432.76 Hz), 69.58 (∆ν1/2 ) 1981 Hz). Anal. Calcd for
C19H31N5Mo: C, 53.64; H, 7.34; N, 16.46. Found: C, 53.50; H,
7.23; N, 16.38. Mo(NBut)2Cl257 may also be used in the preparation
of 2 with similar results.
General Considerations for Single-Crystal X-ray Diffraction.
Single crystals of 1-4 were grown at -35 °C in an MBraun inert
atmosphere glovebox. All but a small portion of the mother liquor
was removed, and the crystals were removed from the glovebox in
a sealed vial. The crystals were rapidly coated in Paratone N and
mounted on a glass fiber. The mounted crystal was placed under a
cold stream of nitrogen from an Oxford “Cryostream” low-
temperature device. Data were collected on a Bruker-AXS, Inc.,
SMART CCD diffractometer utilizing a PC running Windows NT.
The data collection was done on a Bruker-AXS, Inc., 3-circle
goniometer (ø set to 54.78°). The source was a water-cooled Mo
X-ray tube (λ ) 0.71073 Å) operating at 50 kV/40 mA. A single-
crystal graphite monochromator selected the wavelength of light
prior to being collimated. The cell was determined using ω-θ scans
(-0.3° scan width) with 3 sets of 20 frames. The initial cell was
found by repeated least squares and Bravais lattice analysis. Full
data sets were collected using ω-θ scans in four runs. The fourth
run duplicates the first 50 frames of the first run to allow analysis
of peak intensity changes resulting from crystal degradation; no
correction was necessary for any of the structures reported.
Absorption corrections were applied to the data. Using the initial
cell, data were integrated to hkl/intensity data using the Bruker-
AXS, Inc., program package SAINT. The final unit cell was
determined by SAINT using all the observed data. The structures
were solved and refined using the SHELXTL program developed
by G. M. Sheldrick and Bruker-AXS, Inc. A full listing of atomic
Preparation of W(NBut)2(dpma) (3). A solution of H2dpma
(0.1902 g, 1.00 mmol) in 5 mL of toluene was cooled to near frozen.
The cold solution of ligand was added to a cold solution of
W(NBut)2(NHBut)2 (0.4702 g, 1.00 mmol) in 5 of mL toluene. The
(57) Schoettel, G.; Kress, J.; Osborn, J. A. J. Chem. Soc., Chem. Commun.
1989, 1062-1063.
3616 Inorganic Chemistry, Vol. 43, No. 12, 2004

Transition Metal-Imido Bonding in M(NBut)2(dpma)
Table 3. Data from the Single Crystal X-ray Diffraction Experiments for Compounds 1-4
Cr(NBut)2-
Mo(NBut)2-
W(NBut)2-
Mo[N(2,6-Pri2C6H3)]2-
(dpma) (1)
(dpma) (2)
(dpma) (3)
(dpma) (4)
empirical formula
fw
C19H31CrN5
381.49
C19H31MoN5
425.43
C19H31N5W
513.34
C35H47MoN5
633.72
temp/K
293(2)
293(2)
293(2)
293(2)
crystal system
space group
a/Å
b/Å
c/Å
monoclinic
P2(1)/n
monoclinic
P2(1)/n
9.451(5)
10.175(5)
22.217(11)
93.569(9)
2132.4(17)
4
monoclinic
P2(1)/n
orthorhombic
Fdd2
19.415(5)
61.473(17)
11.280(3)
10.2007(2)
9.4695(3)
21.7312(5)
94.638(2)
2092.26(9)
4
9.4181(11)
10.2171(12)
22.190(3)
93.271(2)
2131.8(4)
4
â/deg
V/Å3
13463(6)
16
Z
d(calc)/Mg m-3
abs coeff/mm-1
θ range for data/deg
data/params
GOF on F2
R indices [I > 2σ(I)]
1.211
0.557
1.325
0.626
1.599
5.429
1.251
0.419
1.88-28.24
4936/227
0.929
R1 ) 0.0433,
wR2 ) 0.0961
R1 ) 0.0853,
wR2 ) 0.1071
1.84-23.32
3082/227
0.931
R1 ) 0.0723,
wR2 ) 0.1020
R1 ) 0.1697,
wR2 ) 0.1225
1.84-23.26
3074/227
1.090
R1 ) 0.0192,
wR2 ) 0.0466
R1 ) 0.0212,
wR2 ) 0.0473
1.32-23.33
4654/379
0.977
R1 ) 0.0431,
wR2 ) 0.0886
R1 ) 0.0673,
wR2 ) 0.0953
R indices (all data)
coordinates, bond lengths, bond angles, and thermal parameters for
all the structures has been deposited at the Cambridge Crystal-
lographic Data Centre and can be found in the Supporting
Information. Additional data pertaining to the collection and
processing of the four structures can be found in Table 3. In Table
American Chemical Society, Department of EnergysDefense
Programs, and Michigan State University. J.T.C. thanks MSU
for Brubaker and Dye Fellowships. The authors thank Seth
N. Brown for helpful comments on an early version of this
manuscript.
2
3, R1 ) ∑||Fo| - |Fc||/∑|Fo| and wR2 ) {∑w(Fo - Fc2)2/
2 2
∑w(Fo ) }1/2. A partial listing of geometrical parameters for all four
data sets may be found in Table 1 (see Figure 2 for labeling
scheme).
Supporting Information Available: Tables of data on the
X-ray diffraction studies. Full 14N NMR spectra for compounds
1-4. This material is available free of charge via the Internet at
http://pubs.acs.org.
Acknowledgment. The authors appreciate the financial
support of the Office of Naval Research (Young Investigator
Program), Petroleum Research Fund administered by the
IC049644O
Inorganic Chemistry, Vol. 43, No. 12, 2004 3617