The first example of μ2-imido functionality bound to a lanthanide metal center: X-ray crystal structure and DFT study of [(μ-ArN)Sm(μ-NHAr)(μ-Me)AlMe2]2 (Ar = 2,6-iPr2C6H3)

Article abstract of DOI:10.1021/om0206960

Reaction of 3 equiv of 2,6-diisopropylaniline with Sm[N(SiMe₃)₂]₃ affords the dimeric species [Sm(NHAr)₃]₂ (1). X-ray crystallography illustrates that each metal center in 1 engages in an η⁶-arene interaction with the aryl ring of an amide ligand attached to an adjacent samarium. IR spectroscopy indicates that the π-arene interactions are maintained in solution. Reaction of 1 with 4 equiv of trimethylaluminum leads to formation of the bis(μ-imido) complex [(μ-ArN)Sm(μ-NHAr)(μ-Me)AlMe₂]₂ (2). The molecular structure of 2 contains a unique central Sm₂N₂ core which displays extremely short bridging Sm-N distances of 2.152-(8) and 2.271(7) A?, characteristic of an imido complex. Density functional theory (DFT) calculations have been carried out in order to gain a better understanding of the nature of the bonding interactions within complex 2 and indicate that the 5d metal acceptor orbitals play a significant role in stabilizing π-donation from the imido groups to the samarium centers within the Sm₂N₂ core.

Full text of DOI:10.1021/om0206960

4726
Organometallics 2002, 21, 4726-4734
Th e F ir st Exa m p le of a µ2-Im id o F u n ction a lity Bou n d to
a La n th a n id e Meta l Cen ter : X-r a y Cr ysta l Str u ctu r e a n d
DF T Stu d y of [(µ-Ar N)Sm (µ-NHAr )(µ-Me)AlMe2]2 (Ar )
2,6-iP r 2C6H3)1
J ohn C. Gordon,*,2a Garth R. Giesbrecht,2b David L. Clark,2c P. J effrey Hay,2d
D. Webster Keogh,2e Rinaldo Poli,3 Brian L. Scott,2a and J ohn G. Watkin2a
Chemistry (C) Division, Nuclear Materials Technology (NMT) Division,
Theoretical (T) Division, and the Glenn T. Seaborg Institute for Transactinium Science,
Los Alamos National Laboratory, Los Alamos, New Mexico 87545, and Laboratoire de
Synthe`se et d’Electrosynthe`se Organome`talliques (LSEO UMR 5632), Faculte` de Sciences
“Gabriel”, Universite` de Bourgogne, 6 boulevard Gabriel, 21000 Dijon, France
Received August 23, 2002
Reaction of 3 equiv of 2,6-diisopropylaniline with Sm[N(SiMe3)2]3 affords the dimeric species
[Sm(NHAr)3]2 (1). X-ray crystallography illustrates that each metal center in 1 engages in
an η6-arene interaction with the aryl ring of an amide ligand attached to an adjacent
samarium. IR spectroscopy indicates that the π-arene interactions are maintained in solution.
Reaction of 1 with 4 equiv of trimethylaluminum leads to formation of the bis(µ2-imido)
complex [(µ-ArN)Sm(µ-NHAr)(µ-Me)AlMe2]2 (2). The molecular structure of 2 contains a
unique central Sm2N2 core which displays extremely short bridging Sm-N distances of 2.152-
(8) and 2.271(7) Å, characteristic of an imido complex. Density functional theory (DFT)
calculations have been carried out in order to gain a better understanding of the nature of
the bonding interactions within complex 2 and indicate that the 5d metal acceptor orbitals
play a significant role in stabilizing π-donation from the imido groups to the samarium centers
within the Sm2N2 core.
In tr od u ction
Imido-containing complexes have been documented for
the majority of the transition metals, and also the
actinides.19-26 Notable by their absence, however, are
examples of terminal imido complexes of scandium,
yttrium, or the lanthanide elements. There have been
two previous reports of structurally characterized com-
plexes which contain imido ligands bridging multiple
lanthanide metal centers: (i) two µ3-phenylimido ligands
cap the tetranuclear core of the complex [Yb4(µ-η2:η2-
Ph2N2)4(µ3-NPh)2(THF)4],27 and (ii) a µ4-NH ligand lies
at the center of the tetranuclear complex [{(η5-µ2-C9H6-
SiMe2NH)Ln}2(µ3-Cl)(THF)]2(µ4-NH)(THF).28 Following
a report by Evans et al. that trialkylaluminum reagents
were capable of deprotonating anilido ligands,29 we
The chemistry of metal-nitrogen multiply bonded
complexes has witnessed a dramatic surge in interest
in recent years4-10 due to the ability of the MdN
functionality to undergo a wide range of reactivity,
including metathesis of imines, aldehydes, and carbo-
diimides, metallacycle formation with alkenes and
alkynes, dealkylation, and C-H bond activation.11-18
(1) This work was first reported at the 219th National Meeting of
the American Chemical Society, San Francisco, March 26-30, 2000,
Poster 223.
(2) Los Alamos National Laboratory: (a) C-SIC, Mail Stop J 514; (b)
NMT-DO, Mail Stop J 514; (c) G. T. Seaborg Institute, Mail Stop E500;
(d) T-12, Mail Stop B268; (e) C-SIC, Mail Stop G739.
(3) Universite` de Bourgogne.
(4) Mountford, P. Chem. Commun. 1997, 2127.
(5) Danopoulos, A. A.; Green, J . C.; Hursthouse, M. B. J . Organomet.
Chem. 1999, 591, 36.
(6) Schrock, R. R. Tetrahedron 1999, 55, 8141.
(7) Romao, C. C.; Kuehn, F. E.; Herrmann, W. A. Chem. Rev. 1997,
97, 3197.
(8) Che, C. M. Pure Appl. Chem. 1995, 67, 225.
(9) Wigley, D. E. Prog. Inorg. Chem. 1994, 42, 239.
(10) Nugent, W. A.; Mayer, J . M. Metal-Ligand Multiple Bonds;
Wiley-Interscience: New York, 1988.
(11) Zuckerman, R. L.; Krska, S. W.; Bergman, R. G. J . Am. Chem.
Soc. 2000, 122, 751.
(12) Wang, W. D.; Espenson, J . H. Organometallics 1999, 18, 5170.
(13) Birdwhistell, K. R.; Lanza, J .; Pasos, J . J . Organomet. Chem.
1999, 584, 200.
(14) Polse, J . L.; Andersen, R. A.; Bergman, R. G. J . Am. Chem. Soc.
1998, 120, 13405.
(15) Lee, S. Y.; Bergman, R. G. Tetrahedron 1995, 51, 4255.
(16) McGrane, P. L.; J ensen, M.; Livinghouse, T. J . Am. Chem. Soc.
1992, 114, 5459.
(17) Blake, R. E.; Antonelli, D. M.; Henling, L. M.; Schaefer, W. P.;
Hardcastle, K. I.; Bercaw, J . E. Organometallics 1998, 17, 718.
(18) Walsh, P. J .; Hollander, F. J .; Bergman, R. G. J . Am. Chem.
Soc. 1988, 110, 8729.
(19) Brennan, J . G.; Andersen, R. A. J . Am. Chem. Soc. 1985, 107,
514.
(20) Burns, C. J .; Smith, W. H.; Huffman, J . C.; Sattleberger, A. P.
J . Am. Chem. Soc. 1990, 112, 3237.
(21) Arney, D. S. J .; Bruck, M. A.; Huber, S. R.; Wigley, D. E. Inorg.
Chem. 1992, 31, 3749.
(22) Brown, D. R.; Denning, R. G.; J ones, R. H. Chem. Commun.
1994, 2601.
(23) Stewart, J . L.; Andersen, R. A. New J . Chem. 1995, 19, 587.
(24) Straub, T.; Frank, W.; Reiss, G. J .; Eisen, M. S. Dalton Trans.
1996, 2541.
(25) Peters, R. G.; Warner, B. P.; Burns, C. J . J . Am. Chem. Soc.
1999, 121, 5585.
(26) Diaconescu, P. L.; Arnold, P. L.; Baker, T. A.; Mindiola, D. J .;
Cummins, C. C. J . Am. Chem. Soc. 2000, 122, 6108.
(27) Trifonov, A. A.; Bochkarev, M. N.; Schumann, H.; Loebel, J .
Angew. Chem., Int. Ed. Engl. 1991, 30, 1149.
(28) Xie, Z.; Wang, S.; Yang, Q.; Mak, T. C. W. Organometallics 1999,
18, 1578.
10.1021/om0206960 CCC: $22.00 © 2002 American Chemical Society
Publication on Web 09/24/2002

Binding of a µ2-Imido Group to a Lanthanide Center
Organometallics, Vol. 21, No. 22, 2002 4727
chose to examine whether the use of a sterically
encumbered anilido ligand on a lanthanide metal center
would allow the trialkylaluminum-mediated formation
of an imido functionality. As a result, we describe here
the isolation and characterization of a dimeric sa-
marium amido-imido compound which provides clear
structural insight into the behavior of an imido func-
tionality on a lanthanide metal center.30
Resu lts a n d Discu ssion
31
Reaction of Sm[N(SiMe3)2]3 with 3 equiv of 2,6-
diisopropylaniline in toluene produces the brick red
complex [Sm(NHAr)3]2 (1; Ar ) 2,6-iPr2C6H3) in moder-
ate yield (eq 1). Compound 1 is only sparingly soluble
i
Figu r e 1. ORTEP view of [Sm(µ-NHC6H3 Pr2-2,6)(NHC6H3-
iPr2-2,6)2]2 (1) drawn with 30% probability ellipsoids.
Isopropyl methyl groups have been omitted for clarity.
Ta ble 1. Selected Bon d Dista n ces (Å) a n d An gles
i
i
(d eg) for [Sm (µ-NHC6H3 P r 2-2,6)(NHC6H3 P r 2-2,6)2]2
(1)
Sm-N1
2.241(8)
Sm1-N2
2.351(10)
2.596
3.018(10)
2.838(10)
3.013(9)
Sm1-N3
Sm1-C13
Sm1-C15
Sm1-C17
2.273(11)
3.094(10)
2.870(10)
2.881(10)
Sm1-arene(cent)
Sm1-C14
Sm1-C16
Sm1-C18
N1-Sm1-N2
92.6(3) N1-Sm1-N3
112.5(4)
117.3(3)
N1-Sm1-arene(cent) 109.5
N2-Sm1-arene(cent) 104.3
N2-Sm1-N3
N3-Sm1-arene(cent) 117.6
Ta ble 2. Selected Bon d Dista n ces (Å) a n d
i
An gles (d eg) for [(µ-NC6H3 P r 2-2,6)Sm -
i
(µ-NHC6H3 P r 2-2,6)(µ-Me)AlMe2]2 (2)
in toluene or hexanes. The poor solubility of 1 in
nonpolar solvents and its high reactivity toward donor
solvents, coupled with the paramagnetism of Sm(III),
did not facilitate characterization by NMR spectroscopy.
Slow evaporation of a saturated hexanes solution of 1
yielded orange plates that were suitable for X-ray
diffraction. The molecular structure of compound 1 is
presented in Figure 1. A list of relevant bond lengths
and angles is available in Table 1; complete details of
the structural analyses of compounds 1 and 2 are listed
in Table 3. Complex 1 possesses a π-arene-bridged
dimeric structure analogous to that reported for the
yttrium analogue; however, in that case, the X-ray
crystal structure was not of sufficient quality to allow
for a detailed discussion of bond lengths and angles.29
In the solid-state structure of 1, each samarium atom
is ligated by three nitrogen atoms. Additionally, each
metal center engages in an η6-arene interaction with
the aryl ring of an amide ligand attached to an adjacent
samarium. The three nitrogen atoms and the aryl ring
centroid support a tetrahedral geometry around the
metal atom, with the angles anchored by the samarium
atom ranging from 92.6(3) to 117.6°. The distances
between the metal center and the six carbon atoms of
the aryl ring range from 2.838(10) to 3.094(10) Å, with
Sm1-N1
Sm1-N2
Sm1-C25
Al1-C25
Al1-C27
2.271(7)
2.535(8)
2.649(14)
2.094(12)
1.971(11)
Sm1-N1*
Sm1-C1
N1-C1
2.152(8)
2.666(8)
1.387(12)
1.939(12)
Al1-C26
N1-Sm1-N2
N1-Sm1-C25
N1*-Sm1-C1
N2-Sm1-C1
C1-Sm1-C25
127.1(3)
110.8(4)
112.0(3)
95.8(3)
N1-Sm1-N1*
N1-Sm1-C1
N1*-Sm1-N2
N1*-Sm1-C25
N2-Sm1-C25
81.1(3)
31.3(3)
151.6(3)
99.6(4)
75.4(3)
109.1(4)
an average distance of 2.952(10) Å (Sm(1)-arene(cent)
) 2.596 Å). This average Ln-C bond distance is similar
to those reported for the related aryloxide species [Sm-
(O-2,6-iPr2C6H3)3]232 (2.986(8) and 3.016(8) Å for the two
independent molecules in the asymmetric unit) as well
as other known examples of compounds containing
trivalent 4f element-π-arene interactions (e.g. average
Ln-C distances of 2.89(3) and 2.91(6) Å have been
33
reported for (η6-C6Me6)Sm(AlCl4)3 and (η6-C6H6)Sm-
(AlCl4)3,34 respectively, while the intramolecular π-arene
interaction in Yb(O-2,6-Ph2-C6H3)3 exhibits an average
Ln-C distance of 2.978(6) Å35). The Ln-C distances
determined for compound 1 are significantly longer than
the Gd-C(av) distance of 2.630(4) Å found for the
zerovalent complex Gd(η6-t-Bu3C6H3)236 and are slightly
(29) Evans, W. J .; Ansari, M. A.; Ziller, J . W.; Khan, S. I. Inorg.
Chem. 1996, 35, 5435.
(30) After submission of this paper, the synthesis and characteriza-
tion of two related ytterbium imido complexes were reported. See:
Chan, H.-S.; Li, H.-W.; Xie, Z. Chem. Commun. 2002, 652.
(31) Bradley, D. C.; Ghotra, J . S.; Hart, F. A. J . Chem. Soc., Dalton
Trans. 1973, 1021.
(32) Barnhart, D. M.; Clark, D. L.; Gordon, J . C.; Huffman, J . C.;
Vincent, R. L.; Watkin, J . G.; Zwick, B. D. Inorg. Chem. 1994, 33, 3487.
(33) Cotton, F. A.; Schwotzer, W. J . Am. Chem. Soc. 1986, 108, 4657.
(34) Fan, B. C.; Shen, Q.; Lin, Y. H. J . Organomet. Chem. 1989,
377, 51.
(35) Deacon, G. B.; Nickel, S.; MacKinnon, P.; Tiekink, E. R. T. Aust.
J . Chem. 1990, 43, 1245.

4728 Organometallics, Vol. 21, No. 22, 2002
Gordon et al.
Ta ble 3. Cr ysta llogr a p h ic Da ta
1
2
formula
mol wt
temp, K
cryst syst
space group
cryst size, mm
a, Å
C
72H108N6Sm2
C54H88Al2N4Sm
1105.91
203(2)
monoclinic
C2/c
0.06 × 0.12 × 0.12
15.912(1)
18.635(1)
18.941(1)
90
1358.34
203(2)
monoclinic
P21/n
0.21 × 0.21 × 0.21
19.363(2)
9.4027(9)
20.063(2)
90
110.193(2)
90
3428.3(6)
2
b, Å
c, Å
R, deg
â, deg
90.709(2)
90
5616.0(7)
4
γ, deg
V, Å3
Z
D
calcd, g/mL
1.316
1.739
1412
1.308
2.133
2276
abs coeff, mm-1
F(000)
θ range, deg
total no of rflns
no. of indep rflns
GOF
1.26-23.33
1.7-23.4
10 114
7377
4711
1.047
3709
1.366
R1
0.0704
0.0581
wR2
0.0917
0.1396
longer than the distances found in {[PhP(CH2SiMe2-
NSiMe2CH2)2PPh]Ln}2{η6:η6-(C6H5)2} (Ln ) Y,37 Ho38),
in which the lanthanide centers interact with the phenyl
rings of a dianionic bridging biphenyl unit. The bonds
between samarium and the two nitrogens of the termi-
nal amido ligands in 1 (2.241(8) and 2.273(11) Å) are
slightly shorter than the Sm-Nbridging distance (2.351-
(10) Å) and are unremarkable for samarium-amido
bonds.39
The infrared spectrum of [Sm(NHAr)3]2 (KBr plates,
Nujol) exhibits two distinct ν(CdC) stretching modes
in the aromatic region (1590 and 1575 cm-1), consistent
with two different arene environments in the solid state.
Similar features have been observed in the solid-state
and solution infrared spectra of [Sm(O-2,6-iPr2C6H3)3]2
(1586 and 1572 cm-1).32 The solution IR spectrum of 1
(benzene solution) also displays two ν(CdC) stretching
modes in the aromatic region (1590 and 1572 cm-1),
suggestive of the fact that the solid-state η6-bridged
arene structure is maintained in solution. Although a
dimeric structure involving anilido ligands bridging
through nitrogen is also possible, the ν(CdC) stretching
frequencies in this case might be expected to be so
similar that they would not be resolved as two distinct
bands.40,41
Compound 2 is extremely soluble in hydrocarbon
solvents such as hexanes and hexamethyldisiloxane.
However, the similar solubility of the aluminum-
containing byproduct [Me2Al(NHAr)]2 hampers the iso-
lation of 2 in pure form. Multiple recrystallizations from
hexanes failed to separate complex 2 from the alumi-
1
num dimer. H NMR spectroscopy revealed a series of
sharp resonances assignable to the aluminum dimer and
several broad, unassignable resonances, presumably due
to complex 2. Attempts to purify 2 by extraction into
hexamethyldisiloxane resulted in a deep red solution
and a colorless precipitate of [Me2Al(NHAr)]2, which was
removed by filtration. Removal of the solvent and
reexamination by NMR spectroscopy still revealed the
presence of the aluminum complex. Multiple iterations
of the extraction/filtration procedure proved unsuccess-
ful in the isolation of pure 2. At this point, we suspected
that the residual aluminum complex was being gener-
ated in solution as a result of the decomposition of 2.
We have recently observed similar redistribution reac-
tions in related 4f-element complexes supported by
phenoxide ligation.42 However, thermolysis experiments
did not indicate any significant decomposition, suggest-
ing that difficulties in obtaining pure samples of 2 are
due to the similar solubility properties of [Me2Al-
(NHAr)]2 and [(µ-ArN)Sm(µ-NHAr)(µ-Me)AlMe2]2 (2).43
Red crystals of 2 that were suitable for an X-ray
structure determination were obtained from hexanes.
An ORTEP representation of compound 2 is available
in Figure 2; selected bond lengths and angles are
presented in Table 2. The X-ray diffraction study of 2
revealed a dimeric samarium amido-imido complex,
lying on a crystallographic inversion center, in which
the two samarium metal centers are asymmetrically
Treatment of 1 with 4 equiv of trimethylaluminum
in toluene, followed by crystallization from hexanes,
leads to isolation of the trialkylaluminum adduct [(µ-
ArN)Sm(µ-NHAr)(µ-Me)AlMe2]2 (2) in moderate yield
(eq 2).
(36) Brennan, J . G.; Cloke, F. G. N.; Sameh, A. A.; Zalkin, A. J . J .
Chem. Soc., Chem. Commun. 1987, 1668.
(37) Fryzuk, M. D.; Love, J . B.; Rettig, S. J . J . Am. Chem. Soc. 1997,
119, 9071.
(38) Fryzuk, M. D.; J afarpour, L.; Kerton, F. M.; Love, J . B.; Patrick,
B. O.; Rettig, S. J . Organometallics 2001, 20, 1387.
(39) Minhas, R. K.; Ma, Y.; Song, J . I.; Gambarotta, S. Inorg. Chem.
1996, 35, 1866.
(40) Evans, W. J .; Olofson, J . M.; Ziller, J . W. Inorg. Chem. 1989,
28, 4308.
(41) We have also synthesized and characterized the lanthanum
derivative [La(NHAr)3]2; X-ray crystallography shows the same π-arene-
bridged structure in the solid state. 1H and 13C{1H} NMR spectroscopy
are consistent with a dimeric structure being maintained in solution.
Giesbrecht, G. R.; Gordon, J . C.; Clark, D. L.; Scott, B. L. Unpublished
results.
(42) Giesbrecht, G. R.; Gordon, J . C.; Brady, J . T.; Clark, D. L.;
Keogh, D. W.; Michalczyk, R.; Scott, B. L.; Watkin, J . G. Eur. J . Inorg.
Chem. 2002, 723.
(43) Attempts to prepare the lanthanum analogue [(µ-ArN)La(µ-
NHAr)(µ-Me)AlMe2]2 by the reaction of [La(NHAr)3]2 with trimethyl-
aluminum were not successful.

Binding of a µ2-Imido Group to a Lanthanide Center
Organometallics, Vol. 21, No. 22, 2002 4729
i
Figu r e 2. ORTEP view of [(µ-NC6H3 Pr2-2,6)Sm(µ-NHC6H3-
iPr2-2,6)(µ-Me)AlMe2]2 (2) drawn with 30% probability
ellipsoids. Isopropyl methyl groups have been omitted for
clarity.
bridged by two µ2-imido ligands. Each metal center also
bears a terminal -NHAr ligand, the nitrogen lone pair
of which is coordinated to a trimethylaluminum mol-
ecule. Sm-N bond lengths within the central Sm2N2
core are unusually short, as expected for bridging imido
(NR2-) as opposed to bridging amido (NR2-) linkages.
The two unique Sm-N distances within the Sm2N2 core
are 2.152(8) and 2.271(7) Å. In comparison, typical
Sm-N distances for bridging amido ligands are much
longer, as exemplified by the distances of 2.468(7) and
2.614(7) Å in [K(THF)6]2[Sm(µ-NHAr)(NHAr)3]2 (Ar )
2,6-Me2C6H3).29 In fact, the bridging Sm-N distances
in 2 are comparable to, if not shorter than, those found
for typical terminal samarium amido ligands: for
example, 2.284(7) Å in Sm[N(SiMe3)2]3,44 2.331(3) Å in
(η5-C5Me5)2Sm(NHPh)(THF),45 and 2.218(5) and 2.202-
(5) Å in {(iPr2N)2SmCl3[Li(TMEDA)]2}.39
The asymmetry of the Sm2N2 core is further com-
pounded by the presence of a significant interaction
between the ipso carbon of the bridging µ-NAr ligand
and one of the samarium metal centers (Sm(1)-C(1) )
2.666(8) Å). This results in an acute Sm(1)-N(1)-C(1)
angle of 90.2(5)° and a correspondingly obtuse Sm(1*)-
N(1)-C(1) angle of 168.1(8)°. The angles about each
bridging nitrogen atom add up to 357.2°, indicating the
near-planarity of the substituents and, hence, an almost
perfect T-shaped coordination geometry about nitrogen.
Similar interactions with phenyl rings have been ob-
served previously in bis(µ2-imido) complexes of ura-
nium23,26,46,47 and zirconium.21 In contrast to the very
short Sm-N bridging distances, the terminal amido
Sm(1)-N(2) distance is extremely long (2.535(8) Å),
presumably as a result of the nitrogen-coordinated
AlMe3 molecule preventing π-donation of the lone pair
to the metal center. We also note the structural similar-
ity of 2 to the bismuth amido-imido complex [Bi-
(NHAr)(µ-NAr)]2,48 although in the case of the bismuth
F igu r e 3. View of the fully optimized [(µ-PhN)Sm(µ-
NHPh)(µ-Me)AlMe2]2 structure from B3LYP calculations.
complex no interaction with the bridging phenyl ring
was observed.
Compound 2 also maintains a short Sm-C distance
(i.e. aluminum-bound methyl group) of 2.649(14) Å,
which is similar to the Sm-C(alkyl group) distances
previously determined for (ArO)Sm[µ-OAr)(µ-R)AlR2]2
(R ) Me, 2.620(5) and 2.632(5) Å;49 R ) Et, 2.627(4)
and 2.649(4) Å).42 This distance is slightly shorter than
typical Ln-C distances found in complexes containing
bridging methyl groups50-53 and suggests the presence
of an agostic interaction between the aluminum methyl
group(s) and the metal center. There is also a distinct
elongation of the Al-C bond associated with the “ago-
stic” methyl group (Al(1)-C(25) ) 2.094(12) Å vs Al(1)-
C(26) ) 1.939(12) Å and Al(1)-C(27) ) 1.971(11) Å).
DFT calculations, however, suggest that this interaction
may be more accurately described as a bridging methyl
group, on the basis of electronic considerations (see
later).
Th eor etica l Stu d ies on Com p lex 2
Str u ctu r al Resu lts u sin g Realistic Ligan ds. Since,
to the best of our knowledge, complex 2 represents the
first structurally characterized example of a discrete
imido complex of a 4f element, we were interested in
obtaining a more detailed picture of the bonding inter-
actions that stabilize the Sm2N2 core within this mol-
ecule. DFT calculations employing the B3LYP functional
were carried out on the dimeric model species [(µ-PhN)-
Sm(µ-NHPh)(µ-Me)AlMe2]2 (Figure 3), in which the
isopropyl substituents on the bridging and terminal
phenyl groups were replaced by hydrogen atoms. The
(49) Gordon, J . C.; Giesbrecht, G. R.; Brady, J . T.; Clark, D. L.;
Keogh, D. W.; Scott, B. L.; Watkin, J . G. Organometallics 2002, 21,
127.
(50) Evans, W. J .; Chamberlain, L. R.; Ulibarri, T. A.; Ziller, J . W.
J . Am. Chem. Soc. 1988, 110, 6423.
(44) Brady, E. D.; Keogh, D. W.; Gordon, J . C.; Hay, P. J .; Poli, R.;
Scott, B. L.; Watkin, J . G. Unpublished results.
(45) Evans, W. J .; Kociok-Kohn, G.; Leong, V. S.; Ziller, J . W. Inorg.
Chem. 1992, 31, 3592.
(46) Brennan, J . G.; Andersen, R. A.; Zalkin, A. J . Am. Chem. Soc.
1988, 110, 4554.
(47) Schnabel, R. C.; Scott, B. L.; Smith, W. H.; Burns, C. J . J .
Organomet. Chem. 1999, 591, 14.
(51) Evans, W. J .; Boyle, T. J .; Ziller, J . W. J . Am. Chem. Soc. 1993,
115, 5084.
(52) Evans, W. J .; Boyle, T. J .; Ziller, J . W. J . Organomet. Chem.
1993, 462, 141.
(48) Wirringa, U.; Roesky, H. W.; Noltemeyer, M.; Schmidt, H. G.
Inorg. Chem. 1994, 33, 4607.
(53) Biagini, P.; Lugli, G.; Abis, L.; Millini, R. J . Organomet. Chem.
1994, C16.

4730 Organometallics, Vol. 21, No. 22, 2002
Gordon et al.
Ta ble 4. Com p a r ison of Ca lcu la ted a n d
Exp er im en ta l Str u ctu r a l P a r a m eter s for
[(µ-Ar N)Sm (µ-NHAr )(µ-Me)AlMe2]2 (Ar )
2,6-iP r 2C6H3) a n d Rela ted Mod el Com p ou n d s
two equatorial H atoms. Thus, the µ-CH3 ligand is best
described as a bridging methyl group, similar to the
situation observed in Al2Me6, and not as an Al-CH3
group establishing agostic interactions with the Sm
atom. In other words, both C-Al and C-Sm interactions
are established with the same carbon electrons.
bond length (Å)
Ar
Sm-Nbr-R
substituent
SmdNbr Sm-Nbr diff Sm-Nt
(deg)
[(µ-ArN)Sm(µ-NHAr)(µ-Me)AlMe2]2 Species
To probe some of the factors affecting the multiple
bonding of the imido fragments within the dimer, three
additional sets of calculations were performed where
each structure was reoptimized to compare to the
original complex: (a) the AlMe3 groups were removed
entirely, (b) the original dimer was calculated without
5d basis functions on the Sm atoms, and (c) the full
dimer was maintained but the Ph substituents on the
bridging and terminal ligands were replaced by methyl
groups. These results are also summarized in Table 4.
Removing the Lewis acidic AlMe3 groups has surpris-
ingly little effect on the bond lengths in the bridging
Sm2N2 unit, with the bond alternation actually modestly
enhanced. The Sm-N terminal bond becomes 0.2 Å
shorter when the nitrogen is no longer complexed to the
aluminum center, as one would expect, since more
electron density is now available on the nitrogen for
bonding to the samarium center in the absence of the
Lewis acid. This result indicates that the trimethyl-
aluminum groups do not directly affect the electronic
structure of the Sm2N2 core but rather have a steric
function, with the -NH(AlMe3)Ph groups acting as
bidentate, uninegative ligands.
2,6-iPr2-
C6H3
Ph
Ph (no
AlMe3)
exptl
2.153
2.271
0.118 2.535
168
calcd
calcd
2.220
2.206
2.299
2.319
0.079 2.549
0.113 2.318
153
157
Ph (no Sm calcd
5d)
2.308
2.425
0.117 2.544
160
[(µ-MeN)Sm(µ-NHMe)(µ-Me)AlMe2]2 Species
calcd 2.247 2.253 0.006 2.458
Me
Ph
131
161
[(µ-PhNbr)Sm(µ-NH2)(µ-Me)AlH2]2 Species
calcd 2.183 2.346 0.163 2.440
structure was optimized using the basis sets and effec-
tive core potentials (ECPs) on Sm described in the
Experimental Section. The partially occupied 4f5 shell
was incorporated into the “large core” ECP, leaving the
5d and 6s orbitals as the most important valence
orbitals. Recent studies on Ln(NH2)3 and Ln[N(SiH3)2]3
(Ln ) La through Lu) model complexes have shown that
accurate results may be obtained without the inclusion
of the f electrons in the valence shell.54,55 We have also
found that “large core” calculations provide results
comparable to those of calculations which explicitly
included the 4f5 shell for larger complexes such as Ln-
[CH(SiMe3)2]3 (Ln ) La, Sm).56 These calculations also
effectively describe the agostic interactions present in
In contrast, replacing the N-Ph groups with N-Me
groups results in a nearly symmetric Sm-N-Sm link-
age, with the two Sm-N bond lengths differing by less
than 0.01 Å. The Sm-Nbr-C angle (153° in the case of
N-Ph) changes to 131° for the bridging N-Me group,
suggesting that π-overlap with the phenyl ring plays
an important part in maintaining the double-bond
character in the SmdN-Ph bridging bonds. These
results are consistent with those reported by Andersen
et al. for the uranium complexes (MeC5H4)4U2(µ-NR)2,
which exhibit symmetric (R ) SiMe3) or asymmetric
(R ) Ph) bridging imido groups depending on the nature
of the organic constituent.46 In that case, the asymmetry
in the structure of (MeC5H4)4U2(µ-NPh)2 was ascribed
largely to the contributions of an η3-azabenzylic reso-
nance structure in which negative charge was delocal-
ized on the ortho and para positions of the phenyl ring,
although the presence of C-N π-bonding was not
conclusively established.
56
molecules such the aforementioned Ln[CH(SiMe3)2]3
and Sm[N(SiMe3)2]3.57
Comparing some of the key geometrical parameters
between the optimized structure and the experimental
structure (Table 4), we see that the Sm-N bridging
bond lengths are calculated to be 2.220 Å (SmdN) and
2.299 Å (Sm-N) (differing by 0.08 Å), as compared to
the experimental values of 2.153 Å (SmdN) and 2.271
Å (Sm-N) (differing by 0.11 Å). Overall, the calculations
appear to be giving a reasonable description of the
multiple-bond character in the Sm-imido linkage. In
addition, the calculated bond length of the longer Sm-N
terminal bond (2.55 Å) agrees well with the observed
bond length (2.535 Å).
The results of the calculations illuminate the nature
of the Sm-(µ-CH3)-Al interactions, since the accurate
locations of the methyl hydrogen atoms are not available
from the experimental X-ray structure. The coordination
geometry at the bridging carbon atom can be described
as distorted trigonal bipyramidal, with the Sm atom and
one H atom occupying axial positions, while the Al atom
and the two remaining H atoms take up equatorial
positions (see Figure 3). The Al-CH3 moiety is therefore
arranged in a staggered conformation relative to the
Al‚‚‚Sm axis. In addition, relative to the Al-C axis, the
methyl group is tilted away from the Sm atom, as
evidenced by the smaller than tetrahedral Al-C-H
angle to the axial H atom and the larger ones to the
The net charge on the Sm centers from the population
analysis in the most realistic model is +1.14 (i.e. much
less ionic than the formal oxidation state of +3). There
is considerable donation of electron density into the 5d
orbitals, as evidenced by the overall d population of 1.28
electrons. The remaining valence population of 0.56
electron involves the 6s and 6p orbitals. The particular
importance of d orbital participation is illustrated in the
results of Table 4. In the absence of 5d functions on
samarium, a lengthening of 0.09 and 0.11 Å in the
calculated SmdN and Sm-N bond distances is ob-
served. This represents an overall increase in bond
distances of approximately 7% relative to the metric
data obtained by the X-ray study. These results show
that, in their role as π-acids, the 5d orbitals play an
important role in the stabilization of the SmdN frag-
(54) Maron, L.; Eisenstein, O. J . Phys. Chem. A 2000, 104, 7140.
(55) Maron, L.; Eisenstein, O. New J . Chem. 2001, 25, 255.
(56) Clark, D. L.; Gordon, J . C.; Hay, P. J .; Poli, R. Unpublished
results.
(57) Brady, E. D.; Clark, D. L.; Gordon, J . C.; Hay, P. J .; Keogh, D.
W.; Martin, R. L.; Poli, R. Unpublished results.

Binding of a µ2-Imido Group to a Lanthanide Center
Organometallics, Vol. 21, No. 22, 2002 4731
Ch a r t 1
ment (see later). The population analysis also suggests
that it is more accurate to view the bonds involving the
samarium center as covalent bonds, in contrast to the
commonly held view of bonds involving the lanthanides
as being predominantly ionic in nature.
Bon d in g An a lysis a n d Resu lts fr om Mod el Ca l-
cu la tion s. In considering the bonding in the Sm dimers
it is useful to consider the frontier MOs of the bridging
[N-Ph]2- ligand (Chart 1). The nitrogen atom can be
thought to have three electron pairs: the σ lone pair
(nσ) and two p lone pairs, one conjugated with the π
electrons of the phenyl group (π ) and one parallel to
the ring (π|). We would expect these MOs on each
bridging nitrogen atom to play an important role in the
Sm-N bonding. Indeed, we find the highest two oc-
cupied MOs correspond to the symmetric and antisym-
metric combinations of the N pπ orbitals interacting
with the Sm centers. Contour plots of these orbitals are
shown in Figure 4 (designated as MOs 147 and 148). A
bonding analysis of these MOs (Table 5) shows ap-
proximately 3% 5d orbital contribution from each Sm
center for the HOMO (MO 148) and about 8% 5d
contribution from each Sm center for the next highest
MO (MO 147). Therefore, while the compositions of
these MO’s are mainly ligand-based in nature, the
overall metal orbital contribution is essentially domi-
nated by the 5d orbitals on each metal center (roughly
75% of the metal-based contribution from Sma and Smb
is 5d in nature in MO 148, while in MO 147 it is closer
to 90%).
F igu r e 4. Contour plots (left) and ChemDraw representa-
tions (right) of the highest occupied molecular orbitals for
the [(µ-PhN)Sm(µ-NHPh)(µ-Me)AlMe2]2 model from DFT
calculations.
At slightly lower energies one finds the symmetric and
antisymmetric combinations of the N pπ| ligand orbitals
interacting with the Sm centers (MOs 143 and 146),
which also have significant 5d orbital participation (6
and 7%, respectively). Together these four MOs account
for 0.54 electron of the 1.28 5d electrons from the
population analysis. One finds orbitals involving the nσ
ligand combinations at much lower binding energies.
Finally, some of the MOs involved in the bonding of the
terminal NHPh groups to the Sm ions are also found,
denoted as Sm-Nt pπ orbitals in Table 5, at energies
similar to those of the orbitals described above.
Ta ble 5. An a lysis of Selected Or bita ls for
[(µ-P h N)Sm (µ-NHP h )(µ-Me)AlMe2]2
% s, p % d % s, p % d % total
MO ꢀ (au)
descripn
Sma
Sma
Smb Smb ligand
149 -0.049 Sm 6s, 5d
20.0 26.0 21.0 25.0
8.0
(LUMO)
148 -0.189 Sm-Nbr
π
1.0
3.3
1.0
3.2
91.5
(HOMO)
147 -0.206 Sm-Nbr
π
146 -0.215 Sm-Nbr π|
145 -0.220 Sm-Nt π
144 -0.225 Sm-Nt π
143 -0.234 Sm-Nbr π|
0.6
0.3
0.8
0.3
1.9
7.5
6.7
2.3
2.8
6.1
0.8
0.3
0.7
0.3
1.9
7.6
6.5
2.4
2.8
6.1
83.5
86.2
93.8
93.8
84.0
To probe the bonding interactions more clearly,
calculations on simpler model systems were performed
in which all amido and aluminum substituents (except
for the methyl group which bridges the aluminum and
samarium atoms) were replaced with hydrogen atoms.
The N-Ph bridging ligand was maintained. The opti-
mized structure of this simplified model yielded geo-
metrical parameters similar to those of the more
realistic models (Table 4), and the structures are
compared in Figure 5. The simpler model resulted in a
lengthening of the Sm-Cbridging methyl bond length from
2.71 to 2.87 Å and in a shortening of the Sm-Nt bond
length from 2.55 to 2.44 Å, leading to a shift of the Al
atoms to the opposite side of the Sm2(µ-N)2 plane, as
seen in Figure 5b. The configuration adopted by the

4732 Organometallics, Vol. 21, No. 22, 2002
Gordon et al.
Con clu sion s
Reaction of the π-arene-bridged anilido dimer [Sm-
(NHAr)3]2 (1; Ar ) 2,6-iPr2C6H3) with 4 equiv of AlMe3
leads to the formation of the bis(µ2-imido) complex [(µ-
ArN)Sm(µ-NHAr)(µ-Me)AlMe2]2 (2). To the best of our
knowledge this represents the first example of a struc-
turally characterized 4f element complex containing a
discrete imido functionality. Although the isolation of
compound 2 appears to be somewhat problematic, the
structural elucidation of this molecule clearly demon-
strates that the synthesis of species containing lan-
thanide main-group multiple bonds is feasible. DFT
calculations indicate that the metal (acceptor) 5d orbit-
als play a significant role in stabilizing π-donation from
the imido groups within the Sm2N2 core. We are
currently developing new approaches toward the isola-
tion of further novel examples of 4f element complexes
containing LndX multiple bonds (X ) CR2, NR, O). We
envisage that these systems will not only provide us
with new and interesting issues in structure and bond-
ing but may also provide some potentially reactive
synthons with which to carry out a variety of small-
molecule transformations.
F igu r e 5. Views of the fully optimized [(µ-PhN)Sm(µ-
NHR)(µ-Me)AlR′2]2 models: (a) R ) Ph, R′ ) Me; (b) R )
R′ ) H.
simpler model probably reflects the actual electronic
preference of the molecule, since the four bonds to each
samarium atom (namely the Sm-Nt, the Sm-CH3, and
the two Sm-Nb bonds) are closer to the ideal tetrahedral
configuration. The distortion observed in the experi-
mental structure as well in the optimized larger model
is attributable to the aryl-aryl repulsive interactions
between the terminal amido and the bridging imido
groups. This repulsion is also reflected in the longer
Sm-Nt bond for the larger model.
Exp er im en ta l Section
Gen er a l Con sid er a tion s. All manipulations were carried
out under an inert atmosphere of oxygen-free UHP grade
argon using standard Schlenk techniques or under oxygen-
free helium in a Vacuum Atmospheres glovebox. AlMe3 (2.0
M hexanes solution) was purchased from Aldrich and used as
received. 2,6-Diisopropylaniline was purchased from Aldrich,
dried over molecular sieves, and distilled prior to use. Sm-
[N(SiMe3)2]3 was prepared according to a literature proce-
dure.58 Hexane and toluene were deoxygenated by passage
through a column of supported copper redox catalyst (Cu-0226
S) and dried by passing through a second column of activated
alumina. Hexamethyldisiloxane was distilled over sodium
benzophenone and degassed prior to use. C6D6 was degassed,
dried over Na-K alloy, and trap-to-trap distilled before use.
To gain a better understanding of the factors which
favor asymmetric over symmetric bridging imido groups,
an MO analysis of the simplified structure, presented
in Figure 5b, was undertaken. Figure 6 compares the
results of the lower energy asymmetric structure (Figure
6a) with the symmetric structure (Figure 6b), where the
molecule was constrained in Cs symmetry. The two
highest occupied MOs in this model are the same Sm-N
pπ orbitals discussed previously in the full model (see
Figure 4, MOs 147 and 148). The next highest MOs are
the Sm-N pπ| orbitals, as in the previous picture, while
the MOs involving the terminal NHPh ligand, which has
been simplified to NH2, are not present. It is instructive
to compare the changes in orbital energies of the highest
MOs in Figure 6 for the two different structures,
although a rigorous correspondence with the total
energy of the molecule in DFT calculations is problem-
atic. One sees that the HOMO shows only a slight
change. The next occupied orbital is substantially lower
in energy in the asymmetric structure, where it not only
maintains a strong SmadNPh bond interaction but also
shows bonding interactions with Smb. Similarly a much
lower orbital (bottom of Figure 6) involving the C pπ
orbitals of the phenyl ring shows some Cipso-Smb
bonding character. Thus, it appears that when N-Ph
groups are employed, an asymmetric structure is fa-
vored due to (a) the overlap of the N pπ orbital with
the Sma 5d orbital and (b) the additional stabilization
afforded by the formation of a weak Cipso-Smb bond. In
the case of imido groups such as N-SiMe3, these
interactions are not possible and a symmetric structure
is observed.
1H NMR spectra were recorded on
a Bruker AMX 500
spectrometer at ambient temperature. Infrared spectra were
recorded on a Nicolet Avatar 360 FT-IR spectrometer as Nujol
mulls between KBr plates. Elemental analyses were performed
on a Perkin-Elmer 2400 CHN analyzer. Elemental analysis
samples were prepared and sealed in tin capsules in the
glovebox prior to combustion.
Sm (µ-NHC6H3iP r 2-2,6)(NHC6H3iP r 2-2,6)2]2 (1). 2,6-Diiso-
propylaniline (1.95 g, 11.1 mmol) in 10 mL of toluene was
added to a pale yellow toluene solution (50 mL) of Sm-
[N(SiMe3)2]3 (2.31 g, 3.66 mmol). This caused the solution to
turn red-orange. The reaction mixture was stirred at room
temperature for 15 min and then heated to reflux for 15 min,
during which time the solution turned deep red. When the
solution was cooled, a brick red solid precipitated. The solvent
was then removed under vacuum and the precipitate washed
with hexanes to remove HN(SiMe3)2. The resulting solid was
collected by filtration and pumped to dryness (1.94 g, 78%
yield). X-ray-quality crystals were grown by evaporation of
filtrate solutions of 1. The paramagnetism and low solubility
of 1 in noncoordinating solvents did not allow for characteriza-
tion by 1H or 13C{1H} NMR spectroscopy. IR (Nujol, cm-1):
1590 (m), 1575 (m), 1423 (s), 1358 (m), 1340 (m), 1324 (m),
1300 (m), 1250 (s), 1215 (m), 1193 (w), 1148 (w), 1138 (w), 1105
(m), 881 (m), 838 (m), 802 (s), 767 (s), 746 (s). Anal. Calcd for
(58) Evans, W. J .; Golden, R. E.; Ziller, J . W. Inorg. Chem. 1991,
30, 4963.

Binding of a µ2-Imido Group to a Lanthanide Center
Organometallics, Vol. 21, No. 22, 2002 4733
F igu r e 6. MO diagram and contour plots of selected orbitals for the model compound [(µ-PhN)Sm(µ-NH2)(µ-Me)AlH2]2:
(a) optimized tilted geometry; (b) partially optimized geometry in C2v symmetry.
C72H108N6Sm2: C, 63.66; H, 8.01; N, 6.19. Found: C, 63.02;
H, 8.25; N, 5.94.
argon flow. The crystals were immediately placed on a Bruker
P4/CCD/PC diffractometer and cooled to 203 K using a Bruker
LT-2 temperature device. The data were collected using a
sealed, graphite-monochromated Mo KR X-ray source. A
hemisphere of data was collected using a combination of æ and
ω scans, with 30 s frame exposures and 0.3° frame widths.
Data collection and initial indexing and cell refinement were
handled using SMART59 software. Frame integration and final
cell parameter calculations were carried out using SAINT60
software. The final cell parameters were determined using a
least-squares fit to 4402 reflections for 1. The data were
corrected for absorption using the SADABS61 program. Decay
of reflection intensity was not observed.
The structures were solved in the space groups P21/n for 1
and C2/c for 2 using direct methods and difference Fourier
techniques. The initial solutions revealed the samarium and
the majority of all non-hydrogen atom positions. The remaining
atomic positions were determined from subsequent Fourier
synthesis. The amide hydrogen atom positions were found on
the difference maps and refined with the isotropic temperature
factor set to 0.08 Å2. All other hydrogen atom positions were
idealized (C-H ) 0.98 Å for methine, 0.97 Å for methylene,
0.96 Å for methyl, and 0.93 Å for aromatic). The idealized
[(µ-N C 6 H 3 i P r 2 -2,6)S m (µ-N H C 6 H 3 i P r 2 -2,6)(µ-Me )Al-
Me2]2 (2). To an orange slurry of [Sm(NHAr)3]2 (0.403 g, 0.297
mmol) in toluene (50 mL) was added 0.60 mL of a 2.0 M
solution of AlMe3 in hexanes (1.20 mmol). Within a few
minutes the slurry dissolved and the solution became deep red.
Solvent was removed under vacuum to produce a dark brown
tacky solid. Hexanes were added (10 mL) with stirring and
the mixture pumped dry, again yielding a tacky solid. This
procedure was then repeated. Following this, hexanes (20 mL)
was added, the mixture filtered through a Celite pad, and the
resulting filtrate concentrated to approximately 10 mL. This
filtrate was then allowed to slowly evaporate under the
glovebox atmosphere. Once at dryness, an oily orange-red solid
resulted. This was washed with a minimal amount of cold
hexanes (-10 °C) and pumped to dryness (yield 0.170 g). A
0.100 g sample of this solid was redissolved in hexanes (5 mL)
and filtered through a Celite pad, and the resulting orange
solution was allowed to slowly evaporate under the glovebox
atmosphere. Once at dryness, an orange-red waxy crystalline
solid resulted. A single-crystal X-ray diffraction study was
carried out on a portion of this material, after which further
vacuum was pulled on the solid to ensure complete dryness
(yield 0.050 g). Anal. Calcd for C54H88Al2N4Sm2: C, 56.49; H,
7.73; N, 4.88. Found: C, 57.26; H, 8.32; N, 4.90.
(59) SMART, Version 4.210; Bruker Analytical X-ray Systems, Inc.,
Madison, WI 53719, 1996.
(60) SAINT, Version 4.05; Bruker Analytical X-ray Systems, Inc.,
Madison, WI 53719, 1996.
(61) Sheldrick, G. SADABS, first release; University of Gottingen,
Gottingen, Germany.
Cr ysta llogr a p h ic Stu d ies. Crystals of 1 and 2 were
attached to a glass fiber using a spot of silicone grease. The
crystals were mounted from a matrix of mineral oil under

4734 Organometallics, Vol. 21, No. 22, 2002
Gordon et al.
hydrogen atoms were refined using the riding model, with
isotropic temperature factors fixed to 1.2 times the equivalent
isotropic U value of the carbon atom they were bound to. The
final refinement62 included anisotropic temperature factors on
all non-hydrogen atoms. Structure solution, refinement, graph-
ics, and preparation of publication materials were performed
using SHELXTL NT.63 Additional details of data collection and
structure refinement are listed in Table 3.
Ack n ow led gm en t. We acknowledge support from
Laboratory Directed Research and Development at Los
Alamos and from the Office of Basic Energy Sciences,
Division of Chemical Sciences, U.S. Department of
Energy. Los Alamos National Laboratory is operated by
the University of California for the U.S. Department of
Energy under Contract W-7405-ENG-36.
Su p p or tin g In for m a tion Ava ila ble: Listings of frac-
tional atomic coordinates, bond lengths and angles, anisotropic
thermal parameters, and hydrogen atom coordinates for
complexes 1 and 2 and coordinates of the optimized model
structure [(µ-PhN)Sm(µ-NHPh)(µ-Me)AlMe2]2 from DFT cal-
culations. This material is available free of charge via the
Internet at http://pubs.acs.org.
Deta ils of Th eor etica l Ca lcu la tion s. All calculations
were carried out using the B3LYP functional64,65 by using a
“large core” effective potential66,67 on the samarium atom
including shells up to the filled 4s, 4p, and 4d shells in the
core. In addition, the partially filled 4f5 shell, corresponding
to the configuration of Sm(III), was also included in the core
for a total of 51 electrons. The remaining 11 electrons were
treated explicitly, which includes the “outer core” 5s2 5p6
electrons and the valence 6s2 5d1 electrons. A valence [5s, 4p,
3d] basis set was employed. 6-31G basis sets were used for C,
N, Al, and H atoms. All calculations were carried out with
Gaussian 98.68
OM0206960
(68) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; 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.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; 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.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J . L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J . A. Gaussian 98, revision A.9; Gaussian, Inc.: Pittsburgh, PA,
1998.
2
(62) R1
)
∑||Fo|
-
|Fc||/∑|Fo| and wR2
)
[∑[w(Fo
-
Fc2)2]/
∑[w(Fo2)2]]1/2; w ) 1/[σ2(Fo2) + (aP)2], where a ) 0.0135 and 0.061.
(63) SHELXTL, NT Version 5.10; Bruker Analytical X-ray Instru-
ments, Inc., Madison, WI 53719, 1997.
(64) Becke, A. D. J . Chem. Phys. 1993, 98, 5648.
(65) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(66) Dolg, M.; Stoll, H.; Savin, A.; Preuss, H. Theor. Chim. Acta 1989,
75, 173.
(67) Dolg, M.; Stoll, H.; Preuss, H. Theor. Chim. Acta 1993, 85, 441.