Ta(V) complexes of a bulky amine tris(phenolate) ligand: Steric inhibition vs. chelate effect

Article abstract of DOI:10.1016/j.inoche.2004.06.003

The reactivity of a bulky amine tris(phenolate) ligand toward Ta(NMe ₂)₅ was investigated. The reaction leading to the final product LigTa(NMe₂)₂ was slow and proceeded via an unusual dinuclear complex. The X-ray structures of both complexes were solved. Significantly, although the reaction of Ta(OEt)₅ with that ligand precursor is slow as well, no dinuclear intermediate is observed during the reaction course.

Full text of DOI:10.1016/j.inoche.2004.06.003

Inorganic Chemistry Communications 7 (2004) 938–941
www.elsevier.com/locate/inoche
Ta(V) complexes of a bulky amine tris(phenolate) ligand:
steric inhibition vs. chelate effect
a
Stanislav Groysman , Sharon Segal , Israel Goldberg ,
a
a
a,
b,
*
Moshe Kol *, Zeev Goldschmidt
a
Raymond and Beverly Sackler Faculty of Exact Sciences, School of Chemistry, Tel Aviv University, Tel Aviv, 69978 Israel
b
Department of Chemistry, Bar-Ilan University, Ramat Gan, 52900 Israel
Received 9 May 2004; accepted 2 June 2004
Available online 2 July 2004
Abstract
The reactivity of a bulky amine tris(phenolate) ligand toward Ta(NMe2)5 was investigated. The reaction leading to the final
product LigTa(NMe2)2 was slow and proceeded via an unusual dinuclear complex. The X-ray structures of both complexes were
solved. Significantly, although the reaction of Ta(OEt)5 with that ligand precursor is slow as well, no dinuclear intermediate is
observed during the reaction course.
Ó 2004 Elsevier B.V. All rights reserved.
Keywords: Tantalum; Amine tris(phenolate) ligands; Chelating ligands; Dinuclear complex
Me
Me
Me
t-Bu
HO
t-Bu
t-Bu
A number of tetradentate trianionic ligands of tri-
podal geometry have been reported in the last two de-
cades. Noteworthy members are the triamido-amine
[N3N] ligands family, that has been extensively studied
[1,2], and the amine tris(phenolate) ligands family in-
troduced more recently to the early transition metals
realm. Whereas complexes of the former with a variety
of early transition metals were prepared, and their re-
activity potential explored [2,3], the chemistry of the
latter ligands has been investigated less thoroughly [4–8],
being focused mainly on Group IV metals [5,6]. Re-
cently, we and others have begun extending this chem-
istry to Group V metals [7,8].
OH HO
N
OH HO
N
Me
Me
t-Bu
t-Bu
HO
Me
t-Bu
Lig2 H3
Lig1 H3
One ligand of this family, bearing methyl groups on the
2, 4 positions of the phenolate rings (Lig1H3) was shown
to undergo a fast and nearly quantitative reaction with a
variety of Ta(V) metal precursors, such as Ta(NMe2)5
[8b], Ta(OEt)5 [8b], Ta(OMe)5 [7b] and Ta(CH2Ph)5 [8b].
The reactions ultimately led to the formation of octahe-
dral complexes of the Lig1TaX2 type (Fig. 1). Lig1H3 is
the only tripodal ligand of this framework reported thus
far in relation to Group V metals. As the atomic radius of
Ta(V) is smaller than its Group IV analogues, we were
interested to reveal how the steric pressure affect the
binding of this ligand family around tantalum. Therefore,
we decided to explore the reactivity of the sterically
* Corresponding authors. Tel.: +972-3-6407392; fax: +972-3-6409293
(M. Kol).
E-mail addresses: moshekol@post.tau.ac.il (M. Kol), goldz@mail.
biu.ac.il (Z. Goldschmidt).
1387-7003/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.inoche.2004.06.003

S. Groysman et al. / Inorganic Chemistry Communications 7 (2004) 938–941
939
Me
X
X
O
Me
M
N
Me
O
O
Me
Me
Me
Fig. 1. X ¼ OEt, OMe, NMe2, CH2Ph.
crowded analogue, Lig2H3, carrying bulky t-Bu substit-
uents in the ortho (para) positions of the phenolate rings.
This ligand was previously employed in Ti(IV) chemistry
[5,6b], however, the coordination chemistry of this ligand
with group V metals and, in particular with Ta(V) has not
been described before.
Fig. 2. ORTEP representation of B, 50% probability ellipsoids. Se-
ꢀ
lected bond lengths (A): Ta1–O2 1.933(3); Ta1–O3 1.944(3); Ta1–O4
2.029(3); Ta1–N5 2.002(4); Ta1–N6 2.019(3); Ta1–N7 2.387(3).
Our first attempt to prepare Ta(V) complexes of this
ligand relied on Ta(CH2Ph)5, a readily available orga-
nometallic precursor [8b]. However, even after long re-
action times, the products mixture consisted mainly of
the unreacted starting materials. Therefore, we turned to
the more traditional metal precursor, Ta(NMe2)5 [9]. As
previously reported, the reaction of Ta(NMe2)5 with the
less bulky ligand Lig1H3 proceeded smoothly when the
reagents were mixed in ether at RT, and after 2 h, no
starting materials were detected in the reaction mixture
[8b]. In contrast, Lig2H3 led to the formation of several
products. Under the same reaction conditions, the
products mixture consisted of three species: the de-
sired product A, the unreacted ligand precursor Lig2H3,
and an additional product B (Scheme 1). On the other
hand, no traces of Ta(NMe2)5 were observed among the
products. Employing longer reaction times led to a
higher portion of A in the products mixture and when
the reaction was allowed to proceed for three days, no
other products besides A were detected. Therefore, we
propose that for the sterically crowded ligand Lig2H3,
the formation of the expected mononuclear complex A
may take place via some sort of long-lived, relatively
stable intermediate B.
substituents, of the relative intensity 2:1. The dimethy-
lamido groups, lying on the mirror plane, appear as two
distinct signals. Four of the six benzyl protons appear as
an AB system, while the other two, being reflected by the
mirror plane, give rise to a sharp singlet. In addition, A
was characterized by means of X-ray crystallography
[11]. The crystal structure of A reveals a mononuclear
octahedral complex Lig2Ta(NMe2)2 (Fig. 2). As may be
expected, the amine tris(phenolate) ligand binds to the
metal in a tetradentate fashion and the two remaining
dimethylamido groups occupy mutually cis positions. We
have previously reported that the ‘‘labile’’ ligands X in
Lig1TaX2 (X ¼ OEt, NMe2) exhibit different reactivities,
attributed to the trans influence of the phenolate oxygen
vs. the central amine donor. This different trans influence
was expressed in different Ta–X bond lengths, i.e., Ta–
X
trans-amine < Ta–Xtrans-phenolate [8b]. In contrast, the Ta–X
bond lengths in A are very similar (2.002trans-amine vs.
2.019trans-phenolate), probably due to the steric pressure
imposed by three t-Bu groups inside the ‘‘phenolates
pocket’’. To minimize repulsive steric interaction, the top
dimethylamido unit is rotated to a conformation that
seems to reduce its ability for p-donation (for an ‘‘unro-
tated’’ dimethylamido group in B see below). A lesser
degree of p-donation by this dimethylamido nitrogen
(N5) in A may explain a relatively short trans Ta–N7
1
A was characterized by means of H and 13C NMR,
that indicate
a
regular Cs-symmetrical complex
1
Lig2Ta(NMe2)2 [10]. The H NMR spectrum of A ex-
hibits two groups of aryl protons and two groups of t-Bu
ꢀ
bond (2.387 A).
In contrast to A, the nature of B was much less ob-
vious. To identify the structure of this intermediate, we
isolated bright-yellow crystals of B by fractional crys-
tallization from pentane [12]. Its 1H NMR spectrum
displayed two groups of aromatic protons and two
groups of t-Bu substituents, each having a relative in-
tensity of 2:1. The NMe2 region was somewhat more
complicated, featuring four signals of relative intensities
of 4:1:1:1. Single crystals of B were obtained by crys-
tallization from cold pentane, and its X-ray structure
was solved [13,14], revealing an unusual dinuclear
Lig2H3 + Ta(NMe2)5
Lig2 Ta(NMe2)2 (A)
B + Lig2H3
Scheme 1.

940
S. Groysman et al. / Inorganic Chemistry Communications 7 (2004) 938–941
slow
Lig2H3 +Ta(OEt) 5
Lig2Ta(OEt)2
Scheme 2.
In addition to Ta(NMe2)5, Ta(OEt)5 may also serve
as a suitable metal precursor. The differences between
these two are the reduced basicity, as well as the reduced
steric bulk of the ethoxy ligand, in comparison to the
dimethylamino ligand. According to the NMR spectra,
the reaction of the ligand precursor with Ta(OEt)5 (in
ether at RT) led to the slow formation of the expected
mononuclear di(ethoxy) complex Lig2Ta(OEt)2 (C).
However, for this metal precursor, no dinuclear inter-
mediate was detected. Thus, after 2–3 h, the reaction
mixture consisted exclusively of the final product and
both starting materials, in the relative ratios of 2:1:1.
Carrying out the reaction for 2 d leads to a high-yield
formation of white C, contaminated by the traces of the
ligand precursor and the metal precursor (<10%). Pure
Lig2Ta(OEt)2 was obtained upon recrystallization from
pentane in ca. 50% yield [15], and is proposed to be
isostructural with A according to its spectroscopic data
scheme 2.
Fig. 3. ORTEP representation of B, 40% probability ellipsoids. The H
atoms, a disordered pentane molecule, and the disorder in t-Bu groups
ꢀ
are omitted for clarity. Selected bond lengths (A): Ta1–O3 1.947(5);
Ta1–O4 1.928(5); Ta1–N5 2.028(6); Ta1–N6 1.970(6); Ta1–N7
2.067(6); Ta1–N8 2.571(6); Ta2–O9 1.952(6); Ta2–N10 2.000(7); Ta2–
N11 2.027(6); Ta2–N12 2.026(7); Ta2–N13 1.952(6).
In conclusion, we have shown that an amine
tris(phenolate) ligand carrying bulky t-Bu substituents
can wrap around a Ta(V) center, leading to an octahe-
dral Lig2TaX2 (A) complex, provided that a suitable
precursor is employed (Ta(NMe2)5 or Ta(OEt)5). Even
with such a precursor, the reaction is slow and may go
through an intermediate dinuclear complex Lig2Ta2X7
(B, X ¼ NMe2), in which two Ta(V) centers are in dif-
ferent environments. When the ‘‘leaving’’ ligand X is less
easily protonated, and less bulky (OEt vs. NMe2), a
kinetic product analogous to the intermediate B is not
observed in the products mixture. We are currently in-
vestigating the reactivity of these species.
complex, in which two Ta(V) centers are bound to one
amine tris(phenolate) ligand (Fig. 3). The structure of B
may be roughly divided into two segments, intercon-
nected by the ligand backbone, designated as [Ta1] and
[Ta2]. The metal center in [Ta1] is hexa-coordinate, with
an [ONO] ligand fragment and three NMe2 groups each
bound in a mer fashion forming a nearly perfect octa-
hedral geometry. The remaining phenolate group binds
to the second, penta-coordinate [Ta2], along with four
NMe2 groups. The geometry of this segment is nearly
square-pyramidal, with the phenolate oxygen lying in
the basal plane, and one of the dimethylamido nitrogens
(N11) occupying the axial position. The phenolate ox-
ygens in the two segments exhibit a different binding
mode: the Ta–O–C bond angles in the [Ta1] segment are
substantially narrower (142–145°) than their counter-
part in the [Ta2] segment, which is almost linear (174°).
The orientation of the central NMe2 ligand in [Ta1] is
noteworthy, as it may shed light on the inertness of B.
To avoid overlapping, the Me groups of the ‘‘central’’
NMe2 ligand move as far as possible from the ortho t-Bu
substituents. Thus, one of the dimethylamido methyl
groups has to point directly toward the ortho-substituent
of the ‘‘incoming’’ (third) phenolate on the reaction
pathway leading to A. This repulsive interaction may be
responsible for the relative inertness of this intermediate
[14]. Interestingly, the dimethylamido nitrogen (N6)
seems to interact strongly with the metal via p-donation
Supplementary information
Crystallographic data for complexes A and B have
been deposited with the Cambridge Crystallographic
Data Centre as Supplementary Publications No. CCDC
238150 and No. CCDC 238151. Copies of the data can
be obtained free of charge on application to CCDC, 12
Union Road, Cambridge CB2 1EZ, UK (fax: (+44)
1223-336-033; e-mail: deposit@ccdc.cam.ac.uk).
Acknowledgements
This research was supported by the Israel Science
Foundation founded by the Israel Academy of Sciences
and Humanities.
ꢀ
(Ta1–N 1.971 A), thus lengthening the trans Ta–N bond
ꢀ
(2.571 A).

S. Groysman et al. / Inorganic Chemistry Communications 7 (2004) 938–941
941
After 2.5 h the solvent was evaporated, leading to 315 mg of a
crude yellow product. Recrystallization from pentane led to
yellow crystals that contain essentially pure B (75 mg, 18% yield).
1H NMR (C6D6, 200 MHz) d 8.24 (br s, 1H), 7.58 (br s, 1H), 7.55
(d, J ¼ 2:5 Hz, 2H), 6.92 (d, J ¼ 2:5 Hz, 2H), 4.89 (d, J ¼ 12:5
Hz, 2H), 4.44 (d, J ¼ 12:4 Hz, 2H), 3.85 (s, 6H, N(CH3)2), 3.45 (s,
6H, N(CH3)2), 3.37 (s, 6H, N(CH3)2), 3.17 (s, 2H), 2.98 (s, 24 H,
four N(CH3)2), 1.77 (s, 18H, t-Bu), 1.53 (s, 9H, t-Bu), 1.50 (s, 9H,
t-Bu), 1.24 (s, 18H, t-Bu).
References
[1] J.G. Verkade, Accounts Chem. Res. 26 (1993) 483.
[2] R.R. Schrock, Accounts Chem. Res. 30 (1997) 9.
[3] R. Kempe, Angew. Chem., Int. Ed. 39 (2000) 468.
[4] N.V. Timosheva, A. Chandrasekaran, R.O. Day, R.R. Holmes,
Organometallics 20 (2001) 2331.
[5] M. Kol, M. Shamis, I. Goldberg, Z. Goldshmidt, S. Alfi, E.
Hayut-Salant, Inorg. Chem. Commun. 4 (2001) 177.
[6] (a) Y. Kim, J.G. Verkade, Organometallics 21 (2002) 2395;
(b) S.D. Bull, M.G. Davidson, A.L. Johnson, D.E.J. Diane, M.F.
Mahon, Chem. Commun. 14 (2003) 1750.
[13] Crystal data for B: C64H120N8O3Ta, M ¼ 1411:58, triclinic, P
,
ꢀ1
ꢀ
a ¼ 13:7350ð3Þ, b ¼ 14:0700ð3Þ, c ¼ 20:7800ð7Þ A, a ¼
93:6480ð9Þ, b ¼ 104:5320ð9Þ, c ¼ 114:6211ð19Þ°, V ¼ 3468:74ð16Þ
3
A , Z ¼ 2, l ¼ 3:198 mmꢀ1, T ¼ 110 K. Final agreement factors
ꢀ
[7] (a) L. Michalczyk, S. de Gala, J.W. Bruno, Organometallics 20
(2001) 5547;
were: R1 ¼ 0:0614 for 9717 rflns with I > 2rðIÞ, and R1 ¼ 0:1225
for all 15,987 rflns.
(b) Y. Kim, P.N. Kapoor, J.G. Verkade, Inorg. Chem. 41 (2002)
4834.
[14] A somewhat similar structure of a Ta(V) complex with tris[(2-
indonyl)methyl]amine ligand was reported. That tetradentate [3)]
ligand coordinates to the metal in a tridentate [2)] fashion, leaving
[8] (a) S. Groysman, S. Segal, M. Shamis, I. Goldberg, M. Kol, Z.
Goldschmidt, E. Hayut-Salant, J. Chem. Soc. Dalton Trans.
(2002) 3425;
3
dimethylamido groups on the Ta. However, the fourth,
unbound, ligand arm was still protonated in that case. The
authors explained the structure by a reduced p-donating ability of
that ligand J.M. Tanski, G. Parkin, Inorg. Chem. 42 (2003) 264.
[15] 0.17 mmol of Ta(OEt)5 in ether were added to a stirred solution of
an equimolar amount of Lig2H3 (0.17 mmol in ether). The
reaction was stirred at RT for 2.3 h, after which time the solvent
was evaporated. The 1H NMR spectrum of the crude product
consisted of Lig2Ta(OEt)2, Ta(OEt)5, and Lig2H3, in relative
intensities of 2:1:1, thus indicating the formation of Lig2Ta(OEt)2
in ca. 50%. The off-white product was extracted with pentane (ca.
2 mL), filtered through Celite, and left in the freezer overnight.
White powder (72 mg) was separated from the pentane solution,
and dried in vacuo, leading to the pure Lig2Ta(OEt)2 (0.08 mmol,
47%). 1H NMR (C6D6, 200 MHz) d 7.60 (d, J ¼ 2:4 Hz, 1H), 7.52
(d, J ¼ 2:4 Hz, 2H), 6.93 (d, J ¼ 2:2 Hz, 2H), 6.81 (d, J ¼ 2:4 Hz,
1H), 4.90 (q, J ¼ 7:0 Hz, 2H), ca.3.6 (q + br s, J ¼ 7:0 Hz, 8H),
1.77 (s, 9H), 1.64 (s, 18H), 1.47 (t, J ¼ 7:0 Hz, 3H), 1.40 (s, 9H),
1.36 (s, 18H), 0.53 (t, J ¼ 7:0 Hz, 3H).
(b) S. Groysman, I. Goldberg, M. Kol, Z. Goldschmidt, Organo-
metallics 22 (2003) 3793.
[9] D.C. Bradley, I.M. Thomas, Can. J. Chemistry 40 (1962) 1355.
[10] Preparation of A: Evaporation of the pentane solution after
crystallization of B (see [12]) led to 186 mg (0.20 mmol, 64%) of
pure A. 1H NMR (C6D6, 200 MHz) d 7.48 (d, J ¼ 2:4 Hz, 2H),
7.19 (d, J ¼ 2:6 Hz, 1H), 6.98 (d, J ¼ 2:4 Hz, 2H), 6.27 (d, J ¼ 2:4
Hz, 1H), 4.00 (d, J ¼ 13:5 Hz, 2H, ArC(H)H), 3.91 (s, 6H,
N(CH3)2), 3.31 (s, 6H, N(CH3)2), 3.04 (s, 2H), 2.96 (d, J ¼ 13:6
Hz, 2H) 1.61 (s, 18H, C(CH3)3), 1.53 (s, 9H, C(CH3)3), 1.39 (s,
18H, C(CH3)3), 1.23 (s, 9H, C(CH3)3).
[11] Crystal data for A: C48H78N3O3Ta, M ¼ 938:09, monoclinic,
ꢀ
P21=c, a ¼ 15:8440ð2Þ, b ¼ 15:8290ð2Þ, c ¼ 19:9410ð2Þ A,
3
ꢀ1
ꢀ
b ¼ 102:9840ð5Þ°, V ¼ 4873:23ð10Þ A , Z ¼ 4, l ¼ 2:296 mm
,
T ¼ 110 K. Final agreement factors were: R1 ¼ 0:0370 for 9318
rflns with I > 2rðIÞ, and R1 ¼ 0:0529 for all 11,514 rflns.
[12] 212 mg (0.31 mmol) of Lig2H3 in ca. 5 mL of ether were added to
the yellow ether solution of Ta(NMe2)5 (125 mg, 0.31 mmol).