Cyclopentadienyl and imide ligand transfer from zirconium to iridium: Can early transition metal imido compounds be used as imide transfer reagents?

Article abstract of DOI:10.1021/om970827a

Instead of leading to imido group transfer, treatment of [Cp*Ir(thf)₃][OTf]₂ with Cp₂Zr(N^tBu)(thf) (Cp* = C₅Me₅; Cp = C₅H₅) gave cyclopentadienyl transfer from Zr to I

Full text of DOI:10.1021/om970827a

Organometallics 1998, 17, 433-437
433
Cyclop en ta d ien yl a n d Im id e Liga n d Tr a n sfer fr om
Zir con iu m to Ir id iu m : Ca n Ea r ly Tr a n sition Meta l Im id o
Com p ou n d s Be Used a s Im id e Tr a n sfer Rea gen ts?
Patrick L. Holland, Richard A. Andersen,* and Robert G. Bergman*
Department of Chemistry, University of California, Berkeley, California 94720
Received September 19, 1997
Instead of leading to imido group transfer, treatment of [Cp*Ir(thf)3][OTf]2 with
Cp2Zr(NtBu)(thf) (Cp* ) C5Me5; Cp ) C5H5) gave cyclopentadienyl transfer from Zr to Ir.
To characterize the product of this type of reaction more completely, reaction of the
corresponding ethyl-substituted compound [(C5Me4Et)Ir(thf)3][OTf]2 with the imidozirconium
complex was also carried out. This led to [CpEtIrCp][CpZr(NHtBu)(OTf)3(thf)], which has
been crystallographically characterized. An X-ray study of the known pentamethyliridoce-
nium salt [Cp*IrCp][BF4] was also performed. In contrast to the above chemistry, the
reaction of Cp*Ir-OCMe2CMe2O with Cp2Zr(NtBu)(thf) gives the heterobinuclear bis(imido)
complex Cp*Ir(µ-NtBu)2ZrCp2. Thus, in none of these reactions was transfer of the imido
group away from the zirconium center observed.
Terminal imido complexes of high-valent and early
transition metals are increasingly common, but very few
low-valent late transition metal terminal imido com-
plexes have been isolated.1,2 Presently, the most general
route to late transition metal terminal imido compounds
is the addition of 2 equiv of a lithium amide to a metal
dichloride. This method has been demonstrated by our
group for d6 Cp*Ir(NR)3 (Cp* ) C5Me5) and ArOs(NR)4
(Ar ) p-cymene, C6Me6) complexes and has recently
been extended to ArRu(NR) complexes by Burrell and
co-workers.5
iridium, these investigations have resulted in two
unusual ligand transfer reactions that are described
below.
Resu lts a n d Discu ssion
9,10
When a thf solution of [Cp*Ir(thf)3][OTf]2
was
treated with Cp2Zr(NtBu)(thf),7 there was an immediate
color change from yellow to a paler yellow. Analysis of
the products by H NMR spectroscopy always showed
1
singlets at δ 5.74 and 2.26 ppm, corresponding to the
formation of the [Cp*IrCp] cation, but the resonances
due to the Zr-containing products were variable.11 It
was possible, however, to extract the iridocenium cation
from these reaction mixtures by salt metathesis and
aqueous extraction to give pure [Cp*IrCp][BF4] (29%
yield) or [Cp*IrCp][BPh4] (56% yield).
In order to gain insight into the nature of the
products, we performed the analogous reaction of [CpEt-
Ir(thf)3][OTf]2 (CpEt ) C5Me4Et) with Cp2Zr(NtBu)(thf);
it was possible to grow a single crystal of a product from
this reaction. Figure 1 shows an ORTEP diagram of
the ion-pair product, [CpEtIrCp][CpZr(NHtBu)(OTf)3-
(thf)], which crystallized as a diethyl ether solvate. It
was possible to refine the structure adequately despite
a static disorder of the diethyl ether molecules, which
This synthetic strategy has two major disadvan-
tages: (a) 1 equiv of amine is released as a byproduct,
and this can cause purification problems in the case of
nonvolatile arylamines, and (b) the lithium amide is a
good reducing agent, which has rendered other imido
complexes of the late transition metals inaccessible.6 We
hoped that Cp2Zr(NR)(thf), which has been studied
extensively in these laboratories,7 would be a milder,
more general imide transfer reagent. As a first step in
this chemistry, we synthesized various Cp*IrX2 com-
plexes with electronegative X ligands (OTf, OR, Cl) and
treated these complexes with Cp2Zr(NtBu)(thf), hoping
to observe ligand transfer to form the known Cp*IrNt-
Bu3 and a Cp2ZrX2 byproduct.8 Although this strategy
was unsuccessful in transferring the imido ligand to
(8) This type of reaction has been successful for group transfers to
main-group elements, see: (a) Fagan, P. J .; Nugent, W. A. J . Am.
Chem. Soc. 1988, 110, 2310-2312. (b) Fagan, P. J .; Burns, E. G.;
Calabrese, J . C. J . Am. Chem. Soc. 1988, 110, 2979-2981. (c) Dufour,
P.; Dartiguenave, M.; Dartiguenave, Y. D.; Dubac, J . J . Organomet.
Chem. 1990, 384, 61-69. (d) Breen, T. L.; Stephan, D. W. Organome-
tallics 1997, 16, 365-369.
(1) Wigley, D. E. Prog. Inorg. Chem. 1994, 42, 239-482.
(2) Cundari, T. R. J . Am. Chem. Soc. 1992, 114, 7879-7888.
(3) Glueck, D. S.; Wu, J .; Hollander, F. J .; Bergman, R. G. J . Am.
Chem. Soc. 1991, 113, 2041-2054.
(4) Michelman, R. I.; Bergman, R. G.; Andersen, R. A. Organome-
tallics 1993, 12, 2741-2751.
(5) Burrell, A. K.; Steedman, A. J . Organometallics 1997, 16, 1203-
1208.
(9) White, C.; Thompson, S. J .; Maitlis, P. M. J . Chem. Soc., Dalton
Trans. 1977, 1654-1661.
(6) Attempts to make the Co and Rh analogues of the Ir imido
complex have resulted in an apparent reduction of the metal, and no
products could be isolated. Glueck, D. S.; Holland, P. L. Unpublished
observations.
(7) Walsh, P. J .; Hollander, F. J .; Bergman, R. G. Organometallics
1993, 12, 3705-3723.
(10) Maitlis, P. M. Acc. Chem. Res. 1978, 11, 301-307.
(11) We have been unable to reproducibly form a single product from
this reaction. We believe that the product variation is due to the
amount of trace water and/or excess silver reagent; silylation of the
Celite and glassware still gave quantitative Cp transfer but did not
give cleaner Zr-containing products.
S0276-7333(97)00827-3 CCC: $15.00 © 1998 American Chemical Society
Publication on Web 01/17/1998

434 Organometallics, Vol. 17, No. 3, 1998
Holland et al.
Ta ble 1. Im p or ta n t Dista n ces a n d An gles in
[Cp EtIr Cp ][Cp Zr (NHtBu )(OTf)3(th f)] ‚1/2Et2O
Distances (Å)
Ir-Cp (centroid)
Ir-C(Cp) (ave)
Ir-CpEt (centroid)
Ir-C(CpEt) (ave)
Zr-Cp
1.82
2.18(1)
1.80
2.16(1)
2.22
2.52(1)
Zr-O(1)
Zr-O(2)
Zr-O(3)
Zr-O(4)
Zr-N
2.379(8)
2.172(8)
2.254(9)
2.194(8)
2.01(1)
Zr-C(Cp) (ave)
Angles (deg)
Cp-Ir-CpEt
Cp-Zr-O(1)
Cp-Zr-O(2)
179.1
175.2
103.4
Cp-Zr-O(3)
Cp-Zr-O(4)
Cp-Zr-N
102.9
102.5
107.9
O(1)-Zr-O(2)
O(1)-Zr-O(3)
O(1)-Zr-O(4)
O(1)-Zr-N
O(2)-Zr-O(3)
O(2)-Zr-O(4)
75.6(3)
72.4(3)
77.2(3)
76.9(3)
81.1(3)
149.4(3)
O(2)-Zr-N
O(3)-Zr-O(4)
O(3)-Zr-N
O(4)-Zr-N
Zr-N-C(6)
twista
93.4(4)
77.5(3)
149.2(4)
93.8(4)
157.5(9)
8
a
See text for definition.
sandwich structure, as expected for an 18-electron
metallocene: the cyclopentadienyl ligands are parallel
(dihedral angle ) 3.2°), and the Cp(centroid)-Ir-Cp-
(centroid) angle is 179.1°. The anionic zirconium-
containing fragment (Figure 1b) has a four-legged piano-
stool structure, with Cp(centroid)-Zr-X angles of 102-
108°, capped by a tetrahydrofuran ligand trans to Cp.
The Zr-O(thf) distance is quite long, 2.379(8) Å; this is
similar to the distance of 2.393(3) Å between Zr and the
thf ligand trans to Cp in the closely related neutral
complex CpZrCl3(thf)2.12 The amido NH proton was
located in a Fourier map, revealing a trigonal planar
geometry at N: this geometry is indicative of the
expected π bonding between N and Zr.
Why did the cyclopentadienyl ligand transfer? In
terms of hard-soft acid-base theory, the most thermo-
dynamically stable situation results when hard ligands
bind to the early transition metal and soft ligands bind
to the late transition metal: this is exactly what has
happened.13 The iridium atom has attracted as many
soft ligands as possible, leaving the hard oxygen-
containing ligands for the zirconium atom. [Cp*Ir-
(solvent)3]2+ complexes are known to coordinate various
arenes9 and the cyclopentadienyl ligands of ferrocenes
and ruthenocenes,14 and exchange of cyclopentadienyl
ligands at zirconium has been observed,15,16 indicating
that low-barrier pathways for Cp ligand exchange are
accessible.
At the time this work was performed, we were
surprised to discover that no crystal structure of any
iridocene had been reported in the literature.17 Because
of the disorder in the structure described above, we
(12) Erker, G.; Sarter, C.; Albrecht, M.; Dehnicke, S.; Kru¨ger, C.;
Raabe, E.; Schlund, R.; Benn, R.; Rufinska, A.; Mynott, R. J . Orga-
nomet. Chem. 1990, 382, 89-102.
F igu r e 1. (a) ORTEP diagram of [CpEtIrCp][CpZr(NHtBu)-
(OTf)3(thf)]‚1/2Et2O, using 50% probability ellipsoids. The
diethyl ether solvate is not shown. (b) ORTEP diagram of
the [CpZr(NHtBu)(OTf)3(thf)] anion. For clarity, only the
oxygen atom of each triflate ligand is shown and only the
quaternary carbon of the tert-butyl group is shown.
(13) The reactions of [Cp*Rh(thf)3]2+ and [Cp*Co(thf)3]2+ gave
analogous metallocene products, as judged by 1H NMR spectroscopy.
(14) Herberich, G. E.; Englert, U.; Marken, F.; Hofmann, P. Orga-
nometallics 1993, 12, 4039-4045. In some cases, Cp transfer was
observed.
(15) Kondakov, D.; Negishi, E. Chem. Commun. 1996, 963-964.
(16) Peng, M. H.; Brubaker, C. H. J . Organomet. Chem. 1977, 135,
333-337.
lie in columnar channels along the a axis of the crystal
lattice, well-separated from the rest of the atoms. An
ORTEP diagram of the asymmetric unit is shown as
Figure 1a; important bond distances and angles are
given in Table 1. The iridocenium cation has a planar
(17) Very recently, a crystal structure of [Cp*2Ir]+[BPh4]- has been
reported, see: Struchkov, Y. T.; Antipin, M. Y.; Lyssenko, K. A.; Gusev,
O. V.; Peganova, T. A.; Ustynyuk, N. A. J . Organomet. Chem. 1997,
536-537, 281-284. They found a twist angle of 24° and attributed
the deviation from an eclipsed geometry to cation-anion steric
interactions.

Ligand Transfer from Zirconium to Iridium
Organometallics, Vol. 17, No. 3, 1998 435
F igu r e 3. ORTEP diagram of Cp*Ir(µ-NtBu)2ZrCp2, using
50% probability ellipsoids.
F igu r e 2. ORTEP diagram of one of the iridocene cations
in [Cp*IrCp][BF4], using 50% probability ellipsoids.
Ta ble 3. Im p or ta n t Dista n ces a n d An gles in
Cp *Ir (µ-NtBu )2Zr Cp 2
Ta ble 2. Im p or ta n t Dista n ces a n d An gles in
[Cp *Ir Cp ][BF 4]
Distances (Å)
Ir-Zr
2.9160(6)
1.87
2.37
Zr-Cp(2)
Ir-N
2.35
2.046(4)
2.008(4)
distance to centroid (Å) average Ir-C distance (Å)
Ir-Cp*
Zr-Cp(1)
Ir(1)-Cp
Ir(1)-Cp*
Ir(2)-Cp
Ir(2)-Cp*
1.82
1.80
1.82
1.80
2.187(5)
2.173(5)
2.185(6)
2.172(5)
Zr-N
Angles (deg)
85.6(2)
87.6(2)
N-Ir-N′
N-Zr-N′
Ir-N-Zr
92.0(2)
angles (deg)
Zr-Ir-Cp*
N-Ir-Cp*
Ir-Zr-Cp(1)
Ir-Zr-Cp(2)
167.0
137.0
109.8
130.5
N-Zr-Cp(1)
N-Zr-Cp(2)
Cp(1)-Zr-Cp(2)
110.3
112.2
119.7
Cp-Ir(1)-Cp*
179.9
179.3
9
5
Cp-Ir(2)-Cp*
twista (molecule 1)
twista (molecule 2)
a
(µ-NtBu)2ZrCp2 by X-ray crystallography, as shown in
Figure 3 and Table 3. Although our group has crystal-
lographically characterized the very similar complex
Cp*Ir(µ-NtBu)(µ-NPh)ZrCp2,20 the previous structure
had disorder problems, and so the extremely well-
behaved structure (R ) 1.6%) here allows better struc-
tural characterization of this type of heterobimetallic
compound. The four-membered Ir-Zr-N2 ring is puck-
ered (fold angle ) 17.6°), and the nitrogen atoms are
planar, indicating π-donation to the formally electroni-
cally unsaturated Ir and Zr centers. The Zr-Ir distance
is 0.3 Å longer than Zr-Ir single bonds21 and roughly
the same distance as in the several known Cp*Ir(µ-N-
tBu)(µ-X)ZrCp2 complexes,20-22 indicating that there is
little direct Zr-Ir interaction, as expected for Zr(IV)-
Ir(III) complexes. Thus, the puckering is probably steric
See text for definition.
decided to determine the structure of an iridocenium
salt with a more conventional anion. Therefore, a
crystal of [Cp*IrCp][BF4]18 was subjected to X-ray
crystallographic analysis; an ORTEP diagram showing
the two independent molecules in the asymmetric unit
is shown in Figure 2, and important bond distances and
angles are give in Table 2. As above, the Cp(centroid)-
Ir-Cp(centroid) angles are virtually linear (179.9°,
179.3°). Interestingly, in all three independent iridoce-
nium groups (from both crystal structures), the alky-
lated Cp ring is 0.02-0.03 Å closer to Ir than the
unsubstituted Cp ring; the reader is referred to a recent
paper for a more complete evaluation of this phenom-
enon.14 The twist angle, which describes whether the
rings are eclipsed (0°) or staggered (36°), is found to be
5-9° for all three iridocenium groups, showing for the
first time the preference for the eclipsed conformation
in iridium metallocenes.17
t
in origin, as we have observed in other cases where Bu
groups were present in a hindered four-membered
ring.23
Clearly, one Zr atom has completely transferred its
In another attempt to effect imido group transfer to
imido group, leaving [Cp2Zr-OCMe2CMe2O]n; we ob-
serve a white powder that presumably contains this
product. Two mechanistic possibilities for the formation
of the bis(imido) complex are shown in Scheme 1. In
path A, the desired transfer reaction takes place, giving
the Cp*Ir fragment, the pinacolate complex Cp*Ir-
OCMe2CMe2O19 was treated with 1 equiv of Cp2Zr(N-
tBu)(thf) in benzene-d6, giving 50% conversion to one
clean product (by 1H NMR spectroscopy). Addition of a
second equivalent of the zirconium imido complex gave
1H NMR spectra indicative of a virtually quantitative
yield of this product, which has been identified as Cp*Ir-
Cp*IrNtBu and Cp2Zr-OCMe2CMe2O. The iridium
(20) Hanna, T. A.; Baranger, A. M.; Bergman, R. G. Angew. Chem.,
Int. Ed. Engl. 1996, 35, 653-655.
(18) Although we used a different synthetic method, pentamethyl-
iridocenium salts have been isolated before, see: Moseley, K.; Kang,
J . W.; Maitlis, P. M. J . Chem. Soc. A 1970, 2875-2883.
(19) Glueck, D. S. Unpublished work. The study of this compound
and its reactivity are being pursued in our laboratories and will be
published in due course. Holland, A. W.; Glueck, D. S.; McCallum, J .
S.; Bergman, R. G. Work in progress.
(21) Baranger, A. M.; Bergman, R. G. J . Am. Chem. Soc. 1994, 116,
3822-3835.
(22) (a) Baranger, A. M.; Hanna, T. A.; Bergman, R. G. J . Am. Chem.
Soc. 1995, 117, 10041-10046. (b) Hanna, T. A.; Baranger, A. M.;
Bergman, R. G. Ibid., 11363-11364.
(23) Holland, P. L.; Andersen, R. A.; Bergman, R. G. J . Am. Chem.
Soc. 1996, 118, 1092-1104.

436 Organometallics, Vol. 17, No. 3, 1998
Holland et al.
by 1H NMR spectroscopy (THF-d8), the following resonances
were observed: δ 7.5-6.5 (variable number of Cp singlets),
5.74 (s, 5, Cp), 2.26 (s, 15, Cp*), 1.4-1.0 (variable number of
alkyl singlets).
Sch em e 1
Syn th esis of [Cp*Ir Cp][BF4] an d [Cp*Ir Cp][BP h 4] fr om
[Cp *Ir Cp ][Zr Ln ]. A solution of [Cp*IrCp][ZrLn] in 3 mL of
THF (synthesized as described above, using 31 mg of [Cp*IrCl2]2,
41 mg of AgOTf, and 30 mg of Cp2Zr(NtBu)(thf)) was treated
with NaBF4 (14 mg, 0.128 mmol), causing no visible change.
This was removed from the glovebox, and water (1 mL) was
added, causing an immediate loss of color. After addition of
additional water (5 mL), the solution was stirred for several
minutes and then extracted with CH2Cl2 (5 mL) in the air.
The CH2Cl2 extract was washed with 5 mL of brine and 5 mL
of NaHCO3 and dried over Na2SO4, and the solvent was
evaporated to give [Cp*IrCp][BF4] as a white solid that was
spectroscopically identical to a sample of [Cp*IrCp][BF4]
prepared as described below. Yield: 11 mg (29%). An
analogous procedure using 16 mg of [Cp*IrCl2]2, 21 mg of
AgOTf, 15 mg of Cp2Zr(NtBu)(thf), and 14 mg of NaBPh4 gave
[Cp*IrCp][BPh4] as a white solid that was spectroscopically
identical to a sample prepared as described below. Yield: 16
mg (56%).
Syn t h esis of [Cp *Ir Cp ][BF 4] fr om [Cp *Ir Cl2]2 a n d
Na Cp . In the glovebox, a mixture of [Cp*IrCl2]2 (316 mg,
0.397 mmol) and NaCp‚THF (144 mg, 0.899 mmol) in THF (8
mL) was stirred for 1 day at room temperature. The yellow
mixture was treated with NaBF4 (302 mg, 2.75 mmol) and
stirred for an additional day. The volatile materials were
removed in vacuo, and the yellow residue was extracted with
CH2Cl2 (10 mL), filtered, and concentrated to 3 mL. This
orange solution was layered with 3 mL of benzene and allowed
to stand at room temperature for 2 days, giving very pale
yellow needles of [Cp*IrCp]+[BF4]- (0.22 g, 58% yield). 1H
NMR (CD2Cl2): δ 5.53 (s, 5, Cp), 2.23 (s, 15, Cp*). 13C{1H}
NMR (CD2Cl2): δ 95.3 (C5Me5), 81.2 (Cp), 10.7 (C5Me5). IR
(KBr pellet): 3079 (s), 3038 (s), 2983 (m), 2921 (w), 2850 (w),
1606 (w), 1472 (s), 1455 (m), 1405 (s), 1385 (s), 1300 (w), 1050
(br s), 879 (m), 532 (w), 521 (w). Anal. Calcd for [Cp*IrCp]-
[BF4]: C, 37.59; H, 4.21. Found: C, 37.86; H, 4.09.
imido complex subsequently reacts with another equiva-
lent of Cp2Zr(NtBu)(thf) to generate the observed het-
erobimetallic species. Consistent with this mechanism,
a genuine sample of Cp*IrNtBu reacted with Cp2Zr(N-
tBu)(thf) in C6D6 to give Cp*Ir(µ-NtBu)2ZrCp2. How-
ever, the latter reaction proceeded on roughly the same
time scale as the reaction of Cp*Ir-OCMe2CMe2O and
Cp2Zr(NtBu)(thf) to give this product. If pathway A was
being followed and ktot ≈ k2, one would expect an
observable concentration of the intermediate Cp*IrN-
tBu to build up; however, no intermediates were de-
1
tected by H NMR spectroscopy. Therefore, the more
likely mechanism is path B, in which rate-determining
formation of an imido-pinacolate heterobimetallic in-
termediate is followed by rapid trapping by another Zr
imido complex to give the observed product.
Syn t h esis of [Cp *Ir Cp ][BP h 4] fr om [Cp *Ir Cl2]2 a n d
-
Na Cp . This was prepared in an analogous fashion to the BF4
salt, using 219 mg of [Cp*IrCl2]2, 103 mg of NaCp‚THF, and
193 mg of NaBPh4. The product did not crystallize well from
CH2Cl2/benzene, so it was recrystallized by cooling an acetone
solution to -80 °C. Yield: 224 mg (57%). 1H NMR (CD2Cl2):
δ 7.37 (m, 2, Ph), 7.06 (t, 2, J ) 7.4 Hz, Ph), 6.92 (t, 1, J ) 7.2
Hz, Ph), 5.14 (s, 5, Cp), 2.13 (s, 15, Cp*). 13C{1H} NMR (CD2-
Cl2): δ 164.5 (q, J BC ) 49 Hz, ipso Ph), 136.5 (q, J BC ) 1 Hz,
meta Ph), 126.2 (q, J BC ) 3 Hz, meta Ph), 122.3 (para Ph),
95.3 (C5Me5), 80.9 (Cp), 11.0 (C5Me5). IR (KBr pellet): 3108
(m), 3051 (s), 2982 (s), 2918 (w), 1579 (m), 1478 (s), 1453 (m),
1425 (m), 1408 (m), 1387 (m), 1265 (w), 1139 (w), 1034 (m),
999 (w), 855 (m), 737 (s), 707 (s), 611 (s). Anal. Calcd for
[Cp*IrCp][BPh4]: C, 65.81; H, 5.66. Found: C, 65.38; H, 5.65.
Exp er im en ta l Section
The general procedures used in these laboratories have been
described previously.23 [Cp*IrCl2]2 (and [CpEtIrCl2]2),24 Cp2Zr-
(NtBu)(thf),7 and Cp*IrNtBu3 were prepared by published
procedures. Silver trifluoromethanesulfonate (triflate), NaBF4,
and NaBPh4 were purchased from Aldrich and used as
received. General crystallographic protocols for operating the
SMART CCD/area detector system, using SAINT for data
reduction and TeXsan for refinement, have been described
previously.23,25
Typ ica l Syn th esis of [Cp *Ir Cp ][Zr Ln ]. In the glovebox,
THF (2 mL) was added to [Cp*IrCl2]2 (13.5 mg, 17 µmol) and
AgOTf (18.5 mg, 72 µmol). After 5 min of stirring, this mixture
turned yellow. It was filtered through Celite (1 cm), and a
solution of Cp2Zr(NtBu)(thf) (13.5 mg, 37 µmol) in THF (1 mL)
was added, causing an immediate color change from yellow to
a paler yellow. (If excess AgOTf had been used, a brown
precipitate was formed at this stage.) This solution could be
reduced to a solid in vacuo, or the anion could be exchanged
as described below. When the reaction was monitored in situ
Syn th esis of Cp *Ir (µ-NtBu )2Zr Cp 2. A solution of Cp*Ir-
OCMe2CMe2O (45 mg, 0.10 mmol) and Cp2Zr(NtBu)(thf) (76
mg, 0.21 mmol) in benzene (5 mL) was heated to 45 °C for 20
h. Over this time, the color changed from red to olive green.
The volatile materials were removed in vacuo, and the residue
was extracted with THF (15 mL). Filtration to remove the
white solid gave a green solution that was dried in vacuo, to
give a spectroscopically pure sample of Cp*Ir(µ-NtBu)2ZrCp2
(64 mg, 93% yield). This material could be crystallized by
vapor diffusion of pentane into an ether solution to give an
analytically pure material and crystals of X-ray quality. 1H
t
(24) Ball, R. G.; Graham, W. A. G.; Heinekey, D. M.; Hoyano, J . K.;
McMaster, A. D.; Mattson, B. M.; Michel, S. T. Inorg. Chem. 1990, 29,
2023-2025.
(25) Feig, A. L.; Bautista, M. T.; Lippard, S. J . Inorg. Chem. 1996,
35, 6892-6898.
NMR (C6D6): δ 5.89 (s, 10, Cp), 1.77 (s, 18, Bu), 1.32 (s, 15,
Cp*). 1H NMR (THF-d8): δ 5.59 (s, 10, Cp), 1.75 (s, 15, Cp*),
t
1.74 (s, 18, Bu). 13C {1H} NMR (C6D6): δ 108.8 (Cp), 88.8
(C5Me5), 67.6 (CMe3), 36.7 (CMe3), 12.6 (C5Me5). IR (Nujol):

Ligand Transfer from Zirconium to Iridium
Organometallics, Vol. 17, No. 3, 1998 437
1445 (sh), 1361 (m), 1346 (s), 1209 (w), 1182 (s), 1066 (w), 1030
(m), 1012 (s), 996 (m), 790 (s), 770 (s), 687 (m). Anal. Calcd
for Cp*Ir (µ-NtBu)2ZrCp2: C, 48.66; H, 6.27; N, 4.05. Found:
C, 48.69; H, 6.79; N, 3.92.
refinement, a model was used in which three fluorine atoms
of one BF4 group were rotationally disordered; these atoms
were included with one-half occupancy. Anisotropic refine-
ment of these atoms converged with reasonable anisotropic
thermal motion. The final cycle of full-matrix least-squares
refinement was based on 4617 observed reflections (I > 3σ(I))
and 406 variable parameters and converged with unweighted
and weighted agreement factors of R ) 2.24%, Rw ) 2.85%,
Syn th esis a n d X-r a y Cr ysta l Str u ctu r e of [Cp EtIr Cp ]-
[Cp Zr (NHtBu )(OTf)3(th f)]‚1/2Et2O. This compound was syn-
thesized by a method analogous to that for the Cp* analogue
above, using 95 mg of [CpEtIrCl2]2, 179 mg of AgOTf, and 91
mg of Cp2Zr(NtBu)(thf). The product, after removal of the
volatile materials, was extracted into diethyl ether; cooling this
solution gave colorless crystals, one of which was used for data
collection. 1H NMR analysis showed that the product did not
exclusively contain the crystallographically characterized
compound.
and GOF ) 1.28. Using all unique data: R ) 3.18%, Rw
)
3.13%.
X-r a y Cr ysta l Str u ctu r e of Cp *Ir (µ-NtBu )2Zr Cp 2. The
systematic absences and lattice symmetry did not distinguish
between several possible space groups; the choice of Abm2 was
confirmed by successful solution and refinement of the struc-
ture. A semiempirical ellipsoidal absorption correction (Tmax
) 0.862, Tmin ) 0.752) was applied using SADABS. Of the
5472 reflections that were measured, 1979 were unique (Rint
) 2.7%); equivalent reflections were averaged. The molecule
lies on a mirror plane which passes through Ir, Zr, C(1), C(11),
C(21), and C(24); the symmetry-related atoms within the
Inspection of the systematic absences and lattice symmetry
indicated space group P212121; the choice of this space group
was confirmed by successful solution and refinement of the
structure. A semiempirical ellipsoidal absorption correction
(Tmax ) 0.801, Tmin ) 0.507) was applied using SADABS. Of
the 18 963 reflections that were measured, 6446 were unique
(Rint ) 4.7%); equivalent reflections were averaged, but Friedel
mates were not averaged. The non-hydrogen atoms were
refined anisotropically, except for the diethyl ether atoms and
C(13), which became nonpositive-definite when refined aniso-
tropically. The positions of the diethyl ether atoms were
chosen to give reasonable bond lengths, and they were refined
with one-half occupancy, giving reasonable isotropic thermal
parameters on these atoms. Hydrogen atoms were included
in calculated positions, using fixed isotropic thermal param-
eters of approximately 1.2 times that of the atom to which they
were attached, except for the diethyl ether molecule, which
was modeled without hydrogens. The final cycle of full-matrix
least-squares refinement was based on 5043 observed reflec-
tions (I > 3σ(I)) and 529 variable parameters and converged
with unweighted and weighted agreement factors of R )
4.62%, Rw ) 5.14%, and GOF ) 1.55. Using all unique data:
R ) 6.10%, Rw ) 5.45%.
X-r a y Cr ysta l Str u ctu r e of [Cp *Ir Cp ][BF 4]. Recrystal-
lization of [Cp*IrCp][BF4] by slow evaporation of acetone gave
clear blocks. The systematic absences uniquely indicated
space group P21/c. A semiempirical ellipsoidal absorption
correction (Tmax ) 0.986, Tmin ) 0.723) was applied. Of the
15 338 reflections that were measured, 5876 were unique (Rint
) 2.7%); equivalent reflections were averaged. The non-
hydrogen atoms were refined anisotropically, and hydrogen
atoms were placed in calculated position with fixed isotropic
thermal parameters of approximately 1.2 times that of the
atom to which they were attached. During the final stages of
1
molecule are related by the operation (x, /2 - y, z). In initial
refinement, the non-hydrogen atoms were refined isotropically;
at this stage, the correct enantiomorph was determined by
comparing the R values for the correct structure (R ) 2.6%)
to the inverted structure (R ) 6.6%). The non-hydrogen atoms
were refined anisotropically, and hydrogen atoms were located
and their positions were refined using fixed isotropic thermal
parameters of approximately 1.2 times that of the atom to
which they were attached. The final cycle of full-matrix least-
squares refinement was based on 1864 observed reflections (I
> 3σ(I)) and 217 variable parameters and converged with
unweighted and weighted agreement factors of R ) 1.61%, Rw
) 2.00%, and GOF ) 0.77. Using all unique data: R ) 1.64%,
Rw ) 2.02%.
Ack n ow led gm en t. The authors thank the National
Institutes of Health (Grant No. GM-25459) for funding
and Dr. F. J . Hollander for help in determining the
crystal structures, especially that of Cp*Ir(µ-NtBu)2ZrCp2.
Su p p or tin g In for m a tion Ava ila ble: Tables of data col-
lection parameters, metrical data, positional parameters, and
anisotropic thermal parameters for the crystal structures and
an ORTEP diagram of the entire asymmetric unit of [Cp*IrCp]-
[BF4] (21 pages). Ordering information is given on any current
masthead page.
OM970827A