Structural diversity in tris(cyclopentadienyl) complexes of the Group 4 metals: Synthesis of Cp and MeCp complexes of Zr and Hf, and crystallographic characterization of (MeCp)3HfCl and (MeCp)4Zr (MeCp = C5H4CH3)

Article abstract of DOI:10.1016/j.poly.2013.03.007

Previous theoretical studies have predicted structural differences between the tris(cyclopentadienyl) zirconium and hafnium complexes Cp₃MX (Cp = C₅H₅;M = Zr, Hf). Cp₃ZrCl can be isolated via the stoichiometric reaction of Cp₂ZrCl₂ and NaCp, but forming the analogous Hf complex is complicated by ligand redistribution. Partial crystallographic analysis of these two complexes indicates likely structural differences wherein the Zr complex has three η⁵ Cp ligands ("3-5" structure) while the Hf complex has two η⁵ Cp ligands and one η¹ Cp ligand ("2-5,1-1" structure). The use of C₅H₄CH ₃ (MeCp) simplifies both the syntheses and the procurement of suitable crystals. The preparation of a number of MeCp complexes of Zr and Hf is reported along with crystallographic characterization of (MeCp) ₃HfCl, which has a 2-5,1-1 structure and (MeCp)₄Zr, which has a 3-5,1-1 structure.

Full text of DOI:10.1016/j.poly.2013.03.007

Polyhedron xxx (2013) xxx–xxx
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Structural diversity in tris(cyclopentadienyl) complexes of the Group 4 metals:
Synthesis of Cp and MeCp complexes of Zr and Hf, and crystallographic
characterization of (MeCp)3HfCl and (MeCp)4Zr (MeCp = C5H4CH3)
Erick J. Palmer a, Richard J. Strittmatter a, Keith T. Thornley a, Judith C. Gallucci a, Bruce E. Bursten a,b,
⇑
a Department of Chemistry and Biochemistry, The Ohio State University, Columbus, OH 43210, USA
b Department of Chemistry, The University of Tennessee-Knoxville, Knoxville, TN 37996, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Previous theoretical studies have predicted structural differences between the tris(cyclopentadienyl) zir-
conium and hafnium complexes Cp3MX (Cp = C5H5; M = Zr, Hf). Cp3ZrCl can be isolated via the stoichiom-
etric reaction of Cp2ZrCl2 and NaCp, but forming the analogous Hf complex is complicated by ligand
redistribution. Partial crystallographic analysis of these two complexes indicates likely structural differ-
Received 13 July 2012
Accepted 4 March 2013
Available online xxxx
ences wherein the Zr complex has three
g
5 Cp ligands (‘‘3-5’’ structure) while the Hf complex has two g5
Dedicated to the memory of Professor
Michelle M. Millar, wonderful friend and
colleague, who we miss dearly.
Cp ligands and one
g
1 Cp ligand (‘‘2-5,1-1’’ structure). The use of C5H4CH3 (MeCp) simplifies both the syn-
theses and the procurement of suitable crystals. The preparation of a number of MeCp complexes of Zr
and Hf is reported along with crystallographic characterization of (MeCp)3HfCl, which has a 2-5,1-1 struc-
ture and (MeCp)4Zr, which has a 3-5,1-1 structure.
Keywords:
Organometallics
Zirconium
Ó 2013 Elsevier Ltd. All rights reserved.
Hafnium
Tris-cyclopentadienyl
Hapticity
Synthesis
Crystallography
1. Introduction
combination of a2 symmetry (C3v point group). As first pointed
out by Moffitt, and expanded upon by Lauher and Hoffman, this
The coordination of cyclopentadienyl (Cp) ligands to transition
metals is arguably the cornerstone of modern organometallic
chemistry [1]. We have been interested in exploring the electronic
filled orbital is unable by symmetry to interact with any orbitals
of a transition metal but there is an available 5f acceptor orbital
for actinide elements [3].
structure of tris(
g
5-Cp) complexes of the actinide and transition
The first example of the tris(
metal system was afforded by the tetrakis-Cp complex
5-Cp)3Zr( 1-Cp), which has three Cp ligands bonded in pentaha-
g
5-Cp) ligand set in a transition-
elements, in part because of the remarkable contrast between
these elemental families: prior to the development of the rich
chemistry of polymethylated Cp ligands with the f-elements, the
(g
g
pto fashion and the fourth serving as a monohapto alkyl ligand [4].
We will refer to this coordination type as ‘‘3-5,1-1’’, indicating
three g5 ligands and one g1 ligand. Interestingly, both the Ti and
tris(
organoactinide chemistry [2], with examples spanning the early
and later actinide elements. In contrast, the tris(
5-Cp) bonding
g
5-Cp) structural motif was clearly the most common one in
g
Hf homologs show a different structure, namely (g g
5-Cp)2M( 1-Cp)2
mode is very rare in transition metal chemistry, exhibited only
by Zr. As discussed below, it also represents a rare instance in
which the chemistry of Zr and Hf are decidedly different.
(M = Ti, Hf), a ‘‘2-5,2-1’’ motif with only two of the Cp ligands
bound in pentahapto fashion [5]. We have previously explored
the structural chemistry of the tris-Cp chloro complexes Cp3MCl
(M = Zr, Hf) to see whether this structural difference persists in
the absence of a fourth Cp ligand [6]. The crystal structure of
Cp3ZrCl clearly indicates that it has a 3-5 structure (Scheme 1)
analogous to Cp4Zr. Attempts to obtain X-ray quality crystals of
Cp3HfCl were less successful and only a partial structure could be
obtained. That structure indicated a 2-5,1-1 structure (Scheme 1)
for the Hf derivative, seeming to indicate again the unusual struc-
tural differences between homologous Zr and Hf complexes.
The tris(g
5-Cp) ligand set also provides an uncommon elec-
tronic structural contrast between the transition and actinide
elements insofar as the three-fold symmetry of the ligands neces-
sarily generates a filled ligand-based symmetry-adapted linear
⇑
Corresponding author at: Department of Chemistry, University of Tennessee-
Knoxville, Knoxville, TN 37996, USA. Tel.: +1 865 9748 225.
E-mail address: bbursten@utk.edu (B.E. Bursten).
0277-5387/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2013.03.007
Please cite this article in press as: E.J. Palmer et al., Polyhedron (2013), http://dx.doi.org/10.1016/j.poly.2013.03.007

2
E.J. Palmer et al. / Polyhedron xxx (2013) xxx–xxx
high-vacuum line and drybox techniques [13]. Reactions on the
Schlenk and high-vacuum lines were performed under an atmo-
sphere of argon, Ar, which was purified by passage through consec-
utive columns of activated De-Ox catalyst (Johnson and Matthey)
and Drierite. All manipulations performed in the glovebox were
accomplished using a Braun MB150 inert-atmosphere drybox
equipped with a BASF R3-11 catalyst/Linde 13X molecular sieve
system. The drybox atmosphere was monitored using titanocene,
‘‘Cp2Ti’’, as an indicator. All solvents and reagents were purified
using standard purification techniques [14]. Prior to use, all
glassware was dried in an oven at 200 °C.
Scheme 1. The 3-5 (three
g g g
5-Cp ligands) and 2-5,1-1 (two 5- and one 1-Cp)
Elemental analyses (C and H) were obtained and were within
acceptable experimental guidelines. 1H and 13C NMR spectra were ob-
tained on a Bruker DPX-250 or DPX-400 MHz spectrometer controlled
with the proprietary X-WINNMR program. NMR spectroscopy of oxygen
and moisture sensitive compounds was performed using Wilmad
520-PP NMR tubes sealed by screw caps fitted with silicone septa.
Spectra were analyzed using the MESTREC NMR software [15]. Chemical
shifts are referenced to tetramethyl silane, TMS, and reported in ppm
relative to the residual 1H signal of the NMR solvent used.
All mass spectrometry experiments were performed at the Ohio
State University Campus Chemical Instrument Center for Mass
Spectrometry and Proteomics Facility. A Bruker Reflex III MALDI-
TOF spectrometer was used for data collection. The use of the nor-
mal chemical matrices was bypassed as it was shown that the
acidic nature of the matrices was detrimental to the moisture
and acid-sensitive compounds.
bonding motifs for Cp3MX complexes.
We have used relativistic density functional theory to explore
the preferred structures of Cp3MX complexes (M = Zr, Hf; X = H,
CH3, OR, NR2) [7]. On the basis of these calculations, we concluded
that the preferred structure (3-5 or 2-5,1-1) will depend on the
p-donor ability of the ligand X. Thus, we predict that the simple
hydride and methyl complexes should show the structural
differences between Zr and Hf noted above, namely that the Zr
complexes prefer a 3-5 structure whereas the Hf complexes prefer
the 2-5,1-1 motif. As the
we propose an ultimate preference for the 2-5,1-1 structure as the
competition between -donation from the X group and the Cp li-
p-donor ability of the X ligand increases,
p
gands increases. In particular, we found that Cp3ZrOH, Cp3ZrOMe,
Cp3ZrNH2, and Cp3ZrNMe2 should all prefer the 2-5,1-1 structure
over the 3-5 structure.
Single-crystal X-ray diffraction data were obtained using a Non-
ius Kappa CCD Diffractometer equipped with a sealed tube gener-
Experimental verification of these theoretical predictions has
been made difficult by the lack of good structural information on
Cp3MX and related tris(Cp) complexes. Not surprisingly, the struc-
tures of such complexes are highly dependent on the steric bulk of
the Cp ligand. Erker and coworkers have reported the structures of
Cp3ZrCH3 and [Cp3ZrCO][CH3B(C6F5)3], and both have 3-5
structures [8]. Although we have observed that Cp3ZrCl has a
3-5 structure [6], Lukens has reported that (TMS-Cp)3ZrCl
(TMS-Cp = C5H4SiMe3) has a 2-5,1-1 structure from X-ray crystal-
lography [9]; obviously the bulky TMS-Cp ligand could lead to int-
erligand repulsions in the 3-5 structure that are not present in
smaller Cp ligands. Etievant et al. have reported the preparation
of (MeCp)3ZrCl (MeCp = C5H4CH3), and proposed on the basis of
NMR data that it also has a 2-5,1-1 structure [10], although a crys-
tal structure was not obtained. Among trivalent complexes, Evans
et al. recently reported that (Me4Cp)3Sc (Me4Cp = C5Me4H) has a
2-5,1-1 structure [11], whereas Cp3Zr and Cp3Hf with unsubstitut-
ed Cp ligands both exhibit 3-5 structures [12].
We believed that some of the challenges in obtaining crystallo-
graphically suitable crystals of Cp3MX complexes could be circum-
vented by using substituted Cp ligands, although we wanted to
use ligands with less steric bulk than TMS-Cp and Me4Cp. MeCp
seemed particularly attractive insofar as it is not significantly larger
than the Cp ligand, and ring alkylation has been shown to improve
crystallization. In this contribution, we report efforts to synthesize
new tris-Cp complexes with both the Cp and the C5H4CH3 (MeCp)
ligand, and report the structures of two new complexes of the latter
ligand, namely (MeCp)4Zr and (MeCp)3HfCl. The structures of these
new complexes, which are 3-5,1-1 and 2-5,1-1 respectively, con-
tinue the pattern of structural differences between analogous Zr
ator with Mo K
a radiation (0.71073 Å). Experiments were run at
200 K using a cooled stream of nitrogen gas from an Oxford Cryo-
systems Cryostream Cooler.
2.2. Solvents and reagents
Hexanes, purchased from Fischer, were washed with concen-
trated sulfuric acid and distilled water, predried with magnesium
sulfate, MgSO4, and distilled over Na metal into a solvent collection
flask containing activated molecular sieves and sealed with a Kon-
tes stopcock. Petroleum ether, purchased from Fischer, was
washed with sulfuric acid and distilled water, predried with
MgSO4, and freshly distilled over diphosphorus pentoxide, P2O5.
Toluene, purchased from Mallinckrodt, was distilled over Na metal
and distilled into a solvent collection flask containing activated
molecular sieves and sealed with a Kontes stopcock. Tetrahydrofu-
ran, THF, purchased from Fischer, was dried over potassium/benzo-
phenone and distilled into a solvent collection flask containing
activated molecular sieves and sealed with a Kontes stopcock.
Methylene chloride, CH2Cl2, purchased from Fischer, was freshly
distilled over P2O5. Diethyl ether, purchased from Mallinckrodt,
was dried over K/benzophenone in a solvent storage flask and vac-
uum transferred as needed. All solvent collection flasks were
stored under an inert atmosphere in the drybox.
NMR solvents were purchased from Cambridge Isotope Labora-
tories and dried and purified using standard techniques. CDCl3 was
dried over P2O5, distilled into a solvent collection flask sealed with
a Kontes stopcock, degassed by three freeze–pump–thaw (FPT) cy-
cles, and stored in a drybox. The CDCl3 was passed through a col-
umn of basic aluminum immediately before use to remove any
acidic impurities. Benzene-d6 and toluene-d8 were dried over Na
metal, distilled into a solvent collection flask, degassed by three
FPT cycles, and stored in the drybox.
and Hf tris-Cp complexes with
r-only or weak p-donor X ligands.
2. Experimental
2.1. General experimental procedures
Cyclopentadiene and methylcyclopentadiene were purchased
from Sigma–Aldrich Chemicals as the dimers and cracked by distil-
lation through a Vigreaux column at 170 °C to form the monomer
[16]. Na metal was purchased from Sigma–Aldrich Chemicals as
All reactions and manipulations of air and moisture
sensitive compounds were performed using standard Schlenk line,
Please cite this article in press as: E.J. Palmer et al., Polyhedron (2013), http://dx.doi.org/10.1016/j.poly.2013.03.007

E.J. Palmer et al. / Polyhedron xxx (2013) xxx–xxx
3
solid cubes in mineral oil, which was removed with hexanes prior
to use. The titanocene and zirconocene dichloride compounds were
purchased from Sigma–Aldrich Chemicals and used as received.
ZrCl4 and HfCl4 (99.9% Hf) were purchased from Strem Chemicals
and used as received. Lithium dimethyl amide, sodium methoxide
and t-butoxide were purchased from Strem Chemicals and used
without further purification. Bromotrimethylsilane, stabilized by
Cu metal, was purchased from Sigma–Aldrich and was used as
received.
THF was removed in vacuo to yield a gray solid. The solid was col-
lected and transferred to a Soxhlet extractor. The Soxhlet extractor
was placed on the Schlenk line, and the Cp2HfCl2 was extracted
with boiling CH2Cl2 under an Ar atmosphere overnight. The CH2Cl2
solution was stirred over activated charcoal and dried over MgSO4.
The MgSO4 was filtered, and the CH2Cl2 was removed to yield an
off-white solid. The solid was transferred to the drybox and dis-
solved in a minimal amount of toluene. The toluene was placed
in the freezer at ꢀ35 °C overnight. 0.673 g (57%) of Cp2HfCl2 was
collected by filtration, rinsed with 10 mL hexane, and dried in va-
cuo. 1H NMR (C6D6): 5.80 (s).
2.3. Synthetic procedures
2.3.1. Na(C5H5) (NaCp)
2.3.4. Cp3ZrCl
This preparation was performed as a variation on a reported
procedure [17]. 100 mL of cyclopentadiene dimer was placed in a
closed system attached to the Schlenk line and purged with Ar
for 30 min. The Cp dimer was then degassed with three FPT cycles.
This purification step was shown to greatly improve the quality of
the NaCp formed in this reaction. The system was backfilled with
Ar, and the Cp2 was cracked by distillation at 170 °C through a Vig-
reaux column. The Cp monomer was kept in a dry ice/isopropanol
bath to minimize dimerization. About 30 g Na metal were cut into
small chunks and placed into a 250 mL Schlenk flask containing
150 mL of THF. The Cp monomer was then transferred via cannula
into the flask containing the Na metal under a positive flow of Ar.
Reaction immediately occurred as evidenced by the bubbling of
hydrogen gas, and the solution changed to orange, then brown,
and finally pink. The solution was stirred until no more bubbling
was observed. The THF was removed in vacuo to quantitatively
yield white NaCp and excess cyclopentadiene dimer. The flask
was transferred to the drybox, and 150 mL of hexane were added
to the flask. The solid NaCp was gathered by filtration, rinsed with
three 25 mL portions of hexane, and dried in vacuo.
In a drybox, 2.000 g (6.80 mmol) of Cp2ZrCl2 and 0.600 g
(6.80 mmol) of NaCp were placed in a 250 mL round bottom flask,
which was equipped with a Kontes valve, sealed with a Teflon
stopcock, and transferred to a high-vacuum line. The flask was
cooled in a liquid nitrogen bath for 15 min, and then 150 mL of
freshly distilled and degassed THF was transferred into the flask.
The reaction was slowly warmed to ꢀ78 °C and was stirred over-
night, allowing the solution to reach room temperature. The THF
was removed in vacuo to yield a bright yellow solid, and the flask
was transferred into the drybox. The solid was dissolved in 150 mL
and stirred for 4 h. The NaCl precipitate was filtered and rinsed
with three 10 mL portions of toluene. The toluene solution was
concentrated to half its original volume and placed in the freezer
at ꢀ35 °C for 2 days. Yellow crystals were collected by filtration,
rinsed with 10 mL of hexane, and dried in vacuo to yield 1.030 g
(47%) of Cp3ZrCl. 1H NMR (C6D6): 5.67 (s).
2.3.5. Cp3HfCl
0.500 g (1.30 mmol) of Cp2HfCl2 and 0.116 g (1.30 mmol) of
NaCp were mixed as solids in a 250 mL round bottom flask. The
flask was fitted with a Kontes valve, sealed with a Teflon stopcock,
and transferred to a high-vacuum line. The flask was cooled in a li-
quid nitrogen bath for 15 min. 100 mL of THF was condensed into
the flask, and the reaction vessel was slowly warmed to ꢀ78 °C
over several hours. The solution was stirred overnight and allowed
to warm to room temperature to produce a bright yellow solution.
The THF was removed in vacuo to yield a yellow solid. The flask
was transferred into the drybox. The yellow solid was dissolved
in 75 mL of toluene, and the NaCl precipitate was filtered. The vol-
ume of toluene was reduced by half and placed in the freezer at
ꢀ35 °C overnight. Yellow crystals of Cp3HfCl were collected by fil-
tration, rinsed with 10 mL of hexane, and dried in vacuo to yield
0.176 g (33%). 1H NMR (C6D6): 5.69 (s).
2.3.2. Na(C5H4CH3) (NaMeCp)
This preparation was performed with slight modifications to the
synthesis of NaCp. 150 mL of methylcyclopentadienyl, MeCpH, di-
mer was placed in a 250 mL round bottom flask with a magnetic
stir bar. The MeCpH dimer was distilled to remove cyclopentadi-
enyl impurities, and the first 20 mL of distillate was discarded.
The remaining MeCpH dimer was purged with Ar, degassed by
three FPT cycles, and distilled through a Vigreaux column to form
the MeCpH monomer. The MeCpH monomer is much more stable
to dimerization than the CpH monomer, and so storage of MeCpH
in a dry ice/isopropanol bath is unnecessary given the timescale of
this reaction. The MeCpH monomer was transferred via cannula
under a positive flow of Ar into a 250 mL Schlenk flask containing
sodium chunks in 75 mL of THF. Upon addition, the THF solution
turned orange and then became colorless. When no more dihydro-
gen gas evolution was observed, the THF was removed in vacuo to
quantitatively yield NaMeCp as a white solid and excess MeCpH di-
mer. The flask was transferred into the drybox, and 150 mL of hex-
ane were added. The white NaMeCp solid was broken up by
stirring and collected by filtration. The solid NaMeCp was rinsed
with three 25 mL portions of hexane and dried in vacuo. Solid
NaMeCp was stable to storage for over a year when stored under
an inert Ar atmosphere in the drybox.
2.3.6. (MeCp)2ZrCl2
This compound was prepared according to the standard proce-
dure reported in the literature with slight modifications [19]. In a
drybox, a 250 mL Schlenk flask was charged with 100 mL THF.
2.856 g (12.3 mmol) of ZrCl4 was slowly added to the flask with
stirring to form
a white suspension. A solution of 2.502 g
(24.6 mmol) of NaMeCp dissolved in 75 mL of THF was added
dropwise over a period of 30 min. Upon addition, the white sus-
pension progressively changed color from pale yellow to orange
to white again. The reaction was stirred for 2 days, and the THF
was removed in vacuo to yield a white solid. The flask was trans-
ferred out of the drybox and worked up under normal atmospheric
conditions. The white solid was dissolved in 100 mL of CH2Cl2 to
yield a brown solution, which was transferred to a separatory fun-
nel. The CH2Cl2 solution was rinsed with 100 mL of 4 M HCl. Upon
addition of the HCl, the CH2Cl2 solution turned green. This step was
repeated twice with 25 mL portions of HCl. The three HCl washes
were combined and rinsed with 30 mL of CH2Cl2. The two CH2Cl2
layers were combined, stirred over activated charcoal, and dried
2.3.3. Cp2HfCl2
The following preparation was performed using the standard
published procedure [18]. In the drybox, 1.000 g (3.1 mmol) of so-
lid HfCl4 (99.9% Hf) was added slowly to a 250 mL Schlenk flask
containing 125 mL of THF to form a white suspension. A solution
of 0.550 g (6.2 mmol) NaCp dissolved in 50 mL of THF was added
dropwise to the HfCl4 suspension. Upon complete addition, the
solution became cloudy and white and was stirred overnight. The
Please cite this article in press as: E.J. Palmer et al., Polyhedron (2013), http://dx.doi.org/10.1016/j.poly.2013.03.007

4
E.J. Palmer et al. / Polyhedron xxx (2013) xxx–xxx
over MgSO4. The MgSO4 was filtered, and the CH2Cl2 was removed
in vacuo to yield a white solid. Failure to perform the activated
charcoal step always led to impure product as indicated by a green
color. The white solid was transferred to the drybox and dissolved
in a minimal amount of toluene. The toluene solution was placed in
the freezer at ꢀ35 °C for recrystallization. After 2 days, 2.440 g
(63%) of white (MeCp)2ZrCl2 was collected via filtration and rinsed
with 10 mL of hexane. The filtrate was reduced and subjected to a
second recrystallization, which afforded an additional 0.360 g of
(MeCp)2ZrCl2 for a 72% total yield. 1H NMR (C6D6): 2.09 (s, 3H),
5.63 (m, 2H), 5.77 (m, 2H).
1 day. 0.80 g (72%) of yellow (MeCp)3HfCl was collected by filtra-
tion, rinsed with three 10 mL portions of hexane, and dried in va-
cuo. X-ray suitable crystals were grown from a toluene solution
at ꢀ38 °C. 1H NMR (C6D6): 2.15 (s, 3H), 5.32 (m, 2H), 5.63 (m,
2H); 13C NMR (C6D6): 15.4, 110.2, 117.3, 129.8.
2.3.10. (MeCp)2ZrBr2
0.840 g (5.25 mmol) of bromotrimethylsilane was added drop-
wise to a solution of 0.320 g (1.00 mmol) of (MeCp)2ZrCl2 dissolved
in 50 mL of toluene. The reaction was stirred for 2 h, and the vola-
tiles were removed in vacuo to yield 0.397 g (97%) of (MeCp)2ZrBr2
as a white solid. 1H NMR (C6D6): 2.09 (s, 3H), 5.64 (m, 2H), 5.77 (m,
2H).
2.3.7. (MeCp)2HfCl2
In the drybox, 3.922 g (12.2 mmol) of HfCl4 (99.9% Hf) was
added with stirring to a 250 mL Schlenk flask containing 100 mL
of THF to produce a white suspension. 2.500 g (24.5 mmol) of
NaMeCp dissolved in 75 mL of THF was added dropwise to the
HfCl4 solution over a 30 min period. Upon addition, the solution
turned pale yellow and cloudy from NaCl precipitation and was
stirred for 1 day. The THF was removed in vacuo to yield an off-
white solid. The flask was pumped out of the drybox and was
worked up under normal atmospheric conditions. The solid was
dissolved in 100 mL of CH2Cl2 to yield a brown solution, which
was transferred to a separatory funnel. The CH2Cl2 solution was
washed with three portions of 50 mL of 4 M HCl. Upon addition
of HCl, the solution turned bright green. The three HCl washes
were combined and washed with 30 mL CH2Cl2, and the two
CH2Cl2 layers were combined, stirred with activated charcoal to
remove the green color, and dried with MgSO4. The CH2Cl2 was
removed in vacuo to yield a white solid. In the drybox, the solid
was dissolved in a minimal amount of toluene, and the flask was
placed in the freezer at ꢀ35 °C overnight. After 2 days, the crystals
were collected via filtration, rinsed with three 10 mL portions of
hexanes, and dried in vacuo to yield 2.930 g (59%) of (MeCp)2HfCl2.
1H NMR (C6D6): 2.14 (s, 3H), 5.56 (m, 2H), 5.68 (m, 2H).
2.3.11. (MeCp)2HfBr2
2.400 g (15.7 mmol) of bromotrimethylsilane was added drop-
wise to a solution of 2.570 g (6.30 mmol) of (MeCp)2HfCl2 dis-
solved in 75 mL of toluene. The solution was stirred for 2 h, and
the toluene was then removed in vacuo to yield 2.978 g (95%) of
(MeCp)2HfBr2 as a grayish-white solid. 1H NMR (C6D6): 2.14 (s,
3H), 5.56 (m, 2H), 5.68 (m, 2H).
2.3.12. (MeCp)4Zr
In a 150 mL Schlenk flask in the drybox, 0.272 g (2.71 mmol) of
solid NaMeCp was added to 0.410 g (1.00 mmol) of (MeCp)2ZrBr2
dissolved in 30 mL of THF. The solution immediately became clou-
dy and orange. The flask was transferred to a Schlenk line, and the
solution was refluxed overnight. The flask was transferred back to
the glovebox, and the THF was removed in vacuo to yield a yellow
solid. The yellow solid was dissolved in 50 mL of toluene, and the
NaCl precipitate was filtered and rinsed with three 10 mL portions
of toluene. The volume of the toluene was reduced to 25 mL, and
the flask was placed in the freezer at ꢀ35 °C over night. The crys-
tals were collected by filtration, rinsed with 10 mL of hexane, and
dried in vacuo to yield 0.173 g (43%) of (MeCp)4Zr. 1H NMR (C6D6):
2.05 (s, 3H), 5.34 (m, 2H), 5.59 (m, 2H).
2.3.8. (MeCp)3ZrCl
This compound was prepared according to the literature proce-
dure with slight modifications [10]. In a 100 ml Schlenk flask,
0.350 g (3.43 mmol) of NaMeCp dissolved in 20 mL of THF was
added dropwise to a solution of 0.900 g (2.81 mmol) (MeCp)2ZrCl2
dissolved in 40 mL of THF. Upon addition, the clear solution imme-
diately turned orange, and solid NaCl precipitated out of solution.
The reaction was stirred overnight, and the THF was removed in
vacuo to yield a pale yellow solid. 75 mL toluene was added to
the flask, and the solution was stirred for 3 h. The solid NaCl was
filtered and rinsed with three 10 mL portions of toluene. The com-
bined toluene solution was reduced to approximately 30 mL.
30 mL of hexane was added to the toluene solution, and the flask
was placed in the freezer at ꢀ35 °C overnight. 0.570 g (56%) yellow
crystals of (MeCp)3ZrCl were collected via filtration, rinsed with
three 10 mL portions of hexane, and dried in vacuo. 1H NMR
(C6D6): 2.10 (s, 3H), 5.22 (m, 2H), 5.57 (m, 2H); 13C NMR (C6D6):
15.8, 110.3, 116.3, 128.6.
2.3.13. (MeCp)4Hf
In a 150 mL Schlenk flask, 0.500 g (1.00 mmol) of (MeCp)2HfBr2
and 0.460 g (4.50 mmol) of NaMeCp were dissolved in 75 mL of
THF. The solution immediately became cloudy and yellow. The
reaction was stirred for 2 days, and the THF was removed in vacuo
to yield an orange-yellow solid. The solid was dissolved in 50 mL of
toluene, and the solution was filtered. The NaCl precipitate was
rinsed with three additional 10 mL portions of toluene. The volume
of the toluene solution was reduced to about 5 mL, and 30 mL of
hexane was layered onto the toluene solution. The flask was placed
in the freezer at ꢀ35 °C overnight. Yellow crystals were collected
via filtration, rinsed with three 10 mL portions of hexane, and dried
in vacuo to yield 0.150 g (31%) (MeCp)4Hf. 1H NMR (C6D6): 2.33 (s,
3H), 4.77 (m, 2H), 5.37 (m, 2H); 13C NMR (C6D6): 16.4, 102.3, 109.7,
118.9.
2.3.9. (MeCp)3HfCl
2.3.14. (MeCp)3ZrO-tBu
A solution of 0.25 g NaMeCp (2.46 mmol) dissolved in 20 mL of
THF was added dropwise to a solution of 1.00 g (2.46 mmol)
(MeCp)2HfCl2 dissolved in 100 mL of THF. Over the course of addi-
tion, the solution became yellow, and NaCl precipitated. The reac-
tion was stirred for 20 h, and the THF was removed in vacuo to
yield a yellow solid. 50 mL of toluene was added to the reaction
flask and stirred. After 1 h, the solid NaCl was filtered and rinsed
with three 10 mL portions of toluene, and the toluene solution
was then concentrated to 5 mL. 100 mL hexane was then added
to the flask, and the flask was placed in the freezer at ꢀ38 °C for
A solution of 0.026 g (0.275 mmol) of sodium t-butoxide dis-
solved in 15 mL of THF was added dropwise to a solution of
0.100 g (0.275 mmol) of (MeCp)3ZrCl dissolved in 10 mL of THF.
Upon complete addition, the solution became cloudy and changed
color to pale yellow. The reaction was stirred for 4 h, and the THF
was removed in vacuo to yield a yellow oil. 25 mL of hexane was
added to the oil, and the solid NaCl precipitate was filtered. The
hexane was removed in vacuo to yield 0.063 g (57%) of (MeCp)3-
ZrO-tBu as a yellow oil. 1H NMR (C6D6): 1.09 (s, 9H), 2.10 (s, 9H),
5.54 (m, 6H), 5.79 (m, 6H).
Please cite this article in press as: E.J. Palmer et al., Polyhedron (2013), http://dx.doi.org/10.1016/j.poly.2013.03.007

E.J. Palmer et al. / Polyhedron xxx (2013) xxx–xxx
5
2.4. X-ray crystallography
3. Results and discussion
3.1. Cyclopentadienyl complexes
2.4.1. (C5H4CH3)3HfCl
X-ray-quality crystals were grown from a 9:1 hexane:toluene
solution upon slow cooling to ꢀ35 °C overnight, and were gath-
ered and coated with a polyfluorinated hydrocarbon oil. The crys-
tals were large yellow parallelepipeds, which looked good to the
eye, but their diffraction patterns indicated that they were not sin-
gle crystals. Fortunately, a small yellow rectangular plate was cut
from one of these, which did appear to be a single crystal, and so it
was used for data collection. The data collection strategy was set
up to measure a quadrant of reciprocal space with a redundancy
factor of 4.1, which means that 90% of the reflections were mea-
We had originally sought to synthesize the Cp3MCl complexes
of Zr and Hf to determine whether these complexes exhibited
the structural differences observed in the Cp4M complexes. The
present studies confirm that the syntheses of the Cp3MX com-
plexes of Zr and Hf can be accomplished by stoichiometrically con-
trolled transmetallation reactions between the appropriate
metallocene dihalide compounds and NaCp according to Scheme 2.
We have previously shown that the crystal structure of Cp3ZrCl
exhibited a 3-5 arrangement of the three C5H5 ligands [6]. Crystal
structure data for Cp3HfCl suggested a 2-5,1-1 arrangement of the
three C5H5 ligands, although the collected data set was of such
poor quality that bond distances and angles could not be deter-
mined accurately. As was the case for the Cp4M complexes, this
apparent structural difference between the tris-Cp zirconium and
hafnium chloride compounds is highly unusual as analogous zirco-
nium and hafnium compounds are almost always structurally
analogous.
Cp3ZrCl was synthesized by reaction of Cp2ZrCl2 with one molar
equivalent of NaCp in THF at ꢀ78 °C. The cold temperature is re-
quired for stoichiometric control of the reaction; when this reac-
tion was performed at room temperature, a mixture of Cp3ZrCl
with significant amounts of Cp4Zr and Cp2ZrCl2 was obtained.
Reactions performed on a Schlenk line at ꢀ78 °C greatly improved
the ratio of Cp3ZrCl to Cp4Zr and Cp2ZrCl2. However, the best way
to strictly control the stoichiometry of the reaction was via synthe-
sis on the high-vacuum line. During this procedure, solid THF was
condensed directly into a flask containing the Cp2ZrCl2 and NaCp
starting materials as solids. The solid THF was slowly melted to
ꢀ78 °C using a dry ice/isopropanol bath over a period of 4 h. This
slow warming of the THF was found to be the key in strictly con-
trolling the 1:1 reaction of NaCp with Cp2ZrCl2 as was indicated
by 1H NMR. Subsequent workup and recrystallization yielded
Cp3ZrCl in 47%, which was an improvement on our previous yield.
Crystals obtained from the slow cooling of a concentrated tolu-
ene solution to ꢀ35 °C were submitted for X-ray diffraction analy-
sis. As we saw previously [6], and as is the case for other
structurally characterized tris-Cp zirconium compounds, such as
Cp4Zr [4] and Cp3ZrCH3 [8], the crystal structure showed disorder
in the arrangement of the tris-Cp moiety within the crystal lattice
and was, therefore, not refined further. The presence of this disor-
der in all 3-5 tris-Cp zirconium compounds reported thus far can
be explained by visualizing the tris-Cp fragment as forming a cyl-
inder around the zirconium center. The orientation of the three
C5H5 ligands can then adopt a number of different orientations,
which vary little in energy, around the cylinder wall. This observa-
tion is consistent with a small rotational barrier of the tris-Cp frag-
sured at least 4.1 times. A combination of
u and x scans with a
frame width of 1.0° was used. Data integration was done with DEN-
ZO [20], and scaling and merging of the data was done with SCALE-
PACK [20].
The structure was solved for the Hf atom by the Patterson
method in SHELXS-86 [21]. The rest of the atoms were located by
standard Fourier methods. Full-matrix least- squares refinements
based on F2 were performed in SHELXL-93 [22]. For each methyl
group, the hydrogen atoms were added at calculated positions
using a riding model with U(H) = 1.5Ueq(bonded atom). The torsion
angle, which defines the orientation of the methyl group about the
C–C bond, was refined. The hydrogen atom bonded to C(13) was lo-
cated on a difference electron density map, added to the model,
and its positional parameters were then refined. The remaining
hydrogen atoms were included in the model at calculated positions
using a riding model with U(H) = 1.2Ueq(attached atom). The max-
imum peak in the final difference electron density map is 1.15 Å
from the Hf atom. Neutral atom scattering factors were used and
include terms for anomalous dispersion [23]. The PLATON program
[24] was used to calculate the metrical parameters involving ring
centroids.
2.4.2. (C5H4CH3)4Zr
X-ray-quality crystals were grown by a slow cooling of a
concentrated toluene solution at ꢀ35 °C. The data collection
crystal was a clear, yellow rectangular chunk, which was cut
from the end of a rod. The data collection strategy was set up
to measure a quadrant of reciprocal space with a redundancy
factor of 4.7, which means that 90% of the reflections were mea-
sured at least 4.7 times. A combination of
u and x scans with a
frame width of 1.0° was used. Data integration was done with
DENZO [20], and scaling and merging of the data was done with
SCALEPACK [20].
The structure was solved by the Patterson method in SHELXS-86
[21]. Full-matrix least-squares refinements based on F2 were per-
formed in SHELXL-97 [25], as incorporated in the WINGX package
[26]. Most of the hydrogen atoms were added to the model at
calculated positions. The hydrogen atom bonded to C(22) was lo-
cated on a difference electron density map and refined. One of
the methyl groups, C(12) appears to be rotationally disordered
about the C–C bond. For this methyl group, two sets of hydrogen
atoms were located and added to the model with occupancy fac-
tors of 0.5 each and only their Uiso values were refined. For the
other methyl groups, the hydrogen atoms were added using a
riding model with U(H) = 1.5Ueq(bonded carbon atom). The tor-
sion angle, which defines the orientation of the methyl group
about the C–C bond, was refined. The remaining hydrogen atoms
were included in the model at calculated positions using a riding
model with U(H) = 1.2Ueq(attached atom). Neutral atom scatter-
ing factors were used and include terms for anomalous disper-
sion [23].
ment about the Zr–C(r) bond vector in solid Cp4Zr. We have
performed exploratory studies of these dynamic processes in the
solid state via variable temperature CP-MAS 13C NMR [27].
Although Cp3ZrCl can be synthesized and characterized without
interference from Cp2ZrCl2 and Cp4Zr impurities, the synthesis of
Scheme 2. Synthesis of Cp3MX complexes from Cp2MX2 complexes.
Please cite this article in press as: E.J. Palmer et al., Polyhedron (2013), http://dx.doi.org/10.1016/j.poly.2013.03.007

6
E.J. Palmer et al. / Polyhedron xxx (2013) xxx–xxx
pure Cp3HfCl proved to be much more difficult. The most favorable
conditions for the strict control of the 1:1 stoichiometry of the
reaction between Cp2HfCl2 and NaCp were found when the
reaction was performed on a high-vacuum line. As was the case
for Cp3ZrCl, when the reactants were frozen in a solid matrix of
THF, which was then allowed to slowly warm to ꢀ78 °C, the pro-
duction of Cp2HfCl2 and Cp4Hf was minimized. However, recrystal-
lizations of the crude Cp3HfCl always led to the formation of
substantial amounts of Cp2HfCl2 and Cp4Hf, presumably from the
reaction of two molecules of Cp3HfCl according to Scheme 3.
The observation of significant ligand redistribution in Cp3MCl to
form Cp2MCl2 and Cp4M compounds for Hf but not for Zr is consis-
tent with a less sterically congested metal center for the Hf com-
pound. Because Cp3HfCl likely exhibits a 2-5,1-1 geometry, the
addition of the fourth C5H5 ligand is more facile for the Hf complex
than for the 3-5 Zr complex. All crystals of Cp3HfCl obtained for
X-ray diffraction produced diffuse scattering peaks, and useful
structural information was not obtained. These diffuse scattering
peaks could have been the result of cocrystallization of Cp2HfCl2
and/or Cp4Hf with Cp3HfCl within the crystal lattice. Preliminary
equilibrium studies showed that placing Cp2HfCl2 and Cp4Hf to-
gether in solution did not produce noticeable quantities of Cp3HfCl
by 1H NMR [27].
These results show that the synthesis of tris-Cp compounds of
the Group IV transition metals is challenging in several respects.
Stoichiometric control is essential to eliminate the undesired
Cp2MCl2 and Cp4M compounds, which seem to be more thermody-
namically favored than and in equilibrium with the tris-Cp
compounds. When conditions were found that minimized the
presence of the bis- and tetrakis-Cp compounds, the X-ray diffrac-
tion data oftentimes showed disorder amongst the tris-Cp moiety
throughout the crystal lattice. As a result of these problems, the
synthesis of these compounds using the MeCp ligand was
investigated.
reported. Herein, the synthesis and structural characterization of
tris-MeCp compounds of both zirconium and hafnium are
reported.
3.2.1. (MeCp)3ZrCl
(MeCp)3ZrCl was synthesized by reaction of (MeCp)2ZrCl2 with
NaMeCp in a 1:1 M ratio in THF at room temperature, analogous to
Scheme 2. The addition of excess NaMeCp did not yield (MeCp)4Zr
at room temperature. Large crystals of (MeCp)3ZrCl were grown
either by slowly cooling a toluene/hexane solution to ꢀ35 °C or
by slow evaporation of a toluene solution. Unfortunately, despite
good diffraction, the crystals were twinned, which prevented the
correct space group from being assigned to the crystal. As a result,
the X-ray diffraction data suggested
a 3-5 structure for
(MeCp)3ZrCl, but a satisfactory solution to the observed diffraction
pattern was not obtained. Repeated attempts at crystallizations
from different solvent systems failed to produce crystallographi-
cally-acceptable crystals.
3.2.2. (MeCp)3HfCl
(MeCp)3HfCl was synthesized by the reaction of (MeCp)2HfCl2
with NaMeCp in a 1:1 M ratio in THF at room temperature. The
addition of excess NaMeCp did not produce (MeCp)4Hf. X-ray qual-
ity crystals were grown from a 9:1 hexane:toluene solution at
ꢀ35 °C. An ORTEP drawing of the crystal structure of (MeCp)3HfCl
is shown in Fig. 1. Crystallographic data for (MeCp)3HfCl are listed
in Table 1, and bond lengths and angles are listed in Table 2.
The Hf–Cl bond length was determined to be 2.414(1) Å, which
is in good agreement with Hf–Cl distances reported for a number of
bis-Cp hafnium chloride compounds [10,28]. The Hf–C(
distance was determined to be 2.349(3) Å, which is 0.03 Å shorter
that the Hf–C( ) bond in Cp4Hf [5b] and 0.11 Å longer than the
Hf-CH3 bond length found in Cp2Hf(CH3)2 [29]. The Hf–C( ) dis-
r) bond
r
p
tances ranged from 2.450(2) to 2.573(2) Å with an average value
of 2.498 Å, which is in good agreement with other bis-Cp hafnium
3.2. Methylcyclopentadienyl complexes
compounds [30]. The Cent–Hf–Cent and Cl–Hf–C(r) angles were
determined to be 130.3(1)° and 91.6(1)°, respectively, which
The presence of one methyl group on the Cp ligand was ex-
pected to eliminate the problems listed above by eliminating the
formation of the tetrakis-Cp compounds. Because MeCp is a larger
ligand than Cp, formation of (MeCp)4M (M = Zr, Hf) requires reflux-
ing conditions as was reported previously [10]. Also, the presence
of the methyl group should eliminate the disorder of the tris-Cp
fragments within the crystal lattice. Our investigations into the
synthesis and characterization of these tris-MeCp compounds ver-
ify the alleviation of the stoichiometry and disorder problems ob-
served for the tris-Cp compounds. Although the tris-MeCp
chloride compounds of zirconium and hafnium have been reported
previously [10], no structural data besides solution NMR were
agree well with the respective angles in Cp4Hf [5b]. The two
Cent–Hf–C(
103.8(1)°, and the two Cent–Hf–Cl angles were determined to be
106.8(1)° and 106.7(1)°. The plane of the
1-MeCp ligand formed
an angle of 122° with the Hf–C( ) bond vector.
r) angles were determined to be 110.9(1)° and
g
r
The crystal structure of (MeCp)3HfCl shows the presence of two
types of MeCp rings. However, characterization by 1H NMR indi-
cates the presence of only one type of MeCp ligand, which is ob-
served as an AA0BB0X pattern in the NMR spectrum, suggesting
fluxionality of the MeCp ligands in solution. In order to confirm
this
NMR study of (MeCp)3HfCl in solution was performed. The results
g
1-to-g5 MeCp exchange process, a variable-temperature 1H
Scheme 3. Redistribution of Cp3HfX into Cp2HfX2 and Cp4Hf.
Please cite this article in press as: E.J. Palmer et al., Polyhedron (2013), http://dx.doi.org/10.1016/j.poly.2013.03.007

E.J. Palmer et al. / Polyhedron xxx (2013) xxx–xxx
7
Fig. 1. X-ray crystal structure of (MeCp)3HfCl. The non-hydrogen atoms are drawn with 50% probability displacement ellipsoids. The hydrogen atoms are drawn with an
artificial radius.
Table 1
of this experiment indicate that the MeCp ligands are fluxional
Crystallographic data for (MeCp)3HfCl.
down to ꢀ90 °C, the lowest limit allowed by the hardware on the
NMR spectrometer. Although broadening of the MeCp resonances
is observed at lower temperatures, the same broadening is ob-
served for the toluene-d8 peaks and is attributed to the inhomoge-
neity of the magnetic field at lower temperatures; the broadening
Empirical formula
Formula weight (g/mol)
T (K)
k (Å)
Crystal system
Space group
Unit cell dimensions
a (Å)
C18H21HfCl
451.29
200(2) K
0.71073
monoclinic
P21/c
is not caused by the slowing of the
process.
g
1-to-g5 MeCp exchange
13.5017(2)
10.4033(2)
11.3176(2)
95.671(1)
1581.92(5)
4
1.895
872
Characterization of (MeCp)3HfCl by MALDI mass spectrometry
resulted in a peak at 417 m/z, which is attributed to the (MeCp)3Hf
radical cation. The loss of chloride from the molecular ion peak is
not unprecedented in MALDI experiments [31]. The isotopic peak
pattern is consistent with the presence of hafnium metal. No
matrices were used in the MALDI experiments as they were shown
to react with (MeCp)3HfCl.
b (Å)
c (Å)
b (°)
V (Å3)
Z
Dcalc (Mg/m3)
F(000)
Crystal size (mm)
h (°)
0.08 ꢁ 0.12 ꢁ 0.31
2.48–27.46
Index ranges
ꢀ17 6 h 6 17, ꢀ13 6 k 6 13, ꢀ14 6 l 6 14
29,487
3.2.3. (MeCp)3ZrO-tBu
Reflections collected
Independent reflections (Rint
Refinement method
Data/restraints/parameters
Goodness-of-fit (GOF) on F2
Because our theoretical calculations [7] suggested that p-donor
)
3615 (0.051)
ligands could effectively compete for electron donation to the me-
tal center, which would favor the 2-5,1-1 geometry even for zirco-
nium compounds, the tris-MeCp t-butoxide compound of Zr was
synthesized. (MeCp)3ZrO-tBu was prepared in 57% yield from the
reaction of (MeCp)3ZrCl and NaO-tBu in a 1:1 M ratio at room tem-
perature. Addition of excess NaO-tBu led to the formation of
(MeCp)2Zr(O-tBu)2.
full-matrix least-squares on F2
3615/0/187
1.043
Final R indices [I > 2
R indices (all data)
r(I)]
R1 = 0.0219, wR2 = 0.0452
R1 = 0.0355, wR2 = 0.0483
1.647 and ꢀ0.790
Largest difference peak and
hole (e/Å3)
Unfortunately, (MeCp)3ZrO-tBu is a yellow oil and, therefore,
could not be characterized by X-ray crystallography. 1H NMR at
room temperature indicated only one type of MeCp ligand, which
appeared as an AA0BB0X pattern in the NMR spectrum; these
NMR data are consistent with either a 3-5 structure or a 2-5,1-1
structure that is fluxional on the NMR timescale. Variable-
temperature 1H NMR in solution indicated that if (MeCp)3ZrO-
Table 2
Bond lengths and angles in the crystal structure of (MeCp)3HfCl.
Bond length (Å)
Hf–Cl
Bond angle (°)
tBu exhibits a 2-5,1-1 structure, the barrier for
g
1-to-g5 MeCp
2.414(1)
2.349(3)
2.450–2.573(2)
2.498
Cent–Hf–Cent
Cent1–Hf–Cl
Cent2–Hf–Cl
130.3(1)
106.7(1)
106.8(1)
110.9(1)
103.8(1)
Hf–C(
Hf–C(
Hf–C(
r
)
interconversion is of such low energy that the process could not
be frozen out even at ꢀ90 °C. As a result of these findings, the syn-
thesis and characterization of (MeCp)3ZrO-tBu did not verify nor
p
p
) (range)
) (avg)
Cent1–Hf–C(
Cent2–Hf–C(
r
r
)
)
Hf–Cent
2.193(2), 2.197
disprove the p-competition model advanced previously.
Please cite this article in press as: E.J. Palmer et al., Polyhedron (2013), http://dx.doi.org/10.1016/j.poly.2013.03.007

8
E.J. Palmer et al. / Polyhedron xxx (2013) xxx–xxx
Fig. 2. X-ray crystal structure of (MeCp)4Zr. The non-hydrogen atoms are drawn with 50% probability displacement ellipsoids. All hydrogen atoms except the hydrogen atom
attached to C(22) are omitted for clarity.
Table 3
It should be noted that the synthesis of (MeCp)3ZrN(CH3)2 was
Crystallographic data for (MeCp)4Zr.
attempted by the reaction of (MeCp)3ZrCl with LiN(CH3)2 in a
1:1 M ratio. However, 1H NMR suggested that (MeCp)3ZrN(CH3)2
Empirical formula
C24H28Zr
was not formed.
Formula weight (g/mol)
407.68
T (K)
200(2)
k (Å)
Crystal system
Space group
Unit cell dimensions
a (Å)
b (Å)
c (Å)
0.71073
monoclinic
P21/n
3.2.4. (MeCp)4M complexes
Etievant and coworkers reported the synthesis of (MeCp)4M
(M = Zr, Hf) by refluxing (MeCp)2MCl2 in excess NaMeCp in toluene
for 2 days [10]. In our attempts to crystallize tris-MeCp zirconium
halide compounds, it was serendipitously discovered that the tet-
rakis-MeCp compounds could be synthesized using less harsh con-
ditions starting from the (MeCp)2MBr2 compounds. Although
addition of excess NaMeCp to (MeCp)3ZrCl resulted in no forma-
tion of (MeCp)4Zr, we found from 1H NMR that the reaction of ex-
cess NaMeCp with (MeCp)2ZrBr2 resulted in the formation of large
amounts of (MeCp)4Zr. Refluxing excess NaMeCp with (MeCp)2-
ZrBr2 in THF for several hours resulted in complete conversion to
(MeCp)4Zr.
X-ray-quality crystals of (MeCp)4Zr were grown by slowly cool-
ing a concentrated toluene solution to ꢀ35 °C. The crystal structure
of (MeCp)4Zr is shown in Fig. 2. Crystallographic data are summa-
rized in Table 3, and bond lengths and angles are listed in Table 4.
The crystal structure shows that (MeCp)4Zr exhibits a 3-5,1-1
structure similar to that observed for Cp4Zr [4].
7.902(1)
17.080(2)
14.066(1)
92.460(4)
1896.7(4)
4
1.428
848
b (°)
V (Å3)
Z
Dcalc (Mg/m3)
F(000)
Crystal size (mm)
h (°)
Index ranges
Reflections collected
Independent reflections (Rint
Refinement method
Data/restraints/parameters
Goodness-of-fit (GOF) on F2
0.15 ꢁ 0.23 ꢁ 0.27
2.79–27.46
ꢀ10 6 h 6 10, ꢀ22 6 k 6 22, ꢀ18 6 l 6 18
38,853
)
4346 (0.035)
Full-matrix least-squares on F2
4346/0/239
1.062
Final R indices [I > 2
R indices (all data)
r
(I)]
R1 = 0.0248, wR2 = 0.0568
R1 = 0.0355, wR2 = 0.0598
0.513 and ꢀ0.287
Largest difference peak and
hole (e/Å3)
Although Cp4Zr and (MeCp)4Zr exhibit similar 3-5,1-1
structures, there are noticeable differences between the common
bond lengths of the two structures. The Zr–C(
(MeCp)4Zr was determined to be 2.513(2) Å, which is 0.066 Å
longer than the Zr–C( ) bond length in Cp4Zr. The average
Zr–C( ) distance is 2.615 Å in (MeCp)4Zr, which is 0.035 Å longer
than the average Zr–C( ) distance found in Cp4Zr [4]. The elonga-
tion of the bond lengths in (MeCp)4Zr relative to the respective
bond lengths in Cp4Zr is not unreasonable as the MeCp ligand
increases the steric interactions around the metal center. A closer
r) bond length in
Table 4
Bond lengths and angles in the crystal structure of (MeCp)4Zr.
r
Bond length (Å)
Bond angle (°)
p
Zr–C(
Zr–C(
Zr–C(
Zr–C(
Zr–C(
Zr–C(
Zr–C(
Zr–C(
r
)
2.513(2)
2.555(2)–2.679(2)
2.614(2)
2.540(2)–2.642(2)
2.592(2)
2.534(2)–2.728(2)
2.638(2)
Cent1–Zr–Cent2
Cent2–Zr–Cent3
Cent1–Zr–Cent3
Cent–Zr–Cent (avg)
117.73(3)
118.61(3)
115.34(3)
117.2
100.01(5)
101.80(5)
97.18(4)
99.7
p
p)1 (range)
p)1 (avg)
p)2 (range)
p)2 (avg)
p)3 (range)
p)3 (avg)
p) (avg)
Cent1–Zr–C(
Cent2–Zr–C(
Cent3–Zr–C(
r
r
r
)
)
)
inspection of the Zr–C(
length range is much greater in (MeCp)4Zr than in Cp4Zr. The
shortest and longest Zr–C( ) in (MeCp)4Zr were 2.534(2) and
p) distances reveals that the Zr–C(p) bond
2.615
Cent–Zr–C(r) (avg)
p
2.728(2) Å, respectively; the corresponding bond lengths in Cp4Zr
were 2.517 and 2.634 Å. This observation is consistent with a
greater gearing of the MeCp ligands in (MeCp)4Zr influenced by
steric interactions of the methyl groups.
115.3° for an average value of 117.2°, which is very similar to the
average value of 116.7° determined for Cp4Zr. The three Cent–
The four MeCp ligands were planar to within 0.003 Å. The three
Cent–Zr–Cent angles were determined to be 117.7°, 118.6° and
Zr–C(
r) angles were 97.2°, 100.1°, and 101.8° for an average value
of 99.7°, virtually identical to the average for Cp4Zr.
Please cite this article in press as: E.J. Palmer et al., Polyhedron (2013), http://dx.doi.org/10.1016/j.poly.2013.03.007

E.J. Palmer et al. / Polyhedron xxx (2013) xxx–xxx
9
In addition to (MeCp)4Zr, (MeCp)4Hf was synthesized by reac-
tion of excess NaMeCp with (MeCp)2HfBr2 in THF at room temper-
ature. Unlike (MeCp)4Zr, complete formation of (MeCp)4Hf
required no additional heating. This reactivity difference may be
caused by a structural difference between the two compounds,
namely a 2-5,2-1 structure for (MeCp)4Hf as is observed for Cp4Hf.
Unfortunately, X-ray-quality crystals were not obtained for
(MeCp)4Hf, and, therefore, a crystal structure of (MeCp)4Hf could
not be obtained.
The crystal structures of (MeCp)3HfCl and (MeCp)4Zr show
these compounds to have the same arrangements of g1 and g5
MeCp ligands as their Cp counterparts. This suggests that while
the presence of a methyl group on the Cp ligand can influence bond
lengths, the steric demands introduced by the larger MeCp ligand
is not enough to change the preferred arrangement of the Cp li-
gands around the metal center. As a result, the utilization of the
MeCp ligand proved to be useful in improving reaction yields, con-
trolling stoichiometry and eliminating disorder within the crystal
lattices without changing the overall geometry of the compounds.
data can be obtained free of charge via http://www.ccdc.cam.ac.uk/
conts/retrieving.html, or from the Cambridge Crystallographic
Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44
1223 336 033; or e-mail: deposit@ccdc.cam.ac.uk.
References
[1] G. Wilkinson, J. Organomet. Chem. 100 (1975) 273.
[2] (a) R.J. Strittmatter, B.E. Bursten, J. Am. Chem. Soc. 113 (1991) 552;
(b) B.E. Bursten, R.J. Strittmatter, Angew. Chem. Int. Ed. 30 (1991) 1069;
(c) C.J. Burns, M.S. Eisen, in: L.R. Morss, N.M. Edelstein, J. Fuger (Eds.),
Chemistry of the Actinide and Transactinide Elements, vol. 5, Springer,
Dordrecht, 2006, p. 2799.
[3] (a) W. Moffitt in, L.T. Reynolds, G. Wilkinson, J. Inorg. Nucl. Chem. 2 (1956)
246;
(b) J.W. Lauher, R. Hoffmann, J. Am. Chem. Soc. 98 (1976) 1729.
[4] R.D. Rogers, R. Vann Bynum, J.L. Atwood, J. Am. Chem. Soc. 100 (1978)
5238.
[5] (a) Ti: J.L. Calderon, F.A. Cotton, B.G. DeBoer, J. Takats, J. Am. Chem. Soc. 93
(1971) 3592;
(b) . Hf:R.D. Rogers, R. Vann Bynum, J.L. Atwood, J. Am. Chem. Soc. 103 (1981)
692.
[6] R.J. Strittmatter, Ph.D. Dissertation, The Ohio State University, 1990.
[7] E.J. Palmer, B.E. Bursten, Polyhedron 25 (2006) 575.
[8] T. Brackemeyer, G. Erker, R. Fröhlich, Organometallics 16 (1997) 531.
[9] W. Lukens, Acta Crystallogr., Sect. C 51 (1995) 10.
[10] P. Etievant, B. Gautheron, G. Tainturier, Bull. Soc. Chim. Fr. (1978) 292.
[11] S. Demir, T.J. Mueller, J.W. Ziller, W.J. Evans, Angew. Chem. Int. Ed. 50 (2011)
515.
[12] (a) Zr: W.W. Lukens Jr., R.A. Andersen, Organometallics 14 (1995) 3435;
(b) . Hf:V.V. Burlakov, P. Arndt, A. Spannenberg, U. Rosenthal, Acta Crystallogr.,
Sect. E 67 (2011) m629.
4. Concluding remarks
Tris-Cp zirconium and hafnium compounds were synthesized by
the reaction of Cp2MCl2 with one equivalent of NaCp. These prepara-
tions proved to be difficult because of the stoichiometric control nec-
essary to obtain pure Cp3MCl compounds, which led to low yields.
For zirconium, pure Cp3ZrCl was prepared, but X-ray diffraction stud-
ies showed disorder of the tris-Cp fragment within the crystal lattice.
The synthesis of pure Cp3HfCl was hindered because upon
recrystallization, Cp2HfCl2 and Cp4Hf co-crystallized with Cp3HfCl
and X-ray-quality crystals could not be obtained.
[13] (a) W.L. Jolly, The Synthesis and Characterization of Inorganic Compounds,
Prentice Hall, Prospect Heights, IL, 1970;
(b) A.J. Gordon, R.A. Ford, The Chemist’s Companion, Wiley, New York, 1972;
(c) D.F. Shriver, The Manipulation of Air-Sensitive Compounds, Krieger,
Malbar, FL, 1969.
[14] D.D. Perrin, W.L.F. Armarego, D.R. Perrin, Purification of Laboratory Chemicals,
Pergamon, Oxford, 1988.
[15] Mestrelab Research SL: www.mestrec.com.
[16] G. Wilkinson, Org. Synth. (1963) 4.
[17] R.B. King, F.G.A. Stone, Inorg. Synth. 7 (1963) 99.
[18] (a) R.K. Freidlina, E.M. Brainina, A.N. Nesmayanov, Dokl. Akad. Nauk SSSR 138
(1961) 1369;
The problems of low yields, stoichiometry control, and disorder
found in the crystal structures of tris-Cp compounds of Zr and Hf
were overcome by synthesizing the respective MeCp compounds,
which were shown to be structurally analogous to their tris-Cp
counterparts. The larger MeCp ligand was beneficial in controlling
the 1:1 stoichiometry of the reaction between (MeCp)2MCl2 with
NaMeCp, which dramatically increased yields. Additionally, the
presence of the methyl group on the Cp ligand mitigated the prob-
lem of disorder of the tris-Cp fragment within the crystal lattice for
those structures with 3-5 geometries.
(b) E. Samuel, R. Setton, C.R. Hebd. Seances Acad. Sci. Ser. C C254 (1962) 308;
(c) P.M. Druce, B.M. Kingston, M.L.F. Lappert, T.R. Spalding, R.C. Srivastava, J.
Chem. Soc. A (1969) 2106.
[19] R.F. Jordan, R.E. LaPointe, P.K. Bradley, N. Baenziger, Organometallics 8 (1989)
2892.
[20] Z. Otwinowski, W. Minor, Methods Enzymol. Macromol. Crystallogr., Part A
276 (1997) 307.
[21] G.M. Sheldrick, Acta Crystallogr., Sect. A 46 (1990) 467.
[22] G.M. Sheldrick, SHELXL-93, Universität Göttingen, Germany, 1993.
[23] International Tables for Crystallography, vol. C, Kluwer Academic Publishers,
Dordrecht, 1992.
(MeCp)3HfCl was shown by X-ray crystallography to exhibit a 2-
5,1-1 static structure, as was suggested for Cp3HfCl. Solution VT
NMR studies indicate facile
g
1-to-g5 MeCp interconversion at
[24] A.L. Spek, J. Appl. Crystallogr. 36 (2003) 7.
[25] G.M. Sheldrick, Acta Crystallogr., Sect. A 64 (2008) 112.
[26] L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837.
[27] E.J. Palmer, Ph.D. Dissertation, The Ohio State University, 2005.
[28] (a) M.D. Bruce, G.W. Coates, E. Hauptman, R.M. Waymouth, J.W. Ziller, J. Am.
Chem. Soc. 119 (1997) 11174;
temperatures as low as ꢀ90 °C. Somewhat surprisingly, (MeCp)4Zr
was shown to exhibit a 3-5,1-1 structure analogous to Cp4Zr de-
spite the increased bulk of the MeCp ligand. These structures com-
plement and are consistent with our prior theoretical studies,
which predicted that Cp3Hf(halide) complexes will favor a 2-5,1-
1 geometry, whereas Cp3Zr(alkyl) complexes will favor the 3-5
binding motif.
(b) H.G. Woo, R.H. Heyn, T.D. Tilley, J. Am. Chem. Soc. 114 (1992) 5698;
(c) B.J. Grimmond, J.Y. Corey, N.P. Rath, Organometallics 18 (1999) 404;
(d) R.D. Rogers, M.M. Benning, L.K. Kurihara, K.J. Moriarty, M.D. Rausch, J.
Organomet. Chem. 293 (1985) 51;
(e) C.J. Cardin, D.J. Cardin, D.A. Morton-Blake, H.E. Parge, A. Roy, J. Chem. Soc.,
Dalton Trans. (1987) 1641;
(f) R.M. Shaltout, J.Y. Corey, N.P. Rath, J. Organomet. Chem. 503 (1995) 205;
(g) H.G. Alt, A. Reb, W. Milius, A. Weis, J. Organomet. Chem. 628 (2001) 169;
(h) G.M. Diamond, M.L.H. Green, N.A. Popham, A.N. Chernega, J. Chem. Soc.,
Dalton Trans. (1993) 2535;
(i) F. Lindenberg, U. Muller, A. Pilz, J. Sieler, E. Hey-Hawkins, Z. Anorg. Allg.
Chem. 622 (1996) 683.
Acknowledgement
Acknowledgment is made to the Donors of The Petroleum Re-
search Fund, administered by the American Chemical Society, for
support of this research (Grant 38438-AC3).
[29] W.E. Hunter, D.C. Hrncir, R. Vann Bynum, R.A. Penttila, J.L. Atwood,
Organometallics 2 (1983) 750.
[30] D.J. Cardin, M.F. Lappert, C.L. Raston, Chemistry of Organo-Zirconium and -
Hafnium Compounds, John Wiley & Sons, New York, 1986.
[31] E.A.V. Ebsworth, D.W.H. Rankin, S. Cradock, Structural Methods in Inorganic
Chemistry, 2nd ed., Blackwell Scientific, Boston, 1987.
Appendix A. Supplementary data
CCDC 892042 and 892043 contains the supplementary crystal-
lographic data for (MeCp)3HfCl and (MeCp)4Zr, respectively. These
Please cite this article in press as: E.J. Palmer et al., Polyhedron (2013), http://dx.doi.org/10.1016/j.poly.2013.03.007