Aqueous chemistry of the five-coordinate zirconocene dichloride PhP(CH2CH2-η5-C5H4)2ZrCl2: X-ray structure of the sulfato complex PhP(CH2CH2-η5-C

Article abstract of DOI:10.1016/j.poly.2006.06.031

ES-MS of dilute solutions of (bcep)ZrCl₂ (1) (bcep = PhP(CH₂CH₂-η⁵-C₅H₄)₂) in acetonitrile-water indicates the presence of three species in solution: [(bcep)Zr(H₂O)

Full text of DOI:10.1016/j.poly.2006.06.031

Polyhedron 26 (2007) 406–414
www.elsevier.com/locate/poly
Aqueous chemistry of the five-coordinate zirconocene dichloride
PhP(CH2CH2–g5-C5H4)2ZrCl2: X-ray structure of the sulfato complex
PhP(CH2CH2–g5-C5H4)2Zr(g2-SO4)
*
James R. Butchard, Owen J. Curnow
Department of Chemistry, University of Canterbury, Private Bag 4800, Christchurch, New Zealand
Received 3 April 2006; accepted 28 June 2006
Available online 14 July 2006
Abstract
ES-MS of dilute solutions of (bcep)ZrCl2 (1) (bcep = PhP(CH2CH2–g5-C5H4)2) in acetonitrile–water indicates the presence of three
species in solution: [(bcep)Zr(H2O)2]2+ (2), [{(bcep)Zr(l-OH)}2]2+ (3), and the singly deprotonated form of dimer 3, [{(bcep)Zr}2(l-O)-
(l-OH)]+ (4). In aqueous solutions, reactivity studies investigated largely by 31P NMR spectroscopy support the presence of 2, 3 and 4.
One equivalent of fluoride was ꢀfound to react immediately with 2, but slowly with 3 and 4, to give the monofluoro complex [(bcep)ZrF]+
which was isolated as the BPh4 salt. Addition of aqueous sodium sulphate gave insoluble crystals of the sulfato complex (bcep)Zr(SO4)
which was characterised by X-ray crystallography. Addition of acids HX (X = Br, NO3, HSO4, HCO3) to aqueous solutions of 1 gave
species observed by ES-MS corresponding to [(bcep)ZrX]+.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Keywords: Zirconocene; Phosphine; Sulfato complex; Fluoro complex; Aqueous chemistry
1. Introduction
[5] and zirconium [6] complexes. The number of bis(cyclo-
pentadienyl) phosphine ligands capable of chelation is very
small [7,8]. The ligand C6H11PðCHMe-2-PPh2C5H3ꢀÞ2 [7]
could exhibit phosphine chelation, but this has not yet been
demonstrated. We recently communicated the preparation
of PPh(CH2CH2C5H5)2 (bcepH2) and the synthesis of fer-
rocene complexes derived from this compound in which
the P atom is not coordinated to the Fe center [8]. The P
atom can, however, coordinate to a second metal center
as in the complex trans-PdCl2{(bcep)Fe}2 [8].
The aqueous chemistry of zirconocenes has been investi-
gated by a small number of workers (in contrast to that of
the titanocenes [9]) with the identification of species in solu-
tion largely dependent upon crystallographic characterisa-
tion of the compounds isolated in the solid state. Addition
of two equivalents of water to Cp2Zr(C7H7SO3)2 gives the
toluenesulfonatodiaqua cation [Cp2Zr(C7H7SO3)(H2O)2]+
[10], whereas addition of three equivalents of water to the
bistriflate complex Cp2Zr(CF3SO3)2 (THF) gives the tri-
aqua dication [Cp2Zr(H2O)3]2+ [11] and addition of six
The chemistry of zirconocene dichlorides is currently
one of the most significant areas of research in organome-
tallic chemistry [1]. The area of functionalised bis(cyclo-
pentadienyl) ligands capable of chelation is an active
field of organometallic chemistry [2–4]. Notable zirconoc-
enes are the complexes MeN(CH2CH2C5H4)2ZrCl2 and
{C5H3N(CH2C5H4)2-2,6}ZrCl2 in which amine and pyri-
dine coordination, respectively, is proposed on the basis
of 1H NMR spectroscopy and XPS studies [2]. A crystallo-
graphically characterized hydrolysis product of {C5H3N-
(CH2C5H4)2-2,6}ZrCl2 is the cationic chloro-aqua complex
[{C5H3N(CH2C5H4)2-2,6}ZrCl(H2O)]+ in which the pyri-
dine substituent is coordinated to the metal center [3].
Ansa-metallocenes with a P bridging atom directly
bonded to the C5 ring are known for a number of iron
*
Corresponding author. Tel.: +64 3 364 2819; fax: +64 3 364 2110.
E-mail address: owen.curnow@canterbury.ac.nz (O.J. Curnow).
0277-5387/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2006.06.031

J.R. Butchard, O.J. Curnow / Polyhedron 26 (2007) 406–414
407
equivalents gives a 30% yield of the dinuclear hydroxo-
bridged complex [Cp2Zr2(l-OH)2(H2O)6]4+ in which one
Cp ligand has been hydrolysed from each Zr atom [12].
Other zirconocene dichloride hydrolysis products have been
reported by Thewalt and co-workers in which one Cp ligand
has been hydrolysed and there are oxo/hydroxo bridges:
{CpZr(NO3)2}2(l-OH)2 was isolated from nitric acid in
chloroform [13] and [Cp3Zr3(l3-O)(l-OH)3(l-PhCOO)3]+
was isolated from sodium benzoate in water/chloroform
[14]. Fischer and co-workers obtained the dicationic dinu-
clear di-l-hydroxo complex [{Cp2Zr(l-OH)(n-PrCN)}2]2+
by addition of n-PrCN to a mixture they formulated as
[{Cp2Zr(H2O)}2O][BPh4]2 Æ pH2O and/or [Cp2Zr(OH)-
(H2O)n][BPh4] Æ mH2O (this was prepared by addition of
Na[BPh4] to Cp2ZrCl2 dissolved in water) [15]. A related
neutral dinuclear di-l-hydroxo complex, [Cp2Zr-
(l-OH)(CF3COO)]2, was obtained by Klima and Thewalt
by treatment of Cp2ZrCl2 with CF3COOH in a CHCl3/
H2O two-phase system [16]. Erker and co-workers reported
that treatment of [Cp2ZrCH3(THF)]+ with excess water in
CH2Cl2/THF gave the trinuclear cluster [(Cp2Zr)3-
(l-OH)3(l3-O)]+ in which the cyclopentadienyl ligands are
still intact [17]. Using chloride potentiometry, Toney and
Marks studied the aqueous chemistry of Cp2ZrCl2: the rate
of the second chloride loss was found to be 1.31 hꢀ1 for 1–
3.5 mM solutions with an equilibrium [Clꢀ]:[Zr] ratio of
1.75 [18]. Thus, significant amounts of [Cp2ZrCl(H2O)x]+
and [Cp2Zr(H2O)x]2+ are present in these solutions. Some
aqueous polymerisation studies of styrene and methyl meth-
acrylate using Cp2ZrCl2 have also been reported [19]. Hill-
house and Bercaw have reported some stoichiometric
reactions of bis(pentamethylcyclopentadienyl)-zirconium
and -hafnium hydrides to give a variety of hydroxo and
oxo complexes [20].
evaporation of a wet methanol solution of 1; the ES-MS of
a CH3CN/H2O solution of 1 with added HCl(aq) gave
peaks corresponding to the expected species [(bcep)-
Zr(OH)]+ and [(bcep)ZrCl]+ in addition to species of the
formulae [(bcep)Zr(OH)O]+ and [(bcep)ZrClO]+ (no peak
corresponding to the crystallographically characterised cat-
ion [(bcep)ZrCl(H2O)]+ has been observed).
In this paper, we detail further aqueous chemistry of
complex 1 which has been studied using 31P NMR spec-
troscopy, ES-MS, and X-ray crystallography.
2. Results and discussion
Positive-ion ES-MS of dilute (ca. 10ꢀ4 M) solutions of
complex 1 in 1:1 acetonitrile–water shows the presence of
three species (Fig. 1). The masses are consistent with the
hydroxo complex [(bcep)Zr(OH)]+ (5) (derived from the
diaqua dicationic complex [(bcep)Zr(H2O)2]2+ (2)) and its
dimer [{(bcep)Zr(l-OH)}2]2+ (3), as well as the singly-
deprotonated form of 3, [{(bcep)Zr}2(l-O)(l-OH)]+ (4)
(Scheme 1). Although the diaqua dication [(bcep)Zr-
(H2O)2]2+ (2) is not observed directly by ES-MS, this is not
surprising: under ES-MS conditions, as the solvent evapo-
rates and the charge density increases, the dication 2 can
both lose solvent and reduce charge by the loss of H3O+ to
give the hydroxo complex 5. There is no evidence in these
dilute solutions for any chloro complexes such as the crystal-
lographically-characterised cation [(bcep)ZrCl(H2O)]+ [21].
The aqueous chemistry of 1 was investigated with reac-
tivity studies that were followed by 31P NMR spectroscopy.
Results are summarised in Table 1. Three major species, in a
14:2:1 ratio, are observed. These are proposed to be the dia-
qua dication [(bcep)Zr(H2O)2]2+ (2) (d 32.32), and the dinu-
clear complexes 3 (d 33.14) and 4 (d 34.36). The assignment
of the resonance at d 32.32 as the diaqua dication 2 rather
than the mononuclear hydroxo complex 5, as is observed
under ES-MS conditions, or the hydroxo-aqua complex
We recently reported the preparation and structure of
(bcep)ZrCl2 (1) along with some initial chemical investiga-
tions [21]: crystals of [(bcep)ZrCl(H2O)]Cl were isolated by
Fig. 1. ES-MS of (bcep)ZrCl2 (1) in H2O/CH3CN (cone voltage of 20 V).

408
J.R. Butchard, O.J. Curnow / Polyhedron 26 (2007) 406–414
Scheme 1. (Bcep)ZrCl2 (1) in H2O.
Table 1
31P NMR spectra of aqueous solutions of 1a
Solvent
[(bcep)2Zr2(O)(OH)]+
[{(bcep)Zr(OH)}2]2+
[(bcep)Zr(H2O)2]2+
H2O
D2O
1:1 H2O/D2O
34.36 (0.11)
34.25 (0.10)
34.34 (0.10)
33.14 (0.18)
33.01 (0.22)
33.18 (0.05)
33.15 (0.05)
33.12 (0.05)
33.09 (0.05)
33.18 (0.02)
33.15 (0.04)
33.12 (0.04)
33.09 (0.08)
33.03 (0.22)
32.32 (0.71)
32.46 (0.68)
32.40 (0.70)
2:1 H2O/D2O
34.34 (0.11)
32.40 (0.71)
25:75 H2O/CH3CN
50:50 H2O/CH3CN
65:35 H2O/CH3CN
75:25 H2O/CH3CN
33.16 (0.15)
33.11
29.12 (0.63)
w1/2 ꢁ 83 Hz
31.38
w1/2 ꢁ 50 Hz
32.09 (0.66)
w1/2 = 23 Hz
32.15 (0.69)
w1/2 = 9 Hz
32.68 (0.85)
32.68 (1.00)
33.01
34.10 (0.12)
34.17 (0.11)
34.36 (0.01)
33.09 (0.22)
33.12 (0.20)
33.14 (0.14)
H2O + 10HCl (5 min)
H2O + 10HCl (30 min)
a
Chemical shift in ppm, integral given in parentheses.
[(bcep)Zr(OH)(H2O)]+ (6) is based on the following obser-
vations: addition of one half equivalent of NaOH(aq) rap-
idly converts 2 to 3 and 4 (more than one equivalent of
hydroxide rapidly leads to hydrolysis of bcep from the com-
plex) whereas addition of aqueous HCl converts 3 and 4 to 2
over several minutes – a possible equilibrium between 3 and
its monomer 5 (or a complex such as 6) would not be
affected by addition of acid or base. Balzarek et al. have
reported that the monomer/dimer equilibrium between
[Cp4Mo2(l-OH)2]2+ and [Cp2Mo(OH)(H2O)]+ (equivalent
to an equilibrium between 3 and 6) is independent of pH
[22]. Dicationic species such as 2 also have precedence in
the form of the triaqua complex [Cp2Zr(H2O)3]2+ [11].
Dissolving complex 1 in 0.1 M HCl(aq) gives only 2.
The assignment of the resonance at d 33.14 to the di-l-
hydroxo complex 3, with the phosphorus atoms trans about
the Zr2(OH)2 core, is supported by its 31P NMR spectrum in
2:1 D2O/H2O which shows four peaks in a 4:2:2:1 ratio
(Fig. 2). The downfield resonance at 32.94 ppm is assigned
to [{(bcep)Zr(l-OD)}2]2+, the resonances at 32.91 and
32.88 ppm to the mixed species [{(bcep)Zr}2(l-OH)-
(l-OD)]2+, and the resonance at 32.85 ppm to [{(bcep)Zr-
(l-OH)}2]2+. Complex 4 could also give four peaks in this
solvent mixture, but they would be in a 2:2:1:1 ratio.
Accidental degeneracy of two of these peaks could give a
1:3:2 pattern, unfortunately, we have been unable to resolve
the resonances at ca. 34 ppm sufficiently well to confirm this
unambiguously. The resonance due to 2 is essentially

J.R. Butchard, O.J. Curnow / Polyhedron 26 (2007) 406–414
409
Fig. 2. 31P NMR spectrum of complex 1 dissolved in a 2:1 D2O/H2O mixture.
unchanged, as would be expected for rapid proton exchange
in this complex. In a 1:1 solvent mixture, both 3 and 4
would have four peaks in a 1:1:1:1 ratio; this is observed
for 3 but, again, is unresolved for 4. The related complex
[{Cp2Zr(n-PrCN)(l-OH)}2]2+, has been crystallographi-
cally characterized and also found to have the n-PrCN
substituents trans about the Zr2(OH)2 core [15]. The obser-
vation of this isotope effect deserves some comment. The
average isotope effect on the 31P NMR chemical shift is
0.045 ppm per H/D substitution, which is quite significant
considering that the substitution is made three bonds away
from the P atoms. A comparison can be made with P(CH3)3
versus P(CD3)3 in which the chemical shifts are ꢀ61.9 and
ꢀ64.9 ppm, respectively [23]. This is an average isotope
effect of 0.33 ppm per H/D exchange, for which the P atom
is two bonds from the H/D atoms.
solution for which a stable pH reading can be taken after
about 30 min. For a 10ꢀ3 M solution, the equilibrium pH
is 3.3. This is lower than the expected pH calculated
using the 31P NMR peak ratios, which gives a pH of
3.7, but is higher than would be expected if 2 was in fact
a hydroxo species like 5 or 6, in which case, the expected
pH would be less than 3.0. Similarly, a 10ꢀ2 M solution
gives a pH of 2.2, rather than 2.7. The discrepancy could
be accounted for by an equilibrium between 2 and 6,
which would be expected to be rapid and which could
also provide a proton exchange mechanism. Attempts to
measure the free chloride concentration using
a
chloride-ion selective electrode were prevented by chronic
electrode fouling.
To confirm the absence of chloro complexes, Ag(NO3)-
(aq) was added to the aqueous solution of 1. Although the
expected precipitate of AgCl rapidly formed, there was no
change in the 31P NMR spectrum. Thus, the presence of
chloro complexes can be ruled out.
Further evidence in support of the proposed structures
for 2, 3 and 4 is provided by the treatment of aqueous 1
with aqueous HClO4. One equivalent immediately precipi-
tates out only 3 and 4, as is expected from their larger size
and hydrophobic exterior (the Zr2O2 cores are sterically
shielded by the four cyclopentadienyl rings and two phenyl
groups, as evidenced by the slow rate of proton exchange in
the mixed D2O/H2O solvent systems). The concentration
of 2 is unaffected and it seems reasonable to suggest that
hydrogen bonding with the solvent keeps this species in
solution.
In an attempt to observe
a
species such as
[(bcep)ZrCl(H2O)]+ (7), aqueous NaCl was added to an
aqueous solution of 1. Although no new peaks were
observed in the 31P NMR spectrum and the peaks assigned
to 3 and 4 remained unchanged, the peak assigned to 2 at
ca. 32 ppm broadened from 6 Hz with no added chloride to
60 Hz at 1.2 M chloride. A similar broadening effect was
seen upon addition of acetonitrile (Table 1). We believe
that this broadening is due to the formation of an equilib-
rium in which the 7 is in rapid equilibrium with the dication
2. The chloroaqua cation may also have a largely non-
coordinated phosphorus atom as in complex 8 since a
non-coordinated P atom would be expected to have a
Potentiometry measurements were also carried out to
support the proposed mixture of compounds. Complex
1 dissolves in water over several minutes to give an acidic

410
J.R. Butchard, O.J. Curnow / Polyhedron 26 (2007) 406–414
chemical shift considerably upfield of 2 and similar to that
of the ferrocene complex (bcep)Fe at ꢀ16.8 ppm [8].
plex cannot be isolated, even with only one equivalent of
thiocyanate [21]. This difference in chemistry can be attrib-
uted to the p-donor effect of fluoride which disfavours
(relative to isothiocyanate) coordination of an additional
ligand.
The fluoro complex 9 was precipitated from aqueous
solution as the tetraphenylborate salt. It is important to
do this reaction in the presence of added acid (as HCl(aq)),
otherwise, an increase in pH as Fꢀ substitutes OHꢀ in 3 and
4 eventually leads to hydrolysis of the zirconocene. Spectro-
scopic evidence indicates that there is no water coordinated
in the solid state: no tOH in the infrared spectrum or addi-
tional proton resonances in the 1H NMR spectrum (CD3CN
solvent) were observed and TGA/DSC shows no changes
below 175 ꢁC – above 175 ꢁC, there are significant mass
losses (>25%). In solution, however, a weakly coordinated
water molecule cannot be ruled out. Fluoride is a much bet-
ter p-donor ligand than chloride and, thus, electronic factors
may disfavour coordination of water. As would be expected
for ES-MS conditions, the highest mass ion (401 m/e (90Zr
isotopomer) for [PhP(CH2CH2C5H4)2ZrF]+) does not con-
tain coordinated water.
Treatment of an acidified aqueous solution of (bcep)-
ZrCl2 with sulfate gives a precipitate of (bcep)Zr(SO4)
(11) that is insoluble in common solvents. Fortunately,
the precipitate appears over several minutes and X-ray
quality crystals were obtained directly from the reaction
mixture. The infrared S@O stretch was observed at
1263 cmꢀ1, similar to other Zr(IV) complexes [24]. Under
dilute conditions, complex 11 can be observed by ES-MS
at 479 m/e as a bisulfato complex, [(bcep)Zr(SO4H)]+. A
single crystal of 11 was subjected to an X-ray crystallo-
graphic analysis. The structure contains one independent
molecule in the asymmetric unit. The structure is similar
to other mononuclear (bcep)Zr complexes: the zirconium
atom is approximately trigonal bipyramidal with the two
cyclopentadienyl rings and one oxygen atom of the sulfato
ligand occupying the equatorial positions and the axial
positions occupied by the phosphorus atom and another
sulfato oxygen atom. The sulfato ligand is O,O0 chelated.
Fig. 3 shows an ORTEP plot of 11 with the atomic
Unlike the addition of chloride, the addition of one
equivalent of fluoride to aqueous solutions of 1 immedi-
ately converts 2 to the monofluoro species [(bcep)ZrF-
(H2O)x]+ (9) (x = 0 or 1), while 3 and 4 react only slowly
to give the same product (Scheme 2). Following the reac-
tion by 31P NMR spectroscopy, a doublet is observed for
9 at 27.3 ppm with a P–F coupling constant of 67 Hz. A
small doublet of doublets is also observed at 18.3 ppm with
coupling constants of 95 and 84 Hz. This compound is
assigned as the difluoro complex PhP(CH2CH2C5H4)2ZrF2
(10). This compound disappears after several minutes and
attempts to isolate a difluoro complex failed: only the oxide
of the free ligand is observed in solution and a white precip-
itate that forms after addition of a second equivalent of
fluoride contains no carbon or hydrogen. ES-MS of
aqueous 1 with added CsF(aq) gives cation 9 as the major
species with the difluoride observed as a Cs+ adduct,
[PhP(CH2CH2C5H4)2ZrF2Cs]+ (m/e = 553) with a relative
intensity of 8%. Significant peaks for hydrolysis products
[PHPh(CH2CH2C5H5)2]+ (75%) and [C20H22FPO]+ (50%)
are also observed. The chemistry with fluoride contrasts
with that of isothiocyanate in which the diisothiocyanato
complex readily forms, but the monoisothiocyanato com-
Scheme 2. Reaction of aqueous 1 with fluoride.

J.R. Butchard, O.J. Curnow / Polyhedron 26 (2007) 406–414
411
Table 2
labelling scheme and selected bond distances and angles.
Table 2 gives the crystallographic data and structure refine-
ment details.
Crystallographic data and structure refinement parameters for
(bcep)Zr(SO4) (11)
Empirical formula
Formula weight
Crystal colour
Crystal size (mm)
Temperature (K)
C20H21O4PSZr
479.62
colourless
0.39 · 0.12 · 0.07
93(2)
˚
Wavelength (A)
0.71073
Crystal system
Space group
Unit cell dimensions
monoclinic
P21/c
˚
a (A)
11.6805(13)
8.9775(13)
17.800(2)
90
˚
b (A)
The Zr–O distances in 11 are very similar with the axial
˚
c (A)
˚
distance of 2.2010(12) A only very slightly longer than the
a (ꢁ)
˚
equatorial distance of 2.1944(12) A. These distances are
b (ꢁ)
c (ꢁ)
F(000)
Volume (A )
Z
Dcalc
Absorption coefficient (mmꢀ1
h Range (ꢁ)
Limiting indices
93.110(2)
90
976
1863.8(4)
4
1.709
similar to those found in other chelated Zr(IV) sulfato
complexes, for example, in [Zr(SO4)4]4ꢀ, the Zr–O distances
3
˚
˚
range from 2.180(4) to 2.220(5) A [25] and in [{CpCo-
(P(OEt)2O)3}Zr(NO3)2(SO4)]ꢀ, the Zr–Osulfato distances
˚
are 2.232(3) and 2.164(4) A [24]. The Zr–P distance of
2.7793(5) A is in the middle of the range of distances found
in other (bcep)Zr complexes: 2.7810(8), 2.7499(7) and
2.794(2) A for (bcep)ZrCl2, (bcep)Zr(NCS)2 and
)
0.812
˚
1.75–33.41
ꢀ15 6 h 6 16
ꢀ11 6 k 6 13
ꢀ23 6 l 6 23
23474
˚
[(bcep)ZrCl(H2O)]Cl, respectively [21]. Similarly, the Zr–
CNT distances and the CNT–Zr–CNT angle are typical
of (bcep)Zr complexes reported to date [21]. The trans
angle, P–Zr–O2, of 139.54(3)ꢁ is small compared to the
other (bcep)Zr complexes in which the smallest trans angle
of 148.55(8)ꢁ is found in (bcep)Zr(NCS)2 [11]. However, this
is simply due to the chelating nature of the sulfato ligand
since the P–Zr–O1 angle of 75.68(3)ꢁ is actually larger than
the corresponding angle in (bcep)Zr(NCS)2 (71.14(8)ꢁ).
Notably, the trans angles in the five-coordinate di-l-
hydroxo complexes [Cp2Zr(CF3COO)(l-OH)]2 [16] and
[{Cp2Zr(nPrCN)(l-OH)}2]2+ [15] are similar (141.9ꢁ and
139.5ꢁ, respectively) to those found in 11. The ligand con-
Reflections collected
Independent reflections (>2r)
R indices (I > 2r(I))
4733
R1 = 0.0269,
wR2 = 0.0589
R1 = 0.0366,
wR2 = 0.0656
0.477/ꢀ0.422
R indices (all data)
ꢀ3
˚
Largest difference in peak/hole (e A
)
formation is similar to that found in [(bcep)ZrCl(H2O)]+
in which ax = ꢀ69.8ꢁ and 95.9ꢁ (ax = Cl–Zr–CNT(x)–
C(x2), (CNT(x) = centroid of atoms C(x3) to C(x7)) and
b = 2.6ꢁ (b = Cl–Zr–P–C11) [21]. The corresponding angles
in 11 are a2 = ꢀ67.3ꢁ, a3 = 98.7ꢁ and b = ꢀ3.1ꢁ. The differ-
ence in jaxj of 31.4ꢁ corresponds to staggered cyclopenta-
dienyl rings. The most significant difference is that in
[(bcep)ZrCl(H2O)]+ the phenyl plane lies between the
chloro ligand and a cyclopentadienyl ring which forces
one ring to be further from the equatorial chloro ligand
than the other: CNT–Zr–Cl angles of 120.0ꢁ and 111.9ꢁ.
In 11, the phenyl ring is approximately perpendicular to
the ZrPO2 plane (Zr–P–C(11)–C(12) = ꢀ94.7ꢁ) and the
CNT–Zr–O1 angles are consequently similar to each other
(116.44ꢁ and 115.42ꢁ).
2.1. Other electrospray mass spectrometry studies
Addition of acids HX (X = Br, NO3, HSO4, HCO3) to
10ꢀ4 M CH3CN/H2O solutions of 1 eliminates the peaks
due to the dinuclear complexes 3 and 4 and gives only
peaks corresponding to 2 and [(bcep)ZrX]+ (Table 3). Since
the structure of (bcep)Zr(SO4) contains a chelated sulfato
ligand, it seems reasonable that the complexes [(bcep)Zr-
(HCO3)]+, [(bcep)Zr(HSO4)]+, and [(bcep)Zr(NO3)]+
would also contain chelated ligands to give five-coordinate
Fig. 3. ORTEP of (bcep)Zr(SO4) (11) with 40% thermal ellisoids. Selected
bond distances and angles are Zr–O1 = 2.1944(12), Zr–O2 = 2.2010(12),
Zr–P = 2.7793(5), Zr–CNT(C23–C27) = 2.225(1), Zr–CNT(C33–C37) =
2.223(1), S–O1 = 1.5341(12), S–O2 = 1.5209(12), S–O3 = 1.4382(15), S–
˚
O4 = 1.4412(16) A, O1–Zr–O2 = 63.87(4)ꢁ, O1–Zr–P = 75.68(3)ꢁ, O2–Zr–
P = 139.54(3)ꢁ, Zr–P–C11 = 125.05(5)ꢁ, Zr–P–C21 = 105.91(6)ꢁ, Zr–P–
C31 = 108.57(6)ꢁ, CNT–Zr–CNT = 128.13ꢁ, CNT(2)–Zr–O1 = 116.44ꢁ,
CNT(2)–Zr–O2 = 100.24ꢁ, CNT(2)–Zr–P = 96.82ꢁ, CNT(3)–Zr–O1 =
115.42ꢁ, CNT(3)–Zr–O2 = 102.91ꢁ, CNT(3)–Zr–P = 94.62ꢁ.

412
J.R. Butchard, O.J. Curnow / Polyhedron 26 (2007) 406–414
Table 3
tively. Unless otherwise stated, spectra were measured at
ambient temperature with residue solvent peaks as internal
standard for 1H and 13C{1H} NMR. 31P{1H} NMR chem-
ical shifts are reported relative to external 85% H3PO4,
positive shifts representing deshielding. ES-MS were col-
lected on a VG Platform MS8 and m/e are reported for
the 90Zr isotopomer. IR spectra were obtained on a Shima-
dzu FTIR-8201PC spectrophotometer. Elemental analyses
were done by Campbell Microanalysis Services, University
of Otago, Dunedin.
Summary of ES-MS experiments with H2O/CH3CN solutions of 1 at low
cone voltage (20 V)
Reagent added
None
m/e (rel. int.)
Assignment
399(100)
797(6)
[{(bcep)Zr(OH)}n]n+ (n = 1, 2)a
[{(bcep)Zr}2(O)(OH)}]+
[(bcep)Zr(OH)]+
CO2(aq)
HBr
399(10)
443(100)
399(45)
463(100)
479(8)
399(13)
415(65)
463(10)
479(100)
399(35)
479(100)
399(15)
444(100)
553(8)
[(bcep)Zr(CO3H)]+
[(bcep)Zr(OH)]+
[(bcep)ZrBr]+
[(bcep)ZrOBr]+
HBr (after 1 h)
[(bcep)Zr(OH)]+
[(bcep)Zr(OH)O]+
[(bcep)ZrBr]+
4.2. Speciation of 1 in H2O, D2O and CD3CN solutions by
31P NMR spectroscopy
[(bcep)ZrBrO]+
H2SO4
HNO3
CsF
[(bcep)Zr(OH)]+
[(bcep)Zr(SO4H)]+
[(bcep)Zr(OH)]+
In a typical NMR experiment, 1 (2.5 mg) was dissolved
in the solvent (0.6 mL) to give a 10ꢀ2 M solution in Zr.
H2O solutions were locked with a D2O insert. A 10 s delay
was used to ensure accurate determination of integrals. The
results for a variety of solvent mixtures are given in Table 1.
[(bcep)Zr(NO3)]+
[PhP(CH2CH2C5H4)2ZrF2Cs]+
[PhP(CH2CH2C5H4)2ZrF]+
[C20H22PFO]+
401(100)
328(50)
295(75)
[PHPh(CH2CH2C5H5)2]+
4.3. Reactions of 1 followed by 31P NMR spectroscopy
a
Ratio of approximately 1:2 for n = 1 and 2, respectively.
(a) Addition of one half equivalent of NaOH(aq) gave
only peaks at d 34.4 and 33.1 in the ratio 1:1.6.
(b) Addition of one equivalent of AgNO3(aq) immedi-
ately gave a white precipitate of AgCl. The 31P reso-
nances were unchanged.
(c) Addition of one equivalent of HClO4(aq) gave an
immediate precipitate with loss of the 31P resonances
of the dinuclear complexes while the resonance due to
[(bcep)Zr(H2O)2]2+ was unchanged.
zirconocene complexes. After 1 h, the solution containing
added HBr has additional peaks corresponding to
[(bcep)ZrOBr]+ (479 m/e, rel. int. = 100) and [(bcep)ZrO-
(OH)]+ (415 m/e, rel. int. = 65), which may contain oxi-
dised phosphine, while [(bcep)ZrOH]+ and [(bcep)ZrBr]+
are now much weaker with relative intensities of 13 and
10, respectively. Similar behaviour was observed for solu-
tions containing added HCl [21].
(d) Addition of NaCl broadens the resonance of the dia-
qua dication while the other resonances are unaf-
fected: [Zr] = 1.4 · 10ꢀ2; [Clꢀ] (w1/2): 2.8 · 10ꢀ2 M
(6 Hz), 5.4 · 10ꢀ2 M (8 Hz), 16.8 · 10ꢀ2 M (14 Hz),
60 · 10ꢀ2 M (27 Hz), 120 · 10ꢀ2 (60 Hz).
3. Conclusions
In this paper, we have described some aqueous chemis-
try of the five-coordinate zirconocene dichloride complex 1.
Its speciation in water and water/acetonitrile solvents was
investigated by 31P NMR and ES-MS studies and shown
to give three species in solution. Some reactivity studies
were also carried out: treatment of aqueous solutions of 1
with various acids HX produced cationic complexes
[(bcep)ZrX]+ which were observed by ES-MS; addition of
fluoride allowed the isolation of the monofluoro complex
9, but not the difluoro complex 10; addition of sulphate
allowed the isolation of the sulfato complex 11 which was
crystallographically characterised. These reactions thus
demonstrate the potential to carry out reactions of this
zirconocene in aqueous solutions.
4.4. Speciation and reactions of 1 in CH3CN/H2O solutions
by ES-MS
In a typical experiment, a dilute solution of the reagent
was added to 10ꢀ4 M solutions of 1 in 1:1 CH3CN/H2O.
The reagents and results are summarised in Table 3.
4.5. Preparation of [PhP(CH2CH2C5H4)2ZrF]BPh4 (9)
Complex 1 (0.0339 g, 0.0745 mmol) was dissolved in
water and HCl(aq) (0.55 mmol) added after the complex
had dissolved. One equivalent of NaF (1 mL of a
0.0313 g/10 mL solution, 0.0745 mmol) was slowly added
and the solution stirred at ambient temperature for 1 h.
4. Experimental
2
4.1. General considerations
31P{1H} NMR (H2O): d 27.3 (d, JPF = 62 Hz) (s). An
aqueous solution of NaBPh4 (0.0291 g, 0.0850 mmol) was
slowly added to give immediately a white precipitate, this
was stirred for 1 h before collection by filtration and dry-
ing in air to give 0.050 g (93% yield) of 9. 1H NMR
(CD3CN): d 7.75 (m, 3H, Ph), 7.52 (m, 2H, Ph), 7.27 (m,
Solvents were distilled prior to use. Complex 1 was pre-
1
pared by the published procedure [21]. H, 13C{1H} and
31P{1H} NMR data were collected on a Varian XL-300
spectrometer operating at 300, 75 and 121 MHz, respec-

J.R. Butchard, O.J. Curnow / Polyhedron 26 (2007) 406–414
413
8H, BPh4ꢀ), 6.98 (t, J = 7.5 Hz, 8H, BPh4ꢀ), 6.83 (t,
J = 7.2 Hz, 4H, BPh4ꢀ), 6.53 (m, 2H, Cp), 6.20 (m, 2H,
Cp), 6.13 (m, 2H, Cp), 5.97 (m, 2H, Cp), 3.2–2.7 (m, 8H,
equipped with a Bruker SMART CCD area detector (using
the program SMART [26]). Processing was carried out by use
of the program SAINT [26] which applied Lorentz and polar-
isation corrections to three-dimensionally integrated dif-
fraction spots. The program SADABS [26] was utilised for
the scaling of diffraction data, the application of a decay
correction, and empirical absorption correction based on
redundant reflections. The structure was solved by the
direct methods procedure [27] and refined by least-squares
methods on F2 with anisotropic thermal parameters for all
non-hydrogen atoms. Hydrogen atoms were added as rid-
ing contributors, at calculated positions with isotropic
thermal parameters based on the attached carbon atom.
Crystal data and structure refinement parameters are given
in Table 2.
CH2); 13C{1H} NMR (CD3CN): d 137.0 (s, BPh ꢀ),
4
133.7 (d, JPC = 9 Hz, Ph), 132.3 (s, Ph), 130.1 (d,
JPC = 9 Hz, Ph), 126.7 (m, BPh4ꢀ), 123.7 (s, Cp), 122.9
(s, BPh4ꢀ), 113.0 (s, Cp), 111.9 (s, Cp), 104.3 (s, Cp),
1
2
32.7 (d, JPC = 23 Hz, PCH2), 25.5 (d, JPC = 11 Hz,
PCH2CH2), (i-Ph and i-Cp not observed); 31P{1H} NMR
2
(CD3CN): d 24.3 (d, JPF = 55 Hz). Anal. Calc. for
C44H41BFPZr: C, 73.22; H, 5.73. Found: C, 72.41; H
5.59%.
4.6. Attempted preparation of (bcep)ZrF2 (10)
Complex 1 (0.015 g, 0.033 mmol) was dissolved in D2O
and one equivalent of NaF added (1.4 mg, 0.033 mmol).
The 31P resonance at 32.77 ppm, assigned as the dication
2, rapidly disappeared with the formation of a doublet
due to the monofluoro complex 9 and a small doublet of
doublets at 18.31 ppm (2JPF = 95 and 84 Hz), assigned as
the difluoro complex 10. This doublet of doublets disap-
peared after several minutes as the resonances due to the
dinuclear complexes 3 and 4 slowly converted to that of
9. Addition of a second equivalent of NaF rapidly gave
an insoluble white precipitate and resonances consistent
with the oxide of the free ligand (31P NMR: d 40.28,
40.12 and 39.96 in a ratio of 1:2:1).
5. Supplementary material
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic Data
Centre, CCDC reference number 299554. Copies of this
information may be obtained free of charge from The
Director, CCDC, 12 Union Road, Cambridge, CB2 1EZ,
UK, fax: +44 1223 366 033, e-mail: deposit@ccdc.ac.uk
or on the web: http://www.ccdc.cam.ac.uk.
Acknowledgements
We thank Professor Brian K. Nicholson, University of
Waikato, for assistance with the ES-MS work and Dr. Cees
Lensink, Industrial Research Laboratories, Lower Hutt,
New Zealand, for the TGA/DSC results.
4.7. Preparation of PhP(CH2CH2C5H4)2ZrSO4 (11)
Complex 1 (0.011 g, 0.024 mmol) was dissolved in aque-
ous HCl (3 mL of 0.1 M HCl). After 20 min, Na2SO4
(0.0034 g, 0.024 mmol) was added. Over several minutes,
small colourless plates of the product 11 appeared
(0.01 g, 87%). The product was found to be insoluble in
common solvents. IR (KBr) (cmꢀ1): 1263 (s), 1155 (s) 935
(s), 918 (s). Anal. Calc. for C20H21O4PSZr: C, 50.08; H,
4.41. Found: C, 50.06; H, 4.62%.
References
[1] (a) P. Binger, S. Podabrin, in: E.W. Abel, F.G.A. Stone, G.W.
Wilkinson (Eds.), Comprehensive Organometallic Chemistry II, vol.
4, Elsevier, 1995, p. 439 (Chapter 7);
(b) E.J. Ryan, in: E.W. Abel, F.G.A. Stone, G.W. Wilkinson (Eds.),
Comprehensive Organometallic Chemistry II, vol. 4, Elsevier, 1995, p.
465 (Chapter 8);
4.8. Potentiometry experiments
(c) E.J. Ryan, in: E.W. Abel, F.G.A. Stone, G.W. Wilkinson (Eds.),
Comprehensive Organometallic Chemistry II, vol. 4, Elsevier, 1995, p.
483 (Chapter 9);
(d) E. Hey-Hawkins, in: E.W. Abel, F.G.A. Stone, G.W. Wilkinson
(Eds.), Comprehensive Organometallic Chemistry II, vol. 4, Elsevier,
1995, p. 501 (Chapter 10);
(e) S. Gambarotta, J. Jubb, J. Song, D. Richeson, in: E.W. Abel,
F.G.A. Stone, G.W. Wilkinson (Eds.), Comprehensive Organome-
tallic Chemistry II, vol. 4, Elsevier, 1995, p. 543 (Chapter 11);
(f) A.S. Guran, R.F. Jordan, in: E.W. Abel, F.G.A. Stone, G.W.
Wilkinson (Eds.), Comprehensive Organometallic Chemistry II, vol.
4, Elsevier, 1995, p. 589 (Chapter 12).
Dissolution of complex 1 (0.227 g, 0.5 mmol) in H2O
(50 mL) to give a 10ꢀ2 M solution was followed by pH
measurement. Total dissolution occurred within 10 min
and a stable pH reading (pH 3.3) was obtained after about
30 min. Neutralisation required two equivalents of NaO-
H(aq) with a white precipitate forming at this pH (7–8).
Attempts to follow the dissolution process or to measure
chloride-ion equilibria using a chloride-ion selective elec-
trode were prevented by chronic electrode fouling.
[2] C. Qian, J. Guo, C. Ye, J. Sun, P. Zheng, J. Chem. Soc. D (1993) 3441.
[3] K.-H. Thiele, Ch. Schliessburg, B. Neumuller, Z. Anorg. Allg. Chem.
¨
621 (1995) 1106.
4.9. Crystal structure determination
[4] (a) For examples, see: U. Siemeling, Chem. Rev. 100 (2000) 1495;
(b) D.W. Macomber, W.P. Hart, M.D. Rausch, Adv. Organomet.
Chem. 21 (1982) 1;
X-ray crystallographic data for 11 were collected from a
single crystal sample which was mounted on a glass fibre.
Data were collected using an APEX-II diffractometer
(c) N.J. Coville, K.E. du Plooy, W. Pickl, Coordin. Chem. Rev. 116
(1992) 1;

414
J.R. Butchard, O.J. Curnow / Polyhedron 26 (2007) 406–414
(d) P.J. Shapiro, Coordin. Chem. Rev. 231 (2002) 67;
(e) C. Qian, Z. Xie, Y. Huang, J. Organomet. Chem. 323 (1987)
285;
(f) J. Gra¨per, R.D. Fischer, G. Paolucci, J. Organomet. Chem.
471 (1994) 87;
(e) H.G. Alt, M. Jung, J. Organomet. Chem. 568 (1998) 127;
(f) G.K. Anderson, M. Lin, Organometallics 7 (1988) 2285.
[7] P. Barbaro, A. Togni, Organometallics 14 (1995) 3570.
[8] O.J. Curnow, G. Huttner, S.J. Smail, M.M. Turnbull, J. Organomet.
Chem. 524 (1996) 267.
(g) C. Qian, D. Zhu, J. Chem. Soc. D (1994) 1599;
(h) G. Paolucci, F. Ossola, M. Bettinelli, R. Sessoli, M. Benetollo,
G. Bombieri, Organometallics 13 (1994) 1746;
(i) T. Kauffmann, J. Ennen, H. Lhotak, A. Rensing, F. Steinse-
ifer, A. Woltermann, Angew. Chem., Int. Ed. Engl. 19 (1980) 328;
(j) T. Kauffmann, J. Ennen, K. Berghus, Tetrahedron Lett. 25
(1984) 1971.
[9] K. Doppert, Naturwissenschaften 77 (1990) 19, and references
therein.
[10] W. Lasser, U. Thewalt, J. Organomet. Chem. 302 (1986) 201.
[11] U. Thewalt, W. Lasser, J. Organomet. Chem. 276 (1984) 341.
[12] W. Lasser, U. Thewalt, J. Organomet. Chem. 311 (1986) 69.
[13] W. Lasser, U. Thewalt, J. Organomet. Chem. 275 (1984) 63.
[14] U. Thewalt, K. Do¨ppert, W. Lasser, J. Organomet. Chem. 308 (1986)
303.
[15] H. Aslan, S.H. Eggers, R.D. Fischer, Inorg. Chim. Acta 159 (1989) 55.
[16] S. Klima, U. Thewalt, J. Organomet. Chem. 354 (1988) 77.
[17] M. Niehues, G. Erker, O. Meyer, R. Fro¨hlich, Organometallics 19
(2000) 2813.
[5] (a) T. Mizuta, T. Yamasaki, H. Nakazawa, K. Miyoshi, Organomet-
allics 15 (1996) 1093;
(b) D. Seyferth, H.P. Withers, Organometallics 1 (1982) 1275;
(c) T. Mizuta, M. Onishi, K. Miyoshi, Organometallics 19 (2000)
5005;
(d) C.H. Honeyman, T.J. Peckham, J.A. Massey, I. Manners, J.
Chem. Soc., Chem. Commun. (1996) 2589;
(e) C.H. Honeyman, D.A. Foucher, F.Y. Dahman, R. Rulkens, A.J.
Lough, I. Manners, Organometallics 14 (1995) 5503;
(f) H.P. Withers, D. Seyferth, J.D. Fellmann, P.E. Garrou, S. Martin,
Organometallics 1 (1982) 1283;
[18] J.H. Toney, T.J. Marks, J. Am. Chem. Soc. 107 (1985) 947.
[19] (a) B.N. Patra, M. Bhattacharjee, J. Poly. Sci., Part A: Poly. Chem. 43
(2005) 3707;
(b) B.N. Patra, M. Bhattacharjee, J. Poly. Sci., Part A: Poly. Chem.
43 (2005) 3797.
[20] G.L. Hillhouse, J.E. Bercaw, J. Am. Chem. Soc. 106 (1984) 5472.
[21] J.R. Butchard, O.J. Curnow, S.J. Smail, J. Organomet. Chem. 541
(1997) 407.
(g) T.J. Peckham, A.J. Lough, I. Manners, Organometallics 18 (1999)
1030;
(h) A.G. Osborne, R.H. Whiteley, R.E. Meads, J. Organomet. Chem.
193 (1980) 345;
[22] C. Balzarek, T.J.R. Weakley, L.Y. Kuo, D.R. Tyler, Organometallics
19 (2000) 2927.
(i) H. Stoeckli-Evans, A.G. Osborne, R.H. Whiteley, J. Organomet.
Chem. 194 (1980) 91;
[23] P.J. Desrosiers, R.S. Shinomoto, T.C. Flood, J. Am. Chem. Soc. 108
(1986) 7964.
(j) I.R. Butler, W.R. Cullen, S.J. Rettig, Organometallics 6 (1987) 872.
[6] (a) J.H. Shin, B.M. Bridgewater, G. Parkin, Organometallics 19
(2000) 5155;
[24] Q.-F. Zhang, T.C.H. Lam, X.-Y. Li, E.Y.Y. Chan, W.-Y. Wong,
H.H.Y. Sung, I.D. Williams, W.-H. Leung, Chem. Eur. J. 11 (2005)
101.
(b) N. Leyser, K. Schmidt, H.-H. Brintzinger, Organometallics 17
(1998) 2155;
[25] Y.-L. Fu, Z.-W. Xu, J.-L. Ren, S.W. Ng, Acta Crystallogr. E 61
(2005) 1520.
(c) J.H. Shin, T. Hascall, G. Parkin, Organometallics 18 (1999) 6;
(d) C.J. Schaverien, R. Ernst, W. Terlouw, P. Schut, O. Sudmerijer,
P.H.M. Budzelaar, J. Mol. Catal. A: Chem. 128 (1998) 245;
[26] APEX-II v 2.02, SMART, XPREP, SAINT, SADABS, Bruker AXS Inc.,
Madison, Wisconsin, USA, 2005.
[27] SHELXTL v 6.12, Bruker AXS Inc., Madison, Wisconsin, USA, 2000.