Synthesis of C2 and Cs symmetric zinc complexes supported by bis(phosphinimino)methyl ligands and their use in ring opening polymerisation catalysis

Article abstract of DOI:10.1039/b207358g

Synthetic routes to zinc complexes supported by sterically demanding bis(phosphinimino)methyl ligands are reported. Two aryl-substituted ligand precursors have been utilised, [CH₂(Ph₂P=NC₆H₂Me₃-2,4,6) ₂], 1, and [CH₂(Ph₂P=NPh)(Ph₂P=NC₆H₂ Me₃-2,4,6)], 2. The second of these is the first example of an asymmetric bis(phosphinimino)methane. These ligands are converted to three-coordinate RZnX complexes (R = 1; X = Me, N(SiMe₃)₂; R = 2; X = Me) with either C₂ or C_s symmetry in solution by reaction with either ZnMe₂ or Zn[N(SiMe₃)₂]₂ in toluene. All three derivatives have been shown to exist as three-coordinate monomers in the solid state by X-ray diffraction analysis. Protonolysis of these compounds with the bulky phenol 2,4-^tBu₂C₆H₃OH or triphenylmethanol resulted in the isolation of a series of three-coordinate aryloxy or alkoxyzinc derivatives. In contrast, reaction with less sterically demanding alcohols resulted in protonation of the bis(phosphinimino)methyl ligand. A similar result was obtained from reaction of RZnMe (R = 1) and triphenylsilanol and the product is the first example of a bis(triorganosiloxy)zinc compound to be structurally characterised. All the compounds have been examined for activity in ring opening polymerisation catalysis of rac-lactide. The aryloxy and triphenylmethoxy derivatives are active catalysts; however, no evidence of true 'living' behaviour or stereocontrol of diastereomer insertion has been observed.

Full text of DOI:10.1039/b207358g

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis of C2 and Cs symmetric zinc complexes supported by
bis(phosphinimino)methyl ligands and their use in ring opening
polymerisation catalysis
Michael S. Hill*† and Peter B. Hitchcock
School of Chemistry, Physics and Environmental Science, University of Sussex, Falmer,
Brighton, UK BN1 9QJ
Received 26th July 2002, Accepted 11th October 2002
First published as an Advance Article on the web 18th November 2002
Synthetic routes to zinc complexes supported by sterically demanding bis(phosphinimino)methyl ligands
are reported. Two aryl-substituted ligand precursors have been utilised, [CH (Ph P᎐NC H Me -2,4,6) ], 1,
᎐
2
2
6
2
3
2
and [CH (Ph P᎐NPh)(Ph P᎐NC H Me -2,4,6)], 2. The second of these is the first example of an asymmetric
᎐
᎐
2
2
2
6
2
3
bis(phosphinimino)methane. These ligands are converted to three-coordinate RZnX complexes (R = 1;
X = Me, N(SiMe3)2; R = 2; X = Me) with either C2 or Cs symmetry in solution by reaction with either ZnMe2 or
Zn[N(SiMe3)2]2 in toluene. All three derivatives have been shown to exist as three-coordinate monomers in the
solid state by X-ray diffraction analysis. Protonolysis of these compounds with the bulky phenol 2,4-tBu2C6H3OH
or triphenylmethanol resulted in the isolation of a series of three-coordinate aryloxy or alkoxyzinc derivatives. In
contrast, reaction with less sterically demanding alcohols resulted in protonation of the bis(phosphinimino)methyl
ligand. A similar result was obtained from reaction of RZnMe (R = 1) and triphenylsilanol and the product is the
first example of a bis(triorganosiloxy)zinc compound to be structurally characterised. All the compounds have been
examined for activity in ring opening polymerisation catalysis of rac-lactide. The aryloxy and triphenylmethoxy
derivatives are active catalysts; however, no evidence of true ‘living’ behaviour or stereocontrol of diastereomer
insertion has been observed.
been attributed to a chain end control mechanism in which the
Introduction
sterically demanding β-diketiminate ligand amplifies the influ-
Several recent reports have described the synthesis of three- and
ence of the sterogenic centre of the last inserted monomer.1a
four-coordinate zinc complexes supported by sterically demand-
Our current efforts in this area are prompted by these obser-
vations and are a continuation of our program of study that
ing β-diketiminate ligands.1 Much of this work is motivated by
a drive toward the realisation of metalorganic catalysts that will
affect ring opening polymerisation (ROP) of cyclic ethers or
seeks to apply bulky and kinetically stabilising ligand systems
to the ‘taming’ of reactive metal centres.5–7 Historically, hetero-
esters and exert some level of stereocontrol upon successive
atom-substituted methyl ligands, (X)nH3ϪnCϪ [e.g. X = R3Si,
monomer coordination and insertion.2 Among the most prom-
n = 1–3;8 X = R2P(), R2P(), n = 1,29], have been at the fore-
inent are the alkoxyzinc β-diketiminate complexes, I, first
reported by Coates and coworkers and demonstrated to effect
front of such an approach and we were struck by the resem-
blance of the β-diketiminate ligand of I to anions produced by
the deprotonation of bis(diphenylphosphinimino)methanes in
which the phosphinimino nitrogen atoms bear aryl substituents
remarkable stereospecific monomer insertion in the living ROP
of ,-lactide.1a The biocompatible polyester poly(lactide)
(PLA) has been identified as a particularly important material
due to the renewable nature of its feedstock. As the monomer is
with comparable steric demands. We were also intrigued to dis-
cover whether variation in catalyst symmetry would affect any
most commonly available in racemic form, the tacticity and
observed stereocontrol in ROP. Similar principles have been
mechanical properties of the resultant polymer are dependent
effective in the design of catalysts for living α-olefin polymeris-
ation where, for example, isospecific insertion of propylene
has been achieved by judicious design of catalysts with overall
upon the relative rates of diastereomer insertion and the result-
ant sequence of stereocentres after ROP.3,4
C2, Cs or C1 symmetry.10 A similar approach utilising unsym-
metrical β-diketiminate ligands has very recently been des-
cribed for living CO2/epoxide polymerisation.1e We now wish to
describe the application of the bis(diphenylphosphinimino)-
methyls derived from the symmetrical methane 1 and the
unsymmetrical methane 2 to the synthesis of three-coordinate
zinc alkoxides with overall C2 or Cs symmetry, and our pre-
liminary observations of their use in the ring opening polymer-
For example, polymerisation of rac-lactide by I when R = iPr
isation of racemic lactide.
and Ar = 2,6-diisopropylphenyl results in heterotactic PLA,
containing successive pairs of -(RR)- and -(SS)- stereo-
sequences. This alternating enchainment of - and -lactide has
† Present address: Department of Chemistry, Imperial College of
Science, Technology and Medicine, Exhibition Road, South
Kensington, London, UK SW7 2AY. E-mail: mike.hill@ic.ac.uk
4694
J. Chem. Soc., Dalton Trans., 2002, 4694–4702
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b207358g

View Article Online
Table 1 Selected bond lengths (Å) and angles (Њ) for compounds 2–5
Results and discussion
The bis(mesityl)-substituted bis(diphenylphosphinimino)-
methane 1 was synthesised according to the method reported
by Bochmann and coworkers in which bis(diphenylphosphino)-
methane (dppm) is oxidised with two equivalents of mesityl
azide at 60 ЊC.11 The unsymmetrical compound 2 may be syn-
thesised by careful oxidation of dppm with a single equivalent
of mesityl azide at room temperature followed by addition
of a toluene solution of phenyl azide and heating to 60 ЊC
(Scheme 1).
2
3a
4a
5
Zn–N(1)
Zn–N(2)
Zn–C
P(1)–N(1)
P(2)–N(2)
P(1)–C(1)
P(2)–C(1)
—
—
—
1.573(3)
1.555(3)
1.816(4)
1.820(3)
1.988(4)
1.992(5)
1.932(7)
1.606(5)
1.611(5)
1.715(6)
1.705(6)
2.011(3)
2.028(3)
1.951(4)
1.609(3)
1.628(3)
1.719(4)
1.722(4)
1.977(4)
1.994(4)
1.907(4)b
1.634(4)
1.629(4)
1.717(5)
1.699(5)
N(1)–Zn–C
N(2)–Zn–C
N(1)–Zn–N(2)
Zn–N(1)–P(1)
Zn–N(2)–P(2)
—
—
—
—
—
121.7(3)
134.0(3)
104.3(2)
122.1(3)
114.5(3)
110.8(3)
112.3(3)
133.2(4)
123.5(16)
132.2(16)
103.94(12)
107.84(16)
106.63(16)
106.97(16)
107.66(17)
129.9(2)
123.56(17)c
128.14(17)d
108.29(16)
111.5(2)
120.1(2)
115.2(2)
N(1)–P(1)–C(1) 106.37(15)
N(2)–P(2)–C(1) 106.95(15)
109.5(2)
123.7(3)
P(1)–C(1)–P(2)
122.25(19)
a Values given for Zn(1)-containing molecule only. b Value for Zn–N(3).
c Value for N(1)–Zn–N(3). d Value for N(2)–Zn–N(3).
Scheme 1
Although the intermediate phosphiniminophosphine II was
not isolated, its formation was inferred from the molecular ion
at m/z 518 observed in the 70 eV mass spectrum of a sample
taken from the reaction mixture prior to addition of phenyl
azide. The 31P{1H} NMR spectrum of the same sample indi-
cated quantitative consumption of dppm and showed two
doublet signals typical of an AX spin pattern at Ϫ7.4 and Ϫ4.9
ppm, with a homonuclear 2JPP coupling of 7.3 Hz. Compound 2
was isolated as a pale yellow powder after washing with pentane
and could be obtained as analytically pure, pale yellow crystals
after crystallisation from diethyl ether. Its formation was
apparent from the appearance of a molecular ion peak at m/z
608 in the mass spectrum and the observation of a pair of
doublets at Ϫ9.5 and 0.5 ppm (2JPP = 18.2 Hz) as the only peaks
in the 31P{1H} NMR spectrum. Compounds analogous to the
intermediate II have been reported previously and have been
shown, upon careful oxidation with the appropriate chalcogen,
to yield asymmetric disubstituted methane derivatives contain-
ing both iminophosphino and chalcogenophosphino functions
[i.e. CH (Ph P᎐NR)(Ph P᎐E); E = O, S].12 Compound 2 is, how-
ever, to the best of our knowledge, the first example of a
bis(phosphinimino)methane bearing dissimilar nitrogen-bound
substituents. The unsymmetrical structure was confirmed by a
single crystal X-ray diffraction analysis performed on a sample
crystallised by slow evaporation of a diethyl ether solution in
the open atmosphere. The molecular structure of 2 along with
the numbering scheme used is illustrated in Fig. 1; selected bond
lengths and angles are given in Table 1 and details of the X-ray
analysis are given in Table 3 (see Experimental). The structure
consists of pairs of molecules that are hydrogen bonded
through the mesityl-bound nitrogen to a water molecule lying
Fig. 1 Molecular structure of 2. Water of crystallisation and H atoms
omitted for clarity (20% probability ellipsoids).
ature addition of one equivalent of dimethylzinc to toluene
solutions of 1 and 2, respectively. Reaction under these condi-
tions resulted in the generation of one equivalent of methane
and the isolation of 3 and 4 in high yield as colourless, air-
sensitive crystalline solids after crystallisation from toluene
(Scheme 2).
The amidozinc derivative [{CH(Ph P᎐NC H Me -2,4,6) }-
᎐
2
6
2
3
2
ZnN(SiMe3)2], 5, was obtained similarly by addition of a tolu-
ene solution of 1 to Zn[N(SiMe3)2]2. Compounds 3–5 have been
completely characterised by microanalysis, mass spectrometry
᎐
᎐
2
2
2
on a crystallographic 2-fold axis. The P᎐N [P(1)–N(1) 1.573(2)
᎐
Å] bond of the mesityl-containing molecular fragment is
slightly longer than that of the corresponding phenyl-linked
P(2)–N(2) measurement [1.555(3) Å], reflecting the additional
intermolecular interaction. Both distances are, however, com-
parable to values determined in other structurally characterised
phosphiniminomethanes and bis(phosphinimino)methanes.13
The zinc methyl compounds [{CH(Ph P᎐NC H Me -2,4,6) }-
᎐
2
6
2
3
2
ZnMe], 3, and [{CH(Ph P᎐NPh)(Ph P᎐NC H Me -2,4,6)}-
᎐
᎐
2
2
6
2
3
ZnMe], 4, were obtained straightforwardly by room temper-
Scheme 2
J. Chem. Soc., Dalton Trans., 2002, 4694–4702
4695

View Article Online
and multinuclear NMR spectroscopy. C2 symmetry in solution
was deduced from the simplicity of the collated NMR data for
compounds 3 and 5. The respective 31P{1H} NMR spectra con-
sisted of singlets at 29.1 and 33.1 ppm, representing downfield
shifts of 44.2 and 48.2 ppm from the resonance observed for
free 1. In the 13C{1H} NMR spectra the methanide carbon
centres in both 3 and 5 appear as triplets centred at 16.0 and
17.7 ppm, respectively, with associated 1JPC values of 143.5 and
134.2 Hz for coupling to the two equivalent phosphorus nuclei.
The asymmetric chelate structure of 4 is evident from the
observation of two doublets (2JPP = 8.5 Hz) in the 31P{1H}
NMR spectrum which display similar downfield chemical shifts
of 23.7 and 26.1 ppm. The identity of the bis(phosphin-
imino)methyl derivatives 3, 4, and 5 was confirmed by X-ray
diffraction analysis and the molecular structures are illustrated
in Figs. 2, 3 and 4 respectively. Selected bond lengths and angles
are given in Table 1 and details of the X-ray analyses presented
in Table 3.
The crystal structures of both 3 and 4 contain two independ-
ent molecules of similar geometry. In each case the molecular
parameters discussed relate to the Zn(1)-containing molecule.
As in the analogous trimethylsilyl-substituted complex [ZnMe-
{HC(Ph P᎐NSiMe ) }]14 and in several of the recently
᎐
2
3 2
described alkyl and dialkylamido zinc derivatives of bulky
β-diketiminate ligands,1 all three compounds are monomeric
and comprise a six-membered metallacyclic structure in which
the ligand behaves as a bidentate chelate and the zinc centres
are three coordinate. All three molecules contain zinc in a dis-
torted trigonal planar environment. The high degree of
coplanarity is indicated by the sums of angles subtended by the
bonds around zinc which are 360.0(8), 359.6(6) and 360.0(6)Њ
for 3, 4 and 5 respectively. The increased steric demands of the
bulky (Me3Si)2N group of 5 cause the N–P–C–P–N–Zn chelate
to adopt a twist-boat form which minimises interaction with the
mesityl substituents while the methyl derivatives 3 and 4 take on
distorted boat conformations (Fig. 5). This distortion is rather
more pronounced in the two independent molecules of 4
and decreases the Zn(1)–C(1) separation to less than 2.8 Å. A
number of structurally characterised bis(diphenylphosphin-
imino)methyl complexes have been shown to contain direct
metal to methanide carbon bonds.15 The distortion observed in
4 (and also compound 8, vide infra) indicate that the reduced
steric demands introduced by replacement of a mesityl sub-
stituent by phenyl are sufficient to allow a similar interaction
to develop in zinc derivatives of 2. This interaction is best
regarded as electrostatic in origin, given the closed shell con-
figuration of the Zn2ϩ ion. Comparable distortions have not
been observed in metal derivatives of β-diketiminate ligands
where charge is delocalised via planar sp2 carbon rather than
the tetrahedral P() centres of anions derived from 1 and 2. As
a consequence the central carbon atom of such bis(diphenyl-
phosphinimino)methyl ligands exhibits a higher degree of
carbanionic character. Recent calculations performed on the
Fig. 2 Molecular structure of the Zn(1)-containing molecule of 3.
H atoms omitted for clarity (20% probability ellipsoids).
i
square planar nickel complex [NiBr{HC(Ph P᎐NC H Pr -
᎐
2
6
3
2
2,6)2}] have shown that comparable negative charge resides on
the methanide carbon centre and the nitrogen donors of the
ligand chelate.11b The ligand frameworks of 3, 4 and 5 neverthe-
less provide evidence for considerable delocalisation of charge
via the P–N linkages. The P–C(1) bonds are shortened in com-
parison to typical P–C σ bonds while the P–N bond distances
are elongated compared to the corresponding values in 2 (Table
1) and other structurally characterised bis(phosphinimino)-
methanes [e.g. for H C(Ph P᎐NSiMe ) P–C(1) = 1.825(1), P–N
᎐
2
2
3 2
= 1.536(2) Å; P–C–P = 124.94(15)Њ].13a The decreased Zn–C(1)
distance of 4 is also accompanied by a lengthening of the
Zn(1)–N(1) and Zn(1)–N(2) bond lengths [2.011(3), 2.028(3) Å]
in comparison to the analogous measurements in 3 [1.988(4),
1.992(5) Å]. These values are, however, longer than those
usually observed in β-diketiminato derivatives of zinc bearing
similar co-ligands {e.g. in [HC(CMeNC6H3 Pr2-2,6)2ZnMe],1d
i
Fig. 3 Molecular structure of the Zn(1)-containing molecule of 4.
H atoms omitted for clarity (20% probability ellipsoids).
the Zn–N distances are 1.940(18) and 1.9428(18) Å while
in the analogous (Me3Si)2N derivative they are 1.949(2) Å1a}
but shorter than in the three-coordinate bis(3-tert-butylpyra-
zolyl)hydroborato derivative [{η2-BH2(3-tBupz)2}ZnC(CH3)3]
[2.040(5), 2.045(6) Å].16
Synthesis of bis(diphenylphosphinimino)methylzinc alkoxides
Three-coordinate alkoxyzinc β-diketiminates have been success-
fully applied in the living ROP of ,-lactide.1 We thus sought
to investigate conditions under which analogous complexes
supported by bis(diphenylphosphinimino)methyl ligands could
be prepared from compounds 3–5. Attempted protonolysis of 3
with a single equivalent of 2,6-di-tert-butyl phenol in toluene
gave no reaction even under reflux conditions due, most likely,
to the extremely hindered nature of the phenolic hydroxy
group. Conversely, treatment of 3 with equimolar quantities of
either 2,4-di-tert-butylphenol or triphenylmethanol in toluene
Fig. 4 Molecular structure of 5. H atoms omitted for clarity (20%
probability ellipsoids).
4696
J. Chem. Soc., Dalton Trans., 2002, 4694–4702

View Article Online
Fig. 5 N–P–C–P–N–Zn chelate cores of (a)3, (b)4, (c)5 and (d)8, showing the respective boat and twist-boat conformations. For 3, 4 and 8 (boat
conformation) the respective angles defined by the P(1)–C(1)–P(2) and N(1)–Zn–N(2) least square planes and the P(1)–N(1)–P(2)–N(2) plane are for
3: 14.84, 37.06Њ; 4: 48.63, 59.56Њ; and 8: 57.69, 71.36Њ. For 5 (twist-boat) the angle between the least squares planes defined by P(1)–C(1)–P(2)–N(2)
and Zn–N(1)–N(2)–P(1) is 42.15Њ.
Fig. 6 Molecular structure of 6. H atoms omitted for clarity (30%
probability ellipsoids).
Scheme 3
at room temperature resulted in the clean elimination of a
single equivalent of methane and the formation of the zinc
aryloxy and alkoxy derivatives 6 and 7 in good yield (Scheme 3).
Compounds 6 and 7 displayed similar spectroscopic features
to those described for the C2 symmetric precursors 3 and 5.
In both cases retention of a symmetrical chelated [CH-
(Ph2PNC6H2Me3-2,4,6)2]Ϫ anion was confirmed by the observ-
ation of a single resonance in the 31P{1H} NMR spectrum at
33.0 ppm for 6 and 32.6 ppm for 7. Similar treatment of 4 with
Ph3COH also provided the asymmetrically chelated alkoxyzinc
derivative 8 which was identified by the observation of two
doublets, at 28.1 and 27.0 ppm (2JPP = 7.8 Hz), in the 31P{1H}
NMR spectrum.
The molecular structures of compounds 6–8 were deter-
mined by single crystal X-ray diffraction studies. These con-
firmed the monomeric nature of all three compounds. The
results are illustrated in Figs. 6–8 and selected bond lengths and
angles are given in Table 2. A summary of the crystallographic
analyses is provided in Table 4 (see Experimental). An
approximately planar, distorted trigonal N2ZnO coordination
environment was confirmed for compounds 6, 7 and 8 with
respective sums of angles around the metal of 358.7, 358.7 and
356.85Њ. The slightly increased perturbation to planar coordin-
ation that is indicated by the last of these figures may be attrib-
uted to the pronounced boat conformation adopted by the
Fig. 7 Molecular structure of 7. H atoms omitted for clarity (20%
probability ellipsoids).
chelate ligand in 8. This permits a close Zn–C(1) contact of
2.464(4) Å [Fig. 5(d)], which is considerably shorter than that
observed in the parent methyl derivative 4 and, most likely, a
result of the increased electronegativity of the triphenyl-
methoxy substituent. Comparable effects upon the individual
P–N, P–C and Zn–N bond lengths around the chelate ring are
also observed. The Zn–N [1.928(3), 1.937(2) Å] and Zn–O
[1.837(3) Å] bonds of the aryloxy derivative 6 are notably
shorter than in any of the other compounds discussed while
J. Chem. Soc., Dalton Trans., 2002, 4694–4702
4697

View Article Online
Table 2 Selected bond lengths (Å) and angles (Њ) for compounds 6–9
6
7
8
9
Zn–N(1)
Zn–N(2)
Zn–O
1.928(3)
1.937(2)
1.837(3)
1.988(3)
1.970(3)
1.859(2)
2.034(3)
2.010(3)
1.846(3)
2.067(3)
2.081(3)
1.894(3)a
1.887(3)b
1.590(4)
1.594(4)
1.816(5)
1.814(4)
P(1)–N(1)
P(2)–N(2)
P(1)–C(1)
P(2)–C(1)
1.630(3)
1.637(3)
1.712(3)
1.709(3)
1.635(5)
1.634(3)
1.700(3)
1.716(3)
1.611(3)
1.619(3)
1.737(4)
1.728(3)
N(1)–Zn–O
N(2)–Zn–O
130.98(11)
113.53(12)
122.13(11)
124.68(11)
115.06(12)
135.92(12)
115.71(13)c
103.60(13)d
102.42(13)e
114.60(13)f
102.13(13)
121.92(18)
128.39(19)
109.8(2)
N(1)–Zn–N(2)
Zn–N(1)–P(1)
Zn–N(2)–P(2)
N(1)–P(1)–C(1)
N(2)–P(2)–C(1)
P(1)–C(1)–P(2)
Zn–O–C
114.19(11)
114.44(14)
122.76(16)
114.92(16)
110.75(15)
127.7(2)
111.93(11)
120.52(15)
108.26(14)
108.92(15)
116.56(15)
124.6(2)
105.87(13)
98.78(15)
100.01(14)
105.74(17)
105.25(17)
127.7(2)
111.0(2)
123.7(3)
133.9(3)
125.6(2)
128.3(2)
163.03(19)g
168.5(2)h
a Zn–O(1). b Zn–O(2). c N(1)–Zn–O(1). d N(1)–Zn–O(2). e N(2)–Zn–O(1). f N(2)–Zn–O(2). g Zn–O(1)–Si(1). h Zn–O(2)–Si(2).
Scheme 4
formation of dialkoxyzinc derivatives further coordinated by
the neutral bis(phosphinimino)methane 1 (Scheme 4 and see
the siloxy derivative below). This behaviour is probably related
to the size of the hydroxy-bearing reactant that is insuffici-
ently bulky to prevent rapid protonolysis of the target com-
pounds. A similar outcome has previously been observed from
reactions of alcohols and alkylzinc compounds supported
by insufficiently bulky β-diketiminate and tris(pyrazolyl)borate
ligands.1a,16
Fig. 8 Molecular structure of 8. H atoms omitted for clarity (20%
probability ellipsoids).
this latter distance is also less than that observed in previously
characterised three-coordinate zinc derivatives bearing ter-
minal phenolic substituents. For example in [(2,6-tBu2PhO)2-
Zn(PMePh2)], the Zn–O measurements are 1.844(4) and
1.864(4) Å.17 The Zn–O–C(44) angle [133.9(3)Њ] in 6 is also
wider than those of the triphenylmethoxy derivatives 7 and 8,
suggesting some level of O pπ to Zn pπ bonding. Aryloxyzinc
complexes have been discussed in this context previously.18
The work of Coates and Chisholm has demonstrated that
zinc β-diketiminate derivatives bearing larger co-ligands than
isopropoxide or tert-butoxide are inferior for ROP of rac-
lactide.1a,c This is believed to be associated with a slow rate of
initiation which results in polymers with broad molecular
weight distributions and higher than expected molecular
weights. The putative initiating aryloxy and triphenylmethoxy
groups of compounds 6, 7 and 8 were therefore judged to be
excessively bulky. Attempts to synthesise compounds bearing
smaller alkoxy substituents using the protolytic strategy out-
lined in Scheme 3 have, however, proved unsuccessful. Treat-
ment of 3 or 5 with stoichiometric quantities of methanol,
isopropanol or phenol resulted in protonation of the bis(phos-
phinimino)methyl ligand in addition to the desired elimination
of methane or hexamethyldisilazane. This was demonstrated
most clearly by the observation of free bis(phophinimino)-
methane 1 at Ϫ16.4 ppm in the 31P{1H} NMR spectra of the
crude isolated materials. These spectra also indicated that
approximately half of the starting 3 remained unreacted. It
appears likely therefore that in these cases protonation of the
desired alkoxide is more rapid than of 3 itself leading to the
Chisholm et al. have reported that trialkylsiloxyzinc
β-diketiminates and tris(pyrazolyl)borates are also active in the
polymerisation of cyclic ethers and esters.19 As an extension to
the current study we therefore targeted analogous compounds
supported by bis(diphenylphosphinimino)methyl ligands. Reac-
tion of 3 with a single equivalent of triphenylsilanol resulted in
protonation of the bis(phosphiminimo)methyl ligand in an
identical fashion to that outlined above. We have been able to
isolate the product of this reaction however, the bis(triphenyl-
siloxy)zinc derivative 9, which was separated from unreacted 3
by fractional crystallisation and identified as a mononuclear
bis(triphenylsiloxy)zinc complex by an X-ray diffraction analy-
sis. The molecular structure is shown in Fig. 9 and selected
bond distances and angles are given in Table 2. Compound 9 is
the first bis(triorganosiloxy)zinc compound to have been struc-
turally characterised.20 The distorted tetrahedral coordination
at zinc is provided by two terminal OSiPh3 fragments and a
single molecule of the bis(diphenylphosphinimino)methane 1.
The lengths of the Zn–O bonds in 9 [1.894(3), 1.887(3) Å] are
somewhat longer than in the only other structurally character-
ised compounds containing terminal (non-bridging) Zn–O–Si
linkages, the tris(pyrazolyl)borato derivatives [{η3-HB(3-
tBupz)3}ZnOSiMe3] [1.829(2) Å] and [{η3-HB(3-tBupz)-
(3,5(CF3)2pz)2}ZnOSiMe3] [1.792(2) Å].19 The Zn–O–Si angles
observed in both of these compounds [175.0(1) and 175.9(2)Њ,
respectively] are also wider than either the Zn–O(1)–Si(1)
[163.03(19)Њ] or Zn–O(2)–Si(2) [168.5(2)Њ] angles in 9. This is
4698
J. Chem. Soc., Dalton Trans., 2002, 4694–4702

View Article Online
noted carbanionic character of the PCP bridgehead carbon
centre may also direct the behaviour of these systems. For
example, the closely related compound [ZnMe{HC(Ph P᎐
᎐
2
NSiMe3)2}] has been reported to react with cumulenes such
as adamantyl isocyanate by nucleophilic addition of the
methanide to the isocyanate carbon to form a new C–C bond.14
Further studies are therefore in progress to establish the
initial products from reaction of the present zinc complexes
with enantiopure lactide and other simple organic molecules.
A fundamental requirement of these and similar zinc com-
plexes is that the ligand remains bound as an innocent spectator
during chain growth. Although reaction at the methanide
centre need not preclude the possibility of stereoselective
insertion, any behaviour of this nature will require a re-
evaluation of the mode of action of the active species. We have
noted no evidence for a tacticity bias from analysis of the poly-
(lactide) derived from ring opening by 6, 7 or 8 by selective
decoupling of the methine protons in the 1H NMR. In all
cases the data indicate the formation of effectively atactic
poly(lactide).23
Fig. 9 Molecular structure of 9. H atoms omitted for clarity (20%
probability ellipsoids).
probably a reflection of the more crowded and asymmetric
coordination environment provided by the bis(phosphin-
imino)methane and triphenylsiloxy co-ligands.
We are continuing to develop the chemistry of related ligands
(both symmetrical and unsymmetrical) bearing more sterically
demanding aryl substituents and are pursuing metathetical
routes to low-coordinate metal complexes bearing less bulky
alkoxy substituents.
Polymerisation studies
We have conducted preliminary experiments to investigate the
utility of the zinc complexes 3–9 in the polymerisation of
rac-lactide. Attempts to initiate ring opening polymersiation of
100 equivalents of monomer at room temperature in CH2Cl2
(the conditions reported to give the highest degree of stere-
ocontrol for complexes I) in all cases have been unsuccessful
and gave no appreciable evidence of polymer formation even
over a period of several weeks (<1% by 1H NMR). These reac-
tions were repeated in toluene solution at 60 ЊC. The methyl
derivatives 3 and 4, the amide 5 and the siloxide 9 were also
inactive under these conditions. The aryloxy derivative 6 and
both the triphenylmethoxy derivatives 7 and 8, however,
are active for the ROP of rac-lactide requiring 4, 5, and 2 h,
respectively, to effect >95% conversion of 100 equivalents of
monomer. This activity is lower than that reported for other
zinc-based systems,1 but comparable to a recently reported
Sn() catalyst,21 while the enhanced rate of monomer consump-
tion in 8 is most likely related to the decreased steric demands
of the unsymmetrical ligand. These data, taken in isolation,
however, are an unreliable indicator of the nature of the poly-
merisation process. The molecular weights of the isolated
polymers determined by gel permeation chromatography are, in
general, higher than expected for a genuine living process in
which each molecule of zinc catalyst mediates the growth of an
individual polymer chain. The observed polydispersities were
also somewhat erratic, ranging from effectively monodisperse
(Mw/Mn = 1.1) to values in excess of 2 for reactions undertaken
with ostensibly identical reagents and conditions. This behavior
may be associated with the relatively large initiating groups of
6–8 effectively inhibiting monomer coordination. Similar
observations have been made previously with regard to catalysts
based on zinc β-diketiminato derivatives,1 and Sn() triphenyl-
methoxides supported by sterically demanding amidinate
ligands.22 The observed activity may therefore be initiated by
minor impurities in either the catalyst or monomer inputs.
Room temperature addition of a single equivalent of rac-lactide
to 6, 7, or 8 in C6D6 in an NMR tube resulted in complete
Experimental
All reactions were conducted under an atmosphere of dry
argon and manipulated either on a double manifold vacuum
line or in a dinitrogen-filled drybox operating at less than 1 ppm
of O2. Toluene was purified by distillation from molten
sodium. Zn[N(SiMe3)2]2,24 mesityl azide,11 phenyl azide25 and
111 were synthesised by literature procedures. NMR spectra
were recorded at 300.1 (1H), 75.5 (13C), 99.4 (29Si), and 121.5
(31P) MHz in C6D6 unless otherwise stated; intensities of the
quaternary and 29Si signals were enhanced by polarisation
transfer. Chemical shifts of H and 13C NMR were referenced
1
internally to residual solvent resonances. 31P NMR were refer-
enced externally to H3PO4 (85% aqueous solution). Mass
spectra were obtained at 70 eV; in assignments, mes = 2,4,6-
trimethylphenyl. Elemental analyses were performed by SACS
at the University of North London.
[CH (Ph P᎐NPh)(Ph P᎐NC H Me -2,4,6)] 2
᎐
᎐
2
2
2
6
2
3
A toluene (20 mL) solution of mesityl azide (2.51 g, 15.6 mmol)
was added at room temperature to a toluene (30 mL) solution
of dppm (6.00 g, 15.6 mmol) causing immediate gas evolution.
This was stirred for 14 h at room temperature before a toluene
(20 mL) solution of phenyl azide (1.86 g, 15.6 mmol) was added
and the yellow solution heated to 60 ЊC for 6 h. In vacuo
removal of volatiles yielded a yellow tar, which was triturated
with pentane to produce a pale yellow powder. Crystallisation
from diethyl ether at room temperature gave 2 as pale yellow
microcrystals (8.10 g, 85%) A sample of 2 recrystallised in
diethyl ether on the bench top provided pale yellow single
crystals suitable for X-ray diffraction that proved to be a hemi-
water solvate. Anal. calc. for C40H38N2P2: C 78.91, H 6.30, N
4.60. Found: C 78.87, H 6.26, 4.52%. 1H NMR (CDCl3, 298 K):
δ 1.98 (s, 6H, o-Me), 2.18 (s, 3H, p-Me), 3.53 (t, 2H, PCH2,
2JPH = 16.8 Hz), 6.46 (d, 2H, o-Ph), 6.71 (s, 2H, m-mes) 6.93–
7.86 (m, ca. 23H, m,p-Ph, o,m,p-PhP). 13C{1H} NMR (CDCl3,
298 K): δ 21.9 (o-Me), 27.7 (p-Me), 31.4 (CH2P,), 117.5 (p-Ph),
122.8 (Ar), 123.1 (Ar), 128.1 (d, PhP, JPC = 12.8 Hz), 128.9 (Ar),
129.7 (Ar), 130.3 (Ar), 131.4 (Ar), 131.6 (Ar), 132.0 (d, PhP,
JPC = 12.8 Hz), 132.7 (d, PhP, JPC = 10.6 Hz), 134.8 (Ar),
137.3 (o-mes), 144.0 (i-mes), 151.2 (i-Ph). 31P{1H} NMR (298
K): δ Ϫ9.5, 0.5 (2JPP = 18.2 Hz). MS: m/z 608 [15%, Mϩ],
474 [20, Ph2PCH2Ph2PNPh], 398 [100, PhPCH2Ph2PNPh], 367
1
monomer consumption (by H NMR) and the appearance of
several new peaks in the 31P{1H} NMR spectrum, including
resonances close to those attributed to the free ligands 1 and 2.
Although this observation could arise from hydrolysis by
adventitious moisture in the monomer precursor, the inactivity
of all the compounds under study in CH2Cl2 and of 3, 4 and 5
at elevated temperatures mitigates against this [our lactide
(Aldrich) was sublimed prior to use and its integrity ensured by
storage in a drybox for the duration of this study]. The above
J. Chem. Soc., Dalton Trans., 2002, 4694–4702
4699

View Article Online
t
[25], 318 [40, Ph2PNmes], 290 [65, CH2Ph2PNPh], 276 [35,
Ph2PNPh], 241 [10, PhPNmes], 199 [30, PhPNPh], 183 [60], 121
[40, mesH2], 77 [20, Ph].
[{CH(Ph2PNC6H2Me3-2,4,6)2}ZnOC6H3 Bu2-2,4] 6
A solution of 2,4-di-tert-butylphenol (0.28 g, 1.37 mmol) in
toluene (10 mL) was added at room temperature to a solution
of 3 (1.0 g, 1.37 mmol) in toluene (20 mL). The pale yellow
solution was stirred for 14 h before concentration to incipient
crystallisation. Warming to dissolve precipitated solids fol-
lowed by slow cooling to room temperature afforded 6 as large
colourless crystals suitable for an X-ray diffraction study (0.98
g, 78%). Anal. calc. for C57H64N2P2OZn: C 74.38, H 7.02, N
3.04. Found: C 74.35, H 7.09, 2.95%. 1H NMR (298 K): δ 1.32
[{CH(Ph2PNC6H2Me3-2,4,6)2}ZnMe] 3
A solution of ZnMe2 in toluene (3.08 mmol, 1.4 mL of a 2.0 M
solution, Acros) was added at room temperature to a solution
of 1 (2.0 g, 3.08 mmol) in toluene (20 mL). After initial gas
evolution had subsided, the pale yellow solution was stirred for
14 h. The solution was concentrated to incipient crystallisation,
warmed to redissolve precipitated solids and then allowed to
cool slowly to provide large pale yellow crystals of 3ؒ(C7H8)0.25
suitable for X-ray diffraction analysis (1.76 g, 76%). Anal. calc.
for C44H46N2P2Zn(C7H8)0.25: C 72.95, H 6.44, N 3.72. Found: C
72.71, H 6.59, 3.73%. 1H NMR (298 K): δ Ϫ0.59 (s, 3H, ZnMe),
1.52 (s, 1H, CH), 2.11 (s, 6H, p-Me), 2.15 (s, 12H, o-Me), 6.74
(s, 4H, m-mes), 6.95–7.05 (m, 12H, m,p-Ph), 7.74–7.80 (m, 8H,
o-Ph). 13C{1H} NMR (298 K): δ Ϫ16.6 (ZnMe), 16.0 (t, CH2P,
1JPC = 143.5 Hz), 20.8 (p-Me), 21.1 (o-Me), 127.8 (o-mes), 129.3
(o-Ph), 130.5 (p-Ph), 131.9 (m-mes), 132.8 (m-Ph), 135.9
t
t
(s, 9H, Bu), 1.51 (s, 9H, Bu), 1.70 (s, 1H, CH), 2.10 (s, 6H,
p-Me), 2.13 (s, 12H, o-Me), 6.05 (d, 1H, 6-C6H3 Bu2O), 6.67 (s,
t
t
4H, m-mes), 6.75 (d, 1H, 4-C6H3 Bu2O) 6.96 (m, 12H, m,p-Ph),
7.19 (s, 1H, 3-C6H3 Bu2O), 7.75 (m, 8H, o-Ph). 13C{1H} NMR
t
(298 K): δ 18.6 (CH2P), 20.7 (p-Me), 21.5 (o-Me), 29.9
[C(CH3)3], 32.2 [C(CH3)3], 34.0 [C(CH3)3], 35.3 [C(CH3)3],
120.0 (Ar), 122.7 (Ar), 123.0 (Ar), 129.9 (Ar), 130.9 (Ar),
132.8 (Ar), 132.9 (Ar), 133.0 (Ar), 134.6 (Ar), 135.9 (Ar),
136.8 (p-mes), 142.1 (i-mes) 162.1 (i-C6H3 Bu2O). 31P{1H}
t
NMR (298 K): δ 33.0. MS: m/z 918 [15%, Mϩ], 713 [80], 517
[10], 440 [65, Ph2PCH2PNmes], 332 [25, CH3Ph2PNmes], 319
[45, Ph2PNmesH], 240 [40], 191 [100], 134 [40], 91 [85, NPh], 57
[65].
1
(p-mes), 136.8 (dd, i-Ph, JPC = 90.6 Hz), 143.5 (i-mes).
31P{1H} NMR (298 K): δ 29.1. MS: m/z 729 [2%, Mϩ], 713
[8, Mϩ Ϫ CH4], 440 [30, Ph2PCH2PNmes], 332 [14, CH3Ph2-
PNmes], 319 [20, Ph2PNmesH], 210 [35], 183 [20], 164 [5,
PNmes], 143 [50, HNmes], 91 [100, NPh], 65 [20, Zn].
[{CH(Ph2PNC6H2Me3-2,4,6)2}ZnOCPh3] 7
Compound 7 was synthesised by the same general method as
that outlined for 6 from 3 (1.0 g, 1.37 mmol) and triphenyl-
methanol (0.36 g, 1.37 mmol). Crystallisation from toluene
at room temperature afforded 7ؒ(C7H8) as large colourless
crystals suitable for an X-ray diffraction analysis (1.03 g,
70%). Anal. calc. for C69H66N2P2OZn: C 77.70, H 6.25, N 2.63.
[{CH(Ph2PNPh)(Ph2PNC6H2Me3-2,4,6)}ZnMe] 4
Compound 4 was synthesised in an analogous fashion from 2
(1.50 g, 2.47 mmol) and ZnMe2 (2.47 mmol, 1.23 mL, of 2.0 M
solution in toluene) and isolated as pale yellow crystals suitable
for X-ray analysis after crystallisation from toluene (0.97 g,
57%). Anal. calc. for C41H40N2P2Zn(C7H8)0.5: C 72.80, H 6.05,
1
Found: C 77.67, H 6.30, N 2.52%. H NMR (298 K): δ 1.72
1
(s, 1H, CH), 2.13 (s, 12H, o-Me), 2.13 (s, 6H, p-Me), 6.67
(s, 4H, m-mes), 6.96–7.31 (m, 27H, m,p-PhP, o,m,p-PhCO),
7.67 (m, 8H, o-PhP). 13C{1H} NMR (298 K): δ 17.6 (CH2P),
21.0 (p-Me), 22.1 (o-Me), 83.7 (OCPh), 125.6 (Ar), 125.9
(Ar), 127.9 (Ar), 128.8 (Ar), 130.0 (Ar), 131.0 (Ar), 132.6 (Ar),
133.3 (Ar), 135.1 (Ar), 136.4 (p-mes), 142.9 (i-mes) 154.7
(i-OCPh). 31P{1H} NMR (298 K): δ 32.6. MS: m/z 973 [1%,
Mϩ], 895 [1, M Ϫ PhH], 840 [2], 714 [5, CH(Ph2PNmes)Zn],
650 [5, CH2(Ph2PNmes)], 440 [15, Ph2PCH2PNmes], 332 [10,
CH3Ph2PNmes], 260 [45], 183 [70], 91 [100, HNPh].
N 3.82. Found: C 72.82, H 5.97, N 3.80%. H NMR (298 K):
δ Ϫ0.68 (s, 3H, ZnMe), 1.34 (s, 1H, CH), 1.56 (s, 6H, o-Me),
1.98 (s, 3H, p-Me), 6.44 (d, 2H, o-Ph), 6.87–7.18 (m, 23H,
m-mes, m,p-Ph, o,m-PhP), 7.65 (m, 4H, p-PhP). 13C{1H} NMR
(298 K): δ Ϫ13.8 (ZnMe), 19.1 (CH2P), 20.7 (o-Me), 21.5
2
(p-Me), 125.3 (o-mes), 127.5 (d, o-PhP, JPC = 11.3 Hz), 128.2
2
(Ar), 128.3 (d, o-PhP, JPC = 11.3 Hz), 128.5 (Ar), 128.7 (Ar),
3
129.0 (Ar), 130.3 (Ar), 131.0 (Ar), 131.8 (d, m-PhP, JPC = 9.8
Hz), 132.5 (d, m-PhP, 3JPC = 9.8 Hz), 133.3 (Ar), 135.6 (p-mes),
1
135.7 (dd, i-PhP, JPC = 90.6 Hz), 142.2 (i-Ph), 148.9 (i-mes).
31P{1H} NMR (298 K): δ 23.7, 26.1 (2JPP = 8.5 Hz). MS: m/z 687
[30%, Mϩ], 671 [100, Mϩ Ϫ CH4], 487 [10], 370 [15], 330 [30],
290 [20], 198 [20], 183 [70], 122 [50], 77 [25].
[{CH(Ph2PNPh)(Ph2PNC6H2Me3-2,4,6)}ZnOCPh3] 8
Compound 8 was synthesised by the same general method out-
lined for 6 from 4 (1.0 g, 1.45 mmol) and triphenylmethanol
(0.38 g, 1.45 mmol). Crystallisation from warm toluene pro-
vided 8 as colourless crystals suitable for an X-ray diffraction
study (0.70 g, 51%). Anal. calc. for C59H52N2P2OZn: C 76.00, H
[{CH(Ph2PNC6H2Me3-2,4,6)2}ZnN(SiMe3)2] 5
A solution of Zn[N(SiMe3)2]2 (2.0 g, 5.20 mmol) in toluene
(10 mL) was added at room temperature to a solution of 1
(3.38 g, 5.20 mmol) in toluene (25 mL) and the resulting pale
yellow solution stirred for 14 h. Volatiles were removed in vacuo
and the pale yellow solid produced crystallised by slow cooling
of a warm saturated solution to room temperature to afford
5 as large colourless crystals suitable for X-ray diffraction
analysis (3.89 g, 86%). Anal. calc. for C49H61N3P2Si2Zn: C
67.22, H 7.04, N 4.80. Found: C 67.70, H 7.41, 4.41%. 1H NMR
(298 K): δ 0.19 (s, 18H, SiMe3), 1.89 (s, 1H, CH), 2.10 (s, 6H,
p-Me), 2.12 (s, 12H, o-Me), 6.70 (s, 4H, m-mes), 6.98 (m, 12H,
m,p-Ph), 7.79 (m, 8H, o-Ph). 13C{1H} NMR (298 K): δ 5.5
1
5.63, N 3.01. Found: C 76.08, H 5.62, N 2.98%. H NMR
(298 K): δ 1.71 (s, 1H, CH), 1.97 (s, 6H, o-Me), 2.17 (s, 3H,
p-Me), 6.30 (d, 2H, o-Ph), 6.81–7.14 (m, 32H, m-mes, m,p-Ph,
o,m-PhP), 7.84 (m, 8H, o-PhP). 13C{1H} NMR (298 K): δ 18.9
(CH2P), 20.5 (o-Me), 21.2 (p-Me), 82.9 (OCPh), 125.8 (Ar),
126.9 (Ar), 128.3 (Ar), 128.4 (Ar), 128.5 (Ar), 128.7 (Ar), 129.0
(Ar), 129.3 (Ar), 130.4 (Ar), 131.3 (Ar), 131.6 (Ar), 132.4 (Ar),
133.4 (Ar), 135.6 (Ar), 135.7 (Ar), 141.5 (i-Ph), 142.3 (i-mes),
154.7 (OCPh). 31P{1H} NMR (298 K): δ 28.1, 27.0 (2JPP = 7.3
Hz). MS: m/z 670 [80%], 593 [10], 398 [10], 330 [15], 290 [25],
243 [20], 183 [100], 105 [85], 77 [75].
1
(SiMe3), 17.7 (t, CH2P, JPC = 134.4 Hz), 20.7 (p-Me), 21.7
2
(o-Me), 125.6 (o-mes), 127.6 (dd, o-Ph, JPC = 5.7 Hz), 130.4
[{CH2(Ph2PNC6H2Me3-2,4,6)2}Zn(OSiPh3)2] 9
(m-mes), 130.7 (p-Ph), 133.4 (t, m-Ph, JCP = 4.74 Hz) 136.3
1
(dd, i-Ph, JPC = 95.8 Hz), 136.4 (p-mes), 143.0 (i-mes).
A solution of triphenylsilanol (0.57 g, 2.07 mmol) in toluene
(20 mL) was added to a solution of 3 (1.50 g, 2.07 mmol) in
toluene (25 mL) at room temperature. The yellow solution was
stirred for 14 h before removal of solvent to provide a pale
yellow solid. 31P{1H} NMR of this displayed an approximate
1:1 mix of unreacted 3 and a signal characteristic of the
31P{1H} NMR (298 K): δ 33.1. 29Si{1H} NMR (298 K): δ Ϫ4.4.
MS: m/z 874 [2%, Mϩ], 858 [5, Mϩ Ϫ CH4], 712 [40], 440
[15, Ph2PCH2PNmes], 330 [25, CHPh2PNmes], 319 [15,
Ph2PNmesH], 240 [15], 183 [30], 164 [15, PNmes], 146 [30], 91
[100, NPh].
4700
J. Chem. Soc., Dalton Trans., 2002, 4694–4702

View Article Online
Table 3 Selected crystallographic and data collection parameters for compounds 2–5
2
3
4
5
Chemical formula
Formula weight
T /K
C40H38N2P2ؒ½(H2O)
617.67
173(2)
C44H46N2P2Znؒ¼(C7H8)
753.17
173(2)
C41H40N2P2Znؒ½(C7H8)
734.13
173(2)
C49H61N3P2Si2Zn
875.5
173(2)
Crystal system
Space group
a/Å
b/Å
c/Å
α/Њ
β/Њ
γ/Њ
Orthorhombic
Fdd2 (no.43)
26.6838(6)
53.8252(14)
9.4022(2)
90
90
90
16
Triclinic
P1 (no. 2)
Monoclinic
P21/c (no.14)
25.6605(3)
14.7185(2)
20.5047(2)
90
98.474(1)
90
8
Monoclinic
P21/c (no.14)
12.2060(2)
19.8779(4)
19.7239(5)
90
100.129(1)
90
4
¯
11.2619(5)
18.4588(13)
20.2548(13)
112.223(2)
95.261(4)
92.019(4)
4
3870.2(4)
1.29
0.75
Z
V/Å3
13504.0(5)
1.22
0.16
7659.8(2)
1.27
0.76
4711.0(2)
1.23
0.68
Dc/Mg mϪ3
µ/mmϪ1
R1, wR2 [I > 2σ(I )]
R1, wR2 (all data)
Measured/independent reflections/Rint
Reflections with I > 2σ(I )
0.043, 0.088
0.058, 0.095
10024/4977/0.050
4217
0.065, 0.124
0.110, 0.141
20269/9360/0.104
6313
0.057, 0.136
0.089, 0.152
61498/13600/0.062
9703
0.065, 0.143
0.090, 0.155
31199/7435/0.094
5775
Table 4 Selected crystallographic and data collection parameters for compounds 6–9
6
7
8
9
Chemical formula
Formula weight
T /K
C57H64N2P2OZn
920.41
173(2)
C62H58N2OP2ZnؒC7H8
1066.55
173(2)
C59H52N2P2OZn
932.34
173(2)
C79H74N2O2P2Si2Znؒ2(C7H8)
1451.16
173(2)
Crystal system
Space group
a/Å
b/Å
c/Å
α/Њ
β/Њ
γ/Њ
Triclinic
P1 (no. 2)
Monoclinic
P21/c (no.14)
12.8791(2)
26.6119(5)
16.6536(4)
90
100.941(1)
90
4
Triclinic
P1 (no. 2)
Monoclinic
P21/n (no.14)
12.7106(3)
21.5304(4)
29.7106(8)
90
101.289(1)
90
4
¯
¯
12.9453(7)
14.0557(9)
15.6630(9)
94.210(2)
106.701(3)
110.514(3)
2
2508.0(3)
1.22
0.59
10.4249(3)
12.5215(3)
19.3876(5)
96.599(1)
94.940(1)
106.035(1)
2
2397.7(1)
1.29
0.62
Z
V/Å3
5604.1(2)
1.264
0.54
7973.4(3)
1.21
0.43
Dc/Mg mϪ3
µ/mmϪ1
R1, wR2 [I > 2σ(I )]
R1,wR2 (all data)
Measured/independent reflections/Rint
Reflections with I > 2σ(I )
0.048, 0.094
0.07, 0.106
13734/6926/0.053
5141
0.049, 0.103
0.074, 0.113
27200/7853/0.064
5919
0.050, 0.093
0.078, 0.103
20170/6629/0.100
4927
0.059, 0.137
0.092, 0.154
33679/10993/0.072
7871
protonated 1. Compound 9ؒ(C7H8)2 was isolated by fractional
crystallisation from toluene at room temperature as colourless
needles suitable for an X-ray diffraction analysis (0.40 g, 27%
based upon Ph3SiOH). Anal. calc. for C93H90N2P2O2Si2Zn: C
Crystal structure determinations
Data were collected at 173 K on an Nonius Kappa CCD dif-
fractometer, λ(Mo Kα) = 0.71073 Å; details are given in Tables 3
and 4. The structures were solved by direct methods (SHELXS-
97)26 and refined by full matrix least squares (SHELXL-97)27
with non-H atoms anisotropic and H atoms included in riding
mode except in 2 where the hydrogen atoms on C(1) and on the
oxygen atom were located on a difference map and freely
1
76.97, H 6.26, N 1.93. Found: C 76.87, H 6.19, N 1.84%. H
NMR (298 K): δ 2.19 (s, 12H, o-Me), 2.25 (s, 6H, p-Me), 3.91
(t, 2H, CH2P), 6.74–6.67 (s, 4H, m-mes), 6.92–7.42 (m, 27H,
m,p-PhP, o,m,p-PhSiO), 7.76 (m, 8H, o-PhP). 13C{1H} NMR
(298 K): δ 21.6 (o-Me), 22.7 (p-Me), 29.1 (PCH2), 125.8 (Ar),
126.7 (Ar), 127.5 (Ar), 128.3 (Ar), 128.7 (Ar), 129.0 (Ar), 129.3
(Ar), 129.4 (Ar), 130.1 (Ar), 131.3 (Ar), 131.6 (Ar), 132.1 (Ar),
133.1 (Ar), 135.6 (Ar), 136.6 (Ar), 137.1 (Ar), 141.9 (i-Ph),
142.0 (i-mes), 149.9 (OSiPh). 31P{1H} NMR (298 K): δ Ϫ16.4.
MS: m/z 988 [3%], 911 [2], 854 [2], 790 [7], 712 [100], 672 [20],
635 [15], 580 [15], 475 [15], 440 [45], 332 [80], 276 [20], 240 [50],
183 [65], 77 [35].
t
refined. In 6 the C(54) Bu group was disordered and was
refined with SADI distance constraints and with the lower
occupancy sites isotropic. In 9 there were three independent,
poorly defined molecules of toluene solvate. All three were dis-
ordered, with one on a general position and two on inversion
centres. For all three, C atoms were left isotropic and H atoms
omitted, and SADI constraints were applied. A semi-empirical
absorption correction was applied in 3, 4, 5, 6, 7 and 9.
CCDC reference numbers 190727–190734.
Typical polymerisation procedure
See http://www.rsc.org/suppdata/dt/b2/b207358g/ for crystal-
lographic data in CIF or other electronic format.
rac-Lactide (typically 0.5 g) was combined with the zinc com-
plex (0.001 equivalents) in a 25 mL ampoule with a Teflon
Youngs’ stopcock. Toluene (12 mL) was then added and the
sealed ampoule heated to 60 ЊC. Monomer conversion was
monitored by integration of monomer vs. polymer methine
Acknowledgements
We thank the Royal Society for the provision of a University
Research Fellowship (M. S. H.), Dr Tony Avent for 29Si NMR
and Professor Steve Armes and Mr Jon Weaver for assistance
with gel permeation chromatography.
1
resonances in the H NMR (CDCl3) by withdrawal of 0.5 mL
aliquots, quenching with methanol (ca. 0.1 mL) and removal of
volatiles.
J. Chem. Soc., Dalton Trans., 2002, 4694–4702
4701

View Article Online
Einstein and R. G. Cavell, Inorg. Chem., 1990, 29, 808; K. V. Katti,
B. D. Santarsiero, A. A. Pinkerton and R. G. Cavell, Inorg. Chem.,
References
1 (a) M. Cheng, A. B. Attygalle, E. B. Lobkovsky and G. W. Coates,
J. Am. Chem. Soc., 1999, 121, 11583; (b) B. M. Chamberlain,
M. Cheng, D. R. Moore, T. M. Ovitt, E. B. Lobkovsky and
G. W. Coates, J. Am. Chem. Soc., 2001, 123, 3229; (c) M. Cheng,
E. B. Lobkovsky and G. W. Coates, J. Am. Chem. Soc., 1998,
120, 11018; (d ) M. Cheng, N. A. Darling, E. B. Lobkovsky
and G. W. Coates, Chem. Commun., 2000, 2007; (e) M. Cheng,
D. R. Moore, J. J. Reczek, B. M. Chamberlain, E. B. Lobkovsky and
G. W. Coates, J. Am. Chem. Soc., 2001, 123, 8738; ( f ) M. H.
Chisholm, J. C. Huffman and K. Phomphrai, J. Chem. Soc.,
Dalton Trans., 2001, 222; (g) M. H. Chisholm, J. Gallucci and
K. Phomphrai, Inorg. Chem., 2002, 41, 2785; (h) H. Hao, C. Cui,
H. W. Roesky, G. Bai, H.-G. Schmidt and M. Noltemeyer,
Chem. Commun., 2001, 1118; (i) J. Prust, A. Stasch, W. Zheng,
H. W. Roesky, E. Alexopoulos, I. Usón, D. Böhler and
T. Schuchardt, Organometallics, 2001, 20, 3825; (j) D. R. Moore,
M. Cheng, E. B. Lobkovsky and G. W. Coates, Angew. Chem., Int.
Ed., 2002, 41, 2599.
1993, 32, 5919 . For [CH2(Ph2P᎐S)(Ph2PNAr)], see: M. W. Avis,
᎐
M. Goosen, C. Elsevier, N. Veldman, H. Kooijman and A. L. Spek,
Inorg. Chim. Acta, 1997, 264, 43 . For [CH2(Ph2P᎐O)(Ph2PNAr)],
᎐
see: M. Alajarín, C. López-Leonardo, P. Llamas-Lorente and
D. Bautista, Synthesis, 2000, 2085.
13 (a) A. Müller, M. Möhlen, B. Neumüller, N. Faza, W. Massa and
K. Dehnicke, Z. Anorg. Allg. Chem., 1999, 625, 1748; (b) M. W. Avis,
C. J. Elsevier, N. Veldman, H. Kooijman and A. L. Spek,
Inorg. Chem., 1996, 35, 1518.
14 A. Kasani, R. McDonald and R. G. Cavell, Organometallics, 1999,
18, 3775.
15 See, for example, ref. 11 and: M. J. Sarsfield, M. Thornton-Pett
and M. Bochmann, J. Chem. Soc., Dalton Trans., 1999, 3329;
M. J. Sarsfield, M. Said, M. Thornton-Pett, L. A. Gerrard and
M. Bochmann, J. Chem. Soc., Dalton Trans., 2001, 822;
G. Aharonian, K. Feghali, S. Gambarotta and G. P. A. Yap,
Organometallics, 2001, 20, 2616; P. Wei and D. W.
Stephan, Organometallics, 2002, 21, 1308; M. T. Gamer, S. Dehnen
and P. W. Roesky, Organometallics, 2001, 20, 4230; M. T. Gamer and
P. W. Roesky, J. Organomet. Chem., 2002, 647, 123 . A number of
‘pincer carbene’ derivatives have also been reported; for reviews, see:
R. G. Cavell, R. P. Kamalesh Babu and K. Aparna, J. Organomet.
Chem., 2001, 617, 158; W.-P. Leung, Z.-X. Wang, H.-W. Li and
T. C. W. Mak, Angew. Chem., Int. Ed., 2001, 40, 2501.
2 For a general review of catalyst types in lactide polymerisation, see:
B. J. O’Keefe, M. A. Hillmyer and W. B. Tolman, J. Chem. Soc.,
Dalton Trans., 2001, 2215.
3 A. BenBrahim, Chem. Br., 2002, 38, 40.
4 R. E. Drumwright, P. R. Gruber and D. E. Henton, Adv. Mater.,
2000, 12, 1841.
5 M. S. Hill and P. B. Hitchcock, Angew. Chem., Int. Ed., 2001, 40,
4089.
16 I. B. Gorrell, A. Looney, G. Parkin and A. L. Rheingold,
J. Am. Chem. Soc., 1990, 112, 4068; A. Looney, R. Han, I. B.
Gorrell, M. Cornebise, K. Yoon, G. Parkin and A. L. Rheingold,
Organometallics, 1995, 14, 274.
17 D. J. Darensbourg, M. S. Zimmer, P. Rainey and D. L. Larkins,
Inorg. Chem., 1998, 37, 2852.
18 P. P. Power, J. Organomet. Chem., 1991, 400, 49; M. M. Olmstead,
P. P. Power and S. C. Shoner, J. Am. Chem. Soc., 1991, 113, 3379.
19 M. H. Chisholm and N. W. Eilerts, Chem. Commun., 1996, 853;
M. H. Chisholm, N. W. Eilerts, J. C. Huffman, S. S. Iyer, M. Pacold
and K. Phomphrai, J. Am. Chem. Soc., 2000, 122, 11845.
20 Search of Cambridge Structural Database for Zn(OSi)2 fragment,
Daresbury Laboratory, UK National Database Service.
6 M. S. Hill and P. B. Hitchcock, Organometallics, 2002, 21, 220.
7 M. S. Hill and P. B. Hitchcock, Organometallics, 2002, 21, 3258.
8 C. Eaborn and J. D. Smith, J. Chem. Soc., Dalton Trans., 2001, 1541;
C. Eaborn, K. Izod and J. D. Smith, J. Organomet. Chem., 1995, 500,
89; C. Eaborn and J. D. Smith, Coord. Chem. Rev., 1996, 154, 125;
M. F. Lappert and D. S. Liu, J. Organomet. Chem., 1995, 500, 203;
P. J. Davidson, M. F. Lappert and R. Pearce, Acc. Chem. Res., 1974,
7, 209; P. J. Davidson, M. F. Lappert and R. Pearce, Chem. Rev.,
1976, 76, 219.
9 K. Izod, Coord. Chem. Rev., 2002, 227, 153; A. Steiner, S. Zacchini
and P. I. Richards, Coord. Chem. Rev., 2002, 227, 193.
10 For recent reviews, see: G. W. Coates, Chem. Rev., 2000, 100, 1223;
G. W. Coates, J. Chem. Soc., Dalton Trans., 2002, 467; G. W. Coates,
P. D. Hustad and S. Reinartz, Angew. Chem., Int. Ed., 2002, 41,
2237.
11 (a) I. Ugi, H. Perlinger and L. Behringer, Chem. Ber., 1958, 91, 2324;
(b) S. Al-Benna, M. J. Sarsfield, M. Thornton-Pett, D. L. Ormsby,
P. J. Maddox, P. Brès and M. Bochmann, J. Chem. Soc., Dalton
Trans., 2000, 4247.
21 See, for example, A. P. Dove, V. C. Gibson, E. L. Marshall, A. J. P.
White and D. J. Williams, Chem. Commun., 2001, 283.
22 M. H. Chisholm, S. S. Iyer, D. G. McCollum, M. Pagel and
U. Werner-Zwanziger, Macromolecules, 1999, 32, 963.
23 K. B. Aubrecht, M. A. Hillmyer and W. B. Tolman, Macromolecules,
2002, 35, 644.
24 D. J. Darensbourg, M. W. Holtcamp, G. E. Struck, M. S. Zimmer
and S. A. Niezgoda, J. Am. Chem. Soc., 1999, 121, 107.
12 For (phospinimino)methylphosphines, see: K. V. Katti and
R. G. Cavell, Inorg. Chem., 1989, 28, 413; H. Schmidbaur,
G. A. Bowmaker, O. Kumberger, G. Müller and W. Wolfsberger,
Z. Naturforsch, B, 1990, 45, 476; K. V. Katti and R. G. Cavell,
Organometallics, 1988, 7, 2236; K. V. Katti, R. J. Batchelor, F. W. B.
25 I. Ugi, H. Perlinger and L. Behringer, Chem. Ber., 1958, 91, 2330.
26 G. M. Sheldrick, SHELXS-97, Program for the Solution of Crystal
Structures, University of Göttingen, Germany, 1997.
27 G. M. Sheldrick, SHELXL-97, Program for the Refinement of
Crystal Structures, University of Göttingen, Germany, 1997.
4702
J. Chem. Soc., Dalton Trans., 2002, 4694–4702