Synthesis, reactions and X-ray crystal structures of metallacrown ethers with unsymmetrical bis(phosphinite) and bis(phosphite) ligands derived from 2-hydroxy-2′-(1,4-bisoxo-6-hexanol)-1,1′-biphenyl

Article abstract of DOI:10.1016/S0022-328X(98)01170-X

Chlorodiphenylphosphine and 2,2′-biphenylylenephosphorochloridite react with 2-hydroxy-2′-(1,4-bisoxo-6-hexanol)-1,1′-biphenyl to yield the new α,ω-bis(phosphorus-donor)-polyether ligands, 2-Ph₂PO(CH₂CH₂O)₂-C ₁₂H₈-2′-OPPh₂ (1) and 2-(2,2′-O₂C₁₂H₈)P(CH₂CH ₂O)₂-C₁₂H ₈-2′-P(2,2′-O₂C₁₂H₈) (2). These ligands react with Mo(CO)₄(nbd) to form the monomeric metaflacrown ethers, cis-Mo(CO)₄{2-Ph₂PO(CH₂CH₂O) ₂-C₁₂H₈-2′-OPPh₂} (cis-3) and cis-Mo(CO)₄{2-(2,2′-O₂C₁₂H ₈)P(CH₂-CH₂O)₂-C₁₂H ₈-2′-P(2,2′-O₂C₁₂H₈)} (cis-4), in good yields. The X-ray crystal structures of cis-3 and cis-4 are significantly different, especially in the conformation of the metal center and the adjacent ethylene group. The very different ¹³C-NMR coordination chemical shifts of this ethylene group in cis-3 and cis-4 suggest that the solution conformations of these metallacrown ethers are also quite different. Both metallacrown ethers undergo cis-trans isomerization in the presence of HgCl₂. Although the cis-trans equilibrium constants for the isomerization reactions are nearly identical, the isomerization of cis-3 is more rapid. Phenyl lithium reacts with cis-3 to form the corresponding benzoyl complexes but does not react with either trans-3 or cis-4. Both the slower rate of cis-trans isomerization of cis-4 and its lack of reaction with PhLi are consistent with weaker interactions between the hard metal cations and the carbonyl oxygens in both trans-3 and cis-4.

Full text of DOI:10.1016/S0022-328X(98)01170-X

Journal of Organometallic Chemistry 580 (1999) 328–338
Synthesis, reactions and X-ray crystal structures of metallacrown
ethers with unsymmetrical bis(phosphinite) and bis(phosphite) ligands
derived from 2-hydroxy-2%-(1,4-bisoxo-6-hexanol)-1,1%-biphenyl
Maheswaran Hariharasarma, Charles H. Lake, Charles L. Watkins, Gary M. Gray *
Department of Chemistry, CHEM 201 UAB Station, The Uni6ersity of Alabama at Birmingham, 901 South 14th Street, Birmingham,
AL 35294-1240, USA
Received 28 September 1998
Abstract
Chlorodiphenylphosphine and 2,2%-biphenylylenephosphorochloridite react with 2-hydroxy-2%-(1,4-bisoxo-6-hexanol)-1,1%-
biphenyl to yield the new a,v-bis(phosphorus-donor)-polyether ligands, 2-Ph2PO(CH2CH2O)2–C12H8-2%-OPPh2 (1) and 2-(2,2%-
O2C12H8)P(CH2CH2O)2–C12H8-2%-P(2,2%-O2C12H8) (2). These ligands react with Mo(CO)4(nbd) to form the monomeric
metallacrown ethers, cis-Mo(CO)4{2-Ph2PO(CH2CH2O)2–C12H8-2%-OPPh2} (cis-3) and cis-Mo(CO)4{2-(2,2%-O2C12H8)P(CH2-
CH2O)2–C12H8-2%-P(2,2%-O2C12H8)} (cis-4), in good yields. The X-ray crystal structures of cis-3 and cis-4 are significantly
different, especially in the conformation of the metal center and the adjacent ethylene group. The very different 13C-NMR
coordination chemical shifts of this ethylene group in cis-3 and cis-4 suggest that the solution conformations of these
metallacrown ethers are also quite different. Both metallacrown ethers undergo cis–trans isomerization in the presence of HgCl2.
Although the cis–trans equilibrium constants for the isomerization reactions are nearly identical, the isomerization of cis-3 is more
rapid. Phenyl lithium reacts with cis-3 to form the corresponding benzoyl complexes but does not react with either trans-3 or
cis-4. Both the slower rate of cis–trans isomerization of cis-4 and its lack of reaction with PhLi are consistent with weaker
interactions between the hard metal cations and the carbonyl oxygens in both trans-3 and cis-4. © 1999 Elsevier Science S.A. All
rights reserved.
Keywords: Metallcrown ethers; Phosphite; Phosphinite; Carbonyl activation
Almost all of the metallacrown ethers that have been
studied to date are derived from polyethylene glycols
1. Introduction
and thus have symmetrical and conformationally flex-
ible metallacrown ether rings. Metallacrown ethers with
Metallacrown ethers are formed when a,v-bis(phos-
phorus-donor)polyether ligands chelate soft transition
unsymmetrical and conformationally rigid metal-
metals. These complexes bind alkali metal cations to
lacrown ether rings are of interest because they could
form hard–soft bimetallic complexes with the strength
interact quite differently with hard metal cations. The
of the binding depending on both the ring size and
recently synthesized polyether diol, 2-hydroxy-2%-(1,4-
conformation [1–3]. This ability is of interest because
hard–soft bimetallic complexes can function as highly
bisoxo-6-hexanol)-1,1%-biphenyl is a promising precur-
sor for such metallacrown ethers [5]. An added
active catalysts for olefin hydroformylation reactions
1
advantage of this precursor is that the methylene H-
[4].
and 13C{1H}-NMR resonances are well resolved, unlike
those of the polyethylene glycols. This suggests that
NMR spectroscopy could be used to study the solution
* Corresponding author.
E-mail address: gmgray@uab.edu (G.M. Gray)
0022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(98)01170-X

M. Hariharasarma et al. / Journal of Organometallic Chemistry 580 (1999) 328–338
329
Fig. 1. Synthetic scheme for the bis(phosphorus donor)-polyether ligands, 1 and 2.
state conformational features of the metallacrown
ethers derived from this precursor.
NMR spectrum of each ligand due to the inequivalent
phosphorus nuclei. The chemical shifts of these singlets
are assigned on the basis of 1H-coupled 31P-NMR
spectra and by comparison with the chemical shifts of
the 31P-NMR resonances of ligands with similar phos-
phorus environments [6]. The 13C-NMR spectrum of
each ligand exhibits well-resolved resonances for all of
the methylene carbons. The resonances of the carbons
two and three bonds from the phosphorus (CH2OP and
CH2CH2OP, respectively) are doublets and are readily
assigned by their chemical shifts and the relative values
In this paper, we report the synthesis and characteri-
zation of a,v-bis(phosphinite)- and a,v-bis(phosphite)-
polyether ligands derived from 2-hydroxy-2%-(1,4-
bisoxo-6-hexanol)-1,1%-biphenyl. The cis-Mo(CO)4
complexes of both ligands have been prepared, and the
solution state conformational features of the complex
with the bis(phosphite)-polyether ligand have been in-
vestigated using one- and two-dimensional NMR spec-
troscopic techniques. The X-ray crystal structures of
both complexes have been determined, and pertinent
features of these structures are described. The reactions
of both complexes with both PhLi and HgCl2 have
also been studied.
n
of the ꢀ J(PC)ꢀ coupling constants. The assignment of
the resonances for the carbons five and six bonds from
the phosphorus (CH2OAr and CH2CH2OAr, respec-
tively) is consistent with the assignment made for these
resonances in cis-4. The 1H-NMR spectrum of each
ligand exhibits four resonances in the 3.3–4.1 ppm
region, indicating that the protons attached to each
methylene carbon are chemically equivalent. No at-
tempt was made to assign these resonances.
2. Results and discussion
2.1. Synthesis and NMR characterization of the ligands
The ligands, 2-Ph2PO(CH2CH2O)2–C12H8-2%-OPPh2
2.2. Synthesis and NMR characterization of the
(1)
and
2-(2,2%-O2C12H8)P(CH2CH2O)2–C12H8-2%-
metallacrown ethers
P(2,2%-O2C12H8) (2) were prepared by the reactions of
2-hydroxy-2%-(1,4-bisoxo-6-hexanol)-1,1%-biphenyl with
chlorodiphenylphosphine and 2,2%-biphenylylenephos-
phorochloridite ester, respectively, in the presence of
triethylamine (Fig. 1). The 31P{1H}-, 13C{1H}- and
1H-NMR spectra of the crude ligands contained no
unexplained resonances, and the crude ligands were
used in the syntheses of the metallacrown ethers.
The NMR spectra of 1 and 2 show no unexpected
features. Two singlets are observed in the 31P{1H}-
We
initially
prepared
both
Mo(CO)4{2-
Ph2PO(CH2CH2O)2–C12H8-2%-OPPh2} (3) and Mo-
(CO)4{2-(2,2%-O2C12H8)P(CH2CH2O)2–C12H8-2%-P(2,2%-
O2C12H8)} (4) in good yields by the reactions of
Mo(CO)4{nbd} with ligands 1 and 2, respectively, un-
der relatively high-dilution conditions (Fig. 2). Both
complexes were initially colorless, but 4 became col-
ored when stored in the air in the solid state for several
days. This behavior is unusual as most Mo(CO)4-

330
M. Hariharasarma et al. / Journal of Organometallic Chemistry 580 (1999) 328–338
Fig. 2. Synthetic scheme for the metallacrown ether complexes, cis-3 and cis-4.
2.3. Solution conformations of the metallacrown ethers
{bis(phosphite)} complexes are relatively air stable.
Storing this complex in the dark under nitrogen elimi-
nated this problem.
To gain more insight into the solution conformation
of cis-4, we have completely assigned the 1H- and
13C-NMR resonances of the methylene groups and
calculated the geminal and vicinal H–H coupling con-
stants. We would also have liked to carry out this same
analysis for cis-3, however, unlike cis-4, cis-3 does not
exhibit well-resolved resonances for all of the methylene
carbons and protons. In addition, cis-3 slowly isomer-
izes to trans-3 in solution, which prevents accurate 2-D
NMR spectra from being obtained. The 13C{1H}-NMR
resonances of this complex were assigned by compari-
son with those of cis-4.
The 31P{1H}-NMR spectrum of 4 displays a single
AX pattern as expected. Both the coordination chemi-
cal shifts of these resonances (ca. 30 ppm) and the
2
ꢀ J(PP%)ꢀ coupling constant of 48 Hz indicate that only
the cis isomer of 4 (cis-4) is present. In contrast, the
31P-NMR spectrum of 3 displays two AX spectral
patterns. Both the coordination chemical shifts and
coupling constants demonstrate that the major product
(ca. 30 ppm, 34 Hz) is the cis isomer of 3 (cis-3) and the
minor product (ca. 47 ppm, 89 Hz) is the trans isomer
of 3 (trans-3).
The 13C{1H}- and 1H-NMR resonances of the
methylenes in cis-4 were assigned using a variety of
two-dimensional NMR techniques [7]. A 31P{1H} HET-
COR NMR experiment was used to assign the reso-
nances of the methylenes closest to the phosphorus
(CH2OP; H4% & H4) because these are the only
methylene protons that are coupled to the phosphorus.
A phase-sensitive 1H–1H NOESY NMR experiment
was used to assign the proton resonances of the
methylene closest to the 2,2%-biphenoxy group
(CH2OAr; H1 and H1%) because both protons show
NOESY contacts to the 3 proton of the 2,2%-biphenoxy
The 13C-NMR spectra of cis-3 and cis-4 exhibit two
interesting features. The first is that well resolved reso-
nances are observed for all four carbonyls, which is
extremely unusual for cis-Mo(CO)4(P-donor ligand)2
complexes. The carbonyls trans to the phosphorus-
donor groups are chemically inequivalent because the
two phosphorus-donor groups in the ligands are differ-
ent. The inequivalence of the carbonyls cis to both
phosphorus-donor groups cannot be due to this and
instead must be due to an unsymmetrical arrangement
of the metallacrown ether ring relative to the plane
containing the molybdenum and the two phosphorus-
donor groups.
1
group. A H–1H COSY-45 NMR experiment was run
to determine the remaining 1H–1H connectivity, and
A second interesting aspect of the 13C-NMR spectra
of cis-3 and cis-4 is that the coordination chemical
shifts of equivalent methylene 13C-NMR resonances in
cis-3 and cis-4 are quite different. The largest differ-
ences are in the coordination chemical shifts of the
CH2OP resonances (3.7 ppm upfield from that of the
free ligand, 1, in cis-3 vs. 3.8 ppm downfield of that of
the free ligand, 2, in cis-4). These shifts cannot be
explained by electronic effects of the phosphorus donor
substituents and must instead be due to different aver-
age solution conformations of the metallacrown ether
carbons. This hypothesis is consistent with the fact that
the metallacrown ether rings of cis-3 and cis-4 have
very different solid state conformations at the CH2OP
carbon (see Section 2.4).
1
the H–13C connectivity was then determined using a
13C{1H} HETCOR NMR experiment.
2
The magnitudes of the geminal, ꢀ J(H,H)ꢀ, and vici-
3
nal, ꢀ J(H,H)ꢀ, coupling constants of the diastereotopic
OCH2CH2O fragments of cis-4 (Fig. 3) provide signifi-
cant insight into the average conformations [7] about
the OCH2CH2O groups in the metallacrown ether ring.
The geminal coupling constants are ca. 12 Hz for both
OCH2CH2O groups, consistent with the tetrahedral
geometry around each methylene carbon. One moder-
ate vicinal coupling constant of 7–8 Hz and three small
vicinal coupling constants of 3–4 Hz are observed for
each of the OCH2CH2O groups. The magnitudes of the
vicinal coupling constants indicate that the average
solution conformation about the C–C bonds in each of

M. Hariharasarma et al. / Journal of Organometallic Chemistry 580 (1999) 328–338
331
2
Fig. 3. The magnitudes of geminal ꢀ J(H,H))ꢀ and vicinal coupling
3
constants ꢀ J(H,H)ꢀ of the diastereotopic OCH2CH2O fragments of
cis-4.
Fig. 5. ORTEP drawing of the molecular structure of cis-4. The
hydrogens are omitted for clarity, and the thermal ellipsoids are
drawn at the 20% probability level.
the OCH2CH2O groups in the metallacrown ether ring
is gauche. Gauche conformations about these bonds are
also observed in the solid state structure of cis-4, as
discussed below.
by very different phosphorus–molybdenum–phospho-
rus bond angles (102.4(1)° in cis-3 vs. 91.7(1)° in cis-4).
This suggests that the diphenylphosphinite groups in
cis-3 are sterically more demanding than are the diben-
zo[d, f ][1,3,2]dioxaphosphepin groups in cis-4. This con-
clusion is consistent with the expected cone angles of
phosphorus-donor groups [8–11]. The lack of steric
strain in cis-4 is also indicated by the dihedral angles of
the 2,2%-biphenyl groups of the dibenzo[d, f ][1,3,2]-
dioxaphosphepin groups, 44.6 and 47.8°, which are very
similar to each other and are also similar to those in other
dibenzo[d, f ][1,3,2]dioxaphosphepin ligands [12].
2.4. Solid-state structures of the complexes
The molecular structures of cis-3 and cis-4 are shown
in Figs. 4 and 5, respectively. These structures are of
interest because, unlike other metallacrown ethers
whose structures have been reported, these metal-
lacrown ethers are unsymmetrical and sterically
crowded.
The coordination geometry of the molybdenum in
each complex is a distorted octahedron. The distortion
is significantly greater in cis-3 than in cis-4 as indicated
The conformations of the metallacrown ether rings in
cis-3 and cis-4 are the most interesting aspects of the
structures. The biphenyl groups and the adjacent
OCH2CH2O groups have similar confirmations in the
two complexes. The dihedral angles between mean least
square planes of the phenylene rings in cis-3 and cis-4
are quite similar (65.3 and 63.1°, respectively) and are in
the expected range for ortho-disubstituted biphenyl com-
pounds [12]. Moreover, gauche conformations for the
–O2–C13–C14–O3– groups adjacent to the biphenyl
group are observed in both cis-3 (67.3°) and cis-4
(−68.2°). In contrast, the metal centers and the adjacent
OCH2CH2O groups have very different conformations in
cis-3 and cis-4. In cis-3, the –O3–C15–C16–O4– group
has a trans anti conformation (178.2°) while in cis-4, it
has a gauche conformation (−68.4°). These differences
demonstrate that the phosphorus substituents have a
significant influence on the solid state conformations of
the complexes. It is interesting to note that the conforma-
tions of the OCH2CH2O groups in cis-4 are the same in
the solid state and in the average solution state confor-
mation of the metallacrown ether.
Fig. 4. ORTEP drawing of the molecular structure of cis-3. The
hydrogens are omitted for clarity, and the thermal ellipsoids are
drawn at the 20% probability level.

332
M. Hariharasarma et al. / Journal of Organometallic Chemistry 580 (1999) 328–338
absence of HgCl2. Because the isomerization pre-
sumably occurs via coordination of the HgCl2 to both
the metallacrown ether and a carbonyl oxygens, the
relative rates of isomerization suggest that the HgCl2
does not interact as strongly with the carbonyl oxygens
in cis-4.
2.5. Cis–trans isomerization equilibria of the
metallacrown ethers
As discussed above, the reaction of Mo(CO)4(nbd)
and 1 in THF under moderately high dilution condi-
tions yields a mixture of cis-3 and trans-3. Because
Mo(CO)4(nbd) reacts with phosphorus-donor ligands to
exclusively form cis isomers, the trans-3 must form via
isomerization of cis-3. To prove this, we carried out the
reaction of 1 and Mo(CO)4(nbd) in hexanes. Both 1
and Mo(CO)4(nbd) are sparingly soluble in hexanes
while both cis-3 and trans-3 are insoluble. Under these
conditions, only the cis-3 precipitates from the reaction
mixture. Surprisingly, we did not observe formation of
oligomeric complexes in spite of the fact that the reac-
tion was carried out with relatively high concentrations
of 1 and Mo(CO)4(nbd). This suggests that the steric
constraints in 1 favor formation of monomeric metal-
lacrown ethers.
The fact that cis-3 readily undergoes isomerization
while cis-4 does not suggests that the phosphorus sub-
stituents have a significant effect on this isomerization.
To better understand this phenomenon, we have stud-
ied the HgCl2 catalyzed isomerizations [13] of cis-3 and
cis-4, shown in Fig. 6. These studies were carried out by
adding solid HgCl2 to chloroform-d1 solutions of the
complexes in a mole ratio of 1:100. For cis-3, the
cis–trans equilibrium is established in less than 5 min,
which is the time needed to add solid HgCl2 to a freshly
prepared chloroform-d1 solution of cis-3, shake the
mixture vigorously, and record a non-quantitative 31P-
NMR spectrum. The equilibrium constant for the reac-
tion at 294 K, calculated using Keq=[trans-3]/[cis-3], is
3.690.4 [14]. For cis-4, the equilibrium is reached in
20 min. The equilibrium constant for the reaction at
294 K, calculated using Keq=[trans-3]/[cis-3], is 4.29
0.4.
In contrast to the differences in the rates of the
cis–trans isomerization reactions of cis-3 and cis-4, the
cis–trans equilibrium constants for the two metal-
lacrown ethers are not statistically different (assuming a
5% error in the integration of the two resonances in the
quantitative 31P-NMR spectra). The similarity of the
equilibrium constants indicates that they are not
strongly affected by the phosphorus substituents. How-
ever, these equilibrium constants are significantly larger
than those reported for cis-Mo(CO)4{Ph2P(CH2CH2-
O)4CH2CH2PPh2-P,P%}, and cis-Mo(CO)4{Ph2PCH2-
CH2OCH2CH2O-1,2-C6H4–OCH2CH2OCH2CH2PPh2-
P,P%}, 1.0 and 0.23, respectively [13]. This indicates that
the magnitudes of the cis–trans equilibrium constants
are quite sensitive to the nature of the metallacrown
ether ring. The fact that metallacrown ethers with a
bulky 2,2-biphenoxy group adjacent to the phosphorus
have larger cis–trans equilibrium constants than those
with less bulky groups adjacent to the phosphorus is
not surprising because the position of equilibrium for
cis–trans isomerization in cis-Mo(CO)4(phosphorus-
donor)2 complexes is known to parallel the steric re-
quirement of the phosphorus-donor ligand [10,11].
2.6. Reactions of metallacrown ethers with PhLi
Previous studies by Powell’s group [1] have estab-
lished that the coordination of Li+ to both a metal-
lacrown ether and a carbonyl oxygen activates the
cis-carbonyl ligands in cis-Mo(CO)4-metallacrown
ethers to nucleophilic attack by PhLi reagents. This
mechanism is similar to that which we have proposed
for the HgCl2 catalyzed cis–trans isomerization of the
metallacrown ethers [13]. On the basis of the difference
in the rates of the cis–trans isomerization reactions of
cis-3 and cis-4, it seemed likely that these metallacrown
ethers should also react differently with PhLi, and this
is indeed the case. When PhLi is added to a benzene-d6
solution of cis-4, no reaction is observed. In contrast,
when an excess of PhLi is added to a benzene-d6
solution mixture of cis-3 (major) and trans-3 (minor),
the solution first becomes orange, and then an orange
solid slowly precipitates. The IR spectrum of the orange
solution contains three new, strong CO absorptions at
1942, 1873 and 1838 cm−1 (80–100 cm−1 less than the
absorptions due to cis-3). The 31P-NMR spectrum of
the orange solution contains a new AX spectral pattern
that is ca. 20 ppm downfield from that of cis-3 in
addition to the unchanged AX pattern of trans-3. These
The rates of the HgCl2-catalyzed cis–trans isomeriza-
tion for the two complexes are quite different, with that
of cis-3 being significantly faster. The relative rates of
cis–trans isomerization are consistent with the fact that
only cis-3 is observed to isomerize in solution in the
Fig. 6. HgCl2 catalyzed cis–trans isomerization of the cis-metal-
lacrown ethers, cis-3 and cis-4.

M. Hariharasarma et al. / Journal of Organometallic Chemistry 580 (1999) 328–338
333
a Bruker ARX-300 NMR spectrometer with a quad (1H-,
13C-, 19F-, 31P-) 5 mm probe. The 31P{1H}-NMR spectra
were referenced to external 85% phosphoric acid, while
1
both the 13C{1H}- and the H-NMR spectra were refer-
enced to internal SiMe4. In all cases, downfield was
treated as positive. Quantitative 31P-NMR spectra of
chloroform-d solutions of the complexes were acquired
using an inverse-gated 30° pulse sequence with a 30-s
delay between acquisitions. Two-dimensional 1H–1H
COSY-45, 31P{1H} HETCOR, 13C{1H} HETCOR and
1H–1H NOESY spectra of 0.2 M chloroform-d solutions
of the complexes were recorded using a Bruker DRX-400
NMR spectrometer. All two-dimensional data were pro-
cessed using standard Bruker programs. The IR spectra
of dilute benzene solutions of the complexes in the
2200–1800 cm−1 region were taken on a Bruker Vector
22 FTIR spectrometer. Elemental analyses were per-
formed by Atlantic Microlabs, Norcross, GA. A solvent
of crystallization is included in a calculated analysis only
if the solvent was observed in the NMR spectrum of the
analytical sample.
All reactions and purification procedures were carried
out under high-purity nitrogen. All starting materials
were reagent grade and were used as received. THF was
distilled from sodium/benzophenone under high-purity
nitrogen. Triethylamine was distilled from calcium hy-
dride prior to use. Deuterated NMR solvents (chloro-
form-d, benzene-d6 and THF-d8) were opened and
handled under a nitrogen atmosphere at all times. The
starting materials, 2-hydroxy-2%-(1,4-bisoxo-6-hexanol)-
1,1%-biphenyl [5], cis-Mo(CO)4(nbd) [15], and 2,2%-
biphenylylene phosphorochloridite ester [16], were
synthesized using literature procedures. Phenyllithium
solution was purchased from Aldrich Chemical Com-
pany and used from newly opened bottles. All complexes
were stored in the dark under dry nitrogen.
Fig. 7. Reaction of PhLi with the cis-metallacrown ethers, cis-3 and
cis-4.
data indicate that PhLi reacts with cis-3 to give
fac - Mo(CO)3(PhC(O)){2 - Ph2PO(CH2CH2O)2 – C12H8-
2%-OPPh2} as shown in Fig. 7. A unique aspect of this
reaction is that although cis-3 has two chemically in-
equivalent cis-carbonyls that can potentially react with
PhLi, only a single product is formed. This suggests that
one of the cis-carbonyls preferentially reacts with the
PhLi.
The nature of the metallacrown ether obviously has a
dramatic effect on its ability to react with PhLi. Powell
has proposed that the difference in the reactivities of the
carbonyl ligands in bis(phosphite) metallacrown ethers
such as cis-4 and bis(phosphinite) metallacrown ethers
such as cis-3 is due to the fact that phosphites are poorer
electron donors than phosphinites [1d]. However, Powell
has also demonstrated that PhLi can coordinate to
phosphite oxygens [1f]. This could prevent carbonyl
activation in the same manner that additional ether
oxygens do [1]. Our solution and conformational studies
have now demonstrated that the phosphorus substituents
also have a dramatic effect on the solution and solid state
conformations of the metallacrown ethers. These differ-
ences could also affect the ability of a Li+ to coordinate
to a carbonyl ligand.
The fact that trans-3 does not react with PhLi cannot
be due to electronic or coordination effects and must be
due to differences in the conformations of cis-3 and
trans-3. Our studies of the structures of cis- and trans-
metallacrown ethers with bis(phosphine)polyether lig-
ands have demonstrated that ether oxygens in the
trans-metallacrown ethers are significantly farther apart
than they are in the cis metallacrown ethers with the same
ligands [13]. This suggests that trans-3 does not react with
PhLi because the ether oxygens are unable to adopt a
conformation that allows Li+ to coordinate to both the
metallacrown ether oxygens and a carbonyl oxygen.
3.1. 2-Ph2PO(CH2CH2O)2–C12H8-2%-OPPh2 (1)
A
solution of 4.44
g
(20.0 mmol) of
chlorodiphenylphosphine in 75 ml of THF was added to
a solution of 2.74 g (10.0 mmol) of 2-hydroxy-2%-(1,4-
bisoxo-6-hexanol)-1,1%-biphenyl and 2.80 ml (20.0 mmol)
of triethylamine in 100 ml of THF at room temperature
(r.t.). The mixture was stirred for 15 h and then was
filtered to remove the triethylamine hydrochloride pre-
cipitate. The filtrate was evaporated to dryness to give
6.23 g (96.9%) of crude 1 as a colorless oil. 31P{1H}-NMR
(chloroform-d): l 115.31 (ArOP, s), 110.67 (CH2OP, s).
13C-NMR (aliphatic carbons, chloroform-d): l 71.62
3
(CH2CH2OP, d, ꢀ J(PC)ꢀ 7 Hz), 69.33 (CH2OP, d,
2
3. Experimental
ꢀ J(PC)ꢀ 18 Hz), 69.14 (CH2CH2OAr, s), 68.07 (CH2OAr,
s). 1H-NMR (chloroform-d): l 7.50–6.60 (m, 28H,
P-C6H5 and C12H8), 3.70 (m, 2H, CH2OAr), 3.58 (m, 2H,
CH2O), 3.42 (m, 2H, CH2O), 3.33 (m, 2H, CH2OP).
All one-dimensional 31P{1H}-, 13C{1H}- and 1H-NMR
spectra of the ligands and complexes were recorded using

334
M. Hariharasarma et al. / Journal of Organometallic Chemistry 580 (1999) 328–338
yield 1.52 g (91.6%) of crude cis-3 as a light brown
solid. 31P{1H}-NMR (freshly prepared chloroform-d): l
3.2. 2-(2,2%-O2C12H8)P(CH2CH2O)2–C12H8-
2%-P(2,2%-O2C12H8) (2)
2
146.25 (d, ArOP, ꢀ J(PP%)ꢀ 37 Hz), 141.43 (d, CH2OP,
2
ꢀ J(PP%)ꢀ 37 Hz). 13C-NMR (carbonyl and aliphatic
A solution of 8.70 g (31.7 mmol) of 2-hydroxy-2%-
(1,4-bisoxo-6-hexanol)-1,1%-biphenyl, 15.9
carbons, chloroform-d): l 215.02 (trans CO, dd,
g
(63.4
2
2
ꢀ J(PC)ꢀ 29 Hz, ꢀ J(P%C)ꢀ 8 Hz), 214.52 (trans CO, dd,
mmol) of 2,2%-biphenylylenephosphochloridite ester and
13.8 ml (99.0 mmol) triethylamine in 150 ml of freshly
distilled dichloromethane was stirred under nitrogen for
12 h. Then, the solvent was removed under reduced
pressure, and 50 ml of THF was added to the residue.
This mixture was stirred at r.t. for 30 min before being
filtered to remove the insoluble triethylamine hy-
drochloride. The filtrate was evaporated to dryness to
yield 20.4 g (91.6%) of crude 2 as a viscous, colorless oil
that slowly solidified upon standing. 31P{1H}-NMR
(chloroform-d): l 145.21(s, ArOP), 140.79 (s, CH2OP).
13C-NMR (aliphatic carbons, chloroform-d): l 70.87
2
2
ꢀ J(PC)ꢀ 31 Hz, ꢀ J(P%C)ꢀ 9 Hz), 210.40 (cis CO, dd,
2
2
ꢀ J(PC)ꢀ 12 Hz, ꢀ J(P%C)ꢀ 12 Hz), 207.53 (cis CO, dd,
2
2
ꢀ J(PC)ꢀ 10 Hz, ꢀ J(P%C)ꢀ 10 Hz), 70.79 (CH2CH2OP, d,
3
ꢀ J(PC)ꢀ 8 Hz), 69.01 (CH2CH2OAr, s), 68.81 (CH2OAr,
2
1
s), 65.52 (CH2OP, d, ꢀ J(PC)ꢀ 7 Hz). H-NMR (chloro-
form-d): l 7.85–6.55 (m, 28H, P–C6H5 and C12H8),
4.18 (m, 1H, CH2OAr), 3.95 (m, 1H, CH2OAr), 3.77
(m, 1H, CH2O), 3.53 (m, 2H, CH2OP), 3.40–3.23 (m,
3H, CH2O). wCO (benzene-d6): 2025, 1931, 1915, 1905
cm−1
.
3.5. cis-Mo(CO)4{2-(2,2%-O2C12H8)P(CH2CH2O)2–
C12H8-2%-P(2,2%-O2C12H8)} (cis-4)
3
(CH2CH2OP, d, ꢀ J(PC)ꢀ 3 Hz), 69.28 (CH2CH2OAr, s),
2
68.46 (CH2OAr, s), 63.81 (CH2OP, d, ꢀ J(PC)ꢀ 5 Hz).
1H-NMR (chloroform-d):
l 7.51–6.80 (m, 24H,
Solutions of 1.07 g (3.56 mmol) of Mo(CO)4(nbd) in
100 ml of THF and 2.56 g (3.56 mmol) of 2 in 100 ml
of THF were added simultaneously and dropwise to
1000 ml of THF over a 2 h period. The mixture was
stirred for additional 5 h and then evaporated to dry-
ness to yield a light brown, oily residue. The residue
was taken up in a 50 ml of a 1:1 dichloromethane–hex-
anes mixture. This solution was treated with 15 g of
silica gel and then filtered. The silica gel was washed
with two, 50 ml portions of a 1:1 dichloromethane–
hexanes mixture. The filtrate and washes were com-
bined and evaporated to dryness to yield 3.03 g (93.5%)
of cis-4 as a white powder. Recrystallization from ethyl
acetate–hexanes gave analytically pure cis-4 as color-
less crystals. Anal. Calc. for C44H32O12P2Mo: C, 58.04;
H, 3.54. Found: C, 57.94; H, 3.60%. 31P{1H}-NMR
OPO2C12H8 and C12H8), 4.05 (m, 2H, CH2OAr), 3.83
(m, 2H, CH2O), 3.62 (m, 2H, CH2O), 3.32 (m, 2H,
CH2OP).
3.3. Mo(CO)4{2-Ph2PO(CH2CH2O)2–C12H8-2%-
OPPh2} (3) (cis-3 and trans-3)
Solutions of 0.585 g (1.95 mmol) of Mo(CO)4(nbd) in
100 ml of THF and 1.28 g (1.95 mmol) of 1 in 100 ml
of THF were added simultaneously and dropwise to
1000 ml of THF over a 2 h period. The mixture was
stirred for an additional 18 h and then evaporated to
dryness to yield a light brown, oily residue. The residue
was taken up in 50 ml of a 1:1 dichloromethane–hex-
anes mixture, and the resulting solution was treated
with 15 g of silica gel. This mixture was filtered, and the
silica gel was washed with two, 50 ml portions of a 1:1
dichloromethane–hexanes mixture. The filtrate and
washes were combined and evaporated to dryness to
yield 1.11 g (66.9%) of 3 as a white powder. Recrystal-
lization from dichloromethane–hexanes gave analyti-
cally pure 3 · H2O as colorless crystals. Anal. Calc. for
C44H38O9P2Mo: C, 60.84; H, 4.41. Found: C, 60.91; H,
4.37%. 31P{1H}-NMR (freshly prepared chloroform-d):
2
(chloroform-d): l 171.98 (d, ArOP, ꢀ J(PP%)ꢀ 48 Hz),
2
167.85 (d, CH2OP, ꢀ J(PP%)ꢀ 48 Hz). 13C-NMR (car-
bonyl and aliphatic carbons, chloroform-d): l 210.90
2
2
(trans CO, dd, ꢀ J(PC)ꢀ 13 Hz, ꢀ J(P%C)ꢀ 9 Hz), 210.21
2
2
(trans CO, dd, ꢀ J(PC)ꢀ 13 Hz, ꢀ J(P%C)ꢀ 9 Hz), 206.49
2
2
(cis CO, dd, ꢀ J(PC)ꢀ 14 Hz, ꢀ J(P%C)ꢀ 14 Hz), 205.48 (cis
2
2
CO, dd, ꢀ J(PC)ꢀ 14 Hz, ꢀ J(P%C)ꢀ 14 Hz), 70.30
3
(CH2CH2OP, d, ꢀ J(PC)ꢀ 6 Hz), 69.59 (CH2CH2OAr, s),
2
68.80 (CH2OAr, s), 67.63 (CH2OP, d, ꢀ J(PC)ꢀ 11 Hz).
2
major cis isomer: l 146.25 (d, ArOP, ꢀ J(PP%)ꢀ 37 Hz),
1H-NMR (chloroform-d): l 8.01–7.00 (m, 24H, P–
C6H5 and C12H8), 4.54 (ddd, 1H, CH2OAr), 4.35 (ddd,
1H, CH2OAr), 4.25 (ddd, 1H, CH2CH2OAr), 4.23 (m,
2H, CH2OP), 4.06 (ddd, 1H, CH2CH2OAr), 3.88 (ddd,
1H, CH2CH2OP), 3.73 (ddd, 1H, CH2CH2OP).
2
141.43 (d, CH2OP, ꢀ J(PP%)ꢀ 37 Hz); minor trans isomer:
2
l 162.26 (d, ArOP, ꢀ J(PP%)ꢀ 88 Hz), 157.38 (d, CH2OP,
2
ꢀ J(PP%)ꢀ 88 Hz).
3.4. cis-Mo(CO)4{2-Ph2PO(CH2CH2O)2–C12H8-
2%-OPPh2} (cis-3)
3.6. Reaction of PhLi and cis-Mo(CO)4
A mixture of 0.585 g (1.95 mmol) of Mo(CO)4(nbd)
and 1.284 g (1.95 mmol) of 1 in 100 ml of hexanes was
stirred for 2 h at r.t. The mixture was then filtered to
{2-Ph2PO(CH2CH2O)2–C12H8-2%-OPPh2} (cis-3)
A solution of cis-3 (0.085 g, 0.10 mmol) in 0.5 ml of

M. Hariharasarma et al. / Journal of Organometallic Chemistry 580 (1999) 328–338
335
benzene-d6 was prepared in a 5 mm NMR tube under
nitrogen. Phenyllithium (55 ml, 0.10 mmol, 1.8 M in
70:30 cyclohexane–ether) was then added to the tube,
and the 31P-NMR spectrum of the solution was taken.
During this time, a yellow–orange solid slowly precipi-
tated from the solution. This solid was collected and
washed several times with benzene and hexanes to yield
0.072 g (77%) of the crude product as a yellow–orange
powder. This compound was extremely reactive, and all
attempts to recrystallize this material resulted in its
rapid decomposition to give the starting material, cis-3.
CAD4 single-crystal diffractometer. Details of the data
collection of each complex are summarized in Table 1.
Standard peak search and automatic indexing routines
followed by least squares fits of 25 accurately centered
reflections resulted in accurate unit cell parameters for
cis-3 and cis-4.
The analytical scattering factors of each complex
were corrected for both Df % and iDf ¦ components of
anomalous dispersion. Both structures were solved by
the use of Patterson syntheses. Positional and an-
isotropic thermal parameters for all non-hydrogen
31P{1H}-NMR (benzene-d6):
l 174.23 (d, ArOP,
ꢀ J(PP%)ꢀ 35 Hz), 151.11 (d, CH2OP, ꢀ J(PP%)ꢀ 35 Hz).
2
2
Table 1
wCO (benzene-d6): 1942, 1873, 1838 cm−1
.
Experimental data for crystallographic studies of cis-3 and cis-4
cis-3
cis-4
3.7. HgCl2-catalyzed cis–trans isomerization of
cis-Mo(CO)4{2-Ph2PO(CH2CH2O)2–C12H8-2%-
OPPh2}, cis-3, and cis-Mo(CO)4{2-(2,2%-O2C12H8)
P(CH2CH2O)2–C12H8-2%-P(2,2%-O2C12H8)} (cis-4)
Formula
C44H36O8P2Mo
850.65
P1
11.885(3)
12.410(5)
15.899(6)
92.98(3)
99.62(2)
116.49(2)
2
C44H32O12P2Mo
910.60
Pna21
18.267(2)
14.326(2)
15.277(3)
90
90
90
8
Formula weight (Da)
Space group
˚
a (A)
˚
b (A)
A solution of either 0.096 g (0.11 mmol) of cis-3 or
0.10 g (0.11 mmol) of cis-4 in 0.5 ml of chloroform-d
was prepared in a 5 mm, screw-top NMR tube under
nitrogen. The head space of the tube was flushed with
nitrogen, and both 13C{1H}- and 31P{1H}-NMR spec-
tra of the solution were recorded. Then solid HgCl2 (0.3
mg, 0.001 mmol) was added to the solution. After
vigorous shaking, a non-quantitative 31P{1H}-NMR
spectrum was acquired, and the cis and trans reso-
nances were integrated. Additional 31P{1H}-NMR spec-
tra were taken at 5 min intervals until the cis:trans ratio
reached a constant value. Then, a quantitative 31P{1H}-
NMR spectrum was taken. 31P{1H}-NMR of 3 (chloro-
˚
c (A)
h (°)
i (°)
k (°)
Z
Dcalc. (g cm−3
hmax, hmin
kmax, kmin
lmax, lmin
)
1.379
1.513
19, −19
15, 0
15, −15
16, −16
0, −20
Enraf–Nonius
CAD4
Mo–Ka (u=
0.71073)
298
16, 0
Diffractometer
Enraf–Nonius
CAD4
Mo–Ka (u=
0.71073)
298
2.0–45.0
5343
2726
˚
Radiation (A)
T (K)
2q limits (°)
1.0–54.0
9765
9422
2
Reflections measured
Independent reflections
measured
Observed reflections
Rint (%)
form-d): minor cis isomer: l 146.25 (d, ArOP, ꢀ J(PP%)ꢀ
2
37 Hz), 141.43 (d, CH2OP, ꢀ J(PP%)ꢀ 37 Hz); major trans
2
isomer: l 162.26 (d, ArOP, ꢀ J(PP%)ꢀ 88 Hz), 157.38 (d,
6164 (F\6|(F)) 2190 (F\6|(F))
2.02
ꢀ−2q
Semi-empirical
0.45
2
CH2OP, ꢀ J(PP%)ꢀ 88 Hz). 31P{1H}-NMR of 4 (chloro-
2.57
ꢀ−2q
Semi-empirical
0.50
form-d): minor cis isomer: l 171.98 (d, CH2OP,
Scan type
2
2
Absorption correction
Absorption coefficient
ꢀ J(PP%)ꢀ 48 Hz), 167.85 (d, ArOP, ꢀ J(PP%)ꢀ 48 Hz);
2
major trans isomer: l 191.55 (d, CH2OP, ꢀ J(PP%)ꢀ 237
(mm−1
)
2
Hz), l 178.56 (d, ArOP, ꢀ J(PP%)ꢀ 237 Hz).
Min./max. transmission 0.8050/0.8316
0.7993/0.8154
533
No. of variables
497
Refinement method
Full matrix
least-squares
ꢁw(Fo−Fc)2
Full matrix least-
squares
3.8. X-ray structural analysis of cis-Mo(CO)4{2-Ph2
PO(CH2CH2O)2–C12H8-2%-OPPh2} (cis-3) and
cis-Mo(CO)4{2-(2,2%-O2C12H8)P(CH2CH2O)2–C12H8-
2%-P(2,2%-O2C12H8)} (cis-4)
Quantity minimized
Weighing scheme (w−1
Extinction coefficient
R (%)
ꢁw(Fo−Fc)2
)
|2(F)+0.0025F2 |2(F)+0.0005F2
NA
9.5(6)×10−5
3.41(all data)/
2.19(6|)
7.17 (all data)/
3.78(6|)
6.75(all data)/
4.40(6|)
0.82
Hot, saturated dichloromethane–hexanes solutions
of both cis-3 and cis-4 were slowly cooled to −10°C to
yield X-ray quality, colorless, single crystals of each
complex. A suitable crystal of cis-3 was mounted on a
glass fiber with epoxy cement under aerobic conditions.
A single crystal of cis-4 was mounted in a thin-walled
glass capillary under aerobic conditions. Each crystal
was mounted and aligned upon an Enraf–Nonius
Rw (%)
3.41(all data)/
3.05(6|)
0.82
GOF
Max. difference peak (e 0.50
0.31
−3
˚
A
)
Max. difference hole (e
−0.71
−0.27
−3
˚
A
)
Max. mean D/s
0.001/0.000
0.001/0.000

336
M. Hariharasarma et al. / Journal of Organometallic Chemistry 580 (1999) 328–338
Table 2
˚
Selected bond distances (A) with their estimated S.D. for cis-3 and cis-4
˚
˚
Atom 1
Atom 2
Distance (A)
Atom 1
Atom 2
Distance (A)
cis-3
Mo1
Mo1
Mo1
C45
C42
P1
P1
P2
C1
P1
2.530(1)
1.985(3)
1.988(4)
1.132(6)
1.147(6)
1.652(2)
1.822(4)
1.824(4)
1.381(3)
1.421(6)
1.411(4)
1.489(4)
Mo1
Mo1
Mo1
C41
C43
P2
P1
P2
C12
C13
C15
C16
P2
2.508(1)
2.011(4)
2.054(4)
1.138(4)
1.154(6)
1.627(2)
1.824(3)
1.822(3)
1.362(5)
1.487(6)
1.413(5)
1.436(4)
C41
C43
O5
O7
O1
C17
C29
O1
O2
O3
C42
C44
O6
O8
O4
C23
C35
O2
C14
O3
C13
C14
C15
C16
O4
cis-4
Mo1
Mo1
Mo1
C41
C43
P1
P1
P2
C1
P1
2.437(1)
2.002(6)
2.019(5)
1.148(7)
1.139(6)
1.602(4)
1.611(4)
1.634(3)
1.402(6)
1.437(8)
1.399(8)
1.478(9)
Mo1
Mo1
Mo1
C42
C44
P2
P1
P2
C12
C13
C15
C16
P2
2.429(1)
2.039(7)
2.044(6)
1.135(8)
1.126(9)
1.569(5)
1.618(4)
1.612(5)
1.375(7)
1.501(9)
1.405(8)
1.460(8)
C41
C43
O10
O12
O1
O5
O7
O1
O2
C42
C44
O11
O9
O4
O6
O8
O2
C14
O3
O4
C13
C14
C15
O3
C16
atoms were refined. Hydrogen atoms were not located
directly but were input in calculated positions with d
confirmed this. The absolute structure was confirmed
by an p-refinement procedure with p=0.9(2).
˚
(C–H)=0.96 A [17]. All calculations were carried out
®
using the Siemens SHELXTL-PC program package [18].
Selected bond lengths, bond angles and torsion angles
are given in Tables 2–4, respectively [19].
4. Conclusions
Although the unsymmetrical metallacrown ethers,
cis-Mo(CO)4{2-Ph2PO(CH2CH2O)2–C12H8-2%-OPPh2}
(cis-3) and cis-Mo(CO)4{2-(2,2%-O2C12H8)P(CH2CH2-
O)2–C12H8-2%-P(2,2%-O2C12H8)} (cis-4), have identical
metallacrown ether rings, they have very different reac-
tivities. The differences in reactivities of cis-3 and cis-4
are consistent with more effective interaction of the
carbonyl ligands in cis-3 with hard metal cations. The
research described in this paper demonstrates that both
the conformation of the metallacrown ether ring and
the phosphorus substituents affect the interaction of the
carbonyl ligands with the hard metal cations.
3.8.1. cis-Mo(CO)4{2-Ph2PO(CH2CH2O)2–C12H8-
2%-OPPh2} (cis-3)
Unit cell parameters of cis-3 indicate that the crystal
belongs to the triclinic crystal system. Intensity statistics
clearly favored the centrosymmetric space group P1
(no. 2) over the noncentrosymmetric space group P1.
The successful solution and refinement of the crystal
structure later confirmed this.
3.8.2. cis-Mo(CO)4{2-(2,2%-O2C12H8)-
P(CH2CH2O)2–C12H8-2%-P(2,2%-O2C12H8)} (cis-4)
Unit cell parameters of cis-4 indicate that the crystal
belongs to the orthorhombic crystal system. Systematic
absences (h0l for h=2n+1, 0k0 for k=2n+1, 0kl for
k+l=2n+1, h00 for h=2n+1 and 00l for l=2n+1)
require the space group to be either the noncentrosym-
metric Pna21 (no. 33) or the centrosymmetric Pnma
(No. 62). Intensity statistics and unit cell volume fa-
vored the noncentrosymmetric choice Pna21. Successful
solution and refinement of the crystal structure later
5. Supplementary material
X-ray crystallographic data for cis-3 and cis-4 in-
cluding tables of atomic and thermal coordinates, bond
lengths and bond angles, torsion angles and least
squares planes and 1H–1H NOESY and 1H COSY
NMR spectra of cis-4 (21 pages) [19].

M. Hariharasarma et al. / Journal of Organometallic Chemistry 580 (1999) 328–338
337
Table 3
Selected bond angles (°) with their estimated S.D. for cis-3 and cis-4
Atom 1
Atom 2
Atom 3
Bond angle (°)
Atom 1
Atom 2
Atom 3
Bond angle (°)
cis-3
P1
P1
P1
P2
Mo1
Mo1
Mo1
Mo1
Mo1
P2
P1
P1
O4
C15
C13
C12
C1
P2
102.4(1)
94.7(1)
86.9(1)
86.7(1)
91.6(1)
118.6(1)
123.8(1)
111.5(1)
118.4(1)
107.0(2)
108.0(3)
123.7(4)
118.2(3)
P1
P1
P2
P2
Mo1
Mo1
Mo1
P1
O4
O3
O2
O1
Mo1
Mo1
Mo1
Mo1
P2
P2
P1
O1
C16
C14
C12
C1
C41
C43
C41
C43
O4
C35
C17
C1
170.7(1)
85.0(1)
86.7(1)
C42
C44
C42
C44
C29
O1
C23
C16
C16
C14
C11
C6
172.2(1)
120.8(1)
113.9(1)
117.6(1)
125.3(2)
106.6(2)
111.3(3)
116.0(3)
120.9(2)
P2
Mo1
Mo1
Mo1
P2
O3
O2
C15
C13
C7
O2
O1
C2
cis-4
P1
P1
P1
P2
Mo1
Mo1
Mo1
Mo1
Mo1
P2
P1
P1
O4
C15
C13
C12
C1
P2
91.7(1)
90.1(2)
90.1(2)
88.8(2)
86.4(2)
120.8(1)
117.0(1)
119.5(2)
127.4(4)
110.2(6)
107.6(5)
124.7(6)
119.6(5)
P1
P1
P2
P2
Mo1
Mo1
Mo1
P1
O4
O3
O2
O1
Mo1
Mo1
Mo1
Mo1
P2
P2
P1
O1
C16
C14
C12
C1
C41
C43
C41
C43
O4
O8
O5
C1
173.7(2)
89.3(2)
94.6(2)
C42
C44
C42
C44
O7
178.9(2)
115.5(2)
112.8(2)
112.7(1)
126.1(3)
108.5(5)
114.8(6)
113.4(5)
117.7(5)
P2
Mo1
Mo1
Mo1
P2
O3
O2
O1
O6
C16
C16
C14
C11
C6
C15
C13
C7
O2
O1
C2
Table 4
Acknowledgements
Selected torsion angles (°) for cis-3 and cis-4
Atom 1
Atom 2
Atom 3
Atom 4
Torsion angle (°)
The authors thank the Department of Chemistry at
The University of Alabama for supporting this re-
search. M. Hariharasarma thanks the Graduate School
of The University of Alabama at Birmingham for a
Graduate Fellowship.
cis-3
Mo1
P1
P1
O1
C1
C7
C1
C6
C12
C7
C14
O3
C13
C16
O4
C15
C16
89.9
−127.8
−70.3
−175.0
178.0
67.3
−100.6
166.9
O1
C6
O2
O2
C13
O3
O3
C15
O4
P2
C1
C13
C12
O2
C15
C14
O3
C12
C13
C14
C14
C15
C16
C16
O4
References
[1] (a) J. Powell, A. Kuksis, C.J. May, S.C. Nyberg, S.J. Smith, J.
Am. Chem. Soc. 103 (1981) 5941. (b) J. Powell, S.C. Nyberg, S.J.
Smith, Inorg. Chim. Acta 76 (1983) L75. (c) J. Powell, K.S. Ng,
W.W. Ng, S.C. Nyberg, J. Organomet. Chem 243 (1983) C1. (d)
J. Powell, M.R Gregg, A. Kuskis, P. Meindl, J. Am. Chem. Soc.
105 (1983) 1064. (e) J. Powell, M.R. Gregg, A. Kuksis, C.J. May,
S.J. Smith, Organometallics 8 (1989) 2918. (f) J. Powell, A.
Kuskis, C.J. May, P.E. Meindl, S.J. Smith, Organometallics 8
(1989) 2933. (g) J. Powell, M.R. Gregg, P.E. Meindl,
Organometallics 8 (1989) 2942. (h) J. Powell, A. Lough, F.
Wang, Organometallics 11 (1992) 2289.
[2] (a) N.W. Alcock, J.M. Brown, J.C. Jeffery, J. Chem. Soc. Chem.
Commun. (1974) 829. (b) N.W. Alcock, J.M. Brown, J.C. Jef-
fery, J. Chem. Soc. Dalton Trans. (1976) 583. (c) D.H.W.
Thewissen, K. Timmer, J.G. Noltes, J.W. Marsman, R.M. Laine,
Inorg. Chim. Acta 97 (1985) 143. (d) K. Timmer, D.H.W.
Thewissen, Inorg. Chim. Acta 100 (1985) 235. (e) K. Timmer,
178.2
168.6
−61.0
P2
Mo
cis-4
Mo1
P1
P1
O1
C1
C7
C1
C6
C12
C7
C14
O3
C13
C16
O4
C15
C16
130.6
101.3
63.6
O1
C6
O2
O2
C13
O3
O3
C15
O4
P1
C1
C13
C12
O2
C15
C14
O3
C12
C13
C14
C14
C15
C16
C16
O4
179.4
−171.0
−68.2
−84.4
−167.0
−68.4
145.5
P2
Mo1
165.9

338
M. Hariharasarma et al. / Journal of Organometallic Chemistry 580 (1999) 328–338
D.H.W. Thewissen, J.W. Marsman, Recl. Trav. Chim. Pays-Bas,
107 (1988) 248.
and Applications of Organotransition Metal Chemistry, Univer-
sity Science Books, Mill Valley, California, 1987.
[12] (a) M.J. Baker, K.N. Harrison, A.G. Orpen P.G. Pringle, G.J.
Shaw, J. Chem. Soc. Chem. Commun. (1991) 803. (b) M.J. Baker,
P.G. Pringle, J. Chem. Soc. Chem. Commun. (1991) 1292.
[13] C.H. Duffey, Ph.D. Thesis, The University of Alabama at Birm-
ingham, 1996.
[14] [cis] and [trans] were determined by integration of the appropri-
ate peak areas in the quantitative 31P-NMR spectrum.
[15] W. Ehrl, R. Ruck, H. Vahrenkamp, J. Organomet. Chem. 56
(1973) 285.
[3] (a) A. Varshney, G.M. Gray, Inorg. Chem. 30 (1991) 1748. (b) A.
Varshney, M.L. Webster, G.M. Gray, Inorg. Chem. 31 (1992)
2580. (c) G.M. Gray, C.H. Duffey, Organometallics 13 (1994)
1542. (d) G.M. Gray, F.P. Fish, C.H. Duffey, Inorg. Chim. Acta
246 (1996) 229. (e) G.M. Gray, C.H. Duffey, Organometallics 17
(1998) 3550.
[4] S.J. McLain, F.J. Waller, 1984, US Patent 4 432 904.
[5] M. Hariharasarma, G.M. Gray, J. Chem. Crystallogr. 28 (1998)
297.
[6] (a) J.G. Verkade, L.D. Quin, Phoshorus-31 NMR Spectroscopy
in Stereochemical Analysis: Organic Compounds and Metal
Complexes, VCH, Weinheim, 1987. (b) Comprehensive
Organometallic Chemistry. Vol. 3, Pergamon Press, Oxford,
1982.
[16] L.V. Verizhnikov, P.A. Kirpichnikov, Zh. Obshch. Khim. 37
(1967) 1355 (Engl. transl. p. 1281).
[17] M.R. Churchill, Inorg. Chem. 12 (1973) 1213.
[18] Siemens SHELXTL-PC Manual, Release 4.1, Siemens Analytical
Instruments, Madison, WI, 1990.
[7] H. Friebolin, Basic One- and Two-Dimensional NMR Spec-
troscopy, Second Enlarged Edition, VCH, Weinheim, 1993.
[8] D.J. Darensbourg, M.A. Murphy, J. Am. Chem. Soc. 100 (1978)
463.
[9] D.J. Darensbourg, R.L. Kump, Inorg. Chem. 17 (1978) 2680.
[10] C.A. Tolman, Chem. Rev. 77 (1977) 313.
[19] Crystallographic Data has been deposited with the Cambridge
Crystallographic Data Center, CCDC No. 11-4974 for cis-3 and
CCDC No. 11-4975 for cis-4. Copies of this information may be
obtained free of charge from The Director, CCDC, 12 Union
Road, Cambridge, CB1 1EZ UK (Fax: +44-1223-336-033 or
e-mail: deposit@ccdc.cam.ac.uk or www://http:www.ccdc.cam.
ac.uk).
[11] J.P. Collman, L.S. Hegedus, J.R. Norton, R.G. Finke, Principles
.