Photochemistry of ring-coupled group VI bimetallic compounds I: Phosphine substitution without disproportionation. The molecular structures of Mo2(CO)5[P(C6H5)3] (μ-η5,η5-C5H4-CR2-C5H4).

Article abstract of DOI:10.1016/S0022-328X(98)00396-9

Solution photolysis of ring-coupled bimetallic compounds of the formula M₂(CO)₆(μ-η⁵,η⁵-C₅H₄-CR₂-C₅H₄), where M=Mo or W and R=H or CH₃, in the presence of triphenylphosphine or trimethylphosphine yields simple substitution derivatives M₂(CO)₅L(μ-η⁵,η⁵-C₅H₄-CR₂-C₅H₄) or M₂(CO)₄L₂(μ-η⁵,η⁵-C₅H₄-CR₂-C₅H₄). The compounds have been fully characterized by IR and NMR spectroscopy and elemental analysis. Phosphine ligands are found to occupy positions trans to the M-M bond. No evidence was found for disproportionation processes such as those found for similar reactions of the non-ring-coupled species. The molecular structures of two compounds have been determined: (μ-η⁵,η⁵-C₅H₄-CH₂-C₅H₄) Mo₂(CO)₅(PPh₃): monoclinic, P2₁/c, a=22.871(8), b=16.415(6), c=15.957(9) A, β=94.29(3)°, V=5974(4) A³, Z=8, R(F)=8.87%; (μ-η⁵,η⁵-C₅H₄-CH ₂-C₅H₄)Mo₂(CO)₄ (PMe₃)₂: monoclinic, P2₁/c, a=9.843(5), b=15.343(7), c=31.514(6) A, β=97.28(3)°, V=4720(4) A³, Z=8, R(F)=4.38%.

Full text of DOI:10.1016/S0022-328X(98)00396-9

Journal of Organometallic Chemistry 562 (1998) 89–96
Photochemistry of ring-coupled group VI bimetallic compounds I:
phosphine substitution without disproportionation. The molecular
structures of Mo2(CO)5[P(C6 H5)3](v-p5,p5-C5H4–CR2–C5H4) and
Mo2(CO)4[P(CH3)3]2(v-p5,p5-C5H4–CR2–C5H4)1
T.E. Bitterwolf a,*, Abdel Saygh a, Joyce E. Shade b, Arnold L. Rheingold c, Glenn P.A. Yap c,
Louise Liable-Sands c
a Department of Chemistry, Uni6ersity of Idaho, Moscow, ID 83844-2343, USA
b Department of Chemistry, US Na6al Academy, Annapolis, MD 21402, USA
c Department of Chemistry and Biochemistry, Uni6ersity of Delaware, Newark, DE 19716, USA
Received 28 October 1997; received in revised form 14 January 1998
Abstract
Solution photolysis of ring-coupled bimetallic compounds of the formula M2(CO)6(v-p5,p5-C5H4–CR2–C5H4), where M=Mo
or W and R=H or CH3, in the presence of triphenylphosphine or trimethylphosphine yields simple substitution derivatives
M2(CO)5L(v-p5,p5-C5H4–CR2–C5H4) or M2(CO)4L2(v-p5,p5-C5H4–CR2–C5H4). The compounds have been fully characterized
by IR and NMR spectroscopy and elemental analysis. Phosphine ligands are found to occupy positions trans to the M–M bond.
No evidence was found for disproportionation processes such as those found for similar reactions of the non-ring-coupled species.
The molecular structures of two compounds have been determined: (v-p5,p5-C5H4–CH2–C5H4)Mo2(CO)5(PPh3): monoclinic,
3
5
5
˚
˚
P21/c, a=22.871(8), b=16.415(6), c=15.957(9) A, i=94.29(3)°, V=5974(4) A , Z=8, R(F)=8.87%; (v-p ,p -C5H4–CH2–
3
˚
˚
C5H4)Mo2(CO)4(PMe3)2: monoclinic, P21/c, a=9.843(5), b=15.343(7), c=31.514(6) A, i=97.28(3)°, V=4720(4) A , Z=8,
R(F)=4.38%. © 1998 Elsevier Science S.A. All rights reserved.
Keywords: Photolysis; Bi metallic compounds; Substitution reactions; Phosphines
1. Introduction
which the metal–metal bond was retained, precluding a
radical involvement in this important reaction.
The photochemistry of Mo2(CO)6(p5-C5H5)2 in the
absence of ligands is now well understood to involve
both radical and carbonyl-loss pathways as summarized
in Scheme 1 [1]. Some years ago Turaki and Huggins [2]
established that the carbonylation of triply bonded
Mo2(CO)4(p5-C5H5)2 to give Mo2(CO)6(p5-C5H5)2 (and
its microscopic reverse) proceeded by a mechanism in
In the presence of a potential ligand, such as a
phosphine or phosphite, either the 17-electron (17-e)
radical, M(CO)3(p5-C5H5), or the electron deficient in-
termediate, Mo2(CO)5(p5-C5H5)2, could react with the
ligand. A number of investigations have established
that the ultimate products in these reactions are
strongly dependent upon the electronic donor ability
and the cone angle of the phosphine or phosphite [3].
Tyler has proposed a mechanism for these reactions,
Scheme 2, in which a M(CO)3(p5-C5H5), 17-e radical
reacts with a ligand to give M(CO)2L(p5-C5H5). Subse-
quent reaction of this 17-e species with a second ligand
forms a highly reducing 19-e species, M(CO)2L2(p5-
* Corresponding author. Tel.: +1 208 8856552; fax: +1 208
8856173; e-mail: bitterte@osprey.csrv.uidaho.edu
1 In honor of our friend and colleague Prof. R. Bruce King on the
occasion of his 60th birthday.
0022-328X/98/$19.00 © 1998 Elsevier Science S.A. All rights reserved.
PII S0022-328X(98)00396-9

90
T.E. Bitterwolf et al. / Journal of Organometallic Chemistry 562 (1998) 89–96
Scheme 1.
C5H5). The rate dependence of the reaction on the cone
angle of the ligand is attributed to the step in which the
second ligand molecule reacts with M(CO)2L(p5-C5H5)
since formation of a bis(ligand) species would be inhib-
ited by bulky ligands.
2. Results and discussion
Broad band photolysis of the red bimetallic com-
pounds 1–3, Scheme 4, in the presence of
triphenylphosphine ligand or photolysis of 1 with
trimethylphosphine in degassed benzene afforded dark
purple reaction mixtures. After removal of solvent, the
reaction mixtures were cleanly separable by medium
pressure column chromatography to give three bands.
Characterization of these bands, vide infra, established
them to be monosubstituted phosphine derivatives, dis-
ubstituted phosphine derivatives and unreacted starting
material. The proportions of the three bands were
dependent upon the ratio of the bimetallic compound
and phosphine in the initial reaction mixture. Insuffi-
An alternative mechanism that is also consistent with
the observed data is shown in Scheme 3. In this mecha-
nism Mo2(CO)5L(p5-C5H5)2, formed by either carbonyl
loss or radical routes, undergoes photochemical dispro-
portionation to yield [M(CO)2L(p5-C5H5)]+ and
[M(CO)3(p5-C5H5)]−. Reaction of [M(CO)2L(p5-
C5H5)]+ with a second molecule of L yields the bis(li-
gand) cation. Since in this mechanism, L is competing
with
[M(CO)3(p5-C5H5)]−
for
reaction
with
[M(CO)2L(p5-C5H5)]+, disproportionation will be fa-
vored for all but the largest cone angles.
cient
quantities
of
[v-p5,p5-C5H4–CH2–C5H4]
We have previously reported the synthesis and matrix
photochemistry of ring-coupled compounds of the gen-
eral formula M2(CO)6(v-p5,p5-C5H4–CR2–C5H4),
where M=Mo or W and R=H or CH3 [4]. Photolysis
of these compounds in inert gas matrices [5] and in
frozen Nujol [6] indicate that a single carbon monoxide
is lost upon photolysis. Unlike the parent species,
M2(CO)6(p5-C5H5)2, there is no evidence for loss of a
second carbon monoxide to form a triply bonded spe-
cies when the compounds are photolyzed in the absence
of potential ligands, nor is there any evidence for the
formation of metal-based radicals upon photolysis. In
order to evaluate the consequences of ring-coupling on
the reaction chemistries of these Group VI compounds,
we have examined the simple photochemical ligand
substitution chemistry with phosphines. The results of
these studies are reported in this paper.
Mo2(CO)5(PMe3) were isolated for full characteriza-
tion.
Monosubstituted
compounds,
Mo2(CO)5PPh3
(v-p5,p5-C5H4–CH2–C5H4), 4, W2(CO)5PPh3(v-p5,p5-
C5H4–CH2–C5H4), 5, and Mo2(CO)5PPh3(v-p5,p5-C5
H4–CMe2–C5H4), 6, were isolated as red–purple pow-
ders. The IR spectra of all three compounds are very
Scheme 2.

T.E. Bitterwolf et al. / Journal of Organometallic Chemistry 562 (1998) 89–96
91
Scheme 3.
similar and contain four carbonyl stretching bands in
the terminal region, shifted about 50–70 cm−1 to lower
energy relative to the parent compounds reflecting the
lower y-acid character of triphenylphosphine relative to
carbon monoxide. The room temperature (r.t.) 1H-
NMR spectra of 4 and 5 show a complex envelope
associated with the triphenylphosphine group, a well-
defined AB pattern for the methylene group, and eight
cyclopentadienyl ring proton resonances. The spectrum
of 6 differs only in that two methyl resonances were
observed for the CMe2 linkage between the cyclopenta-
dienyl rings. The observed patterns require that all
three compounds have C1 symmetry in solution. The
13C-NMR spectra are also consistent with monosubsti-
tutition to yield an asymmetric molecule. The carbon
spectra of 4 and 5 cleanly resolve five carbonyl reso-
nances, ten cyclopentadienyl ring resonances (two dis-
tinct ipso resonances were resolved) and a resonance for
the methylene carbon. The spectrum for 6 was identical
to those of 4 and 5 except that two methyl resonances
and a quaternary carbon resonance were observed for
the CMe2 linkage. The ortho phenyl carbon resonances
of the triphenylphosphine ligand were broadened and
split into two broad resonances having an integration
of ca. 2:1. No coupling was resolved between the single
para carbon resonance and the phosphorus atom, and
the ipso resonances were very broadened when ob-
served. We attribute the unusual spectral pattern of the
triphenylphosphine phenyl groups to hindered or
slowed rotation about the P–Mo bond creating two
magnetically distinct environments for the phenyl
groups.
X-ray quality crystals of 4 were obtained by vapor
diffusion of petroleum ether into a dichloromethane
solution of 4 in the cold. The ORTEP diagram of 4 is
presented in Fig. 1. Crystallographic data for com-
pounds 4 and 11 are summarized in Table 1, and
selected bond lengths and angles for 4 are presented in

92
T.E. Bitterwolf et al. / Journal of Organometallic Chemistry 562 (1998) 89–96
Scheme 4.
Tables 2 and 3. The structure of 4 was solved in the
P21/c space group with two independent, but chemi-
cally equivalent, molecules in the asymmetric unit. One
of these molecules resolved well, but the other was
found to be statistically disordered over two positions.
As a consequence, the R factor of 8.87% is somewhat
higher than is typical for this class of compounds.
The core structures of 4 and 1, i.e. Mo2(v-p5,p5-
C5H4–CH2–C5H4), are virtually identical with a C2
rotation about an axis running between the bridging
carbon and the midpoint of the Mo–Mo bond. The
triphenylphosphine ligand in 4 occupies a position trans
to the M–M bond. The Mo–Mo bond length of 1 is
ing observed in the phosphine derivative may be at-
tributable to either the reduced y-acid character of
triphenylphosphine relative to carbon monoxide, or to
an indirect effect of the steric bulk of the phosphine.
Variable temperature NMR spectral studies of 1
demonstrate that this compound undergoes a twisting
motion about the C2 axis that averages the magnetic
environment of the ring protons. At r.t. two ring pro-
ton resonances are observed and these split upon cool-
ing to give four resonances, consistent with the frozen
1
out C2 structure. In the case of 4, the r.t. H-NMR
spectrum exhibits five resonances (three resonances
were shown by integration to result from the accidental
overlap of two hydrogen resonances) for ring protons,
while ten separate ring resonances were observed in the
13C-NMR spectrum. This is consistent with the C1
symmetry established by the molecular structure. Ap-
parently the phosphine ligand serves to lock the
molecule into the twisted structure as found for 4,
increasing the barrier to twisting. The NMR and IR
spectra of compound 5 and 6 were similar to those of 4,
thus all of the mono-phosphine compounds in the series
are assumed to have similar structures.
˚
˚
3.140 A while that of 4 is 3.173 A. The bond lengthen-
Photolysis of 1, 2 and 3 in the presence of excess
triphenylphosphine in degassed benzene afforded good
yields of dark purple colored products. Three broad
carbonyl bands were observed for each of these com-
pounds. The r.t. 1H-NMR spectra of 7 and 8 show
broad envelopes of resonances associated with the
triphenylphosphine ligands, a singlet for the methylene
group linking the cyclopentadienyl rings, and four cy-
Fig. 1. Molecular structure of (v-p5,p5-C5H4–CH2–C5H4)
Mo2(CO)5(PPh3).

T.E. Bitterwolf et al. / Journal of Organometallic Chemistry 562 (1998) 89–96
93
Table 1
Table 3
Crystallographic data for (v-p5,p5-C5H4–CH2–C5H4)Mo2(CO)5
(PPh3), 4 and (v-p5,p5-C5H4–CH2–C5H4)Mo2(CO)4(PMe3)2, 11
Selected bond angles of 4 in degrees
Atoms
Angle (°)
4
11
Molecule A
Molecule B
Formula
C34H25Mo2O5P
736.39
Monoclinic
P21/c
C
21H28Mo2O4P2
Formula weight
Crystal system
Space group
Crystal data
598.25
Monoclinic
P21/c
C(7)–C(6)–C(5)
112.1
76.1
81.6
105.1
82.0
77.0
115.0(2)
96.1(13)
80.0(13)
104.0(13)
79.7(9)
C(12)–Mo(2)–C(13)
C(12)–Mo(1)–C(14)
C(16)–Mo(2)–C(15)
C(15)–Mo(2)–P(1)
C(16)–Mo(2)–P(1)
˚
a (A)
22.871(8)
16.415(6)
15.957(9)
94.29(3)
5974(4)
8
9.843(5)
15.343(7)
31.514(6)
97.28(3)
4720(4)
˚
b (A)
77.7(8)
˚
c (A)
i (°)
3
˚
V (A )
and 8 show two sets of phosphorus coupled carbonyl
resonances, five cyclopentadienyl ring resonances (in-
cluding the ipso) and a resonance for the methylene
carbon. The triphenylphosphine phenyl resonances
were similar to those observed for the mono-substituted
derivatives indicating some degree of hindered rotation
about the M–triphenylphosphine bond. The 13C spec-
trum of 9 substitutes a methyl resonance and quater-
nary carbon resonance for that of the methylene group,
but is otherwise identical. 13P-NMR spectra of all three
compounds show a single phosphorus resonance.
X-ray quality crystals could not be obtained for these
compounds, but it is reasonable to assume that the
phosphine ligands are located trans to the M–M bonds
as found for 6.
Z
8
Crystal color, habit
Red block
1.638
9.35
Red block
1.683
12.22
Dcalc (g cm−3
)
v(Mo–Kh) (cm−1
T (K)
)
233
244
Diffractometer
Radiation
Siemens P4
Mo–Kh (u=
Siemens P4
Mo–Kh (u=0.71073
˚
˚
0.71073 A)
A)
2q scan range (°)
4–42
7911
6404
4–50
10692
8321
Reflections collected
Independent reflec-
tions
R(F) (%)a
8.87
17.98
0.81
9.8
4.38
9.23
0.95
15.9
1.03
R(wF2) (%)a
−3
˚
Dz (e A
No/Nv
GOF
)
1.49
1
As noted above, only a small quantity of mono-sub-
stituted trimethylphosphine, 10, was isolated and this
material was characterized by comparison with the IR
spectra of 4, 5, and 6.
a Quantity minimized=R(wF2)−S[w(Fo2−Fc2)2]/S[(wF2o)2]2; R=SD/
S(Fo), D=ꢀ(Fo−Fc)ꢀ.
Upon photolysis of [v-p5,p5-C5H4–CH2–C5H4]
Mo2(CO)6 in the presence of an excess of the
trimethylphosphine ligand and subsequent chromato-
graphic work-up, a dark purple product, 11, was recov-
ered. The IR spectra of 11 contains three terminal
carbonyl stretching bands that are, almost, identical in
position to those of Mo2Fv(CO)4(PMe3)2 reported by
Vollhardt and coworkers [7]. The r.t. 1H- and 13C-
NMR spectra of 11 are similar to those of 7 with the
substitution of trimethylphosphine resonances for those
of triphenylphosphine.
clopentadienyl ring protons resonances. The spectrum
of 9 differs only in that a singlet resonance is observed
for the two methyl groups on the CMe2 linkage. The
spectra indicate that these molecules have C2 symmetry
in solution. 13C-NMR spectra are also consistent with a
C2 symmetry for these molecules. The 13C spectra of 7
Table 2
Selected bond lengths of 4 in angstroms
˚
Atoms
Bond length (A)
X-ray quality crystals of 11 were grown by vapor
diffusion of petroleum ether into a saturated solution of
11 in dichloromethane in the cold. The structure was
solved in the P21/c space group and both of the inde-
pendent, but chemically equivalent molecules in the
asymmetric unit were well resolved. The ORTEP dia-
gram of 11 is presented in Fig. 2. Tables 4 and 5 present
selected bond lengths and angles, respectively. Com-
parison of the Mo–Mo bond length of 11 with that of
the parent compound indicates that considerable
metal–metal bond shortening has resulted from the
substitution of two carbonyl ligands by two
trimethylphosphine ligands. This may be attributed to
Molecule A
Molecule B
Mo(1)–Mo(2)
Mo(1)–C(12)
Mo(1)–C(13)
Mo(1)–C(14)
Mo(2)–C(15)
Mo(2)–C(16)
C(12)–O(12)
C(13)–O(13)
C(14)–O(14)
C(16)–O(16)
C(15)–O(15)
Mo(2)–P(1)
3.173(2)
1.93(2)
2.00(2)
1.92(2)
1.99(2)
1.96(2)
1.17(2)
1.16(2)
1.23(2)
1.14(2)
1.12(2)
2.477(4)
3.190(4)
1.93(3)
2.02(4)
1.94(4)
2.04(3)
1.87(3)
1.15(3)
1.14(4)
1.18(4)
1.29(4)
1.09(3)
2.462(8)

94
T.E. Bitterwolf et al. / Journal of Organometallic Chemistry 562 (1998) 89–96
Table 5
Selected bond angles for 11 in degrees
Atoms
Angles (°)
Molecule A
Molecule B
C(7)–C(6)–C(5)
115.6(5)
102.7(2)
76.3(2)
80.3(2)
79.6
116.2(6)
102.4(2)
78.4(2)
76.5(2)
78.0(2)
76.8(2)
102.0(2)
C(16)–Mo(1)–C(15)
C(15)–Mo(1)–P(1)
C(16)–Mo(1)–P(1)
C(20)–Mo(2)–P(2)
C(21)–Mo(2)–P(2)
C(20)–Mo(2)–C(21)
77.9
99.4(2)
by facilitating rapid recombination of any radicals or
ionic species that might be formed. In essence, the
linkage creates an artificial cage from which the metal
fragments cannot escape. We are continuing to pursue
an understanding of the mechanism of the dispropor-
tion reaction of the non-ring coupled Group VI
compounds.
Fig.
2.
Molecular
structure
of
(v-p5,p5-C5H4–CH2–
C5H4)Mo2(CO)4(PMe3)2.
the strong |-donating and weak y-accepting character
of trimethylphosphine relative to carbon monoxide.
The same effect on Mo–Mo bond length was observed
for (v-p5,p5-C5H4–C5H4)Mo2(CO)4(PMe3)2 (see Table
6 for a comparison of Mo–Mo bond lengths).
4. Experimental
3. Conclusions
All synthesis were carried out under an atmosphere
of dinitrogen using standard Schlenk techniques. All
purification chromatography was performed using
reagent grade solvents, neutral alumina provided by
CAMAG or silica gel (Fisher Scientific). Medium pres-
sure chromatography was carried out using an Altex
medium pressure column packed with 40 micron silica
gel. All preparative solvents were dried and distilled
under nitrogen.
Photolysis of the ring-linked compounds (v-p5,p5-
C5H4–CR2–C5H4)M2(CO)6, where M=Mo or W, and
R=H or Me, with either triphenylphosphine or
trimethylphosphine gives good yields of mono and di-
substituted products. No evidence was found for dis-
proportionation products such as those exclusively
isolated when the analogous non-ring-linked com-
pounds are photolyzed with triphenylphosphine. Since
it is not reasonable that the ring-linkages should signifi-
cantly alter the accessible electronic states of the M–M
bonds, it appears that the role of the linkage may be to
suppress the radical and disproportionation pathways
1H-, 13C- and 13P-NMR spectra were recorded on an
Bruker NR-300 spectrometer (UI) or a QE-300 spec-
trometer (USNA). Chemical shifts were recorded in
ppm relative to tetramethylsilane using appropriate sol-
vent resonances as internal standards. IR spectra were
obtained using a Bio-Rad Qualimatic FTIR spectrome-
ter. HPLC analyses were conducted using a Gow-Mac
HPLC with a 254 nm fixed wavelength detector. The
stationary phase for HPLC was a Whatman Parasil
PXS 10/25 silica gel column and the solvent system was
70:30 mixture of petroleum ether and THF. Elemental
Table 4
Selected bond lengths for 11 in angstroms
˚
Atoms
Bond lengths (A)
Molecule A
Molecule B
Mo(1)–Mo(2)
Mo(1)–C(15)
Mo(1)–C(16)
Mo(1)–P(1)
Mo(2)–C(20)
Mo(2)–C(21)
Mo(2)–P(2)
C(15)–O(15)
C(16)–O(16)
C(20)–O(20)
C(21)–O(21)
3.1341(12)
1.951(7)
1.942(6)
2.405(2)
1.919(7)
1.96137(7)
2.410(2)
1.177(7)
1.166(7)
1.183(7)
1.157(7)
3.1200(13)
1.959(6)
1.934(7)
2.409(2)
1.962(6)
1.937(7)
2.402(2)
1.156(7)
1.171(7)
1.163(7)
1.165(7)
Table 6
˚
Comparisons of M–M bond lengths (A) among related compounds
Compound
Mo–Mo
bond length
(v-p5,p5-C5H4–C5H4)Mo2(CO)6
3.371 [7]
3.220 [8]
3.140 [9]
3.1341
(v-p5,p5-C5H4–C5H4)Mo2(CO)4(Pme3)2
(v-p5,p5-C5H4–CH2–C5H4)Mo2(CO)6
(v-p5,p5-C5H4–CH2–C5H4)Mo2(CO)4(PMe3)2

T.E. Bitterwolf et al. / Journal of Organometallic Chemistry 562 (1998) 89–96
95
4.3. (v-p5,p5-C5H4–CH2–C5H4)W2(CO)5(PPh3), 5
analysis were performed by Desert Analysis of Tucson,
Arizona. Dr Gary Knerr conducted fast atom bom-
bardment mass spectrometry on a VG 7070-HS GC/
MS.
1, 2, and 3 were prepared by the procedures of
Bitterwolf ([4]b) and Feirro ([4]a).
Compound 2 and triphenylphosphine were irradiated
for 30 min and the procedure described in Section 4.1
was followed. Result: red–purple crystals, m.p. 245–
250°C, yield 25%. IR (CH2Cl2): 1976, 1902, 1876 and
.
1820 cm−1 1H-NMR: (l, CDCl3) 7.39 (br, 15H,
4.1. General procedures for photochemical synthesis
triphenylphosphine), 5.65 (m, 2H, Cp), 5.3 (m, 2H, Cp),
5.1 (s, 1H, Cp), 4.18 (m, 2H, Cp), 4.05(s, 1H, Cp), 4.44
and 3.33 (AB quartet, J A–B=14.0 Hz, 2H, CH2) ppm.
13C-NMR: (l, CDCl3), 230 (d, JC–P=16.7 Hz, W–
CO), 224 (d, JC–P=17.85 Hz, W–CO), 223, 218 and
214.4 (W–CO), 135 (broad d, JP–C=60 Hz, ipso-Ph)
133 (broad d about 2:1, o-Ph), 130 (s, p-Ph), 128 (d,
Group VI bimetallic compounds and an appropriate
phosphine were charged into a 400 ml Ace quartz
irradiation vessel and the apparatus was flushed with
dinitrogen. Air-free, dry benzene was added and the
solution was stirred while nitrogen was bubbled
through via a syringe. A medium pressure mercury
lamp was used for broad band irradiation of the reac-
tion mixture. During photolysis, the color of the solu-
tion changed from dark red to dark purple. The
photolysis was stopped when IR spectra showed the
absence of starting material. The reaction mixture was
concentrated with a rotary evaporator and the resulting
solution was placed on a silica gel column and chro-
matographed with petroleum ether or heptane as an
eluant to remove unreacted phosphine ligand. The ma-
terial remaining on the column was stripped from the
column by benzene. After removal of the benzene sol-
vent, the resulting solid was taken up in a 60:40 mixture
of benzene and heptane (or 20:80 for compound 11)
and chromatographed using a medium pressure column
eluting with the same mixture. In all cases, three bands
were eluted from the column which were shown to be
the mono-substituted derivative, the di-substituted
derivative, and unreacted starting material, respectively.
The yields of products are cited relative to starting
material.
JP–C=9 Hz, m-Ph), 116 (ipso-Cp), 105 (ipso-Cp), 98,
94, 89, 88, 87, 85, 84 and 83 (Cp), 27 (CH2) ppm.
31P-NMR: (l, CDCl3) 44.5 (W–PPh3) ppm. Anal. Calc.
C34H25O5W2P: C, 44.77; H, 2.77%. Found: C, 44.42; H,
2.47%.
4.4. (v-p5,p5-C5H4–CMe2–C5H4)Mo2(CO)5(PPh3), 6
See procedure (Section 4.1). Result: red–purple crys-
tals, m.p. 234–236°C, yield 20%. IR (CH2Cl2): 1978,
1
1908, 1883 and 1828 cm−1. H-NMR: (l, CDCl3) 7.35
(br, 15H, triphenylphosphine), 5.55 (m, 2H, Cp), 5.1
(br, 2H, Cp), 4.2 (br, 2H, Cp), 3.8 (s, 2H, Cp), 1.5 (s,
3H, CH3), 1.18 (s, 3H, CH3) ppm. 13C-NMR: (l,
CDCl3) 133 (d, JC–P=10 Hz, o-Ph), 130 (s, p-Ph), 128
(d, JC–P=7 Hz, m-Ph), 119 (s, ipso-Cp), 98, 95.1, 94,
93(br), 89, 87 (d, JC–P=7.5 Hz), 86 (d, JC–P=10 Hz),
85 and 83.5 (br, Cp), 36 (C, bridging CMe2), 30 (s,
CH3), 28 (s, CH3) ppm. Metal carbonyl resonances
were not resolved. 31P-NMR: (l, CDCl3) 74.9 (Mo–
PPh3) ppm. Anal. Calc. C36H29O5Mo2P2: C, 56.56; H,
3.82%. Found: C, 55.43; H, 3.97%.
4.2. (v-p5,p5-C5H4–CH2–C5H4)Mo2(CO)5(PPh3), 4
4.5. (v-p5,p5-C5H4–CH2–C5H4)Mo2(CO)4(PPh3)2, 7
Compound 1, (0.52g, 1.0 mmol) and triphenylphos-
phine (0.26g, 1.0 mmol) were placed in the quartz
irradiation vessel. The reaction mixture was photolyzed
for 110 min and the reaction mixture worked-up ac-
cording to the general procedure described above (Sec-
tion 4.1). Result: red–purple crystals, m.p. 223–226°C,
yield 28%. IR (CH2Cl2): 1979, 1910, 1882 and 1830
See procedure (Section 4.1). Result: red–purple crys-
tals, m.p. 323–326°C, yield 35%. IR (CH2Cl2): 1918(m),
1
1832 and 1800(s) cm−1. H-NMR: (l, CDCl3) 7.35 (br,
30H, triphenylphosphine), 5.62 (m, 2H, Cp), 4.55 (br,
2H, Cp), 4.28 (m, 4H, Cp), 3.38 (s, 2H, CH2) ppm.
13C-NMR: (l, CDCl3) 241 (d, JC–P=10 Hz, CO, Mo–
CO), 234 (d, JC–P=10 Hz, CO, Mo–CO), 137 (broad,
ipso-Ph), 133 (broad d, o-Ph), 129 (s, p-Ph), 128 (d,
m-Ph), 110 (ipso-Cp), 97.5, 89, 88 and 86 (s, Cp), 28 (s,
CH2) ppm. 31P-NMR: (l, CDCl3) 79.25 (Mo–PPh3)
ppm. Anal. Calc. C51H40O4Mo2P2: C, 63.10; H, 4.16; P,
6.39%. Found: C, 63.02; H, 4.60; P, 4.29%
cm−1
.
1H-NMR: (l, CDCl3) 7.35 (br, 15H,
triphenylphosphine), 5.5 (m, 2H, Cp), 5.1 (br, 2H, Cp),
4.57 (s, 1H, Cp), 4.17 (m, 2H, Cp), 3.95(s, 1H, Cp), 3.33
and 3.73 (AB quartet, J A–B=14.5 Hz, 2H, CH2) ppm.
13C-NMR: (l, CDCl3) 230 (d, JP–C=16.6 Hz, Mo–
CO), 223.8 (d, JP–C=17.8 Hz, Mo–CO), 223, 218.6
and 214.4 (Mo–CO), 133 (broad d, o-Ph, about 2:1),
130 (s, p-Ph), 128 (d, JC–P=7 Hz, m-Ph), 116 (ipso-
Cp), 105 (ipso-Cp), 98, 94, 89, 88, 87, 85, 84 and 83
(Cp), 27 (CH2) ppm. 31P-NMR: (l, CDCl3) 74.5 (Mo–
PPh3) ppm. Anal. Calc. C34H25O5Mo2P2: C, 55.45; H,
3.43; P, 4.21%. Found: C, 55.26; H, 3.28; P, 4.02%.
4.6. (v-p5,p5-C5H4–CH2–C5H4)W2(CO)4(PPh3)2, 8
See procedure (Section 4.1). Result: red–purple crys-
tals, m.p. 263–270°C, yield 30%. IR (CH2Cl2): 1911(m),

96
T.E. Bitterwolf et al. / Journal of Organometallic Chemistry 562 (1998) 89–96
1
1824 and 1792(s) cm−1. H-NMR: (l, CDCl3) 7.38 (br,
30H, triphenylphosphine), 5.65 (s, 2H, Cp), 4.60 (s, 2H,
Cp), 4.36 (s, 2H, Cp), 4.21 (d, 2H, Cp), 3.71 (s, 2H,
CH2) ppm. 13C-NMR: (l, CDCl3) 229 (d, JC–P=10 Hz,
CO, W–CO), 225 (d, JC–P=10 Hz, CO, W–CO),137
(broad, ipso-Ph), 133 (broad s, o-Ph), 129 (s, p-Ph), 128
(d, m-Ph), 110 (ipso-Cp), 96, 87, 86.8 and 86 (Cp), 27.8
(CH2) ppm. 31P-NMR: (l, CDCl3) 42 (W–PPh3) ppm.
Anal. Calc. C51H40O4W2P2: C, 53.45; H, 3.53; P, 5.58%.
Found: C, 53.95; H, 4.26; P, 5.12%
group for both was unambiguously determined from
systematic absences in the data. Only small variations
in azimuthal scan intensities were seen (B10%) and
corrections for absorption were ignored. The structures
were solved by direct methods. In both cases, the
asymmetric unit contains two independent, but chemi-
cally similar, molecules. In 4, one of the two indepen-
dent molecules is disordered in two positions in an
approximately equal occupation; we were able to re-
solve all atoms of the two sites except for one of the
bridging CH2 groups. In 4, cyclopentadienyl and phenyl
rings were rigidly constrained and carbon atoms of the
distorted molecule were isotropically refined. Other-
wise, all non-hydrogen atoms in both structures were
anisotropically refined. Hydrogen atoms were treated as
idealized contributions except for those in the disor-
dered molecule of 4 where they were ignored. All
computations used SHELXTL software (version 5.03,
G. Sheldrick, Siemens XRD, Madison, WI).
4.7. (v-p5,p5-C5H4–CMe2–C5H4)Mo2(CO)4(PPh3)2, 9
See procedure (Section 4.1). Result: red–purple crys-
tals, m.p. 323–326°C. IR (CH2Cl2): 1917(m), 1832 and
1798(s) cm−1 1H-NMR: (l, CDCl3) 7.37 (br, 30H,
.
triphenylphosphine), 5.71 (s, 2H, Cp), 4.50 (s, 2H, Cp),
4.25 (s, 4H, Cp), 1.36 (s, 6H, CH3) ppm. 13C-NMR: (l,
CDCl3) 241 (d, JC–P=24.3 Hz, CO, Mo–CO), 234 (d,
JC–P=24.3 Hz, CO, Mo–CO), 133 (d, JC–P=10 Hz,
o-Ph), 129 (s, p-Ph), 128 (d, JC–P=7 Hz, m-Ph), 95.3
(s, ipso-Cp), 88.2, 87.7 and 84.3 (Cp), 31 (CH3) ppm.
31P-NMR: (l, CDCl3) 79.8 (Mo–PPh3) ppm. Anal.
Calc. C53H44O4Mo2P2: C, 63.73; H, 4.45; P, 6.2%.
Found: C, 63.15; H, 5.08; P, 5.98%.
Acknowledgements
Abdel Saygh wishes to thank the government of
Saudi Arabia and the King Abdulaziz Military
Academy for their generous financial support. The Uni-
versity of Delaware acknowledges the National Science
Foundation for their support of the purchase of the
CCD-based diffractometer (Grant CHE-9628768).
4.8. (v-p5,p5-C5H4–CH2–C5H4)Mo2(CO)4(PMe3)2, 11
See procedure (Section 4.1). Result: red–purple crys-
tals, m.p. 250–252°C, yield 35%. IR (CH2Cl2): 1907(m),
1
1817 and 1784(s) cm−1. H-NMR: (l, CDCl3) 5.36 (m,
References
2H, Cp), 5.05 (m, 2H, Cp), 4.58 (m, 4H, Cp), 3.38 (s,
2H, CH2), 1.55(d, 18H, PMe3) ppm. 13C-NMR: (l,
CDCl3) 240 (d, JC–P=25.3 Hz, CO, Mo–CO), 233 (d,
[1] (a) J.R. Knorr, T.L. Brown, J. Am. Chem. Soc. 115 (1993) 4087.
(b) J. Peters, M.W. George, J.J. Turner, Organometallics 14
(1995) 1503–1506.
JC–P=25.3 Hz, CO, Mo–CO), 109.5 (ipso-Cp), 97, 87,
85 and 84 (Cp), 28 (CH2), 21 (d, JC–P=29.9 Hz, PMe3)
ppm. 31P-NMR: (l, CDCl3) 32.6 (Mo–PMe3) ppm.
MS: 600 (M+, 13%), 544 (M+ −2CO, 32%), 149
(100%). Anal. Calc. C21H28O4Mo2P2: C, 42.16; H, 4.73;
P, 10.35%. Found: C, 41.93; H, 4.49; P, 10.66%.
[2] N.N. Turaki, J.M. Huggins, Organometallics 4 (1985) 1766.
[3] (a) A.E. Stiegman, M. Stieglitz, D.R. Tyler, J. Am. Chem. Soc.
105 (1983) 6032–6037. (b) C.E. Philbin, A.S. Goldman, D.R.
Tyler, Inorg. Chem. 25 (1986) 4434.
[4] (a) R. Fierro, T.E. Bitterwolf, A.L. Rheingold, G.P.A. Yap,
L.M. Liable-Sands, J. Organomet. Chem. 524 (1996) 19. (b) T.E.
Bitterwolf, A.L. Rheingold, Organometallics 11 (1991) 3856.
[5] T.E. Bitterwolf, M.L. Baker, P.E. Bloyce, A.K. Campen, A.J.
Rest, J. Chem. Soc. Dalton Trans. (1990) 833.
[6] J.T. Bays, T.E. Bitterwolf, Unpublished observations, 1996.
[7] M. Tilset, K.P.C. Vollhardt, R. Boese, Organometallics 13
(1994) 3146.
[8] J.S. Drage, P.C. Vollhardt, Organometallics 5 (1986) 280.
[9] M. Scheer, Personal communication, 1995.
5. Crystallographic structure determinations
Crystallographic data are collected in Table 1. Crys-
tals of both 4 and 11 were obtained as red blocks and
found to possess monoclinic symmetry. The space
.
.