Nucleophilic behaviour of dioxo- and thiooxophosphorane complexes [MoCp(CO)2{E,P-EP(O)(2,4,6-C6H2tBu3)}]- (E = O, S)

Article abstract of DOI:10.1039/c4dt01942c

The title anions were prepared as (DBU-H)⁺ salts upon reaction of the oxophosphinidene complex (H-DBU)[MoCp(CO)₂{P(O)R}] with either dimethyldioxirane or elemental sulphur (R = 2,4,6-C₆H₂^tBu₃; Cp = η⁵-C₅H₅, DBU = 1,8-diazabicyclo[5.4.0]undec-7-ene). The dioxophosphorane complex failed to react with MeI at room temperature, but reacted readily with (Me₃O)BF₄ to give the phosphonite complex [MoCp{O,P-OP(OMe)R}(CO)₂]. In contrast, the thiooxophosphorane complex reacted with MeI to give the thiolophosphinide derivative [MoCp{S,P-(MeS)P(O)R}(CO)₂], whereas its reaction with (Me₃O)BF₄ gave a mixture of the latter complex and the phosphonothiolate isomer [MoCp{S,P-SP(OMe)R}(CO)₂] in similar amounts. Other electrophiles were added selectively to the terminal O atom of the RPOS ligand. Thus the thiooxophosphorane complex reacted with ClC(O)C₂H₃, [NH₄]PF₆, ClSiMe₃, ClSnMe₃ and [ZrCp₂Cl₂] to give the corresponding derivatives [MoCp{S,P-SP(OX)R}(CO)₂] (X = C(O)C₂H₃, H, SiMe₃, SnMe₃, ZrCp₂Cl). The structure of two of these products (X = C(O)C₂H₃, SiMe₃) was determined by single-crystal X-ray diffraction studies. Density functional theory (DFT) calculations of the title anions and some of their derivatives indicated that attachment of an external electrophile to the terminal O atom of the thiooxophosphorane ligand is favoured under the conditions of charge control, while the sulphur atom is the favoured site under the conditions of orbital control, although it leads to less stable products. This journal is

Full text of DOI:10.1039/c4dt01942c

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Nucleophilic behaviour of dioxo- and
thiooxophosphorane complexes [MoCp(CO)2-
Cite this: Dalton Trans., 2014, 43,
16074
t
{E,P-EP(O)(2,4,6-C6H2 Bu3)}]− (E = O, S)†
María Alonso, M. Angeles Alvarez, M. Esther García, Daniel García-Vivó* and
Miguel A. Ruiz*
The title anions were prepared as (DBU-H)+ salts upon reaction of the oxophosphinidene complex
t
(H-DBU)[MoCp(CO)2{P(O)R*}] with either dimethyldioxirane or elemental sulphur (R* = 2,4,6-C6H2 Bu3;
Cp = η5-C5H5, DBU = 1,8-diazabicyclo[5.4.0]undec-7-ene). The dioxophosphorane complex failed to
react with MeI at room temperature, but reacted readily with (Me3O)BF4 to give the phosphonite complex
[MoCp{O,P-OP(OMe)R*}(CO)2]. In contrast, the thiooxophosphorane complex reacted with MeI to give
the thiolophosphinide derivative [MoCp{S,P-(MeS)P(O)R*}(CO)2], whereas its reaction with (Me3O)BF4
gave a mixture of the latter complex and the phosphonothiolate isomer [MoCp{S,P-SP(OMe)R*}(CO)2] in
similar amounts. Other electrophiles were added selectively to the terminal O atom of the R*POS ligand.
Thus the thiooxophosphorane complex reacted with ClC(O)C2H3, [NH4]PF6, ClSiMe3, ClSnMe3 and
[ZrCp2Cl2] to give the corresponding derivatives [MoCp{S,P-SP(OX)R*}(CO)2] (X = C(O)C2H3, H, SiMe3,
SnMe3, ZrCp2Cl). The structure of two of these products (X = C(O)C2H3, SiMe3) was determined by
single-crystal X-ray diffraction studies. Density functional theory (DFT) calculations of the title anions and
some of their derivatives indicated that attachment of an external electrophile to the terminal O atom of
the thiooxophosphorane ligand is favoured under the conditions of charge control, while the sulphur
atom is the favoured site under the conditions of orbital control, although it leads to less stable products.
Received 27th June 2014,
Accepted 1st September 2014
DOI: 10.1039/c4dt01942c
www.rsc.org/dalton
more detailed study of their chemical behaviour. However, the
coordination chemistry of these molecules is virtually
Introduction
Dioxophosphoranes (RPO2) and thiooxophosphoranes (RPOS) unknown, with just one example described for a dioxophos-
are unstable molecules generated during thermolysis of a large phorane complex.3 In a preliminary study we found that the
variety of suitable organophosphorus precursors. These transi- anionic oxophosphinidene complex (H-DBU)[MoCp(CO)2{P(O)-
ent molecules display a strong electrophilic character located R*}] (1) (R* = 2,4,6-C6H2 Bu3; Cp = η5-C5H5, DBU = 1,8-diazabi-
t
at their positively-polarized phosphorus(V) atom, thus enabling cyclo[5.4.0]undec-7-ene) could be used as a precursor for the
their use as strong phosphorylating agents upon reaction with corresponding dioxophosphorane (H-DBU)[MoCp(CO)2{O,P-
many different organic substrates while, in the absence of OP(O)R*}] (2) and thiooxophosphorane (H-DBU)[MoCp(CO)2-
external reagents, they usually evolve through polymerization {S,P-SP(O)R*}] (3) derivatives (Scheme 1).4 Because of the nega-
or intramolecular activation of C–H bonds.1 Only under very tive charge of the latter complexes, the acid–base chemistry of
favourable conditions can these species be detected directly, as their phosphorus ligands is reversed, so they can now behave
is the case for MePO2 when generated in the gas phase under as nucleophiles. Indeed our preliminary study on the reactivity
high vacuum.2 It might be conceived that substantial stabiliz- of the thiooxophosphorane complex 3 indicated that methyl-
ation of these elusive molecules could be achieved through
coordination to metal centres, which in turn would allow for a
Departamento de Química Orgánica e Inorgánica/IUQOEM, Universidad de Oviedo,
E-33071 Oviedo, Spain. E-mail: garciavdaniel@uniovi.es, mara@uniovi.es
†Electronic supplementary information (ESI) available: A PDF file containing
the complete ref. 33 and details of DFT calculations (drawings, atomic coordi-
nates and energies of compounds 2, 3, 5, 6, and 8). CCDC 995392 (7) and
995393 (9). For ESI and crystallographic data in CIF or other electronic format
see DOI: 10.1039/c4dt01942c
Scheme 1
16074 | Dalton Trans., 2014, 43, 16074–16083
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 1 Selected IRa and 31P{1H}NMRb data for new compounds
ation could occur easily at either the S or O sites, to give unpre-
cedented thiolophosphinide (PR(O)(SMe)) and phosphonothio-
late (PR(OMe)S) derivatives, respectively.4 Therefore it was of
interest to explore the potential of anions 2 and 3 in the gene-
ration of novel P-donor entities. In this paper we give full
details of the above reactions, which we have now extended to
the dioxophosphorane complex 2 and also to other electro-
philic reagents, including the proton and some metal-based
electrophiles. As will be shown, most electrophiles add to the
terminal oxygen atom of the phosphorus ligand of the above
anions, an event favoured on steric, electrostatic and thermo-
dynamic grounds, according to our theoretical calculations.
Addition to the sulphur site of the thiooxophosphorane ligand
is an orbital-favoured event only observed in the methylation
reactions of 3.
Compound
ν(CO)
δP (JPW)
(H-DBU)[MoCp(CO)2{P(O)R*}] (1)c
(H-DBU)[WCp(CO)2{P(O)R*}] (1′)c
1874 (vs), 1790 (s)d 385.0
1868 (vs), 1784 (s) 335.6 (658)
(H-DBU)[MoCp(CO)2{O,P-OP(O)R*}] (2) 1904 (vs), 1803 (s) 39.6
(H-DBU)[MoCp(CO)2{S,P-SP(O)R*}] (3) 1911 (vs), 1816 (s) 83.8
(H-DBU)[WCp(CO)2{S,P-SP(O)R*}] (3′) 1902 (vs), 1803 (s) 43.3 (240)
[MoCp{O,P-OP(OMe)R*}(CO)2] (4)
[MoCp{S,P-(MeS)P(O)R*}(CO)2] (5)
[MoCp{S,P-SP(OMe)R*}(CO)2] (6)
1948 (vs), 1856 (s) 73.1
1963 (vs), 1887 (s) 102.0
1947 (vs), 1862 (s) 122.0
[MoCp{S,P-SP(OC(O)C2H3)R*}(CO)2] (7) 1961 (vs), 1879 (s) 107.4
[MoCp{S,P-SP(OH)R*}(CO)2] (8)
[MoCp{S,P-SP(OSiMe3)R*}(CO)2] (9)
1950 (vs), 1862 (s) 101.2e
1946 (vs), 1858 (s) 97.0
[MoCp{S,P-SP(OSnMe3)R*}(CO)2] (10) 1935 (vs), 1843 (s) 101.0
[MoCp{S,P-SP(OZrCp2Cl)R*}(CO)2] (11) 1938 (vs), 1846 (s) 106.0
a Recorded in dichloromethane solution, with C–O stretching bands
[ν(CO)] in cm−1 b Recorded in CD2Cl2 at 121.50 MHz and 295 K, with
.
31P–183W coupling constants (JPW) in Hz. c Data taken from ref. 5;
DBU = 1,8-diazabicyclo[5.4.0]undec-7-ene. d Recorded in tetrahydrofuran
solution. e Recorded at 162.01 MHz and 263 K.
Results and discussion
Synthesis and structure of dioxo- and thiooxophosphorane
complexes
Upon addition of the chalcogen atom to the oxophosphinid-
ene ligand, the P nucleus undergoes a dramatic nuclear shield-
ing of some 300 ppm, consistent with a change from planar
trigonal to a distorted tetrahedral environment (Table 1). We
note that the chemical shift of 2 is very close to those
measured for the only other dioxophosphorane complexes
described in the literature (δP ca. 40 ppm for [Ru(η6-p-cymene)
{O,P-OP(O)R*}(PR3)]; R = Ph, Cy).3 At the same time, the P–W
coupling in the tungsten derivative 3′ is more than halved with
respect to its precursor 1′, thus reflecting the increased coordi-
nation number of the P and W atoms and formal reduction in
the W–P bond order (from 2 to 1).7 Because of the absence of
any symmetry elements in these molecules, their 1H and 13C
NMR spectra display separate resonances for the pairs of CO
ligands, CH(aryl) moieties or ortho-tBu substituents, often
exhibiting substantial differences in the corresponding coup-
lings to the P atom. Such differences are particularly strong for
the carbonyl ligands, due to their distinct positioning (hence
different C–M–P angles) with respect to the P atom (e.g. 29 vs.
∼0 Hz for JPC in 3).8 These spectroscopic features are also
present in all neutral derivatives of anions 2 and 3 and will not
be further discussed.
In order to gain further support for the structure of these
anions and to better understand their reactivity toward
different electrophiles we have computed the geometric and
electronic structures of the anions in 2 and 3 using density
functional theory (DFT) methods.9 We have also inspected the
topological properties of the electron density as managed by
the atoms in molecules (AIM) theory.10 First we note that the
optimized structures of these anions (Fig. 1 and Table 2) are as
expected from the considerations given above. Thus, the short
MovP distance of 2.287 Å in 1 is enlarged to values of
ca. 2.50 Å, consistent with single Mo–P bonds, while the term-
inal PvO bond lengths of 1.51 Å have remained almost unper-
turbed (cf. 1.518 Å in 1),5 and consistent with a considerable
multiple bond character, close to a formal order of two. We
The oxophosphinidene complex 1 reacts readily at 243 K with
a mild oxygenating reagent as dimethyldioxirane to yield the
corresponding dioxophosphorane derivative (H-DBU)[MoCp-
(CO)2{O,P-OP(O)R*}] (2) as a major product. This compound,
however, was quite air-sensitive and could not be obtained as a
pure material, although it could be used for further reactions.
In a related way, 1 reacts rapidly at 273 K with S8 to give the
thiooxophosphorane derivative (H-DBU)[MoCp(CO)2{S,P-SP(O)-
R*}] (3), a more stable species that could be isolated as a pure
material in a conventional way. These reactions can be easily
extended to the tungsten analogue of 1. Thus, the latter reacts
with S8 to give the corresponding tungsten derivative (H-DBU)-
[WCp(CO)2{S,P-SP(O)R*}] (3′) in good yield. The formation of
the above compounds can be viewed as resulting from
addition of a chalcogen atom (E) to the MovP bond via the
HOMO of the starting oxophosphinidene complex, which has
π(Mo–P) bonding character and displays lobes oriented per-
pendicular to the Mo–P–O plane,5 and is thus expected to
leave the newly incorporated atom E nearly perpendicular to
that plane (Scheme 1), in agreement with the calculations on 2
and 3 described below.
Spectroscopic data in solution for compounds 2, 3 and 3′
(Table 1 and the Experimental section) are fully consistent
with the structure proposed for the corresponding anions. As
expected, these complexes display C–O stretches in the IR
spectrum at frequencies higher than the starting material, and
with a similar pattern (very strong and strong, in the order of
decreasing frequencies), this being characteristic of M(CO)2
oscillators with C–M–C angles somewhat lower than 90°.6 It is
perhaps surprising (based on the relative electronegativity of
O vs. S) that the dioxophosphorane 2 displays bands some
10 cm−1 lower than the thiooxophosphorane analogue 3,
which indicates a more efficient electron donation from the
O,P-bound dioxophosphorane ligand to the metal centre.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 16074–16083 | 16075

View Article Online
Paper
Dalton Transactions
Table 3 Selected atomic charges for the anions of 2 and 3
2
3
Atom
Mulliken
NPA
Mulliken
NPA
Mo
E
P
−0.151
−0.671
1.080
−0.138
−0.983
1.874
−0.4706
−0.266
0.832
−0.349
−0.485
1.424
O(term)
−0.615
−1.081
−0.604
−1.072
items to look at. The relevant atomic charges computed for 2
and 3 are collected in Table 3, these including Mulliken
charges12 and those derived from the NBO analysis (NPA
charges).13 Both schemes give a similar picture: a high and
comparably negative charge at both O atoms in 2 and a nega-
tive charge much higher at O than at the S atom for the thiooxo-
phosphorane complex 3. Thus we conclude that, under the
conditions of charge control, compound 2 might not display a
particular site preference for attachment of an incoming elec-
trophile, but the thiooxophosphorane complex 3 should
display a definite preference for attachment at the terminal O
site, which moreover is likely a more accessible site on steric
grounds. The frontier orbitals of these anions (see the ESI†)
are not very different from each other, except for the corres-
ponding HOMO, which is largely located at the metal site in
the anion of 2, whereas it has a large contribution from a
sulphur lone pair in the anion of 3 (Fig. 2), thus enabling in
the latter case the attachment of an incoming electrophile at
the sulphur position under the conditions of orbital control.
Fig. 1 DFT-optimized structures of the anions in compounds 2 (left)
and 3 (right), with H atoms omitted for clarity.
note that the computed P–O lengths for free P(O)R* (1.501 Å),5
and MePO2 (1.461 Å),2 are only slightly shorter than the above
figures. In contrast, the P–O length for the Mo-bound oxygen
in 2 is substantially longer, as expected for an interaction
approaching the single-bond character. We finally note that
the Mo–P length in the dioxophosphorane complex 2 is ca.
0.07 Å shorter than the corresponding distance in the thio-
oxophosphorane 3, suggesting a stronger binding of the R*PO2
ligand to the metal centre, in agreement with the lower C–O
frequencies of 2 noted above.
Inspection of the topology of electron density involving the
EP(O)R* ligands in anions 2 and 3 gives a picture consistent
with that derived from the above geometrical analysis (see the
ESI†). Thus, the electron density (ρ) at the corresponding
Mo–P bond critical points (bcp) follows the expected order
(0.636 (1) ≫ 0.488 (2) > 0.463 (3) e Å−3), while that at the term-
inal PvO bond remains with a high value of 1.40–1.43 e Å−3
(cf. 1.431 e Å−3 in 1) and with a very high value of ca. 35 e Å−5
for its Laplacian (∇2ρ), which is a feature characteristic of mul-
tiple CO bonds in organic molecules and transition-metal car-
bonyls.11 In contrast, the P–O bond for the Mo-bound O atom
displays a lower electron density at the corresponding bcp
(1.240 e Å−3), with a negative value for its Laplacian (−17.6 e
Å−5), as expected for a covalent single bond.
Methylation and related reactions of complexes 2 and 3
The dioxophosphorane complex 2 failed to react with the soft
methylating agent MeI at room temperature, but reacts readily
with the stronger reagent (Me3O)BF4 to give selectively the
phosphonite complex [MoCp{O,P-OP(OMe)R*}(CO)2] (4),
a
derivative formally resulting from selective attachment of the
Me+ cation to the terminal O atom of the former dioxophos-
phorane ligand (Chart 1). In contrast, methylation of the
thiooxophosphorane complex 3 is strongly dependent on the
Concerning the Lewis-base behaviour of compounds 2 and
3, atomic charges and occupied frontier orbitals are the major
Table 2 Selected bond lengths (Å) and angles (°) for DFT-optimized geometries
Parameter
1a
2
3
5
6
8-S
8-O
Mo–P
Mo–E
P–E
2.287
2.472
2.227
1.580
1.509
78.9
62.2
38.9
131.9
120.1
80.5
84.1
103.2
2.536
2.592
2.094
1.516
64.5
67.3
48.2
128.2
119.2
79.7
82.4
104.7
2.514
2.578
2.283
1.506
64.8
64.8
53.2
125.6
110.1
79.5
80.7
108.0
2.478
2.617
2.049
1.647
62.8
69.9
47.3
125.8
112.0
79.9
88.5
106.3
2.508
2.587
2.343
1.501
60.9
64.3
54.7
125.8
112.0
79.4
80.4
106.6
2.470
2.620
2.046
1.655
62.5
70.2
47.3
121.1
112.3
79.8
87.2
106.7
PvO/P–OC/H
Mo–E–P
E–P–Mo
P–Mo–E
Mo–P–O
E–P–O
OC–Mo–CO
OC–Mo–P
OC–Mo–P
1.518
136.0
85.0
89.4
90.7
a Data taken from ref. 5.
16076 | Dalton Trans., 2014, 43, 16074–16083
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 2 DFT-computed HOMO orbitals for the anions in compounds
2 (left) and 3 (right).
Fig. 3 ORTEP diagram (30% probability) of compound 7 with H atoms
(except the olefinic ones) and tBu groups (except their C1 atoms)
omitted for clarity.
now determined the structure of the related complex 7, which
displays comparable geometrical parameters (Fig.
3 and
Table 4). A salient feature of this structure is the presence of
an S,P-bound phosphonothiolate ligand derived from attach-
ment of the hydrocarbyl group at the O-site of the thiooxophos-
phorane ligand of 3. As a result, the P–O distance (1.656(1) Å)
becomes much longer than the distance of 1.516 Å computed
for the PvO bond in 3 (Table 2), as expected for a single bond
between these atoms. In contrast, the P–S bond (2.0022(6) Å)
becomes slightly shorter than the corresponding length com-
puted for 3 (2.094 Å), and it should be considered as relatively
short for a P–S single bond (expected around 2.12 Å).14 Simi-
larly short P–S lengths have been found previously for different
S,P:P-bridged chalcogenophosphinidene Mo2 complexes, a cir-
cumstance which we have linked to the involvement of the
chalcogen atom in some π(S–P) bonding interaction within the
MoPS ring of these molecules.15 On the other hand, the Mo–P
bond of 3 is not expected to be much perturbed upon alkyl-
ation, and displays a value of 2.3963(6) Å, indicative of a strong
single bond. A final structural feature deserving comment is
the distorted trigonal pyramidal environment around the
P atom, derived from the fact that this atom remains placed
almost in the same plane as the Mo, O3 and C11 atoms
(Σ(X–P–Y) = 354°). Although no S,P-bound phosphonothiolate
complex other than 6 has been structurally characterized, we
note that this local environment is also present in related
O,P-bound phosphinite and phosphonite complexes,4,5 and
also in the S,P-bound thiophosphinite (or phosphinothiolate)
complexes [WCp(SPCltBu)(CO)2],16 [WCp{SPMeN(SiMe3)2}(CO)2]17
and [Mo(C6H7)(SPMe2)(CO)2].18
Chart 1
electrophilic reagent used. Thus, its reaction with MeI takes
place selectively at the sulphur site to give the thiolophos-
phinide derivative [MoCp{S,P-(MeS)P(O)R*}(CO)2] (5), whereas
its reaction with (Me3O)BF4 yields a mixture of 5 and the phos-
phonothiolate complex [MoCp{S,P-SP(OMe)R*}(CO)2] (6) in
similar amounts, with the latter resulting from attachment of
the Me+ cation to the oxygen atom of the thiooxophosphorane
ligand. In view of the considerations made above, it is clear
that 5 is the product of orbital control and thus is preferen-
tially formed when using soft methylating reagents, whereas 6
emerges under the conditions of charge control. In reactions
of the dioxophosphorane 2, interaction with the incoming
electrophile is probably dominated by electrostatic factors and,
since both oxygen atoms of the R*PO2 ligand bear comparable
negative charges, the reason for the observed selective methyl-
ation at the terminal O site would be the superior accessibility
of that site to an external reagent. In line with this hypothesis,
we note that reaction of 3 with acryloyl chloride gives exclu-
sively the corresponding phosphonothiolate derivative [MoCp-
{S,P-SP(OC(O)C2H3)R*}(CO)2] (7), here suggesting the operation
of steric barriers disfavouring the approach of an external
reagent, certainly softer than (Me3O)BF4, to the S site.
Spectroscopic data in solution for the phosphonothiolate
complexes 6 and 7 (Table 1 and the Experimental section) are
fully consistent with their asymmetric solid-state structures
and deserve no particular comment. We just note that they
display 31P NMR resonances some 30 ppm more deshielded
Structure of phosphonite and phosphonothiolate derivatives
The structure of the O-methylated derivative 6 was established
during our preliminary study of the reactivity of 3.4 We have
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 16074–16083 | 16077

View Article Online
Paper
Dalton Transactions
short (1.492(3) Å) and comparable to the corresponding figure
computed for 3. The general environment around the P atom
is also little perturbed, with the Mo, P, O and C31 atoms
remaining almost coplanar (Σ(X–P–Y) = 359°), as found also
for the isomeric phosphonothiolate complexes 6 and 7. As
expected, the S-bound methyl group points away from the
bulky R* substituent at phosphorus, thus yielding a less con-
gested diastereoisomer.
Table 4 Selected bond lengths (Å) and angles (°) for compound 7
Mo1–P1
Mo1–S1
Mo1–C1
Mo1–C2
P1–S1
2.3963(6)
2.5570(6)
1.968(2)
1.956(2)
2.0022(6)
1.656(1)
1.830(2)
1.380(2)
1.186(2)
1.309(3)
C1–Mo1–C2
P1–Mo1–C1
P1–Mo1–C2
P1–Mo1–S1
P1–S1–Mo1
Mo1–P1–S1
Mo1–P1–O3
P1–O3–C3
76.5(1)
105.2(1)
86.8(1)
47.55(2)
62.01(2)
70.44(2)
128.66(4)
126.0(1)
123.7(2)
109.4(2)
P1–O3
P1–C11
O3–C3
C3–O4
C4–C5
O3–C3–O4
O3–C3–C4
In order to find the relative stability of the phosphono-
thiolate or thiolophosphinide derivatives of 3 we computed the
structures of isomers 5 and 6 using DFT methods. The mini-
mized structures (Table 2 and ESI†) were in good agreement
with those determined crystallographically, although the com-
puted bond lengths involving the Mo atoms were somewhat
longer than the experimental data, a common trend of the func-
tionals currently used in DFT computations of transition metal
complexes,9a,19 and the P–S length for 5 also was significantly
longer than the experimental value (2.283 vs. 2.1805(6) Å). The
Gibbs free energy computed for the thiolophosphinide
complex 5 was 21 kJ mol−1 above that of its isomer 6, thus
qualifying it as a kinetic product. This energetic difference is
perhaps mainly derived from the different strength of the new
bonds being formed in each case (the S–C bond energy is
some 85 kJ mol−1 lower than that of an O–C bond).20 More-
over, the electron donation of the phosphonothiolate ligand of
6 seems to be more efficient than that of the isomeric thiolo-
phosphinide ligand in 5, as judged from the respective C–O
stretching frequencies, which are some 15 cm−1 lower for 6
(Table 1).
than that of the starting anion 3, while their C–O stretches are
substantially more energetic, as expected from the reduction
of charge in these neutral derivatives. As for the phosphonite
complex 4, all data point to an O,P-bound structure similar to
that of the sulphur-containing analogues 6 and 7. Actually,
such a coordination geometry has been confirmed crystallo-
graphically for [MoCp{O,P-OP(OC6H4OH)R*}(CO)2],4 a complex
resulting from reaction of 1 with p-benzoquinone, which inci-
dentally remains the only O,P-bound phosphonite complex
structurally characterized so far. Spectroscopic data for the
latter complex are indeed comparable to those of 4, particu-
larly its C–O stretches (1945 (vs), 1865 (s) cm−1) and 31P chemi-
cal shift (δP 74.5 ppm, cf. 73.1 ppm for 4), which again are
some 30 ppm higher than that of the parent anion 2.
Structure of the thiolophosphinide complex 5
The structure of the S-methylated derivative 5 also was estab-
lished during our preliminary study of the reactivity of 3
(Fig. 4).4 No other complex containing a comparable ligand
seems to have been structurally characterized so far. As a
result of the attachment of the methyl group to the sulphur
atom of 3, any π-bonding interaction involving the S atom
vanishes, and accordingly the P–S length in 5 is elongated to a
value of 2.1805(6) Å, while the P–O bond is little perturbed as
judged from the corresponding bond length, which remains
Protonation of compound 3
The thiooxophosphorane ligand in 3 is readily protonated
upon reaction with a weak acid such as [NH4]PF6 in dichloro-
methane solution to give the corresponding neutral derivative
[MoCp{S,P-SP(OH)R*}(CO)2] (8) as a unique product, a process
that can be quantitatively reversed upon reaction of the latter
with DBU. This quite air-sensitive complex displays C–O stretch-
ing bands almost identical to those of the O-methylated deriva-
tive 6, with frequencies significantly lower than those of the
S-methylated derivative 5, thus suggesting that the added
proton is attached to the terminal O site of the anion of 3,
rather than attached to the sulphur atom (Chart 2). The appear-
ance of a broad resonance at ca. 6.8 ppm in the 1H NMR
spectrum of 8 also supports the presence of an OH group.
Further support for the proposed structure of 8 was
obtained by computing the gas-phase structures of the S-proto-
Fig. 4 ORTEP diagram (30% probability) of compound 5 with H atoms
and tBu groups (except their C1 atoms) omitted for clarity.4 Selected
bond lengths (Å) and angles (°): Mo1–P3
= 2.4415(5); Mo1–S1 =
2.4939(5); P3–S1 = 2.1805(6); P3–O3 = 1.492(1); P3–C31 = 1.847(1);
C11–Mo1–C12 = 78.4(1).
Chart 2
16078 | Dalton Trans., 2014, 43, 16074–16083
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
nated (8-S) and O-protonated (8-O) derivatives of the anion of 3
(see Table 2 and ESI†). Indeed the computed structures are
very similar to those of the methylated products 5 and 6
respectively (Table 2), and the Gibbs free energy for 8-S is now
much higher (by 51 kJ mol−1) than that of 8-O, once more
reflecting the relative strengths of the newly formed bonds (the
S–H bond energy is some 95 kJ mol−1 lower than that of an
O–H bond).20 We should stress that no other complex contain-
ing a thiolophosphonous (or phosphonothious) acid ligand
(PR(OH)(SH)) or any of their conjugated anions appears to
have been reported in the literature.
Heterometallic derivatives of compound 3
We have shown previously that reactions of the oxophosphinid-
ene complex 1 with metal-based electrophiles may lead to
products resulting from attachment of the added electrophile
to the terminal O atom of the phosphorus ligand, to give oxo-
phosphinidene-bridged derivatives, or to the metal atom of the
anion, to give products with new heterometallic Mo–M bonds,
with the latter being the preferred outcome when using soft
electrophiles.21 We have now examined reactions of the thiooxo-
phosphorane complex 3 with analogous reagents which, by
taking into account the outcome of the methylation and proto-
nation reactions discussed above, might be expected to take
place at either the S or O atoms of the anion. However, these
reactions instead turned out to occur selectively at the terminal
O atom of the anion irrespective of the electrophile being used.
Compound 3 reacts readily with the relatively hard electro-
philes SiMe3Cl, SnMe3Cl and [ZrCp2Cl2] to give the corres-
ponding phosphonothiolate-like derivatives [MoCp{S,P-SP-
(OSiMe3)R*}(CO)2] (9), [MoCp{S,P-SP(OSnMe3)R*}(CO)2] (10)
and [MoCp{S,P-SP(OZrCp2Cl)R*}(CO)2] (11), all of them dis-
playing a thiooxophosphorane ligand bridging the metal
atoms in a novel μ-S,P:O- fashion (Chart 3). Reactions with
softer electrophiles such as the Au(I) complexes [Au(THT)(PR3)]-
PF6 (R = Ph, Me; THT = SC4H8) yielded similar derivatives, as
judged from the corresponding IR and NMR spectra, but these
products were quite unstable and could not be properly iso-
lated and characterized.
Fig. 5 ORTEP diagram (30% probability) of compound 9 with H atoms
and tBu groups (except their C1 atoms) omitted for clarity.
Table 5 Selected bond lengths (Å) and angles (°) for compound 9
Mo1–P1
Mo1–S1
Mo1–C1
Mo1–C2
P1–S1
P1–O3
P1–C8
2.417(2)
2.544(2)
1.956(7)
1.973(7)
2.011(2)
1.579(4)
1.838(6)
1.680(4)
C1–Mo1–C2
P1–Mo1–C1
P1–Mo1–C2
P1–Mo1–S1
P1–S1–Mo1
Mo1–P1–S1
Mo1–P1–O3
P1–O3–Si1
80.3(3)
107.3(2)
84.6(2)
47.74(5)
62.82(6)
69.45(6)
127.1(2)
146.6(3)
O3–Si1
while at the same time alleviating the steric pressure around the
P atom. For instance, in the cis and trans isomers of the
complex [PdCl2{P(OSiMe3)3}2] the P–O lengths fall in the range
1.55–1.57 Å (O–Si lengths ca. 1.67 Å), while the P–O–Si angles
span a wide range of 136–152°.22 In the same line, the tungsten
complex [W(CO)5{PH(OSiMe3)(CH(SiMe3)2)}] displays values of
ca. 1.60 Å and 142° for these geometrical parameters.23
Spectroscopic data in solution for compounds 9 to 11
(Table 1 and the Experimental section) are fully consistent
with the structure found for 9 in the crystal, and are compar-
able to those of the phosphonothiolate complexes 6 and 7,
thus deserving no detailed analysis. We just note that the C–O
stretching frequencies of these complexes broadly reflect the
expected trend based on the electron withdrawing power of the
electrophile being attached to the O atom of the phosphorus
ligand, by following the sequence 6 (CH3) > 9 (SiMe3) ≫ 11
(ZrCp2Cl) > 10 (SnMe3). The relative order of the tin and zirco-
nium derivatives is not the expected one on the basis of rela-
tive electronegativities of Sn and Zr, and probably reflects a
superior involvement of the latter atom in a π(O→Zr) inter-
action that in the end would remove electron density off the
molybdenum centre.
The structure of the silyl derivative 9 in the crystal (Fig. 5
and Table 5) is very similar to those determined for the hydro-
carbyl derivatives 6 and 7 and need not be discussed in detail,
except for the P–O–Si linkage, which displays a relatively large
P–O–Si angle of 146.6° (cf. 126° for the analogous P–O–C angle
in 7), and a P–O length slightly shorter than that in 7 (1.579(4)
vs. 1.656(1) Å). This is not unusual in the chemistry of silicon,
and might be associated with some participation of the lone
electron pairs at oxygen in π bonding along the P–O–Si chain,
Conclusions
The formation of the title compounds can be viewed as result-
Chart 3
ing from addition of a chalcogen atom (E) to the MovP bond
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 16074–16083 | 16079

View Article Online
Paper
Dalton Transactions
in the parent oxophosphinidene complex [MoCp(CO)2{P(O)-
Preparation of (H-DBU)[MoCp(CO)2{O,P-OP(O)R*}] (2).
R*}]− via the HOMO of the latter, which displays lobes oriented Dimethyldioxirane (1.0 mL of a ca. 0.1 M solution in acetone,
perpendicular to the Mo–P–O plane, and is thus expected to 0.10 mmol) was added to a dichloromethane solution (10 mL)
leave the newly incorporated atom E nearly perpendicular to of compound 1 (0.030 g, 0.045 mmol) at 243 K, and the
that plane, in agreement with their DFT-optimized structures. mixture was stirred for 20 min to give an orange solution. The
The thiooxophosphorane complex 3 bears a negative charge at solvent was then removed under vacuum, the residue was
O much higher than that at the S atom, while its HOMO has a extracted with diethyl ether (15 mL) and the extract was filtered
large contribution from a sulphur lone pair. This favours the using a cannula. Removal of the solvent from the filtrate and
attachment of an incoming electrophile to the sulphur atom washing of the residue with petroleum ether gave compound 2
under the conditions of orbital control, thus explaining the as an air-sensitive, reasonably pure orange powder (0.016 g,
selective formation of the thiolophosphinide complex 5 in the 52%). Attempts to further purify this product resulted in its
1
reaction of 3 with the soft reagent MeI. In contrast, attachment progressive decomposition. H NMR (200.13 MHz): δ 11.78 (s,
of the incoming electrophile to the O atom is favoured under 1H, NH), 7.34 (dd, JHP = 5, JHH = 2, 1H, C6H2), 7.24 (t, JHP
=
the conditions of charge control and is also kinetically JHH = 2, 1H, C6H2), 4.81 (s, 5H, Cp), 3.47–3.20 (m, 6H, CH2),
t
favoured on steric grounds, thus explaining the preferential or 2.79, 1.96 (2m, 2 × 2H, CH2), 1.76, 1.52, 1.32 (3s, 3 × 9H, Bu),
specific formation of phosphonothiolate and related com- 1.75–1.64 (m, 6H, CH2).
plexes in the reactions of 3 with all other electrophilic reagents
Preparation of (H-DBU)[MoCp(CO)2{S,P-SP(O)R*}] (3).
investigated in this work. According to our DFT calculations Sulphur (0.73 mL of a ca. 0.062 N solution in dichloro-
on the H+ and Me+ derivatives of 3, the phosphonothiolate methane, 0.045 mmol) was added to a dichloromethane solu-
complexes invariably are thermodynamically favoured over tion (10 mL) of compound 1 (0.030 g, 0.045 mmol) at 273 K,
their thiolophosphinide isomers, likely as a result of the con- and the mixture was stirred for 2 min to give an orange solu-
siderably higher strength of the newly formed O–H and O–C tion. Solvent was then removed under vacuum, the residue was
bonds relative to the alternative formation of S–H and S–C extracted with diethyl ether (2 × 10 mL) and the extracts were
bonds.
filtered. Removal of solvent from the filtrate and washing of
the residue with petroleum ether gave compound 3 as an
orange powder (0.029 g, 93%). Anal. Calcd for
C34H51MoN2O3PS: C, 58.78; H, 7.40; N, 4.03. Found: C, 59.34;
H, 7.17; N, 4.12. 1H NMR: δ 12.28 (s, 1H, NH), 7.21 (dd, JHP = 5,
JHH = 2, 1H, C6H2), 7.09 (t, JHP = JHH = 2, 1H, C6H2), 4.71 (s, 5H,
Experimental section
General procedures and starting materials
Cp), 3.38–3.28 (m, 6H, CH2), 2.75, 1.94 (2m, 2 × 2H, CH2),
All manipulations and reactions were carried out under a 1.76–1.61 (m, 6H, CH2), 1.71, 1.64, 1.28 (3s, 3 × 9H, Bu). 13C
nitrogen (99.995%) atmosphere using standard Schlenk tech- {1H} NMR: δ 254.6 (d, JCP = 29, MoCO), 247.1 (s, MoCO), 165.7
niques. Solvents were purified according to literature pro- (s, CN), 156.4 [d, JCP = 21, C2(C6H2)], 153.7 [s, C4(C6H2)], 148.4
cedures, and distilled prior to use.24 Compounds (H-DBU)- [s, C2(C6H2)], 144.1 [d, JCP = 46, C1(C6H2)], 121.7 [d, JCP = 10,
[MoCp{P(O)R*}(CO)2] (1) and (H-DBU)[WCp{P(O)R*}(CO)2] (1′) C3(C6H2)], 120.4 [d, JCP = 15, C3(C6H2)], 94.2 (s, Cp), 53.2, 47.9,
t
t
(R* = 2,4,6-C6H2 Bu3; Cp = η5-C5H5; DBU = 1,8-diazabicyclo 38.2, 32.0, 29.0, 26.7, 24.4, 19.4 (8s, CH2), 41.5, 40.1, 34.7 [3s,
[5.4.0]undec-7-ene),5 and dimethyldioxirane,25 were prepared C1(tBu)], 34.8, 34.3, 31.2 [3s, C2(tBu)].
as described previously, and all other reagents were obtained
Preparation of (H-DBU)[WCp(CO)2{S,P-SP(O)R*}] (3′). The
from the usual commercial suppliers and used as received, procedure is analogous to that described for compound 3, but
unless otherwise stated. Petroleum ether refers to that fraction using compound 1′ (0.030 g, 0.040 mmol) as the starting
distilling in the range 338–343 K. Filtrations were carried out material. After a similar workup, compound 3′ was obtained as
through diatomaceous earth unless otherwise indicated. an orange solid (0.030 g, 96%). Anal. Calcd for
Chromatographic separations were carried out using jacketed C34H51N2O3PSW: C, 52.18; H, 6.57; N, 3.58. Found: C, 52.44; H,
1
columns refrigerated by tap water (ca. 288 K) or by a closed 6.38; N, 3.49. H NMR (200.13 MHz): δ 12.11 (s, 1H, NH), 7.21
2-propanol circuit kept at the desired temperature with a cryo- (dd, JHP = 5, JHH = 2, 1H, C6H2), 7.10 (t, JHP = JHH = 2, 1H,
stat. Commercial aluminium oxide (activity I, 150 mesh) was C6H2), 4.80 (s, 5H, Cp), 3.43–3.30 (m, 6H, CH2), 2.78, 1.95 (2m,
degassed under vacuum prior to use. The latter was mixed 2 × 2H, CH2), 1.77–1.70 (m, 6H, CH2) 1.73, 1.64, 1.28 (3s,
afterwards under nitrogen with an appropriate amount of 3 × 9H, tBu).
water to reach the activity desired (activity IV, unless otherwise
stated). IR stretching frequencies of CO ligands were measured (Me3O)BF4 (0.010 g, 0.067 mmol) was added to a dichloro-
in solution, using CaF2 windows. Nuclear magnetic resonance methane solution (10 mL) of compound (0.040 g,
Preparation of [MoCp{O,P-OP(OMe)R*}(CO)2] (4). Solid
2
(NMR) spectra were routinely recorded at 300.13 (1H), 121.50 0.059 mmol) at 273 K, and the mixture was stirred for 5 min to
(31P{1H}) or 75.47 MHz (13C{1H}) at 295 K in CD2Cl2 solution give an orange solution. The solvent was then removed under
unless otherwise stated. Chemical shifts (δ) are given in ppm, vacuum and the residue was chromatographed on alumina at
relative to internal tetramethylsilane (1H, 13C) or external 85% 253 K. Elution with dichloromethane–petroleum ether (1/9)
aqueous H3PO4 (31P). Coupling constants (J) are given in hertz. gave a pink fraction yielding, after removal of solvents, com-
16080 | Dalton Trans., 2014, 43, 16074–16083
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
pound 4 as an orange solid (0.029 g, 91%). Anal. Calcd for on alumina at 285 K. Elution with dichloromethane–petroleum
1
C26H37MoPO4: C, 57.78; H, 6.90. Found: C, 57.54; H, 7.04. H ether (1/3) gave an orange fraction yielding, after removal of
NMR: δ 7.47 (dd, JHP = 6, JHH = 2, 1H, C6H2), 7.37 (t, JHP = JHH
=
solvents, compound 7 as an orange solid (0.026 g, 84%). The
2, 1H, C6H2), 5.05 (s, 5H, Cp), 3.44 (d, JHP = 14, 3H, OMe), 1.64, crystals used in the X-ray study were grown from a concen-
t
1.45, 1.33 (3s, 3 × 9H, Bu). 13C{1H} NMR: δ 253.7 (d, JCP = 30, trated petroleum ether solution of the complex at 253 K. Anal.
MoCO), 246.9 (d, JCP = 3, MoCO), 156.6 [d, JCP = 24, C2(C6H2)], Calcd for C28H37MoO4PS: C, 56.37; H, 6.25. Found: C, 56.33;
1
153.9 [s, C4(C6H2)], 152.3 [d, JCP = 3, C2(C6H2)], 130.0 [d, JCP
=
H, 6.34. H NMR: δ 7.38 (dd, JHP = 7, JHH = 2, 1H, C6H2), 7.29
58, C1(C6H2)], 123.8 [d, JCP = 10, C3(C6H2)], 123.7 [d, JCP = 18, (t, JHP = JHH = 2, 1H, C6H2), 6.42 (dd, JHH = 17, JHP = 1, 1H,
C3(C6H2)], 95.5 (s, Cp), 52.9 (d, JCP = 5, OMe), 39.7, 35.1, 30.0 trans-CH2), 6.07 (ddd, JHH = 17, JHH = 10, JHP = 1, 1H, CH), 5.91
[3s, C1(tBu)], 33.9, 32.6, 30.9 [3s, C2(tBu)].
(d, JHH = 10, 1H, cis-CH2), 4.93 (s, 5H, Cp), 1.63 (s, 18H, Bu),
t
Preparation of [MoCp{S,P-(MeS)P(O)R*}(CO)2] (5). Methyl 1.31 (s, 9H, Bu). 13C{1H} NMR: δ 251.0 (d, JCP = 28, MoCO),
t
iodide (27 μL, 0.43 mmol) was added to a dichloromethane 237.1 (s, MoCO), 160.1 [d, JCP = 10, C(O)C2H3], 158.5 [d, JCP
=
solution (10 mL) of compound 3 (0.030 g, 0.043 mmol) at 27, C2(C6H2)], 152.7 [d, JCP = 5, C2(C6H2)], 156.7 [s, C4(C6H2)],
290 K, and the mixture was stirred for 15 min to give a yellow 133.2 [d, JCP = 59, C1(C6H2)], 132.8 (s, CH2), 129.3 (s, CH),
solution. Solvent was then removed under vacuum and the 123.7 [d, JCP = 11, C3(C6H2)], 123.0 [d, JCP = 18, C3(C6H2)], 94.5
residue was chromatographed on alumina at 253 K. Elution (s, Cp), 41.2 [d, JCP = 3, C1(tBu)], 40.5 [d, JCP = 5, C1(tBu)], 35.1
with tetrahydrofuran–petroleum ether (1/3) gave a yellow frac- [s, C1(tBu)], 33.9, 33.6, 30.9 [3s, C2(tBu)].
tion which yielded, after removal of solvents, compound 5 as a
Preparation of [MoCp{S,P-SP(OH)R*}(CO)2] (8). Solid (NH4)
yellow microcrystalline solid (0.022 g, 92%). Anal. Calcd for PF6 (0.017 g, 0.104 mmol) was added to a tetrahydrofuran solu-
1
C26H37MoPO3S: C, 56.11; H, 6.70. Found: C, 55.65; H, 6.56. H tion (10 mL) of compound 3 (0.036 g, 0.052 mmol) at room
NMR: δ 7.33 (dd, JHP = 6, JHH = 2, 1H, C6H2), 7.20 (t, JHP = JHH
=
temperature, and the mixture was stirred for 10 min to give an
2, 1H, C6H2), 4.59 (s, 5H, Cp), 2.09 (d, JHP = 3, 3H, SMe), 1.67, orange solution. After removal of solvent under vacuum, the
t
1.46, 1.30 (3s, 3 × 9H, Bu). 13C{1H} NMR: δ 242.1 (d, JCP = 40, residue was extracted with petroleum ether and the extract was
MoCO), 239.7 (s, MoCO), 157.0 [d, JCP = 26, C2(C6H2)], 152.3 [s, filtered. Removal of solvent from the filtrate yielded compound
1
C4(C6H2)], 151.0 [s, C2(C6H2)], 143.0 [d, JCP = 28, C1(C6H2)], 8 as an orange, very air-sensitive solid (0.020 g, 71%). H NMR
122.5 [d, JCP = 8, C3(C6H2)], 122.3 [d, JCP = 15, C3(C6H2)], 93.7 (200.13 MHz): δ 7.33 (dd, JHP = 6, JHH = 2, 1H, C6H2), 7.22 (t,
(s, Cp), 41.6, 39.9, 35.0 [3s, C1(tBu)], 34.2, 33.4, 30.9 [3s, JHP = JHH = 2, 1H, C6H2), 6.82 (br, 1H, OH), 4.91 (s, 5H, Cp),
t
C2(tBu)], 19.0 (s, SMe).
1.65, 1.62, 1.29 (3s, 3 × 9H, Bu). 13C{1H} NMR (243 K): δ 251.7
Preparation of [MoCp{S,P-SP(OMe)R*}(CO)2] (6). Solid (d, JCP = 28, MoCO), 241.1 (s, MoCO), 157.8 [d, JCP = 24,
(Me3O)BF4 (0.010 g, 0.067 mmol) was added to a dichloro- C2(C6H2)], 151.5 [d, JCP = 4, C2(C6H2)], 156.3 [s, C4(C6H2)],
methane solution (10 mL) of compound
3
(0.040 g, 133.1 [d, JCP = 62, C1(C6H2)], 122.9 [d, JCP = 11, C3(C6H2)], 122.2
0.058 mmol) at 273 K, and the mixture was stirred for 2 min to [d, JCP = 17, C3(C6H2)], 94.5 (s, Cp), 41.1 [d, JCP = 4, C1(tBu)],
give an orange-yellow solution. Solvent was then removed 40.1 [d, JCP = 5, C1(tBu)], 34.9 [s, C1(tBu)], 33.9, 33.5, 30.9
under vacuum and the residue was chromatographed on [3s, C2(tBu)].
alumina at 253 K. Elution with dichloromethane–petroleum
ether (1/9) gave a yellow fraction which yielded, after removal SiMe3Cl (0.014 μL, 0.108 mmol) was added to a dichloro-
of solvents, compound 6 as an orange microcrystalline solid methane solution (10 mL) of compound (0.036 g,
Preparation of [MoCp{S,P-SP(OSiMe3)R*}(CO)2] (9). Neat
3
(0.016 g, 50%). Elution with tetrahydrofuran–petroleum ether 0.052 mmol) at room temperature, and the mixture was stirred
(1/3) gave a yellow fraction which yielded analogously com- for 10 min to give an orange solution. After removal of the
pound 5 as a yellow microcrystalline solid (0.014 g, 43%). Anal. solvent under vacuum, the residue was extracted with pet-
Calcd for C26H37MoPO3S: C, 56.12; H, 6.70. Found: C, 56.58; roleum ether and the extracts were filtered. Removal of solvent
1
H, 6.55. H NMR: δ 7.37 (dd, JHP = 6, JHH = 2, 1H, C6H2), 7.26 from the filtrate yielded compound 9 as an orange solid
(t, JHP = JHH = 2, 1H, C6H2), 4.91 (s, 5H, Cp), 3.18 (d, JHP = 16, (0.029 g, 91%). The crystals used in the X-ray study were grown
t
3H, OMe), 1.62, 1.57, 1.30 (3s, 3 × 9H, Bu). 13C{1H} NMR: from a concentrated petroleum ether solution of the complex
δ 251.1 (d, JCP = 27, MoCO), 239.4 (s, MoCO), 156.4 [d, JCP = 24, at 253 K. Anal. Calcd for C28H43MoO3PSSi: C, 54.71; H, 7.05.
C2(C6H2)], 156.7 [s, C4(C6H2)], 152.1 [d, JCP = 4, C2(C6H2)], Found: C, 54.62; H, 7.14. 1H NMR (200 MHz): δ 7.31 (dd, JHP = 6,
131.8 [d, JCP = 66, C1(C6H2)], 123.3 [d, JCP = 11, C3(C6H2)], 122.5 JHH = 2, 1H, C6H2), 7.20 (t, JHP = JHH = 2, 1H, C6H2), 4.83 (s, 5H,
t
[d, JCP = 18, C3(C6H2)], 94.4 (s, Cp), 52.7 (d, JCP = 7, OMe), 41.1, Cp), 1.64, 1.61, 1.29 (3s, 3 × 9H, Bu), 0.21 (s, 9H, SiMe). 13C
40.3 [2d, JCP = 4, C1(tBu)], 34.9 [s, C1(tBu)], 33.8, 35.3, 30.9 [3s, {1H} NMR: δ 252.6 (d, JCP = 29, MoCO), 240.7 (s, MoCO), 157.5
C2(tBu)].
[d, JCP = 26, C2(C6H2)], 151.2 [d, JCP = 4, C2(C6H2)], 155.0 [s,
Preparation of [MoCp{S,P-SP(OC(O)C2H3)R*}(CO)2] (7). Neat C4(C6H2)], 137.7 [d, JCP = 60, C1(C6H2)], 122.8 [d, JCP = 11,
acryloyl chloride (10 μL, 0.121 mmol) was added to a dichloro- C3(C6H2)], 122.2 [d, JCP = 17, C3(C6H2)], 94.4 (s, Cp), 41.2 [d, JCP
methane solution (10 mL) of compound
0.052 mmol) at room temperature, and the mixture was stirred 33.7, 30.9 [3s, C2(tBu)], 1.0 (s, SiMe).
3
(0.036 g, = 3, C1(tBu)], 40.3 [d, JCP = 4, C1(tBu)], 35.0 [s, C1(tBu)], 34.2,
for 10 min to give an orange solution. Solvent was then
Preparation of [MoCp{S,P-SP(OSnMe3)R*}(CO)2] (10). The
removed under vacuum, and the residue was chromatographed procedure is analogous to that described for 9, but using solid
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 16074–16083 | 16081

View Article Online
Paper
Dalton Transactions
SnMe3Cl (0.031 g, 0.156 mmol) instead. After a similar
workup, compound 10 was isolated as an orange solid
(0.032 g, 87%). Anal. Calcd for C28H43MoO3PSSn: C, 47.68; H,
Table 6 Crystal data for new compounds
7
9
1
6.14. Found: C, 47.44; H, 6.19. H NMR (200 MHz): δ 7.28 (dd,
Mol. formula
C
28H37MoO4PS
C28H43MoO3PSSi
614.69
Monoclinic
P21/c
Mol. Wt.
Cryst. syst.
Space group
Radiation (λ, Å)
a (Å)
b (Å)
c (Å)
α (°)
596.55
Triclinic
ˉ
P1
JHP = 5, JHH = 2, 1H, C6H2), 7.16 (t, JHP = JHH = 2, 1H, C6H2),
4.78 (s, 5H, Cp), 1.64, 1.61, 1.29 (3s, 3 × 9H, Bu), 0.57 (s, 9H,
t
SnMe). 13C{1H} NMR: δ 253.0 (d, JCP = 28, MoCO), 243.4 (s,
MoCO), 156.8 [d, JCP = 23, C2(C6H2)], 154.1 [s, C4(C6H2)], 150.2
[s, C2(C6H2)], 139.8 [d, JCP = 57, C1(C6H2)], 122.3 [d, JCP = 11,
C3(C6H2)], 121.3 [d, JCP = 16, C3(C6H2)], 94.1 (s, Cp), 41.2 [d,
JCP = 2, C1(tBu)], 40.1 [d, JCP = 5, C1(tBu)], 34.9 [s, C1(tBu)], 34.2,
33.6, 31.0 [3s, C1(tBu)], −1.2 (s, SnMe).
0.71073
1.54184
10.891(2)
10.945(2)
13.589(3)
90.011(3)
100.921(3)
114.499(3)
1441.7(5)
2
1.374
0.613
120(2)
2.05 to 27.15
−17, 13; −14, 14;
0, 17
17 186
6327 (0.0267)
5731
R1 = 0.0228
wR2 = 0.060b
R1 = 0.0267
wR2 = 0.0615b
1.07
0/329
0.47, −0.402
14.9470(3)
10.0787(2)
21.2080(4)
90
106.4360(10)
90
3064.35(10)
4
1.332
5.205
150(2)
3.08 to 68.21
−17, 17; 0, 12;
0, 25
19 119
5568 (0.0476)
4899
R1 = 0.0537
wR2 = 0.1473c
R1 = 0.0616
wR2 = 0.1575c
1.079
β (°)
γ (°)
V (Å3)
Preparation of [MoCp{S,P-SP(OZrCp2Cl)R*}(CO)2] (11). The
procedure is analogous to that described for 9, but using solid
[ZrCp2Cl2] (0.015 g, 0.052 mmol) instead. After a similar
workup, compound 11 was isolated as an orange solid (0.034 g,
Z
Calcd. density (g cm−3
)
Absorp. Coeff. (mm−1
Temperature (K)
θ range (°)
)
Index ranges (h, k, l)
82%). Anal. Calcd for C35H44ClMoO3PSZr: C, 52.65; H, 5.55.
1
Found: C, 52.24; H, 5.39. H NMR (200 MHz): δ 7.30 (dd, JHP
=
No. of reflns. collected
No. of indep. reflns. (Rint
Reflns with I > 2σ(I)
6, JHH = 2, 1H, C6H2), 7.19 (t, JHP = JHH = 2, 1H, C6H2), 6.54,
)
6.49, 4.83 (3s, 3 × 5H, Cp), 1.73, 1.62, 1.30 (3s, 3 × 9H, tBu).
R indexes [data with I > 2σ(I)]a
R indexes (all data)a
X-Ray crystal structure determination for compound 7
Data collection was performed on a Smart-CCD-1000 BRUKER
diffractometer using graphite-monochromated Mo-Kα radi-
ation at 120 K. Cell dimensions and orientation matrixes were
initially determined from least-squares refinements on reflec-
tions measured in 3 sets of 30 exposures collected in 3
different ω regions and eventually refined against all reflec-
tions. The software SMART26 was used for collecting frames of
data, indexing reflections, and determining lattice parameters.
The collected frames were then processed for integration using
the software SAINT,26 and a multi-scan absorption correction
was applied with SADABS.27 Using the program suite WinGX,28
the structure was solved by Patterson interpretation and phase
expansion using SHELXL97,29 and refined with full-matrix
least squares on F2 using SHELXL97. All non-hydrogen atoms
were refined anisotropically. All hydrogen atoms were fixed at
calculated geometric positions except for H(4), which was
located in the Fourier map and refined. All hydrogen atoms
were given an overall isotropic thermal parameter. Crystallo-
graphic data and structure refinement details are collected in
Table 6.
GOF
No. of restraints/params.
0/328
1.398, −0.787
Δρ(max, min), e Å−3
a R = Σ||Fo| − |Fc||/Σ|Fo|. wR = [Σw(|Fo|2 − |Fc|2)2/ Σw|Fo|2]1/2. w =
1/[σ2(Fo2) + (aP)2 + bP] where P = (Fo + 2Fc2)/3. b a = 0.0313, b = 0.4264.
2
c a = 0.0000, b = 34.3529.
given an overall isotropic thermal parameter. Crystallographic
data and structure refinement details are collected in Table 6.
Computational details
All DFT calculations were carried out using the GAUSSIAN03
package,33 in which the hybrid method B3LYP was used with
the Becke three-parameter exchange functional34 and the Lee–
Yang–Parr correlation functional.35 An accurate numerical
integration grid (99 590) was used for all the calculations via
the keyword Int = Ultrafine. Effective core potentials and their
associated double-ζ LANL2DZ basis set were used for the metal
atoms.36 The light elements (P, O, S, C and H) were described
with the 6-31G* basis.37 Geometry optimizations were per-
formed under no symmetry restrictions, using initial coordi-
nates derived from X-ray data of compounds 5 and 6, and
frequency analyses were performed to ensure that minimum
structures with no imaginary frequencies were achieved. Mole-
cular orbitals and vibrational modes were visualized using the
MOLEKEL program.38 The topological analysis of the electron
density was carried out with the Xaim routine.39
X-Ray crystal structure determination for compound 9
Data collection was performed at 150 K on a Nonius Kappa-
CCD single crystal diffractometer using Cu-Kα radiation.
Images were collected at a 29 mm fixed crystal–detector dis-
tance, using the oscillation method, with 1.5° oscillation and
60 s exposure time per image. Data collection strategy was
determined with the program Collect.30 Data reduction and
cell refinement were performed with the programs HKL Denzo
and Scalepack.31 A semi-empirical absorption correction was
applied using the program SORTAV.32 Structure solution and
refinements were performed as described for 7. All non-hydro-
Acknowledgements
gen atoms were refined anisotropically, and all hydrogen We thank the DGI of Spain and the Consejería de Educación
atoms were fixed at calculated geometric positions and were del Principado de Asturias for financial support (projects
16082 | Dalton Trans., 2014, 43, 16074–16083
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
CTQ2012-33187 and IB05-110), and the MEC of Spain for a 16 W. Malisch, K. Grün, U.-A. Hirth and M. Noltemeyer,
grant (to M. A.). We also thank the CMC of the Universidad de J. Organomet. Chem., 1996, 513, 31.
Oviedo for access to computing facilities and the X-ray units of 17 H.-U. Reisacher, W. F. McNamara, E. N. Duesler and
the Universidad de Oviedo and Universidad de Santiago de
Compostela for acquisition of diffraction data.
R. T. Paine, Organometallics, 1997, 16, 449.
18 H. Alper, F. W. B. Einstein, F. W. Harstock and R. H. Jones,
Organometallics, 1987, 6, 829.
19 C. J. Cramer, Essentials of Computational Chemistry, Wiley,
Chichester, U.K., 2nd edn, 2004.
20 J. E. Huheey, E. A. Keiter and R. L. Keiter, Inorganic
Chemistry: Principles of Structure and Reactivity, Harper-
Collins College Publishers, New York, USA, 4th edn, 1993,
p. A-25.
21 M. Alonso, M. A. Alvarez, M. E. García, M. A. Ruiz,
H. Hamidov and J. C. Jeffery, Inorg. Chem., 2010, 49,
11595.
22 T. Gebauer, G. Frenzen and K. Dehnicke, Z. Naturforsch.,
B: Chem. Sci., 1993, 48, 1661.
23 H. Wilkens, A. Ostrowski, J. Jeske, F. Ruthe, P. G. Jones and
R. Streubel, Organometallics, 1999, 18, 5627.
24 W. L. F. Armarego and C. Chai, Purification of Laboratory
Chemicals, Butterworth-Heinemann, Oxford, U. K., 5th edn,
2003.
25 W. Adam, J. Bialas and L. Hadjiarapoglou, Chem. Ber.,
1991, 124, 2377.
26 SMART & SAINT Software Reference Manuals, version 5.051
(Windows NT version), Bruker Analytical X-ray Instruments,
Madison, WI, 1998.
27 G. M. Sheldrick, SADABS, Program for Empirical Absorption
Correction, University of Göttingen, Göttingen, Germany,
1996.
28 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
29 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-
logr., 2008, 64, 112.
30 Collect, Nonius BV, Delft, The Netherlands, 1997–2004.
31 Z. Otwinowski and W. Minor, Methods Enzymol., 1997, 276,
307.
32 R. H. Blessing, Acta Crystallogr., Sect. A: Fundam. Crystal-
logr., 1995, 51, 33.
33 M. J. Frisch, et al., Gaussian 03, Revision B.02, Gaussian,
Inc., Wallingford CT, 2004.
34 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
35 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens.
Matter, 1988, 37, 785.
36 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 299.
37 (a) P. C. Hariharan and J. A. Pople, Theor. Chim. Acta, 1973,
28, 213; (b) G. A. Petersson and M. A. Al-Laham, J. Chem.
Phys., 1991, 94, 6081; (c) G. A. Petersson, A. Bennett,
T. G. Tensfeldt, M. A. Al-Laham, W. A. Shirley and
J. Mantzaris, J. Chem. Phys., 1988, 89, 2193.
38 S. Portmann and H. P. Lüthi, MOLEKEL, An Interactive Mole-
cular Graphics Tool, CHIMIA 2000, 54, 766.
39 J. C. Ortiz and C. Bo, Xaim, Departamento de Química
Física e Inorgánica, Universidad Rovira i Virgili, Tarragona,
Spain, 1998.
Notes and references
1 L. D. Quin, Coord. Chem. Rev., 1994, 137, 525.
2 L. N. Heydorn, P. C. Burgers, P. J. A. Ruttink and
J. K. Terlouw, Chem. Phys. Lett., 2003, 368, 584, and refer-
ences therein.
3 R. Menye-Biyogo, F. Delpech, A. Castel, H. Gornitzka and
P. Rivière, Angew. Chem., Int. Ed., 2003, 42, 5610.
4 M. Alonso, M. A. Alvarez, M. E. García, M. A. Ruiz, H. Hamidov
and J. C. Jeffery, J. Am. Chem. Soc., 2005, 127, 15012.
5 M. Alonso, M. A. Alvarez, M. E. García, D. García-Vivó and
M. A. Ruiz, Inorg. Chem., 2010, 49, 8962.
6 P. S. Braterman, Metal Carbonyl Spectra, Academic Press,
London, U. K., 1975.
7 C. J. Jameson, in Phosphorus-31 NMR Spectroscopy in Stereo-
chemical Analysis, ed. J. G. Verkade and L. D. Quin, VCH,
Deerfield Beach, FL, 1987, ch. 6.
2
8 A general trend established for JXY in complexes of the
type [MCpXYL2] is that |Jcis| > |Jtrans|. See, for instance,
ref. 7 and also B. Wrackmeyer, H. G. Alt and H. E. Maisel,
J. Organomet. Chem., 1990, 399, 125.
9 (a) W. Koch and M. C. Holthausen, A Chemist’s Guide to
Density Functional Theory, Wiley-VCH, Weinheim, Germany,
2nd edn, 2002; (b) T. Ziegler, Chem. Rev., 1991, 91, 651;
(c) J. B. Foresman and Æ. Frisch, Exploring Chemistry with
Electronic Structure Methods, Gaussian, Inc., Pittsburgh,
2nd edn, 1996.
10 (a) R. F. W. Bader, Atoms in Molecules – A Quantum Theory,
Oxford University Press, Oxford, 1990; (b) R. F. W. Bader,
Chem. Rev., 1991, 91, 893.
11 P. Macchi and A. Sironi, Coord. Chem. Rev., 2003, 238–239,
383.
12 R. S. Mulliken, J. Chem. Phys., 1955, 23, 1833.
13 Mulliken population analysis fails to give a useful and
reliable characterization of the charge distribution in many
cases, especially when highly ionic compounds and diffuse
basis functions are involved. Charges calculated according
to the natural population analysis (NPA) do not show these
deficiencies and are more independent of the basis set:
(a) A. E. Reed, R. B. Weinstock and F. J. Weinhold, Chem.
Phys., 1985, 83, 735; (b) A. E. Reed, L. A. Curtis and
F. Weinhold, Chem. Rev., 1988, 88, 899.
14 B. Cordero, V. Gómez, A. E. Platero-Prats, M. Revés,
J. Echeverría, E. Cremades, F. Barragán and S. Alvarez,
Dalton Trans., 2008, 2832.
15 B. Alvarez, M. A. Alvarez, I. Amor, M. E. García, D. García-
Vivó, M. A. Ruiz and J. Suárez, Inorg. Chem., 2012, 51, 7810.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 16074–16083 | 16083