Chemistry of the oxophosphinidene ligand. 2. Reactivity of the anionic complexes [MCp{P(O)R*}(CO)2]- (M = Mo, W; R* = 2,4,6-C6H2tBu3) toward electrophiles based on elements diffe

Article abstract of DOI:10.1021/ic101839k

The anionic oxophosphinidene complexes (H-DBU)[MCp{P(O)R ^*}(CO)₂] (M = Mo, W; R* = 2,4,6-C ₆H₂^tBu₃; Cp = η⁵- C₅H₅, DBU = 1,8-diazabicyclo [5.4.0] undec-7-en

Full text of DOI:10.1021/ic101839k

Inorg. Chem. 2010, 49, 11595–11605 11595
DOI: 10.1021/ic101839k
Chemistry of the Oxophosphinidene Ligand. 2. Reactivity of the Anionic
t
Complexes [MCp{P(O)R*}(CO)2]- (M = Mo, W; R* = 2,4,6-C6H2 Bu3) Toward
Electrophiles Based on Elements Different from Carbon.
†
†
´
´
Marıa Alonso, M. Angeles Alvarez, M. Esther Garcıa,* Miguel A. Ruiz,* Hayrullo Hamidov, and John C. Jeffery
´
ꢀ
ꢀ
Departamento de Quımica Organica e Inorganica/IUQOEM, Universidad de Oviedo, 33071 Oviedo, Spain, and
†School of Chemistry, University of Bristol, Bristol BS8 1TS, U.K.
Received September 7, 2010
The anionic oxophosphinidene complexes (H-DBU)[MCp{P(O)R*}(CO)2] (M = Mo, W; R* = 2,4,6-C6H2tBu3; Cp = η5-C5H5,
DBU = 1,8-diazabicyclo [5.4.0] undec-7-ene) displayed multisite reactivity when faced with different electrophilic
reagents. The reactions with the group 14 organochloride compounds ER4-xClx (E = Si, Ge, Sn, Pb) led to either
phosphide-like, oxophosphinidene-bridged derivatives [MCp{P(OE0)R*}(CO)2] (E0=SiMe3, SiPh3, GePh3,
GeMe2Cl) or to terminal oxophosphinidene complexes [MCp{P(O)R*}(CO)2(E0)] (E0 = SnPh3, SnPh2Cl, PbPh3;
˚
Mo-Pb = 2.8845(4) A for the MoPb compound). A particular situation was found in the reaction with SnMe3Cl, this
giving a product existing in both tautomeric forms, with the phosphide-like complex [MCp{P(OSnMe3)R*}(CO)2]
prevailing at room temperature and the tautomer [MCp{P(O)R*}(CO)2(SnMe3)] being the unique species present
below 203 K in dichloromethane solution. The title anions also showed a multisite behavior when reacting with
transition-metal based electrophiles. Thus, the reactions with the complexes [M0Cp2Cl2] (M0 = Ti, Zr) gave phosphide-like
derivatives [MCp{P(OM0)R*}(CO)2] (M = Mo, M0 = TiCp2Cl, ZrCp2Cl; M = W, M0 = ZrCp2Cl), displaying a bridging κ1,κ1-P,
0
˚
˚
O- oxophosphinidene ligand connecting MCp(CO)2 and M Cp2Cl metal fragments (W-P = 2.233(1) A, O-Zr = 2.016(4) A
for the WZr compound]. In contrast, the reactions with the complex [AuCl{P(p-tol)3}] gave the metal-metal bonded
˚
derivatives trans-[MCp{P(O)R*}(CO)2{AuP(p-tol)3}] (M = Mo, W; Mo-Au = 2.7071(7) A). From all the above results it
was concluded that the terminal oxophosphinidene complexes are preferentially formed under conditions of orbital control,
while charge-controlled reactions tend to give derivatives with the electrophilic fragment bound to the oxygen atom of the
oxophosphinidene ligand (phosphide-like, oxophosphinidene-bridged derivatives).
Introduction
In the context of our studies on the chemistry of binuclear
phosphinidene complexes of the transition metals, we re-
cently found a selective synthetic route to the molybdenum
compound (H-DBU) [MoCp{P(O)R*}(CO)2] (1), the first
anionic oxophosphinidene complex reported in the
literature.5a Our preliminary experiments revealed that the
expectedly high nucleophilicity of this complex would open
Oxophosphinidenes (R-PdO) are unstable molecules of
relevance as intermediates in the synthesis of different organo-
phosphorus products.1,2 They can be isolobally-related to the
electrophilic carbenes,3 and thus might be expected to act as
ligands to many very different metal fragments. However, in
spite of the rich variety of coordination modes available for
these molecules, only a very limited number of complexes
have been described in the literature so far,4and their chemi-
cal behavior has not been explored.
(4) (a) Niecke, E.; Engelmann, M.; Zorn, H.; Krebs, B.; Henkel, G.
Angew. Chem., Int. Ed. Engl. 1980, 19, 710. (b) Hitchcock, P. B.; Johnson, J. A.;
Lemos, M. A. N. D. A.; Meidine, M. F.; Nixon, J. F.; Pombeiro, A. J. L. J. Chem.
ꢀ
Soc., Chem. Commun. 1992, 645. (c) Alvarez, M. A.; García, M. E.; Gonzalez,
R.; Ramos, A.; Ruiz, M. A. Organometallics 2010, 29, 1875. (d) Johnson,
M. J. A.; Odom, A. L.; Cummins, C. C. J. Chem. Soc., Chem. Commun. 1997,
1523. (e) Kourkine, V.; Glueck, D. S. Inorg. Chem. 1997, 36, 5160. (f) Schmitt,
*To whom correspondence should be addressed. E-mail: mara@uniovi.es
(M.A.R.).
(1) (a) Gaspar, P. P.; Qian, H.; Beatty, A. M.; d’Avignon, D. A.; Kao, J.
L.-F.; Watt, J. C.; Rath, N. P. Tetrahedron 2000, 56, 105. (b) Cowley, A. H.;
Gabbaï, F. P.; Corbelin, S.; Decken, A. Inorg. Chem. 1995, 34, 5931. (c) Wang,
K.; Emge, T. J.; Goldman, A. S. Organometallics 1994, 13, 2135.
(2) (a) Stille, J. K.; Eichelberger, J. L.; Higgins, J.; Freeburger, M. E.
J. Am. Chem. Soc. 1972, 94, 4761. (b) Nakayama, S.; Yoshifuju, M.; Okazaki,
R.; Inamoto, N. Bull. Chem. Soc. Jpn. 1975, 48, 546. (c) Quast, H.; Heuschmann,
M. Angew. Chem., Int. Ed. Engl. 1978, 17, 867. (d) Quin, L. D.; Yao, E. U.;
Szewczyk, J. Tetrahedron Lett. 1987, 28, 1077.
€
G.; Ullrich, D.; Wolmershauser, G.; Regitz, M.; Scherer, O. J. Z. Anorg. Allg.
Chem. 1999, 625, 702. (g) Alvarez, C. M.; Alvarez, M. A.; García, M. E.;
ꢀ
Gonzalez, R.; Ruiz, M. A. Organometallics 2005, 24, 5503. (h) Buchholz, D.;
Huttner, G.; Imhof, W. J. Organomet. Chem. 1990, 388, 307.
´
(5) (a) Alonso, M.; Garcıa, M. E.; Ruiz, M. A.; Hamidov, H.; Jeffery, J. C.
J. Am. Chem. Soc. 2004, 126, 13610. (b) Alonso, M.; Alvarez, M. A.; García,
M. E.; Ruiz, M. A.; Hamidov, H.; Jeffery, J. C. J. Am. Chem. Soc. 2005, 127,
15012. (c) Alonso, M.; Alvarez, M. A.; García, M. E.; Ruiz, M. A. Inorg. Chem.
2008, 47, 1252.
(3) Schoeller, W. W.; Niecke, E. J. Chem. Soc., Chem. Commun. 1982, 569.
r
2010 American Chemical Society
Published on Web 11/17/2010
pubs.acs.org/IC

11596 Inorganic Chemistry, Vol. 49, No. 24, 2010
Alonso et al.
Scheme 1
Scheme 2
the way to explore at a great extent the reactivity of the metal-
bound oxophosphinidene ligand,5a-c and we then decided to
study in detail the chemical behavior of this anion. In the first
full report of this wide study, we analyzed the reactivity of the
above molybdenum complex and that of its tungsten ana-
logue (H-DBU)[WCp{P(O)R*}(CO)2] (10) toward several
protic acids and C-based electrophiles (Scheme 1).6
Chart 1
Thus we found that these anionic complexes were easily
protonated at the O atom of the oxophosphinidene ligand to
give the hydroxyphosphide derivatives [MCp{P(OH)R*}(CO)2],
and that the molybdenum anion reacted analogously with
strong C-based electrophiles to give the corresponding alkoxy-
phosphide derivatives [MoCp{P(OR)R*}(CO)2]. In contrast,
the reactions with milder alkylating reagents gave selectively
the corresponding κ2-phosphinite complexes [MoCp{κ2-OP(R)-
R*}(CO)2], as a result of the attack of the electrophile at the
P atom of the oxophosphinidene ligand. According to density
functional theory (DFT) calculations on the Mo anion, the
oxygen atom of the phosphinidene ligand bears the highest
negative charge of the molecule, while the highest occupied
molecular orbital (HOMO) of this complex has substantial
Mo-P π bonding character. Thus, it was concluded that the
phosphinite complexes are formed under conditions of orbit-
al control, while charge-controlled reactions tend to give
alkoxyphosphide derivatives.6
In this paper we analyze the reactivity of the anions1 and 10
toward several electrophilic reagents based on elements dif-
ferent from carbon, either the heavier group 14 elements
(Si to Pb) or some transition and post-transition metals
(Ti, Zr, Au, Ag). As it will be discussed, these reactions
may involve the binding of the electrophilic fragment to
either the O atom of the oxophosphinidene ligand or the
metal atom, thus leading to κ1,κ1-P,O-bridged or terminal
oxophosphinidene complexes, respectively. While the former
reactions are reminiscent of the formation of alkoxyphos-
phide complexes mentioned above, no derivatives involving
the attachment of the electrophile to the P atom of the
oxophosphinidene ligand have been observed in the reactions
now reported.
or room temperature with different organohalide compounds
ER4-xClx (x = 1, 2) of the group 14 elements to give deriv-
atives with terminal or bridging oxophosphinidene ligands
depending on the site of attachment (M or O) of the
electrophilic fragment (Scheme 2).
Thus the reactions with SiR3Cl (R = Ph, Me) give the
corresponding silyloxyphosphide derivatives [MCp-
{P(OSiR3)R*}(CO)2] (M = Mo, R = Me (2), Ph (3);
M = W, R= Me (20)). In a similar way, compound 1 reacts
with GePh3Br or GeMe2Cl2 at 273 K to give the correspond-
ing derivatives [MoCp{P(OE)R*}(CO)2] (E = GePh3 (4),
GeMe2Cl (5)), the process being faster for GeMe2Cl2.
These reactions are reminiscent of the formation of alkoxy-
phosphide complexes in the reactions of 1 with strong
C-based electrophiles,6 and they might analogously be
interpreted as favored on electrostatic grounds (charge
control).
In contrast, the reactions of 1 with SnPh3Cl, SnPh2Cl2,
and PbPh3Cl, that is, with electrophiles based on the
softer members of the group 14 elements, give instant-
aneously the metal-metal bonded complexes [MoCp-
{P(O)R*}(CO)2(SnPh3)] (6), [MoCp{P(O)R*}(CO)2-
(SnPh2Cl)] (7), and [MoCp{P(O)R*}(CO)2(PbPh3)] (8),
respectively, all of them retaining a terminal oxophos-
phinidene ligand. In the same line, the tungsten compound 10
reacts with SnPh2Cl2 and PbPh3Cl to give the correspond-
ing heterometallic derivatives [WCp{P(O)R*}(CO)2-
(SnPh2Cl)] (70) and [WCp{P(O)R*}(CO)2(PbPh3)] (80),
respectively (Scheme 2). The synthesis of the molybdenum-tin
compounds 6 and 7 was described in our preliminary reports
on the chemistry of the oxophosphinidene ligand,5a,c with the
latter complex displaying a remarkablehydrolyticbehavior.5c
Moreover, we note here that while the SnPh3 and PbPh3
Results and Discussion
Reactions of Compounds 1 and 10 with ER4-xClx (E = Si
to Pb). The anionic complexes 1and 10 react rapidly at 273 K
ꢀ
´
(6) Alonso, M.; Alvarez, M. A.; Garcıa, M. E.; Garcia-Vivo, D.; Ruiz,
M. A. Inorg. Chem. 2010, 49, 8962.

Article
Inorganic Chemistry, Vol. 49, No. 24, 2010 11597
δP/ppmb [JPW], (J119SnP), {J117SnP}/Hz
Table 1. IR and 31P{1H}NMR Data for Compounds 2 to 10
compound
νst(CO)a/cm-1
[MoCp{P(OSiMe3)R*}(CO)2] (2)
[WCp{P(OSiMe3)R*}(CO)2] (20)
[MoCp{P(OSiPh3)R*}(CO)2] (3)
[MoCp{P(OGePh3)R*}(CO)2] (4)
[MoCp{P(OGeMe2Cl)R*}(CO)2] (5)
[MoCp{P(O)R*}(CO)2(SnPh3)] (6)
[MoCp{P(O)R*}(CO)2(SnPh2Cl)] (7)
1934 (vs), 1854 (s)
1924 (vs), 1842 (s)
1940 (vs), 1858 (s)
1930 (vs), 1850 (s)
319.3c
272.4 [735]
306.9c
330.6
1935 (vs), 1854 (s)
1947 (m), 1889 (vs)
333.8
468.1(116){116}c
trans: 475.3 (149){149}
cis: 438.5 (540){516}
trans: 420.2 [372](147){147}
cis: 376.0 [422](548){530}
467.5 (JPPb = 177)
412.4 [393], (JPPb = 168)
350.7e
2001(m), 1960 (s), 1899 (vs)a 1999 (vs),
1960 (s), 1906 (s)d
[WCp{P(O)R*}(CO)2(SnPh2Cl)] (70)
1994 (m), 1952 (s), 1891 (vs) a 1993 (vs),
1950 (s), 1901 (m)d
1949 (m), 1893 (vs)
[MoCp{P(O)R*}(CO)2(PbPh3)] (8)
[WCp{P(O)R*}(CO)2(PbPh3)] (80)
[MoCp{P(OSnMe3)R*}(CO)2] (9)
[MoCp{P(O)R*}(CO)2(SnMe3)] (10)
1944 (m), 1885 (vs)
1913 (vs), 1831 (s)
458.0 f
a Recorded in dichloromethane solution, unless otherwise stated. b Recorded in CD2Cl2 solution at 290 K and 121.50 MHz unless otherwise stated;
δ relative to external 85% aqueous H3PO4. c Recorded at 81.03 MHz. d Recorded in petroleum ether solution. e Recorded in C6D6 solution at 290 K and
81.03 MHz. f Recorded in CD2Cl2 solution at 203K and 162.14 MHz.
products are obtained as single (trans) isomers, the SnPh2Cl
products are obtained as an equilibrium mixture of cis and
trans isomers (Chart 1), a matter to be discussed later on.
The subtle effects of the substituents present in the
added electrophile can even counterbalance the formation
of one or another type of products in the above reactions,
as suggested by the nature the SnMe3 derivative of com-
pound 1, that at room temperature exists mainly in its
oxophosphinidene-bridged form [MoCp{P(OSnMe3)R*}-
(CO)2] (9), but in its terminal form [MoCp{P(O)R*}-
(CO)2(SnMe3)] (10) below 203 K. Although we have not
observed this sort of equilibrium for other complexes in
this work, we cannot exclude that similar processes may
take place to a very small extent in other cases, and this
leads us to think that the type of structure actually
displayed by compounds 2 to 10 is not kinetically, but
thermodynamically governed.
Structural Characterization of Compounds 2-5 and 9.
Spectroscopic data in solution for compounds 2-5, 20,
Figure 1. ORTEP drawing (30% probability) of the molecular structure
of compound 8, with all H atoms and Me groups omitted for clarity.
and 9 are comparable to each other (Table 1 and Experi-
mental Section) and to those of the structurally charac-
terized complexes [MoCp(PFR*)(CO)2] and [MoCp-
{P(OC(O)Ph)R*}(CO)2],6 supporting the presence of an
oxophosphinidene ligand bridging the Mo/W and Si/Ge/
Sn stoms in these molecules, and keeping a planar tri-
gonal environment around the P atom. The latter atom then
formally behaves as a 3-electron donor to the Mo/W atom
(alkoxyphosphide-like coordination). Moreover, these
data are also similar to those reported for comparable
phosphide complexes having a supermesityl substituent,
such as [MCp(PHR*)(CO)2] and [MCp(PClR*)(CO)2]
(M= Mo, W),7 and to those for the hydroxyphosphide
derivatives of 1 and 10, [MoCp{P(OH)R*}(CO)2] and
[WCp{P(OH)R*}(CO)2],6 indicating that all those mole-
cules share the same basic structure. The P-donor ligand
in these compounds gives rise to quite deshielded 31P
resonances, in the range 270-350 ppm, a feature ex-
plained by the combination of two effects of known
deshielding influence on P-donor ligands, the presence
of oxygen bound to phosphorus, and the multiple nature
of the metal-phosphorus bond,8 with the latter leading
to a diminished HOMO-LUMO gap and hence to an
increased paramagnetic contribution to the phosphorus
shielding. This also justifies the fact that our complexes
display 31P chemical shifts substantially higher than those
of the mentioned complexes [MoCp(PXR*)(CO)2] (X =
H (266.2 ppm), Cl (259.1 ppm)),7 and within the range
found for the alkoxyphosphide complexes [MoCp{P(OR)R*}-
(CO)2] (290-350 ppm for R = Me, Et, COC2H3, COPh).6
In addition, the 31P NMR spectrum of the tungsten com-
plex 20 exhibits a quite large 31P-183W coupling of 735 Hz,
comparable to that measured in the hydroxyphosphide
complex [WCp{P(OH)R*}(CO)2] (JPW=704 Hz),6 with
these values being consistent with the formulation of
double W-P bonds for these molecules.9
Solid-State Structure of Compound 8. The molecule
of the dicarbonyl complex 8 in the crystal (Figure 1 and
Table 2) displays a distorted four-legged piano stool
(9) (a) Lambert, J. A.; Mazzola, E. P. Nuclear Magnetic Resonance. An
Introduction to Principles and Experimental Methods; Pearson: London, 2003.
(b) Jameson, C. J. In Phosphorus-31 NMR Spectroscopy in Stereochemical
Analysis; Verkade, J. G., Quin, L. D., Eds.; VCH: Deerfield Beach, FL, 1987;
Chapter 6.
€
(7) Malisch, W.; Hirth, U.-A.; Grun, K.; Schmeuber, M. J. Organomet.
Chem. 1999, 572, 207.
(8) Carty, A. J.; MacLaughlin, S. A.; Nucciarone, D. In Phosphorus-31
NMR Spectroscopy in Stereochemical Analysis; Verkade, J. G., Quin, L. D.,
Eds.; VCH: Deerfield Beach, FL, 1987; Chapter 16.

11598 Inorganic Chemistry, Vol. 49, No. 24, 2010
Alonso et al.
Table 2. Selected Bond Lengths (A) and Angles (deg) in Compound 8
perpendicularly to it, as found in the precursor 1. We
should remark that the aryl group is placed specifically cis
to the Cp ligand of the complex. This conformation
around the multiple Mo-P bond is also reproduced in
all derivatives of 1 as noticed previously,6 and is further
confirmed by the present work, it likely being the most
favored one on steric grounds. Concerning the geometry
around the metal atom, we note that the P-Mo-Pb angle
(132.6(1)°) is significantly wider that the C(1)-Mo-C(2)
one (102.0(2)°). This might be in part justified on the basis
of the larger steric requirements of the PbPh3 and P(O)R*
groups, that would take them as far apart from each other
as possible. Additionally, this distortion also follows the
established pattern of angular trans influence,12 which
places the pair of ligands with overall stronger σ-bonding
ability at smaller L-M-Cp(centroid) angles, to maxi-
mize the interaction with the more favorable metal
orbitals.13 In the case of 8, however, this distortion is
essentially limited to the PbPh3 group (Pb-Mo-
Cp(centr.) = 107.2°), while the oxophosphinidene ligand
defines a rather normal P-Mo-Cp(cent.) angle of
120.4°, perhaps because of steric limitations. On the other
hand, the Mo-Pb length of 2.8845(4) A in 8 is only
slightly below the value found in the structurally related
tricarbonyl complex [Mo(PbPh3)Cp(CO)3] (Mo-Pb =
2.90 A).14 This indicates that the trans influence15 of
the oxophosphinidene ligand is comparable to that
(strong one) of the carbonyl ligand. We finally note that,
concerning the environment around the lead atom, the
C-Pb-C bond angles (ca. 107°) are smaller than the
Mo-Pb-C ones (ca. 112°), as expected both on steric
grounds and also from the higher electronegativity of the
Ph groups (compared to the Mo fragment), as predicted
by the Bent model.16
Solution Structure of Compounds 6-8 and 10. The IR
spectra of compounds 6, 8, and 80 (Table 1) display in each
case two C-O stretching bands, with relative intensities
characteristic of trans-dicarbonyl complexes, in agree-
ment with the structure of 8 in the crystal. These struc-
tures are therefore comparable to those of the related
phosphine complexes [MoCp(SnR3)(CO)2L] (R = Me,
Ph; L=PPh3, P(OPh)3, SbPh3),17 and to those of the
carbene complexes [MCp(M0Ph3)(CO)2{C(OR0)R}] (M=
Mo, W; M0 = Ge, Sn).18 All these complexes were found
to exist in solution exclusively as the corresponding trans
isomers, although the carbene complexes displayed two
conformers related by a restricted rotation of the carbene
ligand around the corresponding M-C bond. In contrast,
the SnPh2Cl compounds 7 and 70 display three C-O
stretching bands, indicative of the presence of both cis
and trans isomers, with their ratio being dependent on the
polarity of the solvent. For instance, the IR spectrum of 7
in CH2Cl2 solution displays bands at 2001 (m), 1960 (s),
Mo(1)-P(3)
2.319(1)
2.8845(4)
1.482(3)
102.0(2)
113.0(2)
130.1(2)
116.8(1)
132.6(1)
176.2(4)
175.5(4)
Mo(1)-C(1)
1.983(5)
1.987(5)
1.838(4)
106.1(1)
118.1(1)
111.1(1)
108.6(2)
103.9(2)
108.1(2)
Mo(1)-Pb(1)
Mo(1)-C(2)
P(3)-O(3)
P(3)-C(31)
C(1)-Mo(1)-C(2)
O(3)-P(3)-C(31)
O(3)-P(3)-Mo(1)
C(31)-P(3)-Mo(1)
P(3)-Mo(1)-Pb(1)
O(1)-C(1)-Mo(1)
O(2)-C(2)-Mo(1)
C(111)-Pb(1)-Mo(1)
C(121)-Pb(1)-Mo(1)
C(101)-Pb(1)-Mo(1)
C(111)-Pb(1)-C(121)
C(111)-Pb(1)-C(101)
C(121)-Pb(1)-C(101)
Chart 2
structure, with the PbPh3 and oxophosphinidene groups
in trans arrangement. The Mo-P length of 2.319 (1) A is
substantially shorter than the values of about 2.45 A
characteristic of single Mo-P bonds (e.g., those in Mo(II)
phosphine complexes), but about 0.08 A longer than
the one measured in the anion 1 (2.238(1) A), and sub-
stantially longer than other reference figures for double
Mo-P bonds, such as the values of about 2.20 A for the
complexes [MoCp{P(X)R*}(CO)2] (X = F, C(O)Ph),6 or
the very short Mo-P length of 2.1439(5) A measured in
the diphosphenide complex [Mo(N[tBu]Ar)3(P = PSiiPr3)]
(Ar = 3,5-Me2C6H3).10 At the same time, the P-O length
in 8 (1.482(7) A) is slightly shorter than that measured in
its precursor 1 (1.514(3) A), and is virtually identical to
that measured for the double P-O bond in the terminal
PO complex [Mo(NtBuAr)3(PO)] (1.49(2) A).4d Thus we
conclude that the metal-oxophosphinidene bonding in
compound 8 is best described as an average of the extreme
distributions represented by the canonical forms A and B
in Chart 2. This is in contrast with the anion in compound
1, where the Mo-P and P-O bonds can be described
as essentially double bonds, and therefore might be
adequately represented by just the canonical structure A.
This difference can be understood by considering that on
going from 1 to 8 there is a strong reduction of the
electron density at the metal atom therefore greatly
suppressing the back-donation (Mo to P) responsible
for the double-bond character of the Mo-P interaction.
Of course a similar situation is to be expected for the
related tin complexes 7 and 10. Interestingly, we note that
the above difference concerning the nature of the Mo-P
bond when going from the anion 1 to their neutral deriv-
atives 6-8 or 10 is completely analogous to that existing
between nucleophilic (PdM bonding) and electrophilic
(PfM bonding) phosphinidene complexes.11
In agreement with the multiplicity of the Mo-P and
P-O bonds in 8, the coordination environment around
phosphorus is planar trigonal (sum of angles = 359.9°),
with the Mo, P, O3, and C31 atoms contained in a plane
almost bisecting the molecule, and the aryl group orientated
(12) Poli, R. Organometallics 1990, 9, 1892.
(13) Lin, Z.; Hall, M. B. Organometallics 1993, 12, 19.
(14) Struchkov, Y. T.; Anisimov, K. N.; Osipova, O. P.; Kolobova, N. E.;
Nesmeyanov, A. N. Dokl. Akad. Nauk. SSSR 1967, 172, 107.
(15) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry, 5th ed.;
Wiley: New York, 1988; p 1299.
(10) Piro, N. A.; Cummins, C. C. J. Am. Chem. Soc. 2009, 131, 8764.
(11) (a) Aktas, H.; Slootweg, J. C.; Lammerstma, K. Angew. Chem., Int.
Ed. 2010, 49, 2. (b) Lammertsma, K. Top. Curr. Chem. 2003, 229, 95.
(c) Lammertsma, K.; Vlaar, M. J. M. Eur. J. Org. Chem. 2002, 1127.
(d) Mathey, F.; Tran-Huy, N. H.; Marinetti, A. Helv. Chim. Acta 2001, 84, 2938.
(16) Bent, H. A. Chem. Rev. 1961, 61, 275.
(17) (a) Manning, A. R. J. Chem. Soc., A 1968, 651. (b) George, T. A. Inorg.
Chem. 1972, 11, 77.
(18) (a) Dean, W. K.; Graham, W. A. G. Inorg. Chem. 1977, 16, 1061.
(b) Chan, L. Y. Y.; Dean, W. K.; Graham, W. A. G. Inorg. Chem. 1977, 16, 1067.

Article
Inorganic Chemistry, Vol. 49, No. 24, 2010 11599
and 1899 (vs) cm-1, and their relative intensities are
strongly modified when recorded in petroleum ether
solution, these now appearing at 1999 (vs), 1960 (s), and
1906 (s) cm-1. This indicates an increase of the relative
proportion of the cis isomer in the latter, less polar
solvent, and similar changes were observed in the IR
spectra of the tungsten compound 70 (Table 1).
by recalling that the strong reduction of the electron
density at the metal atom operated upon formation of
these neutral compounds from the anions 1 and 10 should
greatly suppress the back-donation of electron density
from the metal atom to P (π interaction), as noted above,
and therefore should at least greatly reduce the diamag-
netic contribution to the shielding of this nucleus.8 On
the other hand, the cis isomers in compounds 7 and 70
give rise to 31P resonances some 40 ppm below those of
the corresponding trans isomers, a difference difficult to
be justified on simple grounds.
Equilibria between cis and trans isomers is a common
feature of complexes having a four-legged piano stool
structure, and the different factors involved in this iso-
merism have been extensively studied, both in solution
and in the solid state.19,20 For instance, the cis/trans
equilibrium ratio in solution for the complexes [MoCpX-
(CO)2L] (L = P(OMe)3, PPh3), increases in the order X =
I<Br<Cl.20 The presence of bulky groups is expected
to disfavor somewhat the cis isomers for dicarbonyl
complexes; for instance, the complex [MoCpMe(CO)2-
{PN(MeCH2CH2NMe(OMe)}] displays cis and trans
isomers in solution, but the related MoSn complex
[MoCp(CO)2{PN(MeCH2CH2NMe(OMe)}(SnR3)] only
displays the trans isomer.21 Although the steric effects do
not seem to rule the cis/trans isomerism in these com-
plexes,22 the smaller size of the SnPh2Cl group (compared
to that of the SnPh3 or PbPh3 groups) might be the
relevant factor behind the appearance of the cis isomer
just for compounds 7 and 70. This would be so in our
compounds because, as we have discussed above, the
trans influence of the oxophosphinidene ligand might
be comparable to that of the carbonyl ligands, and there-
fore the electronic effects might have only a modest
influence on the cis/trans isomerism in our complexes.
As expected, the cis/trans equilibrium for compounds 7
and 70 is also solvent- and temperature-dependent. Thus,
in the case of 7 the cis isomer is moderately favored by
lowering the temperature, with the cis/trans ratio being
increased from 2/7 at 293 K to 2/5 at 193 K. The solvent
has a much stronger influence on this equilibrium and, in
line with the qualitative trend deduced from the IR data
mentioned above, the cis isomer is favored for the less
polar solvent, with the cis/trans ratio being increased to
about 3/2 in C6D6 solution.
The inspection of the metal-phosphorus couplings
apparent in the 31P resonances of these compounds also
is highly informative. Thus, the tin compounds having a
trans geometry display relatively small two-bond 31P-Sn
couplings (ca. 115 Hz for 6 or 150 Hz for 7 and 70),
whereas the cis isomers of compounds 7 and 70 display
much higher values (ca. 550 Hz with the 119Sn nucleus).
This difference is consistent with the general trend in the
2
absolute values (2Jcis>2Jtrans) found for JXY in com-
plexes of the type [MCp(CO)2XY].9b We also note that
the tungsten compounds 70 and 80 display P-W couplings
of about 400 Hz, almost half the value of 735 Hz
measured in the silyloxyphosphide complex 20, but still
higher than the value of 294 Hz measured for the closely
related phosphine complex trans-[WCp(CO)2(PPh3)-
(SnMe2Cl)].23 Theses differences are consistent with the
substantial (but not complete) reduction in the multi-
plicity of the M-P bond operated upon formation of our
neutral oxophosphinidene complexes, a matter discussed
above on the basis of the structural data for the lead
complex 8.
Interconversion Between Isomers 9 and 10. The reaction
of 1 with SnMe3Cl in tetrahydrofuran (THF) solution is
not complete but reaches to an equilibrium point (eq 1)
that can be shifted to the product side by removal of the
solvent and extraction of the crude reaction mixture with
petroleum ether.
1 þ SnMe3Cl h 9 þ ðH-DBUÞCl
ð1Þ
Spectroscopic data recorded at room temperature for
the neutral product obtained in this way (Table 1 and
Experimental Section) indicate that it exists mainly in its
oxophosphinidene-bridged form [MoCp{P(OSnMe3)R*}-
(CO)2] (9), these being comparable to those of the GePh3
product 4. However, the considerable broadness of the
The oxophosphinidene ligand in compounds 6 to 8 and
10 gives rise to a strongly deshielded resonance in the
ranges 460-475 ppm (Mo) or 415-420 ppm (W) for the
trans isomers, that is, some 80 ppm downfield from the
corresponding resonance in the anionic precursors 1
(385.0 ppm) and 10 (335.6 ppm).6 These shifts are also
substantially higher than those measured for the com-
plexes [Cr(CO)5{P(O)N(iPr)2}] (δP = 319.2 ppm),4a [ReCl-
1
corresponding H and 31P NMR resonances at room
temperature suggested the occurrence of dynamic pro-
cesses in solution. Indeed, upon cooling a CD2Cl2 solu-
tion of this compound, the broad 31P resonance observed
at about 370 ppm at room temperature further broadens
and eventually splits, so that at 243 K two broad reso-
nances at 359.3 and 458.0 ppm are clearly defined, with
relative intensities about 2:1, which we assign respectively
to the isomers 9 and 10 on the basis of their spectroscopic
properties and analogies with the complexes 2 to 8.
Further cooling of the solution increases the relative
amount of the more deshielded resonance, so that below
203 K only the resonance due to the isomer 10 is present in
t
(dppm)2{P(O)CH2 Bu}] (340.1 ppm)4b and [Fe2Cp2-
(μ-CO)2(CO){P(O)R*}] (432.5 ppm),4c these being the
only other terminal oxophosphinidene complexes pre-
viously reported. The strong deshielding effect on the
P nucleus (relative to the parent anions) might be justified
(19) Bala, M. D.; Levendis, D. C.; Coville, N. J. J. Organomet. Chem.
2006, 691, 1919, and references therein.
(20) Faller, J. W.; Anderson, A. S. J. Am. Chem. Soc. 1970, 5852.
(21) (a) Nakazawa, H.; Kishishita, M.; Ishiyama, T.; Mizuta, T.; Miyoshi,
K. J. Organomet. Chem. 2001, 617, 453. (b) Nakazawa, H.; Kishishita, M.;
Yoshinaga, S.; Yamaguchi, Y.; Mizuta, T.; Miyoshi, K. J. Organomet. Chem.
1997, 529, 423.
€
€
(22) Smith, J. M.; Coville, N. J.; Cook, L. M.; Boeyens, J. C. A.
Organometallics 2000, 19, 5273.
(23) Braunschweig, H.; Bera, H.; Geibel, B.; Dorfler, R.; Gotz, D.; Seeler,
F.; Kupfer, T.; Radacki, K. Eur. J. Inorg. Chem. 2007, 3416.

11600 Inorganic Chemistry, Vol. 49, No. 24, 2010
Alonso et al.
Scheme 3. Dissociative Isomerization Proposed for Compound 9
in Solution
Scheme 4
Figure 2. ORTEP drawing (30% probability) of the molecular structure
of compound 120, with all H atoms and Me groups omitted for clarity.
Table 3. Selected Bond Lengths (A) and Angles (deg) in Compound 120
W(1)-C(1)
1.947(7)
1.954(7)
2.233(1)
84.7(3)
90.3(2)
89.3(2)
177.7(2)
Zr(1)-O(3)
2.016(4)
1.563(4)
1.851(5)
100.1(2)
127.4 (2)
132.4(2)
W(1)-C(2)
P(1)-O(3)
W(1)-P(1)
P(1)-C(18)
C(1)-W(1)-C(2)
C(1)-W(1)-P(1)
C(2)-W(1)-P(1)
P(1)-O(3)-Zr(1)
O(3)-P(1)-C(18)
O(3)-P(1)-W(1)
C(18)-P(1)-W(1)
[MoCp{P(OZrCp2Cl)R*}(CO)2] (12), and [WCp-
{P(OZrCp2Cl)R*}(CO)2] (120). In contrast, the anions 1
and 10 react with [AuCl{P(p-tol)3}] to give metal-metal
bonded compounds. In this case, the reaction gives the
corresponding oxophosphinidene complexes [MCp{P(O)R*}-
(CO)2{AuP(p-tol)3}] [M = Mo (13), W(130)], but an equilib-
rium is reached when using THF as solvent, and these
products actually are more conveniently prepared by
reacting the anions 1 and 10 with stoichiometric amounts
of the tetrahydrothiophene complex [Au{P(p-tol)3}-
(THT)][PF6] in the same solvent. Attempts to prepare a
related MoAg species using [AgCl(PPh3)] were unsuccess-
ful, but the reaction of 1 with the perchlorate complex
[Ag(ClO4)(PPh3)] gave a product tentatively formulated
as the corresponding oxophosphinidene derivative [MoCp-
{P(O)R*}(CO)2{(Ag(PPh3)}] on the basis of its spectro-
scopic similarities (ν(CO) 1880 (m), 1815 (vs) cm-1, δP =
428 ppm) with the gold complex 13. However, attempts to
isolate this unstable product as a reasonably pure solid
were unsuccessful.
We should remark that the heterometallic compounds
11 and 12 are (along with 9) almost the first complexes
reported to have an oxophosphinidene ligand bridging
two metal atoms in a μ2-P,O fashion, the only precedent
being found in the MoZr complex [Mo{NR(Ar)}3-
{P(OM)Me}] (M = ZrCp2Me).4d Since a [M0Cp2Cl]þ (M0 =
Ti, Zr) fragment can be considered as a hard electrophile
because of the high oxidation state of the metal atom,
then it is not surprising that such a fragment ended up
bound to the O atom of the oxophosphinidene ligand,
because this is the expected result under conditions of
charge control, as discussed above. In contrast, the
[Au{P(p-tol)3}]þ fragment can be considered as a much
the 31P NMR spectrum. From these data we conclude
that isomers 9 and 10 interconvert rapidly in solution, and
that at room temperature there is only a tiny amount of
isomer 10 in equilibrium with the major, phosphide-like
isomer 9. The equilibrium observed in the synthesis of
this compound in THF solutions (eq 1) suggests that the
interconversion between isomers 9 and 10 in dichloro-
methane solutions also might be dissociative in nature,
that is, with the intermediacy of a small amount of the
anion 1 (Scheme 3). In agreement with this proposal, a
separated experiment showed that compound 9 reacted
with SnPh3Cl in dichloromethane solution at room tem-
perature to quantitatively give the SnPh3 compound 6.
Reactions of Compounds 1 and 10 with Halide Metal
Complexes. The anionic complexes 1 and 10 react readily
with different halide complexes of the transition and post-
transition elements such as the group 4 metallocene
chlorides [MCp2Cl2], and gold(I) complexes of the type
[AuCl(PR3)]. Other halocomplexes having more covalent
M-X bonds such as the carbonylic complexes [MoClCp-
(CO)3], [MBr(CO)5] (M = Mn, Re), and [FeCpCl(CO)2]
failed to react with the molybdenum compound 1 in
dichloromethane or THF solutions. Once again, the
fragments arising from the halide metal complexes can
bind either the oxygen atom of the oxophosphinidene
ligand or the metal atom of the anion, depending on the
reagent being used (Scheme 4).
Thus the reactions with stoichiometric amounts of [M0Cp2Cl2]
(M0 =Ti, Zr) give the corresponding oxophosphinidene-
bridged complexes [MoCp{P(OTiCp2Cl)R*}(CO)2] (11),

Article
Inorganic Chemistry, Vol. 49, No. 24, 2010 11601
δP/ppmb [J/Hz]
Table 4. IR and 31P{1H}NMR Data for Compounds 11-13
compound
νst(CO)a/cm-1
[MoCp{P(OTiCp2Cl)R*}(CO)2] (11)
[MoCp{P(OZrCp2Cl)R*}(CO)2] (12)
[WCp{P(OZrCp2Cl)R*}(CO)2] (120)
[MoCp{P(O)R*}(CO)2{AuP(p-tol)3}] (13)
[WCp{P(O)R*}(CO)2{AuP(p-tol)3}] (130)
1915 (vs), 1839 (s)
1921 (vs), 1841 (s)
1913 (vs), 1830 (s)
1896 (m), 1840 (vs)
1897 (m), 1835 (vs)
361.6
350.3
303.7 [JPW= 724]
443.6 (s, br), 48.4 [d, JPP= 8, AuP] c
393.2 [s, br, JPW = 530], 54.9 [d, JPP = 10, AuP]
a Recorded in dichloromethane solution, unless otherwise stated. b Recorded in CD2Cl2 solution at 290 K and 121.50 MHz unless otherwise stated;
δ relative to external 85% aqueous H3PO4. c Recorded at 81.03 MHz.
Table 5. Selected Bond Lengths (A) and Angles (deg) in Compound 13
Mo(1)-P(1)
2.289(2)
2.7071(7)
2.274(2)
1.861(8)
99.4(3)
82.3(2)
80.7(2)
63.2(2)
67.0(2)
126.0(1)
Mo(1)-C(1)
Mo(1)-C(2)
P(1)-O(3)
1.985(8)
1.974(8)
1.498(5)
Mo(1)-Au(1)
Au(1)-P(2)
P(1)-C(29)
C(2)-Mo(1)-C(1)
C(2)-Mo(1)-P(1)
C(1)-Mo(1)-P(1)
C(1)-Mo(1)-Au(1)
C(2)-Mo(1)-Au(1)
P(1)-Mo(1)-Au(1)
O(3)-P(1)-C(29)
O(3)-P(1)-Mo(1)
C(29)-P(1)-Mo(1)
O(1)-C(1)-Mo(1)
O(2)-C(2)-Mo(1)
114.5(3)
130.4(2)
115.1(2)
173.5(7)
175.4(7)
Spectroscopic data in solution for compounds 11, 12,
and 120 (Table 4) are similar to each other and consistent
with the structure found for 120 in the crystal. Moreover
they are also comparable to those of the silicon and
germanium compounds 2 to 5 (Table 2). Thus, their IR
spectra also display two C-O stretching bands of strong
intensity, although the frequencies are a bit lower than
those of the Si and Ge compounds, an effect that can be
attributed to the lower electron-withdrawing power of the
group 4 elements. In apparent relationship with this, the
31P resonances of these complexes are less shielded than
those of the Si and Ge compounds, they appearing midway
between the latter resonances and those of the starting
anions 1 and 10. Yet, the P-W coupling in 120 (724 Hz)
has a magnitude comparable to the value measured for
the silicon compound 20 (735 Hz), significantly higher
than that in the anion 10 (658 Hz), and in any case
consistent with the double M-P bonds that should be
formulated for all these complexes, also substantiated by
the short P-W length found in the crystal for 120.
Figure 3. ORTEP drawing (30% probability) of the molecular structure
of compound 13, with all H atoms and Me groups (except those of the
p-tol rings) omitted for clarity.
softer acid, and therefore should interact with the anions
1 and 10 under the driving force of the more favorable
orbital interactions. Hence, a metal-metal bonded prod-
uct is more likely to be formed, as observed in the reac-
tions with the tin and lead organohalides.
Structural Characterization of Compounds 11 and 12.
The molecule of the WZr complex 120 in the crystal
(Figure 2 and Table 3) is made up from WCp(CO)2 and
ZrCp2Cl fragments bridged by an oxophosphinidene
ligand through its P and O atoms, respectively. The
coordination geometry around tungsten is of the classical
tree-legged piano stool type, actually very similar to that
of the fluoro- and alkoxyphosphide derivatives of 1
previously reported by us ([MoCp(PXR*)(CO)2], with
X = F, C(O)Ph).6 In particular, the short W-P length of
2.233(1) A and the planar geometry around phosphorus
(sum of angles = 359.9°) is consistent with the formula-
tion of a double W-P bond for this phosphide-like
complex (cf. 2.239(1) A for 1 or about 2.20 A for the
mentioned phosphide complexes). The P-O length of
1.563(4) A is longer than those in the anion 1 (1.514(3) A)
or in the MoPb compound 8 (1.482 (3) A) as expected, but
yet shorter than the single-bond P-O length measured in
the mentioned alkoxyphosphide complex (1.661(4) A),6
and even that in the related MoZr complex [Mo{NR(Ar)}3-
{P(OZrCp2Me)Me}] (1.613(5) A),4d all of this suggesting
the presence of some residual multiplicity in this formally
single P-O bond of compound 120. Finally, the Zr-O
length of 2.016(4) A must be viewed as a normal figure for
a single bond between these atoms (cf. 2.000(4) A in the
mentioned MoZr complex).
Structural Characterization of Compounds 13 and 130.
The molecular structure of the molybdenum compound
13 in the crystal (Figure 3 and Table 5) displays a distorted
four-legged piano stool structure very similar to that of
the lead compound 8, if we just replace the PbPh3 group
by a AuP(p-tol)3 one, with the latter conforming to a
linear coordination geometry around the gold atom. The
geometrical distortions around the Mo atom in 13 are
very similar to those observed for 8 and need not to be
discussed in detail, being reflected in comparable values
for the angles defined by the ligands and the Cp centroid
(Au-Mo-Cp(centr.) = 110.1°, P-Mo-Cp(centr.) =
123.8°), and the angles between ligands (C-Mo-C =
99.4(3)°, Au-Mo-P = 126.0(1)°).
The Mo-P (2.289(2) A) and the P-O (1.498(5) A)
lengths in 13 are respectively slightly shorter and longer
than the corresponding figures in the lead compound 8
(2.319 (1) A and 1.482(7) A). This suggests the retention of
a slightly higher multiplicity in the Mo-P bond for the gold
compound, that might be attributed to the comparatively
smaller reduction of electron density that occurred at the
Mo site (and hence less reduction in the back-donation to

11602 Inorganic Chemistry, Vol. 49, No. 24, 2010
Alonso et al.
phosphorus) upon addition of the gold fragment. This is
in agreement with the C-O stretching frequencies of the
gold compounds, some 50 cm-1 lower than those of the
corresponding lead compounds (see below).
Scheme 5
The Mo-Au length of 2.7071(7) A in 13 is only margin-
ally lower than the corresponding values in the related tri-
carbonyl complexes [MoCp(CO)3{Au(PPh3)}] (2.710(1) A)
and [Mo(η5-C5H4CHO)(CO)3{Au(PPh3)}] (2.7121(5) A).24
From this we conclude once again that the trans influence
of the oxophosphinidene ligand is of comparable magni-
tude to that of a carbonyl ligand. We finally notice the
relatively close approach of the carbonyl ligands to the
gold atom in 13 (Au C separations of 2.533(8) and
2.654(8) A), a feature relatively common in mono and
polynuclear carbonyl complexes of the transition metals
3 3 3
having AuPR3 fragments (e.g., Au C ca. 2.49 A in
3 3 3
[Mo(BnNiPr2)(CO)3{Au(PPh3)}], Bn=η5-C10H9B).25
We trust there are no genuine bonding interactions in
the case of 13, not just because of the linearity of these
carbonyl ligands but also because almost identical posi-
tions of the carbonyl ligands are observed for the lead
compound 8, although the larger intermetallic separation
in that case (ca. 2.9 A) also implies larger and clearly
bound to the oxygen atom of the oxophosphinidene ligand
to give oxophosphinidene-bridged derivatives of the type
[MCp{P(OE)R*}(CO)2], thus resembling the reactions with
strong C-based electrophiles to give alkoxyphosphide deriv-
atives, as expected for interactions proceeding under con-
ditions of charge control (Scheme 5). Reagents based on
softer and milder electron-withdrawing centers [Sn(IV),
Pb(IV), Au(I), Ag(I)] give instead metal-metal bonded
products of the type [MCp{P(O)R*}(CO)2(E)] as a result
of the attachment of the incoming electrophile to the metal
site of the anion. This is in marked contrast with the reactivity
of 1 toward mild C-based electrophiles, the latter resulting in
the attachment of the incoming electrophile to the P atom of
the oxophosphinidene ligand to yield a phosphinite derivate,
as noted above (Scheme 1). These apparently divergent
results might be considered as different evolutions of a
common type of intermediate species by recalling that,
according to the DFT calculations on 1,6 the frontier orbitals
in these anions have M-P π bonding character. Therefore,
under conditions of orbital control (mild and soft electro-
philic reagents) we should expect that a generic electrophile
would interact initially with the M-P bond. The resulting
intermediate species A might afterward rearrange in two
different ways: either by a movement of the electrophilic
fragment into a terminal position at phosphorus (route
followed by the hydrocarbon electrophiles) or into a terminal
position at the metal (route followed by the metallic electro-
philes) (Scheme 5). A delicate balance among the different
bonds being broken and formed seems to favor the formation
of the E-P bonds only for the hydrocarbon fragments, but
steric effects must be also operative, the latter disfavoring the
attachment of the incoming electrophile at the quite con-
gested P site. This might be a determinant factor in the case of
the comparatively bulkier metallic fragments studied in the
present work.
nonbonding Pb C(O) separations of about 2.9 A.
3 3 3
Spectroscopic data in solution for 13 and 130 are
comparable to each other and consistent with the trans-
dicarbonyl structure found for the molybdenum com-
pound in the crystal. The C-O stretching frequencies are
some 50 cm-1 lower than those of the corresponding Pb
compounds of type 8, this being explained by taking into
consideration the great difference in the oxidation states
of the metal atoms involved in each case (I vs IV). The
electron density at the group 6 metal atom in compounds
13 therefore must be intermediate between these in the
anions 1 and 10 and those in the tin and lead compounds 6
to 8 and, according to the discussion made above for the
latter compounds, the diamagnetic contribution to the
shielding of the 31P nucleus would have also an inter-
mediate magnitude. In agreement with this, the 31P reso-
nance in compounds 13 appears some 60 ppm higher than
in corresponding anion, but still some 20 ppm lower than
those of the related tin and lead derivatives (Tables 2 and 4).
Finally, the 31P spectra of compounds 13 also display a
much more shielded resonance corresponding to the gold-
bound phosphine ligand, in a position (ca. 50 ppm) com-
parable to those observed for related MoAu compounds
(e.g., 47.8 ppm for [Mo(BnNiPr2)(CO)3{Au(PPh3)}],25 or
47.7 ppm for [MoCp(CO)2(PMe3){Au(PPh3)}]).26
Concluding Remarks
The reactions of the anionic complexes 1 and 10 with
electrophilic halide compounds (E-Cl) of elements different
from carbon bear some parallelism but also significant
differences with the reactions of these anions with C-based
electrophiles. Reagents based on hard and strong electron-
withdrawing centers [Si(IV), Ge(IV), Ti(IV), Zr(IV)] end up
Upon attachment of an electrophile at the metal site
(to give a terminal oxophosphinidene complex) there is a
strong reduction of the electron density at that site and a
significant reduction in the multiplicity of the M-P bond
because of the concomitant reduction in the back-donation
from M to P. As a result, the P nucleus of the oxophos-
phinidene ligand is significantly deshielded, and a plot of the
31P chemical shifts for these complexes with respect to their
average C-O stretching frequencies relative to the anion 1
(24) Fischer, P. J.; Krohn, K. M.; Mwenda, E. T.; Young, V. G.
Organometallics 2005, 24, 5116.
(25) Braunstein, P.; Cura, E.; Herberich, G. E. J. Chem. Soc., Dalton
Trans. 2001, 1754.
(26) Galassi, R.; Poli, R.; Alessandra, E.; Fettinger, J. C. Inorg. Chem.
1997, 36, 3001.

Article
Inorganic Chemistry, Vol. 49, No. 24, 2010 11603
300.13 (1H), 121.50 (31P{1H}), or 75.47 MHz (13C{1H}) at 290 K
in CD2Cl2 solutions unless otherwise stated. Chemical shifts (δ)
are given in ppm, relative to internal tetramethylsilane (TMS) or
external 85% aqueous H3PO4 solutions (31P). Coupling con-
stants (J) are given in hertz (Hz).
Preparation of [MoCp{P(OSiMe3)R*}(CO)2] (2). Neat SiMe3Cl
(14 μL, 0.108 mmol) was added to a dichloromethane solution
(10 mL) of compound 1 (0.036 g, 0.054 mmol), whereupon the
solution changed immediately from yellow to red. The solvent
was then removed under vacuum, the residue was extracted
with petroleum ether, and the extracts were filtered. Removal
of the solvent from the filtrate yielded compound 2 as a red
microcrystalline solid (0.026 g, 81%). Anal. Calcd for C28H43-
O3MoPSi: C, 57.72; H, 7.44. Found: C, 57.32; H, 7.58. 1H NMR
(200.13 MHz): δ 7.36 (d, JHP = 3, 2H, C6H2), 4.99 (s, 5H, Cp), 1.59
(s, 18H, o-tBu), 1.32 (s, 9H, p-tBu), 0.39 (s, 9H, SiMe).
Figure 4. 31PChemicalshifts (δP/ppm) formolybdenum complexeswith
terminal and bridging oxophosphinidene ligands, plotted against their
average C-O stretching frequency relative to that of the anion 1(ΔνCO/cm-1),
with the type of element (E) bound to O or Mo indicated in each case.
The vertical axis is positioned at the chemical shift of the anion 1 (385 ppm).
All data in CH2Cl2 or CD2Cl2 solutions and taken from this work and
from reference 6.
Preparation of [WCp{P(OSiMe3)R*}(CO)2] (20). The proce-
dure is completely analogous to that described for 2, but now
using compound 10 (0.030 g, 0.040 mmol) and SiMe3Cl (10 μL,
0.077 mmol). This yielded compound 20 as a red microcrystalline
solid (0.021 g, 78%). Anal. Calcd for C28H43O3PSiW: C, 50.15;
1
H, 6.46. Found: C, 50.22; H, 6.54. H NMR (200.13 MHz):
δ 7.37 (d, JHP = 3, 2H, C6H2), 5.04 (s, 5H, Cp), 1.58 (s, 18H,
(taken here as an indicator of electron density at the metal)
reveals a good correlation although not a linear one, there-
fore suggesting that the overall reduction of the electron
density at phosphorus (caused indirectly by the attachment of
the electrophile at the metal site) must be one of the major
factors governing the observed changes in the 31P shifts
(Figure 4). In contrast, the attachment of an electrophile at
the O site of the anion to give an oxophosphinidene-bridged
complex also implies a reduction (even if not so strong) of the
electron density at the metal site, but now a reinforcement of
the M-P bond. The latter can be explained by taking into
account that the electron density removed from the metal is
now being driven toward the P atom, therefore reinforcing the
π component of the M-P bond. As a result of this electron
redistribution the P nucleus of the oxophosphinidene ligand
is significantly shielded, and a similar plot of the 31P chemical
shifts for the corresponding compounds (including the
alkoxyphosphide complexes previously reported by us) reveals
a gross correlation with respect to their average C-O stretch-
ing frequencies, although not as good as the one apparent for
the terminal oxophosphinidene complexes.
o-tBu), 1.31 (s, 9H, p-tBu), 0.40 (s, 9H, SiMe).
Preparation of [MoCp{P(OSiPh3)R*}(CO)2] (3). Solid
SiPh3Cl (0.025 g, 0.082 mmol) was added to a dichloromethane
solution (10 mL) of compound 1 (0.036 g, 0.054 mmol) where-
upon the solution changed immediately from yellow to red. The
solvent was then removed under vacuum, the residue was washed
with petroleum ether (2 ꢀ 4 mL) and then extracted with diethyl
ether. Removal of the solvent from the filtered extracts yielded
compound 3 as a red solid (0.033 g, 79%). Anal. Calcd for
C43H49O3MoPSi: C, 67.17; H, 6.42. Found: C, 67.02; H, 6.49.
1H NMR (200.13 MHz): δ 7.64-7.30 (m, 6H, Ph), 7.48-7.30
(m, 11H, Ph and C6H2), 5.03 (s, 5H, Cp), 1.49 (s, 18H, o-tBu),
1.34 (s, 9H, p-tBu).
Preparation of [MoCp{P(OGePh3)R*}(CO)2] (4). Solid GePh3Br
(0.018 g, 0.045 mmol) was added to a dichloromethane solution
(10 mL) of compound 1 (0.030 g, 0.045 mmol), and the mixture
was stirred at 273 K for 10 min to give an orange solution. Workup
as described for 2 (filtration using a canula) gave compound 4 as an
orange, quite air-sensitive solid (0.031 g, 84%). 1H NMR: δ 7.74-
7.59 (m, 6H, Ph), 7.53-7.38 (m, 9H, Ph), 7.34 (s, 2H, C6H2), 4.88
(s, 5H, Cp), 1.51 (s, 18H, o-tBu), 1.32 (s, 9H, p-tBu).
Preparation of [MoCp{P(OGeMe2Cl)R*}(CO)2] (5). The
procedure is completely analogous to that described for com-
pound 4, but now using GeMe2Cl2 (0.008 g, 0.046 mmol) and a
reaction time of 1 min. This yielded compound 5 as an orange,
quite air-sensitive solid (0.023 g, 79%). 1H NMR (200.13 MHz):
δ 7.32 (s, 2H, C6H2), 4.96 (s, 5H, Cp), 1.56 (s, 18H, o-tBu), 1.32
(s, 6H, GeMe), 1.29 (s, 9H, p-tBu).
Experimental Section
General Procedures and Starting Materials. All manipula-
tions and reactions were carried out under a nitrogen (99.995%)
atmosphere using standard Schlenk techniques. Solvents were
purified according to literature procedures,27 and distilled prior
to use. Petroleum ether refers to that fraction distilling in the
range 338-342 K. Compounds (H-DBU)[MCp(CO)2{P(O)R*}]
(M = Mo (1), W(10); DBU = 1,8-diazabicyclo [5.4.0] undec-
Preparation of [MoCp{P(O)R*}(CO)2(SnPh3)] (6). Solid
SnPh3Cl (0.037 g, 0.096 mmol) was added to a dichloromethane
solution (10 mL) of compound 1 (0.030 g, 0.045 mmol), and the
mixture was stirred for 1 min to give a pale yellow solution. The
solvent was then removed under vacuum, the residue was
extracted with toluene, and the extracts were filtered using a
canula. Crystallization of the filtrate from toluene-petroleum
ether (1:2) gave compound 6 as a yellow microcrystalline solid
(0.035 g, 90%). Anal. Calcd for C43H49O3MoPSn: C, 60.09; H,
7-ene; R* = 2,4,6-C6H2 Bu3)6 and [AuCl{P(p-tol)3}]28 were pre-
t
pared as described previously. All other reagents were obtained
from the usual commercial suppliers and used as received.
Chromatographic separations were carried out using jacketed
columns cooled by tap water (ca. 288 K). Commercial aluminum
oxide (activity I, 150 mesh) was degassed under vacuum prior to
use. The latter was mixed under nitrogen with the appropriate
amount of water to reach the activity desired. Filtrations were
performed using diatomaceous earth. IR stretching frequencies
of CO ligands were measured typically in solution. Nuclear
Magnetic Resonance (NMR) spectra were routinely recorded at
5.75. Found: C, 60.40; H, 5.90. ν(PdO) (Nujol mull) 1184 cm-1
.
1H NMR (200.13 MHz): δ 7.58-7.34 (m, 17H, Ph and C6H2),
4.82 (s, 5H, Cp), 1.64 (s, 18H, o-tBu), 1.35 (s, 9H, p-tBu).
Preparation of [MoCp{P(O)R*}(CO)2(SnPh2Cl)] (7). Solid
SnPh2Cl2 (0.017 g, 0.048 mmol) was added to an stirred dichloro-
methane solution (10 mL) of compound 1 (0.030 g, 0.045 mmol)
at 273 K to rapidly give a pale yellow solution. The solvent was
then removed under vacuum, the residue was extracted with
toluene-petroleum ether (1:2), and the extract was filtered using
(27) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals; Pergamon Press: Oxford, U. K., 1988.
(28) Braunstein, P.; Lehrer, H.; Matt, D. Inorg. Synth. 1990, 218.

11604 Inorganic Chemistry, Vol. 49, No. 24, 2010
Alonso et al.
a canula. Removal of the solvents from the filtrate gave complex
7 as a yellow powder (0.032 g, 86%). Anal. Calcd for C37H44O3-
ClMoPSn: C, 54.34; H, 5.42. Found: C, 54.65; H, 5.54. 1H NMR:
δ 8.10 (m, 1H, C6H2, cis isomer), 7.95 (m, 1H, C6H2, cis isomer),
7.68 (m, 4H, Ph, trans isomer), 7.50-7.38 (m, Ph, cis isomer, and
Ph and C6H2, trans isomer), 5.26 (s, 5H, Cp, cis isomer), 4.96
(s, 5H, Cp, trans isomer), 1.64 (s, 18H, o-tBu, trans isomer), 1.61
(s, 9H, CH3 cis isomer), 1.35 (s, 9H, p-tBu trans isomer), 1.28
(s, 9H, CH3, cis isomer), 1.17 (s, 9H, CH3, cis isomer), (ratio trans/
cis = 3.5). 119Sn{1H} NMR: δ 216.9 [d, J119SnP = 149 Hz, trans
isomer].
solution (10 mL) of compound 1 (0.042 g, 0.063 mmol), and the
mixture was stirred for 1 min to give a red solution. The solvent
was then removed under vacuum, the residue was extracted with
toluene, and the extracts were filtered. Removal of the solvent
from the filtrate gave complex 12 as a red powder (0.045 g, 92%).
Anal. Calcd for C351H44O3ClMoPZr: C, 54.86; H, 5.79. Found:
C, 55.10; H, 5.93. H NMR: δ 7.41 (d, JHP = 2, 2H, C6H2),
6.39 (s, 10H, Cp), 5.12 (s, 5H, Cp), 1.65 (s, 18H, o-tBu), 1.31
(s, 9H, p-tBu).
Preparation of [WCp{P(OZrCp2Cl)R*}(CO)2] (120). Solid
[ZrCp2Cl2] (0.017 g, 0.058 mmol) was added to a dichloro-
methane solution (10 mL) of compound 10 (0.042 g, 0.056 mmol),
and the mixture was stirred for 1 min to give a red solution. The
solvent was then removed under vacuum, the residue was
extracted with toluene-petroleum ether (2:1), and the extracts
were filtered. Removal of the solvents from the filtrate gave
compound 120 as a red powder (0.042 g, 87%). The crystals used
in the X-ray study were grown by slow diffusion of layers of
petroleum ether and toluene into a concentrated CH2Cl2 solu-
tion of the complex at 253 K. Anal. Calcd for C35H44O3Cl-
PWZr: C, 49.21; H, 5.19. Found: C, 48.93; H, 5.08. 1H NMR:
δ 7.41 (d, JHP = 2, 2H, C6H2), 6.42 (s, 10H, Cp), 5.15 (s, 5H, Cp),
1.64 (s, 18H, o-tBu), 1.31 (s, 9H, p-tBu).
Preparation of [WCp{P(O)R*}(CO)2(SnPh2Cl)] (70). The
procedure is completely analogous to that described for com-
pound 7, but now using compound 10 (0.030 g, 0.040 mmol) and
SnPh2Cl2 (0.015 g, 0.042 mmol). This yielded compound 70 as a
yellow powder (0.030 g, 83%). Anal. Calcd for C37H44O3-
1
ClPSnW: C, 49.07; H, 4.90. Found: C, 49.14; H, 5.09. H NMR:
δ 8.07 (m, 1H, C6H2, cis isomer), 7.91 (m, 1H, C6H2, cis isomer),
7.67-7.30 (m, Ph, cis isomer, and Ph and C6H2, trans isomer),
5.34 (s, 5H, Cp, cis isomer), 5.04 (s, 5H, Cp, trans isomer), 1.63
(s, 18H, o-tBu, trans isomer), 1.61 (s, 9H, CH3 cis isomer), 1.35
(s, 9H, p-tBu, trans isomer), 1.28 (s, 9H, CH3, cis isomer), 1.19
(s, 9H, CH3, cis isomer), (ratio trans/cis = 3.5).
Preparation of [MoCp{P(O)R*}(CO)2{AuP(p-tol)3}] (13). A
THF solution (5 mL) of [Au{P(p-tol)3}(THT)]PF6 was prepared
in situ by stirring [AuCl{P(p-tol)3}] (0.016 g, 0.030 mmol) and
TlPF6 (0.010 g, 0.030 mmol) in the presence of THT (tetrahydro-
thiophen, 0.05 mL, excess) for 10 min. The solution was filtered
using a canula and then cooled at 273 K and added to a THF
solution (10 mL) of compound 1 (0.020 g, 0.030 mmol), and the
mixture was stirred at 273 K for 1 min to give a pale-yellow
solution. The solvent was then removed under vacuum, the
residue was extracted with toluene-petroleum ether (2:1), and
the extracts were filtered using a canula. Removal of the solvents
from the filtrate gave complex 13 as a yellow powder (0.027 g,
90%). The crystals used in the X-ray study were grown by the
slow diffusion of layers of petroleum ether and toluene into a
concentrated CH2Cl2 solution of the complex at 253 K. Anal.
Calcd for C35H44O3AuMoP: C, 56.39; H, 5.66. Found: C, 56.08;
Preparation of [MoCp{P(O)R*}(CO)2(PbPh3)] (8). The pro-
cedure is completely analogous to that described for compound
7, but now using PbPh3Cl (0.024 g, 0.045 mmol) and a toluene-
petroleum ether (2:1) mixture for extractions. This yielded
compound 8 as a yellow powder (0.039 g, 91%). The crystals
used in the X-ray study were grown by slow diffusion of a layer
of petroleum ether into a concentrated toluene solution of the
complex at 253 K. Anal. Calcd for C43H49O3MoPPb: C, 54.48;
1
H, 5.21. Found: C, 54.12; H, 5.38. H NMR (200.13 MHz):
δ 7.56-7.30 (m, 17H, Ph and C6H2), 4.86 (s, 5H, Cp), 1.65
(s, 18H, o-tBu), 1.34 (s, 9H, p-tBu).
Preparation of [WCp{P(O)R*}(CO)2(PbPh3)] (80). The pro-
cedure is completely analogous to that described for compound
7, but now using compound 10 (0.030 g, 0.040 mmol) and
PbPh3Cl (0.021 g, 0.040 mmol), and a toluene-petroleum ether
(2:1) mixture for extractions. This yielded compound 80 as a
yellow powder (0.037 g, 90%). Anal. Calcd for C43H49O3PPbW: C,
49.86; H, 4.77. Found: C, 49.74; H, 4.89. 1H NMR (200.13 MHz):
δ 7.58-7.26 (m, 17H, C6H2, Ph), 4.94 (s, 5H, Cp), 1.64 (s, 18H,
o-tBu), 1.34 (s, 9H, p-tBu).
1
H, 5.48. H NMR (200.13 MHz): δ 7.45-7.23 (m, 14H, C6H4
and C6H2), 4.93 (s, 5H, Cp), 2.39 (s, 9H, CH3), 1.62 (s, 18H,
o-tBu), 1.34 (s, 9H, p-tBu).
Preparation of [WCp{P(O)R*}(CO)2{AuP(p-tol)3}] (130).
The procedure is completely analogous to that described for
compound 13, but now using compound 10 (0.020 g, 0.026 mmol).
This yielded complex 130 as a yellow powder (0.026 g, 90%).
Anal. Calcd for C35H44O3AuPW: C, 51.75; H, 5.19. Found: C,
52.02; H, 5.39. 1H NMR: δ 7.45-7.23 (m, 8H, C6H4 and C6H2),
7.25 (m, 6H, C6H4), 5.02 (s, 5H, Cp), 2.38 (s, 9H, CH3), 1.60
(s, 18H, o-tBu), 1.33 (s, 9H, p-tBu).
Preparation of [MoCp{P(OSnMe3)R*}(CO)2] (9). A large
excess of SnMe3Cl (0.108 g, 0.54 mmol) was added to a stirred
dichloromethane solution (10 mL) of compound 1 (0.036 g,
0.054 mmol), whereupon the solution changed immediately
from yellow to orange. The solvent was then removed under
vacuum, the residue was extracted with petroleum ether, and the
extracts were filtered. Crystallization of the concentrated fıltrate
´
X-ray Structure Determination for Compound 8. A single
crystal was coated in parafin oil and mounted on a glass fiber.
X-ray measurements were made using a Bruker SMART CCD
area-detector diffractometer with Mo-KR radiation.29 Inten-
sities were integrated from several series of exposures,30 each
exposure covering 0.3° in ω, and the total data set being a sphere.
Absorption corrections were applied, based on multiple and
symmetry-equivalent measurements.31 The structure was solved
by direct methods and refined by least-squares on weighted F2
values for all reflections (Table 6).32 All non-hydrogen atoms
were assigned anisotropic displacement parameters and refined
at 253 K yielded compound 9 as an orange microcrystalline solid
(0.030 g, 81%). Anal. Calcd for C28H43O3MoPSn: C, 49.95; H,
6.44. Found: C, 50.13; H, 6.58. 1H NMR (200.13 MHz, C6D6):
Isomer 9: δ 7.40 (s, 2H, C6H2), 4.87 (s, 5H, Cp), 1.68 (s, 18H,
1
o-tBu), 1.25 (s, 9H, p-tBu), 0.40 (s, br, 9H, SnMe). H NMR
(CD2Cl2, 203K): Isomer 10: δ 7.44 (s, 2H, C6H2), 4.87 (s, 5H,
Cp), 1.68 (s, 18H, o-tBu), 1.25 (s, 9H, p-tBu), 0.40 (s, br, 9H, CH3).
Preparation of [MoCp{P(OTiCp2Cl)R*}(CO)2] (11). Solid
[TiCp2Cl2] (0.014 g, 0.056 mmol) was added to a dichloro-
methane solution (10 mL) of compound 1 (0.036 g, 0.054 mmol),
and the mixture was stirred for 2.5 h to give a red solution. The
solvent was then removed under vacuum, the residue was
extracted with petroleum ether, and the extracts were filtered.
Removal of the solvents from the filtrate gave complex 11 as a
red, quite air-sensitive powder (0.035 g, 90%). 1H NMR: δ 7.39
(d, JHP = 2, 2H, C6H2), 6.51 (s, 10H, Cp), 5.04 (s, 5H, Cp), 1.63
(s, 18H, o-tBu), 1.32 (s, 9H, p-tBu).
(29) SMART Diffractometer Control Software; Bruker Analytical X-ray
Instruments Inc.: Madison, WI, 1998.
(30) SAINT integration software; Siemens Analytical X-ray Instruments
Inc.: Madison, WI, 1994.
(31) Sheldrick, G. M. SADABS, Program for Empirical Absorption
€
€
Correction; University of Gottingen: Gottingen, Germany, 1996.
(32) SHELXTL program system, version 5.1; Bruker Analytical X-ray
Instruments Inc.: Madison, WI, 1998.
Preparation of [MoCp{P(OZrCp2Cl)R*}(CO)2] (12). Solid
[ZrCp2Cl2] (0.019 g, 0.065 mmol) was added to a dichloromethane

Article
Inorganic Chemistry, Vol. 49, No. 24, 2010 11605
Table 6. Crystal Data for Compounds 8, 120, 13
8 1/2C7H8
3
120
13
mol formula
mol wt
cryst syst
space group
radiation (λ, A)
a, A
C46.5 H53MoO3PPb
993.99
monoclinic
C2/c
0.71073
55.0030(14)
C35H44ClO3PWZr
854.19
monoclinic
P21/c
1.54184
11.1396(3)
C46H55AuMoO3P2
1010.75
monoclinic
P21/c
0.71073
10.01660(10)
b, A
c, A
10.2300(3)
15.0932(4)
17.1725(3)
18.0948(3)
26.9633(4)
17.1911(4)
R, deg
β, deg
90
91.1580(10)
90
98.2870(10)
90
110.9870(10)
γ, deg
90
8490.9(4)
90
3425.30(12)
90
4334.97(13)
V, A3
Z
8
1.555
4.332
173(2)
2.02 to 27.46
-71/71; -13/13; -19/19
43894
4
1.656
10.020
150(2)
3.57 to 69.80
-13/13; 0/20; 0/21
52614
6378(0.1258)
5412
4
1.549
3.781
100(2) K
1.48 to 25.24
-12/11; -32/0; -20/10
20817
calcd density, gcm-3
absorpt. coeff., mm-1
temperature, K
θ range (°)
index ranges (h, k, l)
no. of reflns collected
no. of indep reflns (Rint
reflns with I>2σ(I)
)
9717 (0.0471)
8223
7525 (0.0383)
6081
R indexesa [data with I>2σ(I)]
R indexesa (all data)
GOF
R1 = 0.0315, wR2 = 0.0762b
R1 = 0.0416, wR2 = 0.0806b
1.034
R1 = 0.0510, wR2 = 0.1311c
R1 = 0.0596, wR2 = 0.1453c
1.057
R1 = 0.0404, wR2 = 0.1085d
R1 = 0.0588, wR2 = 0.1498d
1.248
no. of restraints/params
ΔF(max, min), e A-3
227/549
1.827, -1.064
0/388
2.521, -1.196
0/490
1.497, -2.319
P
P
P
P
a R = ||Fo| - |Fc||/ |Fo|. wR = [ w(Fo2 - Fc2)2/ w|Fo|2]1/2. w = 1/[σ2(Fo2) þ (aP)2 þ bP] where P = (Fo þ 2Fc2)/3. b a = 0.0370, b = 29.0672.
2
c a = 0.1054, b = 0.0000. d a = 0.0826, b = 4.0595.
without positional constraints. All hydrogen atoms were con-
strained to ideal geometries and refined with fixed isotropic
displacement parameters. Refinement proceeded smoothly to
give the residuals shown in Table 6. Complex neutral-atom
scattering factors were used.33
X-ray Structure Determination for Compounds 120and 13.
Data collection for these compounds was performed on a
Nonius Kappa CCD single diffractometer, using Cu-KR (120)
or graphite-monochromated Mo-KR (13) radiation. Images
were collected at 29 or 40 mm fixed crystal-detector distance,
using the oscillation method, with 1.5° or 1° oscillation and 60 or
50 s exposure time per image, for compounds 120 or 13 respec-
tively. Data collection strategy was calculated with the program
Collect.34 Data reduction and cell refinements were performed
with the programs HKL Denzo and Scalepack.35 Semiempirical
absorption corrections were applied using the program
SORTAV.36 Using the program suite WinGX,37 the structure
were solved by Patterson interpretation and phase expansion,
and refined with full-matrix least-squares on F2 with SHELXL97.38
All non-hydrogen atoms were refined anisotropically. All
hydrogen atoms were geometrically placed, and they were given
an overall isotropic thermal parameter. Further details of the
data collection and refinements are given in the Table 6.
Acknowledgment. We thank the DGI of Spain and the
ꢀ
Consejerıa de Educacion de Asturias for financial support
´
(Projects CTQ2006-01207 and IB05-110), and the MEC of
Spain for a grant (to M.A.).
Supporting Information Available: A CIF file containing the
crystallographic data for the structural analysis of compounds
8, 120, and 13. This material is available free of charge via the
Internet at http://pubs.acs.org.
(33) International Tables for Crystallography; Kluwer: Dordrecht,
The Netherlands, 1992; Vol. C.
(36) Blessing, R. H. Acta Crystallogr., Sect. A 1995, 51, 33.
(37) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837.
(38) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112.
(34) Collect; Nonius BV: Delft, The Netherlands, 1997-2004.
(35) Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307.