Reactivity of the unsaturated manganese dihydrides [Mn2(μ-H) 2(CO)6(μ-L2)] [L2 = (EtO) 2POP(OEt)2, Ph2PCH2PPh2, Me2PCH2PMe<

Article abstract of DOI:10.1016/j.jorganchem.2010.12.016

The dimanganese hydride complexes [Mn₂(μ-H) ₂(CO)₆(μ-L₂)] [L₂ = (EtO) ₂POP(OEt)₂ (tedip), Ph₂PCH₂PPh ₂ (dppm)] react with primary and secondary s

Full text of DOI:10.1016/j.jorganchem.2010.12.016

Journal of Organometallic Chemistry 696 (2011) 1736e1748
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Reactivity of the unsaturated manganese dihydrides [Mn2(m-H)2(CO)6(m-L2)]
[L2 ¼ (EtO)2POP(OEt)2, Ph2PCH2PPh2, Me2PCH2PMe2] toward silicon and tin
hydrides
,
b
M. Angeles Alvarez a, M. Pilar Alvarez a, Remedios Carreño a, Miguel A. Ruiz a , Claudette Bois
*
a Departamento de Química Orgánica e Inorgánica/IUQOEM, Universidad de Oviedo, E-33071 Oviedo, Spain
b Laboratoire de Chimie Inorganique et Matériaux Moléculaires, Université P. et M. Curie, 4 Place Jussieu, 75252 Paris cedex 05, France
a r t i c l e i n f o
a b s t r a c t
Article history:
The dimanganese hydride complexes [Mn2(
m
-H)2(CO)6(
m
-L2)] [L2 ¼ (EtO)2POP(OEt)2 (tedip), Ph2PCH2PPh2
Received 20 October 2010
Received in revised form
9 December 2010
(dppm)] react with primary and secondary silanes H2SiPhR (R ¼ Ph, Me, H) to give the corresponding
derivatives [Mn2(m-H2SiPhR)(CO)6(m-L2)] having a silane molecule displaying a relatively unusual m- :
k2 k2
coordination mode (averaged values are ca. MneH ¼ 1.59 Å, HeSi ¼ 1.69 Å and MneSi ¼ 2.381 Å, when
R ¼ Ph and L2 ¼ tedip). These complexes display in solution cis and/or trans arrangement of the bridging
silane relative to the diphosphorus ligands (and facial and/or meridional arrangements of the corre-
sponding carbonyl ligands), depending on the bridging groups. The novel unsaturated dihydride
Accepted 9 December 2010
Keywords:
Manganese
Metalemetal interactions
Hydride ligands
Diphosphine ligands
Tin hydrides
[Mn2(
[Mn2(
m
m
-H)2(CO)6(
-Cl)2( -dmpm)(CO)6] and 5 equiv of Li[BH2Me2] in tetrahydrofuran followed by addition of water.
-H)2(CO)6(
-L2)] (L2 ¼ tedip, dppm, dmpm) react with HSnPh3 to give
different mixtures of products strongly dependent on the particular reaction conditions. We have thus been
able to isolate and characterize fivenew types of dimanganese-tin derivatives: [Mn2( -SnPh2)2(CO)6( -L2)],
[Mn2( -H)( -Ph2SnO(H)SnPh2)(CO)6(
-L2)] (average values are MneSn ¼ 2.54 Å, SneO ¼ 2.11 Å, when
L2 tedip), [Mn2( -H)( -L2)], [Mn2( -H)( -L2)], and
k1 2-HSnPh2)(CO)6( k1 1-O(H)SnPh2)(CO)6(
[Mn2( -H)(SnPh3)(CO)7(
m
-dmpm)] (dmpm ¼ Me2PCH2PMe2) has been prepared through the reaction of
m
The dihydride complexes [Mn2(
m
m
Silicon hydrides
m
m
m
m
m
¼
m
m
-
:k
m
m
m
-
:
k
m
m
m
-L2)] (MneMn ¼ 3.237(1) Å, MneSn ¼ 2.642(1) Å, when L2 ¼ dppm).
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
bonds. The latter species in principle should be able to react with
these reagents under mild conditions, thus leading to products
difficult or impossible to obtain from electron-precise binuclear
precursors, because the latter usually require more forcing condi-
tions (i.e. heating or photochemical induction) to react with these
mild reagents.
In the last decades the interest in the chemistry of metallic
complexes containing bonds to semimetallic p-block elements (E)
such as silicon and tin has grown remarkably, with many of the
synthetic and catalytic reactions under study involving the cleavage
of EeH bonds by transition metals. For instance, the oxidative
addition of EeH bonds to a transition-metal center is not only
a widespread synthetic method to obtain complexes containing
metalesilicon bonds [1], but also a fundamental step involved in
several transition-metal catalyzed processes of relevance such as
hydrosilylation [2], dehydrocoupling of organotin and organo-
silicon hydrides [3], and redistribution reactions of silanes [4],
among others. While the reactivity of mononuclear complexes with
silicon and tin hydrides has been examined in detail, the number of
studies involving binuclear complexes is more scarce, especially in
the case of unsaturated complexes having multiple metalemetal
Some time ago we reported a preliminary study revealing
that the reactions of the unsaturated dihydride complexes
[Mn2(
m
-H)2(CO)6(
m
-L2)] (L2 ¼ (EtO)2POP(OEt)2, tedip (1a) [5],
Ph2PCH2PPh2, dppm (1b) [6], Chart 1) with H2SiPh2 and HSnPh3
could take place readily under mild conditions and involved in all
cases some kind of EeH bond activation by the dimanganese center
[7]. Thus, 1b reacted with diphenylsilane to give [Mn2(
(CO)6( -dppm)], a complex with the silane molecule displaying
a relatively unusual
k2 k2 coordination mode. In contrast, the
reactions with HSnPh3 were more complex and gave stannylene
-SnPh2) or hydrostannyl-bridged ( -HSnPh2) products, among
other uncharacterized species at the time [7]. Again, the stannyl
complexes displayed relatively uncommon
k1 2-bridging
m-H2SiPh2)
m
m- :
(m
m
a
m- :k
mode involving a tricentric MneHeSn interaction. These results
were in marked contrast with the reactions of the triosmium
* Corresponding author. Fax: þ34 985103446.
E-mail address: mara@uniovi.es (M.A. Ruiz).
0022-328X/$ e see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2010.12.016

M.A. Alvarez et al. / Journal of Organometallic Chemistry 696 (2011) 1736e1748
1737
Chart 1.
dihydride [Os3(m-H)2(CO)10] (a cluster having an unsaturated
Os2( -H)2 center comparable to those in compounds 1) with
m
H2SiPh2 and HSnMe3 to give products having only terminal silyl or
stannyl groups [8], and this prompted us to further study the
reactions of our dimanganese hydrides with organosilicon and tin
hydrides. In this paper we give full details of the reactions reported
in our preliminary studies, which we have further extended to
other silanes and, in the case of the complex reactions with HSnPh3,
Chart 2.
isomers upon increasing bulkness of the bridging groups. Inci-
dentally, we note that all the H3SiPh derivatives were quite
unstable and would decompose in the absence of excess of silane to
give the starting dihydrides 1a and 1b. Under the above interpre-
tation, we could predict that the use of a silane with intermediate
steric requirements should accordingly yield a mixture of isomers
with intermediate mer/fac ratios. Indeed, the tedip-bridged hydride
1a reacted with H2SiPhMe to give a mixture of the corresponding
mer (2a.2) and fac (3a.2) isomers in a 5:1 ratio. Surprisingly, no
reaction was observed between 1b and H2SiPhMe, even in refluxing
toluene and by using a large excess of silane.
to the novel unsaturated hydride [Mn2(m-H)2(CO)6(m-dmpm)] (1c),
the latter having a diphosphine ligand (dmpm ¼ Me2PCH2PMe2)
with smaller steric demands and higher electron donor ability than
the tedip and dppm ligands. Since our preliminary report, only
a
few related studies involving unsaturated hydride-bridged
complexes and silicon or tin hydrides have appeared. Suzuki et al.
studied the reactions of several secondary and tertiary silanes with
the 28-electron complexes [Fe2Cp*2(m-H)4] [9] and [Ru2Cp*2(m-H)4]
We should note that, except for compounds 2a.1 and 2b.1, all
other silane derivatives of the hydrides 1a and 1b were obtained as
oily species and could not be isolated as solids, nor the mer and fac
isomers could be separated in each case. In addition, all these
complexes were found to be rather unstable species in solution,
especially in the absence of excess silane, and slowly would
decompose to give back the corresponding starting dihydrides.
Presumably, different polysilanes and polysiloxanes are formed in
these processes, as revealed by 29Si NMR inspection of some of
these solutions after decomposition, although we did not investi-
gate further this matter.
[10], to give different silylene- and silyl-bridged derivatives, with
the m- :
k2 k2 coordination mode of a silane ligand being found only
for the bulky H2SiiPr2 and H2SitBu2 molecules [9b,10b]. More
recently, we have reported the reaction of the 30-electron complex
[Mo2Cp2(
-SnPh3)(CO)2], a molecule displaying a triorganostannyl group
unusually bridging two metal atoms exclusively through its tin
atom, and [Mo2Cp2( -H)(HPCy2)(SnPh3)(CO)2], a molecule following
m-H)(m-PCy2)(CO)2] with HSnPh3 to give [Mo2Cp2(m-PCy2)
(m
m
from an unexpected combination of SneH bond cleavage and PeH
bond formation steps [11].
2. Results and discussion
2.2. Characterization of the silane complexes 2 and 3
2.1. Reactions of the dihydride complexes 1a,b with organosilanes
The structure of compound 2a.1 has been confirmed by an X-ray
diffraction study and is shown in the Fig. 1, while the most relevant
bond distances and angles are collected in the Table 1. The molecule
is made up of two meridional (or T-shaped) Mn(CO)3 fragments
Compounds 1a,b do not react with tertiary silanes HSiR3 (R ¼ Ph,
Et) under heating (refluxing toluene solutions) or under exposure
to the visibleeUV light. In contrast, these unsaturated hydrides
react with primary and secondary silanes H2SiPhR (R ¼ Ph, Me, H)
at room temperature or under moderate heating to give the
corresponding derivatives [Mn2(
a silane molecule displaying a relatively unusual
nation mode, and two possible isomeric forms: mer-[Mn2(
H2SiPhR)(CO)6( -L2)] (2) and fac-[Mn2( -H2SiPhR)(CO)6( -L2)] (3).
m
-H2SiPhR)(CO)6(
m
- :
-L2)] having
m
k2 k2 coordi-
m-
m
m
m
These isomers differ in the relative positioning of the CO ligands
around each manganese center (meridional or facial) and in the
relative positioning of the bridging ligands (trans or cis) (Chart 2).
The ratio of mer to fac isomers present in the corresponding reac-
tion mixtures was found to be dependent on the steric require-
ments of the silane. Thus, for the bulkier diphenylsilane molecule,
only the mer isomers 2a.1 and 2b.1 were observed in solution. On
the other extreme, the fac isomers were the species dominant when
using a silane with small steric requirements such as H3SiPh. Thus,
the dppm dihydride 1b gave exclusively the corresponding fac
isomer 3b.3, while the ratio mer to fac was found to be 1:6 for the
tedip-bridged isomers 2a.3 and 3a.3. This influence can be under-
stood by considering that the repulsive steric interactions between
the bridging diphosphorus and silane ligands should be much
higher in the fac isomer, hence the preferential formation of mer
Fig. 1. ORTEP diagram (30% probability) of compound 2a.1, with Et and Ph groups
(except the ipso C atoms in the latter) omitted for clarity.

1738
M.A. Alvarez et al. / Journal of Organometallic Chemistry 696 (2011) 1736e1748
Table 1
Table 2
IR data for new compounds.
Selected bond lengths (Å) and angles (ꢀ) for compound 2a.1.
Mn(1)eMn(2)
Mn(1)eSi(1)
Mn(1)eP(1)
Mn(1)eC(1)
Mn(1)eC(2)
Mn(1)eC(3)
Mn(1)eH(1)
Si(1)eH(1)
Mn(2)eSi(1)
Mn(2)eP(2)
Mn(2)eC(4)
Mn(2)eC(5)
Mn(2)eC(6)
Mn(2)eH(2)
Si(1)eH(2)
2.9375(8)
2.388(1)
2.202(1)
1.784(4)
1.828(4)
1.832(4)
1.59(1)
1.669(6)
2.376(1)
2.189(1)
1.783(5)
1.828(4)
1.837(4)
1.59(1)
Si(1)eMn(1)eMn(2)
P(1)eMn(1)eMn(2)
P(1)eMn(1)eSi(1)
C(1)eMn(1)eSi(1)
C(1)eMn(1)eP(1)
Mn(1)eSi(1)eMn(2)
C(10)eSi(1)eC(20)
Si(1)eMn(2)eMn(1)
P(2)eMn(2)eMn(1)
P(2)eMn(2)eSi(1)
C(4)eMn(2)eP(2)
Mn(1)eH(1)eSi(1)
Mn(2)eH(2)eSi(1)
H(1)eSi(1)eH(2)
51.75(3)
88.45(3)
139.06(5)
124.4(1)
96.3(1)
76.13(4)
106.7(2)
52.13(3)
85.63(3)
136.68(5)
94.7(1)
94.4(2)
93.1(1)
158.0(6)
174.1(2)
177.9(1)
Compound
n
(CO)a/cmꢁ1
[Mn2(
mer-[Mn2(
-tedip)] (2a.1)
mer-[Mn2( -H2SiPh2)(CO)6
-dppm)] (2b.1)
mer-[Mn2( -H2SiPhMe)(CO)6
-tedip)] (2a.2)
fac-[Mn2( -H2SiPhMe)(CO)6
-tedip)] (3a.2)
mer-[Mn2( -H2SiPhH)(CO)6
-tedip)] (2a.3)
fac-[Mn2( -H2SiPhH)(CO)6
-tedip)] (3a.3)
fac-[Mn2( -H2SiPhH)(CO)6
-dppm)] (3b.3)
[Mn2( -SnPh2)2(CO)6
-tedip)] (4a)
[Mn2( -SnPh2)2(CO)6
-dppm)] (4b)
[Mn2( -H){ -Ph2SnO(H)
SnPh2}(CO)6( -tedip)] (5a)
[Mn2( -H){ -Ph2SnO(H)
SnPh2}(CO)6( -dppm)] (5b)
[Mn2( -H){ -Ph2SnO(H)
SnPh2}(CO)6( -dmpm)] (5c)
[Mn2( -H)( -HSnPh2)(CO)6
-tedip)] (6a)
[Mn2( -H)( -HSnPh2)(CO)6
-dmpm)] (6c)
[Mn2( -H){ -O(H)SnPh2}(CO)6
-dppm)] (7b)
[Mn2( -H){ -O(H)SnPh2}(CO)6
-dmpm)] (7c)
[Mn2( -H)( -OH)(CO)6
-dmpm)] (8c)
[Mn2( -H)(SnPh3)(CO)7
-dppm)] (9b)
[Mn2( -H)(SnPh3)(CO)7
-dmpm)] (9c)
m
-H)2(CO)6(
m
-dmpm)] (1c)
2032 (vs), 1988 (vs), 1944 (m), 1915 (vs)
2014 (m), 1974 (vs), 1943 (m)
m
-H2SiPh2)(CO)6
(m
m
2037 (vw), 1995 (s), 1955 (vs), 1925 (s)b
2016 (m), 1976 (vs, br)c, 1947 (s)d
(m
m
(m
m
2048 (m), 2001 (s), 1976 (s, br)c,
1957 (vs, sh), 1936 (m, sh)e
2020 (m), 1985 (vs, br)c, 1944 (s)c,
1940 (s)c
(m
m
(m
m
2050 (vs), 2005 (vs), 1985 (vs, br)c,
1956 (s, br), 1940 (s)c
(m
1.694(3)
H(1)eMn(1)eP(1)
H(2)eMn(2)eP(2)
m
2033 (vs), 1990 (vs), 1963 (m),
1942 (m, sh), 1934 (m), 1919 (w)e
2012 (s), 1981 (vs), 1940 (m, sh),
1928 (vs), 1913 (s, sh)
(m
m
(m
m
2010 (s), 1972 (s), 1925 (s, sh), 1915 (vs)
connected by a metalemetal bond and bridged by tedip and silane
ligands arranged mutually in trans, with the latter displaying
(m
m
m
2031 (vw), 2007 (w), 1957 (vs), 1930 (sh),
1917 (m)d,f
2035 (vw), 1992 (w), 1942 (vs), 1914 (sh),
1907 (s)
a m- :
k2 k2 coordination mode with two tricentric MneHeSi inter-
m
m
m
actions. This is a relatively rare coordination mode of silanes that
has been previously identified only in a reduced number of dinu-
clear complexes of Re [12], Ru [10,13], Fe [9b], and Pt [14].
Overall, the structure of 2a.1 can be related to that of the Mn2Zn
m
m
m
2015 (w), 1991 (w), 1936 (vs), 1902 (m)
m
m
m
2057 (vs), 2015 (vs), 1987 (vs),
1964 (s, sh), 1959 (s), 1930 (s, br)d
2033 (vs), 1988 (vs), 1960 (m), 1942 (s),
1910 (s)
(m
cluster [Mn2Zn(m-H)2(CO)6(Me2NCH2CH2NMe2)(m-tedip)] [15], the
m
m
latter having the same electron count and displaying two hydride
ligands bridging the MneZn edges. Indeed the MneMn distance in
2a.1, 2.9375(8) Å, is consistent with the single metalemetal bond
expected for an electron-precise dimanganese complex, although
somewhat shorter than the value of 3.039(5) Å found in the
mentioned Mn2Zn cluster, or the value of 2.9889(6) Å measured in
(m
m
m
2032 (s), 1997 (s), 1953 (m), 1930 (m),
(m
1920 (vs)d
m
m
2028 (vs), 1990 (vs), 1944 (s), 1917 (s)
(m
m
m
2027 (s), 1995 (vs), 1934 (m), 1905 (vs)
(m
m-H)(m- : -
k2 k2
m
2088 (m), 2021 (s), 2005 (vs), 1977 (s),
1907 (s), 1892 (s)
2079 (m), 2016 (m), 1993 (vs), 1977 (sh),
1969 (sh), 1908 (m), 1890(m)
the related tetrahydroborate-bridged complex [Mn2(
BH4)(CO)6( -dppm)] [7,16], the differences being easily ascribed to
(m
m
m
the distinct bridging groups being present in each case. The
bridging silicon atom in 2a.1 is found at ca. 2.38 Å from the
manganese atoms, a distance within the range of values found in
mononuclear compounds displaying tricentric MneHeSi interac-
tions, typically 2.25e2.46 Å [1,17]. As for the SieH distances, the
values of 1.67(2) and 1.70(2) Å are substantially longer than those
reported for the free organosilanes (ca. 1.48 Å [1]), suggesting
a considerable strength for the tricentric interactions. The MneH
distances [1.59(1) Å] are even shorter than the values of ca. 1.65 Å
measured in the mentioned tetrahydroborate complex and there-
fore suggestive of a very strong interaction too. Indeed, these values
are comparable to those of a MneH bond in a terminal hydride
complex (eg. 1.601(16) Å in [MnH(CO)5] [18]). At the same time, the
angles MneHeSi have rather low values [94(1)ꢀ and 93(1)ꢀ] and
therefore we can classify the corresponding SieHeMn interactions
in 2a.1 as of the side-on type, as opposed to the end-on type
coordination of the BeH bonds described for the mentioned tet-
rahydroborate complex [7,16], the latter being characterized by
more open EeHeMn angles (120e130ꢀ).
Spectroscopic data in solution for compounds 2a.1 and 2b.1 are
similar to each other and consistent with the solid-state structure
just discussed for the tedip-bridged complex. In particular, both
complexes display medium to weak symmetrical CeO stretching
bands above 2000 cmꢁ1 (Table 2), this denoting the presence of T-
shaped Mn(CO)3 oscillators in these molecules [19], and both
display a single resonance in the corresponding 31P{1H} NMR
spectra as expected (Table 3). The symmetrical coordination of the
silane molecule involving agostic-type, tricentric MneHeSi inter-
actions is denoted in each case by the presence of highly shielded
false triplet resonances at ꢁ12.10 (2a.1) or ꢁ13.09 (2b.1) ppm, due
to the chemically equivalent but magnetically inequivalent Mn-
(m
a
b
c
d
e
f
Recorded in CH2Cl2 solution unless otherwise stated.
Recorded in tetrahydrofuran.
Band due to both fac and mer isomers.
Recorded in petroleum ether.
Recorded in toluene.
n
(OeH) 3340 (w) in Nujol mull.
bound hydrogen atoms in each case. Other resonances present in
the 1H spectra of these complexes are in agreement with the
symmetrical structure found for 2a.1 in the crystal and need not to
be discussed in detail.
The spectroscopic features of the fac isomer 3b.3 are signifi-
cantly different from those of the mer isomers just discussed. In the
first place, its IR spectrum (Table 2) displays five strong CeO
stretching bands with a pattern comparable to that of related
dimanganese compounds having facial Mn(CO)3 groups, such as
[Mn2(m-H)(m-BH4)(CO)6(m-dppm)] [16] and [Mn2Au2SnCl2(CO)6(m-
dppm){P(p-tol)3}2] [20]. The stretching band due to the SieH
terminal bond is not observed in this spectrum, most likely
obscured by the CeO stretching bands of the complex [21]. That
bond, however, is denoted by the presence of a resonance at
5.89 ppm in the 1H NMR spectrum [22], with the latter also dis-
playing a broad resonance at ꢁ10.22 ppm with intensity corre-
sponding to 2H atoms, indicative of the presence of a symmetrically
coordinated
m- :k
k2 2-H2SiHPh molecule. Although there are two
possible orientations of the silane molecule with respect to the
diphosphine ligand, only a single conformer is present in the
solutions of 3b.3, presumably the one with the Ph substituent on
the silane pointing away from the bridging diphosphine, more
favored on steric grounds (Chart 2).

M.A. Alvarez et al. / Journal of Organometallic Chemistry 696 (2011) 1736e1748
1739
Table 3
31P{1H}a and 1H NMRb data for new compounds.
d
[P]/ppm
d
[H]/ppm
J/Hzc
J(PP)
P1
42.4d
179.5
63.3
P2
MneHeMn
ꢁ18.02e
MneHeE
OeH
J(HP)
20
J(H1Sn)
J(H2Sn)
1c
2a.1
2b.1
2a.2
3a.2
2a.3
3a.3
3b.3
4a
ꢁ12.10 (ft)
ꢁ13.09 (ft)
ꢁ12.42 (ft)
ꢁ11.02 (br)
ꢁ12.96 (ft)g
ꢁ11.58 (ft)g
ꢁ10.22 (br)e
15
12
14
179.9
171.6
178.9f
12
23
175.2f
63.5
168.4h
45.9h
4b
5a
5b
176.2h
ꢁ24.40 (t)h
ꢁ25.90 (t)h
ꢁ25.50 (t)
2.32
2.50
2.10
13
15
85
75
66.4h
21
5c
36.8
13
90
6a
6c
7b
7c
175.5 (d)h,i
31.01 (AB)j
61.0 (d)h,i
33.5 (AB)j
31.3j
165.3 (d)h,i
27.7 (AB)j
43.8 (d)h,i
30.1 (AB)j
ꢁ18.40 (t)h
ꢁ18.20 (t)k
ꢁ20.40 (dd)h
ꢁ21.75 (t)
ꢁ9.49 (d)
ꢁ9.43 (d)k
88
68
54
65
28
23
22, 37
26
124
142
97
93
32
26
ꢁ2.08
ꢁ2.95
120
8c
ꢁ11.86 (t)k
ꢁ24.84 (dd)e,m
ꢁ24.85 (dd)e,l
ꢁ2.49 (t)
5, 21
14, 21
7, 20
9b
71.1 (d)d,l
38.2 (d)d,l
50.0 (d)d,l
25.7 (d)d,l
70
60
35
47
9cn
a
Recorded at room temperature in C6D6 solution at 121.50 MHz, unless otherwise stated;
d
in ppm relative to external 85% aqueous H3PO4, with multiplicity indicated in
brackets; labeling according to the figure shown below.
b
Recorded at room temperature in C6D6 solution at 300.13 MHz, unless otherwise stated;
d in ppm relative to internal TMS, with multiplicity indicated in brackets; ft stands
for false triplet; in those cases, the values quoted as J(HP) are actually the absolute values of J(HP) þ J(HP0).
c
J (HSn) refers to the average coupling between H and the 119Sn and 117Sn isotopes.
Recorded at 161.9 MHz.
Recorded at 400.13 MHz.
Recorded in toluene-d8 at 253 K.
Recorded in CD2Cl2.
Recorded in CDCl3.
Recorded in CDCl3 at 243 K.
Recorded at 81.01 MHz.
Recorded at 200.13 MHz.
Recorded in CD2Cl2 at 223 K.
Recorded in CD2Cl2 at 263 K.
Data for the isomer B.
d
e
f
g
h
i
j
k
l
m
n
As stated above, the tedip-bridged complexes [Mn2(
m
-H2SiPhH)
ca. 5 equiv of Li[BH2Me2] in tetrahydrofuran at room temperature
to yield a red-greenish solution after 3 h stirring. The subsequent
addition of degassed water to the latter mixture and removal of the
solvents under vacuum gives a dark-red residue yielding ca. 70% of
the purple dihydride complex 1c after chromatographic purifica-
tion at 243 K. Compound 1c can be also obtained analogously by
using K[BH(CHMeEt)3] as reducing agent, but the yield is signifi-
cantly lower (ca. 45%), while the use of Li[HBEt3] failed to give
significant amounts of the desired complex.
Spectroscopic data for 1c are very similar to those of the dppm-
bridged analog 1b [6] and need not to be discussed in detail. We
just note that 1c displays CeO stretching bands some 7 cmꢁ1 lower
than those of 1b, this illustrating the better donor properties of
the dmpm ligand (compared to dppm), while the equivalent
(CO)6( -tedip)] and [Mn2( -H2SiPhMe)(CO)6( -tedip)] are obtained
m
m
m
as mixtures of the corresponding mer and fac isomers, this being
apparent from the complexity of the corresponding IR spectra and
the appearance in each case of separated 31P and 1H NMR reso-
nances that are comparable to those of compounds 2a.1 (mer) or to
those of 3b.3 (fac) (Tables 2 and 3). We notice that the MneHeSi
resonance in the fac isomer 3a.3 displays a large overall coupling to
0
phosphorus (JPH þ JP H ¼ 23 Hz), comparable to the cis couplings
observed in many dimanganese compounds of the type [Mn2(
m-H)
(m-X)(CO)6(m-L2)] (20e30 Hz). These values are much higher than
0
the overall values observed in the mer isomers (cf. JPH þ JP H ¼ 12 Hz
for 2b.3), obviously as a result of the much different HeMneP
angles involved (90ꢀ vs. 180ꢀ).
hydride ligands give
(JPH ¼ 20 Hz) at ꢁ18.02 ppm, comparable to the one observed for
the dppm complex 1b (
¼ ꢁ17.5 ppm, JPH ¼ 21 Hz).
a strongly shielded triplet resonance
2.3. Synthesis of [Mn2(m-H)2(CO)6(m-dmpm)] (1c)
d
The dihydride complex 1c can be prepared by following
a procedure analogous to that previously described for the tedip
and dppm complexes 1a,b [5,6], except that a different reducing
reagent gives the best results in this case. Thus, the chloride-
2.4. Reactions of the dihydride complexes 1aec with HSnPh3
The reactions of compounds 1aec with HSnPh3 are far less
bridged complex [Mn2(m-Cl)2(CO)6(m-dmpm)] is first reacted with
selective than the reactions with silanes just discussed, and they

1740
M.A. Alvarez et al. / Journal of Organometallic Chemistry 696 (2011) 1736e1748
were found to be strongly dependent on the reaction conditions
and the nature of the bridging phosphorus ligand. The Table 4
summarizes the distribution of products (based on yields of iso-
lated species) obtained under different experimental conditions.
The reactions of the tedip-bridged complex 1a with HSnPh3 give
three major species that can be separated through low temperature
chromatography: the stannylene complex [Mn2(
tedip)] (4a), the stannyleneestannyl complex [Mn2(
(H)SnPh2}(CO)6( -tedip)] (5a), and the hydrostannyl complex
[Mn2( -H)( -HSnPh2)(CO)6( -tedip)] (6a) (Chart 3). All of these
m
-SnPh2)2(CO)6(
m-
m
-H){ -Ph2SnO
m
m
m
m
m
products follow from the loss of a Ph group from the organotin
hydride, mostly eliminated as benzene, as shown by GCeMS
analysis of the corresponding reaction mixtures. As expected, the
formation of 6a is favored when using equimolar amounts of
reagents and if carrying the reaction at 273 K, although significant
amounts of this product are obtained even when using excess of
reagent. With the appropriate stoichiometry for ditin derivatives
(2:1), the major products are 4a and 5a if the reaction is carried out
in toluene solution, while compound 5a becomes dominant if using
THF as solvent. Since the formation of the latter complex requires
an external source of oxygen atoms, it is very likely for such
a source to be the traces of water present in the solvent, expected to
be higher in the more hygroscopic THF solvent (compared with
toluene). Under these conditions, another uncharacterized
compound was present in significant amounts in the reaction
mixture. The only significant spectroscopic feature associated to
1
this compound was the presence in the H spectrum of an AA0XX0
0
multiplet at ꢁ22.07 ppm (jJHP þ JHP j ¼ 18 Hz) due to hydride-like
ligands. No resonances for this species could be clearly identified in
the 31P{1H} spectrum of the reaction mixture, perhaps being
obscured by the quite broad resonance of the major product 5a.
Unfortunately, this product decomposed completely upon chro-
matographic workup of the reaction mixture and could not be
characterized.
The reactions of 1b with HSnPh3 are analogous to those of 1a,
except that no hydrostannyl compound isostructural to 6a was
formed at all. Instead, the related hydroxylstannyl complex [Mn2-
Chart 3.
erroneouslyidentifiedasanhydrostannylanalogofthetedipcomplex
6a. Actually, we have now been unable to identify such a stannyl
derivative of the dppm-bridged compound 1b under any conditions,
presumably due to the high reactivity to traces of water of the inter-
mediate species involved (see later).
The reactions of the dihydride 1c and HSnPh3 were also of low
selectivity (Table 4). The main differences with the reactions of 1a
and 1b just discussed are as follows: (a) no stannylene complex of
type 4 could be identified in significant amounts under any
conditions, (b) both hydrostannyl-bridged and hydroxylstannyl-
bridged complexes (5c and 6c) can be isolated in these reactions,
(m-H){m-O(H)SnPh2}(m-dppm)(CO)6] (7b) was obtained under com-
parable circumstances (Chart 3). Thus, the reaction between
stoichiometric amounts of 1b and HSnPh3 in toluene at 273 K gave 7b
as major product. Interestingly, a very small amount of a new type
of product was detected under these conditions. Fortunately, the
latter complex could be generated as the major product by carrying
out the reaction under a CO atmosphere, and it could be identified as
the stannyl heptacarbonyl complex [Mn2(m-H)(SnPh3)(m-dppm)
(CO)7] (9b). At this point we should note that in our preliminary study
of the above reactions [7], the hydroxylstannyl complex 7b was
(c) the novel hydroxide-bridged complex [Mn2(m-H)(m-OH)(m-
Table 4
dmpm)(CO)6] (8c) is formed in significant amounts if using THF as
solvent, and (d) under a CO atmosphere, no product other than
the heptacarbonyl complex 9c is formed in significant amounts
(Chart 3).
Product distribution in the reactions of compounds 1aec with HSnPh3.a
L2
Rr
T/K
solv
t/h
4
5
6
7
8
9
tedip
tedip
tedip
tedip
dppm
dppm
dppm
dppm
dmpm
dmpm
dmpm
1
1
2
2.5
1
1
2
1
2.5
1
293
273
293
293
293
273
293
293
293
273
293
Tol
Tol
Tol
THF
Tol
1
1
2
0.5
1
1
2
4
1
15
ta
40
ta
15
ta
15
50
10
20
70
18
20
Compounds 5 to 7 are formally related to each other through
some simple reactions, so we examined some possible conversions
(Scheme 1). From these we concluded that none of these products
are efficient precursors of any of the others, and therefore must be
formed through independent reaction pathways, to be discussed
later on. Thus, although the hydroxylstannyl compounds 7 are
formally derived from the addition of an oxygen atom to the
hydrostannyl compounds 6, the dmpm-bridged complex 6c did not
react rapidly with oxygen at room temperature, and only a gener-
alized decomposition was observed after several hours. As for the
tedip compound 6a, the reaction with oxygen also required several
hours for completion and yielded a mixture of several uncharac-
terized products. One of the minor species present in the mixture
22
10
70
20
25
25
82
15
Tol
ta
Tol
40
10
Tolb
THF
Tol
55
ta
10
ta
20
ta
20
2.5
2.5
1
THFb
75
a
Given in % of products isolated after column chromatography of the corre-
sponding reaction mixture; ta stands for trace amounts of product, Rr stands for the
molar ratio of HSnPh3 to compound 1 used in each experiment, and solv stands for
solvent, either toluene (Tol) or tetrahydrofuran (THF).
b
Reaction performed under a CO atmosphere.

M.A. Alvarez et al. / Journal of Organometallic Chemistry 696 (2011) 1736e1748
1741
Scheme 1. Hypothetical reactions relating compounds 5 to 7 [Mn2 ¼ Mn2(CO)6(
with L2 ¼ tedip, dppm, dmpm].
m-L2),
showed upfield 1H NMR resonances [
and
those of the hydroxylstannyl complex 7b (Table 3) and therefore it
d
ꢁ2.9 (s, JHSn ¼ 30 Hz,1H, OH)
d
ꢁ23.02 (t, JHP ¼ 31, JHSn z 91, 1H, MneHeMn)] comparable to
is likely to be the corresponding hydroxylstannyl complex
[Mn2(m-H){m- :k m-tedip)] (7a), although we
k1 1-O(H)SnPh2}(CO)6(
were not able to purify this minor product.
We also examined the possibility that the compounds of type 5
might be formed through the reaction of
6 with HOSnPh3
(a possible hydrolysis product of HSnPh3) or from the reaction of 7
with excess of HSnPh3 (Scheme 1). We carried out these tests with
the dmpm compounds 6c and 7c, but no reaction was observed at
room temperature in either case. We also examined the reaction of
the tedip-bridged complex 6a with an excess HOSnPh3, this now
requiring several days for completion at room temperature and
yielding a complex mixture containing compounds 5a, 7a and small
amounts of 4a. From all this we can conclude that neither 6 nor 7
are real precursors of the stannylestannylene complexes of type 5
that are formed rapidly in the reactions of the hydride complexes
1aec with HSnPh3.
Fig. 2. ORTEP diagram (30% probability) of compound 4a, with H atoms omitted for
clarity (taken from Ref. [7]). Selected bond lengths (Å) and angles (ꢀ): Mn(1)eMn0
(1)
¼
3.1045(5), Mn(1)eSn(1)
¼
2.6053(5), Mn0(1)eSn(1)
¼
2.6178(6), Mn(1)eP
(1) ¼ 2.2472(4); Mn(1)eSn(1)eMn0(1) ¼ 72.94(2), Sn(1)eMn(1)eSn0(1) ¼ 106.93(2), C
(10)eSn(1)eC(20) ¼ 98.6(1), P(1)eMn(1)eSn(1) ¼ 91.49(3).
and of a single methylenic resonance for the dppm compound 4b
(see the Experimental section).
2.5. Solid-state and solution structure of compounds 4
2.6. Solid-state and solution structure of compounds 5
The structure of the tedip-bridged complex 4a was confirmed in
our preliminary study by X-ray diffraction and it is depicted in the
Fig. 2. The molecule displays a rhombic planar arrangement of the
Crystals of medium quality for an X-ray diffraction study were
obtained after many unsuccessful attempts for compound 5a. An
ORTEP view of the molecule is shown in the Fig. 3, while the most
relevant bond distances and angles are collected in the Table 5. The
molecule in made up of two T-shaped Mn(CO)3 fragments con-
four metallic atoms similar to those found in the compounds [Re2(
m-
SnPh2)2(CO)8] [23], and [Mn2{ -Sn(Br)Mn(CO)5}2(CO)8] [24]. The
m
manganese atoms are also bridged by the tedip ligand, and
the slightly distorted octahedral coordination environment around
the manganese atoms is completed by three carbonyls in a local
facial arrangement. The intermetallic distance in 4a [3.1045(5) Å] is
similar tothose measured in the mentioned Mn2Sn2 cluster(3.086 Å,
nected through a significantly elongated intermetallic bond
(MneMn ¼ 3.179(4) Å) and bridged symmetrically by two ligands
in a pseudotrans arrangement: a tedip ligand, and a Ph2SnO(H)
SnPh2 bridge. Although the hydrogen atom on this hydroxyl group
was not located in the X-ray study, its presence is denoted by the
corresponding IR and 1H NMR data, to be discussed below. In the
same way, the spectroscopic data clearly indicate the presence in
this molecule of a hydride ligand bridging the manganese atoms
but this atom could not be located either, although its presence can
be guessed from the elongation of the MneMn separation noted
above (cf. MneMn: 2.9038(6) Å in [M2(CO)10], [27]). Both metal
fragments are significantly twisted (by some 27ꢀ) from each other,
thus defining a conformation intermediate between the staggered
and eclipsed conformations of M2L10 molecules, a feature not
unusual in diphosphine-bridged dimanganese complexes (cf. 24ꢀ
[24]) and in the Mn2Au cluster [Mn2(m-AuPPh3)(m-Br)(CO)6(m-
tedip)] (3.090(3)Å, [25]), in agreementwith the singleMneMn bond
to be formulated for this compound on the basis of the EAN rule. On
the otherhand, the average MneSn distance (ca. 2.61 Å) is within the
usual range found for single MneSn bonds (2.55e2.75 Å, [26]).
Spectroscopic data in solution for compounds 4a and 4b are
fully consistent with the highly symmetrical solid-state structure
just discussed. In particular, both compounds exhibit five strong
CeO stretching bands in the IR spectra (Table 2), as expected for
these C2v molecules [19] and display a single resonance in the
corresponding 31P{1H} NMR spectra (Table 3). The presence of
effective symmetry elements relating both sides of the averaged
Mn2P2 plane in each case is denoted in the 1H NMR spectra by the
appearance of a single methyl resonance for the tedip compound 4a
and in the dppm-bridged complex [Mn2(CO)6(m-dppm)2] [28]).
The molecule of 5a was found to crystallize with one molecule of
water for every two molecules of complex, they being interacting
through hydrogen bonds. Indeed, the distance O(7)/O(H2O) of

1742
M.A. Alvarez et al. / Journal of Organometallic Chemistry 696 (2011) 1736e1748
the Re2 anion mentioned above, and the same can be said about the
SneOeSn angles (133.9ꢀ in this Re2 complex, and 127.1(6)ꢀ in 5a).
This similarity is not easy to explain, because the SneO lengths in
a R2SneOeSnR2 group are expected to be significantly shorter than
the corresponding distances in a R2SneO(H)eSnR2 chain, due to
the loss of multiplicity in the SneO bonds in the second case [30].
For instance, the SneO lengths in Ph3SneOeSnPh3 are ca. 1.96 Å
[33a], that is, more than 0.2 Å shorter than the values in [Ph3Sn(m-
OH)]n (2.197(5) and 2.255(5) Å, [33b]). Thus the possibility of
a wrong formulation should not be excluded for this dirhenium
complex even if the 1H NMR spectrum did not show clear reso-
nances for OH or hydride groups at the time [32].
Spectroscopic data for 5a are consistent with its solid-state
structure and provide support for the hydrogen atoms (hydroxyl
and hydride) not located in the X-ray study. Thus, the presence of
the OH group is revealed by the appearance of a characteristic OeH
stretching band at 3340 cmꢁ1 in the IR spectrum of 5a when
recorded in a Nujol emulsion, and also by a 1H NMR resonance at
2.32 ppm. The presence of a hydride ligand symmetrically bridging
the manganese atoms is clearly denoted by the appearance of
a highly shielded triplet resonance at ꢁ24.4 ppm (JHP ¼ 13 Hz)
displaying satellites due to coupling with the 119Sn/117Sn nuclei
(JHSn ca. 85 Hz). Finally, we note that the IR spectrum of 5a en
dichloromethane solution shows five bands, with the CeO
symmetrical stretches above ca. 2000 cmꢁ1 being of weak intensity
(Table 2), in agreement with the retention of the T-shaped Mn(CO)3
fragments in solution.
Spectroscopic data for compounds 5b and 5c are comparable to
those of 5a, indicating that all these compounds are isostructural
(Tables 2, 3 and Experimental section). We just notice that the OH
NMR resonance for 5b displays tin satellites (JHSn ¼ 23 Hz), in
agreement with its positioning between two Sn atoms. We finally
should stress that the hydride ligands in all three compounds
display a relatively low PeH coupling of 13e15 Hz, a value compa-
rable to those measured in the silane complexes of type 2 (mer-
isomers), and significantly lower than the usual values for cis PeH
couplings in our dimanganese complexes (20e30 Hz). This suggests
that the hydride ligand in compounds 5 might be placed in the
average plane defined by the Mn, P, and Sn atoms. Interestingly, the
Fig. 3. ORTEP diagram (30% probability) of compound 5a, with Et and Ph groups
(except the ipso C atoms in the latter) omitted for clarity. Hydrogen atoms were not
located in this study.
2.89(1) Å is indicative of a weak hydrogen bond (O/O separations
in the range 2.70e2.90 Å for weak interactions, [29]). This seems to
be a fundamental element in the stabilization of the crystal lattice,
and justifies the unexpected water capture during the crystalliza-
tion process in dichloromethaneepetroleum ether mixtures.
The most remarkable structural feature in 5a is the presence of
the symmetrical ditin moiety Ph2SneO(H)eSnPh2 that can be
described as a 3-electron donor ligand resulting from the OeSn
coupling between hydroxylstannyl and stannylene fragments. To
our knowledge, this is a quite rare fragment of which we can quote
only a couple of precedents. Actually we have found no examples of
related hydroxyl-bridged groups bound to transition-metal centers,
but can quote the anionic complex [Re2{m-Ph2SnO(Me)SnPh2}
(CO)8]ꢁ, containing a related methoxyl-bridged ditin moiety [30].
Interestingly, the SneO lengths in this anion (ca. 2.13 Å) are almost
identical to those in 5a (ca. 2.12 Å). We can also quote the hydride
complex [Re2(m-H){m-Ph2GeO(H)GePh2}(CO)8] [30], a molecule
strictly comparable to compound 5a after replacing Ge by Sn and
two CO ligands by the tedip bridge. Interestingly, this complex is
formed in the reaction of a dirhenium precursor with HGePh3 in the
presence of water, in agreement with our experimental data dis-
cussed above. An Ir2 complex containing the same Ge2 ligand has
been recently reported as a low-yield product (4%) of the thermal
decomposition of [Ir(H)(GePh3)2(CO)3] [31]. It is worth of note that
this complex was found to crystallize with a water molecule
involved in hydrogen bonding with the Ph2GeO(H)GePh2 ligand.
The structure of compound 5a can be also related to that of
hydride ligand in [Re2(m-H){m-Ph2GeO(H)GePh2}(CO)8], a complex
isoelectronic with compounds 5, was successfully located in that
plane through an X-ray study, a position also consistent with the
results of DFT calculations on that molecule [30].
2.7. Structural characterization of compounds 6e8
The IR spectra of compounds 6a,c show the typical five to six
strong CeO band pattern usually observed in many dimanganese
compounds of the type [Mn2(m-H)(m-X)(CO)6(m-L2)] having two fac-
a complex formulated as [Re2(m-Bu2SnOSnBu2)(CO)8] after an X-ray
diffraction study [32]. Surprisingly, the SneO length in this complex
(ca. 2.13 Å) is almost identical to the values measured for 5a or for
M(CO)3 oscillators, with lower frequencies for the dmpm-bridged
complex 6c, as expected. The inequivalence of both manganese
centers is apparent from the corresponding 31P NMR spectra,
showing two separated resonances corresponding to an AX (6a) or
an AB spin system (6c). For complex 6a, only the more deshielded
resonance displays PeSn satellites (JPSn ca. 414 Hz) and is thus
assigned to the P atom placed further away from the MnHSn edge
(P1 in Table 3), and the same is assumed for compound 6c.
Table 5
Selected bond lengths (Å) and angles (ꢀ) for compound 5a.
Mn(1)eMn(2)
Sn(1)eMn(1)
Sn(1)eO(7)
Mn(1)eP(1)
Mn(1)eC(1)
Mn(1)eC(2)
Mn(1)eC(3)
Sn(2)eMn(2)
Sn(2)eO(7)
Mn(2)eP(2)
Mn(2)eC(4)
Mn(2)eC(5)
Mn(2)eC(6)
3.179(4)
2.549(3)
2.12(1)
2.180(5)
1.75(2)
1.80(2)
1.83(2)
2.539(3)
2.11(1)
2.171(6)
1.76(2)
1.78(2)
1.82(2)
P(1)eMn(1)eMn(2)
P(2)eMn(2)eMn(1)
Sn(1)eMn(1)eMn(2)
Sn(2)eMn(2)eMn(1)
P(1)eMn(1)eSn(1)
P(2)eMn(2)eSn(2)
C(1)eMn(1)eMn(2)
C(1)eMn(1)eC(2)
C(4)eMn(2)eMn(1)
C(4)eMn(2)eC(5)
Sn(2)eO(7)eSn(1)
O(7)eSn(1)eMn(1)
O(7)eSn(2)eMn(2)
83.0(2)
83.2(2)
94.3(1)
95.5(1)
The most remarkable feature in the 1H NMR spectra of
compounds 6 is the presence of two hydride-like resonances in
each case. One of them appears as a strongly shielded triplet
resonance at ca. ꢁ18 ppm with a characteristic cis HeP coupling of
ca. 25 Hz, and is therefore assigned to a hydride ligand bridging
a MneMn edge in an essentially symmetrical way. The second
resonance is less shielded (ca. ꢁ9.5 ppm) and displays coupling to
a single phosphorus atom, and is therefore assigned to a hydrogen
177.2(2)
178.1(2)
176.7(6)
89.5(8)
174.3(7)
93.5(9)
127.1(6)
108.0(3)
105.2(3)

M.A. Alvarez et al. / Journal of Organometallic Chemistry 696 (2011) 1736e1748
1743
atom bridging a SneMn edge, therefore implying the presence of
a tricentric, agostic-like SneHeMn interaction. An unusual feature
of these resonances in compounds 6 is the unexpectedly low HeSn
couplings of ca. 95 Hz, actually lower than the two-bond couplings
of ca. 130 Hz observed for the hydride resonances in the same
compounds. These one-bond couplings are considerably lower than
those measured in related complexes having tricentric SneHeM
interactions, such as [CpMn(CO)2(k
2-HSnPh3)] (ca. 260 Hz, [34])
and [Os3(
m
-H)(
m
-
k1
:
k
2-HSnR2)(CO)10] (ca. 280 Hz, R ¼ CH(SiMe3)2,
[35]). This might be indicative in our case of the presence of
a SneHeMn interaction in the limit of the full cleavage of the
corresponding SneH bond, close to the point of yielding bridging
stannylene and terminal hydride ligands. Unfortunately, all
attempts to obtain X-ray quality crystals of compounds 6a,c (to
confirm this hypothesis) were unsuccessful.
The structure proposed for compounds 7 (Chart 3) is based on
the corresponding spectroscopic data and also on an incomplete
X-ray diffraction study of the dppm complex 7b. Although the
diffraction data were of very low quality and did not allow for
a satisfactory refinement of the structure, the gross structural
features of the molecule were clearly defined: two facial Mn(CO)3
fragments bridged by the dppm ligand with a SnPh2 group asym-
Chart 4.
2.8. Solid-state and solution structure of compounds 9
As discussed below, the compounds 9b,c are present in solution
as mixtures of three isomers that could not be separated from each
other (A to C in Chart 4). Fortunately, one of these isomers is the
dominant species in solution for the dppm-bridged complex 9b, and
it could be crystallized and characterized through an X-ray diffrac-
tion study. The molecular structure of this isomer (Fig. 4 and Table 6,
corresponding to isomer B in Chart 4) can be viewed as derived from
thatof [Mn2(CO)10] afterreplacing two equatorial carbonyls with the
dppm bridging ligand and one more equatorial carbonyl with
a SnPh3 group, trans to phosphorus, with the two metal fragments
being twisted from each other by some 23ꢀ. As found for the Mn2Sn2
hydride compound 5a, the MneMn separation in 9b is relatively
large (3.2370(9) Å), but now the hydride ligand could be located, it
being somewhat asymmetrically placed over that vector
(MneH ¼ 1.61(3), 1.70(3) Å), and positioned cis to the P atoms
(HeMneP ca. 94ꢀ). The axial CO ligands are not colinear with the
intermetallic vector, but rather the corresponding MneC vectors
point to the center of the Mn2H triangle, as usually found for tri-
metrically bridging the Mn atoms (MneMn
¼
3.26 Å,
SneMn ¼ 2.44 Å and 3.27 Å). While the short distance is a bit
shorter than expected for a MneSn bond, the long one is clearly out
of the usual range for MneSn bonds (2.55e2.75 Å, [26]). No clear
indication of the presence of an oxygen atom in the vicinity of these
metal atoms was obtained with these low-quality data. We note,
however, that the related Re2Si complexes [Re2(m-H)(m- :k
k1 1-O(Me)
Si(OMe)2)(CO)7(L)] (L ¼ CO, PMe2Ph) have been characterized
structurally [36] and indeed display Re2SiO rings very similar to the
Mn2SnO ring proposed to be present in compounds 7.
The IR spectra of compounds
7 are similar to those of
compounds 6, supporting the proposal of a similar arrangement of
the CO and L2 ligands, while the latter ligands give rise also to
separate 31P NMR resonances in each case. As found for compounds
6, a strongly shielded resonance at ca. ꢁ21 ppm with the usual cis
PeH couplings is present in the 1H NMR spectra of these
compounds, corresponding to a hydride ligand bridging the Mn
atoms. However, the second resonance in the hydride region of the
spectra of compounds 7 appears as a much more deshielded
resonance at ca. ꢁ2 ppm and exhibits no coupling to phosphorus.
Therefore this resonance is assigned to the hydrogen atom of an
hydroxyl group bridging the Sn and Mn atoms. In agreement with
this proposal, we note that these resonances display a low PeSn
coupling of ca. 30 Hz, close to the value of 21 Hz measured for the
SnO(H)Sn resonance in compound 5b.
centric M2(m-H) interactions of the closed type [38]. Finally, the
SnPh3 group displays no remarkable features, with the MneSn bond
length of 2.6415(7) Å being comparable to the values of 2.63e2.69 Å
measured for several complexes of the type [Mn(SnR3)(CO)5] [39].
The proposed structure for 8c is analogous to the numerous
compounds of the type [Mn2(
m-H)(
m-ER)(CO)6(m-L2)] (E ¼ O, S;
L2 ¼ tedip, dppm) previously synthesized from the reactions of the
dihydrides 1a,b with different alcohols, thiols, aldehydes and
ketones [37], as judged from the similarity in the corresponding
spectroscopic data (Tables 2, 3 and Experimental section) and need
not to be discussed in detail. We just note that, in addition to the 1H
NMR triplet resonance at ꢁ11.86 ppm (JHP ¼ 21 Hz) corresponding
to the bridging hydride ligand, this complex gives rise to a second
triplet resonance substantially more deshielded (
d
¼ ꢁ2.49 ppm)
and displaying much lower coupling to phosphorus (5 Hz), which is
therefore assigned to the H atom of the bridging hydroxide ligand.
We note the similarity between these NMR parameters and those of
the OH group in the hydroxylstannyl compounds 7 described
above. Finally we should also stress that, although compound 8c
might be viewed as the product resulting from the reaction of 1c
with water, this reaction actually does not take place at a significant
extent either in this case or for the related dihydrides 1a,b.
Fig. 4. ORTEP diagram (30% probability) of compound 9b, with H atoms (except the
hydride ligand) and Ph groups (except the ipso C atoms) omitted for clarity.

1744
M.A. Alvarez et al. / Journal of Organometallic Chemistry 696 (2011) 1736e1748
Table 6
Selected bond lengths (Å) and angles (ꢀ) for compound 9b.
Mn(1)eMn(2)
Mn(1)eSn(1)
Mn(1)eP(1)
Mn(1)eC(1)
Mn(1)eC(2)
Mn(1)eC(3)
Mn(1)eH(1)
Mn(2)eP(2)
Mn(2)eC(4)
Mn(2)eC(5)
Mn(2)eC(6)
Mn(2)eC(7)
Mn(2)eH(1)
P(1)eC(8)
3.2370(9)
2.6415(7)
2.275(1)
1.784(4)
1.822(5)
1.819(5)
1.70(3)
2.307(1)
1.862(5)
1.841(5)
1.844(5)
1.810(5)
1.61(3)
Sn(1)eMn(1)eMn(2)
Sn(1)eMn(1)eP(1)
Sn(1)eMn(1)eC(1)
Sn(1)eMn(1)eC(2)
C(1)eMn(1)eMn(2)
C(1)eMn(1)eH(1)
C(4)eMn(2)eMn(1)
C(4)eMn(2)eP(2)
C(5)eMn(2)eP(2)
C(7)eMn(2)eMn(1)
C(7)eMn(2)eH(1)
Mn(1)eH(1)eMn(2)
H(1)eMn(1)eSn(1)
H(1)eMn(1)eP(1)
H(1)eMn(2)eP(2)
98.99(2)
172.4(1)
84.1(1)
84.6(1)
173.1(1)
173(2)
92.7(1)
90.6(1)
175.9(1)
175.1(1)
172(2)
155(3)
89(2)
1.837(4)
1.836(4)
93(2)
94(2)
P(2)eC(8)
The spectroscopic data in solution for 9b (Tables 2 and 3, and
Experimental section) are consistent with the structure found in
the crystal, thus suggesting that the solid-state molecule is likely to
be the major isomer in solution (B in Chart 4). Indeed the other two
isomers (A and C) are expected to be significantly less stable for 9b
because of the steric repulsions derived from the close proximity of
the bulky SnPh3 and PPh2 groups in these cases, and accordingly
they are present only in very tiny amounts in solution, as judged
from the appearance, in the 1H NMR spectra of 9b, of very weak
hydride resonances at ꢁ19.2 and ꢁ22.2 ppm, in addition to the
hydride resonance of the major isomer, with the latter appearing
at ꢁ24.84 ppm and being coupled differently to the inequivalent P
nuclei (JPH ¼ 14 and 21 Hz) as expected.
Scheme 2. Reaction pathways proposed for the reactions of compounds 1aec with
HSnPh3 [Mn2 ¼ Mn2(CO)6( -L2), with L2 ¼ tedip, dppm, dmpm].
m
dehydrogenation, or through benzene elimination. In the first case,
a 32 electron intermediate B (not detected) with bridging stannyl
and hydride ligands would be formed, that would spontaneously
react with CO (formed in small amounts through decomposition) to
give the heptacarbonyl complexes of type 9. Apparently, this is not
a dominant pathway unless CO is deliberately added to the solu-
tion. Yet, we should remark that the sequence 1 / B / 9 is
completely parallel to the recently reported reaction between the
In contrast, isomers A and C are expected to be less disfavored in
the presence of a sterically less demanding diphosphine as dmpm.
Indeed, the hydride area of the 1H NMR spectrum of 9c reveals at
room temperature the presence of three resonances with about
the same relative intensity at ca. ꢁ19.3, ꢁ23.2 and ꢁ24.7 ppm.
A detailed analysis of these spectra, either at room or low
temperature, with the additional help of some selective 31P
decoupling experiments and of a standard 2D-EXSY experiment,
allowed the assignment of the different resonances (see the
Experimental section). The latter experiment also revealed that
isomers A and B convert one into each other but not with the most
asymmetric isomer C (or not so rapidly), and that the hydride
resonance of isomer B (but not that from A) displays a significant
NOE with the ortho hydrogens of the phenyl rings, as expected from
the close proximity of the bridging hydride and SnPh3 groups in
that isomer (Chart 4).
30-electron dimolybdenum hydride [Mo2Cp2(
and HSnPh3 to give the stannyl-bridged complex [Mo2Cp2(
-PCy2)(CO)2] (characterized through and X-ray study), an unsat-
m-H)(
m-PCy2)(CO)2]
m-SnPh3)
(m
urated molecule also reacting with CO to give an electron-precise
derivative having a terminally-bound SnPh3 group [11].
The prevalent evolution of the intermediate A, however, would
be the elimination of a benzene molecule (as noted above, C6H6 has
been actually detected in these reaction mixtures), to produce an
unsaturated hydride C, similar to the intermediate B mentioned
above. However, the presence of a SneH bond in C would now
enable this molecule to reach electronic saturation by a simple
intramolecular rearrangement involving the coordination of the
HeSn bond to a manganese atom, thus leading to the hydrostannyl-
bridged compounds of type 6. Yet, in the presence of an excess of
HSnPh3, the intermediate C would rather add a second molecule of
reactant to give the species D, isoelectronic with A. This might be
followed by new processes of elimination of benzene and hydrogen
(proposing an exact sequence of events here would be exceedingly
speculative) to finally yield the stannylene-bridged compounds of
type 4.
2.9. Reaction pathways in the formation of compounds 4e9
From the results discussed above it is clear that, unlike the
reactions of the unsaturated hydrides 1 with silicon hydrides, those
with HSnPh3 are rather complex events involving several reaction
pathways. Although we have detected no intermediate species in
these reactions, we can make reasonable proposals to account for
the formation of most of the observed products (Scheme 2).
In the first place it is reasonable to assume that, because of the
electronic unsaturation of compounds
1 and their credited
ligandeacceptor behavior, the first step in these reactions would be
the addition of a HSnPh3 molecule (by establishing an agostic-like
SneHeMn interaction) to one of the manganese atoms, with a
concerted movement of one of the bridging hydride ligands into
a terminal position, to give an electron-precise intermediate
A (Scheme 2). This process would be comparable to the addition of
“normal” two-electron donors such as phosphines to the unsatu-
The formation of compounds 5, 7 and 8 require the incorpora-
tion of an oxygen atom at some of the intermediate steps just
described, possibly through reactions with traces of water present
in the solvent, as noted above. As a result, an hydroxyl group is
incorporated into the reaction products bridging either two tin
atoms (5), two manganese atoms (8) or manganese and tin atoms
(7). However, we have no experimental data to identify or make
a sensible guess about the precise steps for the incorporation of this
extra oxygen atom in each case.
rated dirhenium hydrides [Re2(
m-H)2(CO)6(m-L2)] [39]. Intermediate
A
could now evolve in two different ways, either through

M.A. Alvarez et al. / Journal of Organometallic Chemistry 696 (2011) 1736e1748
1745
3. Experimental
as described for 2a.1 yielded compound 2b.1 as a microcrystalline
yellowsolid(0.040g,79%). Anal. Calcd.forC43H34Mn2O6P2Si: C, 61.00;
All reactions were carried out under a nitrogen atmosphere
using standard Schlenk techniques. Solvents were purified
according to literature procedures [40], and distilled under
nitrogen prior to use. Petroleum ether refers to that fraction
distilling in the range 338e343 K. The hydride complexes
H, 4.05. Found: C, 61.19; H, 4.35. 1H NMR:
(t, JHP ¼ 10, 2H, CH2), ꢁ13.09 [false t, jJHP þ JHP j ¼ 12, 2H, MneHeSi].
d
8.1e6.5 (m, 30H, Ph), 3.99
0
3.4. Reaction of compound 1a with H2SiPhMe
[Mn2(
m
-H)2(CO)6(
m
-L2)] (L2
¼
(EtO)2POP(OEt)2, tedip (1a);
Compound 1a (0.050 g, 0.093 mmol) and H2SiPhMe (60 mL,
0.44 mmol) were stirred in petroleum ether (5 mL) at room temper-
ature for 15 h. Workup as described for 2a.1 gave an oily yellow
L2 ¼ Ph2PCH2PPh2, dppm (1b)) were prepared according to litera-
ture procedures [15,6b]. The complex [Mn2(
m
-Cl)2(CO)6(
m
-dmpm)]
-dmpm)]
(dmpm ¼ Me2PCH2PMe2) was prepared from [Mn2(CO)8(
m
residue containing the isomers mer-[Mn2(
tedip)] (2a.2) and fac-[Mn2( -H2SiMePh)(CO)6(
a ratio5:1. Data for 2a.2: 1H NMR:
7.93 (m, 2H, Ph), 7.39e7.27 (m, 3H,
m
-H2SiMePh)(CO)6(
m-
[41], by following the procedure described previously for the
related dppm-bridged complex [42]. All other reagents were
obtained from the usual commercial suppliers and used as received.
Chromatographic separations were carried out using jacketed
m
m-tedip)] (3a.2) in
d
Ph), 4.02 (m, 8H, OCH2), 1.23, 1.16 (2t, JHH ¼ 7, 2 ꢂ 6H, Me), 1.08 (s,
0
3H, SiMe), ꢁ12.42 [false t, jJHP þ JHP j ¼ 14, 2H, MneHeSi]. Data for
columns, kept at the desired temperature with
a
cryostat.
3a.2: 1H NMR:
isomer were obscured by those of the major isomer.
d
ꢁ11.02 [br, MneHeSi]; other resonances for this
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.
IR stretching frequencies of CO ligands were measured either in
solution (using CaF2 windows) or in Nujol mulls (using NaCl
3.5. Reaction of compound 1a with H3SiPh
Compound 1a (0.050 g, 0.093 mmol) and H3SiPh (116
mL,
windows) and are referred to as
(NMR) spectra were routinely recorded at 300.13 (1H), 121.50 (31
{1H}) and 75.47 MHz (13C{1H}) at 290 K in C6D6 solutions unless
n
(CO). Nuclear magnetic resonance
0.93 mmol) were stirred in toluene (5 mL) at room temperature for
2 h. Workup as described for 2a.1 gave an oily yellow residue
P
containing the isomers mer-[Mn2(
and fac-[Mn2( -H2SiPhH)(CO)6( -tedip)] (3a.3) in a ratio 1:6. Data
for 3a.3: 1H NMR (CD2Cl2):
m-H2SiPhH)(CO)6(m-tedip)] (2a.3)
otherwise stated. Chemical shifts (
d
) are given in ppm, relative to
m
m
internal tetramethylsilane (1H and 13C), and external 85% aqueous
d
7.8e7.2 (m, 5H, Ph), 5.85 (t, JHP ¼ 12.5,
H3PO4 solutions (31P). Coupling constants (J) are given in Hertz.
1H, SieH), 3.99 (m, 8H, OCH2), 1.32 (m, 12H, Me), ꢁ11.58 [false t,
jJHP þ JHP j ¼ 23, 2H, MneHeSi]. Data for 2a.3: 1H NMR (CD2Cl2):
0
0
3.1. Preparation of [Mn2(m-H)2(CO)6(m-dmpm)] (1c)
d
ꢁ12.96 [false t, jJHP þ JHP j ¼ 12, 2H, MneHeSi]; other resonances
for this isomer were obscured by those of the major isomer.
A solution of Li[BH2Me2] (10 mL of a 0.5 M solution in petroleum
ether, 5.0 mmol) was slowly added to a dry THF solution (15 mL) of
3.6. Preparation of compound fac-[Mn2(m-H2SiPhH)(CO)6(m-dppm)]
[Mn2(m-Cl)2(CO)6(m-dmpm)] (0.500 g, 1.03 mmol) at room temper-
(3b.3)
ature, and the mixture was stirred for 3 h to give a red-greenish
solution. Degassed water (1 mL, excess) was then added to the
solution, the mixture was stirred for 2 min and the solvent was then
removed under vacuum. The red purple solid thus obtained was
dissolved in the minimum amount of dichloromethane and chro-
matographed at 243Kon an alumina column(activity 2, 30 ꢂ 2.5 cm)
prepared in petroleum ether. Elutionwith tolueneepetroleum ether
(1:1) gave a purple fraction yielding, after removal of solvents under
vacuum, complex 1c as a purple solid (0.302 g, 70%). Anal. Calcd. for
C11H16Mn2O6P2: C, 31.75; H, 3.88. Found: C, 31.81; H, 3.86. 1H NMR
0
Compound 1b (0.040 g, 0.060 mmol) and H3SiPh (40
0.32 mmol) were stirred in toluene (5 mL) at 323 K for 2 h. Workup
mL,
as described for 2a.1 yielded complex 3b.3 as an oily yellow residue.
1H NMR (400.13 MHz):
d
7.8e6.6 (m, 25H, Ph), 5.89 (t, JHP ¼ 12.5,1H,
SieH), 3.39, 3.00 (2 ꢂ dt, JHH ¼ 15, JHP ¼ 11, 2H, CH2), ꢁ10.22 [br, 2H,
MneHeSi].
3.7. Preparation of [Mn2(m-SnPh2)2(CO)6(m-tedip)] (4a)
(400.13 MHz):
d
0.95 [false t, jJHP þ JHP j ¼ 7.5, 12 H, Me], 0.52 (t,
-H).
Compound 1a (0.100 g, 0.186 mmol) and HSnPh3 (0.120 g,
0.372 mmol) were stirred in toluene (20 mL) for 2 h to give a yellow
solution. Solvent was then removed from the solution under vacuum,
and the oily residue was dissolved in petroleum ether and chroma-
tographed on an alumina column (activity 4) at 243 K. Elution with
the same solvent gave a yellow fraction yielding, after removal of the
solvent under vacuum, compound 6a as a yellow solid (0.027 g, 18%).
Elution with dichloromethaneepetroleum ether (1:8) gave another
yellow fraction, yielding analogously 0.015 g (15%) of compound 5a.
Elution with a 1:4 solvent mixture gave finally a third yellow fraction,
yielding, after removal of solvents, compound 4a as a yellow solid
(0.040 g, 40%). Data for 4a: Anal. Calcd. for C38H40Mn2O11P2Sn2: C,
JHP ¼ 11, 2H, CH2), ꢁ18.02 (t, JHP ¼ 20, 2H,
m
3.2. Preparation of mer-[Mn2( -H2SiPh2)(CO)6(m-tedip)] (2a.1)
m
Compound 1a (0.050 g, 0.093 mmol) and H2SiPh2 (170
mL,
0.93 mmol) were stirred in toluene (5 mL) at room temperature for
8 h. The resulting yellow solution was filtered and the solvent was
removed under vacuum. The residue was then washed with cold
petroleum ether (253 K, 2 ꢂ 2 mL) and dried under vacuum to give
compound 2a.1 as a yellow microcrystalline solid (0.057 g, 85%).
The crystals used in the X-ray study were grown by the slow
diffusion of a layer of petroleum ether into a concentrated toluene
solution of the complex. Anal. Calcd. for C26H32Mn2O11P2Si: C,
42.18; H, 3.73. Found: C, 42.03; H. 3.81. 1H NMR (CDCl3):
d 8.02 (m,
4H, Ph), 7.9 (m, 4H, Ph), 7.68e7.16 (m, 12H, Ph), 3.75 (m, 4H, OCH2),
3.3 (m, 4H, OCH2), 1.01 (t, JHH ¼ 7, 12H, Me).
43.35; H, 4.48. Found: C, 43.42; H. 4.70. 1H NMR:
d 7.79 (false d,
JHH ¼ 7, 4H, Ph), 7.59e7.00 (m, 6H, Ph), 3.83 (m, 8H, OCH2), 0.99 (t,
0
JHH ¼ 7, 12H, Me), ꢁ12.10 [false t, jJHP þ JHP j ¼ 15, 2H, MneHeSi].
3.8. Preparation of [Mn2(m-H){m-Ph2SnO(H)SnPh2}(CO)6(m-tedip)]
(5a)
3.3. Preparation of mer-[Mn2(m-H2SiPh2)(CO)6(m-dppm)] (2b.1)
Compound 1a (0.100 g, 0.186 mmol) and HSnPh3 (0.163 g,
0.465 mmol) were stirred in tetrahydrofuran (15 mL) for 25 min to
give an orange solution that was filtered. The solvent was then
Compound 1b (0.040 g, 0.06 mmol) and H2SiPh2 (110
0.60 mmol) were stirred in toluene (5 mL) at 328 K for 1.5 h. Workup
mL,

1746
M.A. Alvarez et al. / Journal of Organometallic Chemistry 696 (2011) 1736e1748
removed from the solution under vacuum and the oily reddish
residue was dissolved in petroleum ether and chromatographed on
an alumina column (activity 3) at 253 K. Elution with dichlor-
omethaneepetroleum ether (1:8) gave a yellow fraction yielding,
after removal of solvents, 0.030 g (20%) of compound 6a. Elution
with a 1:1 solvent mixture gave another yellow fraction yielding,
after removal of solvents, compound 5a as a yellow solid (0.102 g,
50%). The crystals used in the X-ray diffraction study were grown by
slow diffusion of a layer of petroleum ether into a dichloromethane
solution of the complex at 253 K. Anal. Calcd. for
C38H42Mn2O12P2Sn2: C, 41.49; H, 3.85. Found: C, 41.80; H, 3.94. 1H
JHP ¼ 12, 2H, CH2), ꢁ2.08 (s, J119SnH z J117SnH ¼ 32, 1H, OH), ꢁ20.40
[dd, JHP ¼ 22, 37, 1H, -H].
m
3.12. Reaction of compound 1c with excess HSnPh3
Compound 1c (0.080 g, 0.19 mmol) and HSnPh3 (0.182 g,
0.52 mmol) were stirred in THF (15 mL) at room temperature for 1 h
to give a reddish solution. Solvent was then removed from the
solution under vacuum and the oily residue was then dissolved in
petroleum ether and chromatographed on an alumina column
(activity 2.5) at 253 K. Elution with dichloromethaneepetroleum
ether (3:7) gave a yellow fraction yielding, after removal of solvents,
NMR (CDCl3):
(s, OH), 1.38 (t, JHH
d
7.6e7.1 (m, 20H, Ph), 4.10 (m, 8H, OCH2), 2.32
7, 12H, Me), ꢁ24.4 (t, JHP 13,
220.3 (m, 2CO),
¼
¼
0.025
g
(20%) of compound 6c. Elution with dichlor-
J
119SnH z J117SnH ¼ 85,
m
-H). 13C{1H} NMR (CDCl3):
d
omethaneepetroleum ether (2:3) gave an orange fraction yielding
analogously compound 5c as a orange solid (0.020 g, 10%). Elution
with a 1:1 solvent mixture gave a yellow fraction yielding compound
7c as a dark yellow solid (0.035 g, 25%). Then an orange fraction was
eluted with a 3:1 solvent mixture, containing a trace amount of
compound 9c that was discarded. Finally, elution with neat
dichloromethane gave a yellow fraction yielding compound 8c as
a yellow solid (0.015 g, 20%). Data for 5c: Anal. Calcd. for
C35H36Mn2O7P2Sn2: C, 42.99; H, 3.71. Found: C, 42.63; H, 3.37. 1H
218.7 (m, 4CO), 149.0 [s, C1(Ph)], 135.6 [s, C3(Ph)], 128.9 [s, C4(Ph)],
128.3 [s, C2(Ph)], 61.3 (s, OCH2), 16.0 (s, Me).
3.9. Preparation of [Mn2(
m
-SnPh2)2(CO)6(m-dppm)] (4b) and
[Mn2( -H){ -Ph2SnO(H)SnPh2}(CO)6(m
m
m
-dppm)] (5b)
Compound 1b (0.100 g, 0.150 mmol) and HSnPh3 (0.106 g,
0.300 mmol) were stirred in toluene (20 mL) at room temperature
for 2 h to give an orange solution. The solvent was then removed
under vacuum and the resulting oily residue was extracted with
dichloromethaneepetroleum ether (1:9), and the extracts were
chromatographed on an alumina column (activity 4) at 243 K.
Elutionwith petroleum ether gave a yellow-orange fractionyielding,
after removal of solvents under vacuum, compound 7b as an orange
solid (0.028 g, 20%). Elution with dichloromethaneepetroleum
ether (1:6) gave a second orange fraction yielding analogously
compound 5b as an orange solid (0.018 g, 10%). Finally, elution with
dichloromethaneepetroleum ether (1:4) gave another orange frac-
tionyielding compound 4b as an orange solid (0.072 g, 40%). Data for
4a: Anal. Calcd. for C55H42Mn2O6P2Sn2: C, 54.68; H, 3.50. Found: C,
NMR (CD2Cl2):
d
7.60e7.20 (m, 20H, Ph), 2.70 (s, 1H, OH), 2.75
0
(t, JHP ¼ 10, 2H, CH2), 1.77 (false t, jJHP þ JHP j ¼ 7, 12H, Me), ꢁ25.8
[t, JHP ¼ 14, 1H,
m
-H]. 1H NMR:
d
7.50e7.30 (m, 20H, Ph), 2.10 (s, 1H,
0
OH), 1.72 (t, JHP ¼ 11, 2H, CH2), 1.10 (false t, jJHP þ JHP j ¼ 7, 12H, Me),
ꢁ25.5 [t, JHP ¼ 13, J119SnH z J117SnH ¼ 90,
37.4 (s, -dmpm, 2P). Data for 6c: Anal. Calcd. for C23H26Mn2O6P2Sn:
C, 40.10; H, 3.80. Found: C, 39.87; H, 3.85. 1H NMR (200.13 MHz):
m
-H]. 31P{1H} NMR (CD2Cl2):
d
m
d
8.04e7.05 (m,10H, Ph), 0.84 (d, JHP ¼ 7, 3H, Me), 0.81 (d, JHP ¼ 7, 6H,
Me), 0.6 (d, JHP ¼ 7, 3H, Me), 0.70 (m, 1H, CH2), 0.29 (td, JHP ¼ 13,
JHH ¼ 12, 1H, CH2), ꢁ9.43 (d, JHP ¼ 9.4, J119SnH z J117SnH ¼ 93, 1H,
SneHeMn), ꢁ18.2 [t, JHP ¼ 23, J119SnH z J117SnH ¼ 142, MneHeMn].
Data for 8c: Anal. Calcd. for C11H16Mn2O7P2: C, 30.58; H, 3.73. Found:
54.31; H. 3.35. 1H NMR (CDCl3):
JHP ¼ 11, 2H, CH2). Data for 5b: Anal. Calcd. for C55H44Mn2O7P2Sn2: C,
53.87; H, 3.62. Found: C, 54.19; H, 3.87. 1H NMR (CDCl3):
7.50e7.10
(m, 40H, Ph), 3.9 (t, JHP ¼ 10, 2H, CH2), 2.5 (s, J119SnH z J117SnH ¼ 21,1H,
OH), ꢁ25.9 (t, JHP ¼ 15, J119SnH z J117SnH ¼ 75, 1H, -H).
d
7.60e6.97 (m, 40H, Ph), 3.07 (t,
C, 30.45; H, 3.66. 1H NMR (200.13 MHz):
d 0.79 (false t,
0
0
jJHP þ JHP j ¼ 6.4, 6H, Me), 0.67 (false t, jJHP þ JHP j ¼ 7.2, 6H, Me), 0.1 (q,
d
JHH ¼ JHP ¼ 13, 1H, CH2), ꢁ0.26 (dt, JHH ¼ 13, JHP ¼ 11, 1H, CH2), ꢁ2.49
(t, JHP ¼ 5, 1H, OH), ꢁ11.86 (t, JHP ¼ 21, 1H,
m
-H).
m
3.13. Preparation of [Mn2(m-H){m-O(H)SnPh2} (CO)6(m-dmpm)] (7c)
3.10. Preparation of [Mn2( -H)( -HSnPh2)(CO)6(m-tedip)] (6a)
m
m
Compound 1c (0.036 g, 0.085 mmol) and HSnPh3 (0.030 g,
0.085 mmol) were stirred in toluene (15 mL) at 273 K for 2.5 h to
give a yellow solution. Solvent was then removed from the solution
under vacuum and the oily residue was dissolved in petroleum
ether and chromatographed on an alumina column (activity 2.5) at
253 K. Elution with dichloromethaneepetroleum ether (2:3) gave
a yellow fraction containing very small amounts (<5%) of complex
5c. Elution with a 1:1 solvent mixture gave another yellow fraction
yielding, after removal of solvents under vacuum, compound 7c as
a yellow solid (0.050 g, 82%). Anal. Calcd. for C23H26Mn2O7P2Sn: C,
Compound 1a (0.100 g, 0.186 mmol) and HSnPh3 (0.060 g,
0.186 mmol) were stirred in toluene (20 mL) at 273 K for 1 h to give
a yellow solution. The solvent was then removed from the solution
under vacuum, and the oily residue was dissolved in petroleum
ether and chromatographed on an alumina column (activity 4) at
243 K. Elution with the same solvent gave a yellow fraction yielding,
after removal of the solvent under vacuum, compound 6a as
a yellow solid (0.102 g, 68%). Anal. Calcd. for C26H32Mn2O11P2Sn: C,
38.50; H, 3.98. Found: C, 38.85; H, 4.05. 1H NMR (CDCl3):
d
7.70e7.30 (m, Ph, 4H), 7.30e7.10 (m, 6H, Ph), 4.03, 3.70, 3.40, 3.20
39.19; H, 3.72. Found: C, 38.87; H, 4.04. 1H NMR:
d 7.90e7.10
(4m, 4 ꢂ 2H, OCH2), 1.36, 1.34, 1.06, 0.97 (4t, JHH ¼ 7, 4 ꢂ 3H, Me),
ꢁ9.49 (d, JHP ¼ 29, J119SnH z J117SnH ¼ 97, 1H, MneHeSn), ꢁ18.40 (t,
JPH ¼ 28, J119SnH z J117SnH ¼ 124, 1H, MneHeMn). 31P{1H} NMR
(m, 10H, Ph), 1.68 (dt, JHH ¼ 13, JHP ¼ 9, 1H, CH2), 0.98e0.84 (m, 13H,
Me and CH2), ꢁ2.95 (s, J119SnH z J117SnH ¼ 26, 1H, OH), ꢁ21.75
(t, JHP ¼ 26, J119SnH z J117SnH ¼ 120, 1H,
m
-H). 1H NMR (CD2Cl2):
7.60e7.30 (m, 10H, Ph), 2.30e2.10 (m, ABMX, 2H, CH2), 1.70e1.30
(CDCl3, 243 K):
d
175.5 (d, JPP ¼ 88, J119SnP z J117SnP ¼ 415, 1P,
d
m-tedip), 165.3 (d, JPP ¼ 88, 1P,
m-tedip).
(m, 12H, Me), ꢁ1.91 (s, 1H, OH), ꢁ21.80 (t, JHP ¼ 26, 1H, -H).
m
3.11. Preparation of [Mn2( -H){
m
m-O(H)SnPh2}(CO)6(m-dppm)] (7b)
3.14. Preparation of [Mn2( -H)(SnPh3)(CO)7( -dppm)] (9b)
m
m
The procedure is completely analogous to that described for 6a
but using 1b as starting material. After similar workup, complex 7b
was obtained as a yellow-orange solid (0.098 g, 70%). Anal. Calcd.
for C43H34Mn2O7P2Sn: C, 54.18; H, 3.60. Found: C, 54.49; H, 3.87. 1H
Carbon monoxide was bubbled through a solution of compound
1b (0.057 g, 0.085 mmol) in toluene (20 mL) for 30 s, and then
HSnPh3 (0.030 g, 0.085 mmol) was added, and the mixture was
stirred for 4 h. The solvent was then removed under vacuum, and the
residue was then chromatographed on an alumina column (activity
NMR (CDCl3):
d
8.0e6.6 (m, 30H, Ph), 3.42, 3.24 (AB mult, JHH ¼ 13,

M.A. Alvarez et al. / Journal of Organometallic Chemistry 696 (2011) 1736e1748
1747
Table 7
3) at 253 K. Elution with dichloromethaneepetroleum ether (1:5)
Crystal data for new compounds.
gave a fraction yielding, after removal of solvents under vacuum,
complex 9b as a purple solid (0.050 g, 55%). The crystals used in the
X-ray study were grown by slow diffusion of a layer of petroleum
ether into a saturated dichloromethane solution of 9b at 253 K.
Elution with dichloromethaneepetroleum ether (1:1) gave a frac-
tion yielding analogously compound 7b (0.020 g, 25%) as a yellow
solid. Data for 9b: Anal. Calcd. for C50H38Mn2O7P2Sn: C, 57.67; H,
3.67. Found: C, 57.47; H. 3.61. 1H NMR (400.13 MHz, CD2Cl2, 263 K):
2a.1
5a$1/2H2O
9b
Molecular formula C26H32Mn11O3P2Si C38H42Mn2O12.5P2Sn2 C50H38Mn2O7P2Sn
Molecular weight 720.4
Crystal system
Space group
Radiation (
a (Å)
1107.9
Monoclinic
C 2/c
0.71073
14.131(7)
16.937(4)
37.945(9)
90
90.09(3)
90
9082(5)
8
1041.4
Monoclinic
P 21/a
0.71073
18.531(3)
14.749(3)
16.431(2)
90
91.87(1)
90
4488(1)
4
Monoclinic
P 21/c
l, Å)
0.71073
12.779(3)
11.179(4)
23.277(6)
90
93.87(2)
90
3318(2)
4
b (Å)
c (Å)
d
7.60e7.10 (m, 35H, Ph), 3.70 (t, JHP ¼ 11, 2H, CH2), ꢁ24.84 (dd,
a
b
g
(ꢀ)
(ꢀ)
(ꢀ)
JHP ¼ 14, 21, J119SnH z J117SnH ¼ 35,
m
-H). This spectrum exhibits very
weak resonances at ꢁ19.2 and ꢁ22.2 ppm due to the isomers of type
V (Å3)
C and A respectively.
Z
Calculated density 1.44
1.62
1.54
3.15. Preparation of [Mn2(m-H)(SnPh3)(CO)7(m-dmpm)] (9c)
(g cmꢁ3
)
Absorption
0.91
1.741
1.206
Carbon monoxide was bubbled through a solution of compound
1c (0.036 g, 0.085 mmol) in tetrahydrofuran (15 mL) for 30 s, and
then HSnPh3 was added (0.030 g, 0.085 mmol), and the mixture was
stirred for 2.5 h to give a red solution that was filtered using dry
diatomaceous earth. The solvent was removed under vacuum from
the filtrate, and the residue was then chromatographed on an
alumina column (activity 2.5) at 253 K. Elution with dichlor-
omethaneepetroleum ether (3:7) gave a yellow fraction containing
a tiny amount of compound 6c. Elution with a 7:13 solvent mixture
gave another fraction, containing 3 isomers of compound 9c.
Removal of solvents under vacuum from this fraction yielded
compound 9c (0.050 g, 75%) as a red solid. A final elution with neat
dichloromethane gave a deep-yellow fraction yielding analogously
complex 7c as a minor product (0.010 g,15%). Data for 9c: Anal. Calcd.
for C30H30Mn2O7P2Sn: C, 45.43; H, 3.81. Found: C, 45.21; H, 3.54. 1H
coefficient
(mmꢁ1
)
Diffractometer
Nonius CAD4
Nonius CAD4
291
Nonius CAD4
291
u/2q
Temperature (K) 291
Scan type
ꢁ 2
range (ꢀ)
1e25
u
q
u/2q
q
2e20
1e25
index ranges
(h, k, l)
Independent
reflections
ꢁ15, 15; 0, 13; 0, ꢁ13, 13; 0, 14; 0, 36 ꢁ22, 22; 0, 17; 0,
27
5833
19
7883
4513
2555
Reflections with 3994
[I > 3 (I)]
(I)]a
(I)]b
5457
s
R [I > 3
s
0.036
0.036
0/390
0.064
0.069
0/389
0.031
0.033
0/566
Rw[I > 3
s
Restraints/
parameters
a
R ¼ SjFo ꢁ jFcjj/SFo.
b
Rw ¼ [Sw(Fo ꢁ jFcj)2/SwFo2]1/2
.
NMR (400.13 MHz, CD2Cl2):
Data for isomer A: 2.08 (t, JHP ¼ 11.2, 2H, CH2), 1.62 (d, JHP ¼ 8.1, 6H,
Me),1.21 (d, JHP ¼ 6.3, 6H, Me), ꢁ23.2 (br,1H, -H). Data for isomer B:
2.64 (t, JHP ¼ 10, 2H, CH2),1.72 (d, JHP ¼ 7.2, 6H, Me),1.65 (d, JHP ¼ 8.9,
d 7.67e7.32 (m, 45H, Ph, isomers A, B, C).
hydrogen atoms were found on difference maps and others were
geometrically located. Their positions were not refined and they
were given an isotropic overall thermal parameter, except for the
bridging H in each case, this being refined isotropically with an
individual isotropic thermal parameter. Non hydrogen atoms were
anisotropically refined. Because of the small number of reflections
available for compound 5a, the hydrogen atoms were not located
on difference maps in that compound, and the carbon atoms of the
phenyl groups had to be isotropically refined. The crystal lattice also
contains a water molecule located on a binary axis.
m
6H, Me), ꢁ24.65 (d, br,1H, -H). Data for isomer C: 2.04 (t, JHP ¼ 10.7,
m
2H, CH2),1.59 (d, JHP ¼ 6.7, 3H, Me),1.57 (d, JHP ¼ 5.8, 3H, Me),1.39 (d,
JHP ¼ 8.1, 3H, Me), 1.35 (d, JHP ¼ 7.6, 3H, Me), ꢁ19.3 (t, JHP ¼ 23.3,
J
119SnH z J117SnH ¼ 79,1H,
m
-H).1H NMR (400.13 MHz, CD2Cl2, 223 K):
Hydride data for isomer A: ꢁ23.09 (t, JHP ¼ 25, J119SnH z J117SnH ¼ 65,
1H,
m
-H). Hydride data for isomer B: ꢁ24.85 (dd, JHP ¼ 7, 20,
J119SnH z J117SnH ¼ 47, 1H, -H). Hydride data for isomer C: ꢁ19.25
m
(dd, JHP ¼ 20.5, 26, J119SnH z J117SnH ¼ 78, 1H,
m
-H). 31P{1H} NMR
31.3 (br, -dmpm),
-dmpm). Data for
38.2 (d, JPP ¼ 60, J119SnH z J117SnH ¼ 266, -dmpm), 25.7
(d, JPP ¼ 60, 1P, -dmpm). Data for isomer C: 37.5 (d, JPP ¼ 61,
dmpm), 24.4 (d, JPP ¼ 61, -dmpm).
Least-squares refinements were carried out with approximation
in three blocks to the normal matrix by minimizing the function
(161.98 MHz, CD2Cl2, 223 K): Data for isomer A:
d
m
30.3 (AB mult, br, JPP ¼ 47, J119SnP z J117SnP ¼ 165,
m
P
w(jFoj ꢁ jFcj)2 where Fo and Fc are the observed and calculated
isomer B:
d
m
structure factors. Unit weight was used. Models reached conver-
m
d
m-
P
P
P
P
gence with R ¼ (jFoj ꢁ jFcj)/ jFoj and Rw ¼ [ w(jFoj ꢁ jFcj)2/
w
m
(jFoj)2]1/2 having values listed in Table 1. For 2a.1 and 9b, in the last
stages of refinement, each reflection was assigned a weight unity
3.16. X-ray data collection, structure determination and
refinements for compounds 2a.1, 5a and 9b
and for 5a the weighting scheme was w ¼ w0[1ꢁ(jFoj ꢁ jFcj/6
s
P
(Fo))2]2, where w0 ¼ 1/ r¼1,3ArTr(X) with X ¼ Fo/Fomax. Criteria for
satisfactory complete analysis were the ratios of root-mean-square
shifts to standard deviation being less than 0.1 and no significant
features in the final difference maps.
Unit cell dimensions with estimated standard deviations were
obtained from least-squares refinements of the setting angles of 25
well-centered reflections. Two standard reflections were monitored
periodically; they showed no change during data collection. Crys-
tallographic data and other pertinent information are summarized
in Table 7. Corrections were made for Lorentz and polarization
effects. Empirical absorption corrections (Difabs) [43] were applied.
For compound 9b, and extinction correction was applied.
Acknowledgments
We thank the DGI of Spain for financial support (CTQ2009-
09444) and the MEC of Spain for a grant (to R.C.).
Computations were performed by using CRYSTALS [44] adapted
on a Micro Vax II. Atomic form factors for neutral C, O, Mn, P, Sn, Si
and H were taken from Ref. [45]. Anomalous dispersion was taken
into account. The structures were solved by direct methods and
subsequent Fourier maps. For compounds 2a.1 and 9b, some
Appendix A. Supplementary material
CCDC 796136, 796137 and 796138 contain the supplementary
crystallographic data for compounds 2a.1, 5a and 9b respectively.

1748
M.A. Alvarez et al. / Journal of Organometallic Chemistry 696 (2011) 1736e1748
[17] (a) U. Schubert, Adv. Organomet. Chem. 30 (1990) 151;
(b) H.K. Sharma, K.H. Pannell, Chem. Rev. 95 (1995) 1351;
(c) H. Ogino, H. Tobita, Adv. Organomet. Chem. 42 (1998) 223.
[18] S.J. LaPlaca, W.C. Hamilton, J.A. Ibers, A. Davison, Inorg. Chem. 8 (1969)
1928.
These data can be obtained free of charge from The Cambridge
CrystallographicData Centre viawww.ccdc.cam.ac.uk/data_request/
cif.
[19] P.S. Braterman, Metal Carbonyl Spectra. Academic Press, London, 1975.
[20] X.-Y. Liu, V. Riera, M.A. Ruiz, M. Lanfranchi, A. Tiripicchio, Organometallics 22
(2003) 4500.
[21] J. Sun, R.S. Lu, R. Bau, G.K. Yang, Organometallics 13 (1994) 1317.
[22] See, for example W.D. Wang, R. Eisemberg, Organometallics 11 (1992) 908.
[23] R.D. Adams, B. Captain, M. Johansson, J.L. Smith Jr., J. Am. Chem. Soc. 127
(2005) 488.
[24] H. Preut, H.J. Haupt, Z. Anorg. Allg. Chem. 422 (1976) 47.
[25] V. Riera, M.A. Ruiz, A. Tiripicchio, M. Tiripicchio-Camellini, J. Chem. Soc.,
Dalton Trans. (1987) 1551.
[26] M.S. Holt, W.L. Wilson, J.H. Nelson, Chem. Rev. 89 (1989) 11.
[27] M.R. Churchill, K.N. Amoh, H.J. Wasserman, Inorg. Chem. 20 (1981) 1609.
[28] F.A. Cotton, K.B. Shin, Gazz. Chim. Ital. (1985) 705.
[29] J.E. Huheey, Inorganic Chemistry: Principles of Structure and Reactivity, third
ed. Harper & Row, New York, 1983, p 268.
Appendix. Supplementary material
Supplementary data related to this article can be found online at
doi:10.1016/j.jorganchem.2010.12.016.
References
[1] J.Y. Corey, J. Braddock-Wilking, Chem. Rev. 99 (1999) 175.
[2] I. Ojima, Z. Li, J. Zhu, in: S. Patai, Z. Rappoport (Eds.), The Chemistry of Organic
Silicon Compounds, Wiley, New York, 1998.
[3] (a) J.Y. Corey, Adv. Organomet. Chem. 51 (2004) 1;
(b) P. Braunstein, X. Morise, Chem. Rev. 100 (2000) 3541;
(c) F. Gauvin, J.F. Harrod, H.G. Woo, Adv. Organomet. Chem. 42 (1998) 363.
[4] M.D. Curtis, P.S. Epstein, Adv. Organomet. Chem. 19 (1981) 213.
[5] V. Riera, M.A. Ruiz, A. Tiripicchio, M. Tiripicchio-Camellini, J. Chem. Soc., Chem.
Commun. (1985) 1505.
[6] (a) F.J. García-Alonso, M. García-Sanz, V. Riera, M.A. Ruiz, A. Tiripicchio,
M. Tiripicchio-Camellini, Angew. Chem., Int. Ed. Engl. 27 (1988) 1167;
(b) F.J. García-Alonso, V. Riera, M.A. Ruiz, A. Tiripicchio, M. Tiripicchio-Cam-
ellini, Organometallics 11 (1992) 370.
[7] R. Carreño, V. Riera, M.A. Ruiz, Y. Jeannin, M. Philoche-Levisalles, J. Chem. Soc.,
Chem. Commun. (1990) 15.
[8] F.W.B. Einstein, R.K. Pomeroy, A.C. Willis, J. Organomet. Chem. 311 (1986) 257.
[9] (a) Y. Ohki, H. Suzuki, Angew. Chem., Int. Ed. 19 (1990) 3120;
(b) Y. Ohki, T. Kojima, M. Oshima, H. Suzuki, Organometallics 20 (2001) 2654.
[10] (a) H. Suzuki, T. Takao, M. Tanaka, Y. Moro-oka, J. Chem. Soc., Chem. Commun.
(1992) 476;
(b) T. Takao, S. Yoshida, H. Suzuki, M. Tanaka, Organometallics 14 (1995) 3855;
(c) T. Takao, M. Amako, H. Suzuki, Organometallics 22 (2003) 3855.
[11] M.A. Alvarez, M.E. García, A. Ramos, M.A. Ruiz, Organometallics 25 (2006) 5374.
[12] (a) J.K. Hoyano, M. Elder, W.A.G. Graham, J. Am. Chem. Soc. 91 (1969) 4568;
(b) M. Elder, Inorg. Chem. 9 (1970) 762;
[30] R.D. Adams, B. Captain, C.B. Hollandsworth, M. Johansson, J.L. Smith Jr.,
Organometallics 25 (2006) 3848.
[31] R.D. Adams, E. Trufan, Organometallics 29 (2010) 4346.
[32] B.A. Narayanan, J.K. Kochi, Inorg. Chim. Acta 122 (1986) 85.
[33] (a) C. Glidewell, D.C. Liles, Acta Crystallogr. Sect. B. 34 (1978) 1693;
(b) C. Glidewell, D.C. Liles, Acta Crystallogr. Sect. B. 34 (1978) 129.
[34] U. Schubert, E. Kunz, B. Harkers, J. Willnecker, J. Meyer, J. Am. Chem. Soc. 111
(1989) 2572.
[35] C.J. Cardin, D.J. Cardin, H.E. Parge, J.M. Power, J. Chem. Soc., Chem. Commun.
(1984) 609.
[36] R.D. Adams, J.E. Cortopassi, J.H. Yamamoto, Organometallics 12 (1993) 3036.
[37] F.J. García-Alonso, M. García-Sanz, X.Y. Liu, A. Oliveira, M.A. Ruiz, V. Riera,
C. Bois, J. Organomet. Chem. 511 (1996) 93.
[38] (a) R. Bau, R.G. Teller, S.W. Kirtley, T.F. Koetzle, Acc. Chem. Res. 12 (1979) 176;
(b) D.S. Moore, S.D. Robinson, Chem. Soc. Rev. 12 (1983) 415.
[39] (a) G. Albertin, S. Antoniutti, J. Castro, S. García-Fontán, G. Zanardo, Organo-
metallics 26 (2007) 2918;
(b) D. Christendat, I. Wharf, A.M. Lebuis, I.S. Butler, D.F.R. Gilson, Inorg. Chim.
Acta 329 (2002) 36.
[40] W.L.F. Armarego, C. Chai, Purification of Laboratory Chemicals, fifth ed. But-
terwortheHeinemann, Oxford, U.K., 2003.
(c) W.A.G. Graham, J. Organomet. Chem. 300 (1986) 81.
[13] (a) T. Takao, S. Yoshida, H. Suzuki, Organometallics 24 (2005) 521 and refer-
ences therein;
(b) H. Hashimoto, Y. Hayashi, I. Aratani, C. Kabuto, M. Kira, Organometallics 21
(2002) 1534.
[14] M. Tanabe, D. Ito, K. Osakada, Organometallics 27 (2008) 2258.
[15] V. Riera, M.A. Ruiz, A. Tiripicchio, M. Tiripicchio-Camellini, Organometallics 12
(1993) 2962.
[41] R.B. King, K.S.R. Veer, Inorg. Chem. 23 (1984) 2482.
[42] V. Riera, M.A. Ruiz, J. Chem. Soc., Dalton Trans. (1986) 2617.
[43] N. Walker, D. Stuart, Acta Crystallogr. Sect. A 39 (1983) 158.
[44] D.J. Watkin, J.R. Carruthers, P. Betteridge, CRYSTALS, An Advanced Crystallo-
graphic Program System: Chemical Crystallography Laboratory. University of
Oxford, Oxford, U.K, 1988.
[45] International Tables for X-Ray Crystallography, vol. IV, Kynoch Press,
Birmingham, U.K, 1974.
[16] R. Carreño, V. Riera, M.A. Ruiz, C. Bois, Y. Jeannin, Organometallics 12 (1993)
1946.