Synthesis, solid-state structure and solution behavior of hydrido-bridged adducts between the group 11 [M(PPh3)]+ cations and the triangular cluster anion [Re3(μ-H)4(CO)9(PPh3)]-

Article abstract of DOI:10.1021/om0200252

The reactions of the unsaturated anion [Re₃(μ-H)₄(CO)₉(PPh₃)]^- (1) with the metallic Lewis acids [M(PPh₃)]⁺ (M = Cu, Ag, Au) afford the addition derivatives [Re₃{μ-M(PPh₃)}(μ₃-H)-(μ-H)₃ (CO)₉(PPh₃)], which have been characterized in solution and in the solid state, by X-ray analysis. The copper derivative is obtained also by reaction of [Re₃(μ-H)₃(μs₃-H)(CO)₉]^- (as the [Re(CO)₃(Me₂CO)₃]⁺ salt) with [Cu(PPh₃)₂NO₃]. Treatment with PPh₃ extrudes the [M(PPh₃)]⁺ groups, restoring the starting anion 1. Solid-state structures of the adducts show a butterfly Re₃M metal skeleton with one hydride bridging the Re₃M face and three hydrides spanning the three Re-Re edges. The interaction of the [M(PPh₃)]⁺ cation with 1 leads only to a minor structural perturbation of the unsaturated Re₂(μ-H)₂ moiety. The solid-state structure is maintained in nondonor solvents, while in acetone partial (Au, Ag) or complete (Cu) reversible ionic dissociation of the adducts is observed, affording 1 and solvated [M(PPh₃)]⁺ cations. Experiments performed in different solvents show that the donor power and not the dielectric constant is the key factor responsible for dissociation. For the silver adduct the dissociation in acetone is fast on the NMR time scale. The exchange between 1 and the silver or copper (but not gold) adducts occurs even in CD₂Cl₂.

Full text of DOI:10.1021/om0200252

Organometallics 2002, 21, 2705-2714
2705
Syn th esis, Solid -Sta te Str u ctu r e a n d Solu tion Beh a vior
of Hyd r id o-Br id ged Ad d u cts betw een th e Gr ou p 11
[M(P P h 3)]+ Ca tion s a n d th e Tr ia n gu la r Clu ster An ion
[Re3(µ-H)4(CO)9(P P h 3)]-
Tiziana Beringhelli, Giuseppe D’Alfonso,* Maria Grazia Garavaglia, and
Monica Panigati
Dipartimento di Chimica Inorganica, Metallorganica e Analitica, Via Venezian 21,
20133 Milano, Italy
Pierluigi Mercandelli* and Angelo Sironi
Dipartimento di Chimica Strutturale e Stereochimica Inorganica, Via Venezian 21,
20133 Milano, Italy
Received J anuary 14, 2002
The reactions of the unsaturated anion [Re3(µ-H)4(CO)9(PPh3)]- (1) with the metallic Lewis
acids [M(PPh3)]+ (M ) Cu, Ag, Au) afford the addition derivatives [Re3{µ-M(PPh3)}(µ3-H)-
(µ-H)3(CO)9(PPh3)], which have been characterized in solution and in the solid state, by X-ray
analysis. The copper derivative is obtained also by reaction of [Re3(µ-H)3(µ3-H)(CO)9]- (as
the [Re(CO)3(Me2CO)3]+ salt) with [Cu(PPh3)2NO3]. Treatment with PPh3 extrudes the
[M(PPh3)]+ groups, restoring the starting anion 1. Solid-state structures of the adducts show
a butterfly Re3M metal skeleton with one hydride bridging the Re2M face and three hydrides
spanning the three Re-Re edges. The interaction of the [M(PPh3)]+ cation with 1 leads only
to a minor structural perturbation of the unsaturated Re2(µ-H)2 moiety. The solid-state
structure is maintained in nondonor solvents, while in acetone partial (Au, Ag) or complete
(Cu) reversible ionic dissociation of the adducts is observed, affording 1 and solvated
[M(PPh3)]+ cations. Experiments performed in different solvents show that the donor power
and not the dielectric constant is the key factor responsible for dissociation. For the silver
adduct the dissociation in acetone is fast on the NMR time scale. The exchange between 1
and the silver or copper (but not gold) adducts occurs even in CD2Cl2.
In tr od u ction
was observed,6,7 with coordination of the resulting
anionic fragment on the cluster surface.
With other electrophiles (such as I2 in nondonor
solvents), oxidation of one hydride with hydrogen evolu-
tion was observed, instead of hydride transfer, according
to eq 3 (L ) CO).4
Previous studies showed that the unsaturated trian-
gular cluster anions [Re3(µ-H)4(CO)9(L)]- (46 valence
electrons (VE’s), L ) two-electron donor)1,2 easily react
with electrophiles E+ able to extract one hydride ligand
and to promote the coordination of neutral L′ or anionic
X- ligands, affording saturated derivatives (48 VE’s),
according to eqs 1 and 2 (Scheme 1).1-5 The most used
[Re3(µ-H)4(CO)9(L)]- + 0.5I2 f
[Re3(µ-H)3(µ-I)(CO)9(L)]- + 0.5H2 (3)
[Re3(µ-H)4(CO)9(L)]- + E+ + 2L′ f
[Re3(µ-H)3(CO)9(L)(L′)2] + EH (1)
In all these cases, the thermodynamic site of the
electrophilic attack was one of the hydrides of the
Re(µ-H)2Re moiety. According to a local electron count-
ing, this is the site where not only the anionic charge
[Re3(µ-H)4(CO)9(L)]- + E+ + X- f
[Re3(µ-H)3(µ-X)(CO)9(L)]- + EH (2)
(3) (a) Ciani, G.; D’Alfonso, G.; Freni, M.; Romiti, P.; Sironi, A. J .
Organomet. Chem. 1982, 226, C31. (b) Beringhelli, T.; Ciani, G.;
D’Alfonso, G.; Freni, M.; Sironi, A. J . Organomet. Chem. 1982, 244,
C27. (c) Ciani, G.; D’Alfonso, G.; Freni, M.; Romiti, P.; Sironi, A. J .
Organomet. Chem. 1983, 254, C37. (d) Beringhelli, T.; Ciani, G.;
D’Alfonso, G.; Freni, M.; Sironi, A. J . Chem. Soc., Dalton Trans. 1985,
1507. (e) Beringhelli, T.; Ciani, G.; D’Alfonso, G.; Freni, M.; Moret,
M.; Sironi, A. J . Chem. Soc., Dalton Trans. 1989, 1143.
(4) Ciani, G.; D’Alfonso, G.; Freni, M.; Romiti, P.; Sironi, A. Inorg.
Chem. 1983, 22, 3115.
electrophile E+ was the proton, originating from either
strong or weak Brønsted acids,3 but also ethanolic
iodine4 or the tropylium ion5 could be successfully used.
Also, hydride transfer to the electrophilic carbon atom
of organic molecules containing polarized multiple bonds
(1) Beringhelli, T.; Ciani, G.; D’Alfonso, G.; Molinari, H.; Sironi, A.
Inorg. Chem. 1985, 24, 2666.
(2) Beringhelli, T.; Ciani, G.; D’Alfonso, G.; Freni, M.; Molinari, H.;
Sironi, A. J . Chem. Soc., Dalton Trans. 1986, 2691.
(5) Beringhelli, T.; Ciani, G.; D’Alfonso, G.; Freni, M.; Romiti, P.;
Sironi, A. Inorg. Chem. 1984, 23, 2849.
(6) Beringhelli, T.; Ciani, G.; D’Alfonso, G.; Freni, M.; Moret, M.;
Sironi, A. J . Organomet. Chem. 1988, 339, 323.
10.1021/om0200252 CCC: $22.00 © 2002 American Chemical Society
Publication on Web 05/25/2002

2706 Organometallics, Vol. 21, No. 13, 2002
Beringhelli et al.
Sch em e 1
rather than addition, due to the isolobal relationship
between H+ and [M(PPh3)]+:18,19 for instance, the reac-
tion of [Re2H8(PPh3)4] with Au(PPh3)NO3 afforded [Re2-
{AuPPh3}2H6(PPh3)4].20,21 In the case of ReH7(PPh3)2,
both substitution and addition were observed.22
On the other hand, the possibility that electron
transfer rather than addition/substitution takes place
cannot be ruled out a priori: in the reaction of [Re3(µ-
H)4(CO)10]- with iodine the change from a donor to a
nondonor solvent was sufficient to change the reaction
course from hydride abstraction to redox. The relatively
strong oxidizing ability of silver(I) toward metal hydride
complexes is well-known.23 Moreover, the reduction of
[Au(PPh3)]+ with main-group or transition-metal hy-
drides has been used to promote the growth of large
homo- or heteronuclear gold clusters: see, for instance,
24
the reductions of [Au(PPh3)NO3] with NaBH4 or with
[ReH7(PPh3)2],22 affording, respectively, the [Au9(PPh3)8]3+
and [(AuPPh3)5ReH4(PPh3)2]2+ gold phosphine cluster
cations.
It was therefore of interest to verify the preferred
reaction path in the case of the anion 1. We report here
the complete solid-state and solution characterization
of the 1:1 addition products [Re3{µ-M(PPh3)}(µ3-H)(µ-
H)3(CO)9(PPh3)], which have been obtained for all three
coinage metals, as well as the investigation of their
stability in a donor solvent such as acetone.
but also the unsaturation of the cluster is mainly
localized. Therefore, such a moiety can be described
either as an olefin-like system (containing a formal Red
Re double bond) or, better, as a diborane-like system
(containing a four-center-four-electron bond).
We have now studied the reactions of the unsaturated
anion [Re3(µ-H)4(CO)9(PPh3)]- (1) with other electro-
philes, namely the metallic Lewis acids [M(PPh3)]+,
where M indicates a group 11 metal (Cu, Ag, Au). It is
well-known that these cations are able to interact
reversibly with transition metal hydrides, giving com-
plexes containing one or more hydrides doubly or triply
bridging between the transition and the coinage
metal.8-11 In particular, addition of [M(PPh3)]+ cations
to rhenium polyhydrides was previously observed for the
complexes [ReH5(PPh3)3],12 [ReH5(PMe2Ph)3],13 [ReH3-
(CO)(PMe2Ph)3],14 [Re2H8(PPh3)4],15 and [Re3(µ-H)3(µ3-
ampy)(CO)9]- (Hampy ) 2-amino-6-methylpyridine).16
Particularly significant in this context is also the
addition of [M(PPh3)]+ cations (M ) Ag, Au) to the
unsaturated species [Mn2(µ-H)2(CO)6(µ-dppm)],17 since
it provided the first experimental evidence of a Lewis
base behavior, toward group 11 cations, of an unsatur-
ated neutral M(µ-H)2M moiety.
Resu lts a n d Discu ssion
Syn th esis of th e [Re3{µ-MP P h 3}(µ3-H)(µ-H)3(CO)9-
(P P h 3)] Com p lexes (M ) Cu , Ag, Au ). The treatment
of the triangular cluster anion [Re3(µ-H)4(CO)9(PPh3)]-
(1), in CH2Cl2 solution, with an acetone solution con-
taining a stoichiometric amount of PPh3 and AgOTf (OTf
) CF3SO3) caused an instantaneous shift toward higher
frequencies of all the ν(CO) bands (ca. 10-15 cm-1; see
Table 1). Standard workup (see Experimental Section)
led to the isolation of a sun yellow compound, character-
ized as the complex [Re3{µ-AgPPh3}(µ3-H)(µ-H)3(CO)9-
(PPh3)] (2) on the basis of its spectroscopic data and of
a single-crystal X-ray analysis (Scheme 2, eq 4, M ) Ag).
In other cases, the reaction of transition-metal poly-
[Re3(µ-H)4(CO)9(PPh3)]- + [MPPh3]+ f
hydrides with [M(PPh3)]+ cations afforded substitution
[Re3{µ-MPPh3}(µ3-H)(µ-H)3(CO)9(PPh3)] (4)
(7) (a) Beringhelli, T.; Ciani, G.; D’Alfonso, G.; Freni, M.; Moret, M.;
Sironi, A. J . Organomet. Chem. 1990, 399, 291. (b) Beringhelli, T.;
Ciani, G.; D’Alfonso, G.; Minoja, A. P.; Moret, M.; Sironi, A. Organo-
metallics 1991, 10, 3131.
(8) (a) Salter, I. D. In Comprehensive Organometallic Chemistry II;
Abel, E. W., Stone, F. G. A., Wilkinson, G., Eds.; Pergamon: Oxford,
U.K., 1995; Vol. 10, p 255. (b) Salter, I. D. Adv. Organomet. Chem.
1989, 29, 249.
(9) (a) Mingos, D. M. P.; Watson, M. J . Adv. Inorg. Chem. 1992, 39,
327. (b) Hall, K. P.; Mingos, D. M. P. Prog. Inorg. Chem. 1984, 32,
237.
(10) Chetcuti, M. J . In Comprehensive Organometallic Chemistry
II; Abel, E. W., Stone, F. G. A., Wilkinson, G., Eds.; Pergamon: Oxford,
U.K., 1995; Vol. 10, p 23.
The presence of a phosphine bound to the silver atom
(P(2) in Table 1) was revealed by the two doublets in
the 31P NMR spectrum arising from the coupling with
109Ag and 107Ag, while the phosphine bound to a vertex
of the Re cluster, labeled P(1), was responsible for a
singlet at a δ value quite similar to that of the starting
anion 1. The three hydridic resonances in the ratio 1:1:2
(18) Evans, D. G.; Mingos, D. M. P. J . Organomet. Chem. 1982, 232,
171.
(19) Lauher, J . W.; Wald, K. J . Am. Chem. Soc. 1981, 103, 7648.
(20) Boyle, P. D.; J ohnson, B. J .; Alexander, B. D.; Casalnuovo, J .
A.; Gannon, P. R.; J ohnson, S. M.; Larka, E. A.; Mueting, A. M.;
Pignolet, L. H. Inorg. Chem. 1987, 26, 1346.
(21) See also: Casalnuovo, A. L.; Pignolet, L. H.; van der Velden, J .
W. A.; Bour, J . J .; Steggerda, J . J . J . Am. Chem. Soc. 1983, 105, 5957.
(22) Boyle, P. D.; J ohnson, B. J .; Buehler, A.; Pignolet, L. H. Inorg.
Chem. 1986, 25, 5.
(11) Albinati, A.; Venanzi, L. M. Coord, Chem. Rev. 2000, 200-202,
687.
(12) Moehring, G. A.; Walton, R. A. J . Chem. Soc., Dalton Trans.
1988, 1701.
(13) Sutherland, B. R.; Folting, K.; Streib, W. E.; Ho, D. M.;
Huffman, J . C.; Caulton, K. G. J . Am. Chem. Soc. 1987, 109, 3489.
(14) Luo, S.; Burns, C. J .; Kubas, G. J .; Bryan, J . C.; Crabtree, R.
H. J . Cluster Sci. 2000, 11, 189.
(15) Moehring, G. A.; Fanwick, P. E.; Walton, R. A. Inorg. Chem.
1987, 26, 1861.
(16) Cabeza, J . A.; Riera, V.; Trivedi, R.; Grepioni, F. Organome-
tallics 2000, 19, 2043.
(23) See for instance: (a) Rhodes, L. F.; Zubkowski, J . D.; Folting,
K.; Huffman, J . C.; Caulton, K. G. Inorg. Chem. 1982, 21, 4185. (b)
Rhodes, L. F.; Huffman, J . C.; Caulton, K. G. Inorg. Chim. Acta 1992,
198-200, 639.
(17) Carreno, R.; Riera, V.; Ruiz, M. A.; Bois, C.; J eannin, Y.
Organometallics 1992, 11, 2923.
(24) Cariati, F.; Naldini, L. J . Chem. Soc., Dalton Trans. 1972, 2286.

Adducts between [M(PPh3)]- and [Re3(µ-H)4(PPh3)]-
Organometallics, Vol. 21, No. 13, 2002 2707
Ta ble 1. IR a n d NMR Da ta for th e Title Com p ou n d s (Room Tem p er a tu r e, CD2Cl2)
δ(Ha) (ppm)
(J HP, Hz)
δ(Hb) (ppm)
(J HP, Hz)
δ(Hc) (ppm)
(J HP, Hz)
δ(P(1)) δ(P(2))
compd
(ppm)
(ppm)
IR (νCO) (cm-1
)
1
2
3
-7.73 (d, 1H; 5.6)
-8.91 (s, 1H)
-12.26 (d, 2H; 16.2)
8.1
10.3
9.0
2032 m, 2003 s, 1996 s, 1926 sh, 1906 s
2046 m, 2015 vs, 1949 m, 1934 m, 1921 ms
2047 m, 2016 vs, 1950 m, 1936 m
1925 m, sh
-8.55 (d, 1H; 4.1) -10.82 (dd, 1H; 31)a -11.92 (d, 2H; 16.3)
24.0b
59.3
-8.30 (d, 1H; 4.3)
-7.62 (d, 1H; 70)
-11.15 (d, 2H; 16.1)
-11.31 (d, 2H; 15.8)
4
-8.24 (s, 1H)
-11.84(d, 1H; 27)
9.4
14.4
2048 m, 2017 vs, 1954 m, 1935 m, 1921 m
a
b
1
1J HAg ) 85 Hz. dd, J AgP ) 645 and 560 Hz.
Sch em e 2a
a
Legend: (i) AgOTf + PPh3; (ii) AuCl(PPh3); (iii) Cu(NC-
+
Me)4 + PPh3; (iv) Cu(PPh3)2(NO3).
F igu r e 1. ORTEP drawing of [Re3{µ-Ag(PPh3)}(µ3-H)(µ-
H)3(CO)9(PPh3)] (2) with a partial labeling scheme. Ther-
mal ellipsoids are drawn at the 30% probability level.
Hydrogen atoms are given arbitrary radii.
(see Table 1) in 1H NMR spectrum indicated the
preservation of the Cs symmetry of the starting anion
1. Variable-temperature analysis down to 203 K, in CD2-
Cl2, did not show any significant change. One of the
signals of intensity 1 (-10.82 ppm) appeared as a
doublet of doublets, due to coupling both with silver and
phosphorus atoms (the signal was too broad to allow us
to resolve the couplings with 109Ag and 107Ag, even in a
31P-decoupled spectrum): the high value of J HAg (85 Hz)
indicated that the hydride responsible for this signal
(labeled as Hb in Table 1) was directly bound to the
silver atom. The other two hydridic resonances had
chemical shift and J HP values close to those of 1,
indicating that they had been little affected by the
addition of the silver cation. A 2D NOESY experiment
performed in CD2Cl2 at room temperature showed that
all the hydride resonances but that at -8.55 ppm had
NOE cross-peaks with the protons of the two phos-
phines: this indicated that the hydride responsible for
this resonance, labeled as Ha in Scheme 2, is coordinated
in an anti position with respect to both the Ag(PPh3)
and PPh3 ligands. This means that the Ag(PPh3) frag-
ment is syn with respect to the rhenium-bound PPh3
ligand.
hydride (70 Hz) agrees with previous literature data for
complexes containing analogous M2(µ3-H)(µ-AuPPh3)
fragments with an approximately transoid arrangement
of the hydride and the phosphine ligands around the
gold atom (M ) Mn, 74 Hz;17 M ) Mo, 86 Hz;25 M ) Re,
59 Hz16).26 Selective {31P}-1H decoupling experiments
confirmed that also in this case the 31P resonance at
high field was due to the phosphine ligand bound to the
rhenium cluster. The solid-state structure determined
from a single-crystal X-ray analysis is depicted in Figure
2.
To obtain the corresponding copper derivative accord-
ing to reaction 4, we treated the anion 1 with a CH2Cl2
solution containing stoichiometric amounts of PPh3 and
[Cu(CH3CN)4]BF4. The NMR data of the main reaction
product (Table 1) confirmed the formation of the ex-
pected derivative [Re3{µ-Cu(PPh3)}(µ3-H)(µ-H)3(CO)9-
(PPh3)] (4), whose solid-state structure is shown in
Figure S2 (Supporting Information). In this case, how-
ever, the NMR spectrum of the isolated yellow microc-
rystalline solid showed, besides the signals of 4, also
some unattributed hydridic resonances (the main one
being a doublet at -6.43 ppm, with J HP ) 16 Hz), whose
overall intensity varied in the range 18-35% in re-
peated preparations with slightly different experimental
procedures.
1
These data, as well as the results of the H-31P 2D
HMQC experiment described in the Experimental
Section, agree with the solid-state structure of 2,
depicted in Figure 1.
The analogous gold derivative was obtained according
to the same reaction 4 (M ) Au), by treating the anion
1 with a stoichiometric amount of the [Au(PPh3)]+ cation
(prepared in situ from Au(PPh3)Cl and AgOTf). Also in
this case the reaction was instantaneous and the same
purification procedure afforded a bright orange com-
pound, formulated as [Re3{µ-Au(PPh3)}(µ3-H)(µ-H)3-
(CO)9(PPh3)] (3) on the basis of its IR and NMR data
(Table 1). The large value of J HP for the triply bridging
When the reaction was performed directly in the NMR
1
tube, the H spectra showed the formation of a species
containing one or more MeCN molecules bound to the
(25) Ferrer, M.; Reina, R.; Rossel, O.; Seco, M.; Alvarez, S.; Ruiz,
E.; Pellinghelli, M. A.; Tiripicchio, A. Organometallics 1992, 11, 3753.
(26) Still higher values have been observed for transoid doubly
bridging hydrides in M(µ-H)(µ-AuPPh3) fragments: M ) Cr, 105 Hz;27
M ) Re, 96 Hz.14

2708 Organometallics, Vol. 21, No. 13, 2002
Beringhelli et al.
It is worth noting that the preliminary fragmentation
of [Re4(µ3-H)4(CO)12] (eq 6) was not necessary: the
simultaneous treatment of [Re4(µ3-H)4(CO)12] with 1
equiv of [Cu(PPh3)2(NO3)] and some acetone (30 µL), in
CD2Cl2 at low temperature afforded adduct 4 as the only
hydridic species. Under these conditions, 1H NMR
monitoring showed the intermediate formation of a
species responsible for two hydridic resonances (ratio
1:1, δ -9.42, -11.20 ppm, at 193 K), which at 243 K
gave rise to an averaged signal at δ -10.20. After 2 h
at 253 K this species was completely transformed into
4.
In agreement with these findings, no reaction was
observed between 1 and the neutral species [M(PPh3)-
Cl] (M ) Cu, Ag, Au).
Solid Sta te Str u ctu r e of th e Ad d ition Der iva -
tives [Re3{µ-M(P P h 3)}(µ3-H)(µ-H)3(CO)9(P P h 3)] (2-
4). The molecular structures of compounds 2-4 were
determined by X-ray diffraction analysis. The copper
derivative 4 crystallizes in two polymorphic forms, a
monoclinic and an orthorhombic one (4m and 4o),
isomorphous with the silver and the gold derivatives 2
and 3, respectively (see Table 4). Figures 1 and 2 show
ORTEP drawings of 2 and 3, respectively. Table 2
contains the most relevant bond distances and angles
for 2-4, while Table 3 collects some mean bond param-
eters for the adducts 2-4 and for the parent compound
[Re3(µ-H)4(CO)9(PPh3)]- (1).
Compounds 2-4 exhibit a Re3M butterfly metal core
with all the Re-Re edges and the Re2M face bridged
by one hydrido ligand. Each of the rhenium atoms also
bears three terminal CO ligands with facial coordina-
tion; one triphenylphosphine ligand is bound to Re(3)
(in an axial position) and another one to M(1) (in a
transoid arrangement with respect to the face-bridging
hydride).
F igu r e 2. ORTEP drawing of [Re3{µ-Au(PPh3)}(µ3-H)(µ-
H)3(CO)9(PPh3)] (3) with a partial labeling scheme. Ther-
mal ellipsoids are drawn at the 30% probability level.
Hydrogen atoms are given arbitrary radii.
Cu atom (see the Supporting Information). The same
[Re3{µ-Cu(PPh3)(NCMe)n}(µ3-H)(µ-H)3(CO)9(PPh3)] spe-
cies was obtained upon treating 4 with MeCN.
To obtain a more selective pathway to compound 4,
we attempted two different synthetic approaches. At
first we tried to generate the [Cu(PPh3)]+ cation by
treating [Cu(PPh3)Cl]4 with AgOTf, as in the synthesis
of the gold derivative 3. However, the 1H NMR spectrum
revealed the formation of a complex mixture of hydridic
derivatives in which compound 4 was only a minor
component. Then we explored a fully different approach,
consisting of the reaction of the complex [Cu(PPh3)2-
(NO3)] with the “super-unsaturated” (44 VE’s) anion
[Re3(µ-H)3(µ3-H)(CO)9]- (5),28 as shown in Scheme 2.
This approach (eq 5) was successful only on using the
[Re(CO)3(Me2CO)3]+ salt of 5 (generated through the
reaction of [Re4(µ3-H)4(CO)12] with acetone, according
to eq 6).28a,29 When the reaction was repeated by using
Ignoring the propeller-like conformation adopted by
the two triphenylphosphine ligands, molecules 2-4 (as
well as the parent anion 1) show an approximate Cs
symmetry, the main deviation being related to the
distorted trigonal coordination of the group 11 metal
atom, in the monoclinic phases 2 and 4m (compare the
angles Re(1)-M(1)-P(2) and Re(2)-M(1)-P(2) in Table
2).
[Re3(µ-H)3(µ3-H)(CO)9]- + [Cu(PPh3)2(NO3)] f
[Re3{µ-Cu(PPh3)}(µ3-H)(µ-H)3(CO)9(PPh3)] +
-
NO3 (5)
With reference to the parent anion 1 the Re-Re edge
originally bridged by two hydrides is now additionally
spanned by a µ-MPPh3 group. As a consequence of the
larger size of the bridgehead atom (a group 11 metal vs
hydrogen), the Re(1)-Re(2) bond distance results slightly
lengthened in 2-4 with respect to 1 (the effect is
stronger for the more hindered gold derivative; see Table
3). The main effect on the overall stereochemistry of the
addition of the MPPh3 fragment is a displacement of
the carbonyl ligands on Re(1) and Re(2) away from this
sterically demanding group: in particular the Re(1,2)-
Re(2,1)-C(diagonal) angles are increased for the car-
bonyl ligands lying on the same side of the Re3 plane
bearing the MPPh3 group (C(12) and C(22)) and are
decreased for those on the opposite side (C(13) and
C(23)), with respect to the values in the parent anion 1
[Re4(µ3-H)4(CO)12] + 3Me2CO f
[Re3(µ-H)3(µ3-H)(CO)9]- + [Re(CO)3(Me2CO)3]+ (6)
+
the PPh4 salt of 5, the formation of 4 was negligible,
the anion 1 being the main reaction product, both in
CH2Cl2 and in THF solution. We suppose that the [Re-
(CO)3(Me2CO)3]+ cation formed in reaction 6 plays a key
role in the activation of the [Cu(PPh3)2(NO3)] reactant,
-
by abstracting the NO3 anion and generating the
copper cation able to attack the rhenium hydrido
carbonyl cluster anion (eq 7).
[Re(CO)3(Me2CO)3]+ + [Cu(PPh3)2(NO3)] f
[Re(CO)3(Me2CO)2(NO3)] + [Cu(PPh3)2]+ +
Me2CO (7)
(28) (a) Beringhelli, T.; D’Alfonso, G. J . Chem. Soc., Chem. Commun.
1994, 2631. (b) Horng, H. C.; Cheng, C. P.; Yang, C. S.; Lee, G.-H.
Organometallics 1996, 15, 2543.
(27) Green, M.; Orpen, A. G.; Salter, I. D.; Stone, F. G. A. J . Chem.
Soc., Dalton Trans. 1984, 2497.
(29) Beringhelli, T.; D’Alfonso, G.; Garavaglia, M. G. J . Chem. Soc.,
Dalton Trans. 1996, 1771.

Adducts between [M(PPh3)]- and [Re3(µ-H)4(PPh3)]-
Organometallics, Vol. 21, No. 13, 2002 2709
Ta ble 2. Selected Bon d Dista n ces (Å) a n d An gles
(d eg) for th e Ad d ition Der iva tives
of the Re3 plane, in a syn arrangement. A close inspec-
tion of their molecular structure shows that their anti
isomer should be sterically more hindered, due to some
unavoidable interactions of the rhenium-bound phos-
phine with the diagonal carbonyl on Re(1) and Re(2).
Beh a vior of Ad d u cts 2-4 in Solu tion . The addi-
tion of a good donor molecule such as PPh3 to CH2Cl2
solutions of the adducts 2-4 restored the starting anion
1, according to eq 8.30-32 In agreement with previous
[Re3{µ-M(P P h 3)}(µ3-H)(µ-H)3(CO)9(P P h 3)]
2
3
4m
4o
(M ) Ag) (M ) Au)
(M ) Cu) (M ) Cu)
Re(1)-Re(2)
Re(1)-Re(3)
Re(2)-Re(3)
Re(1)-M(1)
Re(2)-M(1)
Re(1)-C(11)
Re(1)-C(12)
Re(1)-C(13)
Re(2)-C(21)
Re(2)-C(22)
Re(2)-C(23)
Re(3)-C(31)
Re(3)-C(32)
Re(3)-C(33)
Re(3)-P(1)
M(1)-P(2)
2.8398(9) 2.8589(7)
3.1821(9) 3.2003(7)
3.2166(8) 3.1854(7)
3.0145(9) 2.9673(6)
2.9604(9) 2.9260(6)
2.8269(15) 2.8256(7)
3.1817(10) 3.2115(8)
3.2161(9) 3.1796(8)
2.8007(11) 2.8024(13)
2.7425(11) 2.7437(13)
1.933(8)
1.927(8)
1.922(8)
1.922(7)
1.928(8)
1.903(7)
1.910(7)
1.917(7)
1.945(7)
1.939(7)
1.935(9)
1.921(8)
1.936(8)
1.934(8)
1.902(7)
1.899(7)
1.936(7)
1.925(7)
1.932(7)
1.925(7)
1.907(7)
1.924(7)
1.931(7)
1.946(7)
1.925(7)
1.925(7)
1.959(7)
1.909(11)
1.922(12)
1.916(13)
1.915(12)
1.910(12)
1.918(12)
1.905(10)
1.936(12)
1.911(12)
[Re3{µ-M(PPh3)}(µ3-H)(µ-H)3(CO)9(PPh3)] +
nPPh3 f [Re3(µ-H)4(CO)9(PPh3)]- +
+
[M(PPh3)n+1
]
(8)
literature data,31b the complete conversion of the dif-
ferent adducts into 1 required different amounts of
phosphine: 1 equiv of PPh3 was sufficient for the gold
adduct 3 and 2 equiv was necessary for the copper
derivative 4, while in the case of the silver adduct 2 a
still higher amount was required, due to the formations
of mixtures of the [Ag(PPh3)3]+ and [Ag(PPh3)4]+ cations
(see Experimental Section).
2.4958(17) 2.4793(15) 2.4956(19) 2.474(2)
2.3935(19) 2.2822(16) 2.2127(18) 2.214(3)
P(1)-C(41)
P(1)-C(47)
P(1)-C(53)
P(2)-C(59)
P(2)-C(65)
P(2)-C(71)
1.822(6)
1.810(7)
1.826(7)
1.824(7)
1.814(7)
1.840(7)
1.833(6)
1.836(6)
1.836(6)
1.807(6)
1.824(6)
1.820(7)
1.830(6)
1.824(6)
1.840(6)
1.823(6)
1.817(6)
1.831(6)
1.822(8)
1.839(9)
1.849(8)
1.810(9)
1.814(9)
1.825(9)
Re(2)-Re(1)-Re(3) 64.278(10) 63.135(16) 64.404(12) 63.183(18)
Re(1)-Re(2)-Re(3) 63.031(16) 63.673(16) 63.15(2)
Re(1)-Re(3)-Re(2) 52.69(2) 53.192(9) 52.44(3)
Re(2)-Re(1)-M(1) 60.664(18) 60.259(13) 58.33(3)
Re(1)-Re(2)-M(1) 62.589(18) 61.707(10) 60.36(3)
Re(3)-Re(1)-M(1) 109.09(2) 105.759(19) 111.69(3) 108.67(3)
Re(3)-Re(2)-M(1) 109.57(3) 107.153(18) 112.28(4) 111.13(3)
64.341(18)
52.476(13)
58.35(3)
Partial or complete cleavage of the interactions be-
tween 1 and the [M(PPh3)]+ cations resulted also from
the simple dissolution of these adducts in a donor
solvent such as acetone (eq 9). However, significant
60.40(3)
Re(1)-M(1)-Re(2) 56.75(2)
Re(2)-Re(1)-C(11) 99.7(2)
Re(2)-Re(1)-C(12) 138.7(2)
Re(2)-Re(1)-C(13) 129.6(2)
Re(3)-Re(1)-C(11) 158.5(2)
Re(3)-Re(1)-C(12) 110.7(2)
Re(3)-Re(1)-C(13) 91.4(2)
M(1)-Re(1)-C(11) 71.1(2)
M(1)-Re(1)-C(12) 86.0(2)
M(1)-Re(1)-C(13) 159.2(2)
Re(1)-Re(2)-C(21) 99.1(2)
58.034(16) 61.31(3)
61.25(3)
100.8(3)
138.6(3)
130.3(5)
160.4(3)
109.8(3)
93.9(4)
67.4(3)
90.6(3)
155.9(3)
97.5(3)
139.8(3)
[Re3{µ-M(PPh3)}(µ3-H)(µ-H)3(CO)9(PPh3)] +
Me2CO S [Re3(µ-H)4(CO)9(PPh3)]- +
[M(PPh3)(Me2CO)]+ (9)
100.9(2)
139.8(2)
126.8(3)
160.8(3)
110.4(2)
91.3(3)
101.2(2)
139.6(2)
128.9(2)
160.0(2)
109.5(2)
90.9(2)
71.7(2)
66.9(2)
differences were observed in the behavior of the three
adducts, as described in detail in the Supporting Infor-
mation. The main features are discussed below.
87.0(2)
92.9(2)
162.2(2)
97.1(2)
154.9(2)
100.1(2)
Re(1)-Re(2)-C(22) 139.11(19) 140.63(19) 138.9(2)
In the case of copper, equilibrium 9 was completely
driven to the right, as judged by 1H NMR and IR spectra
and by conductance measurements. Upon elimination
of acetone, the adduct was quantitatively restored. In
contrast, in the case of the gold derivative 3 equilibrium
9 implied the dissociation of a very small amount of 3
(ca. 5%). This equilibrium, which was attained im-
mediately after dissolution, was further driven to the
right by slower side reactions, which irreversibly con-
sumed the gold cation, making only partial the re-
formation of 3 upon removal of acetone. Conductivity
measurements suggested that 3 behaves as a weak
electrolyte, but the aforementioned side reactions ham-
pered the evaluation of equilibrium data.
Also for the silver adduct 2, equilibrium 9 implied the
fast dissociation of a very small amount of adduct. The
reaction was completely reversed upon elimination of
acetone. The extent of dissociation increased with dilu-
tion and on increasing the temperature. A plot of the
reciprocal molar conductivity Λ against cΛ, according
Re(1)-Re(2)-C(23) 131.5(2)
Re(3)-Re(2)-C(21) 156.7(2)
Re(3)-Re(2)-C(22) 112.7(2)
Re(3)-Re(2)-C(23) 93.5(2)
M(1)-Re(2)-C(21) 70.5(2)
M(1)-Re(2)-C(22) 84.1(2)
M(1)-Re(2)-C(23) 156.9(2)
Re(1)-Re(3)-C(31) 109.0(2)
Re(1)-Re(3)-C(32) 155.3(2)
129.7(2)
155.2(2)
114.2(2)
91.5(2)
73.2(2)
84.14(19)
161.3(2)
132.24(18) 131.2(3)
156.8(2)
112.8(2)
92.65(19) 91.4(3)
67.0(2)
88.8(2)
154.20(19) 157.0(3)
110.3(3)
155.62(19) 156.8(3)
78.66(18) 83.0(3)
155.9(4)
113.1(3)
68.2(3)
87.4(3)
108.04(19) 108.2(2)
157.2(2)
Re(1)-Re(3)-C(33) 77.67(18) 81.6(2)
Re(1)-Re(3)-P(1) 99.31(4)
Re(2)-Re(3)-C(31) 160.2(2)
Re(2)-Re(3)-C(32) 105.7(2)
97.84(4)
158.2(2)
104.9(2)
99.33(4)
159.2(2)
106.1(2)
81.94(18) 78.1(3)
101.73(4) 104.12(5)
137.49(6) 146.58(8)
158.01(6) 148.29(9)
95.72(5)
160.0(3)
104.7(3)
Re(2)-Re(3)-C(33) 81.32(18) 77.3(2)
Re(2)-Re(3)-P(1) 101.61(4) 105.12(4)
Re(1)-M(1)-P(2)
Re(2)-M(1)-P(2)
135.84(5) 146.08(4)
156.49(5) 147.43(4)
Re(3)-P(1)-C(41) 115.77(19) 114.89(19) 115.11(19) 114.6(3)
Re(3)-P(1)-C(47) 115.4(2)
Re(3)-P(1)-C(53) 114.2(2)
115.0(2)
116.2(2)
112.8(2)
112.5(2)
111.7(2)
115.4(2)
113.9(2)
117.8(2)
115.0(2)
109.0(2)
115.3(3)
116.1(3)
114.3(3)
113.6(3)
111.1(3)
M(1)-P(2)-C(59)
M(1)-P(2)-C(65)
M(1)-P(2)-C(71)
116.6(2)
114.7(2)
108.7(2)
(see Table 3). The effect is, as expected, slightly more
marked for the gold derivative than for the silver and
copper ones. However, the data reported in Table 3
clearly show that the interaction of the parent anion 1
with the group 11 metal phosphine group results only
in a minor structural perturbation of the Re2(µ-H)2
moiety.
(30) Extrusion of [M(PPh3)n]+ groups by treatment of mixed transi-
tion metal-group 11 metal clusters with PPh3 has been widely
reported. See, for instance, refs 15 and 31. For extrusion from
homonuclear clusters, see ref 32.
(31) (a) Alexander, B. D.; Boyle, P. D.; J ohnson, B. J .; Casalnuovo,
J . A.; J ohnson, S. M.; Mueting, A. M.; Pignolet, L. H. Inorg. Chem.
1987, 26, 2547. (b) Evans, J .; Street, A. C.; Webster, M. Organome-
tallics 1987, 6, 794.
In compounds 2-4 the bridging MPPh3 group and the
rhenium-bound triphenylphosphine lay on the same side
(32) van der Velden, J . W. A.; Bour, J . J .; Bosman, W. P.; Noordik,
J . H. Inorg. Chem. 1983, 22, 1913.

2710 Organometallics, Vol. 21, No. 13, 2002
Beringhelli et al.
Ta ble 3. Com p a r ison of (Mea n ) Bon d P a r a m eter s (Dista n ces in Å a n d An gles in d eg) for th e Ad d ition
Der iva tives [Re3{µ-M(P P h 3)}(µ3-H)(µ-H)3(CO)9(P P h 3)]
1
2 (M ) Ag)
3 (M ) Au)
4m (M ) Cu)
4o (M ) Cu)
Re(1)-Re(2)
2.797
3.214
2.840
3.199
2.987
1.93
1.93
1.91
2.50
1.91
1.94
2.40
2.859
3.192
2.946
1.94
1.93
1.91
2.48
1.92
1.92
2.28
2.827
3.199
2.771
1.93
1.93
1.91
2.50
1.92
1.96
2.21
2.826
3.196
2.773
1.91
1.92
1.92
2.47
1.92
1.91
2.21
Re(1,2)-Re(3)
Re(1,2)-M(1)
Re(1,2)-C(11,21)
Re(1,2)-C(12,22)
Re(1,2)-C(13,23)
Re(3)-P(1)
1.90
1.92
1.92
2.48
1.92
1.96
Re(3)-C(31,32)
Re(3)-C(33)
M(1)-P(2)
Re(3)-Re(1,2)-C(11,21)
Re(3)-Re(1,2)-C(12,22)
Re(3)-Re(1,2)-C(13,23)
Re(3)-Re(1,2)-M(1)
Re(1,2)-Re(2,1)-C(11,21)
Re(1,2)-Re(2,1)-C(12,22)
Re(1,2)-Re(2,1)-C(13,23)
Re(1,2)-Re(3)-P(1)
158.5
111.7
95.6
157.6
111.7
92.4
109.3
99.4
138.9
130.6
100.5
107.4
157.8
79.5
158.0
112.3
91.4
106.4
99.0
140.2
128.2
101.5
106.4
157.7
79.4
158.4
111.1
91.8
158.1
111.5
92.7
109.9
99.1
139.2
130.8
99.9
107.6
158.4
80.5
112.0
100.6
139.3
130.6
100.5
107.2
157.4
80.3
96.9
134.0
135.0
100.7
107.4
156.8
80.8
Re(1,2)-Re(3)-C(31,32)
Re(1,2)-Re(3)-C(32,31)
Re(1,2)-Re(3)-C(33)
[Re(1)Re(2)Re(3)]-[Re(1)Re(2)M(1)]
46.6
50.2
39.0
42.9
Ta ble 4. Su m m a r y of Cr ysta l Da ta , Da ta Collection Deta ils, a n d Str u ctu r e Refin em en t P a r a m eter s
2
3
4m
4o
Crystal Data
C45H34AuO9P2Re3
1536.23
orthorhombic
P212121 (No. 19)
13.710(3)
formula
fw
cryst syst
space group
a (Å)
C
45H34AgO9P2Re3
C45H34CuO9P2Re3
1402.80
monoclinic
P21/n (No. 14)
9.317(5)
C45H34CuO9P2Re3
1402.80
orthorhombic
P212121 (No. 19)
13.659(3)
18.135(5)
18.529(5)
1447.13
monoclinic
P21/ n (No. 14)
9.271(3)
28.522(9)
17.394(5)
96.41(2)
4571(2)
b (Å)
c (Å)
18.183(5)
18.528(5)
28.035(9)
17.556(7)
96.91(4)
â (deg)
V (Å3)
4619(2)
4552(3)
4590(2)
Z
4
4
4
4
F(000)
2712
2840
2640
2640
density (g cm-3
)
)
2.103
2.209
2.047
2.030
abs coeff (mm-1
cryst descripn
cryst size (mm)
8.464
11.124
8.535
8.465
yellow needle
0.10 × 0.08 × 0.03
orange prism
0.34 × 0.14 × 0.14
orange plate
0.38 × 0.38 × 0.08
orange prism
0.34 × 0.32 × 0.26
Data Collection
diffractometer
scan mode
θ range (deg)
index ranges
Siemens SMART
ω
1.8 e θ e 25.0
-11 e h e 11
-33 e k e 33
-20 e l e 20
none
1.6 e θ e 25.0
-16 e h e 16
-21 e k e 21
-22 e l e 22
4
1.4 e θ e 25.0
-11 e h e 11
-33 e k e 33
-20 e l e 20
none
1.6 e θ e 25.0
-16 e h e 16
-21 e k e 21
-22 e l e 22
none
intensity decay (%)
abs corr
SADABS
0.403
min rel transmissn factor
no. of measd rflns
0.665
39 419
8052
0.0542, 0.0413
6462
0.376
0.712
40 310
8096
0.0679, 0.0514
6648
40 720
8128
0.0412, 0.0309
7655
39 297
8032
0.0519, 0.0336
6819
no. of indep rflns
a
R
int, Rσ
no. of rflns with I > 2σ(I)
Refinement
8128/542
0.025
no. of data/params
weights (a)b
8052/542
0.030
8128/542
0.030
8096/542
0.025
goodness of fit S(F2)c
R(F), I > 2σ(I)d
1.089
1.075
1.134
1.048
0.0304
0.0678
0.742, -0.747
0.0216
0.0514
0.669, -0.551
0.0285
0.0685
0.764, -0.808
0.0315
0.0664
0.489, -0.705
Rw(F2), all datae
largest diff peak, hole (e Å-3
)
2
o
2
2
2
2
Rint ) ∑|Fo2 - Fmean |/∑|F |; Rσ ) ∑|σ(F2)|/∑|F2|. w ) 1/[σ2(Fo) + (aP)2], where P ) (Fo + 2F2c)/3. c S ) [∑w(Fo - Fc)2/(n - p)]1/2, where
a
2
b
o
o
2
2
4
n is the number of reflections and p is the number of refined parameters. R(F) ) ∑||Fo| - |Fc||/∑|Fo|. Rw(F2) ) [∑w(Fo- Fc)2/∑wFo]1/2
.
d
e
to the Ostwald dilution law,33 provided a value of [1.5-
(2)] × 10-4 M for the constant of equilibrium 9 at room
temperature.
As to the kinetics of reaction 9, the equilibrium was
attained very quickly for all of the three adducts.
However, strong differences among them have been
observed on the NMR time scale. In the case of the silver
adduct 2, the reversible dissociation of the [Ag(PPh3)]+
cation in acetone was so fast to broaden the resonances
of 2 and of the anion 1 above 233 K (Figure 3) (still not
completely dynamically averaged at room temperature).
In contrast, the [Au(PPh3)]+ fragment did not show any
(33) Fuoss, R. M.; Accascina, F. Electrolytic Conductance; Inter-
science: New York, 1959.

Adducts between [M(PPh3)]- and [Re3(µ-H)4(PPh3)]-
Organometallics, Vol. 21, No. 13, 2002 2711
F igu r e 3. Variable-temperature 1H NMR spectra (hydridic
region) of a 4.6 × 10-4 M solution of 2 (isolated crystals) in
acetone-d6. The peaks labeled with asterisks are due to the
anion 1, originating from dissociation of the adduct in the
diluted acetone solution.
lability in acetone: no exchange between 3 and 1 was
revealed by EXSY experiments, even at room temper-
ature.
The exchange between the silver adduct 2 and 1
occurs also in CD2Cl2, even if with a much slower rate
(the resonances of the two species being sharp up to
room temperature; Figure 4a). In passing, we observe
that the cross-peak between the Ag-coupled signal of 2
and the hydridic resonance of 1 at intermediate field (δ
-8.9) completed the attribution of the resonances of 1,
allowing the identification of the resonance of 1 due to
the hydride syn with respect to the phosphine ligand.
Lability in CD2Cl2 was observed also for the copper
derivative 4 (but not for the gold adduct 3, as expected
due to its inertness in acetone). However, the exchange
pattern revealed by 2D experiments concerning 4 (with
mixing times in the range 0.3-0.8 s) was more complex
than that of 2. Indeed the hydrides Ha and Hb of both
species 1 and 4 had exchange cross-peaks with all the
other resonances of intensity 1, as shown in Figure 4b.
Since in 1 alone the exchange between Ha and Hb does
not occur, on the NMR time scale of this experiment,
such exchange must be mediated by the Ha/Hb exchange
within adduct 4. A 2D EXSY experiment, performed
with a shorter mixing time (0.05 s), supported this view,
since it showed exchange cross-peaks only between the
Ha and Hb signals of 4 (Figure 4c). This exchange must
occur through an intramolecular pathway, since any
dissociative process would imply 4/1 exchange, not
observed with this shortest mixing time. A planar
“lozenge” structure, with Ha and Hb in equivalent
positions, bridging on the two Re-Cu interactions,
might constitute a reasonable transition state for this
F igu r e 4. 1H 2D-EXSY experiments acquired at room
temperature in CD2Cl2 on equimolar mixtures of 1 and the
silver or copper adducts, respectively, in the hydridic
region: (a) 1 and 2, τm ) 0.3 s (the asterisks mark the
signals of 1); (b) 1 and 4, τm ) 0.8 s(the diamonds mark
unidentified impurities); (c) 1 and 4, τm ) 0.05 s.

2712 Organometallics, Vol. 21, No. 13, 2002
Beringhelli et al.
exchange. However, further investigation into this
unusual process is in progress. At longer times the
dissociation of the [Cu(PPh3)]+ fragment causes the 4/1
exchange, as observed for the silver adduct 2.
perturbation with respect to the parent anion 1. The
position of equilibrium 9 cannot be taken as a measure
of the strength of the interaction (copper is weaker than
silver and gold), since such a position depends also on
the strength of the M-acetone bond, whose formation
drives the cleavage of the adducts.
The [M(PPh3)]+ cations exhibit different labilities: the
gold derivative is inert even in acetone, while the copper
and silver adducts are labile (on the NMR time scale)
also in CH2Cl2, at room temperature. Examples of
intermolecular exchange of [M(PPh3)]+ moieties in solu-
tion have already been reported,8 and also the fact that
gold complexes tend to be less dynamic has been
previously pointed out.31b,35
In the related adduct between [Mn2(µ-H)2(CO)6(µ-
dppm)] and [Ag(PPh3)]+, Riera et al. observed the shift
of the cation between the two hydrides of the Mn(µ-
H)2Mn moiety and suggested a dissociative pathway for
such exchange.17 In the present case, such a shift would
afford the anti isomer, which is expected to be less
stable, for steric reasons, and it is not observed. The
exchange between the hydrides Ha and Hb observed for
the copper derivative 4 most likely involves a different,
intramolecular, mechanism. Further investigation on
this subject is in progress.
Con clu sion s. The results reported above have dem-
onstrated that the hydrides of the unsaturated Re(µ-
H)2Re moiety are the sites of electrophilic attack by the
[M(PPh3)]+ cations. However, different from what occurs
with other electrophiles, no hydride (eqs 1 and 2) or
electron (eq 3) transfer takes place, and the reactions
stop at the initial stage: i.e., the formation of the
hydride-bridged acid-base adduct. This likely results
both from the low tendency of group 11 atoms to give
hydridic derivatives and from the stabilization of the
Re3H-‚‚‚MPPh3 interactions for the formation of the
+
butterfly Re3M metal skeletons. Moreover, the cleavage
of the adducts 2-4 by reaction with phosphines (eq 8)
demonstrated that in this case it is the [M(PPh3)]+
cation that would act as a scavenger of the L′ ligands,
and not the Re3 unsaturated fragment, as required by
eq 1.
The M(PPh3) fragment can be considered to be isolo-
bal with a hydrido ligand,18,19 and experimentally it has
been found that in almost all compounds containing just
one Au(PPh3) fragment this fragment occupies a position
similar to that of the hydrido ligand in the related
hydrido derivative.8,19 On these grounds the adducts
2-4 could provide a model of the kinetic product
resulting from the attack of the proton on the hydride
of the unsaturated anions [Re3(µ-H)4(CO)9L]-. From this
point of view these adducts should be formulated as
triangular clusters containing a bridging η2-HM(PPh3)
ligand,14 isolobally analogous to an (until now unknown)
bridging η2-H2 ligand.
The nature of these adducts in solution requires a
short discussion. In principle they could be viewed either
as Lewis acid-base adducts or as tight ionic couples
between the group 11 cations and the triangular cluster
anion. Accordingly, the (partial or complete) dissociation
in acetone could be attributed either to the Lewis
basicity of the solvent or to its high dielectric constant,
favoring the cleavage of the ionic couple. We observed
that the dissociation of the copper adduct was complete
not only in acetone but also in a donor solvent with a
low dielectric constant such as THF, while it was
negligible in a solvent such as nitromethane, with a high
dielectric constant but a low donor capability. This rules
out the hypothesis of an ionic interaction. Moreover, the
high values of the coupling constants observed in CD2-
Cl2 solution between the triply bridging hydride and the
31P atoms of the [M(PPh3)]+ groups (and with the Ag
isotopes in the case of compound 2) fully support the
idea of a true covalent interaction.
Exp er im en ta l Section
The reactions were performed under nitrogen, using the
Schlenk technique, and solvents were deoxygenated and dried
by standard methods. Literature methods were used for the
preparation of [Re4(µ3-H)4(CO)12],36 [PPh4](5),29 [Au(PPh3)Cl],37
and [Cu(PPh3)2NO3].38 [Cu(CH3CN)4][BF4] was obtained by a
literature method,39 using HBF4 instead of HPF6. [PPh4](1) was
obtained quantitatively by treating the [PPh4]+ salt of 5 with
a stoichiometric amount of PPh3 in CH2Cl2 at room tempera-
ture. PPh3 (Merck) and AgOTf (Aldrich) were used as received.
The NMR spectra were acquired on Bruker AC200 and
DRX300 spectrometers, the former equipped with an external
BSV3 unit. IR spectra were acquired on a Bruker Vector22
FT spectrometer. Microanalyses were performed at the mi-
croanalytical laboratory of the University of Milan.
Syn th esis of [Re3{µ-Ag(P P h 3)}(µ3-H)(µ-H)3(CO)9(P P h 3)]
(2). A solution of 32.4 mg (0.0229 mmol) of [PPh4](1) in CH2-
Cl2 (5 mL) was treated at room temperature with 1.53 mL of
an equimolar solution (0.015 M) of PPh3 and AgOTf in CH2-
Cl2/acetone (3/1). The solution instantaneously became dark
yellow, and IR monitoring showed the quantitative formation
of adduct 2. Addition of freshly distilled Et2O to the solution
(after reduction of the volume to ca. 1 mL, under vacuum)
caused the precipitation of an off-white solid, identified as
[PPh4]OTf. The bright yellow solution was separated and
evaporated to dryness. Crystallization of the residue from CH2-
Cl2/n-hexane afforded a yellow solid (isolated yield 18.5 mg,
0.0128 mmol, 56%, after drying in vacuo). Anal. Calcd for
C
45H34AgO9P2Re3: C, 37.32; H, 2.37. Found: C, 37.4; H, 2.4.
Single crystals suitable for X-ray analysis were obtained by
low-temperature (-25 °C), slow diffusion of n-hexane into a
concentrated CH2Cl2 solution. IR and NMR data: see Table
1.
However, such interactions must be considered quite
weak, since a solvent with a donor power as poor as that
of acetone is sufficient to extrude (partially or com-
pletely) the [M(PPh3)]+ cations. A similar behavior was
previously observed for other weak adducts, such as
those between the Mn(µ-H)2Mn moiety and [M(P-
Ph3)]+.17,34 The weakness of such adducts is substanti-
ated also by the X-ray structures, which show little
1H-31P HMQC experiments (7.05 T, 300 K), optimized for
J HP ) 4 and 15 Hz, were performed on a solution prepared by
(35) Albinati, A.; Lehner, H.; Venanzi, L. M.; Wolfer, M. Inorg. Chem.
1987, 26, 3933.
(36) J ohnson, J . R.; Kaesz, H. D. Inorg. Synth. 1978, 18, 60.
(37) Bruce, M. I., Nicholson, B. K.; Bin Shawkataly, O. Inorg. Synth.
1989, 26, 324.
(38) Gysling, H. J . Inorg. Synth. 1979, 19, 92.
(39) Kubas, G. J . Inorg. Synth. 1979, 19, 90.
(34) No reaction was observed between [Mn2(µ-H)2(CO)6(µ-dppm)]
and [Au(PPh3)]+ in THF, but it did occur upon removal of the solvent
and dissolution in CH2Cl2.17

Adducts between [M(PPh3)]- and [Re3(µ-H)4(PPh3)]-
Organometallics, Vol. 21, No. 13, 2002 2713
dissolving 8.0 mg (0.0055 mmol) of 2 in 0.5 mL of CD2Cl2.
These experiments showed that (i) Ha has a (small) correlation
only with the rhenium-bound phosphorus P(1) due to the
coupling measured in the 1D spectrum (see Table 1), (ii) the
triply bridging hydride Hb has a small correlation (not sizable
in the 1D spectrum) also with P(1), in addition to the strong
correlation with P(2), and (iii) the two hydrides Hc bridging
on the lateral edges have a small correlation (not sizable) also
with P(2), in addition to the strong coupling with P(1).
Syn th esis of [Re3{µ-Au (P P h 3)}(µ3-H)(µ-H)3(CO)9(P P h 3)]
(3). A solution of Au(PPh3)Cl (26.8 mg, 0.0542 mmol) in CH2-
Cl2 (1 mL) was treated with 380 µL of a 0.1454 M solution of
AgOTf in acetone. The precipitate (AgCl) was separated by
slow decantation, and 340 µL of the solution (about 0.0134
mmol of [Au(PPh3)]+) was added to a solution of [PPh4](1) (19.0
mg, 0.0134 mmol) in CH2Cl2 (3 mL). The color changed
immediately from yellow to orange, and IR monitoring showed
the quantitative formation of 3. The solution was treated as
described in the previous synthesis, affording orange microc-
rystals of 3 (15 mg, 0.0097 mmol, 73% isolated yields after
of PPh3 (2.0 mg, 0.0078 mmol) showed the quantitative
formation of 1. 31P NMR showed, besides the signal of 1, also
a broad signal at δ 29.2 ppm, attributable to dynamic averag-
ing32 of the signals of free phosphine and of the [Au(PPh3)2]+
cation (δ 45.4).41
(c) Ad d u ct 4. Addition of 1 equiv of PPh3 (8.5 µL of a 0.126
M solution of PPh3 in CD2Cl2) to a solution of 4 (1.4 mg, 0.001
mmol) in CD2Cl2 (0.5 mL) caused the formation of an equimo-
lar mixture of 1 and 4. Addition of another 1 equiv of PPh3
completed the conversion of 4 into 1. 31P NMR (183 K) showed,
besides the signal of 1, also a signal at δ 1.87 ppm, attributable
to the [Cu(PPh3)3]+ cation.31b
1H 2D NMR In vestiga tion of th e Beh a vior of Ad d u cts
2-4 in CD2Cl2 in th e P r esen ce of a n Equ im ola r Am ou n t
of An ion 1. Three samples were prepared by dissolving in CD2-
Cl2 roughly equimolar amounts (typically 7 × 10-3 mmol) of
1
[PPh4](1) and of one of the adducts 2-4. H 2D EXSY phase-
sensitive experiments (7.05 T)42 were performed for each
sample at 300 K with different mixing times (τm) and relaxation-
delay values. For both the Ag and Cu derivatives, the experi-
ments performed with τm ) 0.3 s showed chemical exchange
with compound 1, while for the Au derivative, even with the
mixing time τm ) 0.8 s, no exchange was detected. For the
copper derivative 4, experiments with τm ) 0.8 and 0.05 s were
also performed. For the observation of NOE effects between
the hydrides and the phenylic hydrogens of the phosphines
(as described above) further experiments were performed on
the mixture of 1 and 2, on using longer relaxation delays and
mixing times (2.0 and 1.0 s, respectively). Typically 16 fid’s
were acquired on a spectral width of 7184 Hz (4496 Hz for 4)
using 1K data points for each of the 256 experiments. Shifted
sine-bell functions were applied in both dimensions before
Fourier transform and after zero-filling to 1K in F1.
crystallization from CH2Cl2/n-hexane). Anal. Calcd for C45H34
-
AgO9P2Re3: C, 35.16; H, 2.23. Found: C, 35.4; H, 2.4.
Syn th esis of [Re3{µ-Cu (P P h 3)}(µ3-H)(µ-H)3(CO)9(P P h 3)]
(4). (a ) F r om th e An ion 1. A CH2Cl2 solution of [PPh4](1)
(14.4 mg, 0.0102 mmol) was treated at room temperature with
425 µL of an equimolar solution (0.024 M) of [Cu(CH3CN)4]-
[BF4] and PPh3 in CH2Cl2. The solution turned dark yellow.
The reaction mixture was stirred for about 5 min, concentrated
to a small volume, and treated with Et2O, causing the
precipitation of [PPh4][BF4]. The yellow solution was separated
and evaporated to dryness, and the residue was crystallized
from CH2Cl2/n-hexane. The yellow solid so obtained contained
again some impurity, on the basis of NMR analysis (the main
signals being a doublet at -6.43 ppm, J HP ) 16 Hz, and two
singlets at -8.60 and -12.67 ppm; overall integrated intensity
ca. 20% of the whole intensity in the hydride region). Pure 4
was obtained by slow diffusion of n-hexane into a concentrated
CH2Cl2 solution (4.3 mg, 0.003 mmol, 29%).
(b) F r om th e An ion 5. A solution of [Re4(µ3-H)4(CO)12] (42.9
mg, 0.0346 mmol) in CH2Cl2 (6 mL) was treated with 250 µL
of acetone at room temperature. The solution was stirred for
ca. 60 min until the red-brown color completely disappeared
and the solution became bright yellow. Addition of [Cu(PPh3)2-
(NO3)] (22.7 mg, 0.0346 mmol) caused a sudden change of the
color at first to red and then to yellow. The product was
purified by chromatography on a silica gel column, by elution
with CH2Cl2/n-hexane (1/1). Concentration under vacuum
afforded a yellow precipitate, which was isolated and crystal-
lized by slow diffusion of n-hexane into CH2Cl2 solution (21.0
mg of crystals, 0.0149 mmol, isolated yield 43%). Anal. Calcd
for C45H34AgO9P2Re3: C, 38.50; H, 2.44. Found: C, 38.4; H,
2.5.
X-r a y Diffr a ction Str u ctu r a l An a lysis. (a ) Collection
a n d Red u ction of X-r a y Diffr a ction Da ta . Suitable crystals
were mounted in air on a glass fiber tip onto a goniometer
head. Single-crystal X-ray diffraction data were collected on a
Siemens SMART CCD area detector diffractometer, using
graphite-monochromated Mo KR radiation (λ ) 0.710 73 Å)
at room temperature (293(2) K). Unit cell parameters were
initially obtained from about 100 reflections (5° < θ < 20°)
taken from 45 frames collected in three different ω regions
and eventually refined against a large amount of reflections
(about 8000). A full sphere of reciprocal space was scanned by
0.3° ω steps, collecting 1800 frames each at 40 s exposure and
keeping the detector at 3.88(2) cm from the sample. Intensity
decay was monitored by recollecting the initial 100 frames at
the end of data collection and analyzing the duplicate reflec-
tions. The collected frames were processed for integration
(SAINT43); an empirical absorption correction was made on
the basis of the symmetry-equivalent reflection intensities
collected (SADABS44). Crystal data and data collection param-
eters are summarized in Table 4.
Rea ction s of Ad d u cts 2-4 w ith P P h 3. (a ) Ad d u ct 2. A
solution of 2 (13.8 mg, 0.0095 mmol) in CD2Cl2 (0.5 mL) was
treated at room temperature with 2.5 mg of PPh3 (0.0095
mmol). The 1H NMR spectrum (183 K) showed the hydridic
resonances of both 2 and 1, in a ca. 1:1 ratio. 31P NMR showed,
besides the signals of 2 and 1, also two doublets of doublets
(ratio ca. 1.2:1, one component of the lowest field doublet
partially overlapping with signals of 1 and 2), attributable
respectively to [Ag(PPh3)4]+ (δ 5.68, J (107AgP) ) 223 Hz) and
[Ag(PPh3)3]+ (δ ca. 11.5, J (107AgP) ) ca. 300 Hz).40 Addition of
another 1 equiv of PPh3 did not cause the complete disappear-
ance of 1 (residual amount ca. 25%), and 31P NMR showed the
increase of the signal of [Ag(PPh3)4]+ with respect to that of
[Ag(PPh3)3]+ (ratio ca. 2:1).
(b) Str u ctu r e Solu tion a n d Refin em en t. The structures
were solved by direct methods (SIR9745) and subsequent
Fourier synthesis; they were refined by full-matrix least
squares on F2 (SHELX9746) using all reflections. Scattering
factors for neutral atoms and anomalous dispersion corrections
were taken from the internal library of SHELX97. Weights
were assigned to individual observations according to the
(41) Sykes, A. G.; Mann, K. R. J . Am. Chem. Soc. 1990, 112, 7247.
(42) Bodenhausen, G.; Kogler, H.; Ernst, R. R. J . Magn. Reson. 1984,
58, 370.
(43) SAINT; Bruker AXS, Madison, WI, 1993-1999.
(44) Sheldrick, G. M. SADABS: Program for Empirical Absorption
Correction; University of Go¨ttingen, Go¨ttingen, Germany, 1996.
(45) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna,
R. J . Appl. Crystallogr. 1999, 32, 115.
(b) Ad d u ct 3. The 1H NMR spectrum (300 K) of a CDCl3
solution containing 3 (8.0 mg, 0.0052 mmol) and a slight excess
(40) Alyea, E. C.; Malito, J .; Nelson, J . H. Inorg. Chem. 1987, 26,
4294.
(46) Sheldrick, G. M. SHELX97: Program for Crystal Structure
Refinement; University of Go¨ttingen, Go¨ttingen, Germany, 1997.

2714 Organometallics, Vol. 21, No. 13, 2002
Beringhelli et al.
formula w ) 1/[σ2(Fo2) + (aP)2], where P ) (Fo2 + 2F2c)/3; the
parameter a was chosen to give a flat analysis of variance in
terms of F2o. Anisotropic displacement parameters were as-
signed to all non-hydrogen atoms. Hydride ligands were
initially evidenced by a difference Fourier map; their positions
were in agreement with the local stereogeometry around the
metal centers. They were eventually constrained in more
accurate positions (calculated employing the program HY-
DEX,47 with dRe-H ) 1.85 Å for both edge- and face-bridging
ligands) and refined with a common (refined) isotropic dis-
placement parameter. Hydrogen atoms of the triphenylphos-
phine ligands were placed in idealized positions (dC-H ) 0.93
Å) and refined riding on their parent atom with an isotropic
displacement parameter 1.2 times that of the pertinent carbon
atom. For compounds 3 and 4o the correctness of the absolute
structure was checked, and the Flack parameters48 for the
correct enantiomer were 0.007(6) and 0.007(9), respectively.
The final difference electron density maps showed no features
of chemical significance, with the largest peaks lying close to
the metal atoms. Final conventional agreement indexes and
other structure refinement parameters are listed in Table 4.
Ack n ow led gm en t. This work has been supported
by the Italian CNR (CSMTBO) and by the MIUR
(Project Metal Clusters: Basic and Functional Aspects).
Thanks are due to Mr. Pasquale Illiano for the careful
acquisition of the 2D NMR data.
Su p p or tin g In for m a tion Ava ila ble: Text giving details
on the reaction of 1 with Cu(NCMe)4+/PPh3 monitored in the
NMR tube and on the behavior of adducts 2-4 in acetone
solutions, conductivities of acetone solutions of 2 and 4 (Table
S1), time evolution of 3 in acetone (Figure S1); ORTEP
drawings for 4m and 4o (Figures S2 and S3), and final atomic
coordinates, anisotropic displacement parameters, and tables
giving all bond distances and angles for 2, 3, 4m , and 4o. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(47) Orpen, A. G. J . Chem. Soc., Dalton Trans. 1980, 2509.
(48) Flack, H. D. Acta Crystallogr. 1983, A39, 876.
OM0200252