Unusual C - H allylic activation in the {PtII(cod)} fragment bonded to a {Pt2(μ-S)2} core

Article abstract of DOI:10.1021/om034312x

Complexes [{Pt₂(μ₃-S)₂(dppp) ₂}Pt(cod)]Cl₂ (1) and [{Pt₂(μ₃-S) ₂(cod)₂}Pt(dppp)]Cl₂ (3), where dppp = 1,3-bis(diphenylphosphino)propane and cod = 1,5-cyclooctadiene, have been synthesized by reacting [Pt₂(μ-S)₂(dppp)₂] and [PtCl₂(cod)] (1:1), and [Pt(SH)₂(dppp)] and [PtCl ₂-(cod)] (1:2), respectively. Complex 1 has not allowed substitution of cod by the chelating dppp ligand. Remarkably, the reaction of 1 with methoxide anion yields [{Pt₂(μ₃-S)₂(dppp) ₂}Pt(C₈H₁₁)]Cl (2), which entails deprotonation of cod instead of the nucleophilic attack of CHaO^- on the olefmic bond. In addition, replacement of the deprotonated cod ligand in 2. by dppp has not been achieved. A combination of experimental data and DFT calculations in 2 is consistent with the binding of C₈H₁₁^- to platinum(II) by means of one η²-alkene and one η¹-allyl bond. The structures of 1 and 2 have been confirmed by single-crystal X-ray diffraction. Analogous to 1, the reaction of 3 with sodium methoxide causes the subsequent deprotonation of the two cod ligands, yielding [{Pt₂(μ₃-S)₂(cod)(C₈H ₁₁)}Pt(dppp)]Cl (4) and [{Pt₂(μ₃-S) ₂(C₈H₁₁)₂}Pt(dppp)] (5). In contrast to 1, replacement of cod by dppp in 3 and 4 leads to 1 and 2, respectively. Also, the substitution of one C₈H₁₁^- ligand by dppp in 5 leads to 2. On the basis of DFT calculations, with inclusion of solvent effects, the factors governing the chemical behavior of the {Pt(cod)}²⁺ fragment bonded to a [Pt₂(μ-S) ₂L₄] (L₂ = dppp, cod, or C₈H ₁₁^-) metalloligand are discussed.

Full text of DOI:10.1021/om034312x

2522
Organometallics 2004, 23, 2522-2532
Articles
Un u su a l C-H Allylic Activa tion in th e {P tII(cod )}
F r a gm en t Bon d ed to a {P t2(µ-S)2} Cor e
Rube´n Mas-Balleste´,† Paul A. Champkin,‡ William Clegg,‡
Pilar Gonza´lez-Duarte,*,† Agust´ı Lledo´s,*,† and Gregori Ujaque†
Departament de Qu´ımica, Universitat Auto`noma de Barcelona,
E-08193 Bellaterra, Barcelona, Spain, and School of Natural Sciences (Chemistry),
University of Newcastle, Newcastle upon Tyne NE1 7RU, U.K.
Received November 21, 2003
Complexes [{Pt2(µ3-S)2(dppp)2}Pt(cod)]Cl2 (1) and [{Pt2(µ3-S)2(cod)2}Pt(dppp)]Cl2 (3), where
dppp ) 1,3-bis(diphenylphosphino)propane and cod ) 1,5-cyclooctadiene, have been syn-
thesized by reacting [Pt2(µ-S)2(dppp)2] and [PtCl2(cod)] (1:1), and [Pt(SH)2(dppp)] and [PtCl2-
(cod)] (1:2), respectively. Complex 1 has not allowed substitution of cod by the chelating
dppp ligand. Remarkably, the reaction of 1 with methoxide anion yields [{Pt2(µ3-S)2(dppp)2}-
Pt(C8H11)]Cl (2), which entails deprotonation of cod instead of the nucleophilic attack of
CH3O- on the olefinic bond. In addition, replacement of the deprotonated cod ligand in 2 by
dppp has not been achieved. A combination of experimental data and DFT calculations in 2
-
is consistent with the binding of C8H11 to platinum(II) by means of one η2-alkene and one
η1-allyl bond. The structures of 1 and 2 have been confirmed by single-crystal X-ray
diffraction. Analogous to 1, the reaction of 3 with sodium methoxide causes the subsequent
deprotonation of the two cod ligands, yielding [{Pt2(µ3-S)2(cod)(C8H11)}Pt(dppp)]Cl (4) and
[{Pt2(µ3-S)2(C8H11)2}Pt(dppp)] (5). In contrast to 1, replacement of cod by dppp in 3 and 4
leads to 1 and 2, respectively. Also, the substitution of one C8H11- ligand by dppp in 5 leads
to 2. On the basis of DFT calculations, with inclusion of solvent effects, the factors governing
the chemical behavior of the {Pt(cod)}2+ fragment bonded to a [Pt2(µ-S)2L4] (L2 ) dppp, cod,
or C8H11-) metalloligand are discussed.
In tr od u ction
reactivity of cationic platinum-olefin complexes, which
show an enhanced electrophilicity at the olefinic moi-
ety,4,3 has often been hampered by the lability of the
unsaturated ligand in these complexes.5 With respect
to the {MII(cod)} fragment in palladium or platinum
complexes, the reaction with a wide range of bases has
led in most cases to the nucleophilic attack of these
species to the olefinic bonds,6 base-promoted deproto-
nation being observed only if the attacking species is a
tertiary amine.7
Activation of the C-H bond in olefins bound to metal
centers is relevant in organometallic chemistry because
of its significance in many catalytic processes.1 Fre-
quently, this activation is achieved by reacting the
metal-olefin complex with a base. However, this reaction
may cause not only deprotonation of the olefin but,
alternatively, the nucleophilic attack of the base to the
double bond.2 Both reactions are depicted in Scheme 1
for a {Pt(η2-propene)} fragment, where for the depro-
tonation reaction only the more usual C-H allylic
activation has been taken into account. While the
nucleophilic attack at olefinic bonds coordinated to
palladium(II) and platinum(II) has been well studied,
unlike palladium, examples of base-promoted olefin
deprotonation reactions in neutral platinum-olefin com-
plexes are very scarce.3 In addition, the study of the
In the quest to explore the remarkable chemistry of
the {Pt2(µ-S)2} core, its influence on the reactivity of a
{PtII(cod)} fragment attracted our interest. The striking
(3) Bandoli, G.; Dolmella, A.; Fanizzi, F. P.; Di Masi, N. G.; Maresca,
L.; Natile, G. Organometallics 2002, 21, 4595.
(4) Bandoli, G.; Dolmella, A.; Di Masi, N. G.; Fanizzi, F. P.; Maresca,
L.; Natile, G. Organometallics 2001, 20, 805.
(5) (a) Albietz, P. J .; Yang, K.; Lachicotte, R. J .; Eisenberg, R.
Organometallics 2000, 19, 3543. (b) Fusto, M.; Giordano, F.; Orabona,
I.; Ruffo, F.; Panunzi, A. Organometallics 1997, 16, 5981.
(6) (a) Macchioni, A.; Bellachioma, G.; Cardaci, G.; Travaglia, M.;
Zuccacia, C.; Milani, B.; Corso, G.; Zangrado, E.; Mestroni, G.;
Carfagna, C.; Formica, M. Organometallics 1999, 18, 2677. (b) Hoel,
G. R.; Stockland, R. A.; Anderson, G. K.; Ladipo, F. T.; Braddock-
Wilking, J .; Rath, N. P.; Mareque-Rivas, J . C. Organometallics 1998,
17, 1155.
* Corresponding authors.
† Universitat Auto`noma de Barcelona.
‡ University of Newcastle.
(1) Crabtree, R. H. The Organometallic Chemistry of the Transition
Metals; J ohn Wiley and Sons: New York, 2001.
(2) (a) Hahn, C.; Morvillo, P.; Herdtweck, E.; Vitagliano, A. Orga-
nometallics 2002, 21, 1807. (b) Vicente, J .; Chicote, M.-T.; MacBeath,
C.; Ferna´ndez-Baeza, J .; Bautista, D. Organometallics 1999, 18, 2677.
(c) Hamed, O.; Henry, P. M. Organometallics 1997, 16, 4903.
(7) (a) Peters, J . C.; Harkins, S. B.; Brown, S. D.; Day, M. W. Inorg.
Chem. 2001, 40, 5083. (b) Dahan, F.; Agami, C.; Levisalles, J .; Rose-
Munch, F. Chem. Commun. 1974, 505.
10.1021/om034312x CCC: $27.50 © 2004 American Chemical Society
Publication on Web 04/17/2004

Unusual C-H Allylic Activation
Organometallics, Vol. 23, No. 11, 2004 2523
Sch em e 1
nucleophilicity of the bridging sulfur atoms together
with the flexible hinge angle between the two PtIIS2
planes in [L2{Pt(µ-S)2Pt}L2] (L ) phosphine) metallo-
ligands account for (a) their outstanding behavior as
precursors to a large family of homo- and heterometallic
complexes;8 (b) their spontaneous evolution in the
presence of organic electrophilic agents, such as CH2-
Cl2;9 (c) the cascade of reactions following protonation
of the bridging sulfur atoms in {Pt2S2};10 and (d) the
first evidence of fast intramolecular S-H‚‚‚S proton
transfer in a transition metal complex.11 Thus, on the
basis of the high electron density on the sulfur atoms
in [L2{Pt(µ-S)2Pt}L2] compounds, it was reasonable to
expect that their coordination to a {PtII(cod)} fragment
should have an effect on the electrophilicity of the cod
ligand and possibly modify the reactivity of the {PtII-
(cod)} fragment toward attack by nucleophiles.
of the [(dppp){Pt(µ-S)2Pt}(dppp)] metalloligand on the
reactivity of the platinum-bonded olefin.
Resu lts a n d Discu ssion
Exp er im en ta l Resu lts. The sequence of reactions
carried out in this study is shown in Scheme 2. NMR
parameters and mass determinations for the complexes
1-5 thus obtained are given in Table 1. Details of the
synthetic pathways followed are given in the Experi-
mental Section.
Complex 1 was formed as the only product of the
following reaction in benzene solution:
[(dppp){Pt(µ-S)2Pt}(dppp)] + [PtCl2(cod)] f
[{Pt2(µ3-S)2(dppp)2}Pt(cod)]Cl2 (1) (A)
Remarkably, all attempts to replace the cod ligand
in 1 by dppp or chloride anions were unsuccessful. Thus,
different reaction conditions, such as the solvent and
temperature of the reaction as well as the molar ratio
of the reagents, never led to the already known [Pt3(µ3-
S)2(dppp)3]Cl2 complex.10 These observations are con-
sistent with previously reported results for related
compounds containing the {RhI(cod)} fragment bonded
to a {Pt2(µ-S)2} core, [{Pt2(µ3-S)2(PPh3)4}Rh(cod)]+ and
[{Pt2(µ3-S)2(-diop)2}Rh(cod)]+.13 However, in contrast to
the striking stability toward substitution of the cod
ligand in 1, this complex reacts with NaCH3O or Na2S
under mild conditions, affording [{Pt2(µ3-S)2(dppp)2}Pt-
(C8H11)]Cl (2). While this formula agreed well with the
ESI MS and MALDI TOF parameters of 2 in solution,
the coordinative behavior of the deprotonated cod
ligand, (C8H11)-, was deduced from the DEPT-135 NMR
spectrum, displaying five different CH and three dif-
ferent CH2 groups. 13C-195Pt couplings have been
observed for the signals at δ ) 83.29 (1J Pt-C ) 144 Hz)
and 81.35 (1J Pt-C ) 176 Hz), which can thus be assigned
In this paper we report the synthesis and character-
ization of [{Pt2(µ3-S)2(dppp)2}Pt(cod)]Cl2 (1) and [{Pt2-
(µ3-S)2(cod)2}Pt(dppp)]Cl2 (3), whose cationic species
contain one or two {PtII(cod)} fragments, respectively.
Both cations, in contrast with the general behavior of
cationic olefin complexes of platinum, do not undergo
spontaneous loss of the unsaturated ligand and, thus,
have allowed exploration of their reactivity toward a
strong nucleophile such as NaCH3O. Unlike previous
reports on the attack of NaCH3O on the {PtII(cod)}
fragment,12 the above reactions have not promoted the
nucleophilic attack of the methoxide anion, but only
deprotonation of cod, thus leading to [{Pt2(µ3-S)2(dppp)2}-
Pt(C8H11)]Cl (2), [{Pt2(µ3-S)2(cod)(C8H11)}Pt(dppp)]Cl (4),
and [{Pt2(µ3-S)2(C8H11)2}Pt(dppp)] (5). In addition, the
replacement of cod by dppp in complexes 1-4 has been
studied. Overall, a combination of experimental data
and theoretical calculations shows the noninnocent role
(8) (a) Fong, S.-W. A.; Hor, T. S. A. J . Chem. Soc., Dalton Trans.
1999, 639, and references therein. (b) Capdevila, M.; Carrasco, Y.;
Clegg, W.; Coxall, R. A.; Gonza´lez-Duarte, P.; Lledo´s, A.; Ram´ırez, J .
A. J . Chem. Soc., Dalton Trans. 1999, 3103. (c) Yam, V. W.-W.; Yeung,
P. K.-Y.; Cheung, K.-K. Angew. Chem Int. Ed. Engl. 1996, 35, 739.
(9) Mas-Balleste´, R.; Capdevila, M.; Champkin, P. A.; Clegg, W.;
Coxall, R. A.; Lledo´s, A.; Me´gret, C.; Gonza´lez-Duarte, P. Inorg. Chem.
2002, 41, 3218.
(10) Mas-Balleste´, R.; Aullo´n, G.; Champkin, P. A.; Clegg, W.;
Me´gret, C.; Gonza´lez-Duarte, P.; Lledo´s, A. Chem. Eur. J . 2003, 9,
5023.
(11) Aullo´n, G.; Capdevila, M.; Clegg, W.; Gonza´lez-Duarte, P.;
Lledo´s, A.; Mas-Balleste´, R. Angew. Chem., Int. Ed. 2002, 41, 2776.
(12) (a) Aucott, S. M.; Slawin, A. M. Z.; Wollins, J . D. J . Chem. Soc.,
Dalton Trans. 2000, 2559. (b) Bhattacharyya, P.; Slawin, A. M. Z.;
Smith, M. B. J . Chem. Soc., Dalton Trans. 1998, 2467. (c) Slawin, A.
M. Z.; Smith M. B.; Wollins, J . D. J . Chem. Soc., Dalton Trans. 1996,
3659. (d) Giordano, F.; Vitagliano, A. Inorg. Chem. 1981, 20, 633. (e)
Bamberi, G.; Forsellini, E.; Gonziani, R. J . Chem. Soc., Dalton Trans.
1972, 525.
(13) (a) Brunner, H.; Weber, M.; Zabel, M. J . Organomet. Chem.
2003, 684, 6. (b) Briant, C. E.; Gilmour, D. I.; Luke, M. A.; Mingos, D.
M. P. J . Chem. Soc., Dalton Trans. 1985, 851.

2524 Organometallics, Vol. 23, No. 11, 2004
Mas-Balleste´ et al.
Sch em e 2
Ta ble 1. NMR P a r a m eter s a n d Ma ss Deter m in a tion s for Com p lexes 1-5
mass data (ESI
and/ or MALDI)
calculated
MW
δ(31P),
1J Pt-P,
Hz
δ(1H) of cod ligand,
δ(13C) of cod ligand,
complex
ppma
ppma
ppma
[Pt2(µ-S)2(dppp)2] 1279 (ESI)
1279.1
1582.4
1581.4
-0.1
2615
3069
3006
1
2
790.9b (ESI)
-4.1
-6.3
4.54 (CH), 2.53 (CH2)
96.9 (CH), 29.8 (CH2)
790.5c (ESI)
1581.1 (MALDI)
4.27 (CH), 4.04 (CH), 3.09
83.3 (CH,1J Pt-C ) 144 Hz),
(CH), 2.81 (CH), 2.27 (CH), 81.3 (CH,1J Pt-C ) 176 Hz),
2.82 (CH2), 2.01 (m, CH2)
75.2 (CH,1J Pt-C ) 598 Hz),
54.2 (CH), 26.3 (CH), 33.9
(CH2), 29.0 (CH2), 28.3 (CH2)
[Pt(SH)2(dppp)]
3
640.1d (ESI)
638.9b (ESI)
673.6
1278.1
-1.3
-2.6
2760
3193
5.49 (CH), 4.99 (CH), 2.5
(m, CH2)
5.3-5.1 (br, CH), 4.8-4.6
(br, CH)
101.0 (CH), 98.8 (CH), 30.1
(CH2)
4
638.7c (ESI)
1277.1
-3.7 (m)
96.4-95.0 (m, CH), 85.2 (CH),
84.5 (CH), 83.4 (CH), 83.2
(CH), 79.0 (CH), 78.3 (CH),
54.6 (CH), 54.5 (CH), 28.0
(CH), 27.3 (CH) 34.5 (CH2),
34.2 (CH2), 30.5-29.7 (m,
CH2), 28.9 (CH2), 28.8 (CH2),
28.5 (CH2), 28.1 (CH2)
85.1-83.2 (m, CH), 79.2-78.0
(m, CH), 73.1-72.3 (m, CH),
54.4-54.1 (m, CH), 26.8-25.3
(m, CH); 35.6-34.3 (m, CH2),
29.4-27.8 (m, CH2)
1277.1 (MALDI)
5
638.9c (ESI)
1276.1
-6.1 (m)
5.2-3.8 (br, m, CH)
t1276.9 (MALDI)
a
b
All 31P, 1H, and DEPT-135 13C NMR spectra were recorded in d6-dmso solution. Abbreviations: m, multiplet; br, broad.
M
2+/2.
c Protonated species as a result of the formic acid added in ESI-MS measurements. [(dppp)Pt(SH)]+ fragment.
d
to a platinum(II)-coordinated double bond, and for the
signal at δ ) 75.18 (1J Pt-C ) 598 Hz) corresponding to
a σ-coordinated carbon atom. These features are con-
-
sistent with (C8H11
)
being bound to platinum(II) by
means of one η2-alkene and one η1-allyl bond. Full
structural characterization of 2 has required a combina-
tion of crystallography and theoretical calculations, both
discussed below. On the basis of the three sets of data
(NMR, X-ray, and theoretical calculations) it has been
deduced that the isomer labeled as allyl A in Figure 1
corresponds to 2 either in the solid phase or in solution.
Interestingly, the reaction leading from 1 to 2 can
easily be reversed. Thus, addition of a stoichiometric
amount of dilute HCl to a solution of 2 converts it to 1.
-
F igu r e 1. Possible binding modes of the (C8H11
in a [L2Pt(C8H11)] complex.
) ligand
2
The H NMR spectrum of a solution of 2 after addition
of DCl instead of HCl reveals an allylic position for the
added hydrogen atom and thus confirms that the base-

Unusual C-H Allylic Activation
Organometallics, Vol. 23, No. 11, 2004 2525
promoted deprotonation of cod in 1 corresponds to a
C-H allylic activation. Additionally, we have observed
-
that the behavior of the anionic deprotonated (C8H11
)
ligand in 2 toward replacement by dppp or chloride
anions is comparable to that of neutral cod in 1, both
being substitutionally inert ligands.
Complex 3, an analogue of 1, but with one dppp ligand
replaced by cod, has been obtained in accordance with
the reaction:
[Pt(SH)2(dppp)] + 2[PtCl2(cod)] f
[{Pt2(µ3-S)2(cod)2}Pt(dppp)]Cl2 (3) + 2HCl (B)
which requires the slow addition of a solution of [Pt-
(SH)2(dppp)] to a solution of [PtCl2(cod)], so that an
excess of the latter reagent is always present in the
reaction medium. Otherwise, a mixture of complexes 1
and 3 is obtained, formation of 1 probably being due to
the reaction
2[Pt(SH)2(dppp)] + [PtCl2(cod)] f
[{Pt2(µ3-S)2(dppp)2}Pt(cod)]Cl2 (1) + 2SH2 (C)
Overall, reactions A and B lead to pure complexes 1
and 3, respectively, under the experimental conditions
here described. A mixture of both 1 and 3 has been
obtained only if reactions B and C are allowed to proceed
simultaneously, which requires the presence of [Pt(SH)2-
(dppp)] as well as a [Pt(SH)2(dppp)]/[PtCl2(cod)] ratio
greater than 1:2. Notably, the reaction of [Pt2(µ-Se)2-
(PPh3)4] with [PtCl2(cod)], which is comparable to reac-
tion A, affords a mixture of [{Pt2(µ3-Se)2(PPh3)4}Pt-
(cod)]2+ and [{Pt2(µ3-Se)2(cod)2}Pt(PPh3)2]2+, closely
related to 1 and 3, respectively.14
The acid-base reactions involving 3 compare well
with those of 1. Thus, the addition of 1 equiv of NaCH3O
to a solution of 3 yields [{Pt2(µ3-S)2(cod)(C8H11)}Pt-
(dppp)]Cl (4), which, after subsequent additions of base,
evolves to [{Pt2(µ3-S)2(C8H11)2}Pt(dppp)] (5). Within the
same context of the acid-base chemistry, the reactions
from 3 to 4, and eventually to 5, are easily reversed by
addition of the corresponding amount of dilute hydro-
F igu r e 2. Labeling of the carbon atoms of cod ligand in 3
and possible isomers of 4 and 5.
the presence of the [(dppp){Pt(µ-S)2Pt}(dppp)] moiety
is responsible for the outstanding stability of 1 and
2.
chloric acid. With respect to the binding modes of the
-
(C8H11
)
ligands in 4 and 5, 13C DEPT 135 NMR data
NMR characterization of complexes 4 and 5 requires
consideration of their corresponding isomeric forms,
which are depicted in Figure 2. In addition, the NMR
parameters of the related complexes 1-3 provide a
reference for a comparative analysis. All NMR data are
given in Table 1. Thus, the 13C{1H} and 1H NMR spectra
provide evidence that the carbon and hydrogen atoms
of cod in 3, in contrast to 1, are not symmetrically
equivalent, as shown by the two groups of signals
assigned to the CH groups. Consequently, deprotonation
clearly indicate that they all bind to platinum(II) by
means of one η2-alkene and one η1-allyl bond, as already
found in 2. As theoretical calculations for the latter
complex and other model compounds show that the allyl
A isomeric form (Figure 1) has the greatest stability
(described below), the same disposition has been as-
-
sumed for all (C8H11
)
ligands in 4 and 5.
The replacement reaction of cod by dppp ligands in 3
and 4 shows marked differences from the results
observed for the same reaction with 1 and 2. Thus, the
reaction of 3 with dppp promotes replacement of one of
the two cod ligands and leads to 1. That of 4 with dppp
entails substitution of the uncharged cod ligand only
and yields 2. These results indicate that 1 and 2 are
particularly stable complexes, not only substitutionally
inert, but also the only products obtained in the reac-
tion of 3 and 4 with dppp. Comparison of both fami-
lies of complexes (1, 2 vs 3, 4) clearly indicates that
of cod in 3 to afford 4 gives rise to two isomers as long
-
as the binding behavior of (C8H11
)
corresponds to the
allyl A form exclusively. This assumption is fully
validated by the 13C DEPT 135 NMR spectrum of 4,
which displays two sets of signals corresponding to the
deprotonated (C8H11)- ligand, each set including five CH
and three CH2 groups. These data not only denote two
different dispositions of the (C8H11)- ligand in 4 but also
show that their binding to platinum occurs by means
of η2-alkene and η1-allyl coordination modes. The choice
of the allyl A with respect to the allyl B form is discussed
(14) Yeo, J . S. L.; Vittal, J . J .; Henderson, W.; Hor, T. S. A. Inorg.
Chem. 2002, 41, 1194.

2526 Organometallics, Vol. 23, No. 11, 2004
Mas-Balleste´ et al.
Ta ble 2. Rela tive En er gies (k ca l/m ol) for th e
Differ en t Isom er s of Com p lexes [L2P t(C8H11)]
theoretical values, the vinyl isomer can be discarded,
in agreement with the spectroscopic data indicating that
the product of the deprotonation reaction contains a Pt-
allyl group. Moroever, the allyl A isomer can be proposed
gas phase
solutiona
allyl
(A)
allyl
(B)
allyl
(A)
allyl
(B)
as the observed isomer for complex 2, and the preference
vinyl
vinyl
-
for this binding mode of the (C8H11
)
ligand can be
2 Cl-
0.0
0.0
0.0
0.0
0.0
5.4
4.6
7.8
5.5
7.3
31.7
29.8
34.7
27.4
32.3
0.0
0.0
0.0
0.0
0.0
3.1
4.2
7.1
4.4
6.4
28.9
27.3
32.9
27.3
32.5
2 SH-
extended to complexes 4 and 5.
Sta bility of [L2P t(cod )]z (z ) 0, +2) Com p lexes
tow a r d cod Disp la cem en t. The cod ligand is usually
a somewhat labile ligand, which can be displaced easily
by ligands with higher coordinating ability such as
bidentate phosphines. However, this is not the case in
compound 1. To elucidate the reasons for the surprising
stability of the cod ligand in 1 toward substitution by
dppp, the reaction energy (∆ELE) for the ligand exchange
reaction (D) has been calculated for a series of eight
related [L2Pt(cod)]z complexes, taking H2P-(CH2)3-PH2
(dhpp) as a model of the dppp ligand.
[Pt2(µ-S)2(PH3)4]
2 PH3
2 SH2
a
Calculated in methanol (ꢀ ) 32.63) using the COSMO con-
tinuum solvation model.
below. The coexistence of isomers 4.1 and 4.2 (Figure
2) in solution is corroborated by the 31P{1H} NMR of 4
showing two superimposed double doublets.
Following the same lines of reasoning as above,
deprotonation of the two cod ligands in 3 should give
rise to 5 in six isomeric forms, as shown in Figure 2.
The coexistence in solution of all or several of these
[L2Pt(cod)]z + dppp f [L2Pt(dppp)]z + cod (D)
isomers accounts for the multiplet signal displayed for
-
each carbon nucleus of (C8H11
)
in the 13C{1H} NMR
The complexes of the series are neutral (z ) 0) when L
ligands are SH-, I-, Br-, or Cl-, and cationic (z ) +2)
when L ligands are PH3, SH2, [Pt2(µ-S)2(dhpp)2] (com-
plex 1), or [Pt2(µ-S)2(dhpp)(cod)] (complex 3). Consider-
ation of a series of [L2Pt(cod)] complexes with ligands
L of different natures should enable us to put the
energetic values for the ligand exchange obtained for
complexes 1 and 3 into a general context. The computed
∆ELE values are shown in Table 3.
spectra. In parallel, the 31P{1H} NMR spectrum of 5
consists of a broad signal.
Th eor etica l Stu d y
-
Bin d in g Mod e of th e (C8H11
)
Liga n d . Deproto-
nation of the cod ligand can occur via a vinylic or an
allylic proton, thus giving rise to different coordination
-
modes of the (C8H11
)
ligand. Whereas in the former
case only one isomer can be obtained (vinyl in Figure
1), the abstraction of an allylic proton can generate two
isomers, depending on which of the two terminal allyl
carbon atoms is bonded to the metal (allyl A and allyl
B in Figure 1). Thus, three different isomers must be
considered. The relative stabilities of these isomers in
the gas phase and in methanol solution were evaluated
for a series of [L2Pt(COD)]z (z ) 0, L2 ) 2SH-, 2Cl-;
z ) +2, L2 ) [Pt2(µ-S)2(PH3)2], 2PH3, and 2SH2) com-
plexes (Table 2). The relative stabilities of the three
isomers are practically the same for all the complexes
studied, both in the gas phase and in solution. The total
charge of the complex does not affect the relative
stabilities: very similar relative energies among the
isomers are obtained for the neutral and dicationic
species. The vinyl isomer is highly unstable compared
to both allylic isomers. Regarding the two allylic iso-
mers, for all the complexes studied allyl A is more stable
than allyl B by a range of 4.5-8 kcal/mol in the gas
phase and 3-7 kcal/mol in solution. The coordination
geometry of the platinum center in isomer B shows a
high distortion from planarity. In contrast, isomer A
presents a less strained structure. These structural
features could explain the higher stability of isomer A.
The relative energies calculated for a model of com-
plex 2 (L2 ) [Pt2(µ-S)2(PH3)2]) fully agree with the
complete series of the studied complexes. From these
The results indicate that substitution of cod by dppp
is, in all studied cases, highly favored on thermodynamic
grounds. The process is highly exothermic, with ∆ELE
ranging from -18.7 to -37.5 kcal/mol. The substitution
of a cod ligand by dppp is more exothermic for cationic
complexes than for neutral ones; this statement agrees
well with the higher lability of olefinic ligands generally
observed for cationic Pt-alkene complexes.5 Our results
justify the replacement of cod by dppp ligands observed
in 3 and 4, but they cannot explain the stability toward
cod displacement exhibited by complex 1. Indeed they
suggest that the factors hampering the exchange of the
cod ligand by dppp in 1 should be of a kinetic nature.
The bond dissociation energies (BDEs) for cod and
dppp ligands in this series of complexes were also
evaluated (see Table 3). Determination of these ther-
modynamic parameters can provide a mechanistic in-
sight, because BDEs give an approximate estimation of
the energy barrier for a dissociative ligand-exchange
process. The results presented here show that the Pt-
cod bond energies are higher for cationic [L2Pt(cod)]
complexes than for neutral ones. High and similar
values of BDEPt-cod have been calculated for cationic
complexes with L2 ) [Pt2(µ-S)2(dhpp)2] and [Pt2(µ-S)2-
(dhpp)(cod)], despite the very different behavior toward
the cod/dppp exchange observed in 1 and 3. The high
BDEPt-cod obtained indicates that the dissociative mech-
Ta ble 3. Rea ction En er gies for th e cod /d p p p Liga n d Exch a n ge Rea ction a n d P t-cod a n d P t-d p p p Bon d
Dissocia tion En er gies in [L2P t(cod )] Com p lexes (in k ca l/m ol)
L2
2SH-
2I-
2Br-
2Cl-
[Pt2(µ-S)2(dhpp)2]a
[Pt2(µ-S)2(dhpp)(cod)]
2PH3
2SH2
∆ELE
-18.7
42.2
60.7
-23.7
57.4
81.1
-23.2
65.5
88.7
-22.5
73.6
96.1
-28.2
78.9
107.1
-28.7
81.7
110.4
-33.6
116.5
150.1
-37.5
140.1
177.6
BDE(Pt-cod)
BDE(Pt-dppp)
a
dhpp ) H2P(CH2)3PH2.

Unusual C-H Allylic Activation
Organometallics, Vol. 23, No. 11, 2004 2527
Ta ble 4. Rea ction En er gies (in k ca l/m ol) for MeO-
Nu cleop h ilic Atta ck or Dep r oton a tion Rea ction s
in [L2P t(cod )] Com p lexesa
L2
2SH- 2Cl- [Pt2(µ-S)2(PH3)4] 2PH3 2SH2
nucleophilic
attack
-28.4 -41.3
-50.8
-62.9 -69.2
deprotonation -12.7 -23.9
-40.0
-52.5 -58.5
a
Calculated in methanol (ꢀ ) 32.63) using the COSMO con-
tinuum solvation model.
barrier. This insight could be extended to the behavior
of the reported complexes containing the {Pt2(µ3-S)2Rh-
(cod)}2+ core, where cod has shown an enhanced stabil-
ity toward ligand-exchange reactions.13
Dep r oton a tion ver su s Nu cleop h ilic Atta ck on
th e cod Liga n d . As far as the reactivity of complexes
1 and 3 toward MeO- is concerned, the reaction
proceeds with the abstraction of an allylic proton instead
of the common nucleophilic attack of the methoxide
anion on the double bond. DFT calculations using a
continuum solvation model were carried out to try to
understand this unexpected reactivity. The inclusion of
solvent effects is mandatory to obtain a reliable estima-
tion of reaction energies, given the changes in the
charges experienced by the species throughout the
reaction. The reaction energy for both processes, depro-
tonation (reaction E) and nucleophilic attack (reaction
F) in methanol (ꢀ ) 32.63), for a set of complexes of
formula [L2Pt(cod)]z (z ) 0, L2 ) 2SH--, 2Cl-; z ) +2,
L2 ) [Pt2(µ-S)2(PH3)2], 2PH3, and 2SH2) were evaluated
(Table 4).
F igu r e 3. Molecular models of complexes [{Pt2(µ-S)2-
(cod)2}Pt(dppp)]2+ (A), [{Pt2(µ-S)2(dppp)2}Pt(cod)]2+ (B), and
[Pt3(µ3-S)2(dppp)3]2+ (C). Arrows indicate the space avail-
able for the attack of an additional ligand.
anism is highly unfavorable in 1 and 3 and suggests an
associative mechanism for the replacement of cod by
dppp. This mechanistic proposal agrees with others
found in the literature; for instance, the ligand exchange
of olefins in d8 square-planar complexes is supposed to
go through an associative mechanism.15 Nevertheless,
associative mechanisms are subject to the achievability
of a transition state with five ligands coexisting in the
coordination sphere of the metal center. In the [L2Pt-
(cod)]z complexes under study, when L is bulky, as for
the [Pt2(µ-S)2(dppp)2] metalloligand, the transition state
for the associative mechanism would be highly unstable
or may not even exist. Thus, the high steric hindrance
due to [Pt2(µ-S)2(dppp)2] in 1 hampers the associative
mechanism for the replacement of cod ligand by dppp
to yield [Pt3(µ3-S)2(dppp)3] and confers a remarkable
kinetic stability toward cod dissociation on 1. Molecular
[L2Pt(cod)]z + CH3O- f [L2Pt(C8H11)]z-1+ CH3OH
(E)
[L2Pt(cod)]z + CH3O- f [L2Pt(C8H12OCH3)]z-1 (F)
These results show that, for all the different com-
plexes studied, both reactions are highly exothermic, the
cationic presenting a higher exothermicity than the
neutral species. Comparing the reaction energies for
both processes, the nucleophilic attack of methoxide
anion on a cod double bond is thermodynamically more
favorable than deprotonation for all the studied species.
Nevertheless, these differences in the reaction energy
are lower for cationic species than for neutral ones.
Given these tendencies, although the nucleophilic attack
is always thermodynamically favored, the deprotonation
of the cod ligand becomes more feasible for cationic
complexes. These features suggest that the experimen-
tally observed reactivity for 1 and 3 toward MeO-,
giving rise to deprotonation, is a kinetically controlled
process.
models of the reactant ([{Pt2(µ3-S)2(dppp)2}Pt(cod)]+2
)
and product ([Pt3(µ3-S)2(dppp)3]+2) for the cod/dppp
exchange reaction illustrate the difficulties encountered
by dppp when approaching the platinum atom bound
to cod and [Pt2(µ-S)2(dppp)2] metalloligand simulta-
neously (Figure 3).
The cod/dppp ligand exchange is experimentally
observed in 3, bearing the metalloligand [Pt2(µ-S)2-
(dppp)(cod)]. It is likely that, when cod has substituted
one dppp ligand, the lower bulkiness of the metallo-
ligand allows the associative mechanism to take place.
The different hindrance imposed by the presence of cod
or dppp in the [Pt2(µ-S)2(dppp)(L)] fragment can be
inferred from Figure 3. In conclusion, the theoretical
study indicates that, despite the replacement of cod by
dppp in 1 to yield [{Pt2(µ3-S)2(dppp)2}Pt(dppp)]2+ being
an exothermic process, the reaction is kinetically unfa-
vorable because neither associative nor dissociative
mechanisms are possible with a reasonable energetic
To evaluate the aforementioned hypothesis, the reac-
tion profiles for the nucleophilic attack by the methoxide
anion and the deprotonation of the cod ligand were
calculated for two different [L2Pt(cod)] complexes, one
neutral (L2) 2Cl-) and one of a cationic nature (L2 )
[Pt2(µ-S)2(PH3)4]). To study methoxide attack, the Csp2
-
Omethoxide distance was selected as the reaction coordi-
nate. The reaction profiles for the nucleophilic attack
found for both complexes in the gas phase and in
methanol do not present any barrier. This is not a
surprising result, which had already been found by
Norrby et al. in a thorough theoretical study of nucleo-
(15) (a) Canovese, L.; Visentin, F.; Uguagliati, P.; Crociani, B. J .
Chem. Soc., Dalton Trans. 1996, 1921. (b) J ohnson, L. K.; Killian, C.
M.; Brookhart, M. J . Am. Chem. Soc. 1995, 117, 6414. (c) van Asselt,
R.; Elsevier, C. J .; Smeets, W. J . J .; Spek A. L. Inorg. Chem. 1994, 33,
1521.

2528 Organometallics, Vol. 23, No. 11, 2004
Mas-Balleste´ et al.
F igu r e 4. Deprotonation reaction profiles in gas phase and solution for complexes [PtCl2(cod)] and 1.
philic attack on cationic (η3-allyl)palladium complexes.16
In this study, no transition state could be located for
the reaction between an anionic nucleophile and a
cationic (allyl)palladium complex. The authors con-
cluded that this reaction does not have an energy barrier
in the gas phase and that the barrier observed in real
reactions can largely be attributed to the desolvation
energy of the reacting ions. For the addition of anionic
nucleophiles to cationic Pd-allyl complexes a transition
state was found in solution only when constructing a
two-dimensional potential energy surface.16 The explo-
ration of the bidimensional potential surface for the
reactions under study greatly increases the computa-
tional requirements, and it is beyond the scope of this
work.
The deprotonation reaction profiles in the gas phase
and in solution for both complexes are shown in Figure
4. The selected reaction coordinate for this process is
the Hallyl-Omethoxide distance, which varies from 1.6 to
1.0 Å. In the case of the neutral [Cl2Pt(cod)] complex,
the reaction has no barrier in the gas phase, whereas
it shows an energy barrier of ca. 6 kcal/mol in solution.
For the cationic [Pt2(µ3-S)2(PH3)4}Pt(cod)]2+ complex, the
reaction does not present any barrier, either in the gas
phase or in solution. To obtain more accurate reaction
profiles, it would also be necessary to construct a two-
dimensional potential energy surface in solution. Nev-
ertheless, the exploration of these potential surfaces also
demands too much computer time and was not per-
formed.
Although the obtained results are not definitely
conclusive, they can provide qualitative ideas about the
unexpected reactivity exhibited by the cod ligand in 1.
The fact that in solution the cationic species do not
present a barrier for proton abstraction, but neutral
species do, suggests that deprotonation of 1 may be
easier than that of [Cl2Pt(cod)]. Moreover, although the
nucleophilic attack is always thermodynamically more
favorable than deprotonation, the energy differences
between the two processes are reduced for cationic
species, also suggesting that the deprotonation reaction
is more feasible for cationic compounds. These results
are consistent with the hypothesis that the observed
reactivity of complex 1 with methoxide anion is a
F igu r e 5. Molecular structure of 1 with the key atoms
labeled and 50% probability ellipsoids. H atoms, Cl-anions,
and CH3Cl solvent molecules have been omitted.
Ta ble 5. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for 1
Pt(1)-S(1)
Pt(2)-S(1)
Pt(3)-S(1)
Pt(1)-P(1)
Pt(2)-P(3)
Pt(3)-C(55)
Pt(3)-C(59)
C(55)-C(62)
2.375(4)
2.372(3)
2.342(3)
2.261(3)
2.263(3)
2.195(13)
2.218(10)
1.346(16)
Pt(1)-S(2)
Pt(2)-S(2)
Pt(3)-S(2)
Pt(1)-P(2)
Pt(2)-P(4)
Pt(3)-C(58)
Pt(3)-C(62)
C(58)-C(59)
2.365(3)
2.381(3)
2.326(3)
2.260(4)
2.264(3)
2.198(10)
2.216(12)
1.399(16)
S(1)-Pt(1)-S(2)
S(1)-Pt(3)-S(2)
P(3)-Pt(2)-P(4)
79.82(10)
81.29(11)
93.72(11)
S(1)-Pt(2)-S(2)
P(1)-Pt(1)-P(2)
79.55(10)
92.59(13)
consequence of kinetic factors. It is likely that the
bulkiness of the [Pt2(µ-S)2(dppp)2] metalloligand causes
steric interactions with the incoming nucleophile and
also plays a role in favoring deprotonation.
Molecu la r Str u ctu r es
X-r a y Cr ysta l Str u ctu r e of [{P t2(µ3-S)2(d p p p )2}-
P t(cod )]Cl2 (1). The crystal structure of this complex
consists of [{Pt2(µ3-S)2(dppp)2}Pt(cod)]2+ cations (Figure
5, Table 5) and Cl- counterions held together by
electrostatic interactions, and chloroform solvent mol-
ecules. The cation has no crystallographic symmetry,
but has an approximately D3h Pt3S2 core. The molecular
(16) Hagelin, H.; Akermark, B.; Norrby, P.-O. Chem. Eur. J . 1999,
5, 902.

Unusual C-H Allylic Activation
Organometallics, Vol. 23, No. 11, 2004 2529
Ta ble 6. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for 2
Pt(1)-S(1)
Pt(3)-S(1)
Pt(2)-P(2)
Pt(3)-C(30)
2.3592(11)
2.3789(11)
2.2519(12)
2.103(10)
Pt(2)-S(1)
Pt(1)-P(1)
Pt(3)-C(29)
Pt(3)-C(34)
2.3705(11)
2.2558(11)
2.064(13)
2.230(10)
S(1)-Pt(1)-S(1′)
S(1)-Pt(3)-S(1′)
P(2)-Pt(2)-P(2′)
81.42(5)
80.60(5)
95.72(6)
S(1)-Pt(2)-S(1′)
P(1)-Pt(1)-P(1′)
80.95(5)
94.93(6)
However, the binding of the deprotonated cod ligand to
platinum deserves special consideration, as disorder of
this ligand and the solvent molecules across a mirror
plane (which contains all three Pt atoms, bisects each
diphosphine ligand, and relates the two bridging sulfide
ligands to each other) resulted in serious overlap of the
atomic arrangements (Figure 7) and hampered reliable
determination of the C-C distances and positions of H
atoms (see the Experimental Section for details of the
disorder treatment). As shown in Figure 8, the C(29)-
C(30) distance of 1.38 Å in the deprotonated cod ligand
is consistent with a double bond and thus with one η2-
alkene binding mode of (C8H11)- to platinum. Moreover,
the Pt(3)-C(34) distance of 2.23 Å is compatible with a
second coordination mode of either allylic (allyl A type)
or vinylic nature. Concomitantly, the position of C(33),
and thus the C(32)-C(33) and C(33)-C(34) distances,
could be interpreted as indicative of an η1-vinyl binding
mode for the cod ligand, though the difference in these
bond lengths is small (and insignificant statistically, in
view of the relatively large standard uncertainties for
these disordered atoms). In contrast, NMR data for
[{Pt2(µ3-S)2(dppp)2}Pt(C8H11)]+ are consistent with η1-
allyl rather than η1-vinyl bonding, and DFT calculations
show not only that the Pt(η1-vinyl) coordination is
strongly disfavored with respect to η1-allyl but also that
the allyl A isomer (Figure 1) is the energetically
preferred form. Comparison of the structural data
obtained by crystallography and DFT (Figure 8) shows
that this apparent inconsistency is mainly due to the
observed position of C(33), which is particularly affected
by overlap with a solvent molecule in the disorder model
and hence is especially unreliable. We thus deduce that
the binding of (C8H11)- to platinum probably occurs by
means of η2-alkene and η1-allyl coordination modes in
the solid state and in solution.
F igu r e 6. Molecular structure of 2 with the key atoms
labeled and 50% probability ellipsoids. H atoms, ClO4
anions, and CH2Cl2 solvent molecules have been omitted.
-
structure of the trinuclear cation can be considered as
formed by one [(dppp){Pt(µ-S)2Pt}(dppp)] metalloligand
linked to a {Pt(cod)} moiety through two sulfur atoms.
The central Pt3S2 unit consists of a triangle of platinum
atoms capped above and below by two sulfur atoms, thus
describing a distorted trigonal bipyramid. These two
triply bridging sulfido ligands lie essentially equidistant,
1.524 Å above and 1.516 Å below the Pt3 plane.
However, deviation from the ideal geometry is evidenced
by the dihedral angles between pairs of the three PtS2
planes, which range from 114.4° to 124.8°. The geomet-
ric features of the Pt3S2 unit compare well with those
observed in the [{Pt2(µ3-S)2(dppp)2}Pt(C8H11)]+ deriva-
tive, described below, and with related complexes
containing the P6Pt3S2 core with monodentate and
chelating phosphine ligands.8b,10,17 Moreover, the overall
structure of [{Pt2(µ3-S)2(dppp)2}Pt(cod)]2+ bears a close
resemblance to that of its selenium analogue containing
a Pt3Se2 core.14
Each platinum atom in [{Pt2(µ3-S)2(dppp)2}Pt(cod)]2+
has square-planar coordination, distortion from the ideal
geometry being mainly due to the reduction of the
S-Pt-S angles from 90°. The two {PtS2(dppp)} subunits
show an antisymmetric arrangement of the PtP2C3
rings, both in a chair conformation, which become
symmetry-equivalent in solution according to NMR
data.
Con clu d in g Rem a r k s
The chemistry of [{Pt2(µ3-S)2(dppp)2}Pt(cod)]2+ has
shown to be remarkable on the basis of (a) the high
stability of the cod ligand toward substitutiton by dppp
and (b) the ease of proton abstraction from the cod
ligand as a result of the reaction with the methoxide
anion. The preference of cod for deprotonation rather
than the usual attack of the nucleophile on the olefinic
bond is also shown in the reaction of [{Pt2(µ3-S)2(cod)2}-
Pt(dppp)]2+ and [{Pt2(µ3-S)2(cod)(C8H11)}Pt(dppp)]+ with
sodium methoxide. Overall, the unprecedented behavior
of cod in three cationic complexes containing a [{Pt2-
(µ3-S)2}Pt(cod)]2+ moiety (as 1, 3, and 4) reveals the
strong influence of the {Pt2(µ-S)2} core on the reactivity
of the {Pt(cod)}2+ fragment. According to DFT calcula-
tions on a series of [L2Pt(cod)]z species, the competition
between proton abstraction and nucleophilic attack is
highly dependent on the charge of the complex, the
X-r a y Cr ysta l Str u ctu r e of [{P t2(µ3-S)2(d p p p )2}-
P t(C8H11)](ClO4) (2). The crystal structure of this
complex consists of [{Pt2(µ3-S)2(dppp)2}Pt(C8H11)]+ cat-
-
ions (Figure 6, Table 6) and ClO4 counterions held
together by electrostatic interactions, and solvent mol-
ecules. The main structural features of the trinuclear
cation can be described analogously to the parent [{Pt2-
(µ3-S)2(dppp)2}Pt(cod)]2+ complex, both consisting of a
central Pt3S2 unit with a trigonal bipyramidal geometry.
Distortion from the ideal geometry is slightly greater
for [{Pt2(µ3-S)2(dppp)2}Pt(C8H11)]+, where the dihedral
angles between PtS2 planes range from 112.7° to 125.7°.
(17) (a) Li, Z.; Mok, H. F.; Batsanov, A. S.; Howard, J . A. K.; Hor,
T. S. A. J . Organomet. Chem. 1999, 575, 223. (b) Pilkington, M. J .;
Slawin, A. M. Z.; Williams, D. J .; Woollins, J . D. J . Chem. Soc., Dalton
Trans. 1992, 2425. (c) Bushnell, G. W.; Dixon, K. R.; Ono, R.; Pidcock,
A. Can. J . Chem. 1984, 62, 696.

2530 Organometallics, Vol. 23, No. 11, 2004
Mas-Balleste´ et al.
F igu r e 7. Representation of the overlap of atomic arrangements corresponding to the {Pt(C8H11)} fragment in 2.
nature of the terminal ligands bound to the central
{Pt2S2} core.
Exp er im en ta l Section
Ma ter ia ls a n d Meth od s. All reactions were carried out
under an atmosphere of pure dinitrogen, and conventionally
dried and degassed solvents were used throughout. These were
Purex Analytical Grade from SDS. Solutions of NaOMe in
MeOH (4.74 M) and DCl in D2O (35 wt %, 99.9% D) were
purchased from Aldrich. Metal complexes of formula [PtCl2-
F igu r e 8. (A) Structural parameters obtained by X-ray
(dppp)] and [PtCl2(cod)] were prepared according to published
diffraction for complex 2. Numbers in bold characters are
inconsistent with the proposal of an allyl A structure for
the deprotonated cod ligand. (B) Structural parameters
calculated by means of DFT calculations for the allyl A
isomer.
methods.18 The synthesis of [Pt(SH)2(dppp)]10 and [(dppp)Pt-
(µ-S)2Pt(dppp)]8b,9 has already been reported.
Complexes 1-5 have been characterized by NMR and mass
spectroscopy. 1H, 13C{1H}, and 31P{1H} NMR spectra were
recorded from samples in (CD3)2SO solution at room temper-
ature, using a Bruker AC250 spectrometer. 13C and 1H
chemical shifts are relative to SiMe4, and 31P chemical shifts
to external 85% H3PO4. The 2H NMR spectra of the product
reaction of 2 with DCl were recorded from (CH3)2SO solution
using a Bruker Avance500 spectrometer. The ESI MS mea-
former reaction appearing as more feasible in cationic
(as 1 and 3) than in neutral platinum complexes.
Concerning the substitution of cod (in neutral or
anionic form) by dppp, the five complexes obtained
behave differently. Those containing the {Pt2(µ2-S)2-
(dppp)2} moiety (as 1 and 2) are substitutionally inert.
By contrast, those based on the {Pt2(µ2-S)2(dppp)L}
moiety with L ) cod or C8H11- (as 3, 4, and 5) undergo
the substitution reaction readily. However, DFT calcu-
lations show that replacement of cod by dppp in 1 or 3
is an exothermic process. Consequently, kinetic effects
should account for the reluctance of cod to be replaced
by dppp in 1 and 2. These effects can be attributed to
the steric hindrance imposed by the {Pt2(µ2-S)2(dppp)2}
unit, which disfavors an associative pathway for the
ligand exchange in 1. This factor, combined with the
high binding energy we have found for the cod ligand
in a series of cationic platinum complexes, precludes the
substitution of cod by dppp in 1.
surements were performed on
a VG Quattro Micromass
instrument. Experimental conditions are given elsewhere.9
MALDI mass spectra were obtained on a Voyager DE-RP
(PerSeptive Biosystems, Framighan) time-of-flight (TOF) mass
spectrometer equipped with a nitrogen laser (337 nm 3 ns
pulse). The accelerating voltage in the ion source was 20 kV.
Data were acquired in the reflector mode operation with a
delay time value of 60 ns. Experiments were performed using
the matrix 2,5-dihydroxybenzoic acid [DHB, Aldrich]. Matrix
solutions were prepared by dissolving 10 mg of DHB in 10 mL
of CH3CN/H2O 50% (v/v). Equal volumes (1 µL + 1 µL) of
matrix solution and diluted samples (10-2 M in CH3CN) were
mixed and spotted onto the stainless steel sample plate. The
mixture was dried in air before being introduced into the mass
spectrometer. Elemental analyses were performed on a Carlo-
Erba CHNS EA-1108 analyzer. However, as already found in
some related phosphine platinum complexes,9-11,19 microana-
lytical data referring to the carbon content in complexes 1, 3,
and 4 were unsatisfactory.
Finally, the unusual reactivity exhibited by the cod
ligand in complexes containing the [{Pt(µ3-S)2Pt}Pt-
(cod)]2+ moiety and the dependence of such reactivity
on electronic and steric factors involving the {Pt2S2} core
allow envisaging new perspectives. On the basis of the
results here reported it can be proposed that the
coordination of [Pt2(µ-S)2L4] metalloligands to a transi-
tion metal MLn′ fragment can be an effective tool for
modifying its chemical properties. In addition, these
modifications could be modulated by a fine-tuning of the
Syn t h esis of [{P t 2(µ3-S)2(d p p p )2}P t (cod )]Cl2 (1). To a
stirring solution of [Pt2(µ3-S)2(dppp)2] (0.5 g, 0.39 mmol) in
benzene (100 mL) was added solid [PtCl2(cod)] (0.15 g, 0.39
mmol). The resulting clear solution was allowed to stir for 5 h
at room temperature, during which time complex 1 formed as
(18) Brown, M. P.; Puddephatt, R. J .; Rashidi, M.; Seddon, K. R. J .
Chem. Soc., Dalton Trans. 1977, 951.
(19) Capdevila, M.; Gonza´lez-Duarte, P.; Foces-Foces, C.; Herna´ndez
Cano, F.; Mart´ınez-Ripoll, M. J . Chem. Soc., Dalton Trans. 1990, 143.

Unusual C-H Allylic Activation
Organometallics, Vol. 23, No. 11, 2004 2531
Ta ble 7. Cr ysta llogr a p h ic Da ta for 1 a n d 2
a white solid. Then, it was filtered off and washed with diethyl
ether. Yield: 78%. X-ray quality crystals were grown by slow
evaporation of a solution of 1 in CHCl3. Anal. Calcd for
1
2
+
formula
C
62H64P4Pt3S22+‚2Cl-‚
C
62H63P4Pt3S2 ‚
C
62H64P4Pt3S2Cl2: C, 45.04; H, 3.90; S, 3.88. Found: C, 44.21;
3CHCl3
2011.4
ClO4-‚ CH2Cl2
1765.8
H, 3.70; S, 3.74.
fw
Syn th esis of [{P t2(µ3-S)2(d p p p )2}P t(C8H11)]Cl (2). To a
suspension of 1 (0.5 g, 0.30 mmol) in benzene (75 mL) was
added an equimolar amount of NaMeO in methanol (63 µL of
a 4.74 M solution). After 8 h of stirring at room temperature
the suspension was concentrated to ca. 20 mL under reduced
pressure. Addition of diethyl ether to the filtrated solution
yielded a pale yellow solid. Yield: 62%. X-ray quality crystals
were obtained in CH2Cl2 after exchanging chloride by perchlo-
rate anions. Anal. Calcd for C62H63P4Pt3S2Cl: C, 46.06; H, 3.93;
S, 3.97. Found: C, 45.62; H, 4.21; S, 3.63.
cryst syst
space group
a (Å)
b (Å)
c (Å)
monoclinic
P21/n
monoclinic
P21/m
13.5107(7)
20.5840(10)
28.7743(14)
99.474(2)
7893.1(7)
4
12.8372(12)
14.8717(14)
19.6316(19)
106.160(2)
3599.8(6)
2
120
24 256
10 158
â (deg)
V (Å3)
Z
T (K)
160
42 657
no. of reflns measd
no. of unique reflns 10 305
Syn th esis of [{P t2(µ3-S)2(cod )2}P t(d p p p )]Cl2 (3). A solu-
tion of [Pt(SH)2(dppp)] (0.22 g, 0.33 mmol) in benzene (50 mL)
was added dropwise to a stirring solution of [PtCl2(cod)] (0.25
g, 0.67 mmol) in the same solvent (50 mL) and the reaction
mixture left to stir for 6 h at room temperature. After this
time the solution was concentrated to ca. 25 mL and the yellow
solid thus formed was filtered off and dried. Yield: 45%. Anal.
Calcd for C43H50P2Pt3S2Cl2: C, 38.28; H, 3.74; S, 4.75. Found:
C, 36.94; H, 3.89; S, 4.56.
Syn th esis of [{P t2(µ3-S)2(cod )(C8H11)}P t(d p p p )]Cl (4).
To a solution of 3 (0.10 g, 0.07 mmol) in benzene (25 mL) was
added a methanol solution of NaMeO (17 µL of a 4.74 M
solution). The resulting solution was left to stir for 4 h, after
which time it was concentrated to ca. 5 mL and filtered.
Addition of diethyl ether to the filtrate caused precipitation
of complex 4 as a pale yellow solid. Yield: 28%. Anal. Calcd
for C43H49P2Pt3S2Cl: C, 39.35; H, 3.76; S, 4.88. Found: C,
37.67; H, 4.38; S, 5.19.
Rint
0.078
0.046
0.119
0.966
0.041
0.035
0.099
1.013
R (F, F2 > 2σ)
Rw (F2, all data)
S (F2, all data)
max., min. el dens
1.61, -1.57
1.58, -1.71
(e Å-3
)
amount of DCl in D2O (13 µL of a 9.3 M DCl solution) instead
of HCl. The solid thus obtained was analyzed by 31P{1H} NMR,
confirming complex 1 as the only product in both cases. In
addition, the 2H NMR spectrum of the compound obtained
after addition of DCl shows a resonance at 2.47 ppm.
Rea ction of [{P t2(µ3-S)2(cod )(C8H11)}P t(d p p p )]Cl (4) or
[{P t2(µ3-S)2(C8H11)2}P t(d p p p )] (5) w ith HCl. A solution
containing 20 mg of 4 (0.015 mmol) or 5 (0.016 mmol) in CH3-
CN (15 mL) was acidified by addition of 20 or 40 µL,
respectively, of 1.0 M HCl in water. After 15 min of stirring
at room temperature the solvent was evaporated to dryness.
Characterization of the solid residue by 31P{1H} NMR indi-
cated that complex 3 was the only end product in both acid-
base reactions.
X-r a y Cr ysta llogr a p h y. Data for complexes 1 and 2 were
measured on Bruker SMART 1K CCD diffractometers.20 For
1, Mo KR radiation (λ ) 0.71073 Å) was used, while the small
size and poorly diffracting nature of the crystals of 2 neces-
sitated the use of synchrotron radiation (λ ) 0.6900 Å) at
CCLRC Daresbury Laboratory SRS station 9.8.21 Crystal-
lographic data are given in Table 7. Both structures were found
to contain solvent, some of which could be modeled with
ordered atomic sites and some of which is disordered; for 1,
three ordered chloroform molecules were located, and other
highly disordered solvent was handled by the SQUEEZE
procedure in the program PLATON,22 while for 2, one ordered
dichloromethane molecule was located (on a mirror plane), and
other disordered solvent was fitted with a number of partially
occupied atom sites giving no recognizable molecular geometry.
For each structure, only the ordered solvent is included in the
formulas given in Table 7. The cation of 2 lies on a crystal-
lographic mirror plane, across which the deprotonated COD
ligand lies. The atoms of the two disorder components and
those of some of the disordered solvents overlap each other,
rendering the geometry of these moieties unreliable. Restraints
were applied to the anisotropic displacement parameters of
the ligand, but no geometrical constraints or restraints were
used for these atoms. The perchlorate anion is also disordered
across a mirror plane. H atoms were included in ideal positions
for ordered parts of both structures. Refinement was on all
unique F2 values in each case.23 The largest features in final
Syn th esis of [{P t2(µ3-S)2(C8H11)2}P t(d p p p )] (5). The
same procedure as that followed for complex 4 yielded complex
5 from 3 (0.10 g, 0.074 mmol) and a methanol solution of
NaMeO (43 µL of a 4.74 M solution). Yield: 31%. Anal. Calcd
for C43H48P2Pt3S2: C, 40.47; H, 3.79; S, 5.02. Found: C, 40.00;
H, 4.15; S, 4.63.
Evolu tion w ith Tim e of a Solu tion of [{P t2(µ3-S)2-
(d p p p )2}P t(cod )]Cl2 (1) or [{P t2(µ3-S)2(d p p p )2}P t(C8H11)]-
Cl (2) in th e P r esen ce of d p p p . To a solution of complex 1
or 2 in either CH2Cl2, CH3CN, or MeOH was added a 5:1 molar
excess of dppp at room temperature. This solution was
monitored for 48 h by taking one aliquot every 12 h, evaporat-
ing it to dryness, and recording the 31P{1H} NMR spectrum
in d6-dmso. The above experiments were also performed in the
same solvents but under reflux conditions. NMR data revealed
that the possible ligand exchange reactions did never occur.
Rea ction betw een [{P t2(µ3-S)2(cod )2}P t(d p p p )]Cl2 (3)
a n d d p p p . To a solution of 3 (25 mg, 0.02 mmol) in CH2Cl2
(15 mL) was added dppp (8 mg, 0.02 mmol). After 2 h of
stirring at room temperature the solvent was removed in
vacuo, leaving the product as a white solid. The 31P{1H} NMR
spectrum of this solid in d6-dmso solvent showed that complex
1 was the only compound present in solution.
Rea ction betw een [{P t2(µ3-S)2(cod )(C8H11)}P t(d p p p )]-
Cl (4) w ith d p p p . To a solution of 0.02 g of 4 (0.015 mmol) in
CH2Cl2 (15 mL) was added 8 mg (0.02 mmol) of dppp. After 2
h of stirring at room temperature the resulting solution was
analyzed by 31P{1H} NMR as indicated above. The NMR data
showed that complex 2 was the only compound present in
solution.
Rea ction betw een [{P t2(µ3-S)2(d p p p )2}P t(C8H11)]Cl (2)
a n d HCl or DCl. To a solution of 2 (0.20 g, 0.12 mmol) in
benzene (25 mL) was added an equimolar amount of HCl in
water (30 µL of a 4 M HCl solution). After 15 min of stirring
at room temperature the solid formed was filtered off and
analyzed by 31P{1H} NMR. These data confirmed that the only
product obtained was complex 1. Yield: 73%. The same
experiment was carried out but with addition of an equimolar
(20) SMART and SAINT software; Bruker AXS Inc.: Madison, WI,
2001.
(21) Cernik, R. J .; Clegg, W.; Catlow, C. R. A.; Bushnell-Wye, G.;
Flaherty, J . V.; Greaves, G. N.; Burrows, I.; Taylor, D. J .; Teat, S. J .;
Hamichi, M. J . Synchrotron Rad. 1997, 4, 279.
(22) Spek, A. L. PLATON; University of Utrecht: The Netherlands,
2001.
(23) Sheldrick, G. M. SHELXTL, version 6; Bruker AXS Inc.:
Madison, WI, 2001.

2532 Organometallics, Vol. 23, No. 11, 2004
Mas-Balleste´ et al.
difference syntheses lie close to heavy atoms and disordered
parts of the structures.
effects were taken into account by means of the COSMO
solvation model.29 Energies were calculated with methanol
(ꢀ ) 32.63) as solvent, keeping the optimized geometry for the
isolated species (single-point calculations).
Com p u ta tion a l Deta ils. Calculations were performed
using the GAUSSIAN98 series of programs.24 Geometry
optimizations were done using the density functional theory
(DFT) with the hybrid B3LYP functional.25 Effective core
potentials (ECP) and their associated double-ú LANL2DZ basis
set were used for platinum, phosphorus, sulfur, iodine, bro-
mine, and chlorine atoms,26 adding an extra series of d-
polarization functions in the case of P, S, I, Br, and Cl.27 The
6-31G basis set was used for the C and H atoms.28 Solvent
Ack n ow led gm en t. Financial support from the Min-
isterio de Ciencia y Tecnolog´ıa of Spain (projects
BQU2001-1976 and BQU2002-04110-CO2-02), the
EPSRC (UK), and Ineos Acrylics is gratefully acknowl-
edged. R.M.B. is indebted to the Universitat Auto`noma
de Barcelona for a predoctoral scholarship.
(24) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J . L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J . A. Gaussian 98 (Revision A.7); Gaussian, Inc.: Pittsburgh,
PA, 1998.
Su p p or tin g In for m a tion Ava ila ble: For 1 and 2, details
of crystal structure determination, atomic coordinates, bond
lengths and angles, and displacement parameters. This mate-
rial is available free of charge via the Internet at
http://pubs.acs.org. Observed and calculated structure factor
lists are available from the authors upon request.
OM034312X
(27) Ho¨llwarth, A.; Bo¨hme, M.; Dapprich, S.; Ehlers, A. W.; Gobbi,
A.; J onas, V.; Ko¨hler, K. F.; Stegman, R.; Veldkamp, A.; Frenking, G.
Chem. Phys. Lett. 1993, 208, 237.
(28) Hehre, W. J .; Ditchfield, R.; Pople, J . A. J . Chem. Phys. 1972,
56, 2257.
(25) (a) Becke, A. D. J . Chem. Phys. 1993, 98, 5648. (b) Lee, C.; Yang,
W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(26) (a) Hay, P. J .; Wadt, W. R. J . Chem. Phys. 1985, 82, 299. (b)
Hay, P. J .; Wadt, W. R. J . Chem. Phys. 1985, 82, 270.
(29) Barone, V.; Cossi, M. J . Phys. Chem, 1998, 102, 1995.