Synthesis, structure, and protonation studies of Cp*MH3(dppe) (M = Mo, W). Pseudo-trigonal-prismatic vs pseudo-octahedral structures for half-sandwich group 6 M(IV) derivatives

Article abstract of DOI:10.1021/om961002o

The compounds Cp*MH₃(dppe) (M = Mo, 1; W, 2) are accessible in good yields from reacting the corresponding compound Cp*MCl₄ and LiAlH₄ in toluene/Et₂O followed by methanolysis. The X-ray structure of 1 shows a pseudo-trigonal-prismatic geometry which is unprecedented for half-sandwich CpML₅-type compounds. Protonation with HBF₄·Et₂O in Et₂O at low temperature affords [Cp*MH₄(dppe)]⁺BF₄^- salts (M = Mo, 3; W, 4). While 3 spontaneously decomposes, even at low temperatures, in coordinating solvents and CH₂Cl₂, 4 is stable at room temperature in MeCN. An X-ray structure of 4 is consistent with a classical tetrahydrido species with a distorted pseudo-pentagonal-bipyramidal structure. The low-temperature NMR properties, J_HD ≤ 1 Hz for 4-d₃, and the T_1(min) value for the hydride resonance are also consistent with this structural assignment. Decomposition of 3 in MeCN at room temperature selectively affords [Cp*M0H₂(MeCN)(dppe)]⁺BF₄^-, 5. The NMR properties of this complex indicate a fluxional structure with inequivalent H and P nuclei and are consistent with a pseudo-trigonal-prismatic structure analogous to that of the precursor 1. Further protonation of 5 in MeCN or direct protonation of 1 with excess acid in MeCN affords two isomers of complex [Cp*MoH(MeCN)₂(dppe)](BF₄)₂, 6 and 7. Thermal treatment in MeCN slowly converts 7 into 6 initially, but both isomers further transform into a third isomer, 8, upon prolonged heating. The structure of 6 has been elucidated by X-ray crystallography and consists of a highly distorted pseudo-octahedral geometry with relative trans MeCN ligands. The structures of 7 and 8 and mechanisms of the interconversions between the various isomeric structures are proposed on the basis of the solution NMR studies.

Full text of DOI:10.1021/om961002o

Organometallics 1997, 16, 1581-1594
1581
Syn th esis, Str u ctu r e, a n d P r oton a tion Stu d ies of
Cp *MH3(d p p e) (M ) Mo, W). P seu d o-Tr igon a l-P r ism a tic
vs P seu d o-Octa h ed r a l Str u ctu r es for Ha lf-Sa n d w ich
Gr ou p 6 M(IV) Der iva tives
Brett Pleune,1a Rinaldo Poli,*,1b and J ames C. Fettinger1a
Department of Chemistry and Biochemistry, University of Maryland,
College Park, Maryland 20742, and Laboratoire de Synthe`se et d’Electrosynthe`se
Organome´tallique, Faculte´ des Sciences “Gabriel”, Universite´ de Bourgogne,
6 Boulevard Gabriel, 21100 Dijon, France
Received November 27, 1996X
The compounds Cp*MH3(dppe) (M ) Mo, 1; W, 2) are accessible in good yields from reacting
the corresponding compound Cp*MCl4 and LiAlH4 in toluene/Et2O followed by methanolysis.
The X-ray structure of 1 shows a pseudo-trigonal-prismatic geometry which is unprecedented
for half-sandwich CpML5-type compounds. Protonation with HBF4‚Et2O in Et2O at low
temperature affords [Cp*MH4(dppe)]+BF4- salts (M ) Mo, 3; W, 4). While 3 spontaneously
decomposes, even at low temperatures, in coordinating solvents and CH2Cl2, 4 is stable at
room temperature in MeCN. An X-ray structure of 4 is consistent with a classical
tetrahydrido species with a distorted pseudo-pentagonal-bipyramidal structure. The low-
temperature NMR properties, J HD e 1 Hz for 4-d3, and the T1(min) value for the hydride
resonance are also consistent with this structural assignment. Decomposition of 3 in MeCN
at room temperature selectively affords [Cp*MoH2(MeCN)(dppe)]+BF4-, 5. The NMR
properties of this complex indicate a fluxional structure with inequivalent H and P nuclei
and are consistent with a pseudo-trigonal-prismatic structure analogous to that of the
precursor 1. Further protonation of 5 in MeCN or direct protonation of 1 with excess acid
in MeCN affords two isomers of complex [Cp*MoH(MeCN)2(dppe)](BF4)2, 6 and 7. Thermal
treatment in MeCN slowly converts 7 into 6 initially, but both isomers further transform
into a third isomer, 8, upon prolonged heating. The structure of 6 has been elucidated by
X-ray crystallography and consists of a highly distorted pseudo-octahedral geometry with
relative trans MeCN ligands. The structures of 7 and 8 and mechanisms of the intercon-
versions between the various isomeric structures are proposed on the basis of the solution
NMR studies.
In tr od u ction
work has been carried out on monohydride systems.17-40
The transient 17-electron species, [M-H]+•, usually
decomposes by deprotonation,24 the outcome being
Transition metal polyhydride systems have been the
focus of much recent attention. The dichotomy between
classical and nonclassical hydrides2-4 and possible
mechanisms of hydride fluxionality5 have been studied
extensively. A few mononuclear transient 17-electron
polyhydride species have been produced by either
chemical or electrochemical oxidation of the saturated
polyhydride precursors,6-16 whereas a greater body of
(11) Costello, M. T.; Walton, R. A. Inorg. Chem. 1988, 27, 2563-
2564.
(12) Herrmann, W. A.; Theiler, H. G.; Herdtweck, E.; Kiprof, P. J .
Organometal. Chem. 1989, 367, 291-311.
(13) Bianchini, C.; Laschi, F.; Peruzzini, M.; Ottaviani, F.; Vacca,
A.; Zanello, P. Inorg. Chem. 1990, 29, 3394-3402.
(14) Roullier, L.; Lucas, D.; Mugnier, Y.; Antin˜olo, A.; Fajardo, M.;
Otero, A. J . Organometal. Chem. 1991, 412, 353-362.
(15) Westerberg, D. E.; Rhodes, L. F.; Edwin, J .; Geiger, W. E.;
Caulton, K. G. Inorg. Chem. 1991, 30, 1107-1112.
(16) Zlota, A. A.; Tilset, M.; Caulton, K. G. Inorg. Chem. 1993, 32,
3816-3821.
(17) Sanders, J . R. J . Chem. Soc., Dalton Trans. 1973, 748-749.
(18) Sanders, J . R. J . Chem. Soc., Dalton Trans. 1975, 2340-2342.
(19) Gargano, M.; Giannoccaro, P.; Rossi, M.; Vasapollo, G.; Sacco,
A. J . Chem. Soc., Dalton Trans. 1975, 9-12.
(20) Pilloni, G.; Schiavon, G.; Zotti, G.; Zecchin, S. J . Organometal.
Chem. 1977, 134, 305-318.
(21) Treichel, P. M.; Molzahn, D. C.; Wagner, K. P. J . Organometal.
Chem. 1979, 174, 191-197.
X Abstract published in Advance ACS Abstracts, March 15, 1997.
(1) (a) University of Maryland. (b) Universite´ de Bourgogne.
(2) Hamilton, D. G.; Crabtree, R. H. J . Am. Chem. Soc. 1988, 110,
4126-4133.
(3) Luo, X.-L.; Crabtree, R. H. Inorg. Chem. 1990, 29, 2788-2791.
(4) Van Der Sluys, L. S.; Eckert, J .; Eisenstein, O.; Hall, J . H.;
Huffman, J . C.; J ackson, S. A.; Koetzle, T. F.; Kubas, G. J .; Vergamini,
P. J .; Caulton, K. G. J . Am. Chem. Soc. 1990, 112, 4831-4841.
(5) Gusev, D. G.; Berke, H. Chem. Ber. 1996, 129, 1143-1155 and
references therein.
(6) Klinger, R. J .; Huffman, J . C.; Kochi, J . K. J . Am. Chem. Soc.
1980, 102, 208-216.
(7) Allison, J . D.; Cameron, C. J .; Wild, R. E.; Walton, R. A. J .
Organomet. Chem. 1981, 218, C62-C66.
(8) Rhodes, L. F.; Zubkowski, J . D.; Folting, K.; Huffman, J . C.;
Caulton, K. G. Inorg. Chem. 1982, 21, 4185-4192.
(9) Bruno, J . W.; Caulton, K. G. J . Organometal. Chem. 1986, 315,
C13-C16.
(10) Detty, M. R.; J ones, W. D. J . Am. Chem. Soc. 1987, 109, 5666-
5673.
(22) Chen, L.; Davies, J . A. Inorg. Chim. Acta 1990, 175, 41-45.
(23) Roullier, L.; Lucas, D.; Mugnier, Y.; Antin˜olo, A.; Fajardo, M.;
Otero, A. J . Organometal. Chem. 1990, 396, C12-C16.
(24) Ryan, O. B.; Tilset, M.; Parker, V. D. J . Am. Chem. Soc. 1990,
112, 2618-2626.
(25) Ryan, O. B.; Smith, K.-T.; Tilset, M. J . Organometal. Chem.
1991, 421, 315-322.
(26) Ryan, O. B.; Tilset, M.; Parker, V. D. Organometallics 1991,
10, 298-304.
(27) Ryan, O. B.; Tilset, M. J . Am. Chem. Soc. 1991, 113, 9554-
9561.
S0276-7333(96)01002-3 CCC: $14.00 © 1997 American Chemical Society

1582 Organometallics, Vol. 16, No. 8, 1997
Pleune et al.
determined by the basic properties of the solvent vs
M-H as well as by the reactivity of the depronotated
product M•. In particular, when the starting compound
M-H captures the proton, the resulting [MH2]+ species
may either remain as a stable (classical or nonclassical)
hydride product or lead to the replacement of one or
more H2 molecules with as many molecules of a mono-
dentate ligand (usually a coordinative solvent). A study
of the protonation of transition metal hydrides, there-
fore, gives useful information on the possible nature of
the ultimate oxidation products. In a recent study of
the oxidation and protonation of the monohydride
system CpMoH(CO)2(PR3) (R ) Me, Ph) in acetonitrile,41
we have shown that proton transfer from the 17-electron
[CpMoH(CO)2(PR3)]+• transient to the starting unoxi-
dized hydride or the direct protonation of the latter by
HBF4 is followed by an irreversible H2 loss, which drives
either reaction to the selective formation of the solvato
complex [CpMo(MeCN)(CO)2(PR3)]+. Polyhydrides show
the same general reactivity patterns as monohydrides;
for instance, treatment of Cp*Ru(PPh3)H3 with
Cp2Fe+PF6- in MeCN leads to H2 evolution and forma-
tion of [Cp*Ru(PPh3)(MeCN)2]+.15
With the ultimate goal in mind of generating stable
17-electron polyhydride species for further activation
and reactivity studies, we have synthesized new com-
pounds of the general class (η5-ring)MH3L2 (ring )
cyclopentadienyl ligand, M ) group 6 metal, L )
tertiary phosphine), previously reported examples of
which include CpMoH3(dppe),42 CpWH3(PMe3)2,43 (η5-
C5H4Pri)MoH3L2 (L ) PMe3, PMe2Ph or L2 ) dppe),44
Cp*MoH3(PMe3)2,45 and CpMoH3(PMe2Ph)2.46 Oxida-
tion studies of these derivatives had not previously been
reported, while protonation studies were limited to (η5-
C5H4Pri)MoH3(PMe3)2 with HCl and water to generate
[(η5-C5H4Pri)MoO(PMe3)2]+.44 In this contribution, we
will examine the protonation of the new polyhydride
species Cp*MH3(dppe) (M ) Mo, 1; W, 2). Oxidation
studies of these compounds will be documented in a
later contribution. Part of these studies have been
previously communicated.47
Exp er im en ta l Section
All manipulations were carried out under an inert atmo-
sphere of dinitrogen or argon by the use of Schlenk-line or
glovebox techniques. Methanol was degassed by three freeze-
pump-thaw cycles prior to use. Other solvents were dried
by conventional methods (Et2O from Na/K/benzophenone,
toluene and heptane from Na, MeCN from CaH2, and CH2Cl2
from P4O10) and distilled under dinitrogen prior to use.
Deuterated solvents were dried over molecular sieves and
degassed by three freeze-pump-thaw cycles prior to use. 1H-
and 31P{1H}-NMR measurements were carried out on Bruker
AF200, WP200, or AM400 spectrometers; the peak positions
are reported with positive shifts downfield of TMS, as calcu-
lated from the residual solvent peaks (1H), or downfield of
external 85% H3PO4 (31P). For each 31P-NMR spectrum, a
sealed capillary containing 85% H3PO4 was immersed in the
same NMR solvent used for the measurement and this was
used as the reference. 31P{selective-1H}-NMR measurements
were carried out on Bruker WP200 or AM400 spectrometers.
1H{31P}-NMR measurements were carried out on a Bruker
AMX500 spectrometer. Values of the longitudinal relaxation
times T1 were obtained from the slopes of linear plots of ln
[2Ieq/(Ieq - Iτ)] vs τ, where Iτ is the peak intensity as measured
by the standard inversion-recovery-pulse (180-τ-90) se-
quence and Ieq is the peak intensity at τ ) ∞. EPR measure-
ments were carried out at the X band microwave frequency
on a Bruker ER 200 D spectrometer upgraded to ESP 300,
equipped with a cylindrical ER/4103 TM 110 cavity. The
elemental analyses were carried out by Desert Analytics,
Tucson, AZ, and by Atlantic Microlab, Norcross, GA. LiAlH4,
dppe, grade I acidic alumina, and HBF4‚Et2O (Aldrich) were
used without further purification. CDFCl2,48 Cp*MoCl4,49 and
50
Cp*WCl4 were prepared according to literature procedures.
Syn th esis of Cp *MoH3(d p p e) (1). Cp*MoCl4 (1.007 g,
2.70 mmol) was slurried in a 70 mL:15 mL toluene/diethyl
ether solvent mixture at room temperature. The ligand dppe
(1.074 g, 2.70 mmol) and LiAlH4 (1.00 g, 26.3 mmol) were
added to the stirring solution at room temperature. The
mixture began to slowly evolve H2 gas as the solution devel-
oped an orange color within 30 min. The mixture was stirred
for an additional 12 h at room temperature. Dropwise addition
of MeOH (ca. 10 mL) at room temperature caused vigorous
evolution of H2 gas; the orange solution darkened. The
reaction mixture was stirred an additional 30 min, followed
by evaporation of the mixture to dryness. The residue was
extracted with heptane (150 mL) and filtered through Celite.
The resulting yellow-orange solution was evaporated to ca. 2
mL, precipitating a yellow powder, which was washed with
with cold (-80 °C) heptane and dried under vacuum. Yield:
1.107 g (65%). Single crystals were obtained from a saturated
warm heptane solution upon cooling to room temperature.
Anal. Calcd for C36H42MoP2: C, 68.4; H, 6.7. Found: C, 67.6;
H, 7.0. 1H-NMR (C6D6, δ): 7.8-7.0 (m, 20H, dppe-Ph), 1.85
(m, 4H, dppe-CH2), 1.83 (s, 15H, Cp*), -5.27 (t, J PH ) 42.0
Hz, 3H, MoH). 31P-NMR (C6D6, δ): 91.8 (s). 31P{selective-
1H}-NMR (C6D6, δ): 91.8 (quartet, J PH ) 41.0 Hz).
(28) Roger, C.; Hamon, P.; Toupet, L.; Rabaaˆ, H.; Saillard, J .-Y.;
Hamon, J .-R.; Lapinte, C. Organometallics 1991, 10, 1045-1054.
(29) Hamon, P.; Toupet, L.; Hamon, J .-R.; Lapinte, C. Organo-
metallics 1992, 11, 1429-1431.
(30) Tilset, M. J . Am. Chem. Soc. 1992, 114, 2740-2741.
(31) J ia, G.; Lough, A. J .; Morris, R. H. Organometallics 1992, 11,
161-171.
(32) Smith, K.-T.; Tilset, M. J . Organometal. Chem. 1992, 431, 55-
64.
(33) Amatore, C.; Frau´sto da Silva, J . J . R.; Guedes da Silva, M. F.
C.; Pombeiro, A. J . L.; Verpeaux, J .-N. J . Chem. Soc., Chem. Commun.
1992, 1289-1291.
(34) Skagestad, V.; Tilset, M. J . Am. Chem. Soc. 1993, 115, 5077-
5083.
(35) Smith, K.-T.; Rømming, C.; Tilset, M. J . Am. Chem. Soc. 1993,
115, 8681-8689.
(36) J ime´nez-Tenorio, M.; Puerta, M. C.; Valerga, P. Organometallics
1994, 13, 3330-3337.
(37) Kaim, W.; Reinhardt, R.; Sieger, M. Inorg. Chem. 1994, 33,
4453-4459.
(38) Pedersen, A.; Tilset, M. Organometallics 1994, 13, 4887-4894.
(39) Blaine, C. A.; Ellis, J . E.; Mann, K. R. Inorg. Chem. 1995, 34,
1552-1561.
Syn th esis of Cp *WH3(d p p e) (2). Cp*WCl4 (1.46 g, 3.17
mmol) was slurried in a 100 mL:15 mL toluene/diethyl ether
solvent mixture at room temperature. The ligand dppe (1.26
g, 3.17 mmol) and LiAlH4 (1.20 g, 31.6 mmol) were added to
(40) Hembre, R. T.; McQueen, J . S.; Day, V. W. J . Am. Chem. Soc.
1996, 118, 798-803.
(41) Quadrelli, E. A.; Kraatz, H.-B.; Poli, R. Inorg. Chem. 1996, 35,
5154-5162.
(42) Aviles, T.; Green, M. L. H.; Dias, A. R.; Romao, C. J . Chem.
Soc., Dalton Trans. 1979, 1367-1371.
(43) Green, M. L. H.; Parkin, G. J . Chem. Soc., Dalton Trans. 1987,
1611-1618.
(47) Fettinger, J . C.; Pleune, B.; Poli, R. J . Am. Chem. Soc. 1996,
118, 4906-4907.
(44) Grebenik, P. D.; Green, M. L. H.; Izquierdo, A.; Mtetwa, V. S.
B.; Prout, K. J . Chem. Soc., Dalton Trans. 1987, 9-19.
(45) Abugideiri, F.; Kelland, M. A.; Poli, R.; Rheingold, A. L.
Organometallics 1992, 11, 1303-1311.
(46) Abugideiri, F.; Fettinger, J . C.; Pleune, B.; Poli, R.; Bayse, C.
A.; Hall, M. B. Organometallics 1997, 16, 1179-1185.
(48) Siegel, S. S.; Anet, F. A. J . Org. Chem. 1988, 53, 2629-2630.
(49) Keogh, D. W.; Poli, R. In New Synthetic Methods of Organo-
metallic and Inorganic Chemistry; Herrmann, W. A., Ed.; Georg
Thieme Verlag: Stuttgart, Germany.
(50) Murray, R. C.; Blum, L.; Liu, A. H.; Schrock, R. R. Organo-
metallics 1985, 4, 953-954.

Studies of Cp*MH3(dppe) (M ) Mo, W)
Organometallics, Vol. 16, No. 8, 1997 1583
the stirring solution at room temperature. Slow evolution of
H2 gas was witnessed, as the solution developed an orange
color within 30 min. The reaction mixture was stirred an
additional 12 h, and the solvent was evaporated to dryness
under vacuum. The residue was extracted into heptane (200
mL), and the solution was filtered through Celite. The orange
solution was evaporated to ca. 2 mL, precipitating a crystalline
orange powder. The powder was washed with with cold (-80
°C) heptane and dried under vacuum. Yield: 712 mg (32%).
Heptane (50 mL) was added to the remaining residue, and 10
mL of methanol was added dropwise at room temperature. H2
gas vigorously evolved, and the solution developed a red-orange
color. The reaction mixture stirred for an additional 30 min
and was filtered through Celite. The solution was evaporated
to ca. 5 mL, precipitating a second crop of orange powder (498
mg). Combined yield: 1.21 g (53%). The second crop is
slightly contaminated by the pentahydride compound
Cp*WH5(η1-dppe) (see Results). Spectroscopically pure prod-
uct is obtained from this crop by recrystallization from hot
1H}-NMR (CD3CN, δ): 78.1 (t, J PH ) 52 Hz). 31P{1H}-NMR
(CD2Cl2, -80 °C, δ): 82.8 (d, J PP ) 20.8 Hz), 75.3 (d, J PP
20.7 Hz).
)
P r oton a tion of Cp *MoH3(d p p e) in Aceton itr ile. HBF4‚
Et2O (7 µL, 0.040 mmol) was added to a suspension of
Cp*MoH3(dppe) (25 mg, 0.039 mmol) in 0.5 mL CD3CN via
microsyringe at room temperature. The insoluble yellow
starting compound began to dissolve as the solution visibly
evolved H2, turning the solution from yellow-orange to red over
a period of 60 min. The solution was monitored via 1H- and
31P-NMR spectroscopies showing a mixture of 5 and two
isomers of compound [Cp*MoH(dppe)(MeCN)2](BF4)2 (6 and
7). According to the Cp* resonances in the 1H-NMR spectrum,
compounds 5, 6, and 7 were approximately in a 6:1:3 ratio.
Syn th esis of tr a n s-[Cp *MoH(d p p e)(MeCN)2](BF 4)2 (6).
HBF4‚Et2O (15 µL, 0.085 mmol) was added to a suspension of
Cp*MoH3(dppe) (54 mg, 0.085 mmol) in 1 mL of CD3CN via
microsyringe at room temperature. After 60 min, the yellow
starting material had dissolved completely. Slow diffusion of
Et2O into this solution at -20 °C produced red crystals after
24 h. Yield: 15 mg (21%). A suitable crystal obtained in this
manner was used for X-ray analysis. 1H-NMR (CD3CN, δ):
7.9-7.1 (m, dppe-Ph, 20H), 3.3-2.4 (m, 4H, dppe-CH2), 2.05
(s, 6H, CH3CN), 1.82 (s, 15H, Cp*), -4.05 (dd, 1H, MoH, J HP′
) 10 Hz, J HP′′ ) 84 Hz). Upon broad-band 31P decoupling, the
1H-NMR resonance at δ -4.05 collapsed into a singlet. 31P-
NMR (CD3CN, δ): 52.6 (d, dppe, J PP ) 26.0 Hz), 76.2 (d, dppe,
J PP ) 26.0 Hz).
heptane. Anal. Calcd for
C36H42WP2: C, 60.0; H, 5.9.
Found: C, 60.6; H, 5.9. 1H-NMR (C6D6, δ): 7.8-7.0 (m, 20H,
dppe-Ph), 2.1-1.9 (m, 4H, dppe-CH2), 1.93 (s, 15H, Cp*),
-5.90 (t, 3H, WH, J PH ) 44.6 Hz, J WH ) 54.4 Hz). 31P-NMR
(C6D6, δ): 67.9 (s, J WP ) 134.4 Hz). 31P{selective-1H}-NMR
(C6D6, δ): 67.9 (quartet, J PH ) 41.5 Hz).
Syn th esis of [Cp *MoH4(d p p e)]BF 4 (3). Cp*MoH3(dppe)
(836 mg, 1.30 mmol) was dissolved in 40 mL of diethyl ether.
HBF4‚Et2O (185 µL, 1.48 mmol) was added to the solution via
microsyringe at room temperature. An off-white precipitate
immediately developed. The supernatant liquid was removed
by filter-cannula, and the product was washed with diethyl
ether (3 × 10 mL) and dried under vacuum. Yield: 845 mg
(90%). Anal. Calcd for C36H43BF4MoP2: C, 60.02; H, 6.02.
Found: C, 58.18: H, 5.09. 1H-NMR (CDFCl2, -60 °C, δ): 7.9-
7.0 (m, 20H, dppe-Ph), 2.25-2.20 (m, 4H, dppe-CH2), 1.93
(s, 15H, Cp*), -3.56 (t, 4H, M-H, J PH ) 36.0 Hz). 31P-NMR
(CDFCl2, -60 °C, δ): 72.9 (s). 31P{selective-1H}-NMR (CD-
FCl2, -60 °C, δ): 72.9 (quintet, J PH ) 34.0 Hz). A better
elemental analysis could not be obtained because of the
decomposition of the compound upon attempted recrystalliza-
tion (see Results).
Syn th esis of [Cp *MoH(d p p e)(MeCN)2](BF 4)2 (7). HBF4‚
Et2O (41 µL, 0.237 mmol) was added to a suspension of
Cp*MoH3(dppe) (101 mg, 0.160 mmol) in 2 mL of CH3CN at
room temperature, and the mixture was allowed to stir for 60
min. The solution was transferred onto a degassed chroma-
tography column (12.6 cm × 2.3 cm) made up of acidic alumina
in CH3CN. Elution with CH3CN yielded a yellow band, which
1
was collected and shown by H-NMR to contain compound 5.
The eluant was then switched to a 10:1 CH3CN/H2O solvent
mixture, yielding a red band, which was collected. The solvent
was removed under reduced pressure, and the residue was
redissolved in 5 mL of fresh CH3CN. Concentration to ca. 0.5
mL by evaporation under reduced pressure, and addition of
10 mL of Et2O yielded the product as a red-orange precipitate.
Yield: 31 mg (22%). Anal. Calcd for C40H46MoP2N2B2F8: C,
54.17; H, 5.19. Found: C, 54.29; H, 5.19. 1H-NMR (CD3CN,
δ): 8.0-7.0 (m, 20H, dppe-Ph), 3.4-2.9 (m, 4H, dppe-CH2),
2.05 (s, 3H, CH3CN), 1.95 (s, 3H, CH3CN), 1.63 (s, 15H, Cp*),
-2.18 (dd, 1H, M-H, J HP' ) 22 Hz, J HP" ) 81 Hz). Upon broad-
band 31P decoupling, the 1H-NMR resonance at δ -2.18
Syn th esis of [Cp *WH4(d p p e)]BF 4 (4). Cp*WH3(dppe)
(102 mg, 0.142 mmol) was dissolved in 10 mL of diethyl ether.
HBF4‚Et2O (20 µL, 0.16 mmol) was added to the solution via
microsyringe at room temperature. An off-white precipitate
immediately developed. The ether solvent was removed by
filter-cannula, and the product was washed with diethyl ether
(3 × 5 mL) and dried under vacuum. Yield: 99 mg (86%).
Single crystals were grown from slow diffusion of a layer of
diethyl ether into a saturated solution in CH3CN at room
temperature. Anal. Calcd for C36H43BF4P2W: C, 53.49; H,
5.36. Found: C, 53.03; H, 5.40. 1H-NMR (CD2Cl2, δ): 7.8-
7.3 (m, 20H, dppe-Ph), 2.4-2.0 (m, 4H, dppe-CH2), 2.05 (s,
15H, Cp*), -2.73 (t, 4H, M-H, J PH ) 32.6 Hz, J WH ) 37.7
Hz). 31P-NMR (CD2Cl2, room t, δ): 51.6 (s, J WP ) 126.6 Hz).
31P{selec-1H}-NMR (CD2Cl2, δ): 51.6 (quintet, J PH ) 28.4 Hz).
collapsed into a singlet. 31P-NMR (CD3CN, δ): 75.5 (d, J PP
)
32.6 Hz), 67.1 (d, J PP ) 32.6 Hz). 31P{selective-1H}-NMR
(CD3CN, δ): 75.5 (dd, J PP ) 32.6 Hz, J P'H ) 25.0 Hz), 67.1
(dd, J PP ) 32.6 Hz, J P"H ) 77.5 Hz).
Th er m a l Tr ea tm en t of Com p ou n d s 6 a n d 7. F or m a -
tion of Com p ou n d 8. (a ) F r om a Mixtu r e of 6 a n d 7. A
solution of compounds 6 and 7 was prepared in situ by
protonation of Cp*MoH3(dppe) (59 mg, 0.094 mmol) in CD3CN
(0.5 mL), as described above. Heating this mixture to 70 °C
for 105 min with 1H- and 31P-NMR monitoring showed an
initial decrease of 7 and an increase of 6, followed by a decrease
and ultimate disappearance of 6 and formation of 8 as the only
product (see Results). Compound 8: 1H-NMR (CD3CN, δ)SP-
CLN 7.9-7.1 (m, dppe-Ph, 20H), 3.3-2.4 (m, 4H, dppe-CH2),
31P-NMR (CDFCl2, -95 °C, δ): 48.6 (d, J PP ) 26.3 Hz, J PW
139.5 Hz), 54.7 (d, J PP ) 26.3 Hz, J PW ) 139.5 Hz).
)
Syn th esis of [Cp *MoH2(d p p e)(MeCN)]BF 4 (5). Com-
pound 3 (110 mg, 0.153 mmol) was dissolved in 2 mL of CH3CN
at room temperature. The solution was stirred for 30 min; 10
mL of Et2O was added to precipitate the product. The solution
was filtered, and the yellow-brown residue was washed with
Et2O (3 × 5 mL). The product was dried under vacuum.
Yield: 95 mg (82%). Anal. Calcd for C38H44MoP2NBF4‚H2O:
C, 58.70; H, 5.96. Found: C, 58.2; H, 5.8. 1H-NMR (CD3CN,
δ): 7.8-7.0 (m, 20H, dppe-Ph), 3.0-2.5 (m, 4H, dppe-CH2),
1.99 (s, 3H, CH3CN), 1.69 (s, 15H, Cp*), -5.50 (dd, 2H, M-H,
J P′H ) 51 Hz, J P′′H ) 52 Hz). A broad resonance at δ 2.2 (br),
assigned to H2O, is also observed. Upon broad-band 31P
1.85 (s, 15H, Cp*), -4.08 (dd, 1H, MoH, J HP′ ) 10 Hz, J HP′′
)
84 Hz). 31P-NMR (CD3CN, δ): 50.2 (d, J PP ) 25.5 Hz), 76.3
(d, J PP ) 25.5 Hz). The MeCN resonance is masked by the
multiplet of the solvent. A saturated solution in CD2Cl2 shows
a single MeCN resonance at δ 2.00 (6H).
(b) F r om P u r e 6. A solution of 6 (25 mg) in CD3CN (0.028
mL) was heated to 70 °C for 30 min. The hydride region of
1
1
decoupling, the H-NMR resonance at δ -5.50 collapsed into
the H-NMR spectrum showed the disappearance of all of the
a singlet. 31P{1H}-NMR (CD3CN, δ): 78.1 (s). 31P{selective-
resonances due to compound 6 and the growth of the reso-

1584 Organometallics, Vol. 16, No. 8, 1997
Pleune et al.
Ta ble 1. Cr ysta l Da ta for All Com p ou n d s
1
4
6
formula
fw
C36H42MoP2
632.58
C36H43BF4P2W
808.30
C40H46B2F8MoN2P2
886.29
space group
a, Å
b, Å
P21/c
P21/n
Pna21
22.245(2)
10.6805(9)
16.9237(11)
10.5920(6)
29.010(2)
11.3044(9)
114.402(5)
3163.3(4)
4
14.3353(13)
16.3703(11)
14.8550(14)
100.45(2)
3428.2(5)
4
c, Å
b, deg
V, Å3
Z
4020.9(6)
4
1.464
0.475
d
calcd, Mg/m3
1.328
0.539
1.566
3.509
m(Mo KR), mm-1
radiation (monochromated in incident beam)
temp, K
Mo KR (λ ) 0.71073 Å)
153(2)
153(2)
153(2)
final R indices [I > 2σ(I)]
R1a
0.0545
0.1006
0.0569
0.1073
0.0338
0.0735
wR2b
R indices (all data)
R1a
0.0763
0.1080
0.0810
0.1148
0.0461
0.0798
wR2b
a
b
2
R1 ) ∑||Fo| - |Fc||/∑|Fo|. wR2 ) [∑w(|Fo| - |Fc|)2/∑w|Fo| ]1/2; w ) 1/σ2(|Fo|).
Ta ble 2. Selected Bon d Dista n ces (Å) a n d An gles
(d eg) for Com p ou n d 1a
nances of compound 8. No indication of the formation of 7
was evidenced.
Mo(1)-P(1)
Mo(1)-P(2)
Mo(1)-CNT
Mo(1)-C(1)
Mo(1)-C(2)
Mo(1)-C(3)
2.3560(12)
2.3797(12)
2.006(5)
2.333(5)
2.353(5)
2.347(5)
Mo(1)-C(4)
Mo(1)-C(5)
Mo(1)-H(1)
Mo(1)-H(2)
Mo(1)-H(3)
2.321(4)
2.342(5)
1.65(5)
1.57(5)
1.55(5)
X-r a y Cr ysta llogr a p h y. (a ) Com p ou n d 1. An orange
parallelepiped crystal with dimensions of 0.50 × 0.225 × 0.125
mm was placed and optically centered on the Enraf-Nonius
CAD-4 diffractometer. The crystal cell parameters and ori-
entation matrix were determined from 25 reflections in the
range 17.6° < θ < 20.7° and confirmed with axial photographs.
Data (5852 reflections) were collected (Mo KR) with ω/2θ scans
in the (h,k,-l quadrant over the range 2.2 < θ < 25.0° with
a scan width of (0.71 + 0.47 tan θ)° and variable scan speed
of 2.0-2.3° min-1, with each scan recorded in 96 steps with
the outermost 16 steps on each end of the scan being used for
background determination. Six standard reflections well-
dispersed in reciprocal space were monitored at 1 h intervals
of X-ray exposure. Minor variations in intensity were ob-
served, and the data were not corrected for decay. An
absorption correction was applied based upon crystal faces
with transmission factors ranging from 0.8854 to 0.9382. Data
were corrected for Lorentz and polarization factors and reduced
to Fo2 and σ(Fo2). Intensity statistics and systematic absences
clearly determined the centrosymmetric monoclinic space
group P21/c (no 14). Averaging of equivalent reflections led
to 5554 unique intensities (R(int) ) 0.0288). The structure
was determined by direct methods with the successful location
of all non-hydrogen atoms and refined by full-matrix least-
squares cycles. A subsequent difference-Fourier map revealed
the location of several of the remaining hydrogen atoms within
the molecule. Since many were missing, all of the hydrogen
atoms bonded to carbon atoms were placed in calculated
positions, with the aromatic hydrogen atoms at dCH ) 0.950
Å and UH ) 1.2UC, the CH2 hydrogen atoms at dCH ) 0.990 Å
and UH ) 1.2UC, and CH3 hydrogen atoms at dCH ) 0.980 Å
and UH ) 1.5UC. The initial configuration of the CH3 groups
was determined with a rotational difference-Fourier map about
the central carbon atom and was further optimized under the
constraint of tetrahedral geometry. After several cycles of
refinement, all of the non-hydrogen atoms were refined aniso-
tropically and another difference-Fourier map revealed three
possible locations for hydrides bound to the central Mo atom.
Two of these were the highest peaks in the map (0.75 and 0.68
e Å-3), while the third potential hydride was the fifth highest
peak (0.45 e Å-3). These three hydrides were allowed to refine
freely (x, y, z, U). The structure was refined to convergence
(∆/σ e 0.001), and a final difference-Fourier map was feature-
less with |∆F| e 0.57 e Å-3. The function minimized during
P(1)-Mo(1)-P(2)
P(1)-Mo(1)-CNT
P(1)-Mo(1)-H(1)
P(1)-Mo(1)-H(2)
P(1)-Mo(1)-H(3)
P(2)-Mo(1)-CNT
P(2)-Mo(1)-H(1)
P(2)-Mo(1)-H(2)
81.21(4)
140.3(1)
60(2)
66(2)
96(2)
132.6(1)
70(2)
119(2)
P(2)-Mo(1)-H(3)
CNT-Mo(1)-H(1)
CNT-Mo(1)-H(2)
CNT-Mo(1)-H(3)
H(1)-Mo(1)-H(2)
H(1)-Mo(1)-H(3)
H(2)-Mo(1)-H(3)
68(2)
106(2)
102(2)
114(2)
123(3)
135(3)
66(3)
a
CNT is the centroid of atoms C(1)-C(5).
refinement parameters are collected in Table 1, and selected
bond distances an angles are listed in Table 2.
(b) Com p ou n d 4. A colorless plate with dimensions of
0.225 × 0.175 × 0.038 mm was placed and optically centered
on the Enraf-Nonius CAD-4 diffractometer. The crystal cell
parameters and orientation matrix were determined from 25
reflections in the range 15.7° < θ < 17.8° and confirmed with
axial photographs. Data (18 511 reflections with indices (h,-
k,l and (h,(k,-l) were collected as described above for
compound 1. No decay correction was necessary. An absorp-
tion correction was applied based upon crystal faces with
transmission factors ranging from 0.5551 to 0.8680. Data were
corrected for Lorentz and polarization factors and reduced to
2
Fo and σ(Fo2). Intensity statistics and systematic absences
clearly determined the centrosymmetric monoclinic space
group P21/n (no 14), nonstandard setting for P21/c. Averaging
of equivalent reflections led to 6006 unique intensities (R(int)
) 0.1053). The structure was determined by direct methods
with the successful location of the W and P atoms, along with
several C atoms. Refinement by alternating full-matrix least-
squares cycles and difference-Fourier maps revealed the
location of the remaining non-hydrogen atoms. After several
cycles of refinement, all of the non-hydrogen atoms were
refined anisotropically. Hydrogen atoms attached to carbon
atoms were placed in calculated positions, as described above
for compound 1. Four hydride H atoms were located from the
highest peaks near the tungsten atom, with W-H distances
ranging from 1.3 to 1.6 Å. These four positions are tentative
at best and more speculative than definitive. They were
initially refined freely, but were found to move continuously
and were, therefore, subsequently restrained in their refine-
ment in such a way that the W-H distance be near 1.6 Å and
2
the full-matrix least-squares refinement was Σw(Fo - Fc2),
where w ) 1/[σ2(Fo2) + (0.0297P)2 + 6.5319P] and P ) (max-
2
(Fo ,0) + 2Fc2)/3. An empirical correction for extinction was
found to be negative and not applied. The crystal data and

Studies of Cp*MH3(dppe) (M ) Mo, W)
Organometallics, Vol. 16, No. 8, 1997 1585
Ta ble 3. Selected Bon d Dista n ces (Å) a n d An gles
(d eg) for Com p ou n d 4a
Ta ble 4. Selected Bon d Dista n ces (Å) a n d An gles
(d eg) for Com p ou n d 6a
W(1)-P(1)
W(1)-P(2)
W(1)-CNT
W(1)-C(1)
W(1)-C(2)
W(1)-C(3)
2.477(2)
2.508(2)
1.990(9)
2.296(8)
2.338(9)
2.372(8)
W(1)-C(4)
W(1)-C(5)
W(1)-H(1)
W(1)-H(2)
W(1)-H(3)
W(1)-H(4)
2.344(9)
2.291(8)
1.58(5)
1.57(4)
1.67(4)
1.65(4)
Mo(1)-P(1)
Mo(1)-P(2)
Mo(1)-N(51)
Mo(1)-N(61)
Mo-CNT
2.541(2)
2.554(2)
2.137(6)
2.122(6)
1.998(7)
2.368(7)
Mo(1)-C(2)
Mo(1)-C(3)
Mo(1)-C(4)
Mo(1)-C(5)
Mo(1)-H(1)
2.247(6)
2.270(7)
2.380(6)
2.418(6)
1.69(6)
Mo(1)-C(1)
P(1)-W(1)-P(2)
P(1)-W(1)-CNT
P(1)-W(1)-H(1)
P(1)-W(1)-H(2)
P(1)-W(1)-H(3)
P(1)-W(1)-H(4)
P(2)-W(1)-CNT
P(2)-W(1)-H(1)
P(2)-W(1)-H(2)
P(2)-W(1)-H(3)
P(2)-W(1)-H(4)
78.72(7)
162.8(2)
70(3)
76(3)
78(2)
CNT-Mo(1)-H(1)
CNT-Mo(1)-H(2)
CNT-Mo(1)-H(3)
CNT-Mo(1)-H(4)
H(1)-W(1)-H(2)
H(1)-W(1)-H(3)
H(1)-W(1)-H(4)
H(2)-W(1)-H(3)
H(2)-W(1)-H(4)
H(3)-W(1)-H(4)
97(3)
CNT-Mo(1)-P(1)
CNT-Mo(1)-P(2)
159.5(2)
121.6(2)
P(1)-Mo(1)-H(1)
P(2)-Mo(1)-N(51)
P(2)-Mo(1)-N(61)
P(2)-Mo(1)-H(1)
59(2)
84.7(2)
84.9(2)
138(2)
107(3)
111(2)
93(3)
134(4)
75(4)
80(4)
128(4)
60(4)
CNT-Mo(1)-N(51) 103.1(2)
CNT-Mo(1)-N(61) 104.8(2)
CNT-Mo(1)-H(1)
P(1)-Mo(1)-P(2)
P(1)-Mo(1)-N(51)
P(1)-Mo(1)-N(61)
101(2)
N(51)-Mo(1)-N(61) 151.8(2)
74(2)
78.78(6) N(51)-Mo(1)-H(1)
75.0(2)
77.3(2)
87(2)
83(2)
118.4(2)
127(3)
74(3)
57(2)
131(3)
N(61)-Mo(1)-H(1)
a
CNT is the centroid of atoms C(1)-C(5).
147(3)
Resu lts
a
CNT is the centroid of atoms C(1)-C(5).
Syn th eses a n d NMR Ch a r a cter iza tion . The com-
pounds CpMH3(dppe) (M ) Mo, 1; W, 2) are produced
by the reaction of Cp*MCl4 and LiAlH4 in the presence
of dppe. The reaction presumably occurs via the forma-
tion of a monodentate phosphine adduct Cp*MCl4(η1-
dppe), see eq 1, as indicated by the EPR resonance
(doublet, g ) 1.960, aP ) 27.5 G) obtained by mixing
Cp*MoCl4 and dppe without the aluminum reagent. The
the four atoms define a plane and have similar U values,
leading to smooth convergence (∆/σ e 0.001) with R(F) )
8.10%, Rw(F2) ) 11.48%, and GOF ) 1.159 for all 6006 unique
reflections [R(F) ) 5.69% and Rw(F2) ) 10.73% for those 4677
data with Fo > 4σ(Fo)]. The H-W-H angles and the direction
of the H4 plane, however, were not restrained. Relative to the
model without the four hydride H atoms, the final residuals
are lower by ca. 0.1% (R1) or 0.2% (wR2). Although the final
hydride locations are acceptable in both position and distance
from the W atom, no reliability should be placed on them. A
final difference-Fourier map possessed many peaks near the
tungsten atom, within 1.2 Å, with the largest peak, |∆F| e 1.29
1. LiAlH4
Cp*MCl4 dppe8 Cp*MCl4(η1-dppe) 2. MeOH 8
Cp*MH3(dppe)
(1)
e Å-3
.
The weighting scheme used was w ) 1/[σ2(Fo2) +
M ) Mo, 1
W, 2
(0.0389P)2 + 9.0199P], P being defined as above. An empirical
correction for extinction was attempted but found to be
negative and not applied. The crystal data and refinement
parameters are collected in Table 1, and selected bond
distances an angles are listed in Table 3.
corresponding W derivative shows only a broad EPR
resonance with no discernible P coupling. The addition
of dppe to the Cp*MCl4 compounds is a rapid and
quantitative reaction, both in toluene and in THF. The
reduction by LiAlH4, on the other hand, proceeds slowly,
even when the Li reagent is solubilized by the addition
of ether.
(c) Com p ou n d 6. A red block with dimensions of 0.325 ×
0.175 × 0.125 mm was placed and optically centered on the
Enraf-Nonius CAD-4 diffractometer. The crystal cell param-
eters and orientation matrix were determined from 25 reflec-
tions in the range 15.8° < θ < 16.2° and confirmed with axial
photographs. Data (5439 reflections) were collected (Mo KR)
in the -h,(k,-l quadrant as described for the above cases.
No decay correction was applied. Six ψ-scan reflections were
collected over the range 6.5° < θ < 10.9°; the absorption
correction was applied with transmission factors ranging from
0.2592 to 0.2795. Data were corrected for Lorentz and
It is interesting to observe that a spontaneous reduc-
tion occurs for the system in eq 1, whereas the same
synthetic procedure on the related complex Cp*MoCl4-
(PMe3) produces the pentahydride product of formal
oxidation, Cp*MoH5(PMe3).51 Depending on the condi-
tions, the reaction leads to contamination of the product
with minor amounts of what appears to be Cp*MH5(η1-
dppe) (M ) Mo or W), as suggested by a doublet hydride
resonance at δ -3.06 (J PH ) 46.2 Hz) for Mo or -2.40
(J PH ) 50.9 Hz) for W. The formation of this by-product
can be minimized by using the toluene/Et2O solvent
mixture in the proportions indicated in the Experimen-
tal Section. The use of a larger proportion of Et2O leads
to increased amounts of the pentahydride by-product
(as does the use of THF), whereas the use of neat
toluene leads to unsatisfactorily slow reactions. Longer
reaction times than 12 h led to the formation of
increasing amounts of other uncharacterized by-prod-
ucts. The MeOH quenching procedure is also worth
briefly discussing. The MeOH quencher improves the
yield for both Mo and W systems, but also produces
varying amounts of the pentahydride by-product. Dur-
ing the synthesis of 1, little if any pentahydride is
produced, whereas substantial amounts of this by-
2
polarization factors and reduced to Fo and σ(Fo2). Intensity
statistics and systematic absences clearly determined the
centrosymmetric monoclinic space group Pna21 (no 33). Av-
eraging of symmetry-equivalent reflections led to 2725 unique
data (R(int) ) 0.0591). Direct methods led to the successful
location of Mo, two P, and several F and C atoms. Refinement
by full-matrix least-squares cycles followed by difference-
Fourier maps allowed the location of the remaining non-
hydrogen atoms. Refinement of the Flack absolute structure
parameter resulted in a value of 0.965(9). The structure was
therefore inverted, and the new absolute structure parameter
equaled -0.01(5). The H atom bonded to Mo was located
directly from a difference-Fourier map and refined freely. All
of the remaining H atoms were placed in calculated positions
and handled as described above for compound 1. The structure
was refined to convergence (∆/σ e 0.001), and a final differ-
ence-Fourier map was featureless with |∆F| e 0.42 e Å-3. The
weighting scheme used was w ) 1/[σ2(Fo2) + (0.0302P)2
+
1.6093P], P being defined as above. An empirical correction
for extinction was found to be negative and not applied. The
crystal data and refinement parameters are collected in Table
1, and selected bond distances an angles are listed in Table 4.
(51) Shin, J . H.; Parkin, G. Polyhedron 1994, 13, 1489-1493.

1586 Organometallics, Vol. 16, No. 8, 1997
Pleune et al.
product are produced during the synthesis of 2. There-
fore, the best procedure for the synthesis of 2 involves
isolation of the pure trihydride before being quenched
by MeOH, followed by quenching of the residue to
recover additional amounts of impure 2, which is
subsequently recrystallized.
Both compounds 1 and 2 exhibit the expected NMR
properties. In particular, the hydride ligands display
a single resonance at δ -5.27 (1) or -5.90 (2) with
coupling to two equivalent P nuclei (J PH ) 42.0 Hz for
1 and 44.6 Hz for 2), indicating high fluxional behavior.
The hydride resonance of the tungsten derivative also
shows the expected satellites, with J WH ) 54.4 Hz.
These properties correspond to those reported for other
derivatives of the same stoichiometry.42-46 Compounds
(C5H4Pri)MoH3(PMe3)2 and CpMoH3(PMe2Ph)2, in par-
ticular, had been investigated at temperatures as low
as -90 °C without the observation of a decoalescence
of the hydride resonance.44,46 The same behavior is
observed for compounds 1 and 2, the triplet hydride
resonance remaining sharp at -90 °C.
F igu r e 1. Variable-temperature 1H-NMR spectrum of
compound 4 in CDFCl2.
ever, does not turn into the expected doublet of doublets
for a frozen P2 system and averaged H4 system, thus
indicating that the fast exchange of the four H ligands
may also start to freeze out at this temperature.
Addition of acetonitrile to 3 at room temperature
induces gas evolution and the formation of 5 as the only
observed product, see eq 3. The 31P{1H}-NMR spectrum
The reaction of 1 or 2 with 1 equiv of HBF4‚Et2O in
diethyl ether at room temperature precipitated [Cp*MH4-
(dppe)]BF4 (M ) Mo, 3; W, 4) in high yields, see eq 2.
-
Cp*MH (dppe) + HBF f
[Cp*MH4(dppe)]+BF4
3
4
M ) Mo, 3
W, 4
(2)
[Cp*MoH4(dppe)]+BF4- + MeCN f
- + H2 (3)
Compound 3 is unstable and loses H2 upon dissolution
into any coordinative solvent (vide infra). Dissolution
in CH2Cl2 at room temperature also induced H2 loss and
decomposition to unknown products. The formation of
this compound is related to its immediate precipitation
from the reaction medium. The crude material, which
is probably already contaminated by products of H2 loss,
cannot be recrystallized, and satisfactory analytical data
could, consequently, not be obtained. The NMR proper-
ties of the crude product could, however, be investigated
in CDFCl2 at -60 °C, under which conditions little
decomposition occurs. Compound 4, on the other hand,
is stable even in MeCN at room temperature, and single
crystals for an X-ray analysis (vide infra) could be
grown from MeCN/Et2O.
[Cp*MoH2(MeCN)(dppe)]+BF4
5
of 5 exhibits a singlet at room temperature for the
coordinated dppe ligand, which splits into a triplet (J PH
) 52 Hz) upon selective 1H decoupling, proving that the
compound is a dihydride derivative. The 31P{1H} singlet
is decoalesced at 193 K into two doublets at δ 82.8 and
75.3 (J PP ) 20.8 Hz), indicating that the two phosphorus
atoms are in an inequivalent coordination environment
and that the room temperature spectrum is the result
of a fluxional process. The averaging of the P nuclei is
also noticeable in the line shape of the hydride reso-
nance in the 1H-NMR spectrum: the single hydride
resonance at -5.50 ppm is a binomial triplet (J PH ) 53
Hz) at high temperature (80 °C) but it broadens upon
cooling. At the lowest achieved temperature (-90 °C),
however, the signal has not yet achieved the slow
exchange limit (see Figure 2). The line shape appears
more complicated than the doublet of doublets expected
for two equivalent H nuclei coupled to two inequivalent
P nuclei, thus indicating that decoalescence of two
inequivalent and rapidly exchanging H nuclei is also
taking place.
Compound 5 is also a product of the direct protonation
of 1 in MeCN. This procedure, however, also results in
the formation of two isomers of a product of double
protonation [Cp*MoH(MeCN)2(dppe)](BF4)2, 6 and 7,
both of which have been isolated. The formation of
these double protonation products occurs even when a
substoichiometric amount of acid is slowly added to
compound 1. Redissolution of isolated 5 in MeCN and
protonation with HBF4‚Et2O also leads to the formation
of 6 and 7. Whatever the method of generation of these
two compounds, their initial relative ratio is always ca.
1:3. An intermediate during the transformation of
The 1H-NMR spectra of 3 and 4 exhibit a single
resonance for the four hydride ligands, which is down-
field-shifted (∆δ ) +1.71 for Mo, +3.17 for W) and more
weakly coupled (∆J PH ) -6.0 Hz for Mo, -12.0 Hz for
W; ∆J HW ) -16.7 Hz for W) relative to 1 and 2,
respectively. The 31P{1H}-NMR resonances of 3 and 4
are upfield-shifted (∆δ ) -18.9 for Mo, -16.3 for W)
relative to 1 and 2, respectively, and the P-W coupling
is reduced for 4 relative to 2 (∆J PW ) -8.9 Hz). The
presence of four hydride ligands is indicated by the
quintet pattern in the selectively 1H-decoupled 31P-NMR
spectra of 3 and 4. Cooling to -90 °C has no effect on
1
the line shape of the H- and 31P-NMR resonances of
compound 3. On the other hand, the 31P-NMR reso-
nance of 4 has decoalesced at this temperature into two
doublets at δ 48.6 and 54.7 (J PP ) 26.3 Hz). This
indicates the inequivalence of the two P donors in the
solution structure and agrees with the solid state
structure (vide infra). This decoalescence is also re-
1
flected in the shape of the H-NMR resonance for the
hydride ligands (see Figure 1). This resonance, how-

Studies of Cp*MH3(dppe) (M ) Mo, W)
Organometallics, Vol. 16, No. 8, 1997 1587
F igu r e 3. Representative 1H-NMR spectra for the thermal
treatment of a mixture of compounds 6 and 7 in CD3CN.
The different regions of the spectra are on different vertical
scales.
F igu r e 2. Temperature dependence of the 1H-NMR hy-
dride resonance of compound 5. The solvent is CD3CN for
the spectra at 80 and 20 °C and CD2Cl2 for all the other
spectra.
complex 5 into 6 and 7 could not be observed: proton-
ation in CD2Cl2 at low temperature led to immediate
loss of the hydride signal and formation of uncharac-
terized, hydride-free products.
Compound 6 crystallizes selectively from a solution
containing compounds 5, 6, and 7. Its X-ray structure
1
has been determined (vide infra). The H-NMR spec-
trum of 6 exhibits a doublet of doublets at room
temperature for the hydride resonance at -4.05 ppm,
with one small and one large phosphorus coupling.
Broad-band P decoupling collapses this resonance into
a singlet. The 31P{1H}-NMR spectrum shows two
mutually coupled doublets, consistent with a rigid
structure containing inequivalent phosphorus donors.
Spectroscopically and analytically pure 7 is obtained
by chromatographic separation from the mixture con-
taining 6 and 7 over acidic alumina. While 6 is trapped
by the column, 7 can be eluted by a MeCN/H2O mixture.
The NMR spectroscopic properties of compound 7 are
similar to those of compound 6. The 1H-NMR spectrum
exhibits a doublet of doublets hydride resonance at
-2.18 ppm, with one small and one large phosphorus
coupling, which collapsed into a singlet by broad-band
P decoupling, and two mutually coupled doublets are
observed in the 31P{1H}-NMR spectrum. The elemental
analysis is consistent with the same stoichiometry as
6. The presence of a single hydride ligand is further
confirmed by a 31P{selective-1H}-NMR spectrum: the
two doublets split into doublets of doublets, the down-
field and the upfield resonances showing HP coupling
corresponding to the small and the large HP couplings,
respectively, observed for the hydride resonance in the
1H-NMR spectrum. An important difference between
6 and 7, however, is the observation of two distinct
MeCN resonances for 7, whereas 6 shows a single
resonance for two equivalent MeCN ligands. No dis-
cernible J HP coupling is observed for the 1H-NMR
resonance of coordinated MeCN for any of the com-
pounds described in this contribution, even with the use
of resolution enhancement.
F igu r e 4. A ORTEP view of the molecular geometry for
compound 1. Ellipsoids are drawn at the 30% probability
level.
Warming a mixture of 6 and 7 in CD3CN to 70 °C for
1
2 h with H- and 31P-NMR monitoring initially shows a
decrease in intensity for the resonances of 7 and a
simultaneous increase for the resonances of 6. We
conclude that 6 and 7 are geometric isomers, the former
being thermodynamically preferred. Obviously, the
formation of a nonequilibrium ratio of 6 and 7 by
protonation of 5 indicates that the two isomers must
be formed by kinetically controlled independent path-
ways. Continued thermal treatment of the mixture of
6 and 7, however, reveals the formation of another
product, 8, whose NMR properties are extremely close
to those of 6, see Figure 3. The formation of 8 is
quantitative. Warming a solution of pure 6 also pro-
duces selectively and quantitatively 8, without the
detection of 7. The cation of compound 8 appears,
therefore, to be yet another isomer of [Cp*MoH(MeCN)2-
(dppe)]2+, whose structure must be quite close to that
of 6. Compound 8 resembles 6 also in having a single
resonance for the two MeCN ligands.
Str u ctu r a l Ch a r a cter iza tion s. The single-crystal
X-ray structure of compound 1 is shown in Figure 4.
The position of the three hydrides was directly located
from the difference-Fourier synthesis and refined with-
out contraints. The coordination geometry can be
related to a pseudo trigonal prism, with the center of

1588 Organometallics, Vol. 16, No. 8, 1997
Pleune et al.
Ch a r t 1
the Cp* ligand and two hydride ligands defining one
triangular face and the dppe and the third hydride
ligand defining the opposite triangular face, as il-
lustrated schematically in I, Chart 1. The ubiquitous
geometry for a (ring)MX3L2 stoichiometry, on the other
hand, is the pseudo octahedron, for which three different
stereochemistries (II-IV) are possible. Examples for
each of the pseudo octahedral geometries have previ-
ously been reported, for instance CpMoCl3[P(OCH2)3-
CEt]2 of type II,52 CpMoCl3(L-L) (L-L ) dppe53 or
dmpe54 ) of type III, and CpMoCl3(PMe2Ph)2 of type
IV.55 The only other previously reported solid state
structure of a trihydride compound having the (ring)-
MH3L2 stoichiometry is for CpMoH3(PMe2Ph), which is
of type IV.46
For a regular trigonal prism, identical CNT-Mo-P
angles should be observed. Compound 1 exhibits only
slightly different CNT-Mo-P angles, indicating a small
twist toward an octahedral geometry. The difference
between these angles should be much greater in a
structure of type III (e.g., see the structure of 6 below),
whereas the CNT-Mo-P angles should be much smaller
for a structure of type II. In addition, the Mo atom is
displaced significantly from the plane defined by CNT
and the two P atoms (by 0.283 Å), one hydride ligand
being located on one side of the plane and the other two
hydride ligands being located on the other side, as can
be clearly seen from the top view in Figure 5.
The average Mo-P distance of 2.368(12) Å is far
shorter than those in the isoelectronic CpMoCl3(dppe),53
which is of type III (axial, 2.521(2) Å; equatorial,
2.688(4) Å), and CpMoCl3(PMe2Ph)2 (average 2.554(1)
Å), which is of type IV.56 It is, on the other hand,
similar to the average Mo-P distance in CpMoH3(PMe2-
Ph)2 (2.407(3) Å), also of type IV.46 If these distances
are regulated mostly by steric interactions, the replace-
ment of three Cl ligands with three H ligands is more
important than the replacement of the Cp with Cp*. The
less electronegative H ligands could also have the effect
of expanding the metal orbitals and favoring stronger
Mo-P covalent interactions (both σ and π). The Mo-H
distances average 1.59(5) Å, which compares with
1.52(4) Å for (C5Me4Et)MoH5(PMe3)51 and with 1.64(5)
Å for CpMoH3(PMe2Ph)2.46 The only available Mo-H
distances from neutron diffraction studies appear to be
those of Cp*MoH(CO)3 (1.789(7) Å)57 and CpMoH(CO)3
F igu r e 5. A top view of the compound 1. Drawing
parameters are as for Figure 4.
(1.720(5) Å).58 The metal-hydrogen distances deter-
mined by X-ray diffraction are usually an underestima-
tion of the real internuclear distances, but shorter
Mo-H distances relative to the above mentioned cyclo-
pentadienyl tricarbonyl derivatives are also expected
because of the higher formal oxidation state. The
distance between H(2) and H(3) of 1.71 Å is much longer
than the longest separation reported for a “dihydrogen”
ligand, e.g., 1.357(7) Å for ReH7[P(p-tolyl)3]2;59 thus, 1
can be considered as a classical trihydride complex.
Given the similarity of the atomic radii of Mo(IV) and
W(IV),60 we assume that the structure of 2 is identical
to that of 1.
Compound 3 could not be structurally investigated in
the solid state because the thermal instability prevents
the growth of single crystals, and the solution charac-
terization by NMR does not allow a conclusive structural
assignment. The related tungsten compound 4, on the
other hand, provided suitable single crystals for an
X-ray crystallographic investigation. Unfortunately, the
location of the hydride ligands was not as clear cut as
for the above described compound 1 (see Experimental
Section). The use of restraints on the the W-H dis-
tances and the H4 plane, but allowing free refinement
of the H-W-H angles and the orientation of the H4
plane, led to the determination of the distorted pseudo
pentagonal bipyramid shown in Figure 6. The P atom
is only 0.27 Å away from the least-squares H4 plane,
and the five angles between adjacent bonds in the
pseudo pentagonal plane are relatively similar. The
deviation of the ideal parallelism between the pseudo
pentagonal plane and the Cp* ring is probably caused
by the steric repulsion between the bulky equatorial
PPh2 ligand and the Cp* ring. The axial W-P bond is
almost perfectly perpendicular to the least-squares
pentagonal plane, while the W-CNT vector is tilted from
this plane by ca. 20°. As a result, the P(1)-W-L (L )
equatorial ligand) angles are essentially equivalent
(average 75(4)°), whereas the CNT-W-L angle is
greater for L ) P(2), intermediate for L ) H(2) and H(3)
(52) Poli, R.; Kelland, M. A. J . Organometal. Chem. 1991, 419, 127-
136.
(53) Sta¨rker, K.; Curtis, M. D. Inorg. Chem. 1985, 24, 3006-3010.
(54) Owens, B. E.; Poli, R. Inorg. Chim. Acta 1991, 179, 229-237.
(55) Abugideiri, F.; Gordon, J . C.; Poli, R.; Owens-Waltermire, B.
E.; Rheingold, A. L. Organometallics 1993, 12, 1575-1582.
(56) Abugideiri, F.; Keogh, D. W.; Poli, R. J . Chem. Soc., Chem.
Commun. 1994, 2317-2318.
(58) Koetzle, T. F. Personal communication. See also footnote 7b in
ref 57.
(59) Brammer, L.; Howard, J . A. K.; J ohnson, O.; Koetzle, T. F.;
Spencer, J . L.; Stringer, A. M. J . Chem. Soc., Chem. Commun. 1991,
241-243.
(60) Pauling, L. The Nature of the Chemical Bond, 3rd ed.; Cornell
University Press: Ithaca, New York, 1960.
(57) Brammer, L.; Zhao, D.; Bullock, R. M.; McMullan, R. K. Inorg.
Chem. 1993, 32, 4819-4824.

Studies of Cp*MH3(dppe) (M ) Mo, W)
Organometallics, Vol. 16, No. 8, 1997 1589
are no reported terminal W-H distances from neutron
diffraction studies. Two recently reported bridging
W-H distances determined by neutron diffraction are
1.926(2) Å in [PPN]2[H2W2(CO)8]64 and 1.897(5) Å in
[PPh4][HW2(CO)10].65
The classical nature of all hydride ligands in com-
pound 4 is consistent with the absence of visible HD
coupling in the 4-d3 isotopomer, which was synthesized
by protonation with HBF4 of compound 2-d3. The latter
compound is available by the same procedure described
for compound 2 (see eq 1), except for the use of LiAlD4
and quenching with CH3OD. The hydride signal of this
species is a sharp triplet at δ ) -2.80 (i.e., 0.07 ppm
upfield relative to the hydride resonance of 4 in CD2Cl2).
The fine structure due to coupling to the three D nuclei
is not visible, even upon resolution enhancement. From
the line width of the resonance, it is estimated that J HD
e 1 Hz. The proposed solid state pseudo-pentagonal-
bipyramidal structure, V, agrees with the measured
longitudinal relaxation time (T1) for the hydride 1H
resonance in solution for both compounds 3 and 4. The
minimum value, T1(min), was measured as 59 ms for 3
at 203 K and 83 ms for 4 at 203 K (both at 200 MHz).
Calculations3 based on geometry V with a M-H dis-
tance of 1.71 Å (Mo) and 1.72 Å (W) (data adapted from
the optimized geometries66 of CpMH5(PH3) with the
same pentagonal bipyramidal geometry; the M-H
distances determined by X-ray crystallography are not
reliable and tend to be artificially shorter than the true
internuclear distances) and a 15° bending of the H
ligands away from the equatorial plane (from the
average P(1)-W(1)-L angle of 75° in the structure of
4, coinciding with the same angle in MP2 geometry
optimized CpMH5(PH3)66) led to an estimation of T1 of
172 ms for 3 and 178 ms for 4 at 200 MHz. The H-Cp*
and H-dppe dipolar interactions were neglected in
these calculations. These values are significantly larger
than those determined experimentally. However, the
presence of a bulky PPh2 group in the pentagonal plane
is likely to significantly distort the regular pentagon and
push the hydride ligands closer to each other, reducing
the calculated value of T1. The precision of the X-ray
determined hydride positions is not sufficient to verify
this hypothesis. A reduction of the adjacent H‚‚‚H
distances from 1.94 Å to 1.62 Å for 3 and from 1.95 Å to
1.72 Å for 4 would allow a perfect fit of the data.
Calculations for a pseudo octahedral nonclassical
[Cp*MH2(H2)(dppe)]+ structure, assuming a 1.0 Å H-H
distance for the dihydrogen ligand and 2.3 Å cis H‚‚‚H
contacts, lead to an estimation for T1 of 10 ms (slow H2
rotation limit) or 38 ms (fast rotation limit) at 200 MHz.
However, stretching of the H2 ligand to 1.0 Å and
beyond usually corresponds to restrained H2 rotation.67
Therefore, we conclude that complexes 3 and 4 are
classical tetrahydrido derivatives with a pseudo pen-
tagonal bipyramidal structure.
F igu r e 6. An ORTEP view of the molecular geometry for
the cation of compound 4. Ellipsoids are drawn at the 30%
probability level.
(i.e., the hydride ligands cis relative to P(2)), and smaller
for L ) H(1) and H(4).
Even when the less accurately determined hydride
positions are not considered, the relative positions of the
Cp* and dppe ligands are most consistent with a pseudo
pentagonal bipyramidal geometry. In particular, the W
atom lies essentially on the plane identified by CNT and
by the P(1) and P(2) atoms (deviation ) 0.003 Å), and
the observed CNT-W-P angles are quite different
(162.8(2)° and 118.4(2)°). Possible alternative struc-
tures are based on a capped trigonal prism derived from
the observed structure of 1 (vide supra). In the latter
case, however, simple capping of a rectangular face by
a hydride ligand would generate a structure such as VI,
with similar CNT-W-P angles and with the W atom
substantially off the plane defined by CNT and the two
P atoms (Chart 2). An alternative capped pseudo
trigonal prism where the two P donors occupy different
positions, such as VII, would also lead to greater devia-
tions from coplanarity of the W, P(1), P(2), and CNT
positions. Furthermore, previously reported d0 (ring)-
ML6 polyhydride structures are also based on the pseudo
pentagonal bipyramid, i.e., (C5Me4Et)MoH5(PMe3)51,61
and Cp*ReH6.62 Possible alternatives of nonclassical
dihydrogen complexes seem excluded by NMR (J HD and
T1 measurements, vide infra).
The pseudo-equatorial W-P distance is significantly
longer than the pseudo-axial one. Their average (2.49(2)
Å) is close to the W-PMe3 distances in Cp*WHCl(η2-
CH2PMe2)(PMe3) (2.4805(8) Å)63 and Cp*WH(PPh2)2-
(PMe3) (2.476(1) Å).63 The average W-H distance is
1.62(5), but the significance of this value is nulled by
the restraints used during the refinement. X-ray de-
termined terminal W-H bond lengths are 1.72(5) Å for
(η1-(P):η5-C5Me4CH2PCy2)WH(PMe3)2 (a W(II) com-
pound)63 and 1.55(4) Å for Cp*WH(PPh2)2(PMe3) (a
W(IV) compound).63 To the best of our knowledge, there
The geometry of compound 6 is unambiguously de-
termined by X-ray crystallography, including the posi-
(64) Wei, C.-Y.; Marks, M. W.; Bau, R.; Kirtley, S. W.; Bisson, D.
E.; Henderson, M. E.; Koetzle, T. F. Inorg. Chem. 1982, 21, 2556-
2565.
(61) Abugideiri, F.; Kelland, M. A.; Poli, R. Organometallics 1993,
12, 2388-2389.
(65) Hart, D. W.; Bau, R.; Koetzle, T. F. Organometallics 1985, 4,
1590-1594.
(66) Lin, Z.; Hall, M. B. Organometallics 1993, 12, 4046-4050.
(67) Bautista, M. T.; Cappellani, E. P.; Drouin, S. D.; Morris, R. H.;
Schweitzer, C. T.; Sella, A.; Zubkowski, J . J . Am. Chem. Soc. 1991,
113, 4876-4887.
(62) Herrmann, W. A.; Theiler, H. G.; Kiprof, P.; Tremmel, J .; Blom,
R. J . Organometal. Chem. 1990, 395, 69-84.
(63) Baker, R. T.; Calabrese, J . C.; Harlow, R. L.; Williams, I. D.
Organometallics 1993, 12, 830-841.

1590 Organometallics, Vol. 16, No. 8, 1997
Pleune et al.
Ch a r t 2
Sch em e 1
increasing the basicity at the metal center and favoring
the adoption of classical structures. The above consid-
erations rationalize rather well the observation of a
classical structure for the trihydride title compounds 1
and 2 (formally d2 complexes of Mo(IV) or W(IV)), their
susceptibility to protonation, and the classical structure
for their protonation products 3 and 4 (formally d0
complexes of Mo(VI) or W(VI)).
Compounds 3 and 4 show a markedly different
stability toward reductive elimination of H2 and coor-
dination of a donor molecule: while 3 easily loses H2
even at low temperature in noncoordinating solvents,
compound 4 is stable in MeCN at room temperature.
The substitution of H2 by MeCN in 3 to give 5 presum-
ably takes place via the dihydrogen complex intermedi-
ate [Cp*MH2(H2)(dppe)]+; thus, we can imagine that the
equilibrium on the right hand side of Scheme 1 is shifted
further to the right for W relative to Mo, in line with
the known greater basicity of 5d metals relative to their
4d counterparts.70
F igu r e 7. An ORTEP view of the molecular geometry for
the cation of compound 6. Ellipsoids are drawn at the 30%
probability level.
tion of the hydride ligand which was located and freely
refined. This geometry can be described as a distorted
pseudo octahedron, i.e., derived from a geometry of type
III by replacing the two relative trans hydride ligands
with as many MeCN ligands, see Figure 7. The steric
interaction between the Cp* and equatorial PPh2 groups
causes a severe distortion, which compresses the P(1)
donor against the hydride ligand (P(1)-Mo-H(1) )
59(2)°). As found for the structure of 4 described above,
the CNT-Mo-P angles are quite different from each
other (104.8(2)° and 159.5(2)°) and the Mo atom lies
essentially in the plane defined by the two P atoms and
CNT (deviation 0.033 Å). The Mo-H distance of 1.69(6)
Å compares relatively well with those of compound 1,
while the Mo-P distances are significantly longer than
those for 1 or CpMoH3(PMe2Ph)2,46 see above. This
effect is consistent with either or both an increase of
steric bulk or/and a reduced Mo-P electronic interaction
as a result of the substitution of two H- by two MeCN
ligands.
The high basicity of the systems examined in this
contribution is highlighted by the occurrence of a second
protonation process, followed by further H2 loss and
MeCN coordination, to form different isomers of
[Cp*MoH(MeCN)2(dppe)]2+. The overall process is likely
to take place according to the mechanism outlined in
Scheme 2. The presence of an intermediate of type XV
or XVI, e.g., [Cp*MoH3(MeCN)(dppe)]2+, however, could
not be verified. Evidently, compound [Cp*MoH(η2-H2)-
(MeCN)(dppe)]2+ must be quite electron poor and im-
mediately releases H2. Other transition metal polyhy-
dride systems that have been reported to undergo a
double protonation process are ReH5(PPh3)2(py), OsH6-
Discu ssion
(PPri ) , and OsH2(dppe)2, generating [ReH(MeCN)3-
3 2
Ba sicity a n d P r oton a tion . Tuning the electron
density on the metal center for transition metal poly-
hydride complexes controls, at the same time, the
basicity of the complex and the choice between classical
and nonclassical structures. A greater electron density
favors protonation as well as the adoption of classical
structures, see Scheme 1. The effect of the metal
electron density on the classical/nonclassical structural
issue for cyclopentadienyl derivatives has been exam-
ined theoretically by Hall et al.66,68,69 Cyclopentadienyl
and phosphine coligands are strong electron donors,
(PPh3)(py)]2+ 71
,
[OsH4-2n(PPri ) (MeCN)2+n
]
(n ) 0,
2+
3 2
1),72 and [Os(H2)(dppe)2(MeCN)]2+ 73,74
respectively.
,
(68) Lin, Z.; Hall, M. B. Organometallics 1992, 11, 3801-3804.
(69) Bayse, C. A.; Couty, M.; Hall, M. B. J . Am. Chem. Soc. 1996,
118, 8916-8919.
(70) Collman, J . P.; Hegedus, L. S.; Norton, J . R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry;
University Science Books: Mill Valley, CA, 1987.
(71) Allison, J . D.; Walton, R. A. J . Chem. Soc., Chem. Commun.
1983, 401-403.
(72) Smith, K.-T.; Tilset, M.; Kuhlman, R.; Caulton, K. G. J . Am.
Chem. Soc. 1995, 117, 9473-9480.

Studies of Cp*MH3(dppe) (M ) Mo, W)
Organometallics, Vol. 16, No. 8, 1997 1591
Sch em e 2
crystallographically characterized compounds 1 (tri-
gonal prism) and 6 (octahedron). The NMR properties
indicate fluxional sets of hydride ligands and P donors.
The pseudo-octahedral structure VIII, therefore, is
excluded because it has equivalent hydrides, Chart 3.
Possible alternatives are the isomeric octahedral struc-
ture IX and the pseudo-trigonal-prismatic structure X.
As dicussed more extensively below, structure X fits
with the structural trends of the other compounds
described in this paper, and an easy rationalization of
the coalescence of the P and H resonances is possible.
Structures IX and X allow an easy rationalization of the
protonation results (formation of the two isomers 6 and
7, see below).
As shown by Figure 3, three isomers for complex
[Cp*MoH(MeCN)2(dppe)]2+ have been observed by NMR.
Isomer 6 has been characterized by X-ray crystal-
lography as the pseudo-octahedral mer,trans isomer (see
Figure 7). Several structural possibilities remain for
isomers 7 and 8. However, the inequivalence of the P
donors, as shown by 31P-NMR spectroscopy, excludes the
possibility of symmetrical structures. A few remaining
possibilities are XI-XV, Chart 4. Geometries XI and
XII are based on the pseudo octahedron, while geom-
etries XIII-XV are based on the pseudo trigonal prism.
In XIII, both P donors are as far as possible from the
Cp* ligand, as found for the structure of compound 1
(which presumably minimizes the steric repulsion be-
tween Cp* and dppe ligands), whereas in XIV, one P
donor occupies a position adjacent to Cp*, and both P
donors are adjacent to Cp* in XV. Additional possibili-
ties are generated by exchanging the H and MeCN
positions in XIV.
Str u ctu r a l P r efer en ces. As mentioned in the Re-
sults section, the ubiquitous geometry adopted by (ring)-
ML5 complexes is the pseudo octahedron. To the best
of our knowledge, there are no previous examples of
(ring)ML5 compounds with a geometry based on the
pseudo trigonal prism. For a d2 system, the pseudo-
octahedral geometry should be electronically favored
because the six σ bonds formed by the s, p, and d (eg
set) orbitals are augmented by the two Mo-Cp π bonds
that use the dxz,dyz set, while the two metal electrons
would reside in the dxy orbitals, which can engage in
Mo-Cp δ back-bonding. Calculations at the MP2 level
on the model system CpMoH3(PH3)2 have shown that a
pseudo-trigonal-prismatic structure like I is 8.95 kcal/
mol higher in energy relative to the most favored
structure II, while calculations on the sterically more
hindered CpMoH3(PMe3)2 place structure IV as the most
stable one by 2.92 kcal relative to II.46 When the two
neutral L donors are connected via a backbone, such as
for the dppe ligand, only type II and III structures
within the pseudo-octahedral geometry could be adopted
in principle. Indeed, a geometry that can be described
as distorted type III is adopted by compound 6. The
steric interaction between the bulky dppe and Cp*
ligands, however, certainly disfavors geometry III rela-
tive to I. It is quite possible that the small size of the
three H- ligands in compounds 1 and 2 can be most
easily accommodated in structure I, whereas the ex-
change of two H- with as many MeCN ligands in
compound 6 introduces additional repulsive interactions
(i.e., MeCN-Cp* and MeCN-dppe) that tilt the energy
balance back toward geometry III.
The pattern of J PH for compounds 7 and 8 is similar
to that of 6 (one small and one large J HP constant), the
spectrum of 8 being remarkably close to that of 6. The
observation of equivalent MeCN ligands for 8 rules out
all the alternative structures XI-XV and leaves only
the possibility of a geometry identical with that of 6.
The difference between 6 and 8 must therefore be due
to a conformational change in the coordination periph-
ery (i.e., the Cp* or dppe ligand). The interesting
conclusion from the thermal study (see Figure 3) is that
the protonation process leads selectively to the higher
energy conformer (6), which then converts quantita-
tively to the more stable one (8).
The observation of two distinct MeCN resonances for
compound 7 does not provide any additional decisive
factor for the structural assignment. Additional con-
siderations, however, can be made. As mentioned
above, the pseudo octahedral geometry has an electronic
preference over the pseudo-trigonal-prismatic one,
whereas the pseudo-trigonal-prismatic structure is
probably sterically driven by placing the two encum-
bered phosphorus donors further away from the Cp*
ligand. It is, therefore, reasonable to assume that any
other trigonal-prismatic geometry will not be favored
unless both P donors are as far as possible from the Cp*
ligand. This argument eliminates geometries XIV and
XV from consideration. Among the pseudo-octahedral
alternatives, XII has both P donors in a cis position
relative to the Cp* ligand and is, thus, likely to be less
stable than XI. Structure XI is similar to the structure
observed for isomer 6, except that the positions of the
hydride ligand and one of the MeCN ligands are
Compound 5 presents an interesting case because it
is sterically and electronically midway between the
(73) Earl, K. A.; J ia, G.; Maltby, P. A.; Morris, R. H. J . Am. Chem.
Soc. 1991, 113, 3027-3039.
(74) Schlaf, M.; Lough, A. J .; Maltby, P. A.; Morris, R. H. Organo-
metallics 1996, 15, 2270-2278.

1592 Organometallics, Vol. 16, No. 8, 1997
Pleune et al.
Ch a r t 3
Ch a r t 4
exchanged. Since the steric interaction between Cp*
and dppe pushes the axial P donor close to the ligand
which is located trans relative to the equatorial P donor
(the hydride for 6, a MeCN ligand for XI), structure XI
is probably sterically less favored than 6 because of the
greater bulk of MeCN relative to H. This idea would
seem consistent with the lower thermodynamic stability
of 7 relative to 6. Structure XI, however, is not
consistent with the NMR properties observed for 7, in
particular the pattern of J HP values. In structure XI,
the hydride ligand is located in a cis position relative
to both P nuclei, leading to the expectation of similar
J PH values, whereas the two observed values are quite
different from each other. By exclusion, we are left with
XIII as the most likely structure for compound 7. This
structure fits the observed J PH pattern, one P donor
being at a much smaller angle than the other relative
to the Mo-H bond. As will be seen in a later section,
this assignment also allows a straightforward rational-
ization of the mechanistic details of the protonation and
isomerization processes. Our efforts at growing single
crystals of compounds 7 and 8 have not so far met with
success.
highly fluxional for 1 (n ) 0), fluxional at room tem-
perature but frozen (on the 31P-NMR time scale) at -80
°C for 5 (n ) 1), and frozen at room temperature for 7
(n ) 2).
F lu xion a l Mech a n ism s. The observation of differ-
ent structures for isoelectronic compounds of the same
class (i.e., pseudo trigonal prismatic for 1, 5, and 7;
pseudo octahedral for 6 and 8), naturally induces us to
propose an interconversion between these two structural
types as the hydrogen scrambling mechanism in the
various compounds. For compounds 1 and 2, this
process would indeed easily scramble all hydride ligands
among all possible positions, as shown in Scheme 3. The
process involves subsequent interconversions of geom-
etries of types I and III and consists in subsequent 60°
rotations of triangular faces that can be related to the
Bailar twist of the true octahedral coordination geom-
etry. This mechanism has also been recently proposed
for the hydride scrambling in the isoelectronic CpOsH5
compound as a result of a theoretical study.69
The same mechanism proposed for the hydride scram-
bling in compounds 1 and 2 can account for the hydride
and phosphorus exchange process in 5, see Scheme 4.
It is likely that the two P and the two hydridic H nuclei
in structure XVIII are averaged via the pseudo-
octahedral structure XIX and the symmetric pseudo-
trigonal-prismatic structure XX. The rearrangement
via the intermediates XXI and XXII would only average
the hydride resonances and not the phosphorus reso-
nances (i.e., Ha and Hb exchange their positions, but Pb
always remain the closest donor to the MeCN ligand
and is thereby not exchanged with Pa). The latter
process, at any rate, should be less favored than that
via XIX and XX, because structure XXI is more steri-
One more point of structural discussion about com-
pound 5 is proper at this point. It was previously shown
that possible structures are those illustrated in IX and
X. However, it has now been shown that one isomer of
complex [Cp*MoH(MeCN)2(dppe)]2+ (i.e., the cation of
7) probably adopts a pseudo-trigonal-prismatic struc-
ture. For this reason, we think it more likely that 5
also adopts the pseudo-trigonal-prismatic structure X.
If this is true, the series of compounds [Cp*MoH3-n
-
(MeCN)n(dppe)]n+ show a monotonous trend of fluxion-
ality for the same pseudo-trigonal-prismatic structure:

Studies of Cp*MH3(dppe) (M ) Mo, W)
Organometallics, Vol. 16, No. 8, 1997 1593
Sch em e 3
Sch em e 4
cally crowded than XIX, having the bulkier MeCN
ligand in the crowded plane defined by the Cp* center
and the dppe ligand, in close proximity to the axial P
donor. The Pb-Mo-NCMe angle in XXI should be
significantly smaller than 90° because of the distortion
caused by the Cp*-dppe steric repulsion (e.g., compare
with the structure of 6). This fluxional mechanism
would also be consistent with XIX as the ground state
structure for compound 5.
For the pseudo-pentagonal-bipyramidal compounds 3
and 4, possible scrambling mechanisms would involve
interconversion with a capped pseudo trigonal prism or
with a capped pseudo octahedron. The interconversion
via equilibration with a nonclassical H2 complex and H2
rotation, however, cannot be excluded, especially for the
Mo compound.
Mech a n ism of P r oton a tion of [Cp *MoH2(MeCN)-
(d p p e)]+ a n d Isom er iza tion of [Cp *MoH(MeCN)2-
(d p p e)]2+. The protonation mechanism of 5 must lead
independently to the different [Cp*MoH(MeCN)2(dppe)]2+
cations of 6 and 7, since these are simultaneously
produced and they do not easily interconvert. A simple
scheme, which incorporates the pseudo trigonal prism-
octahedron interconversion shown above in Schemes 3
and 4, leads to a satisfactory rationalization of all the
observations (see Scheme 5). We propose that proton-
ation of 5 occurs independently at the two inequivalent
hydride positions Ha and Hb of structure XVIII (paths
a and b). The protonated product, rather than collaps-
ing to a classical trihydride derivative, loses H2 to yield
intermediates XXIII and XXIV. At this point, these
intermediates can be trapped by MeCN or rearrange to
a pseudo-octahedral environment, followed by MeCN
trapping. Trapping of XXIII would produce a sym-
metrical pseudo-trigonal-prismatic product XXV, which
is not observed. If this path is followed, intermediate
XXV must have a low barrier pathway for interconver-
sion to one of the observed isomers, e.g., 7 (see Scheme
5). Rearrangement of XXIII, on the other hand, leads
to two different intermediates XXVI and XXVII, de-
pending on which direction the dppe ligand swings, see
Scheme 5. Intermediate XXVI is likely to be favored
because the most sterically hindered position (meridi-
onal with the dppe ligand) is vacant, whereas this is
occupied by a MeCN ligand in XXVII. However, both
intermediates yield the same product, cis-[Cp*MoH-
(MeCN)2(dppe)]2+ (i.e., XI), by MeCN trapping. This is
again an unobserved structure, probably because of
unfavorable PPh2(equatorial)-Cp* and PPh2(axial)-
MeCN steric interactions. These repulsions can be
minimized by transformation to 7, which requires only
minimal rearrangements (a pseudo Bailar twist of the
triangular face defined by the two P and one of the two
MeCN donors) and is thus likely to be a fast process.
Protonation of the position Ha in compound 5 according
to the mechanism of Scheme 5, therefore, would lead to
compound 7 as the only product. Direct trapping of
intermediate XXIV by MeCN would lead directly to the
observed pseudo-trigonal-prismatic 7. On the other
hand, rearrangement leads to intermediates XXVIII
and XXIX, depending on the direction of the dppe ligand
swing. The latter intermediate should be more favored
than the former, because the position meridional with
dppe is occupied by the less bulky hydride ligand.
Trapping of XXVIII leads again to XI (which then
transforms to 7), whereas trapping of XXIX leads to
trans-[Cp*MoH(MeCN)2(dppe)]2+, i.e., the cation of 6.
Thus, this scheme rationalizes quite well the forma-
tion of both isomers 6 and 7, the latter one being the
predominant product. The exact ratio between 6 and 7
is expected to depend on several factors, such as the
relative susceptibility of Ha and Hb to protonation, the
relative ratio of direct MeCN trapping of XXIII and
XXIV vs rearrangement, and the kinetic control of the
rearrangement of XXIV to XXVIII and XXIX. Within
this mechanistic hypothesis, the direct formation of 6
would demonstrate that at least a certain amount of
intermediate XXIV must rearrange before being trapped
by MeCN. According to the proposed scheme and
structural assignments, the conversion of 7 into 6

1594 Organometallics, Vol. 16, No. 8, 1997
Pleune et al.
Sch em e 5
involves a further 60° pseudo Bailar twist of the same
triangular face, which leads to the transformation of XI
into the cation of 7. Experimentally, the former process
is found to be much slower than the latter, for reasons
that do not appear obvious. If the cation of 5 adopted
structure XIX as the ground state structure, the obser-
vation of a mixture of 6 and 7 under kinetically
controlled conditions would still be possible: protonation
of positions a and b followed by H2 loss would lead
directly to intermediates XXVI and XXIX, respectively,
which would lead to the two observed products. If this
were the case, however, the observed relative amounts
of 6 and 7 could only be explained by a 3:1 kinetic
preference for protonation of position a relative to b.
MH3L2 class of derivatives. While the observation of
this structure in the ground state is probably fortuitous
and related to the Cp*-dppe steric repulsion and the
presence of the ethylene backbone in the dppe ligand,
the occurrence of pseudo-trigonal-prismatic structures
as intermediates for the hydride scrambling process in
other (ring)MH3L2 systems may be a general phenom-
enon. The stepwise substitution of H- with a MeCN
ligand in the series of complexes [Cp*MoH3-n
-
(MeCN)n(dppe)]n+ (n ) 0, 1, 2) slows down the hydride
and phosphine scrambling processes, while at the same
time disfavoring the pseudo-trigonal-prismatic structure
in favor of the ubiquitous pseudo-octahedral structure.
It is our intent to explore next the chemical and
electrochemical oxidation of compounds 1 and 2 in
various solvents and the reactivity of the resulting
oxidation products.
Con clu sion
The compounds Cp*MH3(dppe) (M ) Mo, W) are
electron-rich systems, easily yielding classical tetrahy-
dride cationic products [Cp*MH4(dppe)]+, which differ
in stability (Mo < W) in accord with the expected greater
electron donating capability of W relative to Mo. Loss
of H2 and coordination of MeCN for the Mo system
triggers a second protonation process to produce two
isomers of [Cp*MoH(MeCN)2(dppe)]2+ under kinetically
controlled conditions, while a third and thermodynami-
cally more stable isomer is accessible thermally. This
work has highlighted for the first time the involvement
of pseudo-trigonal-prismatic structures for the (ring)-
Ack n ow led gm en t. Support for this work by the
Department of Energy (grant No. DEFG059ER14230 to
R.P.) is gratefully acknowledged. We also thank Mr.
J ung-Soo Yi for technical assistance.
Su p p or tin g In for m a tion Ava ila ble: Tables of crystal
data, atomic coordinates, bond distances and angles, and
anisotropic displacement parameters for compounds 1, 4, and
6 (51 pages). Ordering information is given on any current
masthead page.
OM961002O