Proton transfer in aminocyclopentadienyl ruthenium hydride complexes

Article abstract of DOI:10.1021/om990301l

A new ruthenium hydride complex of the aminocyclopentadienyl ligand (Cp-N)RuH(PPh₃)₂ (Cp-N = C₅H₄CH₂CH₂NMe₂, 1) has been prepared and characterized by X-ray diffraction. Protonation of 1 with excess HPF₆ leads to the dicationic derivative [(Cp-NH)RuH₂(PPh₃)₂]-(PF₆) ₂ (2), in which both the metal and the amino substituent have been protonated. Addition of 1 equiv of HBF₄·Et₂O to 1 leads to the complex [(Cp-N)Ru(PPh₃)₂](BF₄) (3), containing a chelating amino cyclopentadienyl ligand after elimination of H₂. However, using (HNEt₃)-(BPh₄) or (HPBu₃)(BPh₄) as protonating agent, it is possible to form [(Cp-NH)RuH(PPh₃)₂]-(BPh₄) (4), which was isolated as yellow crystals of 4·H₂O upon addition of undistilled methanol and characterized by X-ray crystallographic analysis. A fluxional process exchanging the ammonium proton and the hydride without changing the thermodynamic state of the system could be established by 1H NMR, and activation energies of 11 kcal·mol^-1 were calculated for 4·H₂O and the product resulting from in situ addition Of [HNEt₃][BPh₄] to 1, whereas an activation energy of 10.1 kcal·mol^-1 was found for the product resulting from in situ addition of [HPBu₃][BPh₄] to 1. A density functional study (B3PW91) was carried out, and the dihydrogen bond in the model system for 4 was calculated to be 1.545 A?, in excellent agreement with T₁ measurements (1.52 A?). The proposed mechanism for the fluxional process does not involve a proton transfer within the dihydrogen bond.

Full text of DOI:10.1021/om990301l

Organometallics 1999, 18, 3981-3990
3981
P r oton Tr a n sfer in Am in ocyclop en ta d ien yl Ru th en iu m
Hyd r id e Com p lexes
J ose´ A. Ayllon, Stephen F. Sayers, Sylviane Sabo-Etienne,*
Bruno Donnadieu, and Bruno Chaudret*
Laboratoire de Chimie de Coordination CNRS, 205 Route de Narbonne,
F-31077 Toulouse-Cedex 04, France
Eric Clot
Laboratoire de Structure et Dynamique des Syste`mes Mole´culaires et Solides,
Case Courrier 14, Universite´ de Montpellier 2, Place Euge`ne Bataillon,
F-34095 Montpellier-Cedex, France
Received April 26, 1999
A new ruthenium hydride complex of the aminocyclopentadienyl ligand (Cp-N)RuH(PPh3)2
(Cp-N ) C5H4CH2CH2NMe2, 1) has been prepared and characterized by X-ray diffraction.
Protonation of 1 with excess HPF6 leads to the dicationic derivative [(Cp-NH)RuH2(PPh3)2]-
(PF6)2 (2), in which both the metal and the amino substituent have been protonated. Addition
of 1 equiv of HBF4‚Et2O to 1 leads to the complex [(Cp-N)Ru(PPh3)2](BF4) (3), containing a
chelating amino cyclopentadienyl ligand after elimination of H2. However, using (HNEt3)-
(BPh4) or (HPBu3)(BPh4) as protonating agent, it is possible to form [(Cp-NH)RuH(PPh3)2]-
(BPh4) (4), which was isolated as yellow crystals of 4‚H2O upon addition of undistilled
methanol and characterized by X-ray crystallographic analysis. A fluxional process exchang-
ing the ammonium proton and the hydride without changing the thermodynamic state of
1
the system could be established by H NMR, and activation energies of 11 kcal‚mol-1 were
calculated for 4‚H2O and the product resulting from in situ addition of [HNEt3][BPh4] to 1,
whereas an activation energy of 10.1 kcal‚mol-1 was found for the product resulting from in
situ addition of [HPBu3][BPh4] to 1. A density functional study (B3PW91) was carried out,
and the dihydrogen bond in the model system for 4 was calculated to be 1.545 Å, in excellent
agreement with T1 measurements (1.52 Å). The proposed mechanism for the fluxional process
does not involve a proton transfer within the dihydrogen bond.
In tr od u ction
ligand hydrogen-bonded to the acidic proton of a thiolato
pyridinium ligand.2a,d Intermolecular interactions were
also evidenced by Crabtree in the solid state in an
adduct of ReH5(PPh3)3 with indole and in the H3NBH3
adduct,3 whereas Epstein and Berke4 have identified
such interactions in solution. Complexes displaying such
hydrogen bonding have been proposed to be important
intermediates for the protonation of hydrides or the
reverse reaction, namely, the base-promoted heterolytic
splitting of dihydrogen. In this respect, we have recently
described the first observation of a reversible proton
transfer between the dihydride complex trans-RuH2-
(dppm)2 involved in hydrogen bonding with hydrogen
bond donors such as hexafluoro-2-propanol or phenols
and a dihydrogen derivative.5 We have furthermore
demonstrated the influence of hydrogen bonding in
solution on the magnitude of exchange couplings in
The presence of hydrogen bonds between a transition
metal hydride and a hydrogen bond donor containing
an O-H or an N-H group (dihydrogen bonds) has
recently been established by Crabtree1 and Morris2
using various spectroscopic methods as well as solid-
state structures. In these cases, intramolecular hydro-
gen bonding was rendered possible by the presence on
the same complex of a hydride and a ligand containing
the hydrogen bond donor. These structures seem adapted
to proton transfer, and for instance, a dihydrogen
tautomer has been proposed to be involved in the
deuteration reaction of a complex containing a hydride
* Corresponding author. Fax: (33) 5 61 55 30 03. E-mail: Chaudret@
lcc-toulouse.fr.
(1) (a) Crabtree, R. H.; Siegbahn, P. E. M.; Eisenstein, O.; Rheingold,
A.; Koetzle, T. F. Acc. Chem. Res. 1996, 29, 348. (b) Lee, J . C., J r.;
Peris, E.; Rheingold, A. L.; Crabtree, R. H. J . Am. Chem. Soc. 1994,
116, 11014. (c) Peris, E.; Lee, J . C., J r.; Rambo, J . R.; Eisenstein, O.;
Crabtree, R. H. J . Am. Chem. Soc. 1995, 117, 3845. (d) Peris, E.;
Wessel, J .; Patel, B. P.; Crabtree, R. H. J . Chem. Soc., Chem. Commun.
1995, 2175. (e) Yao, W.; Crabtree, R. H. Inorg. Chem. 1996, 35, 3007.
(2) (a) Lough, A. J .; Park, S.; Ramachandran, R.; Morris, R. H. J .
Am. Chem. Soc. 1994, 116, 8356. (b) Park, S.; Ramachandran, R.;
Lough, A. J .; Morris, R. H. J . Chem. Soc., Chem. Commun. 1994, 2201.
(c) Park, S.; Lough, A. J .; Morris, R. H. Inorg. Chem. 1996, 35, 3001.
(d) Schlaf, M.; Lough, A. J .; Morris R. H. Organometallics 1996, 15,
4423.
(3) (a) Wessel, J .; Lee, J . C., J r.; Peris, E.; Yap, G. P. A.; Fortin, J .
B.; Ricci, J . S.; Sini, G.; Albinati, A.; Koetzle, T. F.; Eisenstein, O.;
Rheingold, A. L.; Crabtree, R. H. Angew. Chem., Int. Ed. Engl. 1995,
34, 2507. (b) Patel, B. P.; Yao, W.; Yap, G. P. A.; Rheingold, A. L.;
Crabtree, R. H. J . Chem. Soc., Chem. Commun. 1996, 991.
(4) Shubina, E. S.; Belkova, N. V.; Krylov, A. N.; Vorontsov, E. V.;
Epstein, L. M.; Gusev, D. G.; Niedermann, M.; Berke, H. J . Am. Chem.
Soc. 1996, 118, 1105.
(5) Ayllon, J . A.; Gervaux, C.; Sabo-Etienne, S.; Chaudret, B.
Organometallics 1997, 16, 2000.
10.1021/om990301l CCC: $15.00 © 1999 American Chemical Society
Publication on Web 09/10/1999

3982 Organometallics, Vol. 18, No. 20, 1999
Ayllon et al.
Cp*RuH3(PCy3)6 and used these modifications of ex-
change couplings to demonstrate the transfer of protons
within a dihydrogen bond.6b
The existence of “dihydrogen bonds” having been
established and the ease of proton transfer inside these
associated structures having been demonstrated, it was
interesting to try to study the proton transfer by
imposing a geometry to the hydrogen bond through the
design of the ligand sphere. Calculations by Akorta and
Elguero have demonstrated the possibility of undergoing
proton transfer in a strong electric field.7 J alon et al.
have studied proton tranfer in pyridylphenylphosphine
ruthenium complexes and evidenced by saturation
transfer the existence of an equilibrium between a
proton linked to nitrogen and a hydride.8 In our group
we have chosen to use cyclopentadienyl ligands func-
tionalized with amino groups.9 Thus, the recent reports
concerning the preparation on a multigram scale of
sodium salts of amino-substituted cyclopentadienyl
ligands, NaCp-N (Cp-N ) C5H4CH2CH2NMe2, C5H4CH2-
CH2CH2-NMe2, C5H4CH(CH2CH2)2NMe) allowed an
easy access to both the ligands and their metal com-
plexes.10 In previous studies we used piperidylcyclopen-
tadienyl ligands. This led in iridium chemistry to the
observation of dynamic equilibrium between a proton
linked to the piperidyl moiety and two hydrides,11
whereas in the ruthenium case, a so far not explained
strong decrease of the relaxation time of the hydride
(down to <2 ms) was found.12 To build a more flexible
system, we considered using cyclopentadienyl ligands
accommodating linear amino groups such as C5H4CH2-
CH2NMe2 or C5H4CH2CH2CH2NMe2. However, during
the course of this work, Lau and co-workers reported a
very similar study involving these two ligands coordi-
nated to a ruthenium hydrido dppm fragment.13 In their
case, they have been able to demonstrate by several
techniques the presence of a hydrogen bond between a
proton linked to the amine moiety and a hydride and
to show the formation of a dicationic dihydrogen com-
plex when using an excess of protonating agent. They
have however not observed in their complex the hydride/
proton exchange within the dihydrogen bond, nor could
they obtain in this system the crystal structure of the
monoprotonated species.
F igu r e 1. CAMERON view of (C5H4CH2CH2NMe2)RuH-
(PPh3)2 (1). Selected bond lengths (Å): Ru-H(1), 1.55(3);
Ru-P(1), 2.2630(8); Ru-P(2), 2.2619(8); Ru-C(1), 2.250-
(3); Ru-C(2), 2.227(3); Ru-C(3), 2.239(4); Ru-C(4), 2.239-
(4); Ru-C(5), 2.245(4). Selected bond angles (deg): P(1)-
Ru-P(2), 98.83(3); P(1)-Ru-H(1), 79.2(19); P(2)-Ru-H(1),
85.9(21).
protonation of 1, leading, according to the protonation
reagent, to chelation or formation of a novel hydride
complex [(Cp-NH)RuH(PPh3)2]BPh4‚H2O (4), also char-
acterized by X-ray crystallography. We also report the
dynamic NMR spectra of the latter species. Moreover
density functional calculations (B3PW91) have been
carried out to characterize the conformation of the
products and to elucidate the mechanism of the ex-
change process. With respect to the dppm system
described by Lau and co-workers,13 our experimental
and theoretical study highlights the influence of any
ligand modification (here phosphines) on these sensitive
hydrogen-transfer processes.
We report in this paper the synthesis of the new
hydride complex (Cp-N)RuH(PPh3)2 (Cp-N ) C5H4CH2-
CH2NMe2) (1) and its X-ray crystal structure and the
Resu lts
Syn th esis a n d Ch a r a cter iza tion of th e Neu tr a l
Com p lex 1. The reaction of RuH(OCOMe)(PPh3)3
14
with NaCp-N in methanol at reflux for 45 min leads to
the formation of (Cp-N)RuH(PPh3)2 (1), (Cp-N ) C5H4-
CH2CH2NMe2) in 72% yield after appropriate workup.
The complex was characterized by analytical and spec-
troscopic methods as well as by X-ray crystallographic
analysis (see Figure 1). The most salient spectroscopic
(6) (a) Ayllon, J . A.; Sabo-Etienne, S.; Chaudret, B.; Ulrich, S.;
Limbach, H.-H. Inorg. Chim. Acta 1997, 259, 1. (b) Gru¨ndemann, S.;
Ulrich, S.; Limbach, H.-H.; Golubev, N. S.; Denisov, G. S.; Epstein, L.;
Sabo-Etienne, S.; Chaudret, B. Inorg. Chem. 1999, 38, 2550.
(7) Rozas, I.; Akorta, I.; Elguero, J . Chem. Phys. Lett. 1997, 275,
423.
(8) Caballero, A.; J alon, F. A.; Manzano, B. R. J . Chem. Soc., Chem.
Commun. 1998, 1879.
(9) (a) J utzi, P.; Siemeling, U. J . Organomet. Chem. 1995, 500, 175.
(b) J utzi, P.; Bangel, M.; Neumann, B.; Stammel, H.-G. Organometal-
lics 1996, 15, 4559, and references therein.
(10) (a) MacGowan, P. C.; Hart, C. E.; Donnadieu, B.; Poilblanc, R.
J . Organomet. Chem. 1997, 528, 191. (b) Philippopoulos, A. I.; Had-
jiladis, N.; Hart, C. E.; Donnadieu, B.; MacGowan, P. C.; Poilblanc, R.
Inorg. Chem. 1997, 36, 1842.
(11) Abad, M. M.; Atheaux, I.; Maisonnat, A.; Chaudret, B. Chem.
Commun. 1999, 381.
(12) Castellanos, A.; Ayllon, J . A.; Sabo-Etienne, S.; Donnadieu, B.;
Chaudret, B.; Yao, W.; Kavallieratos, K.; Crabtree, R. H. C.-R. Acad.
Sci., Ser. IIc 1999, 359.
1
features are the observation in the H NMR spectrum
recorded in CD2Cl2 of a hydride near -11 ppm (t, J P-H
34 Hz), of two pseudotriplets for the cyclopentadienyl
protons near 4.3 and 3.8 ppm, and of the methyl groups
attached to nitrogen near 2.1 ppm, as well as the
observation in the 31P NMR spectrum of a singlet at 70.1
ppm. Other 1H and 13C NMR data are given in the
Experimental Section. The infrared spectrum recorded
(13) Chu, H. S.; Lau, C. P.; Wong, K. Y.; Wong, W. T. Organome-
tallics 1998, 17, 2768.
(14) Mitchel, R. W.; Spencer A.; Wilkinson G. J . Chem. Soc., Dalton
Trans. 1973, 846.

Aminocyclopentadienyl Ru Hydride Complexes
Organometallics, Vol. 18, No. 20, 1999 3983
nitrogen resonate at 2.73 ppm. The 31P NMR spectrum
shows a singlet at 64.62 ppm. The mechanism of this
reaction probably involves protonation of the hydride
and fast elimination of dihydrogen. The driving force is
most probably the formation of a stable five-membered
ring, many examples of which have been reported. These
points will be illustrated in the theoretical section.
Ta ble 1. Cr ysta l Da ta for 1 a n d 4
C45H45NP2Ru [C45H46NP2Ru][B(C6H5)4]‚H2O
(1)
(4)
fw
762.79
1.34
5.26
1085.14
1.27
3.67
Fcalcd (g‚cm-3
)
µ (cm-1
F000
)
769.62
monoclinic
Pc
2299.35
monoclinic
P21/c
crystal system
space group
a (Å)
P r oton Tr a n sfer by Am m on iu m or P h osp h o-
n iu m Sa lts. The reaction of 1 with (HNEt3)(BPh4) was
first studied in situ in an NMR tube. In contrast to the
reaction with HBF4, no dihydrogen elimination is
observed when the reaction is carried out at room
temperature, but a new complex is formed: 4‚NEt3,
corresponding to 1 + (HNEt3BPh4). At room tempera-
9.8147(8)
10.030(1)
19.316(2)
100.901(9)
1867
30.462(4)
10.381(1)
18.175(3)
100.24(2)
5656
b (Å)
c (Å)
â (deg)
V (Å3)
Z
2
4
crystal size
crystal color
radiation type
0.3 × 0.3 × 0.2
light yellow
Mo KR
0.2 × 0.1 × 0.07
yellow
1
ture, the signals in the H NMR spectrum at 4.57 and
Mo KR
0.71073
140
3.53 ppm, due to the Cp protons, are shifted compared
to those of 1 and close to the position of the signals of
the Cp protons of 3, whereas the amino methyl groups
resonate at 2.06 ppm, i.e., close to the position of those
of 1. The 31P NMR spectrum displays a single resonance
near 66 ppm. The most interesting spectroscopic feature
is the presence of a very broad peak centered at -0.95
wavelength (Å) 0.71073
temp (K)
160
2θ range (deg)
no. of measured 14534
reflns
2.9-48.4
2.3-42.0
23673
no. of indep
reflns
no. of reflns
used
5531
5872
3182
5208
1
ppm, which integrates for two protons in the H NMR
[I > 3σ(I)]
spectrum (200 MHz). At 400 MHz and 193 K, two broad
peaks are observed at -11.68 and +11.9 ppm, respec-
tively, due to a hydride and an ammonium proton. When
the temperature is slowly increased, the high-field peak
broadens up to a temperature of 253 K, after which it
is no longer visible. The low-field peak shifts to higher
field as the temperature rises and also disappears at a
temperature higher than 253 K. At room temperature,
no signal is visible because of the coalescence. An
increase in temperature could not be carried out because
of decomposition of the product, but as stated above,
upon lowering the field of observation (200 MHz), the
coalescence peak is observed at -0.95 ppm. The system
is fully reversible. This observation indicates the pres-
ence of a two-site exchange between two protons on the
NMR time scale. One is a hydride, whereas the other is
an ammonium proton, hence indicating that protonation
of the complex has occurred. The energy of activation
of the process, derived from the coalescence tempera-
ture, was found to be 11 kcal‚mol-1. This process is only
a two-site exchange and not a thermodynamic proton
transfer. No formation of a dihydride, of a dihydrogen
complex or of the chelate complex 3 was observed.
R
Rw
GOF
0.026
0.032
0.7
0.054
0.048
1.2
on a sample dispersed in Nujol shows the frequency of
the Ru-H stretch at 1941 cm-1. The X-ray crystal
structure of 1 is shown in Figure 1 (see Table 1). The
Ru-H distance is equal to 1.55(3) Å, whereas no
significant change induced by the presence of the amino
group was observed for the other distances compared
to CpRuCl(PPh3)2.15
P r oton a tion of 1. P r oton a tion w ith Excess HP F 6.
Protonation of 1 with excess HPF6‚H2O in methanol at
room temperature leads to the dicationic derivative
[(Cp-NH)RuH2(PPh3)2] (PF6)2 (2) isolated in 69% yield
and in which both the metal and the amine moiety have
been protonated. Interestingly, this reaction produces
exclusively the dihydride isomer, as deduced from the
observation of a triplet near -7.4 ppm (J P-H 24.0 Hz)
1
for the hydrides in the H NMR spectrum. The proto-
nation of the amino group is apparent by the downfield
shift of the methyl groups to ca. 2.8 ppm and by the
observation of the proton linked to nitrogen at 6.6 ppm.
Other interesting spectroscopic features are an AA′BB′
pattern for the Cp protons at 5.08 and 4.81 ppm and
the shift of the 31P NMR signal near δ 61 ppm.
We chose to carry out the same reaction with (HPBu3)-
(BPh4) since the 1H-31P coupling could give us a
supplementary tool to characterize the proton exchange.
The in situ reaction with 1 produces a complex very
similar but not identical to the previous one, 4‚PBu3
corresponding to 1 + (HPBu3BPh4). For example the Cp
protons resonate at 4.72 and 3.33 ppm and the methyl
groups attached to nitrogen at 2.09 ppm. A broad peak
is observed at room temperature at -0.3 ppm (400 MHz,
CD2Cl2) for the rapidly exchanging signals of the hy-
dride and of the proton attached to nitrogen. At low
temperature (<233 K) this signal splits into two signals
observed at 11.7 and -12.1 ppm and attributed respec-
tively to the N-H proton and to the hydride. Both
signals remain in slow exchange since they display the
same relaxation time; this is obvious in Figure 2, which
shows the inversion recovery spectra of 1 + (HPBu3-
BPh4) at 213 K and 400 MHz. The barrier of activation
P r oton a tion of 1 w ith Stoich iom etr ic Am ou n ts
of HBF 4‚Et2O (Sch em e 1). Upon protonation of 1 with
stoichiometric amounts of HBF4‚Et2O in diethyl ether,
a new complex, namely, [(Cp-N)Ru(PPh3)2](BF4) (3),
accommodating a chelating amino cyclopentadienyl
ligand, was prepared, together with traces (ca. 1%) of
the diprotonated [(Cp-NH)Ru(H)2(PPh3)2](BF4)2, similar
1
to 2, as deduced from the observation in H NMR of a
hydride signal at -7.33 ppm (J P-H 24 Hz). 1H NMR data
of the major product show no hydride signal but a
significant modification of the chemical shift of the
cyclopentadienyl protons compared to 1, which resonate
at 4.65 and 3.57 ppm, whereas the methyl groups on
(15) Bruce, M. I.; Wong, F. S.; Skelton, B. W.; White, A. H. J . Chem.
Soc., Dalton Trans. 1981, 1398.

3984 Organometallics, Vol. 18, No. 20, 1999
Sch em e 1. P r oton a tion Rea ction s of 1
Ayllon et al.
F igu r e 2. Inversion recovery spectra of 1 + (HPBu3BPh4)
at 213 K and 400 MHz, showing the low-field amino proton
and the high-field hydride resonances. The delay times
shown are in ms.
F igu r e 3. CAMERON view of [(C5H4CH2CH2NHMe2)-
RuH(PPh3)2](BPh4)‚H2O (4‚H2O). Selected bond lengths
(Å): Ru-H(1), 1.58(13); Ru-P(1), 2.248(2); Ru-P(2), 2.270-
(2); Ru-C(1), 2.244(7); Ru-C(2), 2.235(8); Ru-C(3), 2.239-
(7); Ru-C(4), 2.238(8); Ru-C(5), 2.287(8); N(1)-H(2),
1.02(3). Selected bond angles (deg): P(1)-Ru-P(2), 100.75-
(7); P(1)-Ru-H(1), 80.6(45); P(2)-Ru-H(1), 94.5(46).
of the process was calculated to be 10.1 kcal‚mol-1
,
slightly lower than in the case of 1 + (HNEt3BPh4). The
1
absence of H-P coupling in H and 31P NMR spectra of
the PBu3 group demonstrates that the proton has been
transferred from phosphorus to nitrogen. However, a
fluxional process is visible by 31P NMR spectroscopy at
higher temperatures. The peak attibuted to PBu3 at -29
ppm disappears at temperatures higher than 263 K,
suggesting the involvement of this molecule in the
fluxional proton-transfer process. A low T1 minimum
value (130 ms, 400 MHz) was observed for both the
hydride and the ammonium proton in this system.
Using 1 as a reference for all relaxation effects besides
the dihydrogen bond and a methodology previously
described, we can calculate an approximate value of the
H‚‚‚H separation (1.52 Å) which is consistent with the
presence of a dihydrogen bond in solution.16
It was not possible to isolate pure any complex
incorporating the triethylamine or tributylphosphine
molecules. However addition of diethyl ether to a
dichloromethane reaction solution of 1 + HPBu3BPh4
or 1 + HNEt3BPh4 leads to the precipitation of powders,
which from NMR and microanalytical data definitely
incorporate the phosphine or amine ligand. After several
attempts, crystals could be grown from dichlorometh-
ane/wet methanol, leading to [(Cp-NH)RuH(PPh3)2]-
(BPh4)‚H2O (4‚H2O) and characterized by X-ray diffrac-
tion (see Figure 3 and Table 1). The structure shows a
single Ru-H(1) bond (1.58(13) Å) and a N-H(2) bond
(1.02(3) Å) in agreement with the exclusive protonation
on the amine moiety. No hydrogen bonding, whether
inter- or intramolecular, is present in the structure. All
(16) Moreno, B.; Sabo-Etienne, S.; Chaudret, B.; Rodriguez, A.;
J alon, F.; Trofimenko, S. J . Am. Chem. Soc. 1995, 117, 7441.

Aminocyclopentadienyl Ru Hydride Complexes
Organometallics, Vol. 18, No. 20, 1999 3985
orientation of the chain with respect to the ruthenium
center do compare very well with the experimental data.
P r oton a tion of th e Mon oh yd r id e. In complex A,
there are three different basic sites that can react with
an acid and yield a cationic species (Figure 5). Proto-
nation at site a yields a trans dihydride complex Ba ,
whose optimized geometry is given in the Supporting
Information along with some geometrical parameters.
The two Ru-H bond distances are typical for ruthenium
hydrides (1.612 and 1.606 Å). The particular geometrical
feature of Ba is the larger angle between the two
phosphorus atoms as a consequence of the trans geom-
etry of the hydrides (92.1° for A vs 102.6° for Ba ). This
type of structure would thus be very difficult to obtain
with chelating diphosphine because the P-Ru-P angle
is constrained to remain below a certain value.
Protonation at site b yields a dihydrogen complex Bb
with a H-H bond distance of 0.879 Å (Supporting
Information). As expected, the P-Ru-P angle is not
very large (90.9°), and this kind of structure could be
observed with a chelating diphosphine like dppm.13
Finally, protonation at site c yields the ammonium
complex Bc with a typical N-H bond distance (1.023
Å), and the P-Ru-P angle has not varried much (91.2°),
which indicates that this structure could also be ob-
tained with a chelating diphosphine.
The most stable of the three structures shown in
Figure 5 is Bb, with Bc lying only at 0.3 kcal‚mol-1
above Bb and Ba lying at 2.6 kcal‚mol-1 above Bb.
There is thus not a strong thermodynamic preference
for one site of protonation. The higher energy of Ba may
be due to the more important geometrical reorganization
required and especially the opening of the P-Ru-P
angle. However this difference is not very important,
and we cannot rationalize the differences observed
experimentally for 1 when it is reacted with either HBF4
or HNEt3 by just looking at the relative stability of Ba ,
Bb, and Bc.
F or m a tion of th e Ch ela te Com p lex. Reaction of 1
with HBF4 yields the complex [(Cp-N)Ru(PPh3)2](BF4),
3, containing a chelating amino cyclopentadienyl ligand.
We have optimized the model compound [(Cp-N)Ru-
(PH3)2]+, C, with a chelating amino chain on the Cp ring
(Supporting Information). The geometrical parameters
for C compare very well with the geometry of the related
complex [(Cp-N)Ru(dppm)]+ published by Lau and co-
workers.13 The system {C + H2} is 2.6 kcal.mol-1 less
stable than the dihydrogen complex Bb, indicating that
the later might be observed by protonation at very low
temperature. However the chelate compound is entropi-
cally favored with increasing temperature.
F igu r e 4. Calculated (B3PW91) geometry for A showing
the atom labeling and the various possible sites for proto-
nation. Selected bond lengths (Å): Ru-P, 2.260; Ru-H(1),
1.612; Ru-C(1), 2.274; Ru-C(2), 2.255; Ru-C(3), 2.292;
C(1)-C(6), 1.498; C(6)-C(7), 1.547; C(7)-N, 1.454. Selected
bond angles (deg): P-Ru-P, 92.0; P-Ru-H(1), 81.4; C(1)-
C(6)-C(7), 111.3; C(6)-C(7)-N, 116.9.
the other distances and angles are typical of such half-
sandwich complexes. The NMR spectra of crystals of
4‚H2O are totally similar to those of the mixture 1 +
HNEt3BPh4. This could indicate that neither triethyl-
amine nor water plays a role in the exchange process
even if either of these bases is present in the solid-state
structure of 4 or, more likely, that they play a similar
role, i.e., give additional hydrogen bonding to the
ammonium proton, thus preventing the transfer and
dihydrogen elimination and/or favoring the exchange
process proposed in the theoretical section (vide infra).
In the case of the mixture 1 + HPBu3BPh4, we observe
an easier proton/hydride exchange as well as the disap-
pearance of the PBu3 signal above 263 K, in agreement
with the involvment of the additional base in the proton-
transfer process through hydrogen bonding or transient
re-formation of a phosphonium salt.
Th eor etical Calcu lation s. Str u ctu r e of th e Mon o-
h yd r id e. The optimized geometry for the model com-
pound RuH(PH3)2(Cp-N), A (Cp-N ) C5H4CH2CH2-
NMe2) (Figure 4), is in very good agreement with the
X-ray structure obtained for RuH(PPh3)2(Cp-N), 1 (Fig-
ure 1 and Supporting Information). The bonding of the
substituted Cp ring is well described, although the
calculated Ru-C bond distances are slightly longer than
the experimental ones. However the trend among the
three different bonds (Ru-C1, Ru-C2, and Ru-C3) is
correctly reproduced. The computed Ru-P bond dis-
tance is in excellent agreement with the experimental
value, and the calculated smaller P-Ru-P angle (92.1°
vs 98.83°) is due to the nature of the model phosphine
PH3, which underestimates the steric repulsion between
the substituents on the phosphorus atoms. The calcu-
lated Ru-H1 distance of 1.612 Å is typical of ruthenium-
hydride distances computed by density functional theory
and compares very well with experimental value.17
There are several possible conformations for the CH2-
CH2NMe2 chain on the Cp ring. The conformation
corresponding to the chelate system is a priori energeti-
cally not favored. For the monohydride A this kind of
conformation is 18.1 kcal‚mol-1 less stable than the
conformation shown in Figure 1, while for Bb it is 7.4
kcal‚mol-1 less stable (Bb vs Bb′ in Figure 6). Cp rings
The amino chain on the Cp ring is also qualitatively
well described with, however, two discrepancies among
the C-C bond distances. The C1-C6 and C6-C7 bond
distances are calculated to be smaller than the experi-
mental ones by 0.1 Å. Nevertheless the three angles
(Ru-C1-C6, C1-C6-C7, and C6-C7-N) describing the
(17) (a) Coalter, J . N., III; Spivak, G. J .; Ge´rard, H.; Clot, E.;
Davidson, E. R.; Eisenstein, O.; Caulton, K. G. J . Am. Chem. Soc. 1998,
120, 9388. (b) Rodriguez, V.; Sabo-Etienne, S.; Chaudret, B.; Thoburn,
J .; Ulrich, S.; Limbach, H.-H.; Eckert, J .; Barthelat, J .-C.; Hussein,
K.; Marsden, C. J . Inorg. Chem., 1998, 37, 3475. (c) Delpech, F.; Sabo-
Etienne, S.; Daran, J .-C.; Chaudret, B.; Hussein, K.; Marsden, C. J .;
Barthelat, J .-C. J . Am. Chem. Soc. 1999, 121, 6668.

3986 Organometallics, Vol. 18, No. 20, 1999
Ayllon et al.
F igu r e 5. Calculated geometries and relative energies (kcal‚mol-1) for the products (Ba , Bb, and Bc) resulting from
protonation of A.
F igu r e 6. Mechanism for the formation of the chelate complex C from the protonated complex Bb (relative energies in
kcal‚mol-1).
are known to be fluxional, and rotation leading to
structures such as Bb-r ot (Figure 6) can be obtained
easily. The computed energy difference is only 0.95
kcal‚mol-1between Bb and Bb-r ot.
The chelate complex is then formed by substitution
of the H2 ligand by the amino group through a transition
state TS1 (Figure 6) lying at 23.2 kcal‚mol-1 above Bb.
In the transition state, the nitrogen atom is in a planar
configuration with the perpendicular p orbital pointing
toward the metal center. The P-Ru-P angle of 90.6°
is not very large, and this mechanism might also be
operative with a chelating diphosphine although cer-
F igu r e 7. Calculated geometry for the complex D with
the dihydrogen bond (distances in Å).
tainly less easy.
F or m a tion of a Dih yd r ogen Bon d . When the
protonation occurs at site c, the complex obtained Bc
is not the most stable conformation and the amino chain
obtained from relaxation time measurements. As a
prefers to adopt a geometry where it is pointing toward
consequence of the interaction between the two hydro-
the ruthenium center. The optimized geometry for this
gen atoms, the N-H bond is significantly elongated
(1.079 Å).
complex D is shown in Figure 7, and it is 7.1 kcal‚mol-1
more stable than Bc, in apparent contradiction with the
Exch a n ge of th e Tw o Hyd r ogen Atom s. The two
hydrogen atoms involved in the dihydrogen bond are
equivalent (δ ≈ 0 ppm) at room temperature on the
NMR spectra, but they decoalesce at 253 K to give two
peaks (δ ≈ -12 ppm and δ ≈ 12 ppm). The activation
energy for the fluxional process was calculated to be
around 11 kcal‚mol-1 depending of the acid used for the
protonation (HNEt3+ or HPBu3+). Usually it is assumed
X-ray crystal structure for 4 (see Figure 3). However,
the energy differences between the two structures are
very small, which may explain that the crystal packing
imposes the observed geometry. The main characteristic
of this structure is the presence of a dihydrogen bond
between the hydride and the hydrogen atom bound to
nitrogen. The distance between the two hydrogen nuclei
(1.545 Å) is in very good agreement with the value

Aminocyclopentadienyl Ru Hydride Complexes
Organometallics, Vol. 18, No. 20, 1999 3987
From this complex E the hydrogen atom (H) can be
transferred to the metal through a transition state that
we have not been able to locate. This step would most
probably be the rate-determining step and would ac-
count for the observed activation energy of ca. 11
kcal‚mol-1, which is expected to be significantly influ-
enced by the presence of a base (H2O, NEt3, PBu3). A
geometry corresponding to a hydrogen atom between N
and Ru was calculated to be less stable than D by only
15 kcal‚mol-1; however despite all our attempts we
failed to optimize this structure as a transition state.
From Bc it is now straightforward to consider the
inverse mechanism working again after rotation of the
Cp ring to take the other hydrogen atom from the metal
and to create a new dihydrogen bond where the two
hydrogen atoms have been permuted.
F igu r e 8. Calculated geometry for the transition state
between the chelate complex C and the complex D with
the dihydrogen bond (distances in Å).
for this kind of fluxional process that a dihydrogen
complex is formed as an intermediate. The rotation of
the H2 ligand in this complex accomplishes the exchange
of the two hydrogen atoms. In the present study we have
not been able to obtain complex 4 by reaction of 3 with
H2, nor have we been able to extrude H2 from 4 to obtain
3. The chemistry of the chelate complex is thus appar-
ently disconnected from the chemistry of the complex
with the dihydrogen bond. As a dihydrogen complex is
involved in the formation of the chelate complex (Figure
6), such species cannot be present along the path for
the formation of complex 4; otherwise it would be
possible to go easily from 3 to 4 and vice versa.
In fact we have optimized the transition state con-
necting the two complexes C and D, and it is a η1
dihydrogen complex (Figure 8) lying at 37.2 kcal‚mol-1
above D. The H-H bond distance of 0.775 Å is short
for a dihydrogen complex because the η1 binding mode
does not allow for an efficient back-donation from the
metal that would result in a longer H-H bond. The Ru-
H-H angle of 148.7° is also not common for a dihydro-
gen complex, and the η1 mode of coordination explains
the high energy of this structure. The synergy between
σ donation and π back-donation is not favored, and the
binding of H2 is consequently less strong. This explains
why it is difficult to go chemically from 3 to 4, although
the difference in energy between D and {C + H2} is only
9.4 kcal‚mol-1 (D is the most stable).
Although we have not characterized any transition
state corresponding to the exchange of the two hydrogen
atoms, we propose a possible mechanism for the ex-
change. This mechanism originates from the observation
that no trans dihydride with the amino chain pointing
toward one H could have been optimized. Despite all
our attempts, we always observed transfer of one of the
two hydrides to the nitrogen atom. The optimized
structure for this complex E is lying only 4.7 kcal‚mol-1
above D, and Figure 9 shows in a pictorial way the
proposed mechanism. The hydrogen atom initially on
the nitrogen (H) is transferred to the metal in several
steps.
First the dihydrogen bond is broken to give the
protonated complex Bc. The Cp ring can easily rotate
and the complex Bc-r ot is obtained, which lies 1.7
kcal‚mol-1 above Bc. As in the case of the compound D
with the dihydrogen bond, complex E with the amino
chain pointing toward the metal is more stable by 4.1
kcal‚mol-1 than the structure Bc-r ot with the chain
pointing away from Ru. This might be due to a stabiliz-
ing interaction between the hydrogen atom bound to N
and the ruthenium center.
Discu ssion
We describe in this study the synthesis of a new
ruthenium hydride complex containing an aminocyclo-
pentadienyl ligand (1). This compound contains three
basic sites, namely, the metal, the hydride, and the
amino group, each one able to react or to undergo
hydrogen bonding with an electrophile (in this case the
proton). It is possible to add two protons to it, which
leads to the dicationic dihydride derivative (2) contain-
ing a quaternary ammonium substituent on the cyclo-
pentadienyl rings. Such dicationic electrophilic hydride
complexes may display interesting catalytic properties
and may even be soluble in water. The reactivity of 2
was not studied for the present report but is now under
investigation. Nevertheless, 2 represents the final ther-
modynamic state of the system and gives us a reference
for the present study.
Addition of a strong acid such as HBF4‚Et2O to 1 leads
to 3 after evolution of dihydrogen and chelation of the
aminocyclopentadienyl ligand, whereas the use of softer
acids such as HNEt3BPh4 or HPBu3BPh4 allows the
isolation of 4, in which only the amine moiety of the
aminocyclopentadienyl ligand is protonated. The theo-
retical calculations at the DFT level demonstrate that
this is not due to a thermodynamic preference for the
chelate system. It has previously been shown that the
kinetic site of protonation of complexes such as CpRuH-
(PR3)2 is the hydride, hence leading to a dihydrogen
complex.18 In the present case, elimination of dihydro-
gen is driven by the formation of kinetically inert chelate
complex 3. In the presence of a weaker acid, or perhaps
for steric reasons because of the size of the ammonium
or phosphonium cation, the protonation occurs on the
nitrogen of the pendent amino group to give 4 and does
not lead to the protonation of the hydride. A process
exchanging the hydride with the ammonium proton and
involving an activation energy of 10-11 kcal‚mol-1
(according to the systems) has been evidenced by
variable-temperature NMR experiments. It is interest-
ing to note that 3 and 4 are both stable and that they
do not interconvert. 3 does not react with H2 in mild
conditions, whereas 4 does not lose H2 at room temper-
ature in solution.
DFT calculations show that the presence of a dihy-
drogen bond associating the hydride and the ammonium
(18) (a) J essop, P. G.; Morris, R. H. Coord. Chem. Rev. 1992, 121,
155. (b) Heinekey, D. M.; Oldham, W. J ., J r. Chem. Rev. 1993, 93, 913.

3988 Organometallics, Vol. 18, No. 20, 1999
Ayllon et al.
F igu r e 9. Proposed mechanism for the intramolecular exchange of the two hydrogen atoms involved in the dihydrogen
bond (relative energies in kcal‚mol-1).
proton in D (model for 4) is favored by 7.1 kcal‚mol-1
over the open structure Bc. This is in agreement with
the observation of a low T1 minimum value (130 ms,
400 MHz) observed for both the hydride and the
amonium proton in the case of the reaction of 1 +
HPBu3BPh4, which is consistent with the presence of
such a dihydrogen bond in solution and a short hydride/
proton separation (1.52 Å). However, the crystal struc-
ture evidences the presence in the solid state of a
conformation in which no interaction is present between
the proton on nitrogen and the hydride. Moreover, the
ammonium proton is not involved in any other hydrogen-
bonding interaction, in apparent contradiction with the
fact that, at least in the phosphonium case, the presence
of an additional base will play a role in the exchange
process between the hydride and the ammonium proton.
The latter assumption derives from the broadening of
the phosphine NMR signal during the exchange process
and from the fact that the activation energy of the
process is lower in the presence of PBu3. It is possible
that, as proposed in the theoretical section, the rate-
determining step is the transfer of proton from the
ammonium to ruthenium and back and that this process
is assisted by an additional base present in the vicinity.
the dihydrogen bond and rotation of the Cp ligand will
allow proton transfer on ruthenium to give a monocat-
ionic dihydride, not observed in this sytem, which is
rapidly deprotonated by the amino group before or after
Cp rotation. In the latter case the result of the process
is a proton/hydride exchange. In the related dppm
complexes studied by Lau and co-workers,13 this process
cannot take place and is indeed not observed, whereas
the presence of a dihydrogen bond is demonstrated and
the transfer of proton is shown to lead to dihydrogen
elimination and formation of a chelated species, in
agreement with our theoretical results.
In conclusion this study evidences the complexity of
a system designed to study one phenomenon, namely,
proton transfer within a dihydrogen bond. It demon-
strates the presence of multiple pathways for the
protons to be transferred to and from the metal, the
direct transfer within the dihydrogen bond being high
in energy. This result is probably general for proton-
transferring sytems in which pathways involving sol-
vents or additional functions in the vicinity of the metal
must be considered. Finally it shows the strong involv-
ment of ligands in the reactivity of these electrophilic
hydride complexes which could be used to transfer
protons during catalytic processes. Such possibilities are
presently being explored.
All these observations and calculations lead us to
propose a new mechanism for the exchange process
between the hydride and the ammonium proton which
does not involve proton transfer within the dihydrogen
bond (see Figure 9). Whereas in the solid state no
dihydrogen bond is present, it is likely that in solution
several fluxional pathways are present, one of which will
lead to the thermodynamically most stable species
which contains a dihydrogen bond between the hydride
and the amonium proton. The lifetime of this species is
presumably significant enough to have a strong influ-
ence on the T1 of 4. No proton transfer occurs in the
dihydrogen bond since this would lead occasionally to
dihydrogen elimination and formation of 3. Breaking of
Exp er im en ta l Section
All manipulations were carried out with standard high-
vacuum or dry argon atmosphere techniques. 1H, 13C, and 31P
NMR spectra were recorded on a Bruker AC 200, AM 250, or
AMX 400 spectrometers. The NMR chemical shifts are re-
ported in ppm, relative to Me4Si for 1H and 13C and relative
to 85% H3PO4 for 31P.
Com p u ta tion a l Deta ils. All the calculations were per-
formed with the Gaussian 94 set of programs19 at the B3PW91
level.20,21 Ruthenium was represented with the Hay-Wadt
relativistic core potential (ECP) for the 28 innermost electrons

Aminocyclopentadienyl Ru Hydride Complexes
Organometallics, Vol. 18, No. 20, 1999 3989
and its associated double-ú basis set.22 Phosphorus atoms were
also represented with Los Alamos ECPs and the associated
double-ú basis set23 augmented by polarization d functions.24
A 6-31G(d,p) basis set25,26 was used for the remaining atoms.
Full geometry optimizations within Cs symmetry have been
carried out within the framework of DFT(B3PW91).
Sta r tin g Ma ter ia ls. [RuH(OCOMe)(PPh3)3]14 and the ami-
nocyclopentadienyl ligand10 were prepared according to pub-
lished methods.
(HNEt3)(BP h 4). (HNEt3)(BPh4) was obtained from the
reaction between NEt3 (1.2 mL) and HCl (37% in water, 1.3
mL) in methanol (20 mL), in the presence of Na(BPh4) (5 g)
as colorless crystals that were washed twice with methanol
and dried in vacuo.Yield; 1.60 g (45%). 1H NMR (200 MHz,
CD2Cl2): 7.5-6.9 (m, 20 H, BPh4); 2.23 (q, J H-H ) 7 Hz, 6 H,
HN(CH2CH3)3), 0.87 (t, J H-H ) 7 Hz, 9 H, HN(CH2CH3)3) (the
signal of HN(CH2CH3)3 is averaged with the water at 1.55
ppm, broad). IR (Nujol) in cm-1: 3400, br, NH.
{HP (n -Bu )3}(BP h 4). {HP(n-Bu)3}(BPh4) was obtained from
the reaction between P(n-Bu)3 (3.3 mL) and concentrated HCl
(37% in water, 1.3 mL) in methanol (50 mL), in the presence
of Na(BPh4) (5 g) as colorless crystal that were washed twice
with methanol and dried in vacuo. Yield: 4.06 g (53%). 1H
NMR (200 MHz, CD2Cl2): 7.50-6.90 (m, 20 H, BPh4); 3.87 (d
of septuplets, J P-H ) 488 Hz, J H-H ) 5 Hz, HP(n-Bu)3); 1.6-
0.90 (m, 27 H, HP(n-Bu)3). 31P NMR (81.0 MHz, CD2Cl2): 11.6
(d, J P-H ) 488 Hz). 31P{1H} NMR (81.0 MHz, CD2Cl2): 11.6
(s). IR (Nujol) in cm-1: 2408 (P-H).
P r epar ation of [(CpCH2CH2NHMe2)Ru H2(P P h 3)2](P F6)2
(2). [(CpCH2CH2NMe2)RuH(PPh3)2] (50.0 mg, 0.065 mmol) is
suspended in methanol (12 mL). Addition of an excess of HPF6
(60% in water, 0.05 mL, 0.24 mmol) leads to complete dissolu-
tion of the starting material and, after 5 min, to the appear-
ance of a new white microcrystalline solid. After 1 h, the
precipitate is filtered, washed with methanol (3 × 1 mL), and
1
dried in vacuo. Yield: 40 mg (69%). H NMR (200 MHz, CD2-
Cl2): 7.60-7.20 (m, 30 H, PPh3); 6.6 (s, 1 H, C5H4CH2-
CH2NHMe2), 5.08 (t, 2 Hz, 2 H, C5H4CH2CH2NMe2), 4.81 (t, 2
Hz, 2 H, C5H4CH2CH2NMe2); 2.99 (m, 2H C5H4CH2CH2-
NHMe2); 2.76 (d, 5 Hz, 6 H, C5H4CH2CH2NHMe2); 1.83 (m,
2H C5H4CH2CH2NHMe2); -7.37 (t, J P-H ) 24 Hz, 2 H, RuH ).
13C NMR (50.32 MHz, CD2Cl2): 134.42 (m), 133.52 (d, 162 Hz),
131.41 (d, 162 Hz), 129.11 (d, 162 Hz) (all for PPh3); 106.53 (s,
quaternary C of C5H4CH2CH2NHMe2); 91.46 (d, 179 Hz), 90.26
(dd, 182 and 4.5 Hz, C-H of C5H4CH2CH2NHMe2); 58.55 (t,
140 Hz, C5H4CH2CH2NHMe2), 43.99 (q, 144 Hz, C5H4CH2CH2-
NHMe2) and 22.73 (t, 131 Hz, C5H4CH2CH2NHMe2). 31P{1H}
NMR (81.0 MHz, CD2Cl2): 60.81 (s). IR (Nujol) in cm-1: 1994
(Ru-H). Anal. Calcd for C45H47F12NP4Ru: C, 51.24; H, 4.49;
N, 1.33. Found: C, 51.18; H, 4.41; N, 1.28.
P r ep a r a tion of [(Cp CH2CH2NMe2)Ru (P P h 3)2]BF 4 (3).
HBF4‚OEt2 (9.4 µL of a 85% solution in diethyl ether, 52.4 ×
10-3 mmol) is added to a vigorously stirred solution of [(CpCH2-
CH2NMe2)RuH(PPh3)2] (1) (40 mg, 52.4 × 10-3 mmol) in
diethyl ether (15 mL), leading to the immediate appearance
of a yellow precipitate. The precipitate is filtered, washed twice
1
with diethyl ether, and dried under vacuum. Yield: 95%. H
P r ep a r a t ion of [(Cp CH 2CH 2NMe2)R u H (P P h 3)2] (1).
Methanol (35 mL) is added to [RuH(OCOMe)(PPh3)3] (592 mg;
0.626 mmol) and Na(CpCH2CH2NMe2) (100 mg; 0.626 mmol),
and the mixture is heated under reflux for 45 min. During
this time the solution clears and a small amount of a yellow
precipitate of [RuH2(CO)(PPh3)3] is formed. The yellow suspen-
sion is allowed to cool and filtered. The solution is then
evaporated to dryness. The residue is extracted with diethyl
ether (ca. 20 mL), hexane is added (ca. 15 mL), and the solution
is evaporated to approximately half-volume under reduced
pressure and allowed to stand at room temperature until
crystallization. The supernatant is filtered off, and the yellow
crystals are washed twice with pentane and dried in vacuo.
NMR (200 MHz, CD2Cl2): 7.40-7.10 (m, 30 H, PPh3); 4.65 (t,
1.6 Hz, 2 H, C5H4CH2CH2NMe2), 3.57 (t, 1.7 Hz, 2 H, C5H4-
CH2CH2NMe2); 3.26 (t, 6.7 Hz, 2H), 2.50 (t, 6.4 Hz, 2 H,
C5H4CH2CH2NMe2); 2.73 (s, 6 H, C5H4CH2CH2NMe2). 13C NMR
(50.32 MHz, CD2Cl2): 139.50 (m), 133.68 (d, 161 Hz), 129.18
(d, 162 Hz), 127.74 (d, 162 Hz) (all for PPh3); 92.95 (s,
quaternary C of C5H4CH2CH2NMe2); 82.24 (dd, 175 and 6 Hz),
79.96 (d, 175 Hz, C-H of C5H4CH2CH2NMe2); 57.33 (t, 149
Hz, C5H4CH2CH2NMe2), 43.11 (q, 142 Hz, C5H4CH2CH2NMe2)
and 22.01 (t, 126 Hz, C5H4CH2CH2NMe2). 31P{1H} NMR (81.0
MHz, CD2Cl2): 64.62 (s). Anal. Calcd for C45H44BF4NP2Ru: C,
63.69; H, 5.23; N, 1.65. Found: C, 63.14; H, 5.00; N, 1.66.
P r ep a r a tion of 1 + (HNEt3BP h 4). [(CpCH2CH2NMe2)-
RuH(PPh3)2] (1) (8.0 mg; 10.5 × 10-3 mmol) and (HNEt3)(BPh4)
(4.4 mg; 10.5 × 10-3 mmol) were loaded in a NMR tube, after
which CD2Cl2 was added. After 15 min NMR spectra were
recorded at different temperatures. 1H NMR (200 MHz, 293
K, CD2Cl2): 7.50-7.00 (m, 50 H, PPh3, BPh4); 4.57 (t, 1.7 Hz,
2H, C5H4CH2CH2NMe2), 3.53 (t, 1.7 Hz, 2 H, C5H4CH2CH2-
NMe2); 2.35 (m, 2 H), 2.18 (m, 2 H, C5H4-CH2CH2NMe2); 2.06
(s, 6 H, C5H4CH2CH2NMe2); 2.42 (q, 7 Hz, 6 H, N(CH2CH3)3),
0.96 (t, 7 Hz, 9 H, N(CH2CH3)3); -0.95 (very br, 2 H, RuH and
HNMe2CH2CH2C5H4 coalescence). 1H NMR (400 MHz, CD2Cl2),
293 K, 7.50-7.00 (m, 50 H, PPh3, BPh4); 4.55 (t, 1.7 Hz, 2H),
3.53 (t, 1.7 Hz, 2 H, C5H4CH2CH2NMe2); 2.36 (m, 2 H), 2.17
(m, 2 H, C5H4CH2CH2NMe2); 2.06 (s, 6 H, C5H4CH2CH2NMe2);
2.44 (q, 7 Hz, 6 H, N(CH2CH3)3), 0.97 (t, 7 Hz, 9 H, N(CH2CH3)3);
no signal for RuH or HNMe2CH2CH2C5H4; 273 K, no signal
1
Yield: 344 mg (72%). H NMR (200 MHz, CD2Cl2): 7.50-7.00
(m, 30 H, PPh3); 4.27 (t, 1.7 Hz, 2H, C5H4CH2CH2NMe2), 3.82
(t, 1.7 Hz, 2 H, C5H4CH2CH2NMe2); 2.30 and 2.02 (m, 4 H,
C5H4CH2CH2NMe2); 2.10 (s, 6 H, C5H4CH2CH2NMe2); -11.56
(t, J P-H ) 34.0 Hz, 1 H, RuH ). 31P{1H} NMR (81.0 MHz, CD2-
Cl2): 70.07 (s). 13C NMR (50.32 MHz, C6D6): 141.0 (m), 133.74
(d, 161 Hz), 127.77 (d, 160 Hz), 126.79 (d, 159 Hz) (all for
PPh3); 102.74 (s, quaternary C of C5H4CH2CH2NMe2); 81.77
(dd, 174 and 6 Hz), 80.43 (d, 175 Hz, C-H of C5H4CH2CH2-
NMe2); 63.20 (t, 136 Hz, C5H4-CH2CH2NMe2), 45.18 (q, 129
Hz, C5H4CH2CH2NMe2) and 27.29 (t, 127 Hz, C5H4CH2CH2-
NMe2). IR (Nujol) in cm-1: 1941 (br, Ru-H). Anal. Calcd for
C
45H45NP2Ru: C, 70.85; H, 5.95; N, 1.84. Found: C, 70.78; H,
6.01: N, 1.78.
(19) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
J ohnson, B. G.; Robb, M. A.; Cheeseman, J . R.; Kerth, T.; Petersson,
G. A.; Montgomery, J . A.; Raghavachari, K.; Al-Laham, M. A.;
Zakrzewski, V. G.; Ortiz, J . V.; Foresman, J . B.; Peng, C. Y.; Ayala, P.
Y.; Chen, W.; Wong, M. W.; Andres, J . L.; Replogle, E. S.; Gomperts,
R.; Martin, R. L.; Fox, D. J .; Binkley, J . S.; Defrees, D. J .; Baker, J .;
Stewart, J . P.; Head-Gordon, M.; Gonzalez, C.; Pople, J . A. Gaussian
94, Revision B3; Gaussian Inc.: Pittsburgh, PA, 1995.
(20) Becke, A. D. J . Chem. Phys. 1993, 98, 5648.
(21) Perdew, J . P.; Wang, Y. Phys. Rev. B 1992, 45, 13244.
(22) Hay, P. J .; Wadt, W. R. J . Chem. Phys. 1985, 82, 299.
(23) Wadt, W. R.; Hay, P. J . J . Chem. Phys. 1985, 82, 284.
(24) Ho¨lwarth, A.; Bo¨hme, M.; Dapprich, S.; Ehlers, A. W.; Gobbi,
A.; J onas, V.; Ko¨hler, K.; Stegmann, R.; Veldkamp, A.; Frenking, G.
Chem. Phys. Lett. 1993, 208, 237.
for RuH or HNMe2CH2CH2C5H4; 253 K, -11.6 (very br, T1
)
170 ms, RuH); 4.9 (very br, HNMe2CH2CH2C5H4; 233 K, -11.7
(broad, T1 ) 140 ms, RuH); 4.9 (very br, HNMe2CH2CH2C5H4;
213 K, -11.72 (t, J P-H ) 30 Hz, T1 ) 230 ms, RuH); 6.0 (very
br, HNMe2CH2CH2C5H4; 193 K, -11.68 (t, J P-H ) 30 Hz, T1
) 780 ms, RuH); 11.9 (br, HNMe2CH2CH2C5H4.31P{1H} NMR
(162.0 MHz, CD2Cl2): 293 K, 66.28 (s); 253 K, 67.9 (s); 233 K,
68.8 (s); 213 K, 70.0; 193 K, 70.43 (s).
P r ep a r a tion of 1 + (HP Bu 3BP h 4). [(CpCH2CH2NMe2)-
RuH(PPh3)2] (1) (10.0 mg; 13.1 × 10-3 mmol) and {HP(n-Bu)3}-
(BPh4) (6.8 mg; 13.0 × 10-3 mmol) were loaded in a NMR tube,
after which CD2Cl2 was added. After 15 min, NMR spectra
(25) Hehre, W. J .; Ditchfield, R.; Pople, J . A. J . Chem. Phys. 1972,
56, 2257.
1
were recorded at different temperatures. H NMR (400 MHz,
(26) Harihan, P. C.; Pople, J . A. Theor. Chim. Acta 1973, 28, 213.
CD2Cl2): 293 K, 7.5-6.9 (m, PPh3, BPh4); 4.72 (br, 2 H, C5H4-

3990 Organometallics, Vol. 18, No. 20, 1999
Ayllon et al.
CH2CH2NMe2), 3.33 (br, 2 H, C5H4CH2CH2NMe2); 2.48 (m),
2.31 (m, C5H4CH2CH2NMe2); 2.09 (s, C5H4CH2CH2NMe2); 1.37
(m), 0.94 (m, P(n-Bu)3), -0.3 (very br, NH and Ru-H,
coalescence); 263 K, no signal for NH or Ru-H; 233 K, 11.78
(br, N-H, T1 ) 130 ms), -12.1 (br, Ru-H, T1) 130 ms); 213
K, 11.76 (br, N-H, T1 ) 180 ms); -12.11 (t, J P-H ) 32 Hz,
Ru-H, T1 ) 180 ms); 193 K, 11.72 (br, N-H, T1 ) 330 ms);
-12.12 (t, J P-H ) 32 Hz, Ru-H, T1 ) 430 ms). 31P{1H} NMR
(162.0 MHz, CD2Cl2): 293 K, 63.37 (s), no signal observed for
P(n-Bu)3); 263 K, 63.2 (s, PPh3), -29.0 (br, P(n-Bu)3); 233 K,
63.2 (s, PPh3), -29.4 (br, P(n-Bu)3); 213 K, 63.2 (s, PPh3); -29.7
(s, P(n-Bu)3); 193 K, 63.0 (s, PPh3); -29.8 (s, P(n-Bu)3).
P r ep a r a tion of [(Cp -NH)Ru H(P P h 3)2][BP h 4]‚H2O (4‚
H2O). (Cp-NH)RuH(PPh3)2 (30 mg, 0.054 mmol) and HNEt3-
BPh4 (17 mg, 0.041 mmol) were mixed in a Shlenck tube, and
CH2Cl2 (5 mL) was added. To the resulting yellow solution was
added MeOH (5 mL). The volume of the solution was reduced
to approximately 5 mL and left to stand at room temperature,
thereby inducing precipitation of [(Cp-NH)RuH(PPh3)2][BPh4]‚
H2O as yellow crystals (15 mg, 32%). IR (KBr) in cm-1: 3590,
of refinement and were given isotropic thermal parameters
20% higher than those of the carbon atoms to which they were
attached. The hydride atoms labeled H(1) in each compound
and the hydrogen H(2) connected to the nitrogen atom N(1)
in 4 were isotropically refined. For both structures all non
hydrogen atoms were anisotropically refined, except C(7), C(8),
C(9), and N(1) of 1, for which a disorder has been found and
which have been isotropically refined on two sites with a
occupancy ratio equal to 0.5. The phenyl rings of the counter-
anion present in the structure of 4 have been idealized as rigid
groups due to the low observations/parameters ratio. The
procedures of least-squares refinement were carried out by
minimizing the function ∑w(|Fo| - |Fc|)2, where Fo and Fc are
respectively the observed and calculated structure factors. A
weighting scheme was used in the last refinement cycles,
where weights are calculated from the following expression:
w ) [weight][1 - (∆(F)/6σ(F)]2.28 Models reached convergence
with the formulas R ) ∑(||Fo| - |Fc||)/∑|Fo|, and Rw ) [∑w(||Fo|
- |Fc||)2/∑w(|Fo|)2]1/2
.
The calculations were performed with a CRYSTALS pro-
gram29 running on a PC, and the drawing of the molecules
was realized with the aid of CAMERON.30 The atomic scat-
tering factors were taken from International Tables for X-ray
Crystallography.31
Further details on the crystal structure investigation are
available on request from the Director of the Cambridge
Crystallographic Data Centre, 12 Union Road, GB-Cambridge
CB21EZ, U.K., on quoting the full journal citation.
3050, 1977, 1605, 1578, 1478, 1434. Anal. Calcd for C69H68
-
BNOP2Ru: C, 75.29; H, 6.18; N, 1.27. Found: C, 75.01; H, 6.09;
N, 1.33.
X-r a y An a lysis. Data were collected on a Stoe Imaging
Plate Diffraction System (IPDS) equipped with an Oxford
Cryosystems Cooler Device.
The structures were solved by direct methods (SIR92)27 and
refined by least-squares procedures on Fobs. All hydrogen atoms
were located on difference Fourier maps, but they were
introduced in calculation in idealized positions (d(C-H) ) 0.96
Å). Their atomic coordinates were recalculated after each cycle
Ack n ow led gm en t. The authors thank CNRS for
support and the European Union within the framework
of the Human Capital & Mobility Network on “Localiza-
tion and Transfer of Hydrogen” for a fellowship to J .A.A.
and S.F.S.
(27) Altomare, A.; Cascarano, G.; Giacovazzo, G.; Guargliardi, A.;
Burla, M. C.; Polidori, G.; Camalli, M. J . Appl. Crystallogr. 1994, 27,
435.
(28) Carruthers. J . R.; Watkin, D. J . Acta Crystallogr. 1979, A35,
698.
(29) Watkin, D. J .; Prout, C. K.; Carruthers, R. J .; Betteridge, P.
CRYSTALS Issue 10; Chemical Crystallography Laboratory: Oxford,
U.K., 1996.
(30) Watkin, D. J .; Prout, C. K.; Pearce, L. J . CAMERON; Chemical
Crystallography Laboratory: University of Oxford, Oxford, U.K., 1996.
(31) International Tables for X-ray Crystallography; Kynoch Press:
Birmingham, England, 1974; Vol IV.
Su p p or tin g In for m a tion Ava ila ble: X-ray structural
information for 1 and 4. An X-ray crystallographic file in CIF
format is available. Cartesian coordinates of the optimized
structure for A (in Å). This material is available free of charge
via the Internet at http://pubs.acs.org.
OM990301L