Solvent control in the protonation of [Cp*Mo(dppe)H3] by CF3COOH

FULL PAPER

DOI: 10.1002/ejic.200700021

Solvent Control in the Protonation of [Cp*Mo(dppe)H₃] by CF₃COOH

Pavel A. Dub,^[a,b]Miguel Baya,^[a]Jennifer Houghton,^[a]Natalia V. Belkova,^[b]

Jean-Claude Daran,^[a]Rinaldo Poli,*^[a]Lina M. Epstein,^[b]and Elena S. Shubina*^[b]

Keywords: Molybdenum / Hydrido ligand / Trifluoroacetato ligand / Proton transfer / Dihydrogen bonding

The outcome of the reaction between [Cp*Mo(dppe)H₃]

[dppe = 1,2-bis(diphenylphosphanyl)ethane] and trifluoro-

acetic acid (TFA) is highly dependent on the solvent and the

TFA/Mo ratio. The dihydride compound [Cp*Mo(dppe)-

H₂(O₂CCF₃)] is obtained selectively when the reaction is car-

ried out in aromatic hydrocarbons (benzene, toluene) when

using less than one equivalent of TFA. The dihydride is also

the end product when THF or MeCN are used as solvent,

independent of the TFA/Mo ratio. In benzene/toluene the

the dihydrogen-bonded adduct, [Cp*Mo(dppe)H₃]···HO₂-

CCF₃, the ion-paired proton-transfer product, [Cp*Mo-

(dppe)H₄]⁺···^–O₂CCF₃, and the separated ions, followed by a

slower irreversible transformation to the final dihydride prod-

uct with H₂evolution. The activation parameters of the H₂

evolution and M–OR product formation were determined.

Excess TFA in low-polarity solvents stabilizes the separated

charged species by forming the homoconjugate anion

[CF₃COO(CF₃COOH)_n]^–. The effect of the solvent on the

use of excess acid has a profound effect, resulting in the for- course of the reaction can be interpreted in terms of the dif-

mation of the tetrahydrido complex [Cp*Mo(dppe)H₄]⁺,

which did not further evolve into the dihydrido product.

Monitoring of the reaction by NMR and IR spectroscopy un-

der different conditions (solvent, temperature, TFA/Mo ratio)

reveals the rapid establishment of an equilibrium between

ferent polarity, H-bonding ability, and coordinating power of

the various solvent molecules.

Germany, 2007)

Introduction

It is now well established that a common proton-transfer

pathway from a proton donor HX to a transition-metal hy-

dride complex involves the initial attack of HX on the hy-

dride ligand with formation of a dihydrogen-bonded ad-

duct, MH···HX, followed by proton transfer and pro-

duction of a nonclassical dihydride (dihydrogen complex)

product, M(η²-H₂)⁺. The establishment of a hydrogen-

bonded ion pair between the nonclassical cation and its

counteranion increases the stability of the M(η²-H₂)⁺com-

plex and retards its transformation into a classical polyhy-

dride or organyloxo (X = OR) species (Scheme 1).^[1]

Scheme 1.

The formation of M(η²-H₂)⁺···X^–ion pairs (III, X^–

=

pair stability increases with the proton-accepting ability of

the anion.^[3]

CF₃COO^–, ArO^–, R^FO^–), stabilized by hydrogen bonds be-

tween the dihydrogen ligand and counteranions, was de-

Proton transfer may also be assisted by a second mole-

cule of the proton donor HX, depending on the acid

strength. In this case, the conjugate base X^–in species III–

VI may be present in the form of the homoconjugate anion,

XHX^–. We have recently shown that the irreversible isomer-

ization of [Cp*M(η²-H₂)(dppe)]⁺complexes (M = Fe,^[5]

Ru^[6]) into classical trans-dihydride species occurs upon

dissociation of the M(η²-H₂)⁺···XHX^–ion pair (i.e., III to

IV). This is indicated by the difference in temperature at

tected for [CpRuH(CO)(PCy₃)],^[2,3]PP₃MH₂{PP₃

=

P(CH₂CH₂PPh₂)₃},^[4]and [Cp*MH(dppe)] (M = Fe,^[5]

Ru^[6]) complexes by IR and UV/Vis spectroscopy. The ion-

[a] Laboratoire de Chimie de Coordination, UPR CNRS 8241 liée

par convention à l’Université Paul Sabatier et à l’Institut

National Polytechnique de Toulouse,

205 Route de Narbonne, 31077 Toulouse Cedex, France

Fax: +33-561553003

E-mail: poli@lcc-toulouse.fr

+

which the M(η²-H₂)⁺Ǟ M(H)₂isomerization process can

occur depending on the counteranion; higher temperatures

are necessary to induce isomerization in the presence of

more basic anions (greater ion-pair stability). Additionally,

[b] Nesmeyanov Institute of Organoelement Compounds, Russian

Academy of Sciences,

Vavilov Street 28, 119991 Moscow, Russia

Supporting information for this article is available on the

WWW under http://ww w .eurjic.org or from the author.

Eur. J. Inorg. Chem. 2007, 2813–2826

2813

R. Poli, E. S. Shubina et al.

FULL PAPER

the isomerization rate constant does not depend on the ticular H-bonding) ability and polarity: benzene, toluene,

anion nature.^[7,8]In some cases, the nonclassical to classical THF, CH₂Cl₂, and MeCN. These reactions led to the for-

rearrangement (III to V, or IV to VI) is a reversible pro- mation of a new hydride product with a coordinated trifluo-

cess.^[9]

roacetate anion. We will show how subtle changes in the

In many cases, however, basic anions (X^–= RO^–, protonation medium may drastically affect the reactivity of

RCOO^–) tend to displace H₂from the transition-metal co- [Cp*Mo(dppe)H₃] and the stability of its protonation prod-

ordination sphere, yielding organyloxo products (i.e., VII). uct.

This is the case for the protonation of [(triphos)ReH-

(CO)₂],^[10][Re(CO)H₂(NO)(PR₃)₂],^[11]and the previously

mentioned [CpRuH(CO)(PCy₃)],^[2]for which corresponding

Results

–

M(η²-H₂)⁺X^–complexes could be isolated as BF₄salts.

These evolved into MX species in the presence of fluori-

Since the results of previous investigations into the pro-

nated alcohols or carboxylic acids. This reaction proceeds tonation of [Cp*Mo(dppe)H₃]^[13–15]serve as a basis for the

intramolecularly without the hydrogen-bonded ion-pair new findings described in this paper, we summarize here

dissociation.^[2]Most of the M–OR complexes were only the salient points that emerge from those studies. (i) The

characterized in solution; two examples of isolated com- trihydride complex has a high basicity factor (E_j

=

pounds are [ReH(OC₆H₄NO₂)(CO)(NO)(PMe₃)₂]^[11]and 1.42Ϯ0.02), placing it in the category of the most hydridic

[(triphos)Re(CO)₂(OCOCH₂Cl)].^[10]In addition to the complexes known. (ii) As mentioned in the Introduction,

counteranion nature, temperature and solvent polarity the interaction with HBF₄in noncoordinating or weakly

could be used to control the stability of the M(η²-H₂)⁺spe- coordinating solvents (diethyl ether, THF) leads directly to

cies and the proton-transfer equilibrium position. For ex- the tetrahydride complex [Cp*Mo(dppe)H₄]⁺, without the

ample, the substantial increase in the dielectric constant of detection of a nonclassical intermediate, although its pres-

the Freon mixture (CDCl₂F/CDF₃, 2:1) at low tempera- ence is suggested by the theoretical calculations. (iii)

–

tures was reported to assist the proton transfer to [Cp*Mo(dppe)H₄]⁺BF₄is unstable and rapidly loses dihy-

[Cp*RuH₃(PCy₃)], yielding [Cp*Ru(H)₂(η²-H₂)(PCy₃)]⁺- drogen; in MeCN, this process leads to the formation of the

OR^–.^[12]However, the influence of the medium on the H₂solvent-stabilized complex [Cp*Mo(dppe)H₂(MeCN)]⁺. (iv)

release process was not studied in detail.

The interaction with weaker proton donors [the fluorinated

In contrast to all of the described systems, the alcohols CH₂FCH₂OH, CF₃CH₂OH, and (CF₃)₂CHOH] in

protonation of the trihydridomolybdenum complex low-polarity solvents (e.g., THF, CH₂Cl₂) at low tempera-

[Cp*Mo(dppe)H₃] gave, via the intermediate formation of tures leads to the observation of intermediate hydrogen-

a dihydrogen-bonded complex, the cationic classical tetra- bonded adducts. Infrared spectral analysis, in combination

hydride complex [Cp*Mo(dppe)H₄]⁺(V/VI), without the with theoretical calculations, indicates that one of the hy-

detection of a nonclassical isomer, [Cp*Mo(dppe)(η²-H₂)- dride ligands of [Cp*Mo(dppe)H₃] takes part in hydrogen

H₂]⁺(III/IV). Theoretical calculations, however, located the bonding. (v) Proton transfer from the dihydrogen-bonded

latter complex on the proton-transfer potential energy sur- adduct yields, without the observable intervention of a sec-

face, with a very low energy barrier for the intramolecular ond HA molecule in the rate-determining step, the tetrahy-

rearrangement.^[13]Facile access to a nonclassical isomer is dride product as a 1:1 ion-pair stabilized by a hydrogen

evidenced by the evolution of H₂from [Cp*Mo(dppe)H₄]⁺bond between the cation and the anion, [Cp*Mo(dppe)-

at ambiet temperature. We now report new results on the H₄]⁺···A^–, as established for HA = p-nitrophenol and

protonation of [Cp*Mo(dppe)H₃] using trifluoroacetic acid CF₃CH₂OH. The proton-transfer reaction is equilibrated

(TFA) in various solvents of different coordinating (in par- for these weaker proton donors. (vi) On a longer timescale,

Scheme 2.

2814

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 2813–2826

Solvent Control of Hydride Protonation

FULL PAPER

the tetrahydride product [Cp*Mo(dppe)H₄]⁺, obtained a simulation using the rate constant obtained from the ³¹P

using HA = CF₃CH₂OH, evolves at ambient temperature NMR spectrum and a reasonable guess for the J_HPvalues

by release of two H₂molecules to form the hydride-free gave a reasonable fit above coalescence, suggesting that the

product [Cp*Mo(dppe)(OCH₂CF₃)], probably stabilized by same mechanism is responsible for both H and P exchange

coordination of a second alcohol molecule, [Cp*Mo(dppe)- processes. A possibility for this mechanism is illustrated in

(OCH₂CF₃)(CF₃CH₂OH)]. These observations are summa- Scheme 3. The T₁value for the hydride resonance remains

rized in Scheme 2.

rather high throughout the temperature range (see Support-

ing Information), characteristic of a classical dihydride,

with a minimum of 510 ms before decoalescence. For com-

parison, the T_1minof complex [Cp*Mo(dppe)H₄]⁺is 174 ms

Protonation in Benzene/Toluene: Formation of Compound

[Cp*Mo(dppe)H₂(O₂CCF₃)]

–

for the BF₄salt and 191 ms when obtained by proton

transfer from TFE (these measurements were carried out at

The room-temperature reaction between [Cp*Mo(dppe)-

H₃] and strictly 1 equiv. TFA in toluene led selectively to a

new dihydride complex, [Cp*Mo(dppe)H₂(O₂CCF₃)], as

seen in Equation (1). The parallel formation of free H₂was

400 MHz).^[13]

1

confirmed by H NMR spectroscopic monitoring (charac-

teristic resonance at δ = 4.58 ppm). The new hydride com-

pound was isolated and fully characterized. It is quite stable

in aromatic solvents, decomposing very slowly with genera-

tion of free dppe (identified by its ³¹P resonance at

–12.9 ppm) and other unidentified products that do not

contain hydride ligands. Ca. 13% of the product decom-

posed after 22 hours at room temperature.

[Cp*Mo(dppe)H₃] + TFA Ǟ [Cp*Mo(dppe)H₂(O₂CCF₃)] + H₂

(1)

The compound is characterized by a hydride resonance

centered at –5.09 ppm, with a relative intensity correspond-

ing to two protons, only slightly displaced relative to the

hydride resonance of [Cp*Mo(dppe)H₃] at –5.16 ppm. The

³¹P{¹H} NMR spectrum shows a resonance at 76.9 ppm,

which becomes a triplet when selectively decoupled from

the dppe proton resonance, confirming the presence of two

hydride ligands. The shape of the hydride resonance sug-

gests the occurrence of a dynamic process. A variable-tem-

perature NMR investigation gave further information about

this process and also provided longitudinal relaxation times

for the hydride resonances. The shape of the ¹H resonances

in the hydride region as a function of temperature is shown

in Figure 1 (a). Cooling resulted in broadening and eventual

decoalescence (225 K), yielding two broad resonances in a

1:1 ratio at –4.19 and –6.24 ppm. The more upfield reso-

nance starts to resolve into a doublet at the lowest tempera-

ture, indicating coupling (J ≈ 72 Hz) to another I = 1/2

nucleus, probably one of the two phosphorus donor atoms

that therefore appear to be inequivalent. This inequivalence

was demonstrated by the variable-temperature ³¹P{¹H}

Figure 1.

Variable-temperature

spectra

for

compound

[Cp*Mo(dppe)H₂(O₂CCF₃)] in C₆D₅CD₃: (a) ¹H NMR

(500 MHz); (b) ³¹P{¹H} NMR (202 MHz).

Scheme 3.

The chemical composition of the product was confirmed

NMR investigation (see part b of Figure 1); decoalescence by single-crystal X-ray analysis. The coordination geometry,

was observed at T Ͻ 233 K to yield two resonances in an shown in Figure 2, corresponds to a rather severely dis-

approximate ratio of 1:1. No P–P coupling can be discerned torted octahedron if the Cp* ligand is considered to occupy

from the spectra. The ³¹P NMR data were used in a linesh- a single coordination position at the ring centroid. This dis-

ape analysis, from which the activation parameters of ∆H^‡tortion is caused by steric repulsion between the Cp* and

= (6.4Ϯ0.4) kcalmol^–1and ∆S^‡= (10Ϯ2) calK^–1mol^–1dppe ligands and by the small size of the two hydride li-

were calculated using an Eyring analysis (see electronic gands, which force the pseudoaxial (trans to the Cp* ligand)

1

Supporting Information). The lineshape analysis of the H phosphorus donor to move up into the wedge of the two

resonance was complicated by the lack of knowledge of J_HPMo–H bonds. The CNT–Mo1–P1 angle [122.648(13)°] is

data due to broadness in the low-temperature spectra, but greater than the CNT–Mo1–O1, CNT–Mo1–H1, and

Eur. J. Inorg. Chem. 2007, 2813–2826

www.eurjic.org

2815

R. Poli, E. S. Shubina et al.

FULL PAPER

CNT–Mo1–H2 angles [119.74(4)°, 98.5(10)°, and 96.9(9)°, Table 1. Selected bond lengths [Å] and angles [°] for compound

[Cp*Mo(dppe)H₂(OCOCF₃)].

respectively],

whereas

the

CNT–Mo1–P2

angle

[144.369(14)°] is much smaller than 180°. A similar distor-

tion was observed in the structure of the related

[Cp*Mo(dppe)H(MeCN)₂]²⁺complex.^[14]The other notable

feature is the monodentate coordination of the trifluo-

roacetate ligand, which occupies a pseudoequatorial posi-

tion cis to the P1 and H1 donors and trans to H2. The

Mo1–O1 bond length [2.2008(15) Å] is unexpectedly longer

than those found in other Mo monodentate carboxylate de-

rivatives in the same or lower oxidation state, for example,

2.167(4) Å found in [Cp*Mo^II(CO)(PMe₃)₂(O₂CMe)],^[16]

2.102(3) and 2.092(3) Å found in [CpMo^III(η⁴-

C₄H₆)(O₂CCF₃)₂],^[17]and 2.113(4) and 2.102(4) Å found in

[Cp₂Mo^IV(O₂CPh)₂].^[18]Another interesting feature is the

similar C8–O1 [1.249(3) Å] and C8–O2 [1.221(3) Å] dis-

tances (bond difference ∆ = 0.028 Å). These are not quite

as long or short, respectively, as expected for localized sin-

gle and double bonds. For instance, the recently determined

structure of Ph₃CCH₂COOH shows clearly defined C–O

[1.304(3) Å] and C=O [1.203(3) Å] bonds (∆ = 0.101 Å)

with no obvious sign of hydrogen atom disorder.^[19]The

monodentate carboxylate C–O distances in the mentioned

examples show larger differences between C=O and C–O

bond lengths, although not quite as large as those found

in free carboxylic acids, namely 1.277(8) Å compared with

Mo1–CNT

Mo1–P1

Mo1–H1

P1–C111

P1–C121

P1–C6

1.96462(17)

2.5206(5)

1.60(3)

1.833(2)

1.834(2)

1.844(2)

1.526(3)

1.249(3)

Mo1–O1

Mo1–P2

Mo1–H2

P2–C211

P2–C221

P2–C7

2.2008(15)

2.3962(5)

1.65(3)

1.8459(19)

1.828(2)

1.849(2)

1.547(3)

1.221(3)

C6–C7

C8–O1

C8–C9

C8–O2

CNT–Mo1–O1 119.74(4)

P2–Mo1–P1 81.360(17)

CNT–Mo1–P1 122.648(13)

O1–Mo1–P1 76.82(4)

CNT–Mo1–H1 98.5(10)

O1–Mo1–H1

P2–Mo1–H1

P1–Mo1–H1

H2–Mo1–H1

CNT–Mo1–P2

O1–Mo1–P2

CNT–Mo1–H2

O1–Mo1–H2

P2–Mo1–H2

P1–Mo1–H2

105.6(14)

144.369(14)

89.40(4)

96.9(9)

83.0(10)

64.6(9)

139.8(10)

61.6(10)

72.0(11)

140.5(9)

Compound [Cp*Mo(dppe)H₂(OCOCF₃)] exhibits two

intense overlapping ν^as

bands in the IR spectrum; the

OCO

frequencies and relative intensities of which are solvent de-

pendent (see Table 2). As the solvent polarity increases,

A_low/A_highdecreases and ∆ν = (ν_high– ν_low) increases. The

same behavior was observed previously for [CpM(CO)₂-

(OCOR)] complexes (M = Fe, Ru)^[20]and was explained by

the presence of a Fermi resonance between the fundamental

ν^as_OCOstretching vibration and an overtone or combination

band of very close frequency. The larger intensity ratio ob-

served in CH₂Cl₂could be explained by the different nature

of the solute–solvent interaction, since this solvent could

act as a proton donor and form hydrogen bonds with the

carboxyl group coordinated to the metal.

1.212(8) Å (∆

=

0.065 Å) for [Cp*Mo^II(CO)(PMe₃)₂-

(O₂CMe)],^[16]1.276(5) and 1.255(5) Å compared with

1.200(5) and 1.192(6) Å (∆ = 0.076 and 0.063 Å, respec-

tively) for [CpMo^III(η⁴-C₄H₆)(O₂CCF₃)₂],^[17]and 1.281(8)

and 1.300(8) Å compared with 1.208(8) and 1.210(8) Å (∆

= 0.073 and 0.090 Å, respectively) for [Cp₂Mo^IV(O₂-

CPh)₂].^[18]This combined evidence indicates a significant

ionic contribution to the bond between the Mo center and

Table 2. Characteristics of the ν^as

(OCOCF₃)] in the solid state and in solution.

band of [Cp*Mo(dppe)H₂-

OCO

the monodentate anion, which then continues to promote Solvent

some electronic delocalization between the two C–O bonds.

ν^as

/cm^–1

∆ν

Intensity ratio

OCO

(A_low/A_high

)

Other relevant structural parameters are listed in Table 1.

Solid (KBr)

C₆H₆

C₆D₅CD₃

THF

1700 sh, 1685 s

1700 sh, 1690

1702, 1690

15

10

12

18

1.8

1.2

1.1

1.0

2.2

CH₂Cl₂

1706 sh, 1688

Protonation in Benzene/Toluene: Evolution with Time

The ¹H NMR spectroscopic monitoring of the

[Cp*Mo(dppe)H₃]–TFA reaction in C₆D₆at different TFA/

Mo ratios at room temperature was most enlightening. Fig-

ure 3 (a) shows the spectrum of the starting material before

TFA addition.^[21]

When using a 10-fold excess of TFA, the tetrahydrido

complex [Cp*Mo(dppe)H₄]⁺was the only observed prod-

uct, see Figure 3 (b). This is characterized by a hydride trip-

let resonance at –3.57 ppm (J_HP= 37.9 Hz) in the ¹H NMR

spectrum and by a ³¹P resonance at 72.3 ppm in C₆D₆, quite

Figure 2. ORTEP view of compound [Cp*Mo(dppe)H₂-

(OCOCF₃)]. Hydrogen atoms, except for the two hydride ligands,

are omitted for clarity. The thermal ellipsoids are drawn at the 30%

probability level.

–

close to the values previously reported for the BF₄salt in

CDFCl₂at –60 °C.^[14]In agreement with the literature, de-

composition takes place at room temperature (90% after

2816

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 2813–2826

Solvent Control of Hydride Protonation

FULL PAPER

protonation process is at equilibrium using this particular

proton donor. Both complexes exhibit sharp resonances

when alone and also when mixed together with

[Cp*Mo(dppe)H₂(OCOCF₃)]. Interestingly, the hydride res-

onance of [Cp*Mo(dppe)H₄]⁺broadens again at T Ͻ

240 K, but does not decoalesce. On the other hand, the

–

BF₄salt of this complex was previously reported to retain

sharp lines in the ¹H- and ³¹P NMR spectra down to

–90 °C, indicating rapid equilibration between the inequiva-

lent H sites.^[14]This suggests that the hydride scrambling

process is slowed down by hydrogen bonding between

[Cp*Mo(dppe)H₄]⁺and CF₃COO^–.

The reaction with 3 equiv. TFA resulted in the observa-

tion of a sharp resonance at the chemical shift of the tetra-

hydride complex (indicating essentially complete disappear-

ance of the starting trihydride reagent) and no significant

amount of the dihydride product. The subsequent addition

of 2 equiv. Et₂NH to this solution led to the formation of

Figure 3. ¹H NMR (300 MHz) spectroscopic monitoring of the re-

action between [Cp*Mo(dppe)H₃] and TFA in C₆D₆(290 K): (a)

starting material before the TFA addition; (b) immediately after the

TFA addition (10 equiv.); (c) immediately after the TFA addition

(1 equiv.); (d) after 3.5 h from spectrum (c).

ca. 25 minutes), without yielding any hydride-containing complex [Cp*Mo(dppe)H₂(OCOCF₃)] cleanly and quanti-

product. This decomposition was not investigated in any tatively. When using 1.5 equiv. TFA, rather sharp reso-

further detail.

nances for the tetrahydrido complex and the final dihydrido

1

For the 1:1 reaction, a broad resonance was initially pres- product were initially observed by H NMR spectroscopic

ent in the hydride region at an intermediate position be- monitoring, but the latter was once again the only observ-

tween those of the trihydride and tetrahydride complexes, able species remaining at the end of the reaction (ca. 30

in addition to the final product resonance (Figure 3, c). The minutes at 290 K). When using only 0.5 equiv. TFA, on the

position and shape of the broad resonance depends on the other hand, a broad resonance was again initially present,

TFA/Mo ratio and on the temperature (vide infra). The but its chemical shift was closer to that of the starting hy-

³¹P{¹H} NMR spectrum at room temperature exhibits two dride complex (see Figure 4). This is as expected because

small and broadened resonances at the characteristic posi- the equilibrium mixture of the rapidly exchanging

tions of the trihydride complex (δ = 91.4 ppm) and the tet- [Cp*Mo(dppe)H₃] and [Cp*Mo(dppe)H₄]⁺complexes is

rahydride protonation product (δ = 72.3 ppm),^[22]in ad- now richer in the former. The formation of the final dihy-

dition to the sharp resonance of the [Cp*Mo(dppe)- dride product was slower (equilibrium was attained in ca.

H₂(O₂CCF₃)] product at 76.9 ppm and a minor contami- two hours at 290 K), and ca. 50% of the initial trihydride

1

nant resonance.^[21]This suggests that the broad H reso- remained unreacted. As [Cp*Mo(dppe)H₄]⁺reacted to yield

nance results from a rapid degenerate exchange between the [Cp*Mo(dppe)H₂(OCOCF₃)], the resonance corresponding

starting complex [Cp*Mo(dppe)H₃], which is not to [Cp*Mo(dppe)H₃] sharpened. The addition of excess

completely consumed, and its protonation product, TFA to

a

solution containing [Cp*Mo(dppe)H₂-

[Cp*Mo(dppe)H₄]⁺. This exchange will be addressed in fur- (OCOCF₃)] did not lead to any reaction, in particular no

ther detail in the next section. Given the larger chemical tetrahydride resonance was observed under these condi-

shift difference between the ³¹P resonances, the shape of the tions.

³¹P NMR spectrum is much closer to the slow exchange

limit than that of the ¹H NMR spectrum. After a few hours

at room temperature (or upon brief warming to ca. 40 °C),

the broad band disappears and the resonance correspond-

ing to [Cp*Mo(dppe)H₂(OCOCF₃)] increases in intensity

(Figure 3, d). These observations reveal the presence of an

equilibrium between the tetrahydride complex and the

starting trihydride complex, followed by a slow irreversible

conversion to the final product by H₂loss.

A variable-temperature study of the reaction mixture in

C₆D₅CD₃, carried out immediately after mixing the rea-

gents in a 1:1 ratio, confirmed the earlier assignments. The

hydride resonance of the dihydride product evolved as dis-

cussed previously (Figure 1, a), whereas the broad reso-

nance at –3.63 ppm decoalesced at T Ͻ 270 K to yield two

Figure 4. ¹H NMR (300 MHz) spectroscopic monitoring of the re-

resonances at the expected positions of [Cp*Mo(dppe)H₃]

action between [Cp*Mo(dppe)H₃] and TFA (1:0.5 ratio) in C₆D₆

(290 K): (a) recorded immediately; (b) after 7 min; (c) 28 min; (d)

and [Cp*Mo(dppe)H₄]⁺. The incomplete protonation in the

presence of a stoichiometric amount of TFA shows that the 155 min.

Eur. J. Inorg. Chem. 2007, 2813–2826

www.eurjic.org

2817

R. Poli, E. S. Shubina et al.

FULL PAPER

We also investigated the reaction between [Cp*Mo-

(dppe)H₃] and CF₃COOD (1 equiv.) by ¹H NMR spec-

troscopy in C₆D₅CD₃. The reaction progressed in an essen-

tially identical manner to that when CF₃COOH was used:

immediate observation of the virtual triplet at –5.25 ppm

due to the final product and a broad resonance centered at

ca. –3.6 ppm. The latter disappears upon brief warming to

ca. 40 °C. However, the NMR spectrum shows the forma-

tion of H₂and HD in ca. 3:1 ratio. This is not surprising

since all hydride positions in the tetrahydride complex are

readily scrambled, therefore the elimination of H₂from the

[Cp*Mo(dppe)H₃D]⁺cation will be statistically favored.

The molybdenum product resulting from this reaction

should therefore contain both H and D. The hydride reso-

nance does not show evidence for the presence of isotopo-

mers, probably because the isotope shift is too small. The

formation of deuterated product(s), [Cp*Mo(dppe)-

HD(O₂CCF₃)] and/or [Cp*Mo(dppe)D₂(O₂CCF₃)], how-

ever, was confirmed by a ²H NMR-spectroscopic study,

which revealed a binomial triplet resonance at –5.9 ppm

with J_PD= 7.7 Hz.

Figure 5. ¹H NMR (300 MHz) spectroscopic monitoring of the re-

action between [Cp*Mo(dppe)H₃] and TFA in [D₈]thf (290 K): (a)

before the TFA addition; (b) TFA/Mo = 0.5, 8 min; (c) further

addition of 1 equiv. to sample (b): TFA/Mo = 1.5, 30 min; (d) TFA/

Mo = 10, immediately after the addition; (e) after 13 min from

spectrum (d). The doublet resonance at δ = –3.73 ppm belongs to

the [Cp*Mo(κ¹-dppe)H₅] contaminant.^[14]

study of a 2:1 [Cp*Mo(dppe)H₃]/HBF₄mixture provides

evidence in favor of the unimolecular pathway. As seen in

Figure 6, addition of 0.5 equiv. HBF₄results in a decrease

in the [Cp*Mo(dppe)H₃] resonance and the appearance of

the [Cp*Mo(dppe)H₄]⁺resonance, plus a weak triplet reso-

nance at –5.65 ppm (J_HP= 53.5 Hz), whilst the doublet res-

onance of the [Cp*Mo(κ¹-dppe)H₅] contaminant remains

unchanged. The trihydride and tetrahydride complexes give

separated sharp resonances, contrary to the same situation

in the presence of trifluoroacetate anion in the same solvent

(Figure 5, b). This clearly demonstrates that the anion de-

termines the rate of degenerate exchange. This exchange

process therefore involves the hydrogen-bonded ion pair

[Cp*Mo(dppe)H₄]⁺···^–O₂CCF₃and the dihydrogen-bonded

trihydride [Cp*Mo(dppe)H₃]···HO₂CCF₃, the latter being

in further rapid equilibration with free trihydride and TFA.

Analogous broadening phenomena for hydride signals as a

consequence of dihydrogen bonding and equilibrated pro-

ton transfer were noted previously in other cases, although

the broadening effect was limited to a few Hz in most

cases.^[2,23]

Protonation in Other Solvents: NMR-Spectroscopic Study

The solvent has a significant effect on the course of the

protonation reaction. The behavior in [D₈]thf is similar but

not identical to that in aromatic hydrocarbons (see Fig-

ure 5); use of a (sub)stoichiometric amount of the acid ini-

tially yields a broad hydride resonance (Figure 5, b). Time

evolution at these ratios leads to sharpening of the signals

as the irreversible H₂elimination and [Cp*Mo(dppe)-

H₂(O₂CCF₃)] formation takes place: a similar process to

that shown in Figure 4 for the benzene solution. When the

sample corresponding to Figure 5 (b) was treated with ad-

ditional TFA (total 1.5 equiv.), quantitative formation of

the dihydride product occurred (Figure 5, c). Use of a large

excess (10 equiv.) of CF₃COOH affords a sharp triplet cor-

responding to the tetrahydrido cation and a strong reso-

nance for the [Cp*Mo(dppe)H₂(O₂CCF₃)] product (Fig-

ure 5, d). As in C₆D₆or C₆D₅CD₃, a broad resonance is not

obtained at this ratio. However, with time, the tetrahydride

resonance disappears and the dihydride resonance increases

in [D₈]thf (Figure 5, e), whereas in the analogous benzene

solution decomposition occurs without formation of the di-

hydride product, as stated earlier.

THF was found to be a suitable solvent in which to

probe the nature of the degenerate exchange between

[CpMo(dppe)H₃] and [Cp*Mo(dppe)H₄]⁺, by repeating the

protonation in the presence of HBF₄instead of CF₃COOH

(the BF₄^–salt of the tetrahydride product is soluble in THF,

but precipitates from toluene or benzene). If the exchange

takes place bimolecularly via a symmetrical [Cp*(dppe)-

H₃Mo···H···MoH₃(dppe)Cp*]⁺transition state, the line-

shape of the signal will not depend on the nature of the acid.

On the other hand, a unimolecular pathway involving the

conjugate base of the proton donor should yield a slower

Figure 6. ¹H NMR-spectroscopic study of a [Cp*Mo(dppe)H₃]/

HBF₄mixture in [D₈]thf: (a) starting compound; (b) 21 min after

the addition of 0.5 equiv. HBF₄. The doublet resonance at δ =

–

exchange rate for the BF₄sample, because this is a much

weaker base than CF₃COO^–. The NMR-spectroscopic –3.73 ppm belongs to the [Cp*Mo(κ¹-dppe)H₅] contaminant.^[14]

2818

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 2813–2826

Solvent Control of Hydride Protonation

FULL PAPER

The new triplet resonance at –5.65 ppm in Figure 6 is

assigned to complex [Cp*Mo(dppe)H₂(THF)]⁺. It is associ-

ated with a new ³¹P{¹H} resonance at 70.7 ppm. For com-

parison, the related [Cp*Mo(dppe)H₂(MeCN)]⁺complex,

obtained from the protonation of [Cp*Mo(dppe)H₃] by

HBF₄in MeCN,^[14]displays a proton resonance at

–5.50 ppm (J_P1= 51 Hz, J_P2= 52 Hz) and a phosphorus

resonance at 78.1 ppm. NMR spectroscopic monitoring

shows that this resonance grows while the resonance of the

tetrahydride product simultaneously decreases. Interesting

features to note are (i) this cationic THF adduct is not ob-

served in the protonation experiment with CF₃COOH; (ii)

the rate of disappearance of the tetrahydride complex is

1

Figure 7. H NMR-spectroscopic study of the [Cp*Mo(dppe)H₃]–

TFA reaction in [D₃]MeCN at room temperature: (a) 0.5 equiv.

TFA, recorded immediately; (b) 1 equiv. TFA, recorded immedi-

ately; (c) after 9 min from spectrum (b); (d) 40 min after spectrum

(b).

–

much slower in the presence of the BF₄anion than in the

presence of the CF₃COO^–anion (cf. Figure 6 with Fig-

ure 5); and (iii) conversion to the THF solvent adduct ap-

pears to be much slower than the analogous reaction in

MeCN, which was reported to lead to vigorous gas evol-

ution and quantitative production of the MeCN complex

within 60 minutes.^[14]

[Cp*Mo(dppe)H₂(S)]⁺+ CF₃COO^–Ǟ

[Cp*Mo(dppe)H₂(O₂CCF₃)] + S

(2)

dissociation, [Cp*Mo(dppe)H₂]⁺, is achieved. The coordi-

nating CF₃COO^–anion is an even better trap than THF

and MeCN. The nature of the solvent has a dramatic effect

on two related features: (i) the ability of the CF₃COO^–

anion to attack the cationic complex and yield the neutral

dihydride-trifluoroacetate product and (ii) the rate of de-

generate exchange between [Cp*Mo(dppe)H₃] and

[Cp*Mo(dppe)H₄]⁺. The way in which these two processes

are affected by the solvent properties will be addressed in

more detail in a later section.

When using CH₂Cl₂as the solvent, the initially formed

tetrahydride complex displayed a sharper resonance, even

when as little as 0.5 equiv. of the acid was used. The forma-

tion of minor amounts of [Cp*Mo(dppe)H₂(O₂CCF₃)] was

also detected (ca. 10% for the 1 equiv. experiment). In this

solvent, however, the interpretation of the data is difficult

because of the instability of the starting material, the cat-

ionic tetrahydride intermediate, and the product (24% de-

composition within one hour at room temperature).

Finally, the results of the protonation with TFA (0.5 and

1 equiv.) in [D₃]MeCN at room temperature are shown in

Figure 7. In both cases, the immediate formation of

[Cp*Mo(dppe)H₂(O₂CCF₃)] and [Cp*Mo(dppe)H₄]⁺was

observed. When using 0.5 equiv., the resonances corre-

IR Studies

As equilibrium positions and reaction rates are expected

sponding to the tetrahydride complex and trihydride resid- to strongly depend on the solvent’s ability to influence hy-

ual starting compound are separate and relatively sharp (see drogen bonding and ion pairing equilibria, we considered

Figure 7, a): a similar situation to that seen in CD₂Cl₂using that additional useful information could be gathered from

TFA as the proton donor or in [D₈]thf using HBF₄. This infrared investigations, especially at low temperature where

indicates that the degenerate exchange is slow under these ion pairing and hydrogen bonding are favored. The dielec-

conditions contrary to the situations in [D₈]thf and C₆D₆tric permittivity, ε, increases at ambient temperature in the

using TFA as the proton donor. In turn, this suggests that order toluene (2.4) Ͻ THF (7.8) Ͻ CH₂Cl₂(8.3) ϽϽ

weak ion pairing occurs, if any, between the tetrahydrido MeCN (35.9).^[24]It increases substantially upon cooling for

[25,26]

cation and the trifluoroacetate anion in [D₃]MeCN. The some solvents, for example, values of 15.5 for CH₂Cl₂

conversion of the tetrahydride complex to [Cp*Mo(dppe)- and 11.9 for THF^[27,28]are recorded at 200 K, but not for

H₂(O₂CCF₃)] was relatively fast and quantitative, see parts others, for example, 2.47 at 240 K and 2.71 at 180 K for

b–d in Figure 7. As already noted, when the reaction was toluene.^[26]On the other hand, these solvents have different

carried out in THF the final product was the trifluoroacet- coordinating and acid/base properties. THF is quite an ef-

ate compound rather than the solvent adduct ficient hydrogen-bond acceptor (E_j= 1.04), better than

[Cp*Mo(dppe)H₂(MeCN)]⁺(previously shown to result MeCN (E_j= 0.75),^[29]although the latter is generally a bet-

from the interaction with HBF₄), implying that the ligand ter ligand for transition metals. Benzene and toluene are

exchange shown in Equation (2) (S = THF, MeCN) is com- much weaker bases (E_j= 0.4),^[29]whereas CH₂Cl₂tends to

pletely displaced towards the right.

behave more like a weak proton donor than as a proton

A first comparison of the behavior of [Cp*Mo(dppe)- acceptor.^[24]We start by presenting the spectral features of

H₄]⁺in noncoordinating solvents confirms the complex in- the TFA proton donor in the ν_COregion in various solvents,

–

stability, as previously shown for the BF₄salt, leading to since the TFA/solvent interaction certainly plays an impor-

hydride-free decomposition products.^[14]However, in the tant role in the proton-transfer equilibria involving the hy-

presence of either coordinating solvents or the CF₃COO^–dride complex. The relevant spectra are shown in Figure 8.

anion, efficient trapping of the 16-electron product of H₂Two bands at similar frequencies are observed in CH₂Cl₂

Eur. J. Inorg. Chem. 2007, 2813–2826

www.eurjic.org

2819

R. Poli, E. S. Shubina et al.

FULL PAPER

and toluene (1804 and 1785 cm^–1for the former, 1803 and

1788 cm^–1for the latter). The higher frequency band is due

to the acid monomer, whereas the lower frequency one is

typically assigned to the corresponding dimer.^[30]Note that

the equilibrium is shifted towards the dimer in toluene, as

is evident from the different relative intensities. In THF the

wide, asymmetric ν_COband of the acid monomer is greatly

redshifted to 1780 cm^–1because of the formation of

stronger hydrogen bonds to solvent molecules. The shoulder

at 1762 cm^–1(which becomes a distinct band at 200 K) is

assigned to the CF₃COOH dimer. The low-frequency asym-

metry of the acid bands in all solvents suggests the presence

of higher order TFA associates. Thus THF can form hydro-

gen bonds with CF₃COOH, lowering the activity of the

acid, which is one of the reasons for the evolution of

[Cp*Mo(dppe)H₄]⁺CF₃COO^–into the final dihydride prod-

uct in this solvent, as will be shown later.

Figure 9. Low-temperature (200 K) IR spectra for the reaction be-

tween [Cp*Mo(dppe)H₃] and TFA in CH₂Cl₂: (a) [Cp*Mo(dppe)-

H₃] (0.028 ); (b) TFA (0.014 ); (c) mixture of [Cp*Mo(dppe)H₃]

(0.028 ) and TFA (0.014 ); (d) mixture of [Cp*Mo(dppe)H₃]

(0.028 ) and TFA (0.028 ).

the position of the free CF₃COO^–anion^[31,32]indicates the

formation of either a weakly H-bonded contact ion pair or

a solvent-separated ion pair, consistent with the relatively

high polarity of CH₂Cl₂at low temperatures and the poor

proton-donor ability of the [Cp*Mo(dppe)H₄]⁺cation. In

reference to Scheme 1, this means that species V is exten-

sively dissociated to VI or that the interaction is loose under

these solvent and temperature conditions. There is no sig-

nificant intensity in the region characteristic for the free

homoconjugate anion, [CF₃COO(CF₃COOH)_n]^–. Accord-

ing to the literature, this species should display a low inten-

sity, broad band with a maximum at 1630–1620 cm^–1.^[33]

Given the presence of an excess of the trihydride complex,

most of the acid is presumably consumed and little is left

to yield the homoconjugate anion. The relative intensity of

the 1686 cm^–1band indicates that the proton-transfer equi-

librium is highly shifted to the tetrahydride product, and

Figure 8. IR spectra of trifluoroacetic acid solutions (0.027 ,

pathlength 0.04 mm) in various solvents at 290 K: (a) CH₂Cl₂; (b)

toluene; (c) THF.

In Dichloromethane

Although the starting material and products are unstable the presence of the unreacted trihydride complex is indi-

in dichloromethane at room temperature, they do not de- cated by the shoulder at 1816 cm^–1{cf. the starting spec-

compose at low temperatures and this solvent has many ad- trum of free [Cp*Mo(dppe)H₃] in Figure 9, a}. There are

vantages for infrared studies (it is not a strong proton ac- two additional weak bands in this spectrum. The first one

ceptor, therefore it does not compete with the basic hydride at 1782 cm^–1could be assigned to free CH₃COOH, but it is

complexes for hydrogen bonding, and it has good solvent also consistent with a H-bonded adduct where the acid acts

properties for ionic compounds). An IR spectroscopic mon- as a proton donor (a CF₃COO–H···B interaction does not

itoring at 200 K in the acid ν_COand conjugated anion significantly perturb the CO normal mode). The second

ν^as

vibration region yielded the results shown in Fig- band at 1736 cm^–1indicates the presence of a H-bonded

OCO

ure 9. At this temperature the dihydride complex adduct where the acid acts as a proton acceptor [the

[Cp*Mo(dppe)H₂(O₂CCF₃)], which has

a

ν^as

at CF₃C(OH)O···HA interaction significantly redshifts the

OCO

1692 cm^–1(vide infra), did not form.

CO normal mode].^[34,35]Thus, these bands constitute evi-

A tentative assignment of all observed bands is possible dence for the presence of a 2:1 H-bonded adduct,

on the basis of previous IR studies into the interaction be- [Cp*Mo(dppe)H₃]···HO₂CCF₃···HO₂CCF₃. The simulta-

tween CF₃COOH and various bases,^[30–32]in combination neous presence of a 1:1 adduct, [Cp*Mo(dppe)H₃]···

with the results shown in the previous section. At 200 K HO₂CCF₃, cannot be established because it would also af-

the acid self-association equilibrium is shifted to the dimer, ford a band at ca. 1780 cm^–1. A general picture of the pro-

exhibiting a strong band at 1780 cm^–1with a shoulder at ton-transfer process can be drawn as shown in Scheme 4.

1800 cm^–1(Figure 9, b). The spectrum recorded for the

solution with a MoH₃/TFA ratio of 1:0.5 (Figure 9, c) shift the hydrogen-bonding equilibrium position towards

shows a prominent band at 1686 cm^–1due to the ν^as

the 2:1 adduct. At the same time, equilibria involving the

An increase in the CF₃COOH/MoH₃ratio is expected to

OCO

vibration of the trifluoroacetate anion. Its relatively sharp anionic species (either hydrogen bonded or free) are ex-

nature and almost negligible high frequency shift relative to pected to shift towards the homoconjugate anion (equilibria

2820

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 2813–2826

Solvent Control of Hydride Protonation

FULL PAPER

Scheme 4.

drawn vertically in Scheme 4). On the other hand, an in-

crease in the CF₃COOH/MoH₃ratio will not affect the

equilibria shown horizontally in Scheme 4. Indeed, upon

addition of a second 0.5 equiv. TFA (final MoH₃/TFA ratio

of 1:1), see Figure 9d, the IR spectrum reveals a significant

increase in intensity for the bands assigned to the neutral

2:1 adduct and a dramatic decrease for that assigned to the

free anion. The growth of a broad shoulder at lower fre-

quencies signals the formation of anionic aggregates, as ex-

pected.

In [D₈]Toluene

The low-temperature IR spectral picture in [D₈]toluene

in the presence of a stoichiometric amount of TFA is sim-

ilar to that shown in dichloromethane, see Figure 10. The

band at 1742 cm^–1observed for the 1:1 mixture (Figure 10c)

is assigned to the ν_COof the hydrogen-bonded complex (cf.

1736 cm^–1in CH₂Cl₂), whereas the bands at 1690 and

1674 cm^–1are assigned to the CF₃COO^–anion (ν^as_OCO, cf.

1686 cm^–1in CH₂Cl₂). The complex pattern attained for

Figure 10. IR spectra for the reaction between [Cp*Mo(dppe)H₃]

and TFA in [D₈]toluene: (a) [Cp*Mo(dppe)H₃] (0.027 ) at 200 K;

(b) TFA (0.027 ) at 200 K; (c) 1:1 mixture (0.027 ) at 200 K; (d)

same mixture as (c), after complete conversion at room tempera-

ture.

In THF

At low temperature in this solvent and in the presence of

this absorption is possibly related to the presence of dif- 1 equiv. TFA, the dominant species were the neutral hydro-

ferent hydrogen-bonded species (hydrogen bonding should gen-bonded complexes, as indicated by the major bands at

be favored more in low-polarity toluene). The anion band 1784 and 1744 cm^–1(see Figure 11, c). No significant

did not show a significant increase in intensity at greater amounts of the CF₃COO^–ion were seen. This implies that

MoH₃/TFA ratios (1:0.5 or 1:0.2), indicating that the pro- proton transfer occurs to a smaller degree than in toluene.

ton-transfer equilibrium is shifted further towards the neu- No homoconjugated anion band was observed upon raising

tral hydrogen-bonded species in this solvent in comparison the TFA excess to 5 equiv., in contrast to the results in aro-

with CH₂Cl₂. This agrees with the results of the NMR in- matic hydrocarbon solution. Upon warming, the spectral

vestigation. Upon warming above 270 K, the bands corre- changes were similar to those observed in toluene (see pre-

sponding to the [Cp*Mo(dppe)H₂(O₂CCF₃)] product at vious section), yielding [Cp*Mo(dppe)H₂(O₂CCF₃)], see

1702 and 1690 cm^–1quickly grow, while the shoulder at Figure 11 (d). We also investigated the reverse proton-trans-

–

1674 cm^–1disappears. The final spectrum is shown in part fer reaction by mixing together [Cp*Mo(dppe)H₄]⁺BF₄

d of Figure 10. Monitoring the growth of this band for the {generated in situ at 200 K from [Cp*Mo(dppe)H₃] and

1:1 reaction in C₆H₆at 298 K (see Figure S3 in the Support- HBF₄} and CF₃COONa at 200 K in THF. This experiment

ing Information) yielded a first-order rate constant of confirms the reversible nature of the proton-transfer pro-

(1.8Ϯ0.1)ϫ10^–3s^–1. In the presence of a 3–5-fold excess of cess and the position of equilibrium. The essentially com-

acid, the broad, low intensity ν^as_OCOband of the homocon- plete disappearance of the trifluoroacetate anion ν^as

OCO

jugated anion was observed at ca. 1640 cm^–1. No formation bands at 1702 and 1690 cm^–1and the growth of the ν_CO

of [Cp*Mo(dppe)H₂(O₂CCF₃)] was observed under these band at 1752 cm^–1corresponding to the H-bonded ad-

conditions, in agreement with the NMR spectroscopic data. duct^[36]were observed at 200 K in the presence of either 1

Eur. J. Inorg. Chem. 2007, 2813–2826

www.eurjic.org

2821

R. Poli, E. S. Shubina et al.

FULL PAPER

or 2 equiv. [Cp*Mo(dppe)H₄]⁺BF₄. Subsequent warming Information) giving the following activation parameters:

–

to 290 K led once again to [Cp*Mo(dppe)H₂(OCOCF₃)] ∆H^‡= (31.8Ϯ0.5) kcalmol^–1and ∆S^‡= (36Ϯ2) e.u. The

formation.

large positive activation entropy suggests that the transition

state is dissociative in nature. At low temperatures, the re-

arrangement becomes too slow; for instance, reaction rate

constants of 5.2ϫ10^–8and 5.8ϫ10^–15s^–1are calculated at

250 and 200 K, respectively.

Table 3. Rate constants for the [Cp*Mo(dppe)H₂(O₂CCF₃)] forma-

tion reaction in THF; [MoH₃] = 0.025 .

T /K

TFA/Mo

k /10^–3s^–1

280

290

298

308

1

0.5

1

3

5

1

0.064Ϯ0.0015

0.63Ϯ0.01

1.7Ϯ0.1

1.8Ϯ0.1

1.9Ϯ0.1

1.80Ϯ0.05

13.7Ϯ0.2

Figure 11. Low-temperature (200 K) IR spectra for the reaction of

[Cp*Mo(dppe)H₃] with TFA in THF: (a) [Cp*Mo(dppe)H₃]

(0.07 ); (b) TFA (0.07 ); (c) 1:1 mixture (0.07 ) at 200 K; (d)

same mixture as (c), after complete conversion at room tempera-

ture.

Discussion

Since the NMR-spectroscopic study shows that this con-

The collective observations that will be detailed now are

version also occurs in the presence of excess acid, a kinetics summarized in Scheme 5.

investigation was carried out at 298 K using different TFA/

Proton transfer to [Cp*Mo(dppe)H₃], via the H-bonded

Mo ratios. The reaction was found to be zero order with adduct A to yield the classical tetrahydride complex,

respect to the acid (see Table 3). Interestingly, the rate con- [Cp*Mo(dppe)H₄]⁺, C/D, is the only reaction that may oc-

stant in THF is identical to that measured in benzene (vide cur at low temperatures. The formation of the 2:1 H-

supra) within experimental error. The study was also ex- bonded adduct AЈ was also evidenced by low-temperature

tended to different temperatures (using a TFA/Mo ratio of IR spectroscopy, but one molecule of TFA is sufficient for

1:1), with the subsequent Eyring analysis (see Supporting proton transfer to occur, as also indicated by previous ki-

Scheme 5.

2822

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 2813–2826

Solvent Control of Hydride Protonation

FULL PAPER

netics studies.^[13]The proton transfer probably proceeds via quantitative formation of the final product E. However, ex-

a nonclassical intermediate (species B in Scheme 5), al- cess acid favors ion-pair dissociation through formation of

though we have not found any direct experimental evidence the thermodynamically more stable homoconjugate anion,

to support this proposition. Our recently reported theoreti- thereby stabilizing the proton-transfer product D and pre-

cal calculations^[13]indicate that the nonclassical complex cluding the H₂evolution.

lies in a very shallow minimum with a low barrier for con-

In solvents with proton-accepting ability (THF, MeCN)

version to the classical tetrahydride product, with the isom- the activity of the acid is reduced by hydrogen bonding with

erization taking place without counteranion dissociation. solvent molecules, formation of the homoconjugated anion

The

competitive

dihydrogen

evolution

yielding is not observed, and the reaction yields E even in the pres-

[Cp*Mo(dppe)H₂(O₂CCF₃)] occurs only above 250 K and ence of excess acid. The attainment of the same formation

is possible (in noncoordinating solvents) by the reverse reac- rate constant in THF and in benzene and the zero-order

tion between D and CF₃COO^–. This forms the ion pair C dependence on TFA agree with the proposed unimolecular

which subsequently isomerizes to the nonclassical species B. generation of E from the same intermediate B. Further-

The formation of [Cp*Mo(dppe)H₄]⁺is reversible below more, the large positive activation entropy is consistent with

250 K in contrast to the irreversible formation of trans- a dissociative mechanism (i.e., H₂must leave before the tri-

Cp*MH₂(dppe)⁺(M = Fe, Ru).^[6,8]One important differ- fluoroacetate anion can coordinate), as may be expected

ence between these systems is the reversibility of the non- from the electronically saturated nature of system B. The

classical/classical isomerization for [Cp*Mo(dppe)H₄]⁺(ac- rate of formation in MeCN was not determined with accu-

cording to the DFT calculations), whereas this process is racy, but the NMR spectroscopic data in Figure 7 allow us

irreversible for [Cp*M(dppe)H₂]⁺(M = Fe, Ru). However, to estimate a rate constant of ca. 2ϫ10^–3s^–1, that is, close

proton transfer to the hydride ligand is less thermodynami- to that determined in THF and benzene (see Table 3), if we

cally favorable for [Cp*Mo(dppe)H₃] than for assume a first-order decay process is in operation. The high

[Cp*Fe(dppe)H]; a significant amount of TFA remained polarity of MeCN favors proton transfer and ion-pair

unreacted (in the H-bonded form) for a TFA/Mo ratio of dissociation through the stabilization of the charge-sepa-

0.5 in CH₂Cl₂at 200 K (see Figure 9), whereas it was com- rated species, but the instability of [Cp*Mo(dppe)H₄]⁺leads

pletely consumed under the same solvent, temperature, and to its ultimate decomposition by H₂elimination. In this

stoichiometry conditions in the presence of [Cp*Fe(dppe)- case, the process may be solvent-assisted, leading to the sol-

H].^[5]These results contrast with the higher basicity factor vent adduct G, as previously established by the protonation

of [Cp*Mo(dppe)H₃] (E_j= 1.42Ϯ0.02)^[13]relative to with HBF₄.^[14]This phenomenon also occurs in THF as

[Cp*Fe(dppe)H] (E_j= 1.36Ϯ0.02).^[7]The corresponding verified by using HBF₄. However, the CF₃COO^–anion has

proton transfer to [Cp*Ru(dppe)H] (E_j= 1.39), however, is a stronger coordinating power than both THF and MeCN,

also an equilibrium process.^[6]The basicity factor correlates and the ultimate product is once again E.

with the strength of the hydrogen bond (formation en-

Species A and C/D exhibit a dynamic exchange process,

1

thalpy), whereas the protonation equilibrium is determined as revealed by a broad H resonance at room temperature

by the free energy of the proton-transfer process. Obviously, which decoalesces at 270 K in C₆D₅CD₃for the mixture

the two parameters do not necessarily correlate quantita- containing 1 equiv. TFA. This results from the fast and di-

tively.

rect unimolecular proton transfer involving the dihydrogen-

Both the A i C/D equilibrium position and rate of the bonded complex A and the H-bonded ion pair C. It is not

irreversible [Cp*Mo(dppe)H₂(O₂CCF₃)] formation depend possible, unfortunately, to use this information to derive an

on the temperature, the solvent, and the MoH₃/TFA ratio. accurate value for the proton-transfer rate constant because

Polar solvents (e.g., dichloromethane at low temperatures, of the complexity of the coupled equilibria and the irrevers-

acetonitrile) weaken the hydrogen bond in C and lead to ible transformation to product E. Assuming, however, that

the formation of a solvent-separated ion pair (species D in only species A and C are present in an approximately equi-

Scheme 5), which stabilizes the tetrahydrido complex molar ratio, an activation barrier ∆G^‡of 11.7 kcalmol^–1

against the H₂loss. The proton-transfer equilibrium is may be estimated for the proton transfer in C₆D₅CD₃at

shifted to species D in CH₂Cl₂even in the presence of stoi- 270 K, from the coalescence temperature and the chemical

chiometric amounts of the acid. Low-polarity solvents (e.g., shift difference.^[37]This activation barrier is lower than that

toluene) provide less stabilization for the charged species, of the proton transfer by HFIP in toluene (∆G^‡_{293 K}

=

thus the proton-transfer equilibrium is shifted to the left 15.8 kcalmol^–1),^[13]in agreement with the higher proton-do-

when near stoichiometric amounts of the acid are used. The nating ability of TFA. It is also lower than ∆G^‡_{270 K}for the

tetrahydride product exists as an ion pair stabilized by hy- proton transfer from HFIP to [Cp*Fe(dppe)H] in CH₂Cl₂

drogen bonding to the CF₃COO^–anion (species C in (14.7 kcalmol^–1) or from PFTB to [CpRuH(CO)(PCy₃)] in

Scheme 5), as reflected for instance in a reduced rate of hy- hexane (16.1 kcalmol^–1), calculated for 270 K using the

dride site exchange measured by ¹H NMR spectroscopy be- published activation parameters.^[3,5]IR kinetic data (vide

low 240 K. The formation of hydrogen-bonded ion pairs supra) give ∆G^‡_{270 K}= 22.1 kcalmol^–1for the dihydrido-tri-

of 1:1 composition was also confirmed for weaker proton fluoroacetate formation. The irreversibility of the H₂evolu-

donors.^[13]At room temperature and above, the competitive tion process renders [Cp*Mo(dppe)H₂(O₂CCF₃)] the sole

irreversible H₂elimination process eventually leads to the end product.

Eur. J. Inorg. Chem. 2007, 2813–2826

www.eurjic.org

2823

R. Poli, E. S. Shubina et al.

FULL PAPER

Scheme 6.

A final point of interest is a comparison between the re- mer (dihydrogen complex); it is stable at low temperatures

sults shown here and those reported previously for the de- and can exist as solvent-separated or contact ion pair. How-

composition of the proton-transfer product obtained with ever, the use of suitable conditions of solvent, temperature,

TFE, which led to a diamagnetic hydride-free decomposi- and hydride/acid ratio led to the selective formation of a

tion product, [Cp*Mo(dppe)(OCH₂CF₃)(TFE)_x] (x = 0 or new product, the dihydrido complex [Cp*Mo(dppe)-

1). The mechanism is likely to be the same for both systems, H₂(O₂CCF₃)]. The trifluoroacetate anion is a sufficiently

leading to a first H₂elimination assisted by the coordina- strong ligand to coordinate to the cationic complex and sat-

tion of the alkoxide ligand (see Scheme 6). Why does urate its coordination sphere after loss of H₂, even in the

[Cp*Mo(dppe)H₂(O₂CCF₃)] show no tendency to expel a presence of coordinating solvent molecules such as MeCN

second H₂molecule and form the hypothetical compound or THF. However, it remains coordinated in a monodentate

[Cp*Mo(dppe)(O₂CCF₃)], in which the trifluoroacetate li- fashion and, contrary to the CF₃CH₂O^–anion, is not cap-

gand would bind in

a chelating fashion, whereas able of inducing the elimination of the residual hydride li-

[Cp*Mo(dppe)H₄]⁺(^–OCH₂CF₃) leads to a hydride-free gands as H₂to form a hydride-free product.

product without the observation of even trace amounts of

a dihydride intermediate? A possible rationalization is that

the stronger π-donating ability of the oxygen lone pairs in

Experimental Section

the fluorinated alkoxide ligand may assist the elimination

of the second H₂molecule by stabilizing the resulting 16-

electron half-sandwich Mo^IIproduct. By contrast, the lone

pair of the trifluoroacetate oxygen donor atom is not suffi-

ciently basic and the second oxygen atom is presumably not

capable of assisting the H₂elimination process, because of

either thermodynamic or kinetic reasons.

General: All manipulations were performed under argon by stan-

dard Schlenk techniques. All solvents were dried with an appropri-

ate drying agent (Na/benzophenone for benzene, toluene or THF;

Na for pentane; CaH₂for CH₂Cl₂) and were freshly distilled under

argon prior to use. C₆D₆, C₆D₅CD₃, [D₈]thf, CD₃CN, and CD₂Cl₂

(Euriso-Top) for NMR spectroscopic experiments and C₆D₅CD₃

(Aldrich) for IR spectroscopic experiments were degassed by three

freeze-pump-thaw cycles, and then purified by vacuum transfer at

room temp. [Cp*Mo(dppe)H₃] was synthesized according to the

literature.^[14]

Conclusions

We explored the protonation of [Cp*Mo(dppe)H₃] by tri-

fluoroacetic acid in various solvents which display different

proton-accepting ability, polarity, and coordinating power

toward transition metals: dichloromethane, benzene/tolu-

ene, THF, and MeCN. The nature of the solvent and the

amount of excess acid determined the nature of the product

by delicately controlling the position of the proton transfer

and ion pairing equilibria. In agreement with previous in-

vestigations, the classical tetrahydrido cation was produced

following the initial formation of a dihydrogen-bonded in-

Spectroscopic Studies

NMR Investigations: Samples of [Cp*Mo(dppe)H₃] in each solvent

were prepared under argon in 5-mm NMR tubes. The ¹H- and

³¹P{¹H} NMR data were collected with Bruker AMX300 and

Bruker AV500 spectrometers, operating at 300.1 or 500.3 MHz and

121.5 or 202.5 MHz, respectively. The temperature was calibrated

using a methanol chemical-shift thermometer; the accuracy and

stability was Ϯ1 K. All samples were allowed to equilibrate at every

temperature for at least 3 min. The spectra were calibrated with the

residual solvent resonance (¹H) and with external 85% H₃PO₄

termediate without the observation of a nonclassical tauto- (³¹P). The conventional inversion-recovery method (180–τ–90) was

2824 Eur. J. Inorg. Chem. 2007, 2813–2826

www.eurjic.org

Solvent Control of Hydride Protonation

FULL PAPER

used to determine the variable-temperature longitudinal relaxation

time T₁. Standard Bruker software was used for the calculation of

the longitudinal relaxation time.

Table 4. Crystal data and structure refinement for compound

[Cp*Mo(dppe)H₂(O₂CCF₃)].

Empirical formula

Formula weight

Temperature

C₃₈H₄₁F₃MoO₂P₂

744.59

180(2) K

IR-Spectroscopic Investigations: The IR measurements were per-

formed with the “Infralum 801” FTIR spectrometer using CaF₂

cells of 0.04-cm path length. All IR measurements were carried out

by the use of a home-modified cryostat (Carl Zeiss Jena) in the

190–290 K temperature range. The cryostat modification allowed

operation under an inert atmosphere and the transfer of the rea-

gents (premixed either at low or room temp.) directly into the cell

precooled to the required temperature. The accuracy of the tem-

perature adjustment was Ϯ1 K. This setup was used both for the

variable-temperature studies and for the kinetics investigations.

Wavelength

0.71073 Å

Crystal system

Space group

monoclinic

P2₁/n

Unit cell dimensions

a = 11.2103(4) Å

b = 15.0429(7) Å

c = 20.1563(7) Å

3392.1(2) Å³

4

α = 90°

β = 93.675(3)°

γ = 90°

Volume

Z

Density (calculated)

Absorption coefficient

F(000)

1.458 Mgm^–3

0.531 mm^–1

1536

Synthesis of [Cp*Mo(dppe)H₂(O₂CCF₃)]: A CF₃COOH solution

(9.5 µL, 0.13 mmol) in THF (3 mL) was slowly added to a solution

of [Cp*Mo(dppe)H₃] (80 mg, 0.13 mmol) in THF (15 mL). The re-

action mixture was stirred at room temp. for ca. 1.5 h. The solvent

was evaporated, and the resulting solid was recrystallized from pen-

tane. Yield: 75 mg (78%). ¹H NMR (300.13 MHz, C₆D₆, 290 K): δ

= –5.09 (Mo–H₂), 1.79 (s, 15 H, Cp*), 3.2–2.5 (m, 4 H, –CH₂–

CH₂–), 7.10–7.77 (m, 20 H, Ph) ppm. ³¹P{¹H} NMR (121.5 MHz,

Crystal size

Theta range for data collection 2.96–30.03°

Index ranges

Reflections collected

Independent reflections

Completeness to θ = 30.00°

Absorption correction

Max. / min. transmission

Refinement method

Data / restraints / parameters

Goodness-of-fit on F²

Final R indices [I Ͼ 2σ(I)]

R indices (all data)

0.57ϫ0.3ϫ0.28 mm

–15 Յ h Յ 15, –12 Յ k Յ 21, –28 Յ l Յ 28

31116

9836 [R_int= 0.0477]

99.1%

semi-empirical from equivalents

0.9726 / 0.7281

full-matrix least-squares on F²

9836 / 0 / 428

1

C₆D₆, 290 K): δ = 76.9 (s) ppm. The H hydride and the ³¹P reso-

nances decoalesce at low temperatures (see Results section). IR

1.082

(KBr): ν = 1682 and 1702 (sh) (ν^as_OCO), 1818 and 1850 (Mo–H₂)

˜

R₁= 0.0403, wR₂= 0.1022

R₁= 0.0480, wR₂= 0.1072

1.193 and –1.187 eÅ^–3

cm^–1. C₃₈H₄₁F₃MoO₂P₂(744.6): calcd. C 61.29, H 5.56; found C

61.26, H 5.54. Crystals were grown by slow diffusion of a pentane

Largest diff. peak and hole

layer into

a saturated solution of the mixture containing

[Cp*Mo(dppe)H₃] and ca. 1 equiv. of TFA in toluene.

Acknowledgments

Kinetics of the [Cp*Mo(dppe)H₂(O₂CCF₃)] Formation: The reagents

were mixed at low temperature (ca. 270 K), and the solution was

transferred into the cryostat. The kinetics data were obtained by

following the increase of the band at 1692 cm^–1corresponding to

the [Cp*Mo(dppe)H₂(O₂CCF₃)] complex at the desired tempera-

ture. The first-order rates were obtained by plotting ln(A_ϱ– A_t)

versus time t.

We thank the European Commission, HYDROCHEM program

(contract HPRN-CT-2002-00176) for support of this work. Ad-

ditional support from the Centre National de la Recherche Sci-

entifique (CNRS, program PICS), France, from the Russian Foun-

dation for Basic Research (RFBR) (05-03-22001, 05-03-32415) and

the Division of Chemistry and Material Sciences of RAS, Russia,

is also gratefully acknowledged. N. V. B. thanks Russian Science

Support Foundation for an individual grant. M. B. thanks the Min-

isterio de Educación y Ciencia (MEC) for a post-doctoral fellow-

ship.

X-ray Analysis of Compound [Cp*Mo(dppe)H₂(O₂CCF₃)]: A single

crystal was mounted under inert perfluoropolyether on the tip of

a glass fiber and cooled in the cryostream of the Oxford-Diffraction

XCALIBUR CCD diffractometer. Data were collected using the

monochromatic Mo-K_αradiation (λ = 0.71073). The structure was

solved by direct methods (SIR97)^[38]and refined by least-squares

procedures on F²using SHELXL-97.^[39]All hydrogen atoms at-

tached to carbon were introduced in the calculation in idealized

positions and treated as riding models. The two hydride ligands

were located in difference Fourier syntheses, and they were freely

refined with isotropic thermal parameters. The molecule was drawn

with the help of ORTEP32.^[40]Crystal data and refinement param-

eters are shown in Table 4.

[1] N. V. Belkova, E. S. Shubina, L. M. Epstein, Acc. Chem. Res.

2005, 38, 624.

[2] N. V. Belkova, A. V. Ionidis, L. M. Epstein, E. S. Shubina, S.

Gruendemann, N. S. Golubev, H. H. Limbach, Eur. J. Inorg.

Chem. 2001, 1753.

[3] N. Belkova, M. Besora, L. Epstein, A. Lledós, F. Maseras, E.

Shubina, J. Am. Chem. Soc. 2003, 125, 7715.

[4] E. Gutsul, N. Belkova, G. Babakhina, L. Epstein, E. Shubina,

C. Bianchini, M. Peruzzini, F. Zanobini, Russ. Chem. Bull.

2003, 52, 1204.

CCDC-632317 contains the supplementary crystallographic data

for this paper. These data can be obtained free of charge from The

Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/

data_request/cif.

[5] N. V. Belkova, E. Collange, P. Dub, L. M. Epstein, D. A. Le-

menovskii, A. Lledós, O. Maresca, F. Maseras, R. Poli, P. O.

Revin, E. S. Shubina, E. V. Vorontsov, Chem. Eur. J. 2005, 11,

873.

[6] N. V. Belkova, P. A. Dub, M. Baya, J. Houghton, Inorg. Chim.

Acta 2007, 360, 149.

[7] N. V. Belkova, P. O. Revin, L. M. Epstein, E. V. Vorontsov, V. I.

Bakhmutov, E. S. Shubina, E. Collange, R. Poli, J. Am. Chem.

Soc. 2003, 125, 11106.

[8] M. Baya, O. Maresca, R. Poli, Y. Coppel, F. Maseras, A.

Lledós, N. V. Belkova, P. A. Dub, L. M. Epstein, E. S. Shubina,

Inorg. Chem. 2006, 45, 10248.

Supporting Information (see also the footnote on the first page of

this article): Table of longitudinal relaxation times at different tem-

peratures for the hydride resonance of [Cp*Mo(dppe)H₂(O₂CCF₃)]

and figures showing IR spectra of this compound in various sol-

vents, the IR-spectroscopic study of its formation kinetics in tolu-

ene, and an Eyring plot of the rate constants in THF.

Eur. J. Inorg. Chem. 2007, 2813–2826

www.eurjic.org

2825

R. Poli, E. S. Shubina et al.

FULL PAPER

[9] M. S. Chinn, D. M. Heinekey, J. Am. Chem. Soc. 1990, 112,

5166.

[10] E. S. Shubina, N. V. Belkova, E. V. Bakhmutova, E. V. Vo-

rontsov, V. I. Bakhmutov, A. V. Ionidis, C. Bianchini, L. Mar-

velli, M. Peruzzini, L. M. Epstein, Inorg. Chim. Acta 1998, 280,

302.

[11] N. V. Belkova, E. S. Shubina, E. I. Gutsul, L. M. Epstein, I. L.

Eremenko, S. E. Nefedov, J. Organomet. Chem. 2000, 610, 58.

[12] S. Gründeman, S. Ulrich, H.-H. Limbach, N. S. Golubev, G. S.

Denisov, L. M. Epstein, S. Sabo-Etienne, B. Chaudret, Inorg.

Chem. 1999, 38, 2550.

[22]

[23]

These two resonances are sufficiently sharp for detection only

on a high-field instrument (³¹P resonance at 202 MHz).

E. S. Shubina, N. V. Belkova, A. N. Krylov, E. V. Vorontsov,

L. M. Epstein, D. G. Gusev, M. Niedermann, H. Berke, J. Am.

Chem. Soc. 1996, 118, 1105.

C. Reichardt, Solvents and Solvent Effects in Organic Chemis-

try, 3rd ed., WILEY-VCH, Weinheim, 2003.

S. O. Morgan, H. H. Lowry, J. Phys. Chem. 1930, 34, 2385.

S. Sharif, G. S. Denisov, M. D. Toney, H.-H. Limbach, J. Am.

Chem. Soc. 2006, 128, 3375.

C. Carvajal, K. J. Tolle, J. Smid, M. Szwarc, J. Am. Chem. Soc.

1965, 87, 5548.

D. J. Metz, A. Glines, J. Phys. Chem. 1967, 71, 1158.

A. V. Iogansen, Theor. Experim. Khim. 1971, 7, 302.

J. Z. Dega-Szafran, M. Grundwald-Wyspianska, M. Szafran,

Spectrochim. Acta, Ser. A 1991, 47, 543.

L. M. Epstein, A. N. Krylov, E. S. Shubina, J. Mol. Struct.

1994, 322, 345.

[24]

[25]

[26]

[27]

[13] N. V. Belkova, P. O. Revin, M. Besora, M. Baya, L. M. Epstein,

A. Lledós, R. Poli, E. S. Shubina, E. V. Vorontsov, Eur. J. Inorg.

Chem. 2006, 2192.

[28]

[29]

[30]

[14] B. Pleune, R. Poli, J. C. Fettinger, Organometallics 1997, 16,

1581.

[15] J. Andrieu, N. V. Belkova, M. Besora, E. Collange, L. M. Ep-

stein, A. Lledós, R. Poli, P. O. Revin, E. S. Shubina, E. V. Vo-

rontsov, Russ. Chem. Bull. 2003, 52, 2679.

[31]

[16] J. H. Shin, D. G. Churchill, G. Parkin, J. Organomet. Chem. [32] E. S. Shubina, A. N. Krylov, N. V. Belkova, L. M. Epstein,

2002, 642, 9.

A. P. Borisov, V. D. Mahaev, J. Organomet. Chem. 1995, 493,

[17] E. Le Grognec, R. Poli, P. Richard, Organometallics 2000, 19,

275.

3842.

[33] W. Klemperer, G. C. Pimentel, J. Chem. Phys. 1954, 22, 1399.

[34] J. P. Castaneda, G. S. Denisov, S. Y. Kucherov, V. M. Schreiber,

A. V. Shurukhina, J. Mol. Struct. 2003, 660, 25.

[35] G. V. Gusakova, G. S. Denisov, A. L. Smolyanskii, Optika i

Spektroskopiya 1972, 32, 922.

[18] M. A. A. F. D. C. T. Carrondo, M. J. Calhorda, M. B. Hurst-

house, Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1987,

43, 880.

[19] T. Steiner, J. Chem. Crystallogr. 1999, 29, 1235.

[20] L. M. Epstein, L. N. Saitkulova, G. P. Zol’nikova, D. N.

Kravtsov, Metalloorganicheskaya Khimiya 1991, 4, 1368.

[36] This band is slightly shifted relative to that obtained upon the

interaction of [Cp*Mo(dppe)H₃] with TFA, because of the ef-

–

[21] The starting material contains

a

minor amount of the

fect of the Na⁺and BF₄present in solution.

[Cp*Mo(κ¹-dppe)H₅] contaminant, which is characterized by a

doublet resonance at –3.09 ppm in the ¹H NMR spectrum and

by a singlet at 57.2 ppm in the ³¹P{¹H} NMR spectrum. This

is the typical by-product of the [Cp*Mo(dppe)H₃] synthesis^[13]

and is present in variable amounts in different samples. It is

rather difficult to completely remove this impurity, but its pres-

ence does not interfere with the observed transformations since

it is less basic than the trihydride complex (thus is not con-

sumed when using substoichiometric amounts of TFA),

whereas it is transformed into the trihydride complex with H₂

evolution by an excess amount of the acid.

[37] J. Sandström, Dynamic NMR Spectroscopy, Academic Press,

London, 1982.

[38] A. Altomare, M. Burla, M. Camalli, G. Cascarano, C. Giacov-

azzo, A. Guagliardi, A. Moliterni, G. Polidori, R. Spagna, J.

Appl. Crystallogr. 1999, 32, 115.

[39] G. M. Sheldrick, SHELXL97. Program for Crystal Structure

Refinement, University of Göttingen, Göttingen, Germany,

1997.

[40] L. J. Farrugia, J. Appl. Crystallogr. 1997, 30, 565.

Received: January 7, 2007

Published Online: April 17, 2007

2826

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 2813–2826

Article Doi

Solvent control in the protonation of [Cp*Mo(dppe)H3] by CF3COOH

DOI: 10.1002/ejic.200700021

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejic.200700021

Full text of DOI:10.1002/ejic.200700021

Products guided by the article

R&D Labs maybe for 1010693-68-1

Relevant to this article

Hot Product