A R T I C L E S
Ellis et al.
change of -28 kcal/mol using our revised value of ∆GH- for
CpMo(PMe3)(CO)2H. The previous ∆G value of -3.5 kcal/
mol would require an extremely large T∆S correction, 25 kcal/
(CO)2(PMe3)H, and Cp*Mo(CO)2(PMe3)H have been cross
referenced and can now be placed on a common absolute scale.
Experimental Section
-
mol. These results lead to the conclusion that the ∆GH values
2
3,36
reported previously for Mo, W, Nb, and Ta hydrides
are
General Procedures and Materials. NMR spectra were recorded
significantly in error. The origin of the error can be traced to
the presumption of an equilibrium for reactions that are far from
equilibrium.
on a Varian 400 MHz spectrometer. Proton chemical shifts were
3
1
recorded relative to residual protons in CD
chemical shifts are reported relative to an external sample of H
0.00 ppm). Isothermal titration calorimetry was performed on a
3
CN (1.93 ppm). P NMR
3
PO
4
(
Within experimental error, the hydride donor ability of CpMo-
Calorimetry Sciences Corporation ITC 4200 calorimeter. Solvents were
reagent grade and were purchased from Aldrich. Acetonitrile was
(
(
PMe3)(CO)2H is the same or greater than that of Cp*Mo-
PMe3)(CO)2H. This is not what one might expect based on
vacuum transferred from CaH
purchased from Aldrich. [CpRe(NO)(CO)
(CO)(CHO) were prepared using literature methods,
2
and stored in a glovebox. Reagents were
electronic effects, because C5Me5 is more electron-donating than
C5H5. However, if the stabilities of the products of hydride
2
](BF ) and CpRe(NO)-
4
3
7-39
as were
)Ph C]-
+
17,40,41
22,42,43
transfer, [CpMo(PMe3)(CO)2(CH3CN)] and [Cp*Mo(PMe3)-
BzNADH,
C
5
R
5
Mo(PMe
C](PF
).45
3
)(CO)
2
H,
[(p-Me
2
NC
H
6 4
2
+
44,45
(
CO)2(CH3CN)] , are considered, it is reasonable that the
6
(PF ),
and [(C
6
5
H )
3
6
Equilibration of BzNADH and [CpRe(NO)(CO)
etonitrile. BzNADH (0.009 g, 0.045 mmol) and [CpRe(NO)(CO)
BF ) (0.017 g, 0.040 mmol) were dissolved in CD CN (0.60 mL) in
2
](BF
4
) in Ac-
product containing the C5Me5 ring may be less stable than that
of the C5H5 ring because of steric interactions between the C5-
Me5 ring and coordinated CH3CN. This would tend to offset
the electronic effects. In the hydride transfer reactions of CpM-
2
]-
(
4
3
an NMR tube. Over the course of 18 h, this reaction came to equilibrium
+
with the formation of BzNAD and CpRe(NO)(CO)(CHO), as deter-
(
CO)2(L)H and Cp*M(CO)2(L)H studied by Bullock, it was
1
mined by comparison of the H NMR spectra with authentic samples
found that the kinetic hydricity of these compounds follows the
order expected on the basis of the electron-donating ability of
the cyclopentadienyl rings22 and not the order observed in this
study for the thermodynamic hydricity. This is likely due to
the fact that solvation of the Mo center occurs after the hydride
transfer reaction and the kinetic barrier arises from formation
17,37-41
prepared using literature methods.
Integrations of these spectra
were used to calculate an equilibrium constant of 830 (K
2
)
+
(
[BzNADH]/[BzNAD]) × ([Re(CO) ]/[ReCHO])) for reaction 2. The
+
reaction of BzNAD with CpRe(NO)(CO)(CHO) was studied in a
similar manner, and an equilibrium constant of 850 was obtained after
4 h.
+
+
Equilibration of Cp*Mo(PMe )(CO) H with BzNAD in Aceto-
of unsolvated [C5R5Mo(PMe3)(CO)2] and not [C5R5Mo-
3
2
+
6
nitrile. [BzNAD](PF ) (0.0155 g, 0.043 mmol) and Cp*Mo(PMe
3
)-
(
PMe3)(CO)2(CH3CN)] (where R ) H or Me). These results
(
CO) H (0.019 g, 0.052 mmol) were dissolved in CD
2
3
CN (0.7 mL) in
illustrate the important role that solvation can play in hydride
an NMR tube. The course of the slow reaction was followed by
transfer reactions. However, the fact that solvation plays a role
1
recording H NMR spectra weekly. The NMR tube was stored in the
-
does not decrease the utility of ∆GH values. Two compounds
glovebox, except for the time required to collect spectra, to protect the
sample from oxygen. The reaction came to equilibrium after 72 days,
-
having the same ∆GH value in any given solvent will have the
same tendency to donate a hydride ion regardless of whether
+
and the equilibrium constant (K
5
) ([BzNADH]/[BzNAD ]) ×
they form M-solvent bonds or not.
As seen from Table 1, the revised values of ∆GH for CpMo-
+
(
[Mo(CD
3
CN) ]/[MoH])) was evaluated using concentrations obtained
from integration of the final H NMR spectrum. An equilibrium constant
of 9 was obtained in this way. A similar reaction in the reverse direction
1
-
(PMe3)(CO)2H and Cp*Mo(PMe3)(CO)2H indicate that these
was not followed to equilibrium, but the formation of Cp*Mo(PMe
CO) H was observed.
Calorimetry. All solutions for calorimetry were prepared in a
glovebox. In a typical experiment, the reaction cell of the calorimeter
was purged with a stream of N for 15 min. Subsequently, it was loaded
by syringe with a 30 mM solution of [(p-Me NC )Ph C](PF ) in dry
acetonitrile (cell volume ) 1.3 mL), and stirring was commenced at
00 rpm. The reference cell of the calorimeter was loaded with neat
acetonitrile. Next, a 250-µL syringe containing a 10.3 mM solution of
CpMo(PMe )(CO) H in acetonitrile was inserted into the cell. After
3
)-
compounds are slightly better hydride donors than NADH model
compounds. They are also moderate hydride donors as compared
to other transition-metal hydrides studied to date, and they are
somewhat less hydridic than the formyl complexes shown in
Table 1. They are much better hydride donors than triaryl-
methane compounds as shown by comparison with the last two
entries of Table 1, and hydride transfer equilibria between these
two classes of compounds are not expected.
(
2
2
2
6
H
4
2
6
3
3
2
Summary
90 min of equilibration at 25.6 °C, 20 10-µL injections of the hydride
solution were added into the cell, with an 8-min delay between
injections. The first data point was discarded, and the average enthalpy/
mol for the subsequent 19 injections was 29.4 ( 0.2 kcal/mol. At least
three independent experiments were performed in this manner for each
The hydride donor abilities of BzNADH and CpMo(CO)2-
(PMe3)H have been reevaluated. Both of these compounds were
found to be significantly better hydride donors than previously
reported. For NADH analogues, this leads to a simple correction
of previously reported hydride donor abilities by -13 kcal/mol.
For the Mo, W, Ta, and Nb hydrides, the situation is more
complex, as our results for CpMo(PMe3)(CO)2H and Cp*Mo-
2
0
(37) Casey, C. P.; Andrews, M. A.; McAlister, D. R.; Rinz, J. E. J. Am. Chem.
Soc. 1980, 102, 1927.
(38) Tam, W.; Lin, G.-Y.; Wong, W.-K.; Kiel, W. A.; Wong, V. K.; Gladysz,
J. A. J. Am. Chem. Soc. 1982, 104, 141.
(
39) Sweet, J. R.; Graham, W. A. G. J. Am. Chem. Soc. 1982, 104, 2811.
(PMe3)(CO)2 indicate that the relative ordering of the com-
(40) Shinkai, S.; Hamada, H.; Kusano, Y.; Manabe, O. J. Chem. Soc., Perkin
Trans. 2 1979, 699.
2
3,36
pounds studied previously may also be incorrect.
of the studies described in this paper, the hydricities of [HM-
As a result
(41) Anderson, A. C.; Berkelnammer, G. J. Am. Chem. Soc. 1958, 80, 992.
(42) Kalck, P.; Pince, R.; Poilblanc, R.; Roussel, J. J. Organomet. Chem. 1970,
24, 445-452.
+
11-16
-
35
(diphosphine)2] complexes,
[HW(L)(CO)4] complexes,
(
43) Asdar, A.; Tudoret, M.-J.; Lapinte, C. J. Organomet. Chem. 1988, 349,
26
transition-metal formyl complexes, NADH derivatives, CpMo-
353-366.
(
44) Alavosus, T. J.; Sweigart, D. A. J. Am. Chem. Soc. 1985, 107, 985-987.
45) McKinley, S. V.; Rakshys, J. W., Jr.; Young, A. E.; Freedman, H. H. J.
Am. Chem. Soc. 1971, 93, 4715-4724.
(
(36) Sarker, N.; Bruno, J. W. Organometallics 2001, 20, 55-61.
2
742 J. AM. CHEM. SOC. VOL. 126, NO. 9, 2004
9