8 Conclusions
This chapter presented studies of C–H activation of sp, sp
2 , and sp
3 hybridized
carbon containing substrates by reactive [Tp
0 RhL] precursors (L ¼ CNneopentyl,
PMe 3 , P(OMe) 3 ). By using the relationship between the kinetics of hydrocarbon
reductive elimination and the competition for C–H activation, the thermodynamics
for the various activations could be determined. Knowledge of the driving force for
a reaction (ΔG
0 ) allows the determination of the relative rhodium–carbon bond
energy. Examination of the trends in M–C bond strength showed four important
features.
First, for the parent hydrocarbons (alkanes, alkenes, alkynes), there is a linear
relationship between the rhodium–carbon bond strength and the strength of the
carbon–hydrogen bond being broken. Second, the range of rhodium–carbon bond
strengths exceeds the range of carbon–hydrogen bond strengths by 38–55%
depending on the spectator L ligand, resulting in a slope for this linear correlation
that is greater than one. This is consistent with a product-driven equilibrium. Third,
for substituted methyl derivatives (i.e., Rh–CH 2 X, X ¼ F, Cl, CN, OR, Ph, vinyl,
keto), the Rh–C bond is about 7 kcal/mol stronger than what would be expected
based upon the C–H bond being broken. This “extra” bond strength was attributed
to an increase in the ionic character of the metal–carbon bond. Fourth, it was found
that a σ-donating L ligand increases the slope of the M–C/C–H correlation, whereas
π-acceptors decrease this slope.
Finally, DFT calculations of these same systems with the same substrates show
good agreement with the experimentally observed trends. For these systems, however, the DFT calculations overestimate the slopes of the correlations by about 10–
12%.
Acknowledgment We thank the U.S. Department of Energy, grant FG02-86ER13569, for their
support of this work.
References
1. Hessell ET, Jones WD (1992) Organometallics 11:1496
2. Blaha JP, Dewan JC, Wrighton MS (1986) Organometallics 5:899
3. Bradley MG, Roberta DA, Geoffroy GL (1981) J Am Chem Soc 103:379
4. Geoffroy GL, Bradley MG (1978) Inorg Chem 17:2410
5. Klahn-Oliva AH, Sineer RD, Sutton D (1986) J Am Chem Soc 108:3107
6. Ghosh CK, Graham WAG (1987) J Am Chem Soc 109:4126
7. Jones WD, Hessell ET (1992) J Am Chem Soc 114:6087
8. Jones WD, Hessell ET (1993) J Am Chem Soc 115:554
9. Jones WD, Wick DD (1999) Organometallics 18:495
10. Wick DD, Northcutt TO, Lachicotte RJ, Jones WD (1998) Organometallics 17:4484
11. Northcutt TO, Wick DD, Vetter AJ, Jones WD (2001) J Am Chem Soc 123:7257
12. Wick DD, Reynolds KA, Jones WD (1999) J Am Chem Soc 121:3974
88
W.D. Jones
This chapter presented studies of C–H activation of sp, sp
2 , and sp
3 hybridized
carbon containing substrates by reactive [Tp
0 RhL] precursors (L ¼ CNneopentyl,
PMe 3 , P(OMe) 3 ). By using the relationship between the kinetics of hydrocarbon
reductive elimination and the competition for C–H activation, the thermodynamics
for the various activations could be determined. Knowledge of the driving force for
a reaction (ΔG
0 ) allows the determination of the relative rhodium–carbon bond
energy. Examination of the trends in M–C bond strength showed four important
features.
First, for the parent hydrocarbons (alkanes, alkenes, alkynes), there is a linear
relationship between the rhodium–carbon bond strength and the strength of the
carbon–hydrogen bond being broken. Second, the range of rhodium–carbon bond
strengths exceeds the range of carbon–hydrogen bond strengths by 38–55%
depending on the spectator L ligand, resulting in a slope for this linear correlation
that is greater than one. This is consistent with a product-driven equilibrium. Third,
for substituted methyl derivatives (i.e., Rh–CH 2 X, X ¼ F, Cl, CN, OR, Ph, vinyl,
keto), the Rh–C bond is about 7 kcal/mol stronger than what would be expected
based upon the C–H bond being broken. This “extra” bond strength was attributed
to an increase in the ionic character of the metal–carbon bond. Fourth, it was found
that a σ-donating L ligand increases the slope of the M–C/C–H correlation, whereas
π-acceptors decrease this slope.
Finally, DFT calculations of these same systems with the same substrates show
good agreement with the experimentally observed trends. For these systems, however, the DFT calculations overestimate the slopes of the correlations by about 10–
12%.
Acknowledgment We thank the U.S. Department of Energy, grant FG02-86ER13569, for their
support of this work.
References
1. Hessell ET, Jones WD (1992) Organometallics 11:1496
2. Blaha JP, Dewan JC, Wrighton MS (1986) Organometallics 5:899
3. Bradley MG, Roberta DA, Geoffroy GL (1981) J Am Chem Soc 103:379
4. Geoffroy GL, Bradley MG (1978) Inorg Chem 17:2410
5. Klahn-Oliva AH, Sineer RD, Sutton D (1986) J Am Chem Soc 108:3107
6. Ghosh CK, Graham WAG (1987) J Am Chem Soc 109:4126
7. Jones WD, Hessell ET (1992) J Am Chem Soc 114:6087
8. Jones WD, Hessell ET (1993) J Am Chem Soc 115:554
9. Jones WD, Wick DD (1999) Organometallics 18:495
10. Wick DD, Northcutt TO, Lachicotte RJ, Jones WD (1998) Organometallics 17:4484
11. Northcutt TO, Wick DD, Vetter AJ, Jones WD (2001) J Am Chem Soc 123:7257
12. Wick DD, Reynolds KA, Jones WD (1999) J Am Chem Soc 121:3974
88
W.D. Jones