5 Thermodynamic Determination of Rhodium–Carbon
Bond Strengths in Tp
0 Rh(PMe 3 )(R)H
As was done previously, the kinetics of reductive elimination, combined with
kinetic competition data, were used to obtain rhodium–carbon bond strengths
with [Tp
0 Rh(PMe 3 )] as the metal fragment. Thermolysis of each compound in
C 6 D 6 at 30
C was found to follow first-order reductive elimination kinetics, giving
Tp
0 Rh(PMe 3 )(C 6 D 5 )D (6-d 2 ). The only exception was the 2-butynyl hydride
Tp
0 Rh(PMe 3 )(CH 2 CCCH 3 )H, which gave the η
2 -butyne complex as confirmed
by X-ray crystallography. This complex could therefore not be employed in the
thermodynamic analysis. In comparison with the earlier case with Tp
0 Rh
(neopentyl)(CH 2 CCCH 3 )H, the elimination of 2-butyne cleanly led to the formation of 2-d 6 . Apparently the stronger donor PMe 3 allows for significant stabilization
of the π-bound alkyne complexes.
Some of the compounds underwent reductive elimination far too slowly at 30
C
for convenient measurement (e.g., alkynes), and therefore, they were conducted at
elevated temperatures (140
C). In addition, since 6-d 2 is unstable at this temperature, C 6 F 5 H was added to trap the metal fragment following reductive elimination.
To compare these barriers to those of the reductive elimination of 6, the temperature
dependence of the rate of elimination for 6 in C 6 D 6 was measured, giving activation
parameters ΔH
{
¼ 32.6 Æ 3.3 kcal mol
À1 and ΔS
{
¼ 10.9 Æ 0.2 kcal mol
À1 K
À1 .
Using these data, the barrier heights could be compared at the same temperature.
Kinetic competitions between a substrate and C 6 H 6 were accomplished by
irradiation of a solution of 4 in a mixture of the two substrates. The samples were
irradiated for only a short time to avoid problems arising from secondary photolysis
of the products. The ratio of the two products could be easily determined by
1 H
NMR spectroscopy, giving the value for ΔΔG
{ . Competition data for methane was
measured vs. pentane and then referred to benzene using the competition between
pentane and benzene: k PhH =k CH 4 ¼ k PhH =k pentane
À
Á
k pentane =k CH 4
À
Á
.
As described above for [Tp
0 Rh(CNneopentyl)], the analysis of the data in Table 2
as in Fig. 1 and using Eq. 7 allows the determination of D rel (Rh–C) for a large
number of substrates. These Rh–C bond strengths can be plotted vs. the
corresponding C–H bond strengths to give the overall trend as shown in Fig. 5.
As before two trends clearly emerge. The first trend is seen joining the unsubstituted
hydrocarbons with a slope of 1.54(4). This compares to the value seen with
CNneopentyl as the ancillary ligand of 1.38(3). The effect of replacing the strong
isocyanide π-acceptor with the strong PMe 3 σ-donor is to increase the slope of the
line. This corresponds to a “stretching out” of the range of Rh–C bond strengths
with the better σ-donor ligand; i.e., the PMe 3 derivative shows a wider range of
selectivity. The second trend seen is in the methyl-substituted derivatives Rh–
CH 2 X. Again, a parallel line is observed with a slope of 1.71(8), which compares
to the slope seen with L ¼ CNneopentyl of 1.40(14). The line is offset vertically by
about 8 kcal/mol, very similar to the values seen with L ¼ CNneopentyl. The range
80
W.D. Jones
Bond Strengths in Tp
0 Rh(PMe 3 )(R)H
As was done previously, the kinetics of reductive elimination, combined with
kinetic competition data, were used to obtain rhodium–carbon bond strengths
with [Tp
0 Rh(PMe 3 )] as the metal fragment. Thermolysis of each compound in
C 6 D 6 at 30
C was found to follow first-order reductive elimination kinetics, giving
Tp
0 Rh(PMe 3 )(C 6 D 5 )D (6-d 2 ). The only exception was the 2-butynyl hydride
Tp
0 Rh(PMe 3 )(CH 2 CCCH 3 )H, which gave the η
2 -butyne complex as confirmed
by X-ray crystallography. This complex could therefore not be employed in the
thermodynamic analysis. In comparison with the earlier case with Tp
0 Rh
(neopentyl)(CH 2 CCCH 3 )H, the elimination of 2-butyne cleanly led to the formation of 2-d 6 . Apparently the stronger donor PMe 3 allows for significant stabilization
of the π-bound alkyne complexes.
Some of the compounds underwent reductive elimination far too slowly at 30
C
for convenient measurement (e.g., alkynes), and therefore, they were conducted at
elevated temperatures (140
C). In addition, since 6-d 2 is unstable at this temperature, C 6 F 5 H was added to trap the metal fragment following reductive elimination.
To compare these barriers to those of the reductive elimination of 6, the temperature
dependence of the rate of elimination for 6 in C 6 D 6 was measured, giving activation
parameters ΔH
{
¼ 32.6 Æ 3.3 kcal mol
À1 and ΔS
{
¼ 10.9 Æ 0.2 kcal mol
À1 K
À1 .
Using these data, the barrier heights could be compared at the same temperature.
Kinetic competitions between a substrate and C 6 H 6 were accomplished by
irradiation of a solution of 4 in a mixture of the two substrates. The samples were
irradiated for only a short time to avoid problems arising from secondary photolysis
of the products. The ratio of the two products could be easily determined by
1 H
NMR spectroscopy, giving the value for ΔΔG
{ . Competition data for methane was
measured vs. pentane and then referred to benzene using the competition between
pentane and benzene: k PhH =k CH 4 ¼ k PhH =k pentane
À
Á
k pentane =k CH 4
À
Á
.
As described above for [Tp
0 Rh(CNneopentyl)], the analysis of the data in Table 2
as in Fig. 1 and using Eq. 7 allows the determination of D rel (Rh–C) for a large
number of substrates. These Rh–C bond strengths can be plotted vs. the
corresponding C–H bond strengths to give the overall trend as shown in Fig. 5.
As before two trends clearly emerge. The first trend is seen joining the unsubstituted
hydrocarbons with a slope of 1.54(4). This compares to the value seen with
CNneopentyl as the ancillary ligand of 1.38(3). The effect of replacing the strong
isocyanide π-acceptor with the strong PMe 3 σ-donor is to increase the slope of the
line. This corresponds to a “stretching out” of the range of Rh–C bond strengths
with the better σ-donor ligand; i.e., the PMe 3 derivative shows a wider range of
selectivity. The second trend seen is in the methyl-substituted derivatives Rh–
CH 2 X. Again, a parallel line is observed with a slope of 1.71(8), which compares
to the slope seen with L ¼ CNneopentyl of 1.40(14). The line is offset vertically by
about 8 kcal/mol, very similar to the values seen with L ¼ CNneopentyl. The range
80
W.D. Jones