Photolysis of Tp′Rh(CNneopentyl)(PhNCNneopentyl) in the presence of ketones and esters: Kinetic and thermodynamic selectivity for activation of different aliphatic C-H bonds

Article abstract of DOI:10.1039/c9dt01802f

The active fragment [Tp′Rh(CNneopentyl)], generated from the precursor Tp′Rh(CNneopentyl)(PhNCNneopentyl), underwent oxidative addition of substituted ketones and esters resulting in Tp′Rh(CNneopentyl)(R)(H) complexes (Tp′ = tris-(3,5-dimethylpyrazolyl)borate). These C-H activated complexes underwent reductive elimination at varying temperatures (24-70 °C) in C₆D₆ or C₆D₁₂. Using previously established kinetic techniques, the relative Rh-C bond strengths were calculated. Analysis of the relative Rh-C bond strengths vs. C-H bond strengths shows a linear correlation with slope R_M-C/C-H = 1.22 (12). In general, α-substituents increase the relative Rh-C bond strengths compared to the C-H bond that is broken.

Full text of DOI:10.1039/c9dt01802f

Dalton
Transactions
View Article Online
View Journal
PAPER
Photolysis of Tp’Rh(CNneopentyl)
PhNCNneopentyl) in the presence of ketones and
(
Cite this: DOI: 10.1039/c9dt01802f
esters: kinetic and thermodynamic selectivity for
activation of different aliphatic C–H bonds†‡
Astrid M. Parsons
and William D. Jones
*
The active fragment [Tp’Rh(CNneopentyl)], generated from the precursor Tp’Rh(CNneopentyl)
PhNvCNneopentyl), underwent oxidative addition of substituted ketones and esters resulting in
Tp’Rh(CNneopentyl)(R)(H) complexes (Tp’ = tris-(3,5-dimethylpyrazolyl)borate). These C–H activated
complexes underwent reductive elimination at varying temperatures (24–70 °C) in C or C 12. Using
previously established kinetic techniques, the relative Rh–C bond strengths were calculated. Analysis of the
relative Rh–C bond strengths vs. C–H bond strengths shows a linear correlation with slope RM–C/C–H
.22 (12). In general, α-substituents increase the relative Rh–C bond strengths compared to the C–H bond
that is broken.
(
6
D
6
6
D
Received 29th April 2019,
Accepted 24th May 2019
=
DOI: 10.1039/c9dt01802f
1
rsc.li/dalton
3
strengths could be determined. Our lab found that the rela-
Introduction
tive thermodynamic stability (ΔG°) can be determined experi-
4
Transition metal-based strategies for C–H functionalization mentally by the thermodynamic cycle shown in Fig. 1. These
have revolutionized the field of organic chemistry. In particu- derived thermodynamic stabilities can be combined with the
lar, C–H functionalization processes can enable for the known C–H bond strengths to provide relative metal–carbon
efficient synthesis of biologically and pharmaceutically rele- bond strengths, which are useful for interpreting the factors
1
vant molecules and feedstock commodity chemicals. One that contribute to product distributions. Related methods to
major challenge for the activation and functionalization of measure these energetic parameters have been applied by
t
t
5
C–H bonds is the ability to control regioselectivity. Often, Wolczanski to ( Bu
3 2
SiO) ( BuSiNH)Ti(R) (R = alkyl, aryl,
t
t
6
modifications to the ligand scaffold provide a means to alter benzyl, and vinyl groups) and ( Bu NH) ( Bu -SiNv)Ta(R)
3
2
3
selectivity for different C–H bonds. Therefore, a detailed (R = Ph, Me, benzyl) systems, where a linear correlation
understanding of the underlying effects of ligand choice on between M–C and C–H bond energies was observed with slope
selectivity for C–H activation would allow for the rational RM–C/C–H = 1.1 and 1.0, respectively.
design of efficient catalytic systems.
In a similar study, Bercaw observed a trend with RM–C/C–H =
7
In general, the process of oxidative addition of C–H bonds 1.29 for Cp*
2
Sc-R complexes (R = alkyl, phenyl, alkynyl).
to coordinatively unsaturated metal centers is photochemically Bryndza and co-workers have shown linear 1 : 1 relationships
or thermally initiated. A number of early studies of C–H acti-
vation revealed the relative thermodynamic stabilities of pro-
ducts formed via oxidative addition to unsaturated metal
2
centers and, from these, the relative metal–carbon bond
Department of Chemistry, University of Rochester, Rochester, New York 14627, USA.
E-mail: jones@chem.rochester.edu
†
This paper is dedicated to our collaborator and friend Prof. Robin Perutz FRS
on the occasion of his 70th birthday, in recognition of his many important and
creative contributions to C–H bond activation.
‡
Electronic supplementary information (ESI) available: Experimental procedures
and characterization data for all reactions. DFT calculated C–H bond strengths.
See DOI: 10.1039/c9dt01802f
Fig. 1 Free energy diagram for competitive C–H activation.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
1
7
between M–X and H–X bond strengths for two different giving rise to C–H oxidative addition products. Previously, it
systems, Cp*Ru(PMe ) (X) and (dppe)Pt(Me)(X), where X is was shown that photolysis of 1 in neat acetone results in the
3
2
8
OH, OR, NR
2
, PR
2
3 2 3
, SiR , and SH. These correlations appear to clean formation of Tp′Rh(CNneopentyl)(CH C(vO)CH )(H)
be unaffected by steric differences. Marks also noted a nearly with a hydride doublet at δ −14.78 featuring a small rhodium–
1
6
linear correlation of Zr–X and Hf–X bond strengths with X–H hydride coupling constant (d, JRh–H = 19.8 Hz). No evidence
9
bond strengths.
for an O-bound enolate was observed. In this prior report, only
Eisenstein and Perutz have examined the above titanium this one example of ketone activation was reported. Herein, we
and Tp′Rh systems computationally, where both M–C and C–H investigate the activation of substituted ketones and esters by
bond strengths can be calculated using DFT. They found rhodium complex, 1. Scheme 1 summarizes the major pro-
reasonably good correlations for the substrates examined (all ducts in the reactions with ketones.
hydrocarbons), with the exception of benzyl and allyl, which
lay above the correlation line. They also found good corre- change from bright to pale yellow after 15 minutes at 0 °C.
Photolysis of 1 in neat diethyl ketone 2a results in a color
1
0
1
lations with slopes in the range 1.9–3.0 for the activation of
H NMR spectroscopic analysis of the reaction mixture in C
reveals the formation of one major species, Tp′Rh
Landis has also summarized and compared a number of (CNneopentyl)(CH CH C(vO)CH CH )(H) (3a), with a hydride
systems where some correlation between M–X and X–H resonance at δ −14.89 (d, JRh–H = 24.0 Hz) (Scheme 1). This
6 6
D
1
1
polyfluoroarenes with a variety of metal complexes (Zr–Ni).
2
2
2
3
1
2
(X includes C) bond strengths has been observed.
magnitude of a Rh–H coupling constant is typical for unsubsti-
Our group has been interested in studying the effect of tuted alkyl hydride species, indicating activation of a methyl
4
common organic functional groups on C–H activation reac- C–H bond rather than an α-keto C–H bond activation. Small
tions using the [Tp′Rh(CNneopentyl)] fragment, which is quantities of o-, m-, and p-carbodiimide activation products
cleanly generated by irradiation of Tp′Rh(CNneopentyl) are also observed as 3 doublets near δ −13.40 and −13.70. The
1
(PhNCN-neopentyl) 1. We have discovered that activation of
H NMR spectrum of the hydride region also shows small
1
-chloropentane and 3-chloropentane results in exclusive C–H
1
3
activation of the terminal methyl groups. When examining
the activation of 2-chloropentane, a mixture of 4-chloropentyl
activation product and Tp′Rh(CNneopentyl)HCl arising from
β-chloride elimination of the 2-chloropentyl activation product
was observed. Studying the activation of various 1-chloro-
alkanes (C –C ) revealed that chlorine substituents have a dra-
1
5
matic effect on the stabilization of the resulting alkyl hydride
1
4
species formed. In a separate study, it was concluded that
reductive elimination of hydrocarbons featuring α-cyano sub-
1
5
stituents required elevated temperatures. The observed stabi-
lization effect was attributed not to the strengthening of the
Rh–C bond due to the electron withdrawing character of the
cyano group, but rather to the formation of a weak α-cyano
C–H bond in the transition state for reductive elimination.
More recently, it was concluded that aryl, vinyl, alkynyl, alkoxy,
CN, and keto α-substitution on Rh-methyl complexes results
in a weakening of the Rh–CH
2
X bond compared to the
Rh–methyl bond. However, when considering the relative
C–H bond strengths (methane vs. H–CH X), it is clear that sub-
1
6
2
stituents have a positive effect on strengthening the metal–
carbon bond. Given these results and in view of the impor-
tance of ketones and esters as building blocks, we became
interested in further understanding how substitution on a
variety of simple ketones and esters affects selectivity for C–H
bond activation.
Results and discussion
C–H activation of ketones
Irradiation of Tp′Rh(CNneopentyl)(PhNCNneopentyl) (1) pro-
vides access to the coordinatively unsaturated fragment [Tp′Rh
Scheme 1 Major products from reaction of [Tp’Rh(CNR)] with various
(CNneopentyl)] which readily reacts with most hydrocarbons ketones.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
1
Fig. 2 H NMR spectrum (500 MHz) of the hydride region of the reac-
tion of 1 with (a) diethyl ketone and (b) methyl ethyl ketone. All spectra
recorded in C D . * indicates o-, m-, and p-carbodiimide activation pro-
6 6
ducts. The small doublet at δ −14.81 (JRh–H = 19.9 Hz) in (a) is assigned
to C–H activation of a trace impurity in the diethyl ketone (e.g.,
acetone).
doublets (∼5%) at δ −14.53 and −15.30 with JRh–H ≈ 22 Hz,
which could be attributed to diastereomers resulting from
α-keto C–H activation, but their low abundance makes this
assignment tentative (Fig. 2a). Attempts to isolate and charac-
terize 3a by treating with either CCl
4 3
or CHBr were
unsuccessful.
Interestingly, activation of methyl ethyl ketone results in
the rapid formation of two major species 3b and 3b′. The
larger hydride resonance at δ −14.79 (d, JRh–H = 20.5 Hz) is
1
Fig. 3 H NMR spectrum (500 MHz) of the hydride region of the reac-
tion of 4 with (a) cyclopentanone, (b) methyl cyclopropyl ketone, and (c)
assigned to Tp′Rh(CNR)(CH C(O)CH CH )(H) (3b). This small
2
2
3
4-phenyl-2-butanone. All spectra recorded in C D12. * indicates o-, m-,
6
Rh–H coupling constant is typical for α-substituted and p-carbodiimide activation products.
1
6
alkyl hydride complexes. Much like diethyl ketone activation,
3
b′ is assigned to Tp′Rh(CNR)(CH CH C(O)CH )(H) which
2
2
3
features a hydride resonance at δ −14.90 (d, JRh–H = 24.0 Hz)
with a larger coupling constant for activation of the β-methyl
Activation of cyclopropyl methyl ketone 2d via exchange
C–H bond. Also, two small doublets (∼5%) are seen at with Tp′Rh(CNneopentyl)(C D )(D) results in the formation of
6
11
δ −14.54 and −15.30 with JRh–H ≈ 23 Hz that could arise from major product 3d with a coupling constant of JRh–H = 19.5 Hz,
the diastereomers formed from α-ethyl C–H activation indicative of C–H activation of the methyl group. The majority
(
Fig. 2b).
For higher boiling substrates that could not be readily C–H activation (Fig. 3b). Only traces of products that may indi-
removed under vacuum without heating, selectivity for C–H cate cyclopropyl C–H activation are seen (c.f. for activation of
of the remaining mass balance was composed of carbodiimide
1
18
bond activation was investigated using the cyclohexyl deuter- cyclopropane, H NMR(C D ): δ −14.89, d, J
= 25 Hz).
6
6
Rh–H
ide derivative, Tp′Rh(CNneopentyl)(C
6
D
11)D, 4, which serves as
6
Irradiation of 1 in C D12 followed by exchange with
a thermal precursor to the reactive intermediate [Tp′Rh 4-phenyl-2-butanone 2e yields two major products. One
CNneopentyl)]. Activation of cyclopentanone 2c gave many hydride resonance at δ −15.17 (d, JRh–H = 19.5 Hz) indicates
(
C–H activated products alongside carbodiimide activation. The activation of the methyl group (3e). A second larger doublet is
identity of the major product 3c was assigned based on the observed at δ −14.17 (JRh–H = 24.0 Hz) and is consistent with
coupling constant observed, JRh–H = 20.4 Hz, indicating α-keto C–H activation of the phenyl group. The exact isomer of this
C–H activation. Hydride resonances for additional minor C–H oxidative addition product could not be readily identified, as
activation products appeared at δ −15.19 (JRh–H = 21.1 Hz) and activations of o-, m-, or p-C–H bonds are possible (Fig. 3c), yet
δ −15.29 (JRh–H = 22.5 Hz). Presumably, these resonances only one doublet is observed.
correspond to either β-C–H activation of cyclopentanone or a
6
Exchange of Tp′Rh(CNneopentyl)(C D11)(D) with 4-chloro-2-
diastereomer of the α-keto activation, although these assign- butanone (2f) led to the formation of methyl activation
ments are tentative (Fig. 3a).
product 3f which is accompanied by carbodiimide activation
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
and an unexpected hydride resonance at δ −14.04 (d, JRh–H
=
Table 1 Data for major products from the photolysis of 1 in various
substrates
a
1
0.5 Hz) (Fig. 3d). The data for this latter product are very
similar to those reported for Tp′Rh(CNneopentyl)HCl (−13.40,
1
4
d, J = 11.5 Hz) in C
solvent shift in the position of the resonance. Therefore, the
chemical shift of Tp′Rh(CNneopentyl)HCl in C 12 was con-
firmed by photolysis of 1 and 2-chloropropane in C
6
D
6
.
6
It is likely that the C D12 induces a
6
D
6 12
D .
Formation of this product is consistent with α-keto activation
followed by rapid β-chloride elimination, as observed with
1
4
many other chloroalkanes.
C–H activation of esters
Next, we sought to investigate the regioselectivity for C–H acti-
vation in the presence of ester containing substrates
b
c
1
Entry
Product
3
(%)
Hydride resonance (δ)
JRh–H (Hz)
(Scheme 2 and Table 1, entries 8–12). H NMR analysis of the
photolysis of 1 in methyl acetate shows activation of both
1
2
3
4
5
6
7
8
9
3a
3b
3b′
3c
3d
3e
3f
3g
3g′
3h
3i
88
57
37
36
60
33
34
49
45
85
89
−14.89
−14.79
−14.90
24.0
20.5
24.0
20.4
19.5
19.5
19.5
20.2
24.1
23.9
24.1
α-CH (3g) and methoxy (3g′) C–H bonds (Fig. 4a). Tp′Rh(CNR)
3
(
(
−
CH
d, JRh–H = 20.2 Hz) while 3g′ shows a hydride resonance at δ
14.50 (d, JRh–H = 24.1 Hz). Surprisingly, when the aliphatic
2
C(O)OCH
3
)(H) (3g) shows a hydride resonance at δ −14.55
d
−14.82
d
−15.02
d
−15.17
d
−15.16
chain length is extended as in methyl butyrate (2h), almost
exclusive activation of the butyrate methoxy C–H bond (3h) is
observed as a doublet at δ −14.54 with J = 23.9 Hz (Table 1,
entry 10) (Fig. 4b). Additional evidence for this assignment is
the long half-life for reductive elimination for this product
−14.55
−14.50
−14.54
−14.24
1
1
0
1
a
b
Samples were irradiated for 15 min at 0 °C. Yields determined via
1
integration of H NMR spectra vs. hexamethyldisiloxane as an internal
standard. Hydride resonance reported in C
reported in C D .
6 12
(
vide infra, t1/2 = 37 h, vs. 12 h for an alkyl hydride product).
c
d
6
D
6
.
Hydride resonance
Furthermore, photolysis of 1 in the presence of dimethyl
carbonate produced the expected oxidative addition product 3i
at δ −14.24 (d, JRh–H = 24.1 Hz) as the sole product (Fig. 4c).
The observed coupling constant is consistent with that of 3g′
and 3h.
Substrates bearing C–Br bonds were investigated, however,
1
no C–H activation products were observed by H NMR spec-
troscopy (eqn (1)). Attempts to characterize the species formed
by treating with bromoform yielded dibrominated species 5.
Lastly, C–H activation of aromatic ketones was attempted, but
selectivity was poor and many aromatic C–H bonds alongside
α-methyl activation were observed. Due to the many C–H
1
Fig. 4 H NMR spectrum (500 MHz) of the hydride region of the reac-
tion of esters with 1: (a) methyl acetate, (b) methyl butyrate, and (c) di-
6 6
methylcarbonate. All spectra recorded in C D . * indicates o-, m-, and
Scheme 2 Products from reaction of [Tp’Rh(CNR)] with various esters.
p-carbodiimide activation products.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
bonds activated, analysis of these reaction mixtures was chal- required elevated temperature and was faster than 4-phenyl-2-
lenging and not pursued. Table 1 summarizes the data for the butanone reductive elimination from 3e. Lastly, reductive
reactions of all ketones and esters with 1.
elimination of cyclopentanone from 3c at 40 °C was relatively
fast, suggesting that sterics associated with a secondary
carbon–rhodium bond may play a factor in thermodynamic
stability.
ð1Þ
Competitive kinetic selectivities
Next, the relative selectivity of the coordinatively unsaturated
fragment [Tp′Rh(CNneopentyl)] for C–H bond activation was
determined in a mixture of two substrates (Table 3). Samples
were irradiated for a short time so that the product ratio rep-
resents the kinetic products of the reaction mixture. The rela-
Kinetics of reductive eliminations
2 1
tive competitive rates (k /k ) were determined on the basis of
the relative areas of the corresponding resonances by H NMR
Rates for the reductive elimination of Tp′Rh(CNneopentyl)(R)
H) complexes in either C or C 12 were determined by
1
(
6
D
6
6
D
spectroscopy using eqn (2), where I /I is the integration area
2
1
monitoring the first-order disappearance of the Rh–H reso-
1
of the hydride resonances and n
1
/n
2
is the mole ratio of the
nance by H NMR spectroscopy (Table 2). Reductive elimin-
two substrates in question (subscript 1 refers to the competing
substrate and subscript 2 refers to benzene). Furthermore, the
ation of α-substituted alkyl hydride complex 3b was slower
when compared to 3a and 3b′ at room temperature. Reductive
elimination of methoxy activated C–H bond in 3g′ was 2.4×
times faster than α-substituted 3g at 60 °C. This suggests that
α substitution provides additional stability to the resulting
hydride complex. While extending the aliphatic chain length
in 3h (compared to 3g′) results in near exclusive methoxy C–H
activation, the rate of reductive elimination of methyl butyrate
from 3h is similar to that of methyl acetate from 3g′ at 60 °C.
In contrast, reductive elimination of dimethyl carbonate from
‡
oa
difference in free energies of activation ΔΔG can be calcu-
lated using eqn (3).
Since reductive elimination experiments were conducted at
‡
varying temperatures (Table 2), the ΔG for benzene reductive
elimination at each respective temperature was calculated
1
7
‡
using known activation parameters. Combining ΔG for
re
‡
‡
benzene, ΔG for the substrate of interest, and ΔΔG , ΔG°
re
oa
can be calculated (eqn (4), Table 4). Finally, the relative Rh–C
bond strengths of C–H activated substrates compared to
D(Rh–Ph) were calculated using eqn (5), which includes the
assumption that ΔG° = ΔH° − RT ln(H/H′), where H/H′ is the
ratio of the number of available hydrogen atoms on the
3
i is slow (t1/2 = 111.13 h), showing a strong electronic effect of
the carbonate.
The reductive elimination of 4-chloro-2-butanone from 3f
was conducted at room temperature and a rate of 8.21 (46) ×
−
6
−1
1
0
s
was observed. However for 4-phenyl-2-butanone, elev-
ated temperature was necessary to observe an appreciable
decrease in the corresponding hydride resonance. Reductive
elimination of methyl cyclopropyl ketone from 3d also
Table 3 Kinetic selectivity data determined from competition
experiments
a
Table 2 Rates of reductive elimination of ketones and esters from Tp’
a
6 6
Rh(CNR)(R)(H) (3a–3i) in C D
b
‡
oa
c
(kcal mol−1
)
Entry
Substrates
k /k
2 1
ΔΔG
Product
CH C(vO)CH
T (°C)
k
re (s−
1
)
t
1/2 (h) ΔG (kcal mol−1
‡
re
)
d
1
Benzene: acetone
Benzene: 2a
Benzene: 2b
Benzene: 2b′
Benzene: 2c
Benzene: 2d
Benzene: 2e
Benzene: 2f
Benzene: 2g
Benzene: 2g′
Benzene: 2h
Benzene: 2i
3.71 (19)
1.37 (7)
1.97 (10)
1.76 (9)
20.85 (104)
10.06 (50)
2.44 (12)
—
6.11 (31)
7.51 (38)
17.36 (87)
3.21 (15)
0.73 (3)
0.19 (3)
0.40 (3)
0.34 (3)
1.79 (3)
1.36 (3)
0.53 (3)
—
1.07 (3)
1.19 (3)
1.68 (3)
0.69 (3)
1.11 (2) × 10−
1.65 (7) × 10−
9.51 (54) × 10
5
5
−
5
5
17.3
27.71 (1)
11.70 23.88 (3)
2
3
4
5
6
7
2
3
70
24
24
24
40
70
70
24
60
60
60
60
3
3
3
3
3
3
3
3
3
3
3
a
b
b′
c
d
e
7
202.42 25.57 (3)
10.82 23.84 (2)
18.67 25.49 (1)
24.88 28.19 (5)
36.00 28.45 (2)
23.45 24.29 (3)
90.38 28.21 (1)
37.86 27.63 (2)
36.98 27.61 (2)
111.13 28.34 (1)
−
1.78 (6) × 10
1.03 (1) × 10−
7.74 (52) × 10−
6
6
6
5.35 (12) × 10−
8
e
−
f
8.21 (46) × 10
9
−6
g
g′
h
i
2.13 (3) × 10
10
11
12
5.08 (12) × 10−
6
6
−
5.21 (12) × 10
1.73 (2) × 10−
6
a
b
Each sample was irradiated for 5 minutes at 0 °C. Errors in rate
a
‡
re
Errors are reported as standard deviation. Errors in ΔG are calcu- ratio estimated at 5% for proton NMR integrations, giving σ
G
= (RT/
−
1 c
lated from k as propagated errors, using σ
G
= (RT/kre)σ
k
. The errors are ratio)σratio = 0.05RT ≈ 0.03 kcal mol
.
A positive value denotes that
d
small because G is a log function of rate. Systematic errors are prob- benzene is kinetically favored. Irradiation done at 8 °C, see ref. 9.
ably larger and can be estimated as ±0.1 kcal mol− assuming 10%
1
e
Attempts to obtain kinetic selectivity data for 4-chloro-2-butanone
were unsuccessful.
error in k. Substrates 2c–f in C
6 12
D .
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 4 Kinetic and thermodynamic data for Tp’Rh(CNneopentyl)
a
(
R)(H)
No.
of H ΔΔG
‡
b
c
e
Product
Phenyl
ΔG°
D(R–H)
D
rel(Rh–C)
oa
d
6
6
6
3
3
4
3
3
3
3
3
3
6
0.00
0.00 (5)
0.21 (9)
3.97 (11)
2.42 (11)
4.21 (10)
3.67 (9)
0.27 (13)
112.9
0.00
2
CH C(vO)CH
3
0.71 (3)
0.17 (3)
0.31 (3)
0.37 (3)
1.65 (3)
1.25 (3)
96.00
98.34
91.34
98.94
85.34
91.22
90.22
90.67
93.59
95.95
94.63
96.83
−17.1
−18.5
−23.6
−17.8
−31.0
−21.5
−21.5
—
−19.0
−17.3
−19.1
−15.7
3
3
3
3
3
3
3
3
3
3
3
a
b
b′
c
d
e
0.48 (3) −0.76 (10)
f
—
—
g
g′
h
i
0.98 (3)
1.09 (3)
1.55 (3)
0.08 (9)
0.77 (10)
1.25 (10)
0.63 (3) −0.040 (9)
a
All values are in kcal mol−1
b
.
Error is the summation of the individ-
c
ual errors in associated with eqn (3). C–H bond strengths were calcu-
lated using B3LYP/6-311g. Benzene C–H bond strength is from ref.
2
d
Fig. 5 Plot of relative experimental M–C bond strengths vs. C–H bond
strengths for Tp’Rh(CNneopentyl)(R)(H). Hydrocarbon C–H bond
strengths were calculated using B3LYP/6-311g. Blue squares refer to
e
3.
D
rel(Rh–H) as defined in eqn (5) refers to the Rh–C bond strength
relative to the Rh–Ph bond strength.
2
α-keto or α-O functionalized Rh–CH X products. Red triangles refer to
β-keto Rh–CH CH C(vO)R and phenyl products.
2
2
substrates. This accounts for the statistical contribution to the
free energy.
2
activation of CH bonds is kinetically slow. Consequently, in
alkyl ketones and esters methyl C–H activation strongly
predominates.
ꢀ
ꢁꢀ
ꢁ
ꢀ
k2
k1
I2
I1
n1
n2
¼
ð2Þ
ð3Þ
ꢁ
‡
k2
k1
ΔΔGoa ¼ RT ln
Conclusions
‡
re
‡
oa
‡
re
ΔG° ¼ ΔG ðR′HÞ þ ΔΔG ꢀ ΔG ðPhHÞ
ð4Þ Photolysis of Tp′Rh(CNneopentyl)(PhNCNneopentyl) 1 in the
presence of aliphatic ketones and esters results in C–H acti-
vation products of the type Tp′Rh(CNneopentyl)(R)(H) 3, where
DrelðRhꢀCÞ ¼ ½ΔHðRhꢀRÞ ꢀ ΔHðRhꢀPhÞꢁ
°
¼
½DðRꢀHÞ ꢀ DðPhꢀHÞꢁ ꢀ ΔG
reactions of methyl groups dominate the observed products.
ð5Þ
ꢀ
ꢁ
6
H
Only traces of secondary, α-keto CH
The reductive elimination of RH from Tp′Rh(CNneopentyl)(R)
H) in C or C 12 was monitored and in combination with
2
activation are observed.
þ RT ln #
(
6
D
6
6
D
A plot of the relative Rh–C bond strengths vs. the C–H bond
strengths shows a linear correlation for the studied substrates
with α-keto or α-O CR substitutions (Fig. 5). The positive corre-
2
lation indicates that the Rh–C bond strengths could be inferred
from the corresponding C–H bond strengths in R–H. The
observed slope, RM–C/C–H = 1.22 (12), suggests that the Rh–C
bond strengths vary in proportion to the C–H bond strengths,
as was seen in the case of related systems where R
larger than unity (∼1.4).
competition experiments allowed for the determination of the
relative Rh–C bond strengths. In general, C–H activation pro-
ducts containing α-substituents provide additional stability to
the resulting hydride species. While α-ester substituents do
provide additionally stability, α-keto substituents result in
stronger Rh–C bonds. A positive correlation RRh–C/C–H = 1.22
was observed suggesting that the relative Rh–C bond strength
can be inferred directly from the relative C–H bond strength in
this class of substrates.
is also
M–C/C–H
1
6,19,20
All of the products in these reactions except for two (plus
benzene) possess either α-keto or –CH OR functional groups. The
2
two data points with β-keto groups with D(C–H) ≈ 98 kcal mol−
1
Materials and methods
General procedures
and benzene are seen to lie on a slightly lower line with RM–C/C–H
=
1
.27 (1), as would be expected for activation of a C–H bond with
16
no α-substitution.
All operations and routine manipulations were performed
Of note here is the kinetic preference for terminal methyl under a nitrogen atmosphere or on a high-vacuum line using
C–H bond activation even though α-keto C–H activation is modified Schlenk techniques. Diethyl ketone, ethyl methyl
expected to be thermodynamically preferred. The observation ketone, methyl acetate, methyl butyrate, dimethyl carbonate,
of only trace quantities of resonances that could be attributed 4-phenyl-2-butanone,
4-hydroxy-2-butanone,
cyclopropyl
to the latter are consistent with the general observation that methyl ketone, cyclopentanone, carbon tetrachloride, and bro-
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
moform were purchased from Sigma Aldrich. The synthesis of for instrumentation and chemicals, and the NSF for a
2
1
4
d
-chloro-2-butanone has been previously reported. Benzene- Graduate Fellowship (DGE-1419118 to A. M. P.). Andrew
(Cambridge Isotopes) was distilled under vacuum from a VanderWeide is acknowledged for assistance with DFT
6
dark purple solution of benzophenone ketyl and stored in a calculations.
Schlenk flask under nitrogen atmosphere. Cyclohexane-d12
(Cambridge Isotopes) was degassed by three freeze–pump–
thaw cycles and stored over activated 3 Å molecular sieves.
Other solvents were used directly from an Innovative
Technologies PS-MD-6 solvent system. The synthesis of Tp′Rh
Notes and references
1 (a) J. Wencel-Delord and F. Glorius, Nat. Chem., 2013, 5,
369–375; (b) D. J. Abrams, P. A. Provencher and
E. J. Sorensen, Chem. Soc. Rev., 2018, 47, 8925–8967.
2 (a) A. H. Janowicz and R. G. Bergman, J. Am. Chem. Soc.,
1983, 105, 3929–3939; (b) J. M. Buchanan, J. M. Stryker and
R. G. Bergman, J. Am. Chem. Soc., 1986, 108, 1537–1550;
(c) M. J. Wax, J. M. Stryker, J. M. Buchanan, C. A. Kovac and
R. G. Bergman, J. Am. Chem. Soc., 1984, 106, 1121–1122;
(d) W. D. Jones and F. J. Feher, J. Am. Chem. Soc., 1984, 106,
1650–1663; (e) W. D. Jones and F. J. Feher, Acc. Chem. Res.,
1989, 22, 91–100.
(
CNneopentyl)(PhNCN-neopentyl) (1) has been previously
2
2
reported.
All photolysis experiments were performed using a 200 W
Hg(Xe) arc lamp purchased from Oriel, which was fitted with a
water-filled IR filter and a 324 nm high pass filter. All experi-
1
13
ments were performed at 0 °C. All H and C NMR spectra
were collected on either a Bruker Avance 400 or Avance
5
(
00 MHz spectrometer. Chemical shifts are reported in ppm
δ) referenced to the residual solvent peaks of C (δ = 7.16)
12 (δ = 1.38). Elemental analysis was performed by the
6 6
D
6
and C D
University of Rochester using a PerkinElmer 2400 series II
elemental analyzer in CHN mode. All kinetic plots and least-
square error analysis were done using Microsoft Excel.
3 (a) S. P. Nolan, C. D. Hoff, P. O. Stoutland, L. J. Newman,
J. M. Buchanan, R. G. Bergman, G. K. Yang and
K. S. Peters, J. Am. Chem. Soc., 1987, 109, 3143–3145;
(
(
b) J. Halpern, Inorg. Chim. Acta, 1985, 100, 41–48;
c) E. P. Wasserman, C. B. Moore and R. G. Bergman,
Reactions in benzene-d
6
Science, 1992, 255, 315–318; (d) J. A. M. Simões and
J. L. Beauchamp, Chem. Rev., 1990, 90, 629–688;
Reactions were setup in a nitrogen filled glovebox. To an oven-
dried 5 mm J Young NMR tube was added 1 (6.9 mg,
(
e) C. P. Schaller, C. C. Cummins and P. T. Wolczanski,
0
.010 mmol). Next, 0.5 mL the corresponding R–H substrate
J. Am. Chem. Soc., 1996, 118, 591–611.
W. D. Jones and E. T. Hessell, J. Am. Chem. Soc., 1993, 115,
was added and the NMR tube was capped. The sample was
removed from the glovebox and irradiated at 0 °C for 15 min.
After 15 min, the reaction was removed from the ice bath and
the excess substrate was removed in vacuo. The resulting
4
5
6
7
8
9
5
54–562.
J. L. Bennett and P. T. Wolczanski, J. Am. Chem. Soc., 1997,
19, 10696–10719.
C. P. Schaller and P. T. Wolczanski, Inorg. Chem., 1993, 32,
31–144.
1
6 6
residue was dissolved in C D , heated if necessary, and the dis-
appearance of the hydride resonance was monitored over time
1
1
by H NMR spectroscopy.
A. R. Bulls, J. E. Bercaw, J. M. Manriquez and
M. E. Thompson, Polyhedron, 1988, 7, 1409–1428.
H. E. Bryndza, L. W. Fong, R. A. Paciello, W. Tam and
J. E. Bercaw, J. Am. Chem. Soc., 1987, 109, 1444–1456.
L. E. Schock and T. J. Marks, J. Am. Chem. Soc., 1988, 110,
Reactions in cyclohexane-d12
Reactions were setup in a nitrogen filled glovebox. To an oven-
dried 5 mm J Young NMR tube was added 1 (6.9 mg,
0
.010 mmol). Next, 1 drop the corresponding R–H substrate
7701–7715.
was added followed by 0.5 mL of C D . The NMR tube was
6
12
1
1
0 E. Clot, C. Mégret, O. Eisenstein and R. N. Perutz, J. Am.
Chem. Soc., 2006, 128, 8350–8257.
capped and the sample was removed from the glovebox and
irradiated at 0 °C for 15 min. After 15 min, the reaction was
removed from the ice bath and heated if necessary. The dis-
appearance of the hydride resonances was monitored over
1 (a) E. Clot, M. Besora, F. Maseras, C. Mégret, O. Eisenstein,
B. Oelckers and R. N. Perutz, Chem. Commun., 2003, 490–
491; (b) E. Clot, B. Oelckers, A. H. Klahn, O. Eisenstein and
R. N. Perutz, Dalton Trans., 2003, 4065–4074; (c) E. Clot,
C. Mégret, O. Eisenstein and R. N. Perutz, J. Am. Chem.
Soc., 2009, 131, 7817–7827.
2 J. Uddin, C. M. Morales, J. H. Maynard and C. R. Landis,
Organometallics, 2006, 25, 5566–5581.
3 A. J. Vetter and W. D. Jones, Polyhedron, 2014, 23, 413–
1
time by H NMR spectroscopy.
Conflicts of interest
1
1
1
The authors declare no competing financial interest.
417.
4 A. J. Vetter, R. D. Rieth, W. W. Brennessel and W. D. Jones,
J. Am. Chem. Soc., 2009, 131, 10742–10752.
Acknowledgements
The authors gratefully acknowledge funding from the 15 A. J. Vetter, R. D. Rieth and W. D. Jones, Proc. Natl. Acad.
U.S. Department of Energy (Grant. No. FG02-86ER-13569)
Sci. U. S. A., 2007, 104, 6957–6962.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
1
1
1
1
6 Y. Jiao, M. E. Evans, J. Morris, W. W. Brennessel and 20 Y. Jiao, W. W. Brennessel and W. D. Jones, Chem. Sci., 2014,
W. D. Jones, J. Am. Chem. Soc., 2013, 135, 6994–7004. 5, 804–812.
7 W. D. Jones and E. T. Hessell, J. Am. Chem. Soc., 1992, 114, 21 R. C. Simon, B. Grischek, F. Zepeck, A. Steinreiber, F. Belaj
087–6095. and W. Kroutil, Angew. Chem., Int. Ed., 2012, 51, 6713–6716.
8 D. D. Wick, T. O. Northcutt, R. J. Lachicotte and 22 E. T. Hessell and W. D. Jones, Organometallics, 1992, 11,
W. D. Jones, Organometallics, 1998, 17, 4484–4492. 1496–1505.
9 Y. Jiao, J. Morris, W. W. Brennessel and W. D. Jones, 23 Y.-R. Luo, Comprehensive Handbook of Chemical Bond
J. Am. Chem. Soc., 2013, 135, 16198–16212. Energies, CRC Press, Boca Raton, FL, 2007.
6
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019