Mechanistic Model for Enantioselective Intramolecular Alkene Cyanoamidation via Palladium-Catalyzed C-CN Bond Activation

Article abstract of DOI:10.1021/acs.joc.7b00196

We studied key aspects of the mechanism of Pd-catalyzed C-CN bond activation and intramolecular enantioselective alkene cyanoamidation. An Abboud-Abraham-Kamlet-Taft (AAKT) linear solvation energy relationship (LSER) model for enantioselectivity was established. We investigated the impact of Lewis acid (BPh₃), Lewis base (DMPU), and no additives. BPh₃ additive led to diminished enantioselectivity and differing results in ¹³CN crossover experiments, initial rate kinetics, and natural abundance ¹²C/¹³C kinetic isotope effect measurements. We propose two catalytic mechanisms to account for our experimental results. We propose that the DMPU/nonadditive pathway passes through a κ²-phosphoramidite-stabilized Pd⁺ intermediate, resulting in high enantioselectivity. BPh₃ prevents the dissociation of CN^-, leading to a less rigid κ²-phosphoramidite-neutral Pd intermediate.

Full text of DOI:10.1021/acs.joc.7b00196

Article
pubs.acs.org/joc
Mechanistic Model for Enantioselective Intramolecular Alkene
Cyanoamidation via Palladium-Catalyzed C−CN Bond Activation
Grant B. Frost,† Nicholas A. Serratore,† Jodi M. Ogilvie, and Christopher J. Douglas*
Department of Chemistry, University of Minnesota-Twin Cities, 207 Pleasant Street SE, Minneapolis, Minnesota 55455, United
States
S
* Supporting Information
ABSTRACT: We studied key aspects of the mechanism of Pd-
catalyzed C−CN bond activation and intramolecular enantiose-
lective alkene cyanoamidation. An Abboud−Abraham−Kamlet−
Taft (AAKT) linear solvation energy relationship (LSER) model
for enantioselectivity was established. We investigated the impact of
Lewis acid (BPh3), Lewis base (DMPU), and no additives. BPh3
additive led to diminished enantioselectivity and differing results in
13CN crossover experiments, initial rate kinetics, and natural
abundance 12C/13C kinetic isotope effect measurements. We
propose two catalytic mechanisms to account for our experimental
results. We propose that the DMPU/nonadditive pathway passes
through a κ2-phosphoramidite-stabilized Pd+ intermediate, resulting
in high enantioselectivity. BPh3 prevents the dissociation of CN−, leading to a less rigid κ2-phosphoramidite-neutral Pd
intermediate.
observed upon addition of BPh3. BPh3 also assisted dissociation
of CN− from the (NC)Ni(dippe) complex.35
INTRODUCTION
■
Metal-catalyzed C−C bond activation is an intriguing challenge
in organic synthesis due to the nonpolar C−C σ-bond being
thermodynamically stable and typically unreactive.1−5 Funda-
mental research sparked the development of general C−C σ-
bond activation strategies for homogeneous catalysis including
strain-promoted,6−9 chelation-assisted,10−13 and C−CN14 bond
activation. Pd or Ni oxidative addition into C−CN bonds,
combined with intramolecular or intermolecular migratory
insertion across C−C π-bonds, has proven particularly
attractive for catalytic reaction development. Hallmarks of C−
CN bond activation methodologies include the production of
two new C−C bonds, high atom economy, and retention of the
CN functional group. Variations include alkylcyanation,15
allylcyanation,16 alkynlcyanation,17 arylcyanation,18−21 acylcya-
nation,22−24 cyanoesterification,25−27 and cyanoamida-
tion.25,28−31
Lewis acid additives, typically B, Al, or Zn, are often additives
in catalytic C−CN bond activation.15,17,19,21,23,25 However,
Lewis acid additives have contrasting effects in stoichiometric
processes, making their overall impact on catalysis difficult to
assess. Nolan et al. noted that Lewis acid additives increased the
rate of reductive elimination of TMSCH2−CN from Pd-
(dppp).32 Jones and colleagues demonstrated that BPh3
additive in the activation of Me−CN promoted oxidative
addition with Pd(dippe)33 but favored reductive elimination
with Ni(dippe).34 In their studies of stoichiometric allyl C−CN
bond activation with Ni(dippe), a process involved in DuPont’s
synthesis of adiponitrile, complete chemoselectivity for C−CN
bond activation, versus competing C−H activation, was
Despite the considerable expansion of catalytic C−CN bond
activation methodologies in organic synthesis, mechanistic
study of these reactions has been sparse. In a seminal
mechanistic study of the catalytic variant of allyl C−CN bond
activation, Jones demonstrated a relationship between solvent
polarity and the regioselectivity of the reaction and also
measured activation parameters for C−CN bond cleavage.36
Beyond this, insights into the mechanism of catalytic C−CN
bond activation reactions have been primarily based on X-ray
structures19,25,26 of isolated intermediates as well as insights
from divergent reactivity observed with atypical substrates from
some methodology studies.18,20,30
This void in mechanistic understanding, especially with
regards to enantioselective Pd-catalyzed C−CN bond activation
reactions and our inability to predict the impact of Lewis acid
additives, motivated us to explore these topics. To the best of
our knowledge, CN crossover, initial rate kinetics, and kinetic
isotope effects (KIE) have not been investigated in detail. We
performed these experiments as well as studied solvent effects
for asymmetric alkene cyanoamidation. Our findings support a
mechanistic model in which Lewis acid additive results in an
altered coordination environment around the catalyst during
the enantiodetermining step. We expect our findings will
impact and guide further development of asymmetric C−CN
bond activation processes in fine chemical synthesis.
Received: January 25, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.7b00196
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
The Pd-catalyzed intramolecular enantioselective cyanoami-
dation of cyanoformamide 1, developed by Takemoto et al.,
was selected for study (Scheme 1).29 The reaction has high
We investigated the potential for interactions within the
cybotactic region to affect the enantioselectivity by performing
an additional 10 cyanoamidation reactions with solvents of
varying dielectric constant (ε). All 10 reactions resulted in
complete conversion to oxindole 2 with a significant range in er
(entries 8−17). Specifically, cyanoamidation in cyclohexane
gave 2 with an er of 89:11, identical to that of decalin.
Acetonitrile, with the highest dielectric constant, gave 2 with an
er of 71:29. There was a general trend of nonpolar solvents
leading to higher er; however, single linear regression analysis
of er and dielectric constant gave R2 = 0.665.38−41 We used the
Abboud−Abraham−Kamlet−Taft (AAKT) multiparameter
solvatochromic equation (eq 1), which is a more comprehen-
sive assessment of solvent polarity, to establish a linear
solvation energy relationship (LSER) with the er data (Figure
1).42−44 The three AAKT parameters, π* (polarizability/
Scheme 1. Enantioselective Alkene Cyanoamidation
synthetic potential, forming a chiral oxindole 2, with good
enantioselectivity, good yield, and no observable byproducts.
The reaction conditions are also distinct among catalytic C−
CN bond activation reactions in that a polar/Lewis base
additive, cyclic urea DMPU (dimethyltetrahydropyrimidinone),
is incorporated. We are unaware of any other studies involving
Lewis base additives. Additionally, the highest enantioselectivity
was obtained using the uncommonly deployed decalin as
solvent.
RESULTS AND DISCUSSION
■
Our initial attempts to reproduce the conversion of 1 into
oxindole 2 consistently gave product with an er of 89:11 (Table
1, entry 1, >10 runs).37 Interestingly, decalin in the absence of
Figure 1. Plot of observed log(er) vs calculated log(er) with
multiparameter AAKT regression analysis.
Table 1. AAKT Model LSER for Enantioselectivity
dipolarity), α (hydrogen bond acidity), and β (hydrogen
bond basicity), gave a R2 = 0.798 (see Supporting Information
for parameter values). A Grubb’s test of the residuals led us to
exclude the MeCN data, for which an anomalously low er was
observed.45 The revised LSER gave an improved fit with R2 =
0.922. The resulting parameter coefficients for π*, α, and β
indicated solvent polarizability/dipolarity as the major contrib-
utor to the fit. Single regression analysis with only π* gave a fit
of R2 = 0.917. The discovery of an inverse correlation between
solvent polarizability/dipolarity and er led us to screen solvents
with lower π* values than decalin (π* = 0.11) and cyclohexane
(π* = 0). Heptane (π* = −0.08) and perfluoromethylcyclohex-
ane (PFMCH) (π* = −0.40) as solvents resulted in full
conversion to oxindole 2, but with a lower than predicted er
(entries 18 and 19). This is likely accounted for by the
reactions remaining heterogeneous. All other reactions were
homogeneous at 100 °C.
a
a
entry
solvent
er
entry
solvent
er
b
1
2
decalin
89:11
89:11
85:15
85:15
11
12
13
14
15
16
17
18
19
MEK
83:17
82:18
82:18
79:21
79:21
76:24
71:29
84:16
75:25
decalin
toluene
toluene
dioxane
PhCF3
DCE
3
b
4
c
5
DMPU
NMP
d
6
decalin
80:20
77:23
89:11
85:15
84:16
DMF
d
7
toluene
MeCN
heptane
PFMCH
8
cyclohexane
m-xylene
9
10
2-propanol
*
log er = c + xπ + yα + zβ
( )
(1)
a
b
1
Complete conversion to 2 observed by H NMR. With 100 mol %
c
d
In the previous work, it was conjectured that a Pd+
intermediate resulting from the dissociation of CN− would
not form during the catalytic cycle because the bidentate
phosphine ligands screened did not perform well in comparison
to phosphoramidite L1.31 We synthesized a 13CN-labeled
methyl-substituted cyanoformamide 3* to investigate this
hypothesis.15 A 1:1 mixture of 1 and 3* underwent
cyanoamidation with 100 mol % of DMPU. Unexpectedly,
crossover of the 13CN label was observed in both decalin and
of DMPU. No reaction. With 100 mol % of BPh3.
DMPU at 100 °C gave full conversion with an identical er of
89:11 (entry 2). Toluene gave quantitative conversion and an
er of 85:15 (entry 3). As observed with decalin, the addition of
100 mol % of DMPU did not affect the er (entry 4). DMPU as
the solvent resulted in no reaction of 1 (entry 5). Substituting
100 mol % of Lewis acid BPh3 for DMPU in both decalin and
toluene decreased the er significantly (entries 6 and 7).
B
DOI: 10.1021/acs.joc.7b00196
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
toluene, producing ∼1:1:1:1 distribution of 2, 2*, 4, and 4*
We investigated the turnover-limiting step by determining
the 12C/13C KIE of the reaction with 100 mol % of DMPU or
BPh3 in decalin. Reactions were run on gram scale to >85%
conversion to oxindole 2. The starting cyanoformamide 1 was
then reisolated and analyzed using quantitative 13C NMR
spectroscopy.48,49 Each reaction with BPh3 or DMPU showed
statistically significant, but different, levels of isotopic enrich-
ment at the carbonyl carbon (C1) as well as the CN carbon
(C2) (Figure 2). Negligible enrichments were observed at the
(Table 2, entries 1 and 2).23,29,46 In the absence of DMPU, full
Table 2. 13CN Crossover Experiments
entry
solvent
additive
4*/2*
er of 2
1
2
3
4
5
6
7
decalin
toluene
decalin
toluene
toluene
toluene
toluene
DMPU (100 mol %)
DMPU (100 mol %)
47:53
51:49
49:51
51:49
36:64
35:65
30:70
Figure 2. Map of 13C enrichment after natural abundance measure-
ment of the 12C/13C KIE: 100 mol % of DMPU (left); 100 mol % of
BPh3 (right).
BPh3 (50 mol %)
BPh3 (100 mol %)
BPh3 (200 mol %)
80:20
77:23
73:27
olefinic carbons (C3 and C4). These results indicate that C−
CN bond activation via oxidative addition is the turnover-
limiting step of the reaction.
crossover was also observed (entries 3 and 4). Upon the
addition of 100 mol % of BPh3, the ratio of 2*/4* favored 4*,
showing partial retention of the 13CN label (entry 6). An
experiment with 50 mol % of BPh3 gave a similar result (entry
5). However, more of the 13CN label was retained when 200
mol % of BPh3 was added (entry 7). Heating the mixture of 1
and 3* in the absence of catalyst resulted in no crossover of the
13CN label (see Supporting Information).
The results of our experiments led us to attempt
cyanoamidation under reoptimized conditions (Scheme 2).
Scheme 2. Optimized Alkene Cyanoamidation
We hypothesized that a mechanism for crossover could
involve the coordination of a second L1, facilitating CN−
dissociation from Pd. To test this hypothesis, the crossover
experiment was performed with 4 mol % of L1 (2:1 instead of
4:1 L1/Pd) with the expectation that some of the label would
be retained (see Supporting Information). However, complete
crossover was observed, and the er of 2 was unchanged. When
L1 loading was reduced to 2 mol % (1:1 L1/Pd), negligible
change in er was observed, thus suggesting a mechanism where
only one L1 is ligated to the metal center at the
enantiodetermining step. Crossover studies of oxindoles 2
and 4* showed no crossover of the 13CN label after 24 h at 100
°C, suggesting reductive elimination is irreversible. Likewise,
resubjecting enantioenriched oxindole 2 induced no change in
er, showing the enantiodetermining step is irreversible.
The drastic difference in reactivity observed with the addition
of BPh3 and the lack of evidence that DMPU has a significant
effect on the reaction led us to study the initial rate kinetics of
the reactions under both sets of conditions. We performed
initial rate kinetics with 1H NMR spectroscopy in toluene-d8 at
90 °C (see Supporting Information).47 By varying DMPU
concentrations relative to 1, we observed an inhibitory effect of
increased [DMPU] on the overall rate. DMPU inhibition
matches with our prior observation that DMPU cannot serve as
a solvent for the reaction. In contrast, BPh3 showed a general
increase in the rate of the reaction, following classic saturation
kinetics that plateaued at [BPh3] greater than 100 mol %
relative to 1. However, the increase in rate is only 28% relative
to the rate without added BPh3 (see Supporting Information).
Cyclohexane was used as solvent as it was shown at small scale
(0.1 mmol) to give the same er as decalin. The ligand loading
was decreased from 8 to 2 mol % (1:1 L1/Pd) as it was shown
to have no effect on conversion or er in toluene. The reaction
was scaled to 0.4 mmol, and the concentration was increased to
0.4 M. The revised cyanoamidation conditions resulted in
quantitative yield of 2 with an er of 88:12.
We forward two mechanistic models consistent with our
results. Without BPh3, we propose that η2-coordination of Pd to
the CN of 1 will form A, in accordance with structures from
Jones et al. (Figure 3, left).33 From A, turnover-limiting cis
oxidative addition into the C−CN bond forms intermediate B.
Crossover may then occur by dissociation of CN− from B,
forming cation C. Although the proposed cationic intermediate
is favored with increased solvent polarity, our AAKT LSER of
enantioselectivity indicated that higher er was achieved with
lower π* value solvents, as long as the reaction was
homogeneous at 100 °C. We propose enhancement of
enantioselectivity by intramolecular stabilization of the Pd+
via the secondary coordination of a phenethyl arm of L1,
creating a more rigid bidentate chiral ligand structure. Higher
π* value solvents increase intermolecular interactions between
solvent dipoles and Pd+, likely disrupting the desirable Ph−Pd
coordination.31,50 Trost et al. proposed an analogous secondary
coordination in phosphoramidate/Pd-catalyzed enantioselective
C
DOI: 10.1021/acs.joc.7b00196
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 3. Proposed catalytic cycle for (a) nonadditive/DMPU and (b) BPh3 additive.
trimethylenemethane reactions.51,52 Higher enantioselectivity
achieved via bidentate Pd+ intermediates also finds precedent in
the cationic pathway for the asymmetric Heck reaction.53−55
Considering this insight, reaction optimization with other
substrates should include screening for the most nonpolar
solvent that still provides homogeneous reaction mixtures. In
this highly enantioselective pathway, migratory insertion from
C to D may be in competition with a lower selectivity pathway
to D from B. Irreversible reductive elimination completes the
cycle, forming E and 2 after coordination of 1.
calculated er for the exclusively BPh3-catalyzed pathway would
be 62:38 (see Supporting Information for calculation). This
result suggests that developing chiral catalysts that can
accomplish C−CN bond activation without the use of Lewis
acid will be advantageous to developing more enantioselective
reactions.
CONCLUSION
■
In conclusion, the results obtained from the LSER, 13CN
crossover, kinetics, and 12C/13C KIE experiments led us to
propose a rigid bidentate L1 Pd+ intermediate as the origin of
high enantioselectivity, counter to previous conjecture.11 Lewis
acid BPh3 opens a second operable catalytic cycle that increases
the reaction rate but leads to diminished enantioselectivity, due
to the retention of CN− preventing the formation of the Pd+
intermediate. DMPU inhibits the rate of the reaction but does
not impact enantioselectivity. We have developed improved
mechanistic models for alkene cyanoamidation that will guide
development of asymmetric reactions involving C−CN bond
activation.
The addition of BPh3 opens a new pathway that operates
concurrently with the non-BPh3-catalyzed pathway. The two
pathways operating concurrently is supported by the
diminished, but nonzero observed, KIE as well as the 13C
label being partially, but not completely, retained in crossover
experiments with BPh3 additive. This new pathway begins with
BPh3 coordination to the nitrogen in 1 to form I.33 11B and 13
C
NMR spectra were taken of pure BPh3 and a 1:1 mixture of
BPh3 and 1 in toluene-d8. No change was observed in the
spectra (see Supporting Information), indicating the equili-
brium does not favor the Lewis pair.56 We hypothesize that
coordination of Pd to I is turnover-limiting based on the
relative decrease in magnitude of the 12C/13C isotope
enrichment observed at C1 and C2 in the BPh3 additive KIE
experiment. Also, there is no enrichment observed at C3 or C4,
indicating that the turnover-limiting step did not change to
migratory insertion or reductive elimination. The catalytic cycle
with BPh3 does not include a Pd+ species, reflecting our
observation that the 13CN label in 3* is retained to a larger
extent in the product 4*. Also, the decreased er obtained in
reactions with added BPh3 further supports CN− coordination
during the enantiodetermining step, preventing the formation
of C.21 BPh3 may prevent CN− dissociation due to a stronger
Pd−CN bond upon formation of the Lewis adduct.33 In both
cases, only one L1 is coordinated to Pd, based on the
observation that decreasing the L1/Pd ratio to 1:1 does not
affect the er of the reaction.51
EXPERIMENTAL SECTION
■
General Information. All reactions were carried out using oven-
dried glassware. Cyanoformamide 1 and 3* were prepared by literature
procedures.29 Solvents were dried and degassed prior to use or used
fresh from new bottles. Pd2(dba)3·CHCl3 was purified by literature
procedure from Pd2dba3, which was purchased from commercial
sources.57 Phosphoramidite L1 was prepared according to a literature
procedure.58 All other chemicals were purchased from commercial
vendors and used as received. All cyanoamidation reactions were
carried out in a nitrogen-filled glovebox in 1 dram vials sealed with
PTFE lined caps or 5 mm NMR tubes. Heating was applied by
aluminum blocks. Analytical thin layer chromatography was carried out
using 0.25 mm silica plates. Flash chromatography was performed
using 230−400 mesh (particle size 0.04−0.063 mm) silica gel. 1H
NMR (300 and 500 MHz), 13C NMR (75, 100, and 125 MHz), and
11B NMR (128 MHz) spectra were obtained on FT NMR instruments.
1H NMR spectra were reported as δ values in parts per million relative
to tetramethylsilane (TMS); 13C NMR spectra were referenced to
CDCl3 at 77.16 ppm, and 11B NMR spectra were absolute referenced
to TMS from a 1H NMR spectrum obtained on the same instrument.
The inherent er of the BPh3 pathway can be estimated with
the assumption that the 13CN label is fully retained if all the
material proceeds through the BPh3-catalyzed pathway. The
D
DOI: 10.1021/acs.joc.7b00196
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
General Procedure for 0.1 mmol Scale Cyanoamidation.
Under N2 atmosphere in a glovebox, 0.0276 g (0.1 mmol) of
cyanoformamide 1, 0.0043 g (0.008 mmol) of L1, and 0.0010 g (0.001
mmol) of Pd2(dba)3·CHCl3 were massed in a 1 dram screw top
reaction vial. If DMPU was used in the reaction, 0.0128 g (12 μL) (0.1
mmol) was added via microsyringe. If BPh3 was used, 0.0242 g (0.1
mmol) or other desired quantity was massed in the reaction vial. One
milliliter of the solvent was added via syringe as well as a 5 mm stir bar,
and the reaction vial was sealed. The reactions were heated at 100 °C
for 24 h in an aluminum heating block on a preheated hot plate in a
glovebox. Once completed, the reactions were allowed to cool to room
temperature and removed from the glovebox. Depending on the
solvent boiling point, the reaction mixtures were concentrated either in
phosphoramidite L1 (1.16 mmol) in a glovebox under nitrogen
atmosphere. Decalin was then added to the catalyst (40 mL).
Cyanoformamide 1 was added (7.25 mmol), and the reaction mixture
was diluted with additional decalin until the reaction mixture was 0.1
M (72 mL total). DMPU (7.25 mmol) was added, and the reaction
flask was sealed with a rubber septum and electrical tape. The flask was
then removed from the glovebox; a nitrogen inlet was inserted with
positive nitrogen pressure, and the reaction was heated to 100 °C in an
oil bath.
After 1 h, a 0.1 mL sample was removed and filtered through a plug
of silica gel. The silica gel was rinsed first with hexanes to remove
decalin and then with ethyl acetate to flush off the product and starting
material. The sample was concentrated, and an NMR spectrum of the
crude material was taken to determine the approximate percent
conversion. This process was repeated until the reaction was
determined to be ∼90% complete.
The reaction mixture was removed from the oil bath, diluted with
cold hexanes, and immersed in an ice bath to quickly cool to room
temperature. The reaction mixture was then flushed down a large
column of silica to remove the catalyst. Decalin was removed by first
flushing the column of silica with hexanes (∼250 mL). Then, the
column was rinsed several times with ethyl acetate (∼500 mL
combined) to ensure all the starting material was obtained. The
solution was then concentrated, and silica gel column chromatography
was used to separate the enriched starting material from the product
(90:10 hexane/EtOAc as eluent). Concentration of fractions yielded
enriched 1 for the natural abundance KIE 13C NMR experiment.
Pure samples of the recovered enriched 1 and unreacted substrate 1
were quantitatively analyzed by 13C NMR spectroscopy. Spectra were
obtained on a Bruker Avance III 500 MHz spectrometer (125 MHz,
CDCl3) at 300 K with an inverse-gated decoupling pulse sequence and
calibrated 30° pulses, collecting 256k points total. T1 values were
measured prior to obtaining data, and D1 values were set to 120 s with
acquisition times of 11.5 s. Five acquisitions were obtained for each
sample. Each spectrum was processed using a −1.00 Hz exponential
and a 3.00 Hz Gaussian apodization, as well as a third-order Bernstein
polynomial baseline correction.
vacuo by rotatory evaporation or with a Kugelrohr. The crude oxindole
̈
2 was then dissolved in ∼1 mL of CDCl3 and filtered through a 0.2 μm
syringe filter. 1H NMR was taken to determine conversion to oxindole:
HPLC [Chiralcel OD-H, hexane/2-propanol = 90/10, 1.0 mL/min, λ
1
= 254 nm, retention times (major) 14.3 min, (minor) 17.5 min; H
NMR (500 MHz, CDCl3) δ 1.57 (s, 3H), 2.64 (d, 1H, J = 16.5 Hz),
2.90 (d, 1H, J = 16.5 Hz), 4.93 (s, 2H), 6.78 (d, 1H, J = 7.6 Hz), 7.09
(t, 1H, J = 7.6 Hz), 7.27 (m, 6H), 7.47 (m, 1H).
General Procedure for 13CN Crossover Experiments. Under
N2 atmosphere in a glovebox, 0.024 g (0.087 mmol) of
cyanoformamide 1 and 0.025 g (0.086 mmol) of 13CN-labeled
cyanoformamide 3*, 0.0075 g (0.014 mmol) of L1, and 0.0017 g
(0.0017 mmol) of Pd2(dba)3·CHCl3 were weighed in a screw top 1
dram reaction vial. If DMPU was used in the reaction, 21 μL (0.022 g)
(0.17 mmol) was added via microsyringe. If BPh3 was used, 0.042 g
(0.17 mmol) or other desired quantity was massed in the reaction vial.
Next, 1.74 mL of solvent (toluene or decalin) was added via syringe as
well as a 5 mm stir bar, and the reaction vial was sealed. The reactions
were heated at 100 °C for 24 h in an aluminum heating block on a
preheated hot plate in a glovebox. Once completed, the reactions were
allowed to cool to room temperature and removed from the glovebox.
The reaction mixtures were concentrated in vacuo by rotatory
evaporation. The crude mixture of oxindoles 2, 2*, 4, and 4* was then
dissolved in ∼1 mL of CDCl3 and filtered through a 0.2 μm syringe
1
filter. H NMR spectra were then taken to determine conversion to
oxindole. The extent of crossover was determined of integration of 13C
NMR spectra (see Supporting Information): HPLC [Chiralcel OD-H,
hexane/2-propanol = 90/10, 1.0 mL/min, λ = 254 nm, retention times
(major) 14.3 min, (minor) 17.5 min.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.7b00196.
■
S
Initial Rate Kinetics Method. In a nitrogen-filled glovebox, two
stock solutions were made in toluene-d8: stock A (0.2 M
cyanoformamide, 0.002 M Pd2(dba)3, and 0.016 M L1) and stock B
(0.4 M additive (BPh3 or DMPU)). Next, 1.2 mL of stock A was
added to separate vials, and then stock B was added to make solutions
of 0, 0.1, 0.25, 0.5, 1, and 2 equiv of additive. Toluene-d8 was then
added to bring the total volume to 2.4 mL in all vials. Next, 0.7 mL
from each vial was added to separate NMR tubes so that each reaction
was run in triplicate, and the tubes were sealed with plastic caps and
the caps secured with electrical tape. The tubes were then placed in the
NMR spectrometer autosampler and sequentially monitored via
variable-temperature NMR spectroscopy at 90 °C (calibrated using
pure glycol standard) for 1 h (roughly 10% conversion). Spectra were
processed using the integrals graph function in MestReNova after we
performed phase and baseline corrections. The data were exported to
Microsoft Excel, and the data fit with a linear line to obtain rates. All
data were normalized to obtain mM/s rates based on the integral of
the peak at 3.93 ppm being from the starting material, which is 90% of
the total starting material (as it is one of two peaks corresponding to
rotational isomers that are resolved in the spectra, with the lesser
isomer being 10% of the total). The rates obtained were averaged and
standard deviations obtained. The average rates were plotted with
error bars equivalent to the standard deviation against the
concentration of additive in the experiment (see Supporting
Information for plots).
General procedure, HPLC data, rate kinetics, KIE
measurement (PDF)
Spectra, kinetics plots, LSER, HPLC chromatograms
(PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: cdouglas@umn.edu.
ORCID
Nicholas A. Serratore: 0000-0002-2602-425X
Christopher J. Douglas: 0000-0002-1904-6135
■
Author Contributions
†G.B.F. and N.A.S. contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank National Institutes of Health for funding this work
(R01 GM095559). We thank Prof. Steve Kass (UMN) for
constructive feedback.
General Procedure for Collection of Sample for 12C/13C
Kinetic Isotope Effect Determination. A 250 mL round-bottom
flask was charged with a stir bar, Pd2(dba)3 (0.140 mmol), and
REFERENCES
(1) Jones, W. D. Nature 1993, 364, 676.
■
E
DOI: 10.1021/acs.joc.7b00196
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(2) Rybtchinski, B.; Milstein, D. Angew. Chem., Int. Ed. 1999, 38, 870.
(3) Murakami, M.; Matsuda, T. Chem. Commun. 2011, 47, 1100.
(4) Ruhland, K. Eur. J. Org. Chem. 2012, 2012, 2683.
(5) Murakami, M.; Ishida, N. J. Am. Chem. Soc. 2016, 138, 13759.
(6) Matsuda, T.; Fujimoto, A.; Ishibashi, M.; Murakami, M. Chem.
Lett. 2004, 33, 876.
(42) Kamlet, M. J.; Abboud, J. L. M.; Abraham, M. H.; Taft, R. W. J.
Org. Chem. 1983, 48, 2877 and references therein.
(43) Stephenson, W. K.; Fuchs, R. Can. J. Chem. 1985, 63, 2535.
(44) Laurence, C.; Nicolet, P.; Dalati, M. T.; Abboud, J.-L. M.;
Notario, R. J. Phys. Chem. 1994, 98, 5807.
(45) Likely due to competitive coordination of the cyano group in
acetonitrile to the palladium, disrupting secondary coordination of the
phosphoramidite.
(7) Huffman, M. A.; Liebeskind, L. S. J. Am. Chem. Soc. 1993, 115,
4895.
(46) Bergman, R. G. Acc. Chem. Res. 1980, 13, 113.
(47) We used 90 °C to obtain rate data because the reactions were
too fast at 100 °C.
(48) Rathbun, C. M.; Johnson, J. B. J. Am. Chem. Soc. 2011, 133,
2031.
(49) Singleton, D. A.; Thomas, A. A. J. Am. Chem. Soc. 1995, 117,
9357.
(50) Changing the phenyl group to methyl led to significantly
decreased er; see ref 31.
(51) Trost, B. M.; Bringley, D. A. Angew. Chem., Int. Ed. 2013, 52,
4466.
(8) Murakami, M.; Itahashi, T.; Ito, Y. J. Am. Chem. Soc. 2002, 124,
13976.
(9) Zhou, X.; Dong, G. J. Am. Chem. Soc. 2015, 137, 13715.
(10) Wentzel, M. T.; Reddy, V. J.; Hyster, T. K.; Douglas, C. J.
Angew. Chem., Int. Ed. 2009, 48, 6121.
(11) Dreis, A. M.; Douglas, C. J. J. Am. Chem. Soc. 2009, 131, 412.
(12) Jun, C.-H.; Lee, H. J. Am. Chem. Soc. 1999, 121, 880.
(13) Ko, H. M.; Dong, G. Nat. Chem. 2014, 6, 739.
(14) Tobisu, M.; Chatani, N. Chem. Soc. Rev. 2008, 37, 300.
(15) Nakao, Y.; Yada, A.; Hiyama, T. J. Am. Chem. Soc. 2010, 132,
10024.
(52) Mikhel, I. S.; Ruegger, H.; Butti, P.; Camponovo, F.; Huber, D.;
̈
(16) Nakao, Y.; Yukawa, T.; Hirata, Y.; Oda, S.; Satoh, J.; Hiyama, T.
J. Am. Chem. Soc. 2006, 128, 7116.
Mezzetti, A. Organometallics 2008, 27, 2937.
(53) Dounay, A. B.; Overman, L. E. Chem. Rev. 2003, 103, 2945.
(54) Cabri, W.; Candiani, I.; DeBernardinis, S.; Francalanci, F.;
Penco, S.; Santo, R. J. Org. Chem. 1991, 56, 5796.
(55) Ozawa, F.; Kubo, A.; Hayashi, T. J. Am. Chem. Soc. 1991, 113,
1417.
(56) Wrackmeyer, B. Prog. Nucl. Magn. Reson. Spectrosc. 1979, 12,
227.
(57) Zalesskiy, S. S.; Ananikov, V. P. Organometallics 2012, 31, 2302.
(58) Arnold, L. A.; Imbos, R.; Mandoli, A.; de Vries, A. H. M.; Naasz,
R.; Feringa, B. L. Tetrahedron 2000, 56, 2865.
(17) Hirata, Y.; Tanaka, M.; Yada, A.; Nakao, Y.; Hiyama, T.
Tetrahedron 2009, 65, 5037.
(18) Nakao, Y.; Yada, A.; Satoh, J.; Ebata, S.; Oda, S.; Hiyama, T.
Chem. Lett. 2006, 35, 790.
(19) Nakao, Y.; Ebata, S.; Yada, A.; Hiyama, T.; Ikawa, M.; Ogoshi, S.
J. Am. Chem. Soc. 2008, 130, 12874.
(20) Nakao, Y.; Oda, S.; Hiyama, T. J. Am. Chem. Soc. 2004, 126,
13904.
(21) Watson, M. P.; Jacobsen, E. N. J. Am. Chem. Soc. 2008, 130,
12594.
(22) Nozaki, K.; Sato, N.; Takaya, H. Bull. Chem. Soc. Jpn. 1996, 69,
1629.
(23) Rondla, N. R.; Ogilvie, J. M.; Pan, Z.; Douglas, C. J. Chem.
Commun. 2014, 50, 8974.
(24) Nozaki, K.; Sato, N.; Takaya, H. J. Org. Chem. 1994, 59, 2679.
(25) Hirata, Y.; Yada, A.; Morita, E.; Nakao, Y.; Hiyama, T.; Ohashi,
M.; Ogoshi, S. J. Am. Chem. Soc. 2010, 132, 10070.
(26) Nishihara, Y.; Inoue, Y.; Itazaki, M.; Takagi, K. Org. Lett. 2005,
7, 2639.
(27) Nishihara, Y.; Inoue, Y.; Izawa, S.; Miyasaka, M.; Tanemura, K.;
Nakajima, K.; Takagi, K. Tetrahedron 2006, 62, 9872.
(28) Kobayashi, Y.; Kamisaki, H.; Yanada, R.; Takemoto, Y. Org. Lett.
2006, 8, 2711.
(29) Yasui, Y.; Kamisaki, H.; Takemoto, Y. Org. Lett. 2008, 10, 3303.
(30) Kobayashi, Y.; Kamisaki, H.; Takeda, H.; Yasui, Y.; Yanada, R.;
Takemoto, Y. Tetrahedron 2007, 63, 2978.
(31) Yasui, Y.; Kamisaki, H.; Ishida, T.; Takemoto, Y. Tetrahedron
2010, 66, 1980.
(32) Huang, J.; Haar, C. M.; Nolan, S. P.; Marcone, J. E.; Moloy, K.
G. Organometallics 1999, 18, 297.
(33) Munjanja, L.; Torres-Lop
Organometallics 2016, 35, 2010.
(34) García, J. J.; Arevalo, A.; Brunkan, N. M.; Jones, W. D.
́
ez, C.; Brennessel, W. W.; Jones, W. D.
́
Organometallics 2004, 23, 3997.
(35) Brunkan, N. M.; Brestensky, D. M.; Jones, W. D. J. Am. Chem.
Soc. 2004, 126, 3627.
(36) Swartz, B. D.; Reinartz, N. M.; Brennessel, W. W.; García, J. J.;
Jones, W. D. J. Am. Chem. Soc. 2008, 130, 8548.
(37) We will use enantiomeric ratio in this paper as er directly
compares the activation barriers for the enantiodetermining step of the
reaction.
(38) Teixeira, J.; Silva, A. R.; Branco, L. C.; Afonso, C. A. M.; Freire,
C. Inorg. Chim. Acta 2010, 363, 3321.
(39) Lee, S. B. J. Ferment. Bioeng. 1995, 80, 141.
(40) Cativiela, C.; Garcia, J. I.; Mayoral, J. A.; Salvatella, L. J. Chem.
Soc., Perkin Trans. 2 1994, 847.
(41) Bini, R.; Chiappe, C.; Mestre, V. L.; Pomelli, C. S.; Welton, T.
Org. Biomol. Chem. 2008, 6, 2522.
F
DOI: 10.1021/acs.joc.7b00196
J. Org. Chem. XXXX, XXX, XXX−XXX