Reductive Elimination from Phosphine-Ligated Alkylpalladium(II) Amido Complexes to Form sp3 Carbon-Nitrogen Bonds

Article abstract of DOI:10.1021/jacs.8b00928

We report the formation of phosphine-ligated alkylpalladium(II) amido complexes that undergo reductive elimination to form alkyl-nitrogen bonds and a combined experimental and computational investigation of the factors controlling the rates of these react

Full text of DOI:10.1021/jacs.8b00928

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE
Reductive Elimination from Phosphine-ligated Alkylpalladium(II)
Amido Complexes to Form sp3 Carbon–Nitrogen Bonds
D. Matthew Peacock, Quan Jiang, Patrick S. Hanley, Thomas R. Cundari, and John F. Hartwig
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b00928 • Publication Date (Web): 14 Mar 2018
Downloaded from http://pubs.acs.org on March 14, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 13
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Reductive Elimination from Phosphine-ligated Alkylpalladium(II)
Amido Complexes to Form sp3 Carbon–Nitrogen Bonds
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
D. Matthew Peacock,† Quan Jiang,‡ Patrick S. Hanley,† Thomas R. Cundari,‡* John F. Hartwig†*
†Department of Chemistry, University of California, Berkeley, California 94720, United States
‡Department of Chemistry and Center for Advanced Scientific Computing and Modeling, University of North Texas, Denton,
Texas 76203, United States
Supporting Information Placeholder
from low-valent metal complexes are more limited. Our group re-
ported the first reductive eliminations from alkylpalladium(II) am-
ido complexes (Scheme 1), in part motivated by reports of catalytic
reactions that could occur through these intermediates.10-12 Ben-
zylpalladium(II) and alkylpalladium(II) complexes have been
shown to react by two distinct mechanisms. Four-coordinate,
bisphosphine-ligated benzylpalladium(II) amido complexes un-
dergo reductive elimination by a stepwise mechanism involving in-
itial dissociation of an anionic amido ligand, followed by nucleo-
philic attack of this anion on the coordinated benzyl ligand.13 This
pathway occurs by inversion of configuration at the palladium-
bound carbon atom. In contrast, three-coordinate, N-heterocyclic
carbene (NHC)-ligated palladium(II) complexes bearing the unsta-
bilized alkyl ligand syn-2-methylnorbornyl underwent reductive
elimination by a concerted pathway, resulting in retention of con-
figuration at the palladium-bound carbon atom.14
ABSTRACT: We report the formation of phosphine-ligated al-
kylpalladium(II) amido complexes that undergo reductive elimina-
tion to form alkyl-nitrogen bonds and a combined experimental and
computational investigation of the factors controlling the rates of
these reactions. The free-energy barriers to reductive elimination
from t-Bu3P-ligated complexes were significantly lower (ca. 3
kcal/mol) than those previously reported from NHC-ligated com-
plexes. The rates of reactions from complexes containing a series
of electronically and sterically varied anilido ligands showed that
the reductive elimination is slower from complexes of less electron-
rich or more sterically-hindered anilido ligands than from those
containing more electron rich and less hindered anilido ligands. Re-
ductive elimination of alkylamines also occurred from complexes
bearing bidentate P,O ligands. The rates of reactions of these four-
coordinate complexes were slower than those for reactions of the
three-coordinate, t-Bu3P-ligated complexes. The calculated path-
way for reductive elimination from rigid, 2-methoxyaryl-ligated
complexes does not involve initial dissociation of the oxygen. In-
stead, reductive elimination is calculated to occur directly from the
four-coordinate complex in concert with a lengthening of the Pd–
O bond. To investigate this effect experimentally, a four-coordinate
Pd(II) anilido complex containing a flexible, aliphatic linker be-
tween the P and O atoms was synthesized. Reductive elimination
from this complex was faster than that from the analogous complex
containing the more rigid, aryl linker. The flexible linker enables
full dissociation of the ether ligand during reductive elimination,
leading to the faster reaction of this complex.
Scheme 1. Previous reports of reductive elimination of al-
kylamines from Pd(II) anilido complexes.
INTRODUCTION
Reductive eliminations to form carbon–nitrogen bonds are im-
portant steps in many reactions that produce amines catalyzed by
or mediated by transition-metal complexes. Reactions to form sp2
carbon–nitrogen bonds from aryl- and heteroarylpalladium(II)
complexes have been well studied.1-2 However, reductive elimina-
tions to form the analogous sp3 C–N bonds in alkylamines from
alkylmetal amido complexes are rare, and the factors controlling
the rates and scope of this elementary reaction are poorly defined.
An increased understanding of this fundamental organometallic re-
action could enable the development of new methods to construct
sp3 C–N bonds by catalytic reactions, such as C–H bond function-
alization, olefin functionalization, or nucleophilic substitution.
The reported examples of reductive elimination from NHC-
ligated (syn-2-methylnorbornyl)palladium(II) complexes occurred
with high kinetic barriers (≥ 26 kcal/mol at 90 °C). Therefore, we
have sought to understand the properties of the complexes and lig-
ands that control the rates of reaction and to identify complexes that
undergo reductive elimination more rapidly. We have also sought
to determine whether the rates of concerted reductive elimination
to form sp3 C–N bonds follow the trends previously reported for
concerted reductive eliminations from d8 transition metal centers
that form sp2-sp3 C–C, sp3-sp3 C–C, and sp2 C–N bonds.
Herein, we report the preparation of a series of phosphine-ligated
palladium(II) amido complexes that undergo reductive elimination
to form alkyl–nitrogen bonds. These studies reveal, by both exper-
imental and computational methods, the effects controlling the
Complexes of palladium(IV)3-5 and other high-valent metal cen-
ters6-9 have been reported to undergo reductive elimination to form
sp3 C–N bonds. However, examples of such reductive eliminations
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 13
structures of the Pd(II) anilido intermediates and the barriers to re-
Scheme 2. Preparation of (t-Bu3P)Pd(2-
CH3norbornyl)NHAr 2a-i and reductive elimination to
form alkylamines 3a-i.
1
2
3
4
ductive elimination from these complexes. These results led to the
discovery that the reductive elimination to form alkylamines from
four-coordinate complexes containing particular P,O ligands is fast
and depends on the ability of the Pd–O bond to lengthen or disso-
ciate during reductive elimination.
5
6
7
RESULTS AND DISCUSSION
Alkylpalladium(II) Amido Complexes Ligated by Mono-
phosphines. Palladium(II) complexes ligated by bulky mono-
phosphines, such as tri-tert-butylphosphine (t-Bu3P), undergo re-
ductive elimination reactions that are slow or do not occur at all
from analogous complexes containing less sterically demanding
ligands.2, 15-16 Therefore, we investigated the potential of alkylpal-
ladium complexes ligated by bulky monophosphines to undergo re-
ductive elimination to form the sp3 carbon–nitrogen bond in alkyl-
amines.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis and reactions of monophosphine-ligated complexes.
The preparation of t-Bu3P-ligated alkylpalladium(II) anilido com-
plexes was conducted in two steps from a known complex by the
sequence shown in Scheme 2. Treatment of the olefin-ligated pre-
cursor 1,5-cyclooctadiene (syn-2-methylnorbornyl)palladium(II)
chloride with t-Bu3P at ambient temperature in THF generated the
three-coordinate complex (t-Bu3P)Pd(2-CH3norbornyl)Cl (1). This
complex was stable at ambient temperature and crystallized as red
blocks in 81% yield. The reaction of complex 1 with sodium tert-
butoxide (NaOt-Bu), t-Bu3P, and a series of arylamines at ambient
temperature formed Pd(II) anilido complexes 2a-i in 76-92% yield
with (t-Bu3P)2Pd0 as a minor side product, as determined by 31P
NMR spectroscopy. Although t-Bu3P is not consumed by the reac-
tion, complexes bearing unhindered anilido ligands formed in
lower yields when the reaction was performed without added t-
Bu3P.
aIsolated yield after recrystallization. bConditions unless otherwise
noted: Pd-Cl 1 (0.0399 mmol), t-Bu3P (0.12 mmol), arylamine (0.042
mmol), and NaO-tBu (0.080 mmol) in THF-d8 (0.70 ml). cReactions were
monitored by 1H NMR spectroscopy while heating at 40 °C in a VT NMR
probe for the noted time. dYield determined by 1H NMR spectroscopy inte-
gration with 1,3,5-trimethoxybenzene (TMB) as internal standard; (values
in parentheses are the overall yields of 3a-i based on 1); averages from two
Table 1. In situ molecular weight estimation by DOSY.a
e
experiments. Reactions monitored by 19F NMR spectroscopy in THF; 4-
fluorotoleune (4-FTol) as internal standard. fReaction was heated at 65 °C.
D·1010
expected DOSY
entry complex
error
/(m2s-1)
12.2
Reductive elimination from alkylpalladium(II) amido complexes
ligated by t-Bu3P. Due to the instability of t-Bu3P-ligated Pd(II)
anilido complexes at room temperature, studies of the reductive
elimination to form alkylamines were conducted with complexes
prepared in situ. Complexes 2a-i underwent reductive elimination
at 40-65 °C to form the corresponding norbornylamines 3a-i in
yields of 59% or higher (Scheme 2). Excess t-Bu3P (included dur-
ing the formation of 2a-i) served to trap the palladium-containing
product as (t-Bu3P)2Pd0.
M
M
1
2
3
4
2g
2i
6
528
558
934
626
497
527
943
666
-31
-31
+9
12.5
9.47
8c
11.1
+40
aDOSY experiments performed by 19F NMR spectroscopy at 300 K in
THF-d8/THF (~10%). See the SI for details on the calculations of diffusion
constants and molecular weight estimates.
The reductive-elimination reactions were monitored by 1H or 19
F
NMR spectroscopy at 40-65 °C in THF-d8 or THF. All reactions of
t-Bu3P-ligated Pd(II) anilido complexes occurred with a first-order
decay. The rate constants for reductive elimination (kRE) and the
corresponding free energies of activation (ΔG‡) are shown in Table
2.
The t-Bu3P-ligated Pd(II) anilido complexes were typically un-
stable at ambient temperature and highly soluble in organic sol-
vents, preventing recrystallization and isolation in pure form. How-
ever, the molecular weights (M) of 2g and 2i were approximated by
19F NMR diffusion ordered spectroscopy(DOSY)experiments, and
these molecular weight-values agreed well with the expected val-
ues for monomeric, three-coordinate alkylpalladium(II) amido
Table 2. Effects of anilido substituents on the rate constants
and free energies of activation for reductive eliminations to
form alkyl–nitrogen bonds.a
1
complexes (Table 1, entries 1 and 2). Furthermore, H NMR nu-
clear Overhauser effect spectra (NOESY) of 2i were consistent
with a three-coordinate T-shaped or distorted T-shaped structure.
Specifically, correlations were observed between the endo hydro-
gen on the palladium-bound carbon and the tert-butyl groups on the
phosphine, as well as between the exo methyl group on the alkyl
ligand and the NH proton of the anilido ligand (see SI for spectra
and assignments).
2
ACS Paragon Plus Environment

Page 3 of 13
Journal of the American Chemical Society
entry
Ar (complex)
Ph (2a)
kRE·104 /s-1 ΔGRE‡ /(kcal·mol-1)
1.0E-03
8.0E-04
6.0E-04
4.0E-04
2.0E-04
0.0E+00
(a)
1
2
3
4
5
6
7
8
1
2
2.6
12
23.5
22.6
23.0
23.5
23.4
23.6
24.5
25.5
23.5
4-(OMe)C6H4 (2b)
p-Tol (2c)
kRE = 7.9±0.3 10-4s-1
3
5.4
2.6
3.1
2.3
0.55
2.5
2.7
4
4-FC6H4 (2d)
5
m-Tol (2e)
0
0.05
0.1
[t-Bu3P] /M
0.15
0.2
6
3-(OMe)C6H4 (2f)
3-FC6H4 (2g)
7b
8c
9
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.0E-03
-11
3.1E-03
3.2E-03
2-(t-Bu)C6H4 (2h)
4-F-2-(OMe)C6H3 (2i)
(b)
aReactions performed as shown in Scheme 2; rate constants determined
by fitting a plot of [2] vs time for 5 half-lives of data to the equation for an
exponential decay; averages from two experiments; deviations were within
20% of the average value. bRate constant was determined from the linear fit
of a ln[2g] vs time plot (1.3 half-lives). cThe reaction of 2h was monitored
at 338 K.
-12
-13
-14
-15
T-1 /K-1
The kinetic data for reductive elimination from complexes bear-
ing meta- and para-substituted anilido ligands (entries 1-7) were
analyzed with a Hammett plot. A reasonable correlation (R2 = 0.95)
between Log kX/kH and σ was observed with a ρ value of -2.0 (Fig-
ure 1). This result suggests that complexes with less stabilization of
a partial negative charge on nitrogen react faster than those with
more stabilization of the charge, and the magnitude of ρ suggests
that significant negative charge accumulates on nitrogen. The or-
tho-tert-butyl substituted 2h underwent reductive elimination with
the highest kinetic barrier of any t-Bu3P-ligated complex that was
investigated. Thus, reductive elimination generally occurred fastest
from complexes bearing unhindered, electron-rich anilido ligands.
Further studies were conducted primarily with 4-fluoro-2-methox-
yaniline because the barrier for reductive elimination from the or-
tho-methoxy substituted 2i was similar to that for reductive elimi-
nation from the unsubstituted 2a, but the reaction of 2i could be
followed by 19F NMR spectroscopy and occurred in high yield.
Figure 2. Determination of the order in [t-Bu3P] and activation parameters
for reductive elimination from 2i. Solutions of 2i were prepared as de-
scribed in Scheme 2. (a) Equivalents of t-Bu3P were varied from 1 to 3 with
T = 323 K. (b) T was varied from 313 to 333 K with 3 equivalents of t-Bu3P.
Further insight into the mechanism of reductive elimination was
obtained by measuring the enthalpy and entropy of activation. An
Eyring analysis (40 - 60 °C) of the reductive elimination from 2i
provided the parameters ΔH‡ = 19±1 kcal/mol and a ΔS‡ = –14±4
eu (Figure 2b). The negative value of ΔS‡ is inconsistent with a
pathway involving ligand dissociation prior to reductive elimina-
tion. The magnitude of ΔS‡ is consistent with either an associative
process or a unimolecular process in which the transition state is
more ordered than the ground state. Because the zeroth-order de-
pendence on [t-Bu3P] rules out an associative pathway, these re-
sults support a mechanism for reductive elimination occurring di-
rectly from the three-coordinate complex observed in solution.
1.0
Synthesis of alkylpalladium(II) complexes ligated by t-Bu2PhP.
To dissect the steric and electronic effects on the reductive elimi-
nation to form the alkyl-nitrogen bond from phosphine-ligated pal-
ladium(II), we studied complexes containing di-tert-bu-
tylarylphosphines. Complexes of such ligands would enable the
steric and electronic properties to be varied by substitution on the
aryl group.
y = -1.96x + 0.050
R² = 0.95
0.5
-0.4
-0.2
0.2
0.4
-0.5
-1.0
The preparation of a t-Bu2PhP-ligated alkylpalladium(II) anilido
complex was attempted by the same two step sequence described
for the synthesis of complexes 2a-i. Treatment of cyclooctadiene-
ligated (syn-2-methylnorbornyl)palladium(II) chloride with t-
Bu2PhP at ambient temperature in THF generated (t-Bu2PhP)Pd(2-
CH3norbornyl)Cl (4a), which was isolated as an off-white powder
in 64% yield after recrystallization. Complex 4a reacted with 4-
fluoro-2-methoxyaniline and NaOt-Bu at ambient temperature with
3 equivalents of t-Bu2PhP in THF to produce a complicated mixture
of palladium complexes (Scheme 3). The species that formed in
σ
Figure 1. Hammett plot for substituent effects on the anilido ligand. Data
from entries 1-7 of Table 2.
We sought to determine whether reductive elimination occurs di-
rectly from the three-coordinate complex observed in solution or
from a different intermediate after association or dissociation of a
ligand. The reaction of the ortho-methoxy substituted 2i to form 3i
was determined to be zeroth-order in t-Bu3P, and the yield of 3i did
not vary significantly with the concentration of t-Bu3P (Figure 2a).
These results suggest that the mechanism of reductive elimination
does not involve either association or reversible dissociation of t-
Bu3P prior to the transition state.
highest concentration in this mixture (as determined by 19F or 31
P
NMR spectroscopy) was assigned to be the bimetallic complex 6a,
which formed in approximately 29% yield (based on 2:1 stoichi-
ometry of 4a to 6a). Complex 6a formed in a higher yield of 63%
when the same reaction was performed with LiHMDS as base in
toluene.
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 13
Scheme 3. Formation of bimetallic alkylpalladium(II) am-
Scheme 4. Reductive elimination from bimetallic alkylpal-
ladium(II) amido complex 6a.a
ido complex 6a.a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aIsolated yield after recrystallization. bYield determined by 19F NMR
spectroscopy integration against 4-FTol; average from two experiments.
cConditions: Pd–Cl 4a (0.020 mmol), t-Bu2PhP (0.060 mmol), 4-fluoro 2-
methoxyaniline (0.020 mmol), and NaOt-Bu (0.040 mmol) in THF (2.0 ml).
dConditions: 4a (0.010 mmol), t-Bu2PhP (0.030 mmol), 4-fluoro 2-methox-
yaniline (0.010 mmol), and LiHMDS (0.010 mmol) in Tol (1.0 ml).
The identity of bimetallic amido complex 6a was deduced by
NMR spectroscopy using samples prepared in situ by the reaction
of chloride complex 4a, 4-fluoro-2-methoxyaline, and LiHMDS in
toluene.17 Complex 6a contains one phosphine ligand, two inequiv-
alent alkyl ligands, and two inequivalent anilido ligands. The for-
mation of approximately one equivalent of free t-Bu2PhP per prod-
uct was also observed, which accounts for the 1:2 ratio of phos-
phine to palladium in 6a. Furthermore, the molecular weight of the
complex, as determined by 19F NMR DOSY experiments, was ap-
proximately 943 (Table 1, entry 3). These results are consistent
with the assignment of 6a as a bimetallic complex lacking one of
the starting phosphine ligands (expected M = 934). 1H NMR
NOESY and 1H-31P HMBC experiments suggested that the two pal-
ladium centers are linked by a single bridging anilido ligand (see
the SI for spectra and partial assignments). These experiments are
consistent with the structure drawn in Scheme 3.
aSamples prepared as described in Scheme 3; yields of 3i (based on 4a)
determined by 19F NMR spectroscopy integration against 4-FTol; averages
from two experiments.
To gain insight into the mechanism of reductive elimination from
the t-Bu2PhP-ligated bimetallic complex 6a, the order in phosphine
was measured. The initial rate (r) of formation of 3i was measured
for reactions conducted with 1, 4, or 9 equivalents of added t-
Bu2PhP (Figure 3, Table S-6, and Figure S-5). In contrast to the rate
of reactions with t-Bu3P, the rate of the reaction with t-Bu2PhP was
dependent on the concentration of phosphine. The approximately
half-order dependence is consistent with a mechanism in which 6a
(or other bimetallic species lacking a phosphine ligand) undergoes
fast reversible association of one t-Bu2PhP to form two equivalents
of the monometallic intermediate prior to reductive elimination
(Scheme 4). Due to its analogy with the t-Bu3P-ligated complex 2i,
we proposed that complex 5a is likely the intermediate that under-
goes reductive elimination of the amine.
The observation of a monometallic complex with t-Bu3P as lig-
and but a bimetallic complex with t-Bu2PhP as ligand shows that
increasing steric bulk at phosphorus leads to a decrease in relative
concentration of the bimetallic complex. The greater electron do-
nation by t-Bu3P than by t-Bu2PhP also would be expected favor
the monometallic complex over the bimetallic complex. However,
a mixture of Pd(II) anilido complexes was observed to form from
reactions of (t-Bu2CyP)Pd(2-CH3norbornyl)Cl (4b), 4-fluoro-2-
methoxyaniline, and base (see the SI for details). The electron-do-
nating properties of t-Bu3P and t-Bu2CyP are similar.18, 19 There-
fore, these results suggest that the steric properties of the ancillary
ligand, not the electronic properties, primarily determine whether
monometallic or bimetallic structures are favored.
-12.8
-13
-13.2
-13.4
-13.6
-13.8
-3.5
-2.5 -1.5
ln([t-Bu2PhP])
-0.5
Reductive elimination from bimetallic alkylpalladium(II) amido
complex 6a. Heating of the mixture of complexes described in
Scheme 3 at 65 °C in THF for 3 h formed the norbornylamine prod-
uct 3i in 75% yield (based on 4a) (Scheme 4). Under these condi-
tions, the remainder of the mass balance (with respect to the amine)
was 4-fluoro-2-methoxyaniline, as determined by 19F NMR spec-
troscopy. The yield of 3i greatly exceeds the yield of any of the
individual palladium complexes in the solution, indicating that
multiple species undergo reductive elimination or that multiple spe-
cies equilibrate prior to reductive elimination.
Figure 3. Effect of the concentration of t-Bu2PhP on the rate of formation
(r) of 3i. Experiments were performed at 50 °C in THF. Concentrations of
3i were measured by 19F NMR integration against 4-fluorotoluene. Initial
rates were measured as the slope of a linear fit of [3i] vs time plot for < 15%
yield
Synthesis and Reactivity of Four-Coordinate Alkylpalla-
dium(II) Amido Complexes Bearing 2-Methoxy-
arylphosphines. Because reductive elimination from the t-
Bu2PhP-ligated complexes is a multi-step process involving multi-
ple (likely equilibrating) Pd(II) anilido complexes, this system was
not amenable to dissecting the effects of ancillary ligands on the
rate of reductive elimination. Therefore, we sought to identify a
class of complexes that was monometallic and contained a phos-
phine that could possess varied steric and electronic properties. It
4
ACS Paragon Plus Environment

Page 5 of 13
Journal of the American Chemical Society
was found that the formation of bimetallic species and the dissoci-
ation of the di-tert-butyl arylphosphine ligand could be prevented
by introducing a second coordinating group at the ortho-position of
1
2
3
the arylphosphine.
Synthesis of alkylpalladium(II) complexes ligated by 2-methoxy-
arylphosphines. The (t-Bu2ArP)Pd(2-CH3norbornyl)Cl complexes
7a-d bearing 2-methoxyaryl groups on phosphorus were prepared
from the reactions of (COD)Pd(2-CH3norbornyl)Cl and the corre-
sponding phosphines in THF or DCM (Scheme 5). All four com-
plexes formed as single isomers, with the phosphine cis to the alkyl
ligand and the ether oxygen cis to the chloride. This geometry was
established by NOESY correlations between the alkyl ligand and
the tert-butyl or adamantyl groups on phosphorus.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 5. Preparation of four-coordinate (t-Bu2(2-
(OMe)Ar)P-κ2P,O)Pd(2-CH3norbornyl)NHAr complexes
8a-d and reductive elimination to form 3i.
Figure 4. ORTEP drawing of [t-Bu2(o-anisyl)P-κ2P,O]Pd(2-
CH3norbornyl)(NHAr) 8a with 50% probability ellipsoids. Hydrogen atoms
are omitted for clarity.Selected bond lengths (Å) and angles (°) for 8a: Pd1-
O1 2.288(2), Pd1-P1 2.2881(8), Pd1-C1 2.057(3), Pd1-N1 2.053(3); O1-
Pd1-P1 80.40(6), P1-Pd1-C1 97.38(9), C1-Pd1-N1 92.92(12), N1-Pd1-O1
89.32(9). For the analogous bonds in 8b: Pd1-O1 2.297(2), Pd1-P1
2.2898(9), Pd1-C1 2.042(4), Pd1-N1 2.081(3); O1-Pd1-P1 81.79(6), P1-
Pd1-C1 97.24(10), C1-Pd1-N1 92.98(13), N1-Pd1-O1 88.48(10).
The 1H NMR signals corresponding to the alkyl ligand in com-
plex 8a containing di-tert-butyl(o-anisyl)phosphine were assigned
by 2-dimensional NMR spectroscopy. An analysis of the 1H
NOESY correlations indicated that the alkyl ligand is positioned
such that the endo face is close to the phosphine ligand, and the exo
face is close to the anilido ligand. These results are consistent with
the distorted square planar, solid-state structure obtained by X-ray
diffraction. Therefore, any variation between the solid- and solu-
tion-state structures of this complex is small. Similar correlations
were observed in the NOESY spectrum of the analogous three-co-
ordinate, t-Bu3P-ligated complex 2i. This similarity suggests that
the geometry at the metal and orientation of the ligands in 8a are
similar to those in 2i (see the SI for spectra and assignments). Fur-
thermore, 19F NMR DOSY experiments with complex 8c (which
contains a second fluorine label on the ancillary ligand) confirmed
that this complex, and presumably the other P-O ligated complexes,
is monometallic in solution (Table 1, entry 4).
aConditions: Pd–Cl 7 (0.020 mmol), phosphine (0.040 mmol), 4-fluoro-
2-methoxyaniline (0.020 mmol), and LiHMDS (0.020 mmol) in Tol (2.0
ml). bThe reaction mixture containing 8a-d was heated at 65 °C with stirring
in a Teflon-capped vial for the noted time. cIsolated yields after recrystalli-
zation. dYields determined by 19F NMR spectroscopy integration against 4-
FTol; (values in parentheses are the overall yield of 3i based on 7a-d); av-
erages from two experiments. eConditions: isolated 8a (0.020 mmol) and t-
Bu2(o-anisyl)P (0.060 mmol) heated at 89 °C for 4 hours in Tol-d8 (0.70
ml).
Reductive elimination from four-coordinate alkylpalladium(II)
amido complexes containing rigid P,O ligands Complexes 8a-d
bearing 2-methoxyarylphosphine ligands on palladium underwent
reductive elimination of norbornylamine 3i in yields of 52-83% af-
ter 14 or 48 hours at 65 °C with 2 equivalents of added phosphine
to trap the Pd(0) product (Scheme 7). The yields of these reactions
depended on the substituents on the phosphine ligand; the highest
yield of 83% was obtained from the reaction of complex 8d con-
taining a 5-CF3-subsituent on the phosphine aryl group.
A comparison of the 13C NMR spectra of these complexes to
those of the three-coordinate Pd(II) chloride complexes revealed a
pronounced upfield shift of the resonance corresponding to C1 of
the alkyl ligand: 40.8 to 44.3 ppm for the four-coordinate com-
plexes 7a-d versus 57.9 to 64.2 ppm for the three-coordinate com-
plexes 1 and 4a-b. This difference in chemical shift suggests that
coordination of the ether to palladium causes a significant increase
in electron density at the alkyl ligand.
The reductive eliminations from four-coordinate complexes 8a-
d were 12 to 120 times slower than that from the analogous three-
coordinate, t-Bu3P-ligated complex 2i (Table 3). Previous studies
on concerted reductive elimination reactions from d8 transition
metal centers have led to the conclusion that the rates of reductive
elimination from four-coordinate, square planar complexes are typ-
ically slower than those from analogous three- or five-coordinate
complexes.20-21 Our observation that alkylpalladium amido com-
plexes containing P,O ligands undergo reductive elimination more
slowly than t-Bu3P-ligated complexes is consistent with this gen-
eral trend deduced from other classes of reductive eliminations.
However, reductive elimination still occurs in moderate to good
yield under mild conditions because the ether oxygen atom is a
weak electron donor (vide infra).
The reactions of the Pd(II) chloride complexes 7a-d with 4-
fluoro-2-methoxyaniline and LiHMDS with excess phosphine at
ambient temperature in toluene generated the corresponding Pd(II)
anilido complexes 8a-d in yields of 85-89% by NMR spectroscopy
(Scheme 7). These four-coordinate complexes were moderately
stable at ambient temperature, and samples suitable for analysis by
X-ray diffraction were obtained from crystallization of 8a and 8b
(Figure 4).
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 13
analyze this proposal experimentally, we prepared complexes with
a more flexible linker between the phosphorus and oxygen donors.
Table 3. Effects of substituents on the ancillary ligand on
the rate of reductive elimination.a
1
2
3
4
5
6
7
8
9
Synthesis of an alkylpalladium(II) complex containing a flexible
linker. The Pd(II) chloride precursor (Ad2PCH2CH2OCH3-
κ2P,O)Pd(2-CH3norbornyl)Cl 7e bearing a chelating 2-methoxy-
ethyl group on phosphorus was prepared by the same method as
described previously for the preparation of complexes 7a-d
(Scheme 6). After recrystallization, the sample contained a single
isomer in which the phosphine and alkyl ligands are mutually cis,
as determined by NMR spectroscopy. The aliphatic linker in this
methoxyethylphosphine is more flexible than the aryl linker in the
2-methoxyarylphosphines. The aliphatic ether unit in 7e is also ex-
pected to be a stronger electron donor than the aryl ether units in
7a-d. This assertion is consistent with the higher-field 13C NMR
chemical shift of the palladium-bound alkyl carbon in 7e than that
of the alkyl carbon in 7a-d (36.2 ppm for 7e versus 40.8-44.3 ppm
for 7a-d).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
entry
1b
2c
complex
kRE·105 /s-1
ΔGRE‡ /(kcal·mol-1)
23.8
26.9
26.9
25.4
25.8
25.7
2i
8a
8b
8c
8d
8e
ca. 300
2.7
Scheme 6. Preparation of (Ad2PCH2CH2OCH3-
κ2P,O)Pd(2-CH3norbornyl)NHAr (8e) and reductive elimi-
nation to form 3i.
3
2.9
4d
5e
25
14
6
18
aConditions: Pd-NHAr 8a-e (0.020 mmol) and phosphine (3 equiv) in
THF (0.70 ml) at 65 °C in a VT NMR probe; concentrations were deter-
mined by 19F NMR spectroscopy integration against 4-FTol; rate constants
and activation energies were determined from linear fits of ln[8] vs time
plots (1 half-life). bThe kRE and ΔGRE‡ for 2i at 65 °C were extrapolated from
the data in Figure 2b. cData averaged from two experiments. dComplex 8c
underwent significant (ca. 50%) unproductive decomposition during this
experiment; the rate constant is approximate and an overestimation.
eAttempts to isolate and purify 8d were unsuccessful; a solution of 8d in
THF was generated by the method described in Scheme 2 for this experi-
ment.
aIsolated yield after recrystallization. bYield determined by 19F NMR
spectroscopy integration against 4-FTol; average from two experiments.
cConditions: Pd–Cl 7e (0.134 mmol) and lithium (4-fluoro-2-methoxy-
phenyl)amide (0.136 mmol) in DCM. dConditions: Pd–Cl 7e (0.020 mmol),
phosphine (0.040 mmol), 4-fluoro-2-methoxyaniline (0.020 mmol), and
Small changes to the steric properties of the complexes did not
significantly influence the rate of reductive elimination. The rate of
reaction of the di-tert-butyl(o-anisyl)phosphine-ligated complex 8a
was similar to that of the di-adamantyl(o-anisyl)phosphine-ligated
complex 8b (Table 3, entries 2 and 3), and the solid-state structures
of these two complexes (Figure 4) were similar, as expected. At-
tempts to synthesize and investigate the reactivity of complexes
bearing less sterically-demanding P,O ligands were unsuccessful
(see the SI for details).
e
LiHMDS (0.020 mmol) in Tol (2.0 ml). The reaction mixture containing
8e and phosphine in toluene prepared as described aboved was heated at 65
°C with stirring in a Teflon-capped vial.
The reaction of the 2-methoxyethyl-ligated 7e with 4-fluoro-2-
methoxyaniline and LiHMDS in toluene produced the correspond-
ing Pd(II) anilido complex 8e in 90% yield by NMR spectroscopy
(Scheme 6). Complex 8e was isolated in a lower yield of 49% from
a similar reaction of 7e with amine and NaOt-Bu in THF. The flex-
ibility of the linker between P and O creates the possibility that iso-
meric, κP structures could exist in equilibrium with 8e (vide infra).
However, an analysis of the correlations in the 2-D 1H NOE spec-
trum indicates that a four-coordinate, κ2P,O structure such as 8e is
the dominant species in solution (see the SI). Specifically, correla-
tions were observed between the methoxy group of the ancillary
ligand and the ortho H of the anilido ligand.
The effects of the electronic properties of the phosphine on the
rate of reductive elimination were more readily accessed and were
significant. The rates of reductive elimination from complex 8c
containing the less electron-donating 5-F-substitued phosphine and
complex 8d containing the less electron-donating 5-CF3-
substituted phosphine were 7 and 6 times faster, respectively, than
that from 8a (Table 3, entries 2 and 4). These results suggest that
reductive elimination of alkylamines occurs faster from complexes
bearing less electron-donating ancillary ligands than from com-
plexes bearing more electron-donating ancillary ligands. This result
is similar to that observed previously for other classes of reductive
eliminations from palladium(II) complexes22 and is consistent with
our observation that t-Bu3P-ligated complexes are much more re-
active than the previously reported complexes containing more
electron-donating23-24 NHC ligands.
A crystal of 8e suitable for X-ray diffraction was obtained. The
solid-state structure of 8e was similar to those of 8a and 8b (Figure
5). The Pd–O bond length in 8e was 2.297 Å, which is only slightly
longer than the 2.288 and 2.283 Å Pd-O bond lengths in 8a and 8b,
respectively. Each of the other bond distances to palladium varied
by less than 0.04 Å between the three complexes, and the bond an-
gles around palladium varied by less than 5°. These results suggest
that the steric properties of the ligand in 8e are similar to those of
the ligands in 8a-d.
A Four-coordinate Alkylpalladium(II) Complex Contain-
ing a More Flexible P,O Ligand. Computational studies on the
mechanism of reductive elimination from complexes 8a-d contain-
ing rigid 2-methoxyarylphosphine ligands suggested that the vari-
ation in the length of the Pd–O bond in the ground state and in the
transition state would impact the rate of reaction (vide infra). To
6
ACS Paragon Plus Environment

Page 7 of 13
Journal of the American Chemical Society
Scheme 7. Preparation of the P,N-ligated [Ad2PCH2(2-
C5H4N)-κ2P,N]Pd(2-CH3norbornyl)NHAr (8f).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure
5.
ORTEP
drawing
[Ad2PCH2CH2OCH3-κ2P,O]Pd(2-
CH3norbornyl(NHAr) 8e with 50% probability ellipsoids. Hydrogen atoms
are omitted for clarity. Selected bond lengths (Å) and angles (°): Pd1-O1
2.297(2), Pd1-P1 2.2898(9), Pd1-C1 2.042(4), Pd1-N1 2.081(3); O1-Pd1-
P1 81.79(6), P1-Pd1-C1 97.24(10), C1-Pd1-N1 92.98(13), N1-Pd1-O1
88.48(10).
b
aIsolated yield after recrystallization. By neither GCMS nor 19F NMR
spectroscopy; 8f was also not detected by 19F NMR spectroscopy after heat-
ing.
Effect of the flexibility of the ligand backbone on the rate of re-
ductive elimination. Reductive elimination from complex 8e con-
taining the alkyl backbone was faster than that from complexes 8a-
b containing the more rigid aryl backbone. Complex 8e underwent
reductive elimination in the presence of 2 equivalents of ligand at
65 °C in toluene to form norbornylamine 3i in 92% yield. This yield
was higher than those from the reactions of any of the 2-methoxy-
arylphosphine-ligated complexes (8a-d). The rate constant for re-
ductive elimination from 8e was 6 times larger than that for reduc-
tive elimination from 8b, corresponding to a 1.2 kcal/mol lower
barrier (Table 3, entry 4). We propose that this difference in reac-
tivity is predominately due to the difference in flexibility between
the aliphatic and aryl linkers, and this proposal was corroborated
by computations (vide infra).
Computational Studies
Methods. Computational studies were performed to investigate
the effects of ancillary ligands on reductive elimination from palla-
dium(II) amido complexes. Density functional theory calculations
were conducted using Gaussian0926 with the hybrid exchange func-
tional B3LYP.27 The Pd atom was represented by the Ste-
vens/Basch/Krauss effective core potential and associated triple-ξ
valence basis set.28-30 The remaining main group atoms were repre-
sented by the 6-31+G(d) and 6-311++G(d,p) basis set for geometry
optimizations in the gas phase and single point calculations, respec-
tively. Solvent effects (THF, Ԑ = 7.4257) utilized the SMD31 con-
tinuum model via single point calculations on geometries opti-
mized in the gas phase. Calculations of the free energies of activa-
tion for reductive elimination from palladium(II) amido complexes
ligated by bulky N-heterocyclic carbene (NHC) ligands at the same
level of theory matched the value measured experimentally.14 The
ground states were determined to be minima on their potential en-
ergy surfaces by the absence of imaginary frequencies, while the
transition states were determined to be saddle points by the pres-
ence of a single imaginary frequency.
Effect of chelating group. To investigate how the electron-donat-
ing property of the chelating group influences the rate of reductive
elimination, we prepared [Ad2PCH2(2-C5H4N)-κ2P,N]Pd(2-
CH3norbornyl)(NHAr) 8f containing a picolinyl group on phospho-
rus by the methods previously described (Scheme 7). The 1H NMR
chemical shifts and NOE correlations observed for this complex are
consistent with the four-coordinate structure drawn. Complex 8f
underwent unproductive decomposition at 65 °C in either THF or
toluene, with no detectable formation of norbornylamine 3i. This
result implies that the barrier to reductive elimination from a com-
plex containing a strongly electron-donating fourth ligand (such as
a pyridine) located trans to the alkyl ligand is significantly higher
than that from a complex containing a weak donor (such as an
ether).25
Computational Results. 1. Mechanism of reductive elimination
from four-coordinate complexes. Reactions of the Pd(II) anilido
complexes 8a-e containing P,O ligands were investigated compu-
tationally to understand how the structure of the bidentate ancillary
ligands affects the rate of reductive elimination. The computed
ground state geometries were similar to those measured by X-ray
diffraction (Figure 4, Figure 5, and Figure 6). The computed free
energy barrier of 8a to reductive elimination was identical to that
from the experimental kinetic measurements (26.9 kcal/mol).
The transition state for reductive elimination resembles a migra-
tion of the alkyl ligand to the anilido nitrogen (Figure 7). The P–
Pd–C bond angle is computed to increase from 101 to 127°, and the
C–Pd–N bond angle to decrease from 95 to 57° during the reaction
of complex 8a to form alkylamine 3i. Similar results were observed
for reductive elimination from the other four-coordinate complexes
(see the SI for details). These changes in bond angles are consistent
with those previously reported for the reductive elimination of ar-
ylamines from phosphine- or NHC-ligated arylpalladium(II) amido
complexes.32-34
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 13
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8a
8e
9
9’
Figure 6. Computed ground state structures for [t-Bu2(o-anisyl)P-κ2P,O]Pd(2-CH3norbornyl)(NHAr) 8a and [Ad2PCH2CH2OCH3-κ2P,O]Pd(2-
CH3norbornyl(NHAr) 8e, and [Ad2PCH2CH2OCH3-κP]Pd(2-CH3norbornyl(NHAr) 9 and 9’. Bond lengths in Å, bond angles in degrees. Hydrogen atoms are
omitted for clarity.
suggest that reductive elimination occurs directly from the four-co-
ordinate complexes and through a four-coordinate transition state
(Figure 7).
Table 4. Calculated effects of ancillary ligands on reductive
elimination to form sp3 C–N bondsa
‡
ΔGRE
entry complex
Pd–O /Å
C–Pd–N /°
/(kcal·mol-1)
26.9
Figure 7. Computed transition state structure for reductive elimination of
norbornylamine 3i from [t-Bu2(o-anisyl)P]Pd(2-CH3norbornyl)(NHAr) 8a.
Bond lengths in Å, bond angles in degrees. Hydrogen atoms are omitted for
clarity.
1
2
3
4
5
8a
8b
8c
8d
8e
2.46 (2.89)
2.45 (2.87)
2.47 (2.90)
2.48 (2.88)
2.47 (3.87)
95.0 (56.9)
94.7 (56.5)
95.1 (57.2)
95.4 (57.1)
95.3 (58.6)
25.8
25.8
The distances from the palladium to the coordinated oxygen (Pd–
O) are calculated to increase along the reaction coordinate from the
ground states to the transition states (Figure 7, Table 4). The aver-
age increase in the Pd–O bond for the complexes 8a-d containing
rigid 2-methoxyaryl groups on the phosphine was only 17% of the
starting bond length (Table 4, entries 1-4). These Pd–O distances
are all significantly less than the sum of the van der Waals radii
(1.63 and 1.52 Å for Pd and O, respectively).35 The calculated free
energy barriers for this pathway agreed well with those measured
experimentally (Table 4 and Table 5). Therefore, computations
25.5
23.1
aPd–O and C–Pd–N values are from ground state calculations. Values in
parentheses are from the corresponding optimized transition state struc-
tures. Barriers are calculated with temperatures at 338 K.
The flexibility of the 2-methoxyethyl group in complex 8e cre-
ates the possibility that isomerization occurs to form a three-coor-
dinate species. An isomeric κP ground state structure (9) was lo-
cated in which the O donor of the P,O ligand has dissociated from
8
ACS Paragon Plus Environment

Page 9 of 13
Journal of the American Chemical Society
elimination. Therefore, we propose that the higher reactivity of 8e
the palladium center (Figure 6). This dissociation reaction was cal-
1
2
3
4
5
6
7
8
culated to be both exergonic and fast, with ΔG = -0.8 kcal/mol and
ΔG‡ = 3.2 kcal/mol (relative to 8e) (Figure 8). Furthermore, a con-
former of 9 (related by rotation of the P–Pd bond) was found and
the calculated energy of this conformer was 0.9 kcal/mol lower than
the energy of 9 (complex 9’, Figure 6). Therefore, 8e could isomer-
ize to a three-coordinate species prior to reductive elimination.
However, neither complex 9’ nor other three-coordinate isomers
were observed experimentally by NMR spectroscopy. The many
possible conformers of 9 complicates the process of predicting the
lowest energy species by DFT and the similarity in energy between
these four-coordinate and three-coordinate complexes suggests that
both may be relevant to the mechanism of reductive elimination
from complexes bearing the 2-methoxyethylphosphine ligand.
than of 8a or 8b is primarily a result of the smaller energy required
for the flexible ether ligand in 8e to dissociate from the metal center
during reductive elimination.
The possibility that an interaction between the ether group in the
anilido ligand and the palladium center affects the barrier to reduc-
tive elimination was evaluated. However, the distances from Pd to
the oxygen atoms in the anilido ligands of the transition states for
reductive elimination from 8a-e were over 4.71 Å. Therefore, the
effect of these interactions on the kinetic barrier to reductive elim-
ination is predicted to be small.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Computational results. 2. Mechanism of Reductive elimination
from complexes ligated by t-Bu3P. The t-Bu3P-ligated complexes
2a-i complexes were too unstable to isolate and characterize by X-
ray diffraction. Therefore, the computed structure of the ground
state of the three-coordinate, t-Bu3P-ligated alkylpalladium(II) am-
ido complex 2a bearing a phenyl group on nitrogen was obtained
by altering the structure of the four-coordinate, t-Bu2(o-anisyl)P-
ligated Pd(II) anilido complex 8a (Figure 9). The ortho-anisyl
group on phosphorus in this complex was replaced by a third tert-
butyl group, and the 4-fluoro-2-methoxyphenyl group on nitrogen
was replaced with a phenyl group. The energy of the resulting struc-
ture was then minimized, and the structures of the other t-Bu3P-
ligated complexes 2b-i were obtained by introducing the appropri-
ate substituents on the aryl group and minimizing the energy.
To gain more information from computation about the relation-
ship between the flexibility of the ancillary ligand and the rate of
reductive elimination, the pathway for simultaneous dissociation
and reductive elimination of the amine and the pathway for reduc-
tive elimination after dissociation of the ether oxygen to form 9
were evaluated. A transition state for reductive elimination directly
from the κ2P,O complex 8e was found and was 23.1 kcal/mol above
the ground state (Table 4, entry 5). In contrast to the change in Pd–
O bond distance during formation of the transition state for the re-
action of 8a bearing the rigid o-anisylphosphine, the increase in the
Pd–O distance during the reaction of complex 8e containing the
flexible 2-methoxyethyl group is large (approximately 57% of the
starting bond length). The Pd–O distance in the transition state for
reductive elimination from 8e is longer than what can be considered
a bond; the value of 3.87 Å is substantially greater than the sum of
the van der Waals radii (3.15 Å).36 Thus, reductive elimination and
dissociation of the ether occur simultaneously to generate this tran-
sition state. In addition, the transition states for reductive elimina-
tion from the κP isomers 9 and 9’ were found, and the energies of
these transition state were 0.9 and 1.1 kcal/mol lower, respectively,
than that of the transition state for the reaction of 8e (Figure 8). The
differences in calculated barriers between these three pathways are
likely less than the error in the calculations. Therefore, these com-
putational results strongly suggest that reductive elimination from
8e occurs with dissociation of the oxygen from the palladium cen-
ter, but they cannot distinguish clearly between pathways by which
dissociation occurs prior to or in concert with reductive elimina-
tion. These calculations indicate that the relative stabilities of κ2P,O
and κP isomers of complexes bearing flexible P,O ligands will af-
fect the rate of reductive elimination.
Figure 8. The free energy profile of three possible reductive elimination
pathways. Relative energies are reported at 338 K.
Figure 9. Computed ground state structure (top) and transition state struc-
ture (bottom) of [t-Bu3P]Pd(2-CH3norbornyl)(NHAr) 2i. Bond lengths in Å,
bond angles in degrees. Hydrogen atoms are omitted for clarity.
The computed and experimentally measured barriers to reductive
elimination from 2-methoxyethyl-ligated 8e were lower than those
from the analogous methoxyaryl-ligated 8b (Table 3 and Table 4).
This difference in reactivity is unlikely to result from differences in
the steric or electronic properties of the phosphine because the dif-
ferences in steric properties are small (Table 4, Figure 4, and Figure
6), and the greater electron donation by the ligand in 8e than by the
ligands in 8a-b should lead to an increase in the barrier to reductive
The transition state structures for reductive elimination of alkyl-
amines 3a-i from t-Bu3P-ligated Pd(II) amido complexes 2a-i were
computed (Figure 9 and Table 5). The average computed free en-
ergy barrier for reductive elimination of para- and meta-substituted
amines 3a-g was 22.9±0.7 kcal/mol (entries 1-7). This average
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 13
computed free-energy barrier agrees well with the average experi- strongly electron donating and one ether that is weakly electron do-
1
2
3
4
5
6
7
8
mental free-energy barrier of 23.4±0.6 kcal/mol (Table 2, entries 1-
7) for reductive elimination from the same complexes. However,
the experimentally observed correlation between the electronic
properties of the anilido ligand and the rate of reductive elimination
from 2a-g presented in Table 2 and Figure 1 was not well repro-
duced by DFT, most likely because the variations in computed bar-
riers between these complexes are within the accuracy of the cal-
culations.
nating. An analysis of the factors controlling the rate of reductive
elimination revealed that the fastest reductive eliminations oc-
curred from complexes bearing unhindered and electron-rich ani-
lido ligands. The same trend was observed for reactions of the t-
Bu3P-ligated alkylpalladium(II) complexes reported herein as for
the analogous NHC-ligated complexes previously reported in com-
munication form14 and for arylpalladium(II) complexes that un-
dergo reductive elimination of arylamines.1, 37 DFT calculations
predict that conformational effects, such as rotation around the C–
N bond, also have a significant impact on the rate of reductive elim-
ination.
The computational results suggest that conformational effects
may contribute substantially to the rate of reductive elimination.
The barriers to reductive elimination from conformational isomers
(by rotation of the C–N bond) of meta- and ortho-substituted com-
plexes 2e-i were computed, and this rotation was found to have a
significant effect on the barrier to reductive elimination. In the
cases of complexes 2e and 2g, reductive elimination is calculated
to occur after rotation of the C–N bond; the lower-energy transition
state corresponds to the higher-energy ground state. Rotation of the
C–N bond in ortho-substituted 2h and 2i was predicted to be en-
dergonic by 7.3 and 4.2 kcal/mol, respectively. The computed free
energy barrier to reductive elimination from the more sterically en-
cumbered, ortho-substituted 2h and 2i (25.2 and 23.4 kcal/mol, re-
spectively) agreed exceptionally well with those measured experi-
mentally (25.5 and 23.5 kcal/mol, Table 2). The general agreement
between the computed barriers and experimentally measured barri-
ers for reactions from palladium(II) amido complexes containing a
t-Bu3P ancillary ligand further support the conclusion that the reac-
tion occurs by a concerted pathway for reductive elimination to
form alkyl–nitrogen bonds.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Syn-2-methylnorbornylpalladium(II) anilido complexes contain-
ing monophosphines smaller than t-Bu3P form bimetallic structures
containing one phosphine and one bridging anilide as the dominant
species in solution. Kinetic experiments conducted with added
phosphine suggest that reductive elimination from this bimetallic
complex occurs via reversible formation of a three-coordinate mon-
ometallic species. Our results indicate that this reaction is reversi-
ble when phosphine is present in high concentrations. However, the
unproductive formation of anilide-bridged bimetallic complexes
would likely limit the utility of these complexes as catalysts in the
development of new synthetic methods for the production of alkyl-
amines.
Four-coordinate Pd(II) anilido complexes containing neutral, bi-
dentate 2-methoxyarylphosphine (P,O) ligands are monometallic
and stable at ambient temperature, but they undergo reductive elim-
ination at mild temperatures. The barriers to reductive elimination
from these four-coordinate complexes are 1.6 to 3.2 kcal/mol
higher than that for reductive elimination from the analogous three-
coordinate, t-Bu3P-ligated complex. Computations indicate that
complexes of the rigid P,O ligands undergo reductive elimination
directly from the four-coordinate complexes, but the palladium–
oxygen bond lengthens significantly from the ground state to the
transition state.
Table 5. Effects of anilido substituents on the calculated
barrier to reductive elimination from 2a-i to form 3a-i.a
The reductive eliminations from complexes ligated by 2-meth-
oxyaryl phosphines bearing electron-withdrawing groups were
faster than those from complexes ligated by unsubstituted, o-ani-
sylphosphines. Furthermore, the reductive elimination from the t-
Bu3P-ligated complexes are approximately 40 times faster than that
from the analogous, more electron-rich NHC-ligated complexes,38
corresponding to a difference in free energy of activation of ap-
proximately 2.7 kcal/mol. These results are consistent with the
trend that reductive elimination occurs more rapidly from more
electron-deficient complexes than from more electron-rich com-
plexes, which is consistent with the trends reported for other classes
of concerted reductive eliminations from Pd(II).22
ΔGRE‡ /(kcal·mol-1)a
entry
X
1
2
H (2a)
21.9
4-OMe (2b)
4-Me (2c)
4-F (2d)
22.7
3
22.8
4
23.4
5
3-Me (2e)
3-OMe (2f)
3-F (2g)
22.6 (1.0), 23.7
22.9, 23.4 (0.8)
24.3 (1.6), 24.7
25.2, 32.8 (7.3)
23.4, 26.7 (4.2)
Complexes containing chelating ligands with flexible backbones
have been shown to undergo reductive elimination to form C–C
bonds faster than those with more rigid backbones. Goldberg and
coworkers have reported that the rates of reaction to form sp3 C–C
bonds from (P,P)Pt(IV)Me4 complexes ligated by flexible bisphos-
phines are faster than the rates of the same reaction from complexes
containing ligands with equal bite angles but more rigid linkers.39-
40 The authors proposed that reductive elimination occurs via initial
dissociation of one phosphine donor to form a penta-coordinate in-
termediate, and that the energy difference between hexa-and penta-
coordinate structures is smaller for complexes ligated by flexible
bisphosphine ligands than for complexes ligated by more rigid
bisphosphines.41-42
6
7
8b
2-t-Bu (2h)
2-OMe-4-F (2i)
9
aΔG‡ values for both conformers by C–N rotation reported relative to
the lower ground state; (ΔGisom, the relative energy of the isomeric ground
state, reported in parentheses). bCalculated temperatures at 338 K.
CONCLUSIONS
We have shown that reductive elimination to form sp3 C–N
bonds by a concerted pathway can occur with low barriers from
monomeric three-coordinate phosphine-ligated (syn-2-methylnor-
bornyl)palladium(II) complexes and from four-coordinate analogs
containing chelating ligands comprising one phosphine that is
We have shown that this trend can be taken even further by com-
bining the effects of flexibility in the linker with weak donation by
an ether. The four-coordinate (κ2P,O) and three-coordinate (κP)
isomers of an alkylpalladium(II) anilido complex bearing the flex-
ible P,O ligand Ad2PCH2CH2OCH3 are computed to be similar in
10
ACS Paragon Plus Environment

Page 11 of 13
Journal of the American Chemical Society
6.
4423.
7.
Koo, K.; Hillhouse, G. L., Organometallics 1995, 14 (9), 4421-
Pawlikowski, A. V.; Getty, A. D.; Goldberg, K. I., J. Am. Chem.
energy, and the reductive elimination reaction from this complex is
1
2
3
4
5
6
7
8
faster and higher yielding than that from the analogous complex
containing a rigid linker. Computations suggest that the rigid, aryl
backbones in ortho-anisyl phosphines do not allow for dissociation
of the ether to form the three-coordinate isomer. We suggest that
the effects of weak chelation on the mechanism of reductive elimi-
nation could be elements of design for catalysts that react by reduc-
tive elimination to form alkyl–nitrogen bonds.
Soc. 2007, 129 (34), 10382-10393.
8.
Chem. Soc. 2012, 134 (15), 6571-6574.
9.
Rucker, R. P.; Whittaker, A. M.; Dang, H.; Lalic, G., J. Am.
Lin, B. L.; Clough, C. R.; Hillhouse, G. L., J. Am. Chem. Soc.
2002, 124 (12), 2890-2891.
10.
931.
11.
Chem. 2001, 66 (24), 8127-34.
Catellani, M.; Del Rio, A., Russ. Chem. Bull. 1998, 47 (5), 928-
The trends revealed by the current work suggest strategies to in-
crease the rate of reductive elimination and to prevent unproductive
decomposition pathways. Specifically, we have shown that coordi-
nation by P,O bidentate ligands can stabilize reactive Pd(II) anilido
intermediates without significantly decreasing the yields from re-
ductive elimination. The square planar structures of these com-
plexes are expected to lead to higher barriers to unproductive de-
composition pathways, such as b-hydride elimination. Further-
more, chelation should disfavor decomposition pathways involving
phosphine dissociation or formation of bimetallic, anilide-bridged
complexes. Therefore, we expect that ligand structures like those
in Scheme 5 and Scheme 6 will enable an expansion of the scope
of alkyl and amido groups that undergo reductive elimination.
Lautens, M.; Paquin, J. F.; Piguel, S.; Dahlmann, M., J. Org.
9
12.
Pan, J.; Su, M.; Buchwald, S. L., Angew. Chem. Int. Ed. 2011,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
50 (37), 8647-8651.
13.
Int. Ed. 2010, 49 (4), 793-796.
14.
Am. Chem. Soc. 2012, 134 (37), 15281-15284.
15.
6115-6128.
16.
Organometallics 2003, 22 (13), 2775-2789.
Marquard, S. L.; Rosenfeld, D. C.; Hartwig, J. F., Angew. Chem.
Hanley, P. S.; Marquard, S. L.; Cundari, T. R.; Hartwig, J. F., J.
Low, J. J.; Goddard, W. A., J. Am. Chem. Soc. 1986, 108 (20),
Mann, G.; Shelby, Q.; Roy, A. H.; Hartwig, J. F.,
17.
Attempts to isolate a crystalline sample of complex 6a were
unsuccessful.
18.
19.
Tolman, C. A., Chem. Rev. 1977, 77 (3), 313-348.
The substituent contributions to the Tolman electronic
ASSOCIATED CONTENT
Supporting Information
paramater are 0, 4.3, and, 0.1 for t-Bu, Ph, and Cy, respectively.
20. DiCosimo, R.; Whitesides, G. M., J. Am. Chem. Soc. 1982, 104
(13), 3601-3607.
Experimental details, NMR spectra, crystallographic data, compu-
tational details, and optimized coordinates. This material is availa-
ble free of charge via the Internet at http://pubs.acs.org.
21.
Tatsumi, K.; Nakamura, A.; Komiya, S.; Yamamoto, A.;
Yamamoto, T., J. Am. Chem. Soc. 1984, 106 (26), 8181-8188.
22.
23.
24.
Hartwig, J. F., Inorg. Chem. 2007, 46 (6), 1936-1947.
Gusev, D. G., Organometallics 2009, 28 (22), 6458-6461.
Kelly Iii, R. A.; Clavier, H.; Giudice, S.; Scott, N. M.; Stevens,
AUTHOR INFORMATION
E. D.; Bordner, J.; Samardjiev, I.; Hoff, C. D.; Cavallo, L.; Nolan, S. P.,
Organometallics 2008, 27 (2), 202-210.
25.
BrC5H3N))]Pd(2-CH3norbornyl)(Cl) to test whether decreasing the basicity
of the pyridine donor would enable reductive elimination to occur.
However, only trace formation of 3i was detected in reactions of this
complex (see the SI for details).
Corresponding Authors
jhartwig@berkeley.edu
t@unt.edu
We also synthesized the complex [Ad2PCH2(2-(5-
Present Addresses
P.S.H. Core Research & Development, The Dow Chemical Com-
pany, 1776 Building, Midland, Michigan 48674, United States.
26.
Gaussian 09, Revision D.01, M.J. Frisch, G.W. Trucks, H.B.
Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scal-mani, V.
Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery, Jr.,
J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin,
V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell,
J. C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M.
Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R.
Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli,
J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth,
P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.B.
Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian, Inc., Wallingford
CT, 2013.
27.
28.
5555-5565.
29.
(12), 6026-6033.
30.
Chem. 1992, 70 (2), 612-630.
31.
B 2009, 113 (18), 6378-6396.
32.
425.
33.
Organ, M. G., Chem. Eur. J. 2012, 18 (1), 145-151.
34. Klinkenberg, J. L.; Hartwig, J. F., J. Am. Chem. Soc. 2010, 132
(34), 11830-11833.
Notes
The authors declare no competing financial interests.
ACKNOWLEDGMENT
The authors gratefully acknowledge funding from the National Sci-
ence Foundation Center for Enabling New Technologies through
Catalysis (Grant CHE-1205189). X-ray diffraction crystallography
was performed by Dr. Antonio DiPasquale at the UC Berkeley
CheXRay facility (NIH S10- RR027172). We thank Dr. Hasan Ce-
lik at the UC Berkeley NMR Facility for his assistance with DOSY
Becke, A. D., J. Chem. Phys. 1993, 98 (7), 5648-5652.
Cundari, T. R.; Stevens, W. J., J. Chem. Phys. 1993, 98 (7),
1
and H-31P HMBC NMR spectroscopy experiments. T.R.C. and
Q.J. also acknowledge the National Science Foundation for their
support of the UNT Chemistry CASCaM high performance com-
puting facility through grant CHE-1531468.
Stevens, W. J.; Basch, H.; Krauss, M., J. Chem. Phys. 1984, 81
Stevens, W. J.; Krauss, M.; Basch, H.; Jasien, P. G., Can. J.
REFERENCES
Marenich, A. V.; Cramer, C. J.; Truhlar, D. G., J. Phys. Chem.
1.
Driver, M. S.; Hartwig, J. F., J. Am. Chem. Soc. 1997, 119 (35),
Yamashita, M.; Hartwig, J. F., J. Am. Chem. Soc. 2004, 126 (17),
Pendleton, I. M.; Pérez-Temprano, M. H.; Sanford, M. S.;
Cundari, T. R.; Deng, J., J. Phys. Org. Chem. 2005, 18 (5), 417-
8232-8245.
2.
5344-5345.
3.
Hoi, K. H.; Çalimsiz, S.; Froese, R. D. J.; Hopkinson, A. C.;
Zimmerman, P. M., J. Am. Chem. Soc. 2016, 138 (18), 6049-6060.
4. Pérez-Temprano, M. H.; Racowski, J. M.; Kampf, J. W.;
Sanford, M. S., J. Am. Chem. Soc. 2014, 136 (11), 4097-4100.
5. Camasso, N. M.; Canty, A. J.; Ariafard, A.; Sanford, M. S.,
Organometallics 2017, 36 (22), 4382-4393.
35.
36.
Bondi, A., J. Phys. Chem. 1964, 68 (3), 441-451.
Attempts to locate a transition state structure in which the Pd-O
distance is within the summ of the VdW radii were unsucessful.
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 13
37.
Chem. Soc. 2003, 125 (52), 16347-16360.
38. The rate constants for reductive elimination of 2b and 2c were
calculated at 363 K, assuming an entropy of activation of -14 eu. Solvent
effects are expected to be small.
Yamashita, M.; Cuevas Vicario, J. V.; Hartwig, J. F., J. Am.
1
2
3
4
5
6
7
8
39.
Arthur, K. L.; Wang, Q. L.; Bregel, D. M.; Smythe, N. A.;
O'Neil, B. A.; Goldberg, K. I.; Moloy, K. G., Organometallics 2005, 24
(19), 4624-4628.
40.
Crumpton-Bregel, D. M.; Goldberg, K. I., J. Am. Chem. Soc.
2003, 125 (31), 9442-9456.
41.
Brown, M. P.; Puddephatt, R. J.; Upton, C. E. E., J. Chem. Soc.,
Dalton Trans. 1974, (22), 2457-2465.
9
42.
Roy, S.; Puddephatt, R. J.; Scott, J. D., J. Chem. Soc., Dalton
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Trans. 1989, (11), 2121-2125.
12
ACS Paragon Plus Environment

Page 13 of 13
Journal of the American Chemical Society
1
2
3
4
Insert Table of Contents artwork here
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
ACS Paragon Plus Environment