Mechanistic investigation of the palladium-catalyzed synthesis of allylic silanes and boronates from allylic alcohols

Article abstract of DOI:10.1021/ja309860h

The mechanism of the palladium-catalyzed synthesis of allylic silanes and boronates from allylic alcohols was investigated. ¹H, ²⁹Si, ¹⁹F, and ¹¹B NMR spectroscopy was used to reveal key intermediates and byprod

Full text of DOI:10.1021/ja309860h

Subscriber access provided by FORDHAM UNIVERSITY
Article
A Mechanistic Investigation of the Palladium-Catalyzed
Synthesis of Allylic Silanes and Boronates from Allylic Alcohols
Johanna M. Larsson, and Kalman J. Szabo
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/ja309860h • Publication Date (Web): 30 Nov 2012
Downloaded from http://pubs.acs.org on December 3, 2012
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 free 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 accessible to all
readers and 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.
Journal of the American Chemical Society 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 14
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
A Mechanistic Investigation of the Palladium-Catalyzed Synthesis of
Allylic Silanes and Boronates from Allylic Alcohols
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
Johanna M. Larsson and Kálmán J. Szabó*
Department of Organic Chemistry, Stockholm University, SE-106 91 Stockholm, Sweden
Supporting Information Placeholder
ABSTRACT: The mechanism of the palladium-catalyzed synthesis of allylic silanes and boronates from allylic alcohols was inves-
tigated. 1H-, 29Si-, 19F- and 11B NMR spectroscopy was used to reveal key intermediates and by-products of the silylation reaction.
The tetrafluoroborate counterion of the palladium catalyst is proposed to play an important role in both catalyst activation as well as
the transmetallation step. We propose that BF3 is generated in both processes and is responsible for the activation of the substrate
hydroxyl group. An (η3-allyl)palladium complex has been identified as the catalyst resting state and the formation of (η3-
allyl)palladium complexes directly from allylic alcohols has been studied. Kinetic analysis provides evidence that the turnover
limiting step is the transmetallation and insights into notable similarities between the borylation and the silylation reaction mecha-
nisms enabled us to considerably improve the stereoselectivity of the borylation.
that the commercially available palladium complex
INTRODUCTION
[Pd(MeCN)4](BF4)2 (3) catalyzes the same transformation at
The direct activation of allylic alcohols in transition metal
lower temperatures with shorter reaction times. Importantly,
catalysis has recently emerged as an attractive approach for
we found that replacing the boron source with (SiMe3)2 (2) in
the formation of new carbon-carbon and carbon-heteroatom
this system enables the unprecedented conversion of allylic
bonds.1,2 From an economic perspective, these stable and
alcohols directly to the corresponding silanes in good yields
readily available substrates are preferable to the more com-
(Scheme 2).2a Both reactions proceed with remarkably high
monly used allylic carboxylates, carbonates and halides, which
regio- and stereoselectivity to afford the linear allylic products
are themselves usually prepared from the corresponding allylic
with trans configuration. Previous methods published by
alcohols.1b,c,2f
Miyaura, Tsuji and Lipshutz for the synthesis of allylic silanes
Allylic silanes and allylic boronates are important building
and boronates have required pre-installed leaving groups, such
blocks in modern organic synthesis.3 Their low toxicity and
relatively high stability makes them extremely useful reagents
in the regio- and stereoselective allylation of electrophiles3h,4
as well as useful transmetallating reagents in transition metal-
catalyzed reactions, such as Hiyama or Suzuki-Miyaura cou-
plings.3h,5,6 As a result, considerable effort has gone into the
development of attractive methods for the preparation of or-
gano silanes and boronates using transition metal catalysis.7
as acetates or trifluoroacetates.7b,7d,e
.
e u v.
e
SiM
3)2 ( )
2
1 2
i
q
(
H
O
mo
e
5
l
Pd M CN BF
3
%
[
]
) (
4
(
4)2 ( )
R
R
e
iM
3
R
S
:
e
:
H 1 1
O
,
DM
M
SO
4
H
O
exam es
l
p
50 °C 15 h
8
-
e
i ld
1
52 84
%
y
.
e u v.
n
i
2p
5
1 2
i
B
q
2 ( )
Pd M CN BF
3
4)2 ( )
H
O
mo
l
e
5
%
[
]
) (
4
(
R
R
n
B i
p
R
one- o
t
p
H
O
mo
l
e
: :
DMSO M OH 1 1
5
(
%
)
L
Pd
X
L
reac ons w
ti
ith
6
e ec ro es
t
OH
exam es
l
hil
r
.
l
p
t, 1 5 h
3
p
R
R
-
7
e
i ld
%
y
1
7
0
8
R
B R
O
( )
2
.
e u v.
1 2
i
B2 OR
q
(
)
4
H
O
Scheme 2. Pd(II)-catalyzed silylation and borylation of allylic
alcohols under mild conditions developed by the Szabó
group.2a
e
: :
M H 1 1
SO O
DM
1
°
r -
t
60 C
Scheme 1. Synthesis of allylboronates from allylic alcohols
In our system allylic alcohols underwent silylation much faster
than did allylic acetates2a - a surprising result given the poor
nucleofugicity of the hydroxyl group. This result, and the fact
that silylation and borylation of allylic alcohols have not yet
been studied in detail, inspired us to undertake a mechanistic
investigation. It was previously found that the silylation and
borylation behave similarly in many respects.2a In both reac-
tions [Pd(MeCN)4](BF4)2 was shown to be the most efficient
catalyst source. Addition of more strongly coordinating lig-
using pincer complex catalysis developed by the Szabó group
-
(L=SePh, X=Cl; L=SMe, X=BF4 ).2b,8
We have previously reported the palladium-catalyzed conver-
sion of allylic alcohols to allylic boronates under mild condi-
tions2b,8 (Scheme 1) and shown that these reagents can be used
subsequently in allylation reactions to form stereo-defined
homoallylic alcohols and amines9 as well as cross-coupling
products with unusual regioselectivity.10 Recently, we showed
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 14
ands such as phosphines, LiCl and LiOAc were found to in- decreasing, due to the formation of hexamethyldisiloxane (10)
1
2
3
hibit both reactions. Furthermore, both the silylation and the
borylation proceed faster with allylic alcohols as substrates
than with allylic acetates.
(Scheme 4) which is observed four hours after the start of the
reaction (Figure S1).11
4
5
6
7
We
focused
on
the
catalytic
performance
of
[Pd(MeCN)4](BF4)2, which proved to be the most reactive
catalyst in our previous studies. In particular, attention was
directed to the following aspects:
8
9
•
•
How is the hydroxyl group activated?
What is the key organopalladium intermediate of the
reaction?
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
•
•
What is the turnover limiting step of the reaction?
What are the similarities and differences between the
silylation and borylation reactions?
We initially chose to investigate the silylation reaction, given
the higher stability of the allylic silane products compared to
the allylic boronate products. The silylation is also slower,
making it a more obvious choice for easier spectroscopic
detection of intermediates as well as kinetics experiments.
Figure 1. 29Si NMR spectrum of the crude reaction mixture after
complete conversion of 1a to 4a (Scheme 3).
In this study we show that the tetrafluoroborate anion releases
BF3, which is able to activate allylic alcohol substrates. We
identify (η3-allyl)palladium species as the key intermediate
and find that transmetallation is the rate-limiting step for the
silylation. Similarities between the silylation and borylation
reactions allow for a substantial improvement in the diastere-
oselectivity of the borylation.
Except the 29Si NMR signals of 8-d3, 9 and 10, 15 minutes
after starting the reaction a doublet at 32 ppm (JSi-F = 274 Hz)
was observed (Figure S1), which was identified as the reso-
nance signals of Me3Si-F (7). This was a surprising finding, as
the BF4- catalyst counterion was the only possible fluorine
source present in the reaction mixture. Me3Si-F formed rapidly
at an early stage of the reaction and its amount continuously
increased during the time the reaction was monitored (11
hours, Figure S1).
RESULTS AND DISCUSSION
As a model reaction we chose the silylation of commercially
available 1a, for which the reaction is clean, high yielding and
gives a stable product. Deuterated solvents were used for ease
of analysis (Scheme 3).
.
e u v.
e
SiM
3)2 ( )
H
1 2
i
2
O
q
(
mo
l
e
5
Pd M CN BF
3
e
iM
S
3
%
[
]
(
(
)
)
4 2 ( )
4
a
4
a
1
-
:
e
-
:
DM
d
6 M H d4 1 1
SO
O
82
%
,
50 °C 15 h
so a e
e
i
l t d i ld
y
Scheme 3. Model reaction for studying the silylation reaction
of allylic alcohol 1a.2a
Multinuclear NMR Analysis. We sought initial insights by
examining reaction intermediates and by-products. To start,
the course of the silylation of 1a (Scheme 3) was followed by
29Si NMR spectroscopy. During the reaction the consumption
of (SiMe3)2 (2) and the formation of the silylated product (4a)
were observed throughout (Figure 1 shows the 29Si NMR
spectrum of the mixture at the end of the reaction; see Figure
S1 in the supporting information for changes observed over
time).
Figure 2. 19F NMR spectrum of the crude reaction mixture after
complete conversion of 1a to 4a (Scheme 3).
O
+
+
e
M
e
3S
i
D
M
i
H
O
D
3C O
H
3S OC
3
e
e
iM
S
3
M
i
3S
-
To find out more about the role of BF4 , the reaction was also
monitored by 19F NMR spectroscopy (Figure 2 shows the 19F
NMR spectrum of the mixture at the end of the reaction; see
Figure S2 in the SI for changes observed over time). As boron
has two naturally occurring isotopes, 10B (20% natural abun-
dance) and 11B (80% natural abundance), compounds contain-
ing B-F bonds are easily identifiable in 19F NMR spectroscopy
-
8
9
10
d3
Scheme 4 Condensation of 8-d3 and 9 forming 10.
After 1.5 hours Me3Si-OD3 (8-d3) appeared and after 2.5 hours
a small peak corresponding to Me3Si-OH (9) (11.60 ppm) was
also observed. Thereafter, the concentration of 8 and 9 started
ACS Paragon Plus Environment

Page 3 of 14
Journal of the American Chemical Society
where F-10B and F-11B give rise to two peaks in close proximi-
Intermediate Formation of Allylic Ether 12. Running our
ty to each other in a 1:4 ratio (e.g. see signals of BF4- in Figure
2).12 At the start of the reaction formation of BF3 (11) was
observed.13 After 10 minutes, the concentration reached its
maximum and then started decreasing. After 2 hours the con-
centration of BF3 was too low to be observed. However, the
tetrafluoroborate anion continued being consumed, suggesting
that BF3 is formed throughout the reaction (Figure S2 in the
SI). At the end of the reaction 5 % of BF4- still remained.
model reaction (Scheme 3) at room temperature instead of 50
°C for 15 hours gave a product mixture containing allylic
silane 4a, linear allylic ether 12a-d3 and a small amount of
12b-d3 (Scheme 5). Also if the model reaction at 50 °C is
stopped after 3 hours, before full conversion to 4a, formation
of 12a-d3 and 12b-d3 can be observed by 1H NMR spectrosco-
py. However, after 15 hours they are fully consumed. We
therefore wanted to find out which role these possible inter-
mediates play in the silylation reaction. Running our model
reaction without disilane 2 for 15 hours gave full conversion to
the allylic ethers (Scheme 6). The linear ether (12a-d3) was
formed in 87% NMR yield together with a small amount of
branched ether (12b-d3). The addition of (SiMe3)2 to the crude
mixture of this reaction and heating for a further 15 hours at
50 °C led only to a low conversion (35%) to allylsilane 4a due
to decomposition of the palladium catalyst. By contrast, when
both (SiMe3)2 and a new portion of palladium catalyst 3 (5
mol%) were added to the crude reaction mixture (which was
then heated for 15 hours), ethers 12a-d3 and 12b-d3 were con-
sumed and allylsilane 4a was formed in 69% NMR yield to-
gether with 13 (23% NMR yield) (Scheme 6). It is important
to point out that a much smaller amount of 13 (approximately
2%) was detected during our model reaction (Scheme 3).
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
D
OC
H
O
Figure 3. 11B NMR spectrum of the crude reaction mixture after
complete conversion of 1a to 4a (Scheme 3).
3
mo
5
C
l
%
7
8
%
a
-
d3
12
e
Pd M
N
)
BF
]
(
3
[
(
4)2 ( )
4
D
OC
3
e
- : - :
SO O
d
DM 6 M H d4 1 1
Likewise, the 11B NMR spectrum of this mixture showed a
small amount of BF4- still present and also that the remainder
of the boron containing compounds were in the form of
B(OR)3 (Figure 3). The broad, partially overlapping peaks at
18.5 and 17.8 ppm correspond to the equilibrium between
B(OH)3 and B(OCD3)3 in the DMSO-d6/MeOH-d4 solvent
mixture, as verified by comparing against genuine samples of
B(OMe)3 and B(OH)3. The tetrafluoroborate anion is known to
undergo slow hydrolysis in aqueous solution.14 We therefore
investigated the stability of the counterion of the palladium
catalyst 3 in the solvent mixture. Accordingly, 3 was heated at
50 °C in a 1:1 mixture of DMSO-d6 and MeOH-d4 for 15
hours, at which time 19F NMR analysis of the solution did not
show appreciable changes. Palladium black formation could
not be observed and formation of hydrogen fluoride (HF) was
not detected. The catalytic model reaction was also repeated in
a teflon-lined NMR tube. However, even though the direct
contact of HF with the glass surface was prevented, evolution
of HF could not be observed.
,
C
50 ° 15 h
a
1
2
%
-
12b
d3
.
iM
e u v.
i
q
3)2 ( )
1 2
S
e
2
(
e
iM
S
3
+
mo
5
C
l
%
a
4
13
23
e
Pd M
(
N
BF
]
(
3
[
)
)
4 2 ( )
4
,
69
%
%
50 ° 15 h
C
Scheme 6. Sequential etherification-silylation. The yields are
determined by 1H NMR spectroscopy using an internal stand-
ard.
To test whether Pd(0) might be responsible for the high yield
of by-product 13 in the sequential reaction, 12a was used
directly as a substrate in the silylation (Scheme 7). Under the
standard conditions 12a was converted to 4a in high NMR
yield and the formation of 13 became comparable to that of
our model reaction (6% NMR yield). This finding suggests
that the formation of 13 is promoted by Pd(0) (see discussion
below).
e
iM
S
3
31
%
e
iM
S
3
.
e u v.
e
2
1 2
i
SiM
q
.
e u v.
e
a
4
2
(
)
3 2 ( )
1 2
i
SiM
l
%
]
q
H
O
(
3)2 ( )
mo
N
a
4
5
C
l
%
mo
N
5
C
e
Pd M
BF
]
(
3
D
OC
[
(
)
4
)
4 2 ( )
e
Pd M
BF
3
4
8 %
3
1
9
%
[
(
)4 ( 4)2 ( )
e
M
O
a
-
12
d3
a
1
-
:
e
-
:
DM
SO
d
r
6 M H d4 1 1
,
O
a
12
e
- : - :
d
DM 6 M H d4 1 1
SO O
D
OC
3
t 15 h
,
50 ° 15 h
C
13
2
%
-
12b
d3
6
%
Scheme 7. The model silylation reaction performed using 12a
Scheme 5. The model silylation reaction performed at room
1
1
as a substrate. The yields are determined by H NMR spec-
temperature. The yields are determined by H NMR spectros-
troscopy using an internal standard.
copy using an internal standard.
Having established that allylic ether 12a can be converted to
silane 4a, we sought to establish the extent of ether formation
in our model reaction. Thus, a competition experiment be-
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 14
tween allylic alcohol 1a and allylic ether 12a under the condi-
tions of our model reaction was followed by H NMR spec-
Scheme 3) we demonstrated above, may be of key importance
for the activation of the hydroxyl group. Indeed, when we
1
1
2
3
4
5
6
7
8
9
troscopy for 13 hours (Figure 4). As MeOH-d4 acts as the only
source of -OCD3, the use of non-deuterated 12a enabled easy
differentiation between starting material (12a) and product
(12a-d3) ethers. The allylic alcohol 1a was consumed rapidly
and the conversion of 12a was observed to begin only after 1a
was completely consumed. This shows that silane 4a is not
exclusively produced via ether 12a. In addition, silane 4a is
formed at a higher rate than 12a-d3 suggesting that 4a is main-
ly formed directly from 1a. The finding that ether 12a-d3 is
formed from both 1a and 12a indicates that 12a-d3 may be a
common intermediate in conversion of both 1a and 12a. The
relatively constant concentration of 12a-d3 after 2 hours is the
result of two opposing processes. The concentration of allylic
ether 12a-d3 increases due to the formation from 12a, while at
the same time 12a-d3 is consumed by transformation to silane
4a. The formation of 12a from 12a-d3 is not observed due to
the considerably higher concentration of CD3OD compared to
that of CH3OH.
conducted the reaction shown in Scheme 8 in the presence of
15 mol% of BF3·OEt2, it afforded allylic silane 4a in 72%
NMR yield together with 14% 12a-d3 (Scheme 9). This find-
ing clearly shows that a catalytic amount of BF3 is able to
activate efficiently the hydroxyl group towards palladium-
catalyzed substitution under very similar conditions to our
model reaction.
.
e u v.
e
SiM
3)2 ( )
2
1 2
i
q
(
⋅
mo
15
l
BF OEt
%
e
iM
3
2
S
3
72
H
%
%
O
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
a
4
.
mo
l
a
14
)
2 5
P
d2 db
%
(
)
3
(
a
1
-
:
e
-
:
DM
d
6 M H d4 1 1
SO
O
D
OC
3
14
a-
12
,
50 °C 15 h
d3
Scheme 9. Silylation using Pd2(dba)3 1and BF3·OEt2 as cata-
lysts. The yields are determined by H NMR spectroscopy
using an internal standard.
D
OC
3
28
%
%
.
e u v.
e
2
a
-
d3
1 2
i
SiM
12
q
(
)
3 2 ( )
H
O
⋅
mo
l
15
BF OEt
%
3
2
D
OC
3
a
1
-
e
: - :
6 M H d4 1 1
O
DM
d
SO
40
,
-
12b
d3
50 °C 15 h
no s a on
il l ti
y
Scheme 10. Attempted silylation without palladium catalyst.
1
The yields are determined by H NMR spectroscopy using an
internal standard (A reaction profile can be found in the SI).
As expected, silylation did not occur in the absence of palladi-
um, even in the presence of BF3·OEt2 (Scheme 10). However,
12a-d3 and 12b-d3 did form, presumably either via a less regi-
oselective SN2/SN2’ and/or SN1 mechanism.
The Resting State of the Catalyst. When the course of our
1
model silylation reaction (Scheme 3) was monitored by H
Figure 4. Competitive silylation reaction using 12a and 1a as
starting materials. Lines are given as a guide for the eye.
NMR spectroscopy, a fast forming species [15L2]-X was ob-
served (Figure 5b). The corresponding signals were also ob-
served in previous experiments, such as the silylation of cin-
namyl alcohol 1b (Figure 5c) and allylic ether 12a. Species
[15L2]-X was present at a constant concentration throughout
.
e u v.
e
iM
3)2 ( )
2
1 2
.
i
S
q
(
H
O
D
OC
3
1
mo
a
Pd db
)3 (
14
the reaction. Its H-NMR spectrum displays the characteristic
2 5
l
%
(
2
)
a
-
d3
12
coupling pattern and shift values of (η3-allyl)palladium com-
plexes, such as (15-Br)2 (Figure 5a).16 Based on a possible
intermediacy of (η3-allyl)palladium species in our model reac-
tion (Scheme 3) and the presence of tetrafluoroborate in 3, the
new signals are expected to belong to a (η3-allyl)palladium-
BF4 complex. Therefore, we separately synthesized complex
[15L2]-BF4. Halide abstraction of (15-Br)2 using AgBF4 in
DMSO-d6/MeOH-d4 (Scheme 11) gave (η3-allyl)palladium
a
1
-
:
e
-
:
DM
d
6 M H d4 1 1
SO
O
on
l
orme
2
f
d
%
y
,
50 ° 15 h
C
no s a on
il l ti
y
Scheme 8. Attempted silylation using Pd2(dba)3 as catalyst.
1
The yield is determined by H NMR spectroscopy using an
internal standard. dba = dibenzylideneacetone.
1
Activation of the Hydroxyl Group. Replacement of 3 with 5
mol% Pd2(dba)3 (14) in our model reaction (Scheme 3)
changed the outcome substantially (Scheme 8). Only 2 %
linear ether 12a-d3 was formed and the rest of the allylic alco-
hol 1a remained unreacted. This sub-catalytic process
(Scheme 8) was accompanied with formation of Pd black.
Previous work by Tamaru has shown that Pd(0) can catalyze
allylic alcohol substitutions, albeit always in the presence of a
Lewis acid, such as BEt3.1a,15 We therefore reasoned that BF3,
whose presence in our model reaction (using 3 as catalyst,
complex [15L2]-BF4. The H NMR shift values and coupling
constants of [15L2]-BF4 are in good agreement with those of
the species observed during catalysis (compare Figures 5b-c
with 5d). Therefore we assign the newly observed peaks to
(η3-allyl)palladium complex [15L2]-BF4 or, given the reactivi-
-
ty and weak coordinating ability of BF4 , a very closely related
cationic species.
ACS Paragon Plus Environment

Page 5 of 14
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Figure 5 1H NMR spectra of the silylation of 1a (b.) and 1b (c.) after 1 hour and authentic samples of (η3-allyl)palladium complexes (15-
Br)2 (a.) and [15L2]-BF4 (d.) in DMSO-d6/MeOH-d4 1:1 solutions. ‡std cond. (standard conditions, reaction arrows in 5b-c): 1.2 equiv. 2, 5
mol% 3, DMSO-d6/MeOH-d4 1:1.
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
Fluorine-carbon coupling was not observed for the allylic
state, we wanted to determine whether [15L2]-BF4 was catalyt-
ically competent. We selected acetonitrile complex
[15(MeCN)2]-BF4 for this study because of its higher stability
fragment in complex [15L2]-BF4 by 13C NMR spectroscopy,
-
consistent with the fact that BF4 is a counterion in complex
[15L2]-BF4. The exact nature of this (η3-allyl)palladium com-
plex in DMSO-d6/MeOH-d4 solution has not been determined.
Since BF4- is known to be a weakly coordinating ligand, the
complex is assumed to be cationic. In the reaction mixture
both DMSO-d6 and MeOH-d4 are present and could coordinate
1
to palladium as neutral ligands. A comparison of the H NMR
spectra for the (η3-allyl)palladium complexes synthesized in
MeOH-d4, DMSO-d6 and a 1:1 MeOH-d4/DMSO-d6 mixture
reveals significant differences.
A BF
g
4
-
:
e
-
:
BF4
DM
SO
d
6 M H d4 1 1
O
Pd
Pd
-
r
r
m n.
,
B
t 10
i
L
L
2
- r
15
15
L2 BF
B
[
]
4
(
)
2
Scheme 11. Ligand exchange of complex [15L2]-BF4
In MeOH-d4 solution, all four allylic protons give rise to sharp
peaks. However, in both DMSO-d6 and 1:1 DMSO-d6/MeOH-
d4 solution the two terminal protons give rise to a single broad
peak at room temperature (Figure 6), suggesting a fast η3- η1-
η3 exchange is taking place.17 That this phenomenon is not
observed in MeOH-d4 suggests that the η3- η1- η3 isomerization
is induced by DMSO-d6 and, accordingly, this solvent mole-
cule is coordinated to palladium at least to some extent. Dur-
ing the course of the model silylation reaction (Scheme 3),
[Pd(MeCN)4](BF4)2 was found to convert quickly and quanti-
tatively to the new cationic (η3-allyl)palladium complex. To
examine if this complex corresponds to the catalyst resting
Figure 6. 1H NMR spectra of [15L2]-BF4 in different solvents.
in the solid state compared to the corresponding DMSO com-
plex [15(DMSO)2]-BF4. Performing our model reaction with 5
mol% [15(MeCN)2]-BF4 instead of [Pd(MeCN)4](BF4)2 pro-
duced 4a in 73% yield, showing that the complex is catalyti-
cally active (Scheme 12). However, in this reaction the pre-
synthesized complex [15(MeCN)2]-BF4 was observed to de-
compose slightly faster than the (η3-allyl)palladium species
formed in situ from [Pd(MeCN)4](BF4)2 under our initial con-
ditions (Scheme 3). In the crude reaction mixture after 15
hours with [15(MeCN)2]-BF4 as a catalyst, 24% of the linear
ACS Paragon Plus Environment

Journal of the American Chemical Society
ether intermediate 12a-d3 remained, while all of the (η3-
Page 6 of 14
being fully consumed within minutes (Scheme 14c). In both
reactions (η3-allyl)palladium complex [15L2]-X could be ob-
served to form, however only as a minor product (approxi-
mately 10-15%) and was formed together with a complex
mixture of saturated by-products. Decreasing the amount of
[Pd(MeCN)4](BF4)2 in this reaction did not change the out-
come significantly, although a small amount of 1a was still
present at the end. That Pd(0) reacts cleanly with 1a, in the
presence of a Lewis acid to form [15L2]-X, whereas the corre-
sponding reaction with [Pd(MeCN)4](BF4)2 (3) is sluggish,
suggests that reduction to Pd(0) is key to catalyst activation.
1
2
3
4
5
6
7
8
allyl)palladium catalyst was decomposed.
mo
5
M
l
%
]
)
2
e
N
C
-
BF
15
[
(
4
a a
l t
s
t
C
y
H
O
.
iM
e u v.
1 2
S
i
q
)
Ph
e
2
(
3 2 ( )
e
iM
S
3
a
4
BF4
Pd
a
1
-
d
6
DM
e
SO
-
7
3
%
e
M
e
N
C
N
M
C
:
M
H d4 1 1
O
,
-
BF
]
50 °C 15 h
e
M N
C
15
[
(
)
2
4
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
H
O
r
Scheme 12. Silylation of 1a using [15(MeCN)2]-BF4 as a
catalyst. The yield is determined by H NMR spectroscopy
using an internal standard.
t 1 h
,
1
a.
a
+
no reac on
ti
Pd2 db
)
(
)
3
a
1
-
d
:
e
-
4
DM
6 M H d
SO
O
14
e u v.
e u v.
i
q
2
i
1
q
A possible difference between the reactions catalyzed by
[15(MeCN)2]-BF4 and [Pd(MeCN)4](BF4)2 is that, in the latter
case, BF3 is formed early in the reaction, probably to a large
extent during catalyst activation. Addition of 10 mol% of
BF3·OEt2 to the reaction catalyzed by [15(MeCN)2]-BF4 sub-
stantially improved the longevity of the (η3-allyl)palladium
species and raised the NMR yield of 4a to 91% (Scheme 13).
H
O
⋅
e u v.
2
i
BF3 OE 2
t
q
.
b
+
)
3
a
Pd db
)
(
2
X
Pd
a
1
-
:
e
-
4
DM
d
6 M H d
SO
O
14
L
[
L
r
m n
i
t 1
0
,
e u v.
e u v.
i
q
2
i
1
q
-
L2
]
15
X
u
f ll
convers on o
i
t
-
15
L2
[
X
]
1
The H NMR spectrum of the crude reaction mixture showed
w
or w ou
ith
ith
2
BF Et
t
that 50% of the added (η3-allyl)palladium complex was still
intact after complete consumption of 1a. Since catalyst de-
composition likely occurs by aggregation of Pd(0) leading to
the formation of Pd black, a possible explanation is that BF3
stabilizes the Pd(0) intermediates (see discussion below in
section “Formation of Allylic Silane from the (η3-
Allyl)palladium Complexes”). Furthermore, BF3 could acti-
vate the alcohol and thereby accelerate the oxidative addition
which would lead to a decrease in the concentration of unsta-
ble Pd(0) intermediates.
H
O
e u v.
i
q
⋅
3 O
2
e
c.
Pd M
(
N
BF
)4(
C
+
)
4 2
)
X
Pd
a
1
-
DM
e
d
SO
6
4
3
-
H d
O
L
L
M
-
e u v.
e u v.
1 i
q
rox.
convers on o
1 2
i
q
r
m n
i
t 1
0
,
-
]
15
L2
[
X
a
-
10 15
%
pp
i
t
+
- ro u uc s
d d
b
t
y p
-
X
15
[
L
]
2
Scheme 14 Stoichiometric formation of complex [15L2]-X.
Oxidation of 1a was not observed during the course of our
model reaction (Scheme 3). Similarly, reduction of 3 to palla-
dium black was not observed upon heating at 50 ºC in DMSO-
d6/MeOH-d4 for 15 hours. This suggests 1a and MeOH-d4 do
not behave as reductants for palladium under the applied cata-
lytic conditions. However, the fast formation of Me3Si-F (7,
Figure 1) and BF3 (11) indicates that (Me3Si)2 (2) is involved
in the reduction of Pd(II) to Pd(0). Heating the Pd(II) salt 3
with (SiMe3)2 2 in the reaction solvent produced a dark red
solution in which Me3Si-F and BF3 were detected by 19F and
11B NMR spectroscopy (Scheme 15). After 2 hours palladium
black was observed to form. However, if alcohol 1a was added
after 1.5 hours this led to the instant formation of (η3-
allyl)palladium complex [15L2]-X and the solution returned to
a yellow color.
mo
10
e
-
BF
]
)
2
15
5
l
M
N
%
C
⋅
[
(
4
H
O
mo
BF Et
3 O
l
%
2
.
e u v.
e
2
1 2
i
iM
S
q
(
3)2 ( )
e
iM
S
3
a
4
a
1
e
- : - :
d
DM 6 M H d4 1 1
SO O
91
%
,
C
50 ° 15 h
Scheme 13. Silylation of 1a using [15(MeCN)2]-BF4 and
1
BF3·OEt2 a catalysts. The yield is determined by H NMR
spectroscopy using an internal standard.
Formation of the Active Catalyst. Having identified the
rapid room temperature formation of catalytically competent
complex [15L2]-X in our model reaction (Scheme 3), we
sought to investigate the conditions under which it is formed.
Alcohol 1a was found to be completely inert to the Pd(0)
source Pd2(dba)3 at room temperature (Scheme 14a). After one
e u v.
1
i
q
Pd0
H
O
e u v.
10
i
q
Ph
Ph
e
2
iM
3)2 ( )
S
(
1
a
1
e
M
i F
3S
hour no reaction had taken place according to H NMR spec-
e
N
C
Pd M
BF
)4(
(
)
4 2
X
Pd
7
.
,
r
50 °C 1 5 h
t
troscopy. However, the inclusion of two equivalents of
BF3·OEt2 resulted in full conversion of alcohol 1a to (η3-
allyl)palladium complex [15L2]-X (Scheme 14b). The reaction
was instantaneous and no other by-products were observed.
This led us to develop a new method for the synthesis of (η3-
allyl)palladium complexes directly from allylic alcohols (see
Supporting Information for further details).
3
L
L
-
DM
SO
d
6
4
a
rox.
40
BF3
%
t
pp
-
15
L2 X
[ ]
e
-
H d
M
O
:
convers on o
i
11
1 1
-
X
15
L
]
2
[
Scheme 15. Formation of [15L2]-X by in situ generation of
Pd(0).
Formation of Allylic Silane from (η3-Allyl)-palladium
Complexes. After establishing that (η3-allyl)palladium com-
Using [Pd(MeCN)4](BF4)2 as palladium source under the same
conditions with or without BF3·OEt2 resulted in alcohol 1a
ACS Paragon Plus Environment

Page 7 of 14
Journal of the American Chemical Society
as
f
t
plex [15L2]-X is formed in our model reaction (Scheme 3), we
Ph
re uc ve
d
ti
Ph
1
2
3
4
5
6
7
8
e m na on
li
ti
investigated its reactivity with (SiMe3)2 (2). A stoichiometric
e
i
2
iM
S
(
)
3 2 ( )
reaction18 between [15(MeCN)2]-BF4 and (SiMe3)2 (24 equiv.
to approximate the catalytic conditions) in DMSO-d6/MeOH-
a
4
BF4
Pd
Pd
L
L
e
SiM
3
L
e
M
i F
3S
1
d4 (1:1) at 50 ºC was monitored by H NMR spectroscopy
-
]
15
L2 BF
[
4
7
(Scheme 16a). After 10 minutes a product ratio of 4a:13 of 7:1
was observed, which decreased to 1.3:1 at full consumption of
[15(MeCN)2]-BF4 (20 min). Continued heating did not alter
the product ratio, suggesting that 13 is not formed by protode-
silylation of 4a.
Scheme 17. Allylic silane 4a and 7 were the only silicon con-
taining products observed to form in the reaction depicted in
Scheme 16a.
The stoichiometric reaction between complex [15(MeCN)2]-
BF4 and (SiMe3)2 at 25 °C afforded both 12a-d3 and 4a (3:1
ratio, see SI for more information). This indicates that the
formation of 12a-d3 via (η3-allyl)palladium intermediates is
possible under our catalytic conditions but that higher temper-
atures favor the formation of 4a.
9
ro uc
P
d
t
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
ra o
ti
a.
e u v.
SiM
(
e
2
24
i
)
Ph
.
q
e
iM
S
1 3
Ph
Ph
3)2 ( )
3
a
4
e
- : - :
d
SO O
6 M H d4 1 1
DM
BF4
Pd
m n.
i
1
,
50 °C 20
e
e
M
N
M CN
C
13
e
-
BF
15
M
(
N
C
[
]
)
2
4
ro uc
t
P
d
ra o
ti
.
b
e u v.
e
2
24
i
SiM
Ph
q
)
(
)
-
3 2 ( )
e
iM
S
Ph
Ph
7
1
3
-
:
e
:
DM
d
6 M H d4 1 1
SO
O
a
4
BF4
Pd
⋅
e u v.
BF Et 15
4
i
3 O
q
e
e
(
)
M
N
M CN
2
C
m n.
i
,
50 °C 40
13
e
N
C
-
15
[
M
(
BF
]
)
2
4
Scheme 16. Stoichiometric reactions between [15(MeCN)2]-
BF4 and 2.
The sequential synthesis of 4a via allylic ethers 12a-d3 and
12b-d3 (Scheme 6) indicates that the presence of Pd(0) pro-
motes the formation of 13, which occurs mainly at the end of
the stoichiometric reaction (Scheme 16a). Additionally, 13 is
formed in approximately 2% NMR yield in the catalytic model
reaction (Scheme 3). In the catalytic process (Scheme 3) the
concentration of Pd(0) is low for the greater part of the process
due to a fast alcohol activation. Increasing the catalyst loading
in the model silylation reaction (Scheme 3) also increased the
formation of 13 (see SI, page S6). This result indicates that
trans-β-methylstyrene (13) is formed at the end of the reaction
as the catalyst starts decomposing. The inclusion of BF3·OEt2
in the stoichiometric reaction (Scheme 16b) gave a higher
4a:13 ratio of 7:1, but required a longer reaction time (40 min)
to achieve complete conversion. A possible explanation for the
reduced formation of trans-β-methylstyrene (13) is that BF3
coordinates to Pd(0) as a so-called “Z-type ligand” (σ-acceptor
ligand).19 This is consistent with our observation that a sub-
stantial amount of homogenous Pd(0) can be present in a BF3-
containing DMSO-d6/MeOH-d4 solution at 50 °C without
palladium black formation being observed (Scheme 15).
Figure 7. Overview of the reaction components over time.
The conversion refers to the silylation of 1a in formation of
4a. Lines are given as a guide for the eye.
Relative Concentration of the Reaction Components as a
Function of Time. The key reaction components of our model
system (Scheme 3) were monitored by 1H NMR spectroscopy
for 12 hours (Figure 7). The (η3-allyl)palladium complex
[15L2]-X is formed quickly at the start of the reaction and
remains at a constant concentration (approximately 5 mol%)
until 1a was consumed completely. As shown previously,
allylic ether 12a-d3 forms at the start and it reaches its maxi-
mum concentration after 3.5 hours, coinciding with the alcohol
being consumed. Me3Si-F (7) is formed at the same rate as the
product for the first hour and the rate of its formation is there-
after reduced. When the reaction is finished, its maximum
theoretical conversion of 40% has almost been reached (using
all available fluorides from a 10 mol% loading of BF4-).
Formation of Me3Si-F (7) in the stoichiometic reaction shown
in Scheme 16a demonstrates that the BF4 counterion of
-
[15(MeCN)2]-BF4 takes part in the reaction. We propose that
BF4- helps in the activation of (SiMe3)2 (2) towards transmetal-
lation to give Me3Si-F (7) and a palladium complex with a
trimethylsilyl-ligand (Scheme 17). This likely mimics the
transmetallation step in the first cycle of the catalytic reaction
(Scheme 3). Additionally, that silyl-Pd species were not ob-
served is consistent with a fast reductive elimination in which
4a is formed.
Kinetic Studies: Reagent Order Determination. Initial rate
measurements were carried out to identify the turnover limit-
ing step. The initial rate dependence on the concentration of
hexamethyldisilane 2 (Figure 8a), [Pd(MeCN)4](BF4)2 (3)
(Figure 8b), alcohols 1a and 1b (Figure 8c-d) and methanol
(Figure 8e) was evaluated.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 14
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 8. Dependence of the logarithm of the initial rate on the logarithm of the concentration of (a) (SiMe3)2, (b) [Pd(MeCN)4](BF4)2, (c)
allylic alcohol 1a, (d) allylic alcohol 1b and (e) MeOH-d4 .
The reaction was observed as approximately first order in
hexamethyldisilane and [Pd(MeCN)4](BF4)2 but zero order in
isomeric allylic alcohols 1a-b. Varying the concentration of
methanol had a drastic effect on the rate. The reaction showed
third-order dependence in methanol in the range of [MeOH-d4]
= 15.5-18.6 M. Our kinetic studies, combined with the identi-
fication of the (η3-allyl)palladium complex [15L2]-X as a key
intermediate and the catalyst resting state, are consistent with
the following conclusions about the mechanism of the model
silylation process:
turn-over limiting step of the reaction. A possible explanation
for the third order dependence on methanol is organization of
three methanol molecules by hydrogen bonding in the rate
determining step. Sigman and co-workers20 reported a similar
third-order dependence of water in the Wacker oxidation,
proposing the pre-organization of three water molecules in the
rate-determining step. We performed a solvent isotope effect
experiment to further investigate the role of MeOH in our
reaction. However, we observed no appreciable difference in
rate between the reactions performed in CD3OH and those
performed in CD3OD (see page S12-13).
•
•
C-OH cleavage is not turnover limiting.
Mechanism of the Borylation Reaction. To investigate the
similarities between the borylation and the silylation reactions
(Scheme 2) the analogous borylation of 4a using
bis(pinacolato)diboron (5) was chosen as a model reaction
(Scheme 18).
Complex [15L2]-X, which is rapidly formed from
[Pd(MeCN)4](BF4)2, participates in the turnover lim-
iting step.
•
The turnover limiting step involves (SiMe3)2 (2).
The silylation (and borylation) of allylic alcohols proceed
rapidly when methanol is used as solvent or co-solvent.2a,b,8
The use of both DMSO-d6 and MeOH-d4 as solvents in the
studied catalytic silylation was crucial to obtain a fast and
clean reaction.2a In pure DMSO-d6, the conversion of 1 to 4 is
very slow and if the reaction is performed in pure methanol
by-product formation becomes a problem.2a The rate depend-
ence observed for methanol concentration is interesting from a
practical synthetic perspective, but somewhat difficult to ra-
tionalize. Unfortunately, the poor solubility of (SiMe3)2 in
DMSO-d6 prevented the study of the rate dependence in a
wider MeOH-d4 concentration range. It seems that MeOH-d4
is needed both to dissolve (SiMe3)2 as well as to facilitate the
.
e u v.
B
2p
n
5
H
1 2
i
i
O
q
2 ( )
mo
e
N
C
5
l
Pd M
(
BF
]
(
3
%
[
)
4)2 ( )
4
n
B i
p
a
1
a
6
-
d
:
e
O
-
:
DM
6 M H d4 1 1
SO
91
%
r
.
,
t 1 5 h
Scheme 18. Model reaction for studying the borylation. The
yield is determined by 1H NMR spectroscopy using an internal
standard.
ACS Paragon Plus Environment

Page 9 of 14
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Figure 11. Overview of the reaction components of the
borylation over time. The conversion refers to the borylation
of 1a in formation of 6a.
Figure 9. 19F NMR spectrum of the crude reaction mixture after
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
complete conversion of 1a to 6a (Scheme 18).
Unlike for the silylation reaction, which is accompanied with
formation of Me3Si-F (7), pinB-F (17)23 is not observed ac-
companying the formation of BF3 in the borylation. It is possi-
ble that pinB-F is initially formed and thereafter hydrolyzed to
B(OR)3 (R = H or CD3), which was shown to form. Monitor-
Following the course of the reaction by 19F NMR spectroscopy
revealed a more gradual formation of BF3 compared to the
silylation process. Additionally, a peak corresponding to
BF3OH- was identified21 by comparison with previously re-
ported shift values (Figure 9). The crude reaction mixture was
also studied using 11B NMR spectroscopy. The product (6a)
and the excess B2pin2 gave rise to two partially overlapping
peaks at 32 and 30 ppm (Figure 10). The only by-product
observed containing a Bpin adduct is pinB-OH (16)22. Analo-
gously to the silylation reaction, B(OR)3 (R = H or CD3) is
also formed, but in larger amounts; B(OR)3 is formed mainly
via hydrolysis of pinB-OH and not only from the catalyst
counterion BF4-.
1
ing the model borylation reaction (Scheme 18) by H NMR
spectroscopy (Figure 11) revealed the presence of the same
(η3-allyl)palladium intermediate ([15L2]-X) observed in the
corresponding silylation. It took 30 min for [15L2]-X to reach
its maximum concentration (5 mol%) which was maintained
until the allylic alcohol (1a) was fully consumed. Allylic
ethers 12a-d3 and 12b-d3 were not observed in this reaction.
mo
5
l
%
e
M
-
15
N
]
)
2
BF
C
[
a a s
l t t
C
y
(
4
⋅
BF OEt
%
3
2
mo
10
l
Ph
H
O
.
B
e u v.
i
q
1 2
2p
n
i
2 ( )
5
BF4
Pd
n
B i
p
a
6
e
e
M N
C
M
N
C
a
1
-
DM
SO
O
d
6
:
e
M
-
e
N
C
)
-
BF
]
H d4 1 1
15 M
[
98
%
(
2
4
r
m n
i
t, 15
Scheme 19. Borylation of 1a using [15(MeCN)2]-BF4 and
1
BF3⋅OEt2 as catalysts. The yield is determined by H NMR
spectroscopy using an internal standard.
Using (η3-allyl)palladium complex [15(MeCN)2]-BF4 as a
catalyst in the borylation with the addition of BF3⋅OEt2 result-
ed in a 98% NMR yield of 6a (Scheme 19), demonstrating its
catalytic competence in analogy with the silylation reaction
(c.f. Scheme 13). Additionally, the stoichiometric reaction
between (η3-allyl)palladium complex [15(MeCN)2]-BF4 and
B2pin2 gave clean conversion to 6a (Scheme 20). These results
strongly suggest that [15L2]-X is the catalyst resting state, in
analogy with the silylation reaction discussed above.
Figure 10. 11B NMR spectrum of the crude reaction mixture after
complete conversion of 1a to 6a (Scheme 19).
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 14
1
2
3
4
5
6
e u v
i B i
2p
n
5
24
q
2 ( )
n
B i
p
BF4
Pd
a
6
e
- : - :
d
DM 6 M H d4 1 1
SO O
e
e
N
C
M
N
M
C
°
.
,
0 C 1 5 h
e
N
C
-
15
M
BF
[
]
2
(
)
4
a
convers on o
i
t
6
u
f ll
7
8
9
Scheme 20. Stoichiometric reactions between [15(MeCN)2]-
BF4 and 5.
In summary, in both the silylation and the borylation the
strong Lewis acid BF3 is formed and likely promotes C-OH
cleavage. In both reactions the (η3-allyl)palladium complex
[15L2]-X is formed, which acts as an active catalyst. Both the
silylation and the borylation involve a transmetallation-
reductive elimination sequence of the (η3-allyl)palladium
intermediate to give allylic silane or allylic boronate products.
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
1
Figure 12. Borylation of cis-1c was followed by H NMR spec-
troscopy under standard borylation conditions (see Scheme 21)
with the exception that 10 mol% of 3 was used. Lines are given as
a guide for the eye.
The Stereoselectivity of the Reaction. The high stereoselec-
tivity of the reactions is an important synthetic advantage.
Moreover, the stereochemical outcome of the reaction can
reveal important mechanistic details. Both the formation of the
(η3-allyl)palladium intermediate and the nature of the reduc-
tive elimination in these reactions will influence the configura-
tion of the final products.
Using conditions we previously reported for the stereoselec-
tive silylation of allylic alcohols, cis-1c afforded 4b in 52%
isolated yield (3:1 trans:cis mixture).2a The analogous boryla-
tion proceeded more cleanly and with higher stereoselectivity,
giving 6b in 70% isolated yield and a trans:cis ratio of 5:1
(Scheme 21). The silylation of cis-1c is difficult to monitor by
1H NMR spectroscopy. Therefore, considering the close simi-
larity in the reaction mechanisms, the borylation of cis-1c was
chosen as a model reaction to investigate the stereoselectivity
(Scheme 21). The reaction was initially monitored by 1H NMR
spectroscopy. The loading of [Pd(MeCN)4](BF4)2 was in-
creased to 10 mol% to make observation of the catalyst more
accurate. Both complexes cis-[18L2]-X and trans-[18L2]-X
formed during the reaction (Figure 12),13 reaching their maxi-
mum concentration after 20 min (8 mol% total) and decreasing
thereafter. The trans complex was shown to be the major
stereoisomer formed, likely via oxidative addition with inver-
sion from cis-1c, in accordance with previous studies.24 Also,
allyl boronate product trans-6b was observed to form at a
much higher rate than cis-6b throughout the reaction (Figure
13).
1
Figure 13. Borylation of 1c was followed by H NMR spec-
troscopy under standard borylation conditions (see Scheme
21) with the exception that 10 mol% of 3 was used. The con-
version refers to the borylation of 1c in formation of 6b. Lines
are given as a guide for the eye.
These findings make it reasonable to assume that boronate
trans-6b forms from complex trans-[18L2]-X and cis-6b from
cis-[18L2]-X via reductive elimination with retention of con-
figuration. As the trans-6b:cis-6b ratio is consistently higher
than the trans-[18L2]-X:cis-[18L2]-X, (η3-allyl)palladium
complex trans-[18L2]-X likely undergoes much faster turnover
than cis-[18L2]-X does.
We also studied (by 1H NMR spectroscopy) the stoichiometric
reaction between complex trans-[18(MeCN)2]-BF4 and B2pin2
at room temperature. A 1:1 mixture of trans-6b and cis-6b
was formed in a single step (Scheme 22) with no observed
intermediates, which is indicative of a fast reductive elimina-
tion, in keeping with our previous findings. However, the
scrambling of the stereochemistry is in contradiction with the
results from the catalytic reaction (Scheme 21 and Figure 13).
e
e
COOM
e
M
COO
COOM
.
e u v.
n
i
2p
2 ( ) (
e
iM
/ S
3)2 ( )
5
2
1 2
i
B
q
mo
l
e
3
5
Pd M CN BF
%
[
]
(
(
)
)
4 2 ( )
4
+
-
:
e
-
:
DM
d
6 M H d4 1 1
SO
O
X
X
H
O
rans
c s
i
t
c
c s-
1
i
=
e
so a e
i l t d i ld t
%
e
rans:c s
:
4b
,
X
iM 52
i
3 1
S
,
,
y
3
=
n
so a e
l t
e
i l
y
rans:c s :
d
t i 5 1
,
6b X B i
7
0
%
i
d
,
,
p
Scheme 21. Silylation: The reaction was stirred at 60 °C for
15 h. Borylation: The reaction was stirred at 20 °C for 1.5 h.
ACS Paragon Plus Environment

Page 11 of 14
Journal of the American Chemical Society
Pd(II) complex [15L2]-X could act as a Lewis acid.26 Howev-
er, the fact that the hydroxyl group is transferred to boron
rather than silicon at the start of the reaction supports the
assumption that BF3 or its derivatives are coordinated to the
alcohol in the C-O bond breaking step.
e
M
COO
e
e
M
M
COO
COO
1
2
3
4
5
6
e u v.
B i
2p
n
5
24
i
q
2 ( )
+
n
n
B i
p
B i
p
:
e
:
H 1 1
O
m n.
i
DM
SO
M
BF4
Pd
r
t, 5
6b
rans:c s :
i 1 1
e
N
C
e
M
M
N
C
t
rans-
e
-
M N BF
C
]
( )
2
18
t
[
4
7
u
f ll
convers on o
i
t
6b
8
9
Scheme 22. Stoichiometric reactions between trans-
[18(MeCN)2]-BF4 and 5.
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
A possible explanation is that Pd(0) formed by the reductive
elimination epimerizes complex trans-[18(MeCN)2]-BF4 to
give a mixture of trans- and cis-[18(MeCN)2]-BF4. Bäckvall
and co-workers have proposed that cis-trans isomerization in
related (η3-allyl)palladium complexes can result from nucleo-
philic attack by free Pd(0) on the (η3-allyl)palladium com-
plex.24b We therefore reasoned that the lack of stereoselectivity
in our stoichiometric reaction (Scheme 22) similarly may be
explained by the high relative concentration of Pd(0), which
forms via reductive elimination.
To further investigate the effects of Pd(0) on the stereochemi-
cal outcome of the reaction, 5 mol% Pd2(dba)3 was added to
the borylation of cis-1c under our standard catalytic conditions
(Figure 14). The reaction produced 6b in substantially lower
NMR yield and lower (3:1) trans/cis selectivity, showing that
Pd(0) has an adverse effect on the stereochemistry of the reac-
tion. This suggests that the cis-trans isomerization should be
suppressed by lowering the concentration of free Pd(0). In-
deed, the addition of 30 mol% of BF3·OEt2 to the borylation
reaction increased the trans-6b:cis-6b ratio as well as the
NMR yield (Figure 14). We reasoned that addition of
BF3·OEt2 accelerated the oxidative addition of Pd(0) to the
allylic alcohol substrate. Therefore, the concentration of free
Pd(0) decreased, leading to less isomerization and, therefore,
higher stereoselectivity. In addition, a reduction in catalyst
loading (2.5 mol%) further improved the stereoselectivity to
trans:cis 11:1 (Figure 14). This can be explained by the fact
that two palladium atoms are involved in the Pd(0)-catalyzed
isomerization.24b Thus, a decrease in the total concentration of
palladium discourages isomerization and favors transmetalla-
tion instead.
Figure 14. Investigation into the effects of additives on the stere-
oselectivity of the borylation. Standard borylation conditions: 5
(1.2 equiv.), 3 (5 mol%), DMSO-d6/MeOH-d4 1:1, rt, 1.5 h. The
yields are determined by 1H NMR spectroscopy using an internal
standard.
a
Our study of the stereochemistry of the borylation indicates
that the formation of (η3-allyl)palladium species by oxidative
addition is stereoselective and occurs with inversion. As both
the borylation and the silylation proceeds via (η3-
allyl)palladium species, it is reasonable to assume that the
oxidative addition during the silylation occurs with the same
stereochemistry. Thereafter, the formed complex probably
undergoes Pd(0)-catalyzed isomerization (Figures 12-14,
Schemes 20-21). This is in accordance with the lower stere-
oselectivity observed for the silylation compared to the boryla-
tion (Scheme 20); the former has a longer reaction time and
therefore a greater opportunity for the (η3-allyl)palladium
complex to isomerize.
Proposed catalytic cycle. Based on the above mechanistic
studies we propose a catalytic cycle in Figure 15. The major
initiation pathway is proposed to proceed via complex 3 un-
dergoing two consecutive transmetallations with (SiMe3)2 (2)
in which Me3Si-F and BF3 are formed (Figure 15). Reductive
elimination will then regenerate one equivalent of (SiMe3)2
with respect to the catalyst and a Pd(0) complex with loosely
coordinated ligands. Thereafter [15L2]-X is proposed to form
by oxidative addition of the alcohol to Pd(0). The activation of
the allylic alcohol by BF3 to form free allyl cations prior to the
reaction with Pd(0) cannot be excluded.25 However, due to the
fast formation of [15L2]-X, a direct oxidative addition to the
activated alcohol is considered a more likely reaction pathway.
Our stoichiometric studies show that formation of
(η3-allyl)palladium complex [15L2]-X only occurs if the hy-
droxyl group is activated, for example by BF3 (Scheme 15).
Under catalytic conditions BF3 (11) is formed from the coun-
terion of [Pd(MeCN)4](BF4)2 (3). Alternatively, the cationic
It has been shown that allylic ether 12a-d3 can form both from
the reaction of (η3-allyl)palladium complex [15(MeCN)2]-BF4
with methanol (SI) and without the assistance of palladium
complexes from alcohol 1a by BF3 catalyzed nucleophilic
substitution (Scheme 10). In the Pd-catalyzed reaction mainly
linear ether 12a-d3 and only trace amounts of branched ether
12b-d3 are formed (Schemes 5 and 6). By contrast, in the BF3-
catalyzed process (in the absence of Pd) formation of the
branched ether 12b-d3 is prefered over the linear analog 12a-
d3. This indicates that under catalytic conditions, the predomi-
nant formation of 12a-d3 arises from reaction of [15L2]-X with
MeOH-d4. That 12a-d3 is both formed and then consumed
during the reaction indicates that this process is reversible.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 14
evers e e er orma on
R
ibl th
f
ti
ca a
t l
g
s
t
y
1
res n s a e
ti
t t
2
3
4
5
6
7
8
9
R
CD3OD
F3B
H
O
R
D
OC
3
2 L
12
F3B
R
Pd II L
( )
co o
2
Al h l
H
O
+
Pd 0
( )
+
ac va on
ti ti
+
H
2O
BF3
15L
[
]
2
H
O
x
dditi
a ve
on
id ti
O
s ar ere
t h
t
a
e
iM
S
(
)
3 2
R
1
urnover
t
m n
li iti
g
a a s ac va on
t l ti ti
t
C
y
rans-
me a a on
t ll ti
T
e
iM
)
3 2
+
2
S
e
iM
S
)
3 2
e
+
+
Pd 0 L
BF3
( )
PdL BF
+
M
i
F
2
3S
2
(
[
]
(
)
4 2
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
4
(
4
3
2
7
+
H
BF
2O
L
e uc ve
R d ti
e
M
i F
3S
e m na on
li
i
ti
R
e
iM
S
3
R
4
Pd II L
( )
e
3S
M
i
3 L
Figure 15. Proposed mechanism of the first turn-over of the palladium-catalyzed allylic silylation reaction. The exact nature of the borate
species may vary for subsequent turn-overs as BF4- is converted to BF4-xXx (X = OH, OCD3) throughout the catalysis.
That (η3-allyl)palladium complex [15L2]-X was observed in
both the silylation of allylic alcohol 1a and allylic ether 12a
further suggests that both species are converted by the same
mechanism into 4a. According to the competition experiment
in Figure 4, alcohol 1a is converted to 4a faster than the corre-
sponding ether 12a, possibly because 1a is more easily acti-
vated by the Lewis acid.
for alcohol 1a since increasing the concentration of 1a will not
influence the concentration of complex [15L2]-X.
CONCLUSIONS
Based on the above studies the most important mechanistic
aspects of the synthetically useful silylation and borylation of
allylic alcohols are now well understood:
Complex [15L2]-X has also been shown to react with (SiMe3)2
in formation of 4a (Scheme 16). The next step in the catalytic
cycle is therefore proposed to be transmetallation between the
formed (η3-allyl)palladium complex and (SiMe3)2. Based on
the observation that 4a and Me3Si-F are the only silicon con-
taining compounds formed in a stoichiometric reaction be-
tween complex [15(MeCN)2]-BF4 and (SiMe3)2 (Scheme 16a)
we propose that the counterion of the palladium complex
activates (SiMe3)2 towards transmetallation. The nature of the
counterion varies during the course of the reaction (e.g. BF4-,
•
•
It has been shown that BF3 is formed from the BF4- coun-
terion of catalyst 3 during the silylation and that a cata-
lytic amount of this Lewis acid is able to efficiently acti-
vate the hydroxyl group under our catalytic conditions.
The key intermediate of the silylation reaction is an (η3-
allyl)palladium complex formed by oxidative addition to
the activated allylic alcohol substrate. The (η3-
allyl)palladium complex is also the resting state of the
catalyst.
BF3OD3 , BF3OH-), which is the reason for the formation of
compounds Me3Si-F (7), Me3Si-OCD3 (8) and Me3Si-OH (9)
-
•
•
The turnover limiting step of the silylation is transmetal-
lation of (SiMe3)2 with the (η3-allyl)palladium complex.
This step proceeds with a high stereoselectivity, in par-
ticularly for the borylation process. Interestingly, cleav-
age of the C-OH group is not rate limiting.
in the catalytic reaction.
The stereochemistry of the process (Figures 12-13) indicates
that the favored pathway of the catalytic borylation involves
the introduction of Bpin via transmetallation to trans-[18L2]-
X, which forms selectively during oxidative addition. Subse-
quent reductive elimination produces trans-6b (Figure 13),
which is consistent with trans-4b product forming analogously
as the major isomer in the silylation (Scheme 21).
The silylation and borylation proceed according to very
similar mechanisms. In both reactions BF3 is formed and
is able to activate the alcohol substrate. Both reactions
proceed via (η3-allyl)palladium intermediates.
Our kinetic studies show a rate dependence on both the palla-
dium precursor 3 and (SiMe3)2 (2) in the silylation. Monitoring
of this reaction by H NMR clearly shows that quantitative
formation of complex [15L2]-X is fast. We therefore propose
that transmetallation of [15L2]-X with (SiMe3)2 (2) is the turn-
over limiting step. The fast formation of [15L2]-X, its presence
at a steady 5 mol% (the theoretical maximum), as well as its
catalytic competence, indicate that it is the catalyst resting
state. This is also in agreement with the the zero-order kinetics
Our insights into the mechanism have also allowed a substan-
tial improvement in the stereoselectivity of the palladium-
catalyzed borylation of allylic alcohols by increasing the rate
of oxidative addition and lowering the catalyst loading. Addi-
tionally we have shown that (η3-allyl)palladium complexes can
be prepared conveniently directly from allylic alcohols. The
above mechanistic study will certainly open new synthetic
routes for the application of allylic alcohols in selective organ-
ic synthesis and open new opportunities for easy access to
stereo- and regiodefined allylic silanes and allylic boronates,
1
ACS Paragon Plus Environment

Page 13 of 14
Journal of the American Chemical Society
Hall, D. G. Pure Appl. Chem. 2008, 80, 913-27; (d) Kennedy, J. W. J.;
which are very important reagents in advanced organic synthe-
sis.4,27
Hall, D. G. Angew. Chem. Int. Ed. 2003, 42, 4732-9; (e) Yamamoto, Y.;
Asao, N. Chem. Rev. 1993, 93, 2207-93; (f) Thibaudeau, S.; Gouverneur,
V. Org. Lett. 2003, 5, 4891-3; (g) Wilkinson, S. C.; Lozano, O.; Schuler,
M.; Pacheco, M. C.; Salmon, R.; Gouverneur, V. Angew. Chem. Int. Ed.
2009, 48, 7083-6; (h) Pilarski, L. T.; Szabo, K. J. Curr. Org. Chem. 2011,
15, 3389-414.
(5) (a) Sore, H. F.; Galloway, W. R. J. D.; Spring, D. R. Chem. Soc. Rev.
2012, 41, 1845-66; (b) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95,
2457-83; (c) Denmark, S. E.; Regens, C. S. Acc. Chem. Res. 2008, 41,
1486-99.
1
2
3
4
5
6
7
EXPERIMENTAL DETAILS
Model silylation reaction. The corresponding allylic alcohol
1a (0.15 mmol) and 2 (0.18 mmol) were added to a solution of
DMSO-d6 and MeOH-d4 (0.2 mL/0.2 mL) containing palladi-
um pre-catalyst 3 (0.0075 mmol, 5 mol %). Thereafter, the
reaction mixture was stirred at 50 °C for 15 hours.
8
9
(6) (a) Hatanaka, Y.; Hiyama, T. J. Org. Chem. 1988, 53, 918-20; (b)
Denmark, S. E.; Werner, N. S. J. Am. Chem. Soc. 2008, 130, 16382-93.
(7) (a) Larsson, J. M.; Zhao, T. S. N.; Szabó, K. J. Org. Lett. 2011, 13,
1888-91; (b) Ishiyama, T.; Ahiko, T.; Miyaura, N. Tetrahedron Lett. 1996,
37, 6889-92; (c) Ito, H.; Kawakami, C.; Sawamura, M. J. Am. Chem. Soc.
2005, 127, 16034-5; (d) Tsuji, Y.; Funato, M.; Ozawa, M.; Ogiyama, H.;
Kajita, S.; Kawamura, T. J. Org. Chem. 1996, 61, 5779-87; (e) Moser, R.;
Nishikata, T.; Lipshutz, B. H. Org. Lett. 2009, 12, 28-31; (f) Hartwig, J. F.
Acc. Chem. Res. 2011, 45, 864-73; (g) Mkhalid, I. A. I.; Barnard, J. H.;
Marder, T. B.; Murphy, J. M.; Hartwig, J. F. Chem. Rev. 2010, 110, 890-
931; (h) Hartwig, J. F. Chem. Soc. Rev. 2011, 40, 1992-2002; (i) Hayashi,
T.; Konishi, M.; Ito, H.; Kumada, M. J. Am. Chem. Soc. 1982, 104, 4962-
3; (j) Matsumoto, Y.; Ohno, A.; Hayashi, T. Organometallics 1993, 12,
4051-5.
Model borylation reaction. The corresponding allylic alcohol
1a (0.15 mmol) was added to a solution of 5 (0.18 mmol) and
palladium pre-catalyst 3 (0.0075 mmol, 5 mol %) in DMSO-d6
and MeOH-d4 (0.2 mL/0.2 mL). Thereafter, the reaction mix-
ture was stirred at room temperature for 1.5 hours.
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
ASSOCIATED CONTENT
Experimental procedures, spectroscopic and kinetic data. This
material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
(8) Dutheuil, G.; Selander, N.; Szabó, K. J.; Aggarwal, V. K. Synthesis
2008, 2008, 2293-7.
kalman@organ.su.se
(9) (a) Selander, N.; Sebelius, S.; Estay, C.; Szabó, K. J. Eur. J. Org.
Chem. 2006, 2006, 4085-7; (b) Selander, N.; Kipke, A.; Sebelius, S.;
Szabó, K. J. J. Am. Chem. Soc. 2007, 129, 13723-31; (c) Aydin, J.;
Kumar, K. S.; Sayah, M. J.; Wallner, O. A.; Szabó, K. J. J. Org. Chem.
2007, 72, 4689-97.
(10) Sebelius, S.; Olsson, V. J.; Wallner, O. A.; Szabó, K. J. J. Am. Chem.
Soc. 2006, 128, 8150-1.
(11) Grubb, W. T. J. Am. Chem. Soc. 1954, 76, 3408-14.
(12) (a) Bacon, J.; Gillespie, R. J.; Quail, J. W. Can. J. Chem. 1963, 41,
3063-9; (b) Bacon, J.; Gillespie, R. J.; Hartman, J. S.; Rao, U. R. K. Mol.
Phys. 1970, 18, 561-70.
ACKNOWLEDGMENT
The authors thank the financial support of the Swedish Research
Council (VR) and the Knut och Alice Wallenbergs Foundation.
We thank Dr. Lukasz T. Pilarski, Department of Chemistry-BMC,
Uppsala University, for help with the synthesis of complex 18-Cl
and Rauful Alam, Department of Organic Chemistry, Stockholm
University for help with the synthesis of 14.
(13) Identified by comparison with a genuine sample.
ABBREVIATIONS
pin, pinacolato; dba, dibenzylideneacetone.
(14) (a) Mesmer, R. E.; Palen, K. M.; Baes, C. F. Inorg. Chem. 1973, 12,
89-95; (b) Lennox, A. J. J.; Lloyd-Jones, G. C. J. Am. Chem. Soc. 2012,
134, 7431-41.
(15) Kimura, M.; Futamata, M.; Shibata, K.; Tamaru, Y. Chem. Commun.
2003, 234-5.
(16) Murahashi, S.; Imada, Y.; Taniguchi, Y.; Higashiura, S. J. Org.
Chem. 1993, 58, 1538-45.
(17) (a) Solin, N.; Szabó, K. J. Organometallics 2001, 20, 5464-71; (b)
Breutel, C.; Pregosin, P. S.; Salzmann, R.; Togni, A. J. Am. Chem. Soc.
1994, 116, 4067-8; (c) Faller, J. W.; Mattina, M. J. Inorg. Chem. 1972, 11,
1296-307.
(18) Hayashi, T.; Yamamoto, A.; Iwata, T.; Ito, Y. Journal of the
Chemical Society, Chemical Communications 1987, 398-9.
(19) (a) Braunschweig, H.; Dewhurst, R. D. Dalton Trans. 2011, 40, 549-
58; (b) Amgoune, A.; Bourissou, D. Chem. Commun. 2011, 47, 859-71.
(20) Anderson, B. J.; Keith, J. A.; Sigman, M. S. J. Am. Chem. Soc. 2010,
132, 11872-4.
(21) Mesmer, R. E.; Rutenberg, A. C. Inorg. Chem. 1973, 12, 699-702.
(22) Ahijado Salomon, M.; Braun, T.; Penner, A. Angew. Chem. Int. Ed.
2008, 47, 8867-71.
(23) Wada, R.; Shibuguchi, T.; Makino, S.; Oisaki, K.; Kanai, M.;
Shibasaki, M. J. Am. Chem. Soc. 2006, 128, 7687-91.
(24) (a) Kurosawa, H.; Ogoshi, S.; Kawasaki, Y.; Murai, S.; Miyoshi, M.;
Ikeda, I. J. Am. Chem. Soc. 1990, 112, 2813-4; (b) Granberg, K. L.;
Bäckvall, J. E. J. Am. Chem. Soc. 1992, 114, 6858-63.
REFERENCES
(1) (a) Sundararaju, B.; Achard, M.; Bruneau, C. Chem. Soc. Rev. 2012,
41, 4467-83; (b) Tamaru, Y. Eur. J. Org. Chem. 2005, 2647-56; (c)
Muzart, J. Tetrahedron 2005, 61, 4179-212; (d) Szabó, K. J. Synlett 2006,
2006, 811-24; (e) Bandini, M. Angew. Chem. Int. Ed. 2011, 50, 994-5.
(2) (a) Selander, N.; Paasch, J. R.; Szabó, K. J. J. Am. Chem. Soc. 2010,
133, 409-11; (b) Olsson, V. J.; Sebelius, S.; Selander, N.; Szabó, K. J. J.
Am. Chem. Soc. 2006, 128, 4588-9; (c) Defieber, C.; Ariger, M. A.;
Moriel, P.; Carreira, E. M. Angew. Chem. Int. Ed. 2007, 46, 3139-43; (d)
Mukherjee, P.; Widenhoefer, R. A. Org. Lett. 2010, 12, 1184-7; (e)
Kimura, M.; Futamata, M.; Mukai, R.; Tamaru, Y. J. Am. Chem. Soc.
2005, 127, 4592-3; (f) Ozawa, F.; Okamoto, H.; Kawagishi, S.;
Yamamoto, S.; Minami, T.; Yoshifuji, M. J. Am. Chem. Soc. 2002, 124,
10968-9.
(3) (a) Youngsaye, W.; Lowe, J. T.; Pohlki, F.; Ralifo, P.; Panek, J. S.
Angew. Chem. Int. Ed. 2007, 46, 9211-4; (b) Kong, K.; Moussa, Z.; Lee,
C.; Romo, D. J. Am. Chem. Soc. 2011, 133, 19844-56; (c) Donohoe, T. J.;
Winship, P. C. M.; Tatton, M. R.; Szeto, P. Angew. Chem. Int. Ed. 2011,
50, 7604-6; (d) Wohlfahrt, M.; Harms, K.; Koert, U. Angew. Chem. Int.
Ed. 2011, 50, 8404-6; (e) Trost, B. M.; Dong, G. Nature 2008, 456, 485-8;
(f) Wan, S.; Wu, F.; Rech, J. C.; Green, M. E.; Balachandran, R.; Horne,
W. S.; Day, B. W.; Floreancig, P. E. J. Am. Chem. Soc. 2011, 133, 16668-
79; (g) Langkopf, E.; Schinzer, D. Chem. Rev. 1995, 95, 1375-408; (h)
Hall, D. G.; Editor Boronic Acids: Preparation and Applications in
Organic Synthesis, Medicine and Materials
(25) Emer, E.; Sinisi, R.; Capdevila, M. G.; Petruzziello, D.; De
Vincentiis, F.; Cozzi, P. G. Eur. J. Org. Chem. 2011, 2011, 647-66.
(26) Strukul, G. Top. Catal. 2002, 19, 33-42.
(27) Lachance, H.; Hall, D. G. Org. React. (Hoboken, NJ, U. S.) 2008, 73,
1-573.
Second Edition ed.; Wiley-VCH Verlag GmbH & Co. KGaA, 2011.
(4) (a) Masse, C. E.; Panek, J. S. Chem. Rev. 1995, 95, 1293-316; (b)
Fleming, I.; Barbero, A.; Walter, D. Chem. Rev. 1997, 97, 2063-192; (c)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 14
1
2
Table of contents:
3
4
5
6
A Mechanistic Investigation of the Palladium-Catalyzed Synthesis of Allylic Silanes and Boro-
nates from Allylic Alcohols
7
8
9
Johanna M. Larsson and Kálmán J. Szabó*
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
res n s a e
ti t t
R
F3B
R
g
H
O
a co o
h l
H
O
l
F3B
H
O
ac va on
ti ti
R
PdII
e
iM
S
(
)
3 2
e
iM
urn-over
t
S
(
4 2
)
3 2
Pd0
+
+
PdL BF
)
e
3S
m
n
g
F
BF3
R
M
i
li iti
[
4](
s e
t
p
BF
3S
H
F
2O
ca a s ac va on
t l ti ti
t
y
R
e
M
i
e
iM
S
3
i
PdII
e
M
3S
14
ACS Paragon Plus Environment