WWW.NATURE.COM/NATURE | 1
SUPPLEMENTARY INFORMATION
doi:10.1038/nature14126
Supplementary Information for
Silylation of
C–H bonds in
aromat
ic heterocycles by an E
arth
-
abundant
metal c
atalyst
Anton A. Toutov, Wen-Bo Liu, Kerry N. Betz, Alexey Fedorov, Brian M. Stoltz* &
Robert H. Grubbs*
stoltz@caltech.edu, rhg@caltech.edu
Division of Chemistry and Chemical Engineering, California Institute of Technology
Pasadena, California 91125, United States
Table of Contents
General information.
........................................................................................................
2
Part I. Reaction optimizations, control experiments and trace metal analysis.
..........
4
1. Reaction optimization.
.........................................................................................................
4
2. Control experiments and trace metal analysis.
....................................................................
5
ICP
-MS trace metal analysis.
...............................................................................................
7
Part II. Competition experiments and evaluation of functional group compatibility.
8
1. Competitio
n experiments with thiophene, furan and pyrrole.
.............................................
8
2. Reactions with electron
-deficient heteroarenes.
..................................................................
9
3. Investigation into the radical nature of the KO
t
-Bu
–catalyzed C
–H silylation.
................
10
4. Evaluation of functional group compatibility.
...................................................................
12
Part III. Experimental and analytics.
...........................................................................
14
1. General procedure for KO
t
-Bu
–catalyzed silylation and characterization data.
...............
14
2. Multi
-gram scale syntheses of
2a
,
4h
and
4n
. ...................................................................
42
3. One-
pot Si
-directed
ipso
–substitution/Suzuki
–Miyaura cross-
coupling.
..........................
44
4. Synthesis of a heteroarylsilanol and application in Denmark
–Hiyama cross
-coupling.
... 45
5. Direct C7 lithiation
-borylation by a Si
-blocking group strategy.
......................................
46
6. Synthesis of a sila-
hete
rocycle by inter
-/intramolecular double C
–H silylation.
..............
48
7. C
–H silylation of terthiophene and EDOT.
.......................................................................
48
8. Late stage silylation of active pharmaceutical ingredients (APIs)
. ...................................
50
9. Oxygen-
directed C(sp
2
)–H silylation of anisole derivatives.
............................................
54
10. Direct C(sp
3
)–H silylation
reactions.
...............................................................................
56
Part IV.
GC-FID spectra of the robustness screen.
.....................................................
59
Part V.
1
H NMR and
13
C NMR spectra of new compounds.
......................................
72
SUPPLEMENTARY INFORMATION
2
| WWW.NATURE.COM/NATURE
RESEARCH
Supplementary Information for Toutov, Liu, Betz, Fedorov, Stoltz, and Grubbs
SI-2
General information.
Unless otherwise stated, reactions were performed in
oven-dried brand-new Fisherbrand
scintillation vials
in a nitrogen filled glove
box or in flamed-dried Schlenk flasks under
argon connected on
a Schlenk line using dry, degassed solvents and brand-new stirring
bars. Solvents were dried by passage through an activated alumina column under argon.
1
Reaction progress was monitored by thin-layer chromatography (TLC), UHPLC
-LCMS
or
GC-FID analyses.
TLC was performed using E. Merck silica gel 60 F254 precoated
glass plates (0.25 mm) and visualized by UV fluorescence quenching, phosphomolybdic
acid, or KMnO
4
stain
ing. Silicycle SiliaFlash P60 Academic Silica gel (particle size 40–
63 nm) was used for flash chromatography.
1
H NMR spectra were recorded on Varian
Inova 500 MHz and 600 MHz spectrometers in CDCl
3
or C
6
D
6
and are reported relative
to residual solvent peak at
δ
7.26 ppm or
δ
7.16 ppm respectively.
13
C NMR spectra were
recorded on a Varian Inova 500 MHz spectrometer (125 MHz) in CDCl
3
or C
6
D
6
and are
reported relative to residual solvent peak at
δ
77.16
ppm
or
δ
128.06 ppm respectively.
Data for
1
H NMR are reported as follows: chemical shift (
δ
ppm) (multiplicity, coupling
constant (Hz), integration). Multiplicities are reported as follows: s = singlet, d =
doublet, t = triplet, q = quartet, p = pentet, sept = septet, m = multiplet, br s = broad
singlet, br d = broad doublet, app = apparent. Data for
13
C NMR are reported in terms of
chemical shifts (
δ
ppm). IR spectra were obtained on
a Perkin Elmer Spectrum BXII
spectrometer using thin films deposited on NaCl plates and reported in frequency of
absorption (cm
–1
). UHPLC
-LCMS analyses were obtained on an Agilent 1290 ultra high
performance liquid chromatography/mass spectrometry equipped with an Agilent
EclipsePlus C18 RRHD 1.8
μ
M column. GC-FID analyses were obtained on an Agilent
6890N gas chromatograph equipped with a HP-5 (5%-phenyl)-methylpolysiloxane
capillary column (Agilent).
GC-MS analyses were obtained on an Agilent 6850 gas
chromatograph equipped with a HP-5 (5%-phenyl)-methylpolysiloxane capillary column
(Agilent).
High
resolution mass spectra (HRMS) were obtained on an Agilent 6200
Series TOF with an Agilent G1978A Multimode source in electrospray ionization (ESI+),
atmospheric pressure chemical ionization (APCI+), or mixed ionization mode (MM:
ESI-
(1) Pangborn,
A. M.
; Giardello, M. A.
; Grubbs,
R. H.;
Rosen
, R. K.;
Timmers,
F. J.
Organometallics
,
1996
,
15
, 1518.
WWW.NATURE.COM/NATURE | 3
SUPPLEMENTARY INFORMATION
RESEARCH
Supplementary Information for Toutov, Liu, Betz, Fedorov, Stoltz, and Grubbs
SI-3
APCI+), or were acquired from the California Institute of Technology Mass Spectrometry
Facility.
ICP-MS analysis was conducted at the California Institute of Technology Mass
Spectrometry Facility.
Al
2
O
3
was purchased from Aldrich and activated by storing in a 200 °C oven for one
week and then heating in a round bottom flask
at 200 °C under vacuum (60 millitorr).
Silanes were purchased from Aldrich and distilled before use. KO
t
-Bu was purchased
from Aldrich (sublimed grade, 99.99% trace metals basis
) and used directly.
Heteroaromatic substrates were purchased from Aldrich, TCI,
or Acros, or synthesized
according to literature procedures
.
2
(2) (a)
Kong, A.; Han, X.; Lu, X.
Org. Lett
.
2006
,
8
, 1339. (b)
Islam, S.; Larrosa, I.
Chem.
– Eur. J
.
2013
,
19
, 15093.
(c)
Huestis, M. P.; Fagnou, K.
Org. Lett
.
2009
,
11, 1357.
(d)
Mahadevan, I.; Rasmussen, M.
Tetrahedron
,
1993
,
49
,
7337.
Supplementary Information for Toutov, Liu, Betz, Fedorov, Stoltz, and Grubbs
SI
-
4
Part I
.
R
eaction optimizations
,
control experiments
and
trace metal analysis
.
1
.
Reaction optimization.
P
rocedure
for r
eaction
condition
optimization: In a nitrogen
-
filled glovebox,
base
and
indole
1
(0.2
mmol, 1 equiv) were added to a 2 dram scintillation vial equipped with a
magnetic stirring bar. Next, Et
3
SiH (
97 μL, 0.6
mmol, 3 equiv
, filtered through a short
pad of activated alumina
before use
) and solvent (
0.2 mL,
unless
the
reaction was run
neat) were added. The vial was sealed and
the mixture was
stirred at the indicated
temperature for the indicated time. The
n the
v
ial was removed from the gloveb
ox,
diluted with diethyl ether (1
mL) and concentrated under reduced pressure. The
regioselectivity (C2 silylation product to C3 silylation product: C2:C3)
and yield were
determined by
1
H NMR or GC analysis of the crude mixture using an internal standard.
Table 1. Con
dition optimization
of direct C
–
H silylation of indoles.
entry
a
R
base
solvent
x
t
(h)
C2:C3
b
C2 (%)
b
1
Me
LiO
t
-
Bu
THF
100
16
–
0
2
Me
NaO
t
-
Bu
THF
100
16
–
0
3
Me
NaOEt
THF
100
16
–
0
4
Me
NaOAc
THF
100
16
–
0
5
Me
KOMe
THF
100
16
–
<5
6
Me
KOEt
THF
100
16
–
14
7
Me
KO
t
-
Bu
THF
100
16
>20:1
67
8
Me
KHMDS
THF
100
16
>20:1
44
9
Me
KOAc
THF
100
16
–
0
10
Me
KH
THF
100
72
–
0
11
Me
KOH
THF
100
16
–
0
12
Me
Cs
2
CO
3
THF
100
16
–
0
13
Me
DABCO
THF
100
16
–
0
N
R
base
(x
mol%)
1a:
R
=
Me
1b:
R
=
Bn
N
R
2
N
R
SiEt
3
SiEt
3
C2-silylation
C3-silylation
Et
3
SiH
(3
equiv)
solvent,
25
°C
Supplementary Information for Toutov, Liu, Betz, Fedorov, Stoltz, and Grubbs
SI
-
5
14
Me
TBAF
THF
100
16
–
0
15
Me
CsF
THF
100
16
–
0
16
Me
KF
THF
100
16
–
0
17
c
Me
KO
t
-
Bu
THF
20
60
4:1
98
18
c
Me
KO
t
-
Bu
MeO
t
-
Bu
20
60
>20:1
89
19
c
Me
KO
t
-
Bu
DME
20
60
3.4:1
95
20
c
Me
KO
t
-
Bu
neat
20
48
>20:1
88
21
d
Me
KHMDS
THF
20
72
17:1
75
22
c,e
Bn
KO
t
-
Bu
THF
20
61
>20:1
90
23
c,e
, f
Bn
KO
t
-
Bu
THF
20
96
>20:1
22
24
c,e
Bn
KOTMS
THF
20
72
>20:1
79
a
R
eactions performed with 0.2 mmol of
1
and 0.6 mmol of Et
3
SiH in 0.2
mL of solvent.
b
Determined by
GC analysis of the crude reaction mixture
using an internal standard
.
c
At 45 °C.
d
At 3
5 °C.
e
T
he ratio
of
C2:C3 and yield were d
etermined by
1
H NMR
analysis of the crude reaction mixture.
f
With 50 mol% of
18
-
crown
-
6.
The results from Table 1 reveal that good catalysts for the C
–
H silylation reaction are
categorized by the combination of a bulky basic anion
and a potassium cation: KO
t
-
Bu
proved to be ideal catalyst and operated under neat conditions or in THF
and MeO
t
-
Bu
(Entry 18, 20
and
22), but KHMDS (Entry 21) and KOTMS (Entry 24) were also
effective. The complete lack of reactivity with LiO
t
-
Bu and NaO
t
-
Bu (Entries 1 and 2) as
well as the precipitous drop in reactivity when 18
-
crown
-
6 is added to KO
t
-
Bu (Entry 23
)
lend support to the crucial, albeit unknown, role of the potassium cation. Conversion
roughly correlates with basicity in stoichiometric react
ions (i.e., O
t
-
Bu > OEt > OMe;
Entries 5
–
7). No product was observed
in the absence of catalyst, or
when KH, KOH,
KOAc and Cs
2
CO
3
were employed
(Entries 9
–
12). The organic base DABCO and
common fluoride
-
based
activators for silicon
–
TBAF, CsF, and KF
–
w
ere also
investigated and failed to convert the starting material (Entries 13
–
16).
Headspace GC
-
TCD analysis of successful silylation reactions
indicated the formation of H
2
.
2
.
Control experiments and trace metal analysis.
Careful experiments were
conducted
in order to rule out
catalysis
by adventitious
transition metal impurities in the
reaction mixture [see below (1)
–
(3)].
Supplementary Information for Toutov, Liu, Betz, Fedorov, Stoltz, and Grubbs
SI
-
6
(1) Control reactions with commercially available
KOt
-
Bu, re
-
sublimed KOt
-
Bu, and
freshly
-
prepared
KOt
-
Bu.
T
hree reactions
were performed in parallel (THF, 45 °C, 1
-
methylindole, 20 mol% KO
t
-
Bu, 0.2 mmol scale): a) KO
t
-
Bu (Aldrich,
sublimed grade,
99.99%
, trace metal basis) was used as received; b) KO
t
-
Bu (Aldrich,
sublimed grade,
99.99% trac
e metal basis) was used after re
-
su
blimation by heating the material under
vacuum; and c) KO
t
-
Bu, freshly prepared by reaction of potassium metal with
anhydrous
t
-
BuOH followed by evaporation
of the
t
-
BuOH
and
sublimation
of the solid
, was used.
No appreciable differences in conversion and
selectiv
ity in these
reactions were
observed.
(2) Control reaction with KOt
-
Bu of different grade purchased from different vendors.
Four
reactions were performed in parallel
(THF, 45 °C, 1
-
benzylindole, 20 mol% KO
t
-
Bu, 0.2 mmol scale):
a)
KO
t
-
Bu (
Aldrich,
sublimed grade,
99.99% trace metal basis
)
, b)
KO
t
-
Bu (Strem, 98%), c) KO
t
-
Bu (TCI, >97%), and d) KO
t
-
Bu (Alfa
-
Aesar
, 97%). Th
e
reactions were monitored by UHPLC
-
LCMS
and no
appreciable differences
in
conversion and selectivity
in these four
reactions was
observed (Figure 1
)
.
Figure 1
.
The results of
with
KO
t
-
Bu purchased from different v
endors
(3)
ICP
-
MS trace metal analysis of all the reaction components.
To
provide
further
support
against involvement of adventitious trace metal species in the C
–
H activation
0"
10"
20"
30"
40"
50"
60"
70"
80"
90"
100"
0"
10"
20"
30"
40"
50"
60"
70"
80"
90"
100"
Aldrich"99.99%"
Strem"98%"
TCI">97%"
Alfa"97%"
N
Bn
KO
t
-Bu
(20
mol%)
1b
N
Bn
2b
N
Bn
SiEt
3
SiEt
3
C2-silylation
C3-silylation
Et
3
SiH
(3
equiv)
THF,
45
°C
+
time
(h)
conversion
(%)
Supplementary Information for Toutov, Liu, Betz, Fedorov, Stoltz, and Grubbs
SI
-
7
catalysis, inductively coupled plasma mass spectrometry was performed on samples of
KO
t
-
Bu from different vendors, 1
-
benzylindole starting material, THF, Et
3
SiH and a
standard r
eaction mixture that was run under optimal conditions in the glove box
(
“Rxn
M
ixture
”
in Table 2)
. The results from quantitative analysis revealed that most metal
contaminants were present below the instrument’s lowest limit of detection (i
.
e.
,
in ppt
rang
e or lower).
Microgram per liter (ppb)
quantities of
metal contaminants are given in
Table 2.
Table 2.
ICP
-
MS trace metal analysis
.
500 mg samples each o
f KO
t
-
Bu from four different vendors (Strem, Aldrich, TCI, Alfa
-
Aesar), 1
-
benzylindole, Et
3
SiH
, THF, a
nd
a standard reaction mixture (0.5 mmol scale
mixture, prepared following the general procedure with 103.5 mg of 1
-
Bn
-
indole, 11.2
mg of KO
t
-
Bu
from Aldrich
, 173.5 mg of Et
3
SiH i
n 0.5 mL of THF and stirred in
the
g
lovebox for 72 h.)
were analyzed.
Each s
ample was added to a 50 mL DigiTUBE
digestion tube (SCP Science) followed by addition of 3.0 mL of Plasma Pure nitric acid
(SCP Science) and heating to 75
o
C for 36 hours. After digestion, each sample was
diluted using Milli Q water to 50 mL and sample analysis
was
performed on an Agilent
7900 ICP
-
MS spectrometer. LOD indicates that the analyte concentration is below the
instrument’s
Lowest Limit of Detection
. Values in ppb
(
microgram per liter
)
.
ICPMS Trace Metal Analysis
–
Agilent 7900 (quantities in ppb)
Element
KO
t
-
Bu
Strem
(98%)
KO
t
-
Bu
TCI
(>97%)
KO
t
-
Bu
Alpha
(97%)
KO
t
-
Bu
Aldrich
(99.99%)
THF
HSiEt
3
1
-
Bn
-
indole
Rxn
Mixture
Ti
0.360
0.051
0.138
0.464
LOD
2.073
9.408
31.082
Mn
1.343
1.168
1.338
1.525
LOD
0.177
88.191
LOD
Fe
12.285
10.171
13.080
14.036
1.691
9.531
86.191
LOD
Co
0.005
LOD
0.006
0.008
0.001
0.006
0.416
LOD
Ni
0.064
LOD
0.232
1.418
0.011
LOD
16.540
19.826
Cu
0.134
0.211
1.126
0.366
LOD
0.520
17.936
3.092
Zr
0.038
LOD
LOD
0.633
LOD
0.031
LOD
8.889
Mo
2.005
1.650
1.744
2.243
LOD
LOD
LOD
LOD
Ru
0.002
0.002
0.001
0.008
LOD
0.004
0.146
LOD
Supplementary Information for Toutov, Liu, Betz, Fedorov, Stoltz, and Grubbs
SI
-
8
Rh
LOD
LOD
LOD
0.001
LOD
LOD
LOD
LOD
Pd
0.014
0.006
0.029
0.116
0.002
0.004
0.070
0.593
Ag
0.001
LOD
0.290
0.015
LOD
0.004
0.055
0.013
Os
0.001
LOD
LOD
0.001
LOD
LOD
0.007
0.016
Ir
0.001
0.001
0.002
0.026
LOD
0.001
0.047
0.041
Pt
0.009
0.004
0.002
0.010
LOD
0.001
LOD
LOD
Au
0.017
0.013
0.013
0.023
0.108
0.024
0.738
1.582
Part II.
Competition
experiments and evaluation of functional group compatibility
.
1
.
Competition experiments with thiophene, furan and pyrrole.
To investigate the relative reactivities of nitrogen
-
, oxygen
-
, and sulfur
-
containing
aromatic heterocycles by KO
t
-
Bu
-
catalyzed C
–
H
silylation, two internal competition
experiments were conducted using one equivalent of Et
3
SiH and one equivale
nt of each
heteroarene (Scheme 1
).
Reactions were run to partial consumption of Et
3
SiH and
relative quantities of silylated heteroarene were det
ermined by
1
H NMR analysi
s. Results
demonstrated that for 5
-
membered heteroarenes, the
relative
rate of reactivity trend
s
as:
thiophene
3q
>
furan
3r
>
1
-
methylpyrrole
3x
(Scheme 1
a). This t
rend is corroborated in
the competition between
substituted thioph
ene
3m
and furan
3n
, as shown in Scheme 1
b.
Scheme 1
.
Competition experiments
Procedure
s
for competition experiments
as shown in Figure 1
:
For reaction (a)
: In a
nitrogen
-
filled glove box, KO
t
-
Bu (11.2 mg, 0.1 mmol, 20 mol%),
thiophene
3q
(42.1
mg, 0.5 mmol, 1 equiv),
furan
3r
(34.0 mg, 0.5 mmol, 1 equiv) and 1
-
methylpyrrole
3x
3q
KO
t
-Bu
(20
mol%)
Et
3
SiH
(1
equiv)
THF,
23
°C,
8
h
N
S
O
3r
3x
S
(1
equiv)
(1
equiv)
(1
equiv)
SiEt
3
O
SiEt
3
N
SiEt
3
SI-4q
SI-4r
4x
+
+
+
(a)
SI-4q:SI-4r:4x
=
5:1:0
3m
KO
t
-Bu
(20
mol%)
Et
3
SiH
(1
equiv)
THF,
23
°C,
8
h
S
O
3n
S
(1
equiv)
(1
equiv)
SiEt
3
O
SiEt
3
4m
4n
(b)
4m:4n
=
5:1
n
-C
5
H
11
n
-C
5
H
11
n
-C
5
H
11
n
-C
5
H
11
+
+
Me
Me
+
Supplementary Information for Toutov, Liu, Betz, Fedorov, Stoltz, and Grubbs
SI
-
9
(
40.5 mg,
0.5 mmol, 1 equiv) were added to a 2 dram scintillation vial equipped with a
magnetic stirring bar.
THF (0.3 mL)
and Et
3
SiH (81 μL, 0.5 mmol, 1 equiv
–
fi
ltered
through a short pad of activated alumina before use
) were then added. The vial was
sealed and stirred at 23 °C for approximately 8 hours. The vial was removed from the
glove box,
diluted with diethyl ether (2 mL) and concentrated under reduced pre
ssure.
Analysis of the
crude reaction mixture by
1
H NMR
revealed that the ratio of
SI
-
4
q
:
SI
-
4
r
:
4x
was 5:1:0.
For reaction (b):
In a nitrogen
-
filled glove box, KO
t
-
Bu (11.2 mg, 0.1
mmol, 20 mol%),
2
-
pentylthiophene
3
m
(77.0 mg, 0.5 mmol, 1 equiv),
and
2
-
pe
ntylfuran
3
n
(69.1 mg, 0.5 mmol, 1 equiv) were added to a 2 dram scintillation vial equipped with
a magnetic stirring bar.
THF (0.3 mL)
and Et
3
SiH (81 μL, 0.5 mmol, 1 equiv
–
filtered
through a short pad of activated alumina before use
) were then added. T
he vial was
sealed and stirred at 23 °C for approximately 8 hours. The vial was removed from the
glove box,
diluted with diethyl ether (2 mL) and concentrated under reduced pressure.
Analysis of the
crude reaction mixture by
1
H NMR
revealed that the ratio
of
4m
:
4n
was
5:1.
2
.
Reactions with electron
-
deficient heteroarenes
.
Scheme 2. Examples of unreactive pyridine derivatives
Pyridine derivatives would be expected to react readily if a conventional silyl radical
addition process was operational.
However, the above substrates were unreactive under
the KO
t
-
Bu
–
catalyzed C
–
H silylation conditions.
This observation argues against the
likelihood of an elementary Minisci
-
type mechanism and suggests that the reaction is
proceeding by an a
lternative and p
resently unidentified
pathway.
Het
H
Het
SiEt
3
KO
t
-Bu
(20
mol%)
Et
3
SiH
(3
equiv)
THF,
65
°C,
72
h
pyridine
N
quinoline
N
isoquinoline
N
acridine
N
No
reaction
or
<5%
product
was
observed
under
above
conditons
with:
pyridine,
quinoline,
isoquinoline,
and
acridine