S
1
Supporting Information
for:
Cp* Non
-
innocence Leads to a Remarkably Weak C
−
H Bond via
Metallocene Protonation
Matthew J. Chalkley, Paul H. Oyala,* and Jonas C. Peters*
Division of
Chemistry and Chemical Engineering, California Institute of Technology (Caltech), Pasadena,
California 91125, United States
Contents:
S.1 General Procedures
Pages
S
2
-
S
6
S.2 Synthetic Procedures
Page
s
S
6
-
S
7
S.3 NMR Characterization of New
Species
Pages
S
7
-
S
11
S.
4
Reactivity of Cp*(
exo
-
η
4
-
C
5
Me
5
H)Co
Page
S
11
-
S
18
S.5 Pulse EPR Spectroscopy
Page
S
19
-
S
21
S.
6
CW EPR Spectroscopy
Page
S
2
1
-
S
2
5
S.7 Electrochemistry
Page
S
2
6
-
S
30
S.
8
X
-
ray Crystallography
Page
S
31
S.9 IR Spectroscopy
Page
S
3
2
-
S
3
3
S.10 Thermochemistry
Page
S
3
4
-
S
3
5
S. 11 DFT Calculations
Page
S
3
5
-
S
3
9
S.
1
2
References
Page
S
3
9
-
S
41
S
2
S.1 General Procedures
:
General Considerations:
All manipulations were carried out using standard Schlenk or glovebox
techniques under an N
2
atmosphere. Unless otherwise noted, solvents were deoxygenated and dried by
thoroughly sparging with N
2
gas followed by passage through an activated alumina colum
n in the solvent
purification system by SG Water, USA LLC. Non
-
halogenated solvents were tested with a standard purple
solution of sodium benzophenone ketyl in tetrahydrofuran in order to confirm effective oxygen and
moisture removal. All reagents were pur
chased from commercial vendors and used without further
purification unless otherwise stated. Cp*
2
Co
,
1
Cp*(
η
4
-
C
5
Me
6
Co),
2
[H(OEt
2
)
2
][BAr
F
4
] (HBAr
F
4
, BAr
F
4
=
tetrakis
-
(3,5
-
bis(trifluoromethyl)phenyl
)borate)
,
3
[Fc][BAr
F
4
] (Fc = ferrocenium)
,
4
and [TBA][BD
4
] (TBA =
tetrabutylammonium, BD
4
= borodeuteride)
5
were
synthesized according to a literature procedu
re.
Deuterated solvents were purchased from Cambridge Isotope Laboratories, Inc. C
6
D
6
and MeCN
-
d
3
were
degassed and stored over activated 3
Å molecular sieves prior to use
. Elemental analysis was performed
by the Beckman Institute Elemental Analysis facili
ty at California Institute of Technology.
1
H and
13
C NMR
chemical shifts are reported in ppm relative to tetramethylsilane, using residual solvent resonances as
internal standards. Solid IR measurements were obtained on a Bruker Alpha spectrometer equ
ipped
with
a diamond ATR probe.
EPR Spectroscopy:
X
-
band (9.4 GHz) CW EPR spectra were acquired using a Bruker EMX spectrometer
equipped with a Super High
-
Q (SHQE) resonator using Bruker Win
-
EPR software (ver. 3.0). Spectra were
acquired at 77 K using a vacuum
-
insulated quartz liquid nitrogen immersion dewar inserted into the EPR
resonator.
Pulse EPR Spectroscopy:
All pulse
Q
-
band (
34
GHz)
electron n
uclear double resonance (ENDOR),
hyperfine sublevel correlation (HYSCORE)
, and electron spin echo detected
field
-
swept
spectra were
acquired using a Bruker ELEXSYS E580 pulse EPR spectrometer equipped with a Bruker
D
2
resonator.
Temperature control was achieved using an ER 4118HV
-
CF5
-
L Flexline Cryogen
-
Free VT cryostat
manufactured by ColdEdge equipped with an
Oxford Instruments Mercury ITC temperature controller.
Q
-
band e
lectron spin
-
echo detected EPR (ESE
-
EPR) field
-
swept spectra were acquired using the 2
-
pulse
“
Hahn
-
echo
”
sequence (
휋
/
2
–
휏
–
휋
–
휏
–
echo) where
휏
was held constant. Subsequently, each fie
ld
swept echo
-
detected EPR absorption spectrum was modified using a pseudo
-
modulation function to
approximate the effect of field modulation and produce the CW
-
like 1
st
derivative spectrum.
6
Specific
acquisition parameters:
휋
/
2
= 12 ns;
휋
= 24 ns;
휏
= 160 ns (6 K spectra), 600 ns (10 K spectra); shot
repetition time (srt) = 5 ms (6 K spectra), 2 ms (10 K spectra).
Q
-
band inversion
recovery experiments were conducted using the pulse sequence
휋
–
푇
–
휋
/
2
–
휏
–
휋
–
휏
–
echo
, where
푇
is a variable delay and
휏
is a fixed delay
.
Specific acquisition parameters:
휋
/
2
= 12 ns;
휋
= 24 ns;
휏
= 160 ns
Q
-
band HYSCORE spectra were ac
quired using the 4
-
pulse sequence (
휋
/
2
−
휏
−
휋
/
2
−
푡
1
−
휋
−
푡
2
−
휋
/
2
–
휏
–
echo), where
휏
is a fixed delay, and
푡
1
and
푡
2
are variable delays independently incremented
by
∆
푡
1
and
∆
푡
2
, respectively. Sixteen step phase cycling was utilized. The time domain
spectra were
baseline
-
corrected (third
-
order polynomial), apodized with a Hamming window function, zero
-
filled to
eight
-
fold points, and fast Fourier
-
transformed to yield the frequency domain.
For
2
H
-
1
H
difference
spectra, the time domain of the HYSCORE s
pectrum of the
1
H sample was subtracted from that of the
2
H
S
3
sample, and the same data processing procedure detailed above was used to generate the frequency
spectrum.
Q
-
band ENDOR spectra were acquired using the Davies pulse sequence (
휋
−
푡
푅퐹
−
휋
푅퐹
−
푡
푅
퐹
−
휋
/
2
–
휏
–
휋
–
echo), where
푡
푅퐹
is the delay between MW pulses and RF pulses,
휋
푅퐹
is the length of the RF
pulse. The RF frequency was randomly sampled during each pulse sequence.
Specific acquisition
parameters:
휋
/
2
= 40 ns;
휋
푅퐹
= 15 μs;
푡
푅퐹
= 2 μs;
휋
= 80 ns;
휏
= 300 ; shot repetition time (srt) = 5 ms
In general, the ENDOR spectrum for a given nucleus with spin
퐼
= ½ (
1
H) coupled to the
S
= ½ electron
spin exhibits a doublet at frequencies
(Eq 1)
휈
±
=
|
퐴
2
±
휈
푁
|
(1)
Where
휈
푁
is the nuclear Larmor frequency and
퐴
is the hyperfine coupling. For nuclei with
퐼
≥
1
(
14
N,
2
H), an additional splitting of the
휈
±
(Eq 2)
manifolds is produced by the nuclear quadrupole interaction
(P)
휈
±
,
푚
퐼
=
|
휈
푁
±
3
푃
(
2
푚
퐼
−
1
)
2
|
(2)
In HYSCORE
spectra, these signals manifest as cross
-
peaks or ridges in the 2
-
D frequency spectrum which
are generally symmetric about the diagonal of a given quadrant. This technique allows hyperfine levels
corresponding to the same electron
-
nuclear submanifold to be
differentiated, as well as separating
features from hyperfine couplings in the weak
-
coupling regime (
|
퐴
|
<
2
|
휈
퐼
|
) in the (+,
+
) quadrant from
those in the strong coupling regime (
|
퐴
|
>
2
|
휈
퐼
|
) in the (
−
,
+
) quadrant. The (
−
,
−
) and (+,
−
) quadrants of
these freque
ncy spectra are symmetric to the (+,+) and (
−
,+) quadrants, thus typically only two of the
quadrants are typically displayed in literature.
F
or systems with appreciable hyperfine anisotropy in frozen solutions or solids, HYSCORE spectra
typically do not e
xhibit sharp cross peaks, but show ridges that represent the sum of cross peaks from
selected orientations at the magnetic field position at which the spectrum is collected. The length and
curvature of these correlation ridges allow for the separation and
estimation of the magnitude of the
isotropic and dipolar components of the hyperfine tensor, as shown in Fig.
S1
.
S
4
Figure
S1
:
a)
HYSCORE
powder
patterns for an
S
= 1/2,
I
= 1/2 spin system with an
isotropic
hyperfine
tensor
A. b)
HYSCORE powder patterns
for an
S
= 1/2,
I
= 1/2 spin system with an isotropic hyperfine
tensor
which contains isotropic (
푎
푖푠표
) and dipolar (
푇
) contributions. Blue correlation ridges represent the
strong coupling case; red correlation ridges represent the weak coupling case.
EPR simulations:
All CW and pulse EPR spectra were simulated using the EasySpin
7
suite of programs
with Matlab 2017 u
sing the following Hamiltonian
(Eq 3)
:
퐻
̂
=
휇
퐵
퐵
⃑
0
푔
푆
̂
+
휇
푁
푔
푁
퐵
⃑
0
퐼
̂
+
ℎ
푆
̂
∙
푨
∙
퐼
̂
+
ℎ
퐼
̂
∙
푷
∙
퐼
̂
(3)
In this expression, the first term corresponds to the electron Zeeman interaction term where
휇
퐵
is the
Bohr magneton, g is the electron spin g
-
value matrix with principle components
g
= [g
xx
, g
yy
, g
zz
], and
푆
̂
is
the electron spin operator; the second term corresponds to the nuclear Zeeman interaction term where
휇
푁
is the nuclear magneton,
푔
푁
is
the characteristic nuclear
g
-
value for each nucleus (e.g.
1
H,
2
H,
31
P)
and
퐼
̂
is the nuclear spin operator; the third term corresponds to the electron
-
nuclear hyperfine term,
where
퐴
is the hyperfine coupling tensor with principle components
퐴
= [A
xx
, A
yy
, A
zz
]; and for nuclei with
퐼
≥
1
, the final term corresponds to the nuclear quadrupole (NQI) term which arises from the
interaction of the nuclear quadrupole moment with the local electric field gradient (efg) at the nucleus,
where
푷
is
the
quadrupole
coupling tensor. In the principle axis system (PAS),
푷
is traceless and
parametrized by the quadrupole coupling constant
푒
2
푞푄
/
ℎ
and the asymmetry parameter
휂
such that
(Eq 4)
:
S
5
푷
=
(
푃
푥푥
0
0
0
푃
푦푦
0
0
0
푃
푧푧
)
=
푒
2
푄푞
/
ℎ
4
퐼
(
2
퐼
−
1
)
(
−
(
1
−
휂
)
0
0
0
−
(
1
+
휂
)
0
0
0
2
)
(4)
where
푒
2
푞푄
ℎ
=
2
퐼
(
2
퐼
−
1
)
푃
푧푧
and
휂
=
푃
푥푥
−
푃
푦푦
푃
푧푧
. The asymmetry parameter may have values between 0
and 1, with 0 corresponding to an electric field gradient with axial symmetry and 1 corresponding to a
fully rhombic efg.
The orientations between the hyperfine and NQI tensor principle axis systems and the
g
-
matrix
reference frame are defined by the Euler angles
(α, β, γ).
X
-
Ray Crystallography:
XRD studies were carried out at the Beckman Institute Crystallography Facility
on
a
Bruker
Kappa Apex II diffractometer (Mo K
α
radiation). Structures were solved using SHELXS or SHELXT
and refined against F
2
on all data by full
-
matrix least squares with SHELXL.
8
All of the solutions were
performed in the Olex2 program.
9
The crystals were mo
unted on a glass fiber under
Paratone N oil or
fluorolube.
Electrochemistry:
Electrochemical measurements were carried out in a thick
-
walled one
-
component
electrochemical cell fitted with a Teflon stopcock and tungsten leads protruding from the top of
app
aratus. A CH
instruments 600B electrochemical analyzer was used for data collection. A freshly
-
polished glassy carbon electrode was used as the working electrode, a platinum wire was used as the
auxiliary electrode, and a silver wire as a reference electro
de. The analyte was used in 1 mM
concentration.
The
solvent and the
concentration of the electrolyte are noted with the individual
voltammograms.
After the desired scans were completed, ferrocene (1 mM) was added to serve as an
internal reference
or the kn
own decamethylcobaltocenium
/decamethylcobaltocene
couple was used as
the internal reference
. All reported potentials are referenced to the ferrocenium
/ferrocene
couple
(Fc
+/0
).
Calculations:
All calculations
were performed using the ORCA
4.0
program.
10,11
In cases where crystal
structures were available these coordinates were used as the input
. The
calculations were performed
using the
TPSS
(meta
-
GGA)
12
functional with the
def2
-
SVP
basis set was on C and H and the def2
-
TZV
P
basis set
on Co
13
and Grimme
-
d3 dispersion correction.
14
That optimized structures represented true
stationary points was checked by doing a single
-
point frequency calculat
ions on the optimized structure
.
For species where calculating the EPR parameters was of interest further calculations were performed in
order to verify the robustness of those results. U
sing
the
structures
optimized as described a
bove
additional optimizations were performed with other functionals: BP86 (GGA),
15,16
B3LYP
(hybrid+GGA),
15,17,18
and TPSSH (hybrid+meta
-
GGA).
12,19
In all cases the ring
-
functionalized structures
were found to be minima
by means of a frequency calculation
. For the Co
–
H structures in all cases
(except with TPSS)
there w
as a small negative frequency that did involve motion of the Co
–
H between
the top and bottom ring suggesting that with these other functionals that this structure may only be a
transition state. This is consistent with our previous observations with M06
-
L
20
wher
e a Co
–
H structure
could not be optimized.
21
In all cases the thermochemistry was very similar.
EPR
parameters
for the TPSS
-
optimized structure
were calculated by doing a single point calculation on
the optimized structure
s
using CP(PPP)
22
on
Co
and
def2
-
TZVP
on
C and H
with Grid
6 and TPSSH as the
S
6
functional.
To check the robustness of
this basis set a higher level calculation was also done using
CP(PPP) on Co and EPR
-
III
23
on C and H with grid 7. The results were very similar. Thus the EPR
parameters were also calculated using CP(PPP) on Co and def2
-
TZVP on C and H with TPS
S, BP86, and
B3LYP. Lastly, the EPR parameters for the structures optimized using TPSSH, BP86, and B3LYP were all
calculated via a single point calculation using TPSSH with CP(PPP) on Co and def2
-
TZVP on C and H with
Grid6.
See below for a discussion of th
e results.
For the calculation of
thermochemical properties (reduction potential, p
K
a
, and hydricity)
an additional
solvation calculation was done using the CPCM solvation model with acetonitrile solvent to determine
the solvated internal energy (E
soln
).
24
-
26
Free energies of solvation were approximated using the
difference in gas phase internal energy (E
gas
) and solvated internal energy
(∆
G
solv
≈
E
soln
–
E
gas
) and the
free energy of a species in solution was then calculated using the gas phase free energy (G
g
as
) and the
free energy of solvation (G
soln
= G
gas
+ ∆
G
solv
).
27,28
The calculations of BDFE were done without the
additional solvation correction as there is no change in charge for either reactants or products and it was
found to introduce add
itional error in previous studies.
S.2
Synthetic Procedures:
Cp*(
exo
-
η
4
-
C
5
Me
5
H
)Co
-
[Cp*
2
Co][PF
6
] (
100.0
mg,
0.21
mmol) was added as a solid to a
THF solution (1
5
mL) of [TBA][BH
4
] (
271.2
mg,
1.05
mmol,
5.0
equiv)
in a Schlenk tube
. This
reaction was then
heated to
reflux and allowed to stir overnight
.
The reaction mixture was cooled to room temperature and the
solvent was evaporated. The orange solid was extracted with pentane (8 x 5 mL) and filtered through an
alumina plug.
The solvent was then removed un
der reduced pressure to yield an orange solid.
X
-
ray
diffraction quality crystals were obtained by slow evaporation of a concentrated pentane solution.
(Yield: 40.2 mg, 57.8%).
1
H NMR (RT,
C
6
D
6
, 400 MHz): δ =
2.11 (1H, m
, C(
H
)
-
CH
3
), 1.87 (6H, s
, β
-
C
H
3
), 1.65 (15H, s, Cp*), 1.42 (3H,
d,
4
J(H
-
H) = 6.7 Hz
, C(H)
-
C
H
3
) 0.76 (6H, d,
5
J(H
-
H) = 1.2 Hz
, α
-
C
H
3
).
13
C{
1
H} NMR (RT, C
6
D
6
, 100 MHz): δ =
87.74, 86.14, 53.21, 51.66,
17.08, 11.39, 10.21, 9.55.
IR (Thin Film): 2
7
08 and 2612
cm
-
1
(ν
C
–
H
for the ring
-
bound
C
–
H)
.
UV/Vis (2
-
MeTHF
):
nm
[
cm
-
1
M
-
1
]
): 463
[
220
]
, 565
[
350
]
.
Elemental Analysis: theory [C
:
7
2
.
71
H
:
9.
46
];
found [C
: 72.82
, H
: 9.64
]
See the IR spectra for a further discussion of why two stretches are observed for the C−H mode.
Cp*(
exo
-
η
4
-
C
5
Me
5
D)Co
-
[Cp*
2
Co][PF
6
] (
100
mg,
0.21
mmol)
and
[
TBA
][BD
4
] (
275.4
mg,
1.05
mmol,
5
equiv)
were added to a Schlenk tube as solids
.
A small stir bar and 20 mL of THF were then added.
The
reaction was then heated to reflux overnight. At this point the solvent was remo
ved
in vacuo
and the
solid extracted with
(8 x 5 mL)
pentane. This material was then filtered
through alumina and
the solvent
was removed under reduced pressure
.
(
Y
ield:
36.5
mg,
52
.0
%
)
1
H NMR (RT, C
6
D
6
, 400 MHz): δ = 1.87 (6H, s
, β
-
C
H
3
), 1.65 (15H, s, Cp*), 1.42 (3H,
s, C(D)
-
C
H
3
) 0.76 (6H,
s,
α
-
C
H
3
).
A small residual peak from the protio
-
incorporated material can be observed at 2.11 ppm.
Integration of this peak suggests that there has been ~95% deuterium incorporation.
13
C{
1
H} NMR (RT,
C
6
D
6
, 100 MHz): δ = 51.21 (t,
2
J(C
-
D) = 18.0 Hz
)
.
2
H
{
1
H}
NMR (RT, 90% C
6
H
6
:10% C
6
D
6
, 61.42 MHz): δ = 2.05 (s).
IR (Thin Film): 2001 and 1957 cm
-
1
(ν
C
–
D
for the ring
-
bound C
–
H).
S
7
Cp*(
exo
-
η
4
-
C
5
Me
5
13
Me)Co
-
Cp*
2
Co (50 mg, 0.15 mmol) was dissolved in toluene and chilled to −78 °C
with stirring. To this was added dropwise
13
C
-
MeOTf (75 mg, 0.45 mmol, 3 eq). The reaction was stirred
for one hour at −78 °C followed by warming to room temperature for five hours. The r
eaction was then
filtered to remove [Cp*
2
Co][OTf]. The desired product could then be obtained by evaporation. (Yield:
23.1 mg, 87%)
1
H NMR (RT, C
6
D
6
, 400 MHz): δ = 1.84 (6H, s), 1.65 (15H, s, Cp*), 1.42 (3H, d,
3
J(C
-
H) = 4.5
Hz) 0.76 (6H, s), 0.47 (3H, d,
1
J(C
-
H) = 124.4 Hz).
13
C{
1
H} NMR (RT, C
6
D
6
, 100 MHz): δ = 26.38.
S.3 NMR Characterization of New Species:
Figure S2:
1
H NMR spectrum (400 MHz, C
6
D
6
, 25 °C) of Cp*(
exo
-
η
4
-
C
5
Me
5
H)Co.
S
8
Figure S3:
1
H
-
COSY NMR spectrum (400 MHz, C
6
D
6
, 25 °C) of Cp*(
exo
-
η
4
-
C
5
Me
5
H)Co. This data was used
to establish the chemical shift of the Me groups that are α and β to the quaternary carbon.
Figure S4:
13
C NMR spectrum (100 MHz, C
6
D
6
, 25 °C) of Cp*(
exo
-
η
4
-
C
5
Me
5
H)Co.
S
9
Figure
S
5
:
1
H NMR spectrum (400 MHz, C
6
D
6
, 25 °C) of
Cp*(
exo
-
η
4
-
C
5
Me
5
D)Co. Integration of the residual
1
H signal at 2.11 i approximately 95% deuterium incorporation.
Figure S6:
2
H
{
1
H}
NMR spectrum (61.42 MHz, 90% C
6
H
6
:10% C
6
D
6
, 25 °C) of Cp*(
exo
-
η
4
-
C
5
Me
5
D)Co. In
addition to the expected deuterium incorporation into the
exo
-
position it appears that there is also a
small amount of deuterium incorporation into the methyl group that is attached to the quaternary
carbon.
S
10
Figure S
7
:
13
C{
1
H} NMR spectrum (100
MHz, C
6
D
6
, 25 °C) of Cp*(
exo
-
η
4
-
C
5
Me
5
D)Co. Here we can also see
a small amount of contamination with the protio species at 51.66. The triplet is centered at 51.21
demonstrating the expected change in chemical shift upon deuteration.
Figure S8:
1
H NMR
spectrum (400 MHz, C
6
D
6
, 25 °C) of Cp*(
exo
-
η
4
-
C
5
Me
5
13
Me)Co demonstrating the
13
C
-
1
H coupling evident in the splitting of the peak for the α
-
protons at 0.46 ppm and the smaller coupling to
the γ
-
protons at 1.41 ppm.
S
11
Figure S9:
13
C NMR spectrum (100 MHz,
C
6
D
6
, 25 °C) Cp*(
exo
-
η
4
-
C
5
Me
5
13
Me)Co highlighting the selective
13
C incorporation.
S.
4
Reactivity
of Cp*(
exo
-
η
4
-
C
5
Me
5
H)Co
:
Figure S
10
: Reaction of
Cp*(
exo
-
η
4
-
C
5
Me
5
H)Co with
4MeO
TEMPO.
Cp*(
exo
-
η
4
-
C
5
Me
5
H)Co
(
1.5 mg
,
0.005
mmol
, 1 eq) and
4MeO
TEMPO (
1.9 mg
,
0.010
, 2
.2
eq) were dissolved
in MeCN
-
d
3
. These were then allowed to react at room temperature for thirty minutes
with shaking. A
1
H NMR was then taken. In this the formation of 2 equivalents of
4MeO
TEMPOH is observed
(Figure S
9
)
29
confirmed by integration relative to a benzene internal standard containing 2 equivalents. However,
the
Co product is unclear. On the basis of reactions performed in toluene
, we believe that this
is because of
the high instability of the fulvene product, Cp*(η
4
-
C
5
Me
4
CH
2
)Co,
in coordinating solvents.
S
12
Figure S
11
:
1
H NMR spectrum (400 MHz, MeCN
-
d
3
, RT) of the reaction between Cp*(
exo
-
η
4
-
C
5
Me
5
H)Co
and
4MeO
TEMPO.
Due to the poor solubility of the
reagents in acetonitrile, the limited ability to cool that solvent, and that
the NMR chemical shifts of the fulvene complex have only been previously reported in aromatic solvents
the reaction was performed again in toluene
-
d
8
.
30,31
Cp*
(
exo
-
η
4
-
C
5
Me
5
H)Co (4
mg, 0.0
12
mmol, 1 eq)
and
4MeO
TEMPO (4.5 mg, 0.024 mmol, 1 eq) were each dissolved in minimal
d
8
-
toluene and both added
to a J
-
Young NMR tube
(Figure S
10
)
. Due to broadening at room temperature the reaction mixture was
cooled to −78 °C
and a
1
H (Figure S
1
1
) and
13
C NMR were taken
(Figu
re S1
3
)
.