Zero-Valent Amino-Olefin Cobalt Complexes as Catalysts for Oxygen Atom Transfer Reactions from Nitrous Oxide

Article abstract of DOI:10.1002/anie.201609173

The synthesis and characterization of several zero-valent cobalt complexes with a bis(olefin)-amino ligand is presented. Some of these complexes proved to be efficient catalysts for the selective oxidation of secondary and allylic phosphanes, as well as diphosphanes, even with a direct P?P bond. With 5 mol % catalyst loadings the oxidations proceed under mild conditions (25–70 °C, 7–22 h, 2 bar N₂O) and afford good to excellent yields (65–98 %). In this process, the greenhouse gas N₂O is catalytically converted into benign N₂and added-value organophosphorus compounds, some of which are difficult to obtain otherwise.

Full text of DOI:10.1002/anie.201609173

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201609173
German Edition: DOI: 10.1002/ange.201609173
N O Activation
2
Zero-Valent Amino-Olefin Cobalt Complexes as Catalysts for Oxygen
Atom Transfer Reactions from Nitrous Oxide
Thomas L. Gianetti,* Rafael E. Rodrꢀguez-Lugo, Jeffrey R. Harmer, Monica Trincado,
Matthias Vogt, Gustavo Santiso-Quinones, and Hansjçrg Grꢁtzmacher*
Abstract: The synthesis and characterization of several zero-
valent cobalt complexes with a bis(olefin)-amino ligand is
presented. Some of these complexes proved to be efficient
catalysts for the selective oxidation of secondary and allylic
phosphanes, as well as diphosphanes, even with a direct PꢀP
In 1971 Yamamoto and co-workers reported the first
catalytic oxidation of PPh with nitrous oxide (N O), using
3
2
a hydrido dinitrogen tris(triphenyphosphane) cobalt(I) com-
[
7f]
plex, [CoH(N )(PPh ) ]. Since that initial report, several
2
3 3
[7]
other Co complexes were reported for the same reaction.
bond. With 5 mol% catalyst loadings the oxidations proceed
However, as for the traditional oxygenation with peroxides,
these metal-based catalytic systems are restricted to the
oxidation of stable tertiary tris-aryl/tris-alkyl monophos-
phanes such as PPh3.
under mild conditions (25–708C, 7–22 h, 2 bar N O) and
2
afford good to excellent yields (65–98%). In this process, the
greenhouse gas N O is catalytically converted into benign N2
2
and added-value organophosphorus compounds, some of
which are difficult to obtain otherwise.
The transformation of the industrial waste N O into less
harmful chemicals, while generating useful/added-value prod-
2
ucts is of particular interest since N O has been recently
2
[
8]
P
hosphane oxides find applications mainly as: 1) ligands in
identified as one of the largest global ozone depleting agents
and has a greenhouse gas effect 300-times higher than CO2.
[
8b]
metal-catalyzed cross-coupling reactions (secondary phos-
[
1]
phane oxides, O=PHR ), 2) contact doping for silicon wafers
and nanostructures, 3) photoinitiators, 4) constituents of
phosphorescent organic light emitting diodes (the so-called
Metal complexes which react stoichiometrically with N O can
2
2
[
2]
[3]
be found in the literature, resulting in the oxidation of low-
[
9]
valent metal centers via OAT (oxygen atom transfer),
insertion of into metal–carbon or metal–hydride
[4]
PHOLEDs), and 5) extracting agents for actinides, lantha-
O
[
5]
[9d,10]
[11]
nides and transuranic elements, to name a few. Traditionally
they are prepared from phosphanes by aerobic oxidation or
treatment with peroxides. These methods can be cumbersome
because of low selectivity and functional-group tolerance
which may lead to complicated work-up protocols and/or
generation of chemical waste. Additionally, these methods are
not suitable for highly sensitive phosphanes containing C=C,
bonds,
and metal-mediated NꢀN bond cleavage. Addi-
tionally, the oxygenating power of N O has been successfully
exploited by using heterogeneous and homogeneous cata-
lysts, but especially for the latter the number of examples
remain scarce. Among them, we recently reported the
rhodium-catalyzed dehydrogenative coupling of various alco-
hols using N O as hydrogen acceptor, yielding N and H O as
2
[12]
[
13]
2
2
2
[
6]
[14]
PꢀH or PꢀP reactive bonds. Therefore, a method for the
clean, selective and mild oxidation of a large variety of
phosphanes remains of practical interest.
the only by-products.
Herein we report the synthesis and characterization of
several zero-valent amino-olefin cobalt complexes, which are
suitable catalysts for the selective oxidation of a variety of
phosphanes with nitrous oxide under mild reaction condi-
tions, including highly O /H O sensitive phosphanes, which
[
*] Dr. T. L. Gianetti, Dr. R. E. Rodrꢀguez-Lugo, Dr. M. Trincado,
Dr. M. Vogt, Dr. G. Santiso-Quinones, Prof. Dr. H. Grꢁtzmacher
Department of Chemistry and Applied Biosciences, ETH Zꢁrich,
Laboratory of Inorganic Chemistry
2
2
are difficult to oxidize selectively.
The complex [Co(trop NH)(PPh )] (1) was prepared by
2
3
mixing an equimolar amount of [CoCl(PPh ) ] and bis(5H-
3
3
Vladimir-Prelog-Weg 1, 8093 Zꢁrich (Switzerland)
E-mail: gianetti@inorg.chem.ehtz.ch
dibenzo[a,d]cyclohepten-5-yl)amine (trop NH) in the pres-
2
hgruetzmacher@ethz.ch
ence of an excess of Zn powder as reducing agent (Sche-
me 1a). Deep red crystals were obtained in ca. 70% yield by
layering a DME solution with n-hexane. H NMR spectro-
Dr. J. R. Harmer
Centre for Advanced Imaging, University of Queensland
Brisbane, QLD, 4072 (Australia)
1
scopy revealed the presence of broadened and shifted
0
Dr. R. E. Rodrꢀguez-Lugo
resonances suggesting the formation of a paramagnetic Co
Current address: Laboratorio de Quꢀmica Bioinorgꢂnica, Centro de
Quꢀmica, Instituto Venezolano de Investigaciones Cientꢀficas (IVIC),
Caracas 1020-A (Venezuela)
species (see Figure S1 in the Supporting Information (SI)).
The zero-valent cobalt complex 1 was used as precursor to
prepare the ylidene 2–4 and phosphite 5 analogues (Sche-
me 1b). Addition of the corresponding N-heterocyclic car-
bene (NHC) (TMIY= 1,3,4,5-tetramethylimidazole-2-yli-
dene, DIIY= 1,3-diisopropylimidazole-2-ylidene, EMIY= 1-
Dr. M. Vogt
Current address: Institut fꢁr Anorganische Chemie und Kristallo-
graphie, Universitꢃt Bremen, Leobener Str. NW2, 28359 Bremen
(
Germany)
ethyl-3-methylimidazole-2-ylidene) or P(OPh) to a toluene
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under http://dx.doi.org/10.
3
solution of 1 resulted in the fast exchange of PPh according to
3
3
1
1
1002/anie.201609173.
P{ H} NMR spectroscopy. Addition of seven equivalents of
Angew. Chem. Int. Ed. 2016, 55, 1 – 7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Figure 1. ORTEP drawing of the molecular structure of 1·dme. Thermal
ellipsoids are drawn at a 50% probability level. Hydrogen atoms and
a molecule of dme have been removed for clarity (see Table S1 for
crystallographic parameters).
2
2
significant elongation from an unperturbed C(sp )-ꢀ(sp )
bond (1.34 ꢀ) suggests a strong p-back-donation from the
metal center to the olefins. Zero-valent cobalt complexes
Scheme 1. Synthesis of cobalt complexes 1–5 in 72%, 75%, 80%,
4% and 94% yield, respectively.
6
2
bearing h -olefinic ligands are scarcely discussed in the
[
15]
literature, and structurally well-characterized compounds
[
15e]
PPh -d to a toluene solution of 1 in an J-Young NMR tube
are rare.
3
15
also resulted in a reversible phosphane exchange as suggested
X-band CW-EPR spectra at room temperature and in
frozen solution (120 K) were recorded and simulated with the
EasySpin software for 1, 2, and 3 (Figure 2, Figures S9–S11
2
by the appearance of new resonances in the H NMR
0
[16]
spectrum assigned to the PPh -d bound to Co (see Fig-
3
15
ure S2). The effective magnetic moments of all compounds in
THF solution at room temperature were determined by using
the Evans method, and revealed that the four complexes
and Table S2). In frozen solution, all three complexes gave
rhombic spectra (g ¼
6
g ¼
6
g3) which are all distinguished by
1
2
a well-resolved eight line splitting along g due to the cobalt
3
co
possess one unpaired electron (meff = 1.8–1.9 m ). Additional
hyperfine interaction A that was accurately determined by
B
3
5
9
magnetic susceptibility measurements (SQUID, T= 2–300 K)
of crystalline [Co(trop NH)(PPh )] (1) and [Co(trop NH)-
simulation (I = 7/2, Co). Each cobalt line is further split into
1
4
a triplet due to the nitrogen nucleus ( N, I = 1) of the trop-
2
3
2
N
(
TMIY)] (2) were performed and gave an effective magnetic
based ligand with hyperfine coupling A ffi27 MHz. At low
3
moment of meff = 1.83 and 1.78 m , respectively, which support
field, both the g-values and cobalt hyperfine couplings are not
clearly resolved but were well estimated from simulations
aided by the constraint Aiso = (A + A + A )/3 (same for g )
B
9
the solution-state data and the presence of d cobalt(0)
complexes (see Figure S3).
1
2
3
iso
X-ray quality crystals were grown from DME solutions
layered with n-hexane, allowing for X-ray diffraction studies
of complexes 1–5. The ORTEP diagram of 1 is shown in
Figure 1, while the plots for complex 2–5 along with the
corresponding crystallographic parameters are presented in
the SI (Figures S4–S8, Table S1). In all four complexes, the
cobalt center resides in a distorted tetrahedral coordination
environment, and the CoꢀN1, Coꢀct1(ct2) and C=C bond
trop
lengths are in the range of 2.09–2.14 ꢀ, 1.90–1.92 ꢀ and 1.42–
1
.45 ꢀ, respectively (see Table S1). Typically, the angles ct1-
Co-ct2 and N1-Co-ct1(ct2) strongly deviate from the ideal
tetrahedral angle (1098): For complex 1 ct1-Co-ct2 = 1348 and
N1-Co-ct1 = 938. An interesting feature of the structures of 1–
Figure 2. X-band CW EPR spectrum of 1, 2, and 3 recorded at 120 K;
black: experiment, red: simulation. The principal g values and cobalt
5
is the existence of p-stacking between the proximal trop
hyperfine couplings are: 1: g=[2.275Æ0.002, 2.258Æ0.002,
benzene rings (3.39–3.57 ꢀ). Such intramolecular interactions
contribute to the large ct1-Co-ct2 angles (ca. 1348) and the
distortion from an ideal tetrahedron. The average CꢀC bond
Co
1
2
3
.996Æ0.001], A =[21Æ4, 6Æ4, 158Æ1] MHz; 2: g=[2.346Æ0.006,
Co
.235Æ0.004, 1.994Æ0.001], A =[75Æ15, 39Æ5, 164Æ1] MHz; and
Co
: g=[2.358Æ0.006, 2.230Æ0.004, 1.995Æ0.001], A =[85Æ10,
length of the coordinated olefins (C=C ) is 1.43 ꢀ. This
33Æ5, 155Æ1] MHz.
trop
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 7
These are not the final page numbers!

Angewandte
Communications
Chemie
t
with the isotropic values giso and Aiso determined in solution
the oxidation of Bu PH·EBr to the phosphane oxide adduct
2 3
Co
t
(
Figures S9–S11). The resulting g and A principal values are
( Bu) PH(O)·EBr using molecular oxygen (E = B, Al, Ga or
Typically, such a phosphane oxide is obtained from
( Bu) PX (X = Cl, I) upon hydrolysis and from BuLi and P or
diphenyl phosphonate.
nylvinylphosphane was oxidized (entry 5), while the pendant
terminal olefin remained untouched as confirmed by
2
3
Co
[19a]
given in Figure 2 and the large g anisotropy and large A
couplings indicate a metal-centered radical. The g-value order
In).
t
t
2
4
9
[19b,c]
with g (g , g ) > g (g ) ꢁ 2 is consistent with a d valence
The phosphorus atom in diphe-
?
1
2
k
3
electron configuration for a distorted tetrahedral coordina-
tion environment where the unpaired electron is in an orbital
[
15a,17]
1
with high dz
2
character.
DFT computations on the
H NMR spectroscopy (see the SI). Bis(diphenylphosphano)-
radicals 1, 2, and 3 show that all are metal centered (Co
Mulliken spin population of 1.33) and the computed g and A
methane was converted to the monoxide in good yield and
selectivity (the dioxide is obtained in ca. 10% yield,
Co
[
20]
principal values model well the corresponding experimental
values (Table S2). Thus both experimental and theoretical
entry 6). The substrate trans-1,2-bis(diphenylphosphano)-
ethylene, with three oxygen-sensitive groups, was trans-
formed to the monoxide with high selectivity (entry 7). Only
ca. 1% bis(oxygenation) of both phosphorus atoms was
0
results show that 1, 2 and 3 are Co metalloradicals.
[18a]
[
Co(trop NH)(PPh )] (1) (5 mol%)
was found to react
2
3
[20]
with N O and to be a selective and efficient catalyst for the
observed. There is no spectroscopic evidence of substrate
2
1
31
1
oxygenation of triphenyl phosphite under mild conditions
epoxidation under these reaction conditions ( H, P{ H} and
1
3
1
(
Table 1, entry 1). The scope of the catalytic system was
C{ H} NMR). The highly moisture and air sensitive diphos-
i
extended to different secondary and tertiary phosphanes,
including aromatic, aliphatic and vinylic substrates as well as
diphosphanes (Table 1, entries 2–8). Interestingly, the clean
conversion of the aliphatic and highly flammable secondary
phosphanes such as diphenylphosphane (entry 3) and di-tert-
butylphosphane (entry 4) to their corresponding oxides is
observed without affecting the PꢀH bond as indicated by both
phane Ph P-P(N Pr ) was converted to the monoxide in
2
2 2
moderate yields without affecting either the PꢀP bond or the
1
13
P(NR ) moiety (entry 8), as confirmed by H and C NMR,
2
2
3
1
1
1
as well as P{ H} spectroscopy ( J = 224.6 Hz) and HRMS
(incorporation of only one oxygen atom, [M+Na] = 455.2351
PP
+
uma).
The addition of butylated hydroxytoluene (BHT), 9,10-
dihydroantracene, or triphenylmethyl radical to the reaction
mixture had no significant influence on the conversion rate
3
1
1
1
1
P{ H} NMR ( J = 475 Hz and J = 426 Hz, respectively)
PH
PH
[
18b]
and HR-MS (see the SI).
While di-phenylphosphane can
[
21]
be selectively oxidized with tert-butyl peroxide, to date, no
(see entries 3–5, Table S3).
No reaction was observed
records have been found in the literature for the direct
between the zero-valent cobalt complexes 1–5 and these
radical traps. In the absence of phosphane substrate, all five
t
t
oxidation of ( Bu) PH to (O)PH( Bu) . The closest example is
2
2
complexes reacted with N O, slowly at room temperature but
2
within minutes at 608C, leading to a mixture of products
Table 1: Cobalt-catalyzed oxidation of different phosphorus-based sub-
strates.
which could not be identified (except OPR due to oxidation
3
[
a]
of the phophane ligand of complex 1 or 5, see above). All
attempts to isolate intermediates or to identify in situ formed
organometallic species or other radicals using EPR spectros-
1
31
copy as well as H and P NMR spectroscopy at low or high
temperature were unsuccessful.
Comparison of the different zero-valent Co complexes
Entry
PR3
O=PR3
T [8C],
t [h]
Yield
[%]
[
b]
[Co(trop NH)(L)] showed that among the carbene com-
2
plexes, [Co(trop NH)(TMIY)] (2) had a relatively poor
performance even at elevated temperature (entry 8,
2
1
2
3
4
P(OPh)3
PPh3
O=P(OPh)3
O=PPh3
70, 16
70, 16
25, 16
60, 15.5
69
63
84
66
P(H)Ph2
O
=
P(H)Ph2
Table S3). The analogous [Co(trop NH)(DIIY)] (3) and
2
t
t
P(H) Bu2
O=P(H) Bu2
[Co(trop NH)(EMIY)] (4) showed slow oxidation of P(OPh)
2
3
at room temperature (entries 9 and 11, Table S3), but
afforded an excellent yield at 608C (entries 10 and 12,
Table S3). It appears that the rate of reaction qualitatively
increases with decreasing coordinating strength of the ligand
5
6
70, 16
25, 6.5
86
[
[
c]
73
74
(
PPh < DIIY= EMIY< TMIY). These data suggest a mech-
3
anism in which displacement of L by either the substrate or
N O is necessary. Finally, oxidation of P(OPh) with N O in
d]
2
3
2
7
8
60, 16.5
70, 22
the presence of 10 mol% of Co (CO)8 or in absence of
2
catalyst was found to be very slow emphasizing the role of 1–5
as catalysts (entries 13 and 14, Table S3).
[
e]
The kinetics of the catalytic OAT reaction from N O to
2
65
3
1
PPh were studied by P NMR spectroscopy at 508C (see the
3
SI). At low concentration of phosphane ([PPh ] = 0.05–
3
0
.15m), the data obtained under pseudo-first-order conditions
[
a] Conditions: PR (0.20 mmol), N O (2.0 bar, 20 mL flask), 1 (5 mol%),
3
2
(
excess PN2O = (PN2O) and [1] = [1] ) showed an excellent fit
THF (5 mL). [b] Isolated yield. [c] 10% of full oxidation product. [d] 1% of
dioxide product. [e] P NMR yield, 35% of unreacted substrate.
0
0
3
1
st
to a rate law that is zero-order in phosphane (see Figure 3), 1
Angew. Chem. Int. Ed. 2016, 55, 1 – 7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
activation step (see the SI for computational details). Free
Gibbs energy differences for the most favorable minimum
energy reaction pathway (MERP) are presented in the
Scheme 2. The free Gibbs energies were obtained by adding
3
1
Figure 3. Oxidation of PPh followed by P NMR spectroscopy using
3
PN2O =2.10 bar and different catalyst concentrations ([1]=1.4–7.9
mm).
st
nd
order in [1] and 1 order in (PN2O)0 leading to an overall 2
0
order rate law (Figures S12–S17). Under these conditions, the
kinetics were analyzed over a temperature range of 30 to
nd
6
08C to give activation parameters for the 2 order reaction
ꢀ
ꢀ1
ꢀ
as DH = 12.0 Æ 0.5 kcalmol
and DS = ꢀ32 Æ 2 calm-
ꢀ
1
ꢀ1
ol
K
(see SI). The rather large negative activation entropy
indicates a ternary transition state.
At higher concentration of phosphane ([PPh ] = 0.25–
3
0
.35m) the data slightly deviate from a zero order in
phosphane, although kinetic traces could still be fit reason-
nd
ably to the same overall 2 order rate law (see Figures S18
ꢀ
1
Scheme 2. Free Gibbs energy differences (in kcalmol ) for the activa-
and S19). However, under these conditions a phosphane
inhibition was observed, resulting in an inverse saturation
behavior (see Figure S20).
tion and splitting of N O at the cobalt complex 1 (entitled I-1d) and
the oxidation of PPh . The labels d indicate the doublet, q the quartet
state. Free Gibbs energy=E(pcm+D3(bj)) + G(298 K) with
E(pcm+D3(bj)) calculated by using B3LYP/(6-311G+ +(2d,p) for H,
C, N, O, P; SDD Def2-TZVP for Co), G(298 K) by using B3LYP/(6-
31G(d,p) for H, C, N, O, P; SDD for Co).
2
3
The kinetical analysis suggests that: 1) N O coordination
2
and/or activation is the rate-limiting step, consistent with the
lack of an observable intermediate; 2) the PPh /N O ligand
3
2
st
exchange occurs via an associative step (1 order in N O and
2
ꢀ
large negative DS ); 3) the ligand exchange is best described
as an interchange associative mechanism (loss of phosphane is
required leading to phosphane inhibition at high concentra-
the gas-phase Gibbs contribution at 298 K to the SCF energy,
that include PCM using THF as solvent and the D3(BJ)
dispersion correction (see the SI). Ligand exchange occurs to
tion); 4) the N O activation step requires only one cobalt
2
st
complex (1 order in [Co]); 5) phosphane is not involved in
form a 17-electron N O adduct (I-2A-d, Scheme 2). Both 15-
2
the N O activation transition state (zero order in [PPh ]).
(I-2B-d) and the 19-electron species (I-2C-d) were found
2
3
These observations are consistent with the mechanism
presented in Equation (1), which follows the rate law shown
in Equation (2). This rate law yields two limiting rate laws: if
higher in energy, supporting the I mechanism. Unfortunately,
a
attempts to locate the I transition state were unsuccessful.
a
Oxygen transfer to the metal center occurs via a barrier-less
ꢀ
ꢀ1
k @ k [PPh ], overall second-order kinetics will be observed
transition state (TS-1, DG < 1 kcalmol ), suggesting that
2
ꢀ1
3
with kobs = k [1] (P ) . On the other hand, if k [PPh ] @ k ,
N O coordination may not be required, and forms a distorted
2
1
0
N2O
0
ꢀ1
3
2
overall first-order kinetics with inverse order in phosphane
concentration will be observed.
butterfly terminal cobalt oxo complex that is best described as
[
22–24]
a cobalt oxyl species (I-3-q, see the SI).
The four-
coordinated intermediate I-3 then coordinates phosphane in
the equatorial (I-4A-q) or axial position (I-4B-q; see the SI).
In the last step, the oxygen is transferred from the metal
center to the substrate through a transition state (TS-2) with
ꢀ
ꢀ1
a small activation barrier (DG = 6 kcalmol ) resulting in
[
25]
a phosphane oxide bound intermediate I-5-d.
Other
k k ½1Š½N OŠ
coordination and activation modes of N O as well as
mechanistic pathways in which the phosphane remains
1
2
2
2
r ¼ k
ð2Þ
½PPh Š þ k
ꢀ
1
3
2
bound to the metal center during N O activation were
2
With no further experimental data available, we turned to
found to be more energetically demanding (see the SI,
Figures S23 and S24 and Tables S6–S8).
[
23]
DFT calculations to shed light on a possible cobalt-based N O
2
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 7
These are not the final page numbers!

Angewandte
Communications
Chemie
Comparison of the spin states showed that the doublet
state was the energetically favored state for I-1, I-2A, I-2C
and I-5 (see Tables S4 and S5). For I-2B, I-4A, TS-1A and TS-
Commun. 2015, 36, 553 – 557; c) L. Gonsalvi, M. Peruzzini,
Angew. Chem. Int. Ed. 2012, 51, 7895; Angew. Chem. 2012, 124,
017.
8
[
[
4] For a review about phosphane oxide derivatives for OLEDs, see:
S. O. Jeon, J. Y. Lee, J. Mater. Chem. 2012, 22, 4233.
5] A. M. Safiulina, A. G. Matveeva, T. K. Dvoryanchikova, O. A.
Sinegribova, A. M. Tu, D. A. Tatarinov, A. A. Kostin, V. F.
Mironov, I. G. Tananaev, Russ. Chem. Bull. Int. Ed. 2012, 61, 392.
[6] R. Curic, F. Di Furia, Tetrahedron 1972, 28, 3905.
[7] a) A. N. Chernysheva, E. K. Beloglazkina, A. A. Moiseeva,
R. L. Antipin, N. V. Zyk, N. S. Zefirov, Mendeleev Commun.
2
A the energy difference between doublet and quartet is
ꢀ1
smaller than 10 kcalmol regardless of the functional used,
making the assignment of a preferred spin state unreliable.
Especially, the preferred spin state of the terminal oxo
intermediates I-3 is highly functional-dependent and a spin-
[26]
flip may occur at this point of the reaction coordinate.
In conclusion, five amino-olefin cobalt complexes were
prepared and fully characterized. These new complexes are
unambiguously zero-valent species with a high metalloradical
character. Several of these complexes can promote the clean
oxidation of phosphanes with N O yielding N as the sole by-
2
012, 22, 70; b) E. K. Beloglazkina, A. G. Majouga, A. A.
Moiseeva, N. V. Zyk, N. S. Zefirov, Mendeleev Commun. 2009,
9, 69; c) A. V. Leontꢂev, L. G. Tomilova, A. O. Fomicheva,
M. V. Proskurina, N. S. Zefirov, Russ. J. Gen. Chem. 2001, 71,
48; d) T. Yamada, K. Suzuki, K. Hashimoto, T. Ikeno, Chem.
1
2
2
6
product. In particular, [Co(trop NH)(PPh )] (1) is an effective
2
3
Lett. 1999, 1043; e) H. Arzoumanian, D. Nuel, J. Sanchez, J. Mol.
Catal. 1991, 65, L9; f) A. Yamamoto, S. Kitazume, L. S. Pu, S.
Ikeda, J. Am. Chem. Soc. 1971, 93, 371.
catalyst for the selective and mild oxidation of a wide range of
secondary and tertiary phosphanes bearing aliphatic, aro-
matic, vinylic, and amido substituents as well as diphosphanes.
None of these substrates can be oxygenated with the same
efficiency and, especially, selectivity using standard meth-
[8] a) A. R. Ravishankara, J. S. Daniel, R. W. Portmann, Science
2
2
009, 326, 123; b) J. Hansen, M. Sato, Proc. Natl. Acad. Sci. USA
004, 101, 16109.
[6]
[9] For selected recent examples see: a) T. D. Palluccio, E. V.
Rybak-Akimova, S. Majumdar, X. Cai, M. Chui, M. Temprado,
J. S. Silvia, A. F. Cozzolino, D. Tofan, A. Velian, C. C. Cummins,
B. Captain, C. D. Hoff, J. Am. Chem. Soc. 2015, 135, 11357;
b) O. P. Lam, S. C. Bart, H. Kameo, F. W. Heinemanna, K.
Meyer, Chem. Commun. 2010, 46, 3137; c) H. Z. Yu, G. C. Jia,
Z. Y. Lin, Organometallics 2009, 28, 1158; d) W. H. Harman,
C. J. Chang, J. Am. Chem. Soc. 2007, 129, 15128.
ods. This is of particular advantage for the oxidation of
pyrophoric and air/moisture sensitive phosphanes. The meth-
odology described is easy, clean and circumvents a compli-
cated work-up protocol. DFT calculations indicate that
a tetrahedral cobalt oxyl intermediate may be involved in
the oxygen transfer reaction. The role of the coordinated
olefins on the stability of the low-valent cobalt complexes
remains to be investigated.
[
10] For selected recent examples see: a) L. E. Doyle, W. E. Piers, J.
Borau-Garcia, J. Am. Chem. Soc. 2015, 137, 2187; b) B. Horn, C.
Limberg, C. Herwig, M. Feist, S. Mebs, Chem. Commun. 2012,
4
4
8, 8243; c) T. M. Figg, T. R. Cundari, Organometallics 2012, 31,
998; d) S. Demir, E. Montalvo, J. W. Ziller, G. Meyer, W. J.
Acknowledgements
Evans, Organometallics 2010, 29, 6608; e) N. D. Harrold, R.
Waterman, G. L. Hillhouse, T. R. Cundari, J. Am. Chem. Soc.
2009, 131, 12872; f) H. Z. Yu, G. C. Jia, Z. Y. Lin, Organo-
metallics 2008, 27, 3825; g) P. Chen, S. I. Gorelsky, S. Ghosh, E. I.
Solomon, Angew. Chem. Int. Ed. 2004, 43, 4132; Angew. Chem.
This work was supported by the Swiss National Science
Foundation (SNF) and the ETH Zꢁrich. T.L.G. was supported
by the ETH Zꢁrich Postdoctoral Fellowship Program, co-
funded by the ETH Zurich-Marie Curie action for people
2004, 116, 4224.
(
FEL-14 15-1). J.R.H. thanks the ARC (FT120100421) for
[
[
11] a) J.-P. F. Cherry, A. R. Johnson, L. M. Baraldo, Y.-C. Tsai, C. C.
Cummins, S. V. Kryatov, E. V. Rybak-Akimova, K. B. Capps,
C. D. Hoff, C. M. Haar, S. P. Nolan, J. Am. Chem. Soc. 2001, 123,
financial support and the Australian National Fabrication
Facility for use of equipment. We thank Prof. Markus Reiher,
Prof. Rinaldo Poli, and Prof. T. Don Tilley for helpful
discussions.
7
271; b) A. R. Johnson, W. M. Davis, C. C. Cummins, S. Serron,
S. P. Nolan, D. G. Musaev, K. Morokuma, J. Am. Chem. Soc.
998, 120, 2071; c) C. E. Laplaza, A. L. Odom, W. M. Davis,
C. C. Cummins, J. D. Protasiewicz, J. Am. Chem. Soc. 1995, 117,
999.
1
Keywords: cobalt · N O activation · nitrous oxide ·
2
4
phosphane oxidation
12] For some examples see: a) D. P. Ivanov, L. V. Pirutko, G. I.
Panov, Appl. Catal. A 2012, 415 – 416, 10; b) A. Ribera,
I. W. C. E. Arends, S. de Vries, J. Pꢃrez-Ramꢄrez, R. A. Sheldon,
J. Catal. 2000, 195, 287; c) J. I. Panov, G. A. Sheveleva, A. S.
Kharitonov, V. N. Romannikov, L. A. Vostrikova, Appl. Catal. A
[
1] For selected examples and review on secondary phosphane
oxides in coordination chemistry see: a) P. Nava, H. Clavier, Y.
Gimbert, L. Giordano, G. Buono, S. Humbel, ChemCatChem
1992, 82, 31; d) B. Ohtani, S. Takamiya, Y. Hirani, M. Sudoh, S.
Nishimoto, T. Kagiya, J. Chem. Soc. Perkin Trans. 2 1992, 175.
13] See the following reviews and references therein: a) K. Severin,
Chem. Soc. Rev. 2015, 44, 6375 – 6386; b) V. R. Landaeta, R. E.
Rodriguez-Lugo, Inorg. Chim. Acta 2015, 431, 21 – 47.
14] T. L. Gianetti, S. P. Annen, G. Santiso-Quinones, M. Reiher, M.
Driess, H. Grꢁtzmacher, Angew. Chem. Int. Ed. 2016, 55, 1854;
Angew. Chem. 2016, 128, 1886.
[15] a) S. Deblon, L. Liesum, J. Harmer, H. Schçnberg, A. Schweiger,
H. Grꢁtzmacher, Chem. Eur. J. 2002, 8, 601; b) J. M. Mevs, W. E.
Geiger, Organometallics 1996, 15, 2350; c) S. A. Locke, P. B.
Shevlin, Organometallics 1984, 3, 217; d) T. Kitamura, T. Joh, J.
[
2015, 7, 3848; b) T. M. Shaikh, C. Weng, F. Hong, Coord. Chem.
Rev. 2012, 256, 771; c) L. Ackermann, Synthesis 2006, 1557.
2] O. Hazut, A. Agarwala, I. Amit, T. Subramani, S. Zaidiner, Y.
Rosenwaks, R. Yerushalmi, ACS Nano 2012, 6, 10311.
3] For a recent examples and highlight concerning bis(acyl)phos-
phane oxide photoinitiators, see: a) J. Wang, G. Siqueira, G.
Mꢁller, D. Rentsch, A. Huch, P. Tingaut, J. Levalois-Grꢁtz-
macher, H. Grꢁtzmacher, Chem. Commun. 2016, 52, 2823; b) G.
Mꢁller, M. Zalibera, G. Gescheidt, A. Rosenthal, G. Santiso-
Quinones, K. Dietliker, H. Grꢁtzmacher, Macromol. Rapid
[
[
[
Angew. Chem. Int. Ed. 2016, 55, 1 – 7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Organomet. Chem. 1974, 65, 235; e) M. Vogt, Diss. ETH Nr.
9295, 2010.
16] S. Stoll, A. Schweiger, J. Magn. Reson. 2006, 178, 42.
17] L. A. Hanlan, H. Huber, E. P. Kundig, B. R. McGarvey, G. A.
Ozin, J. Am. Chem. Soc. 1975, 97, 7054 – 7068.
[23] For examples of low-valent late transition metal oxo species, see
the following reviews and references therein: a) A. Gunay, K. H.
Theopold, Chem. Rev. 2010, 110, 1060; b) C. T. Saouma, J. C.
Peters, Coord. Chem. Rev. 2011, 255, 920; c) K. Ray, F. Heims,
F. F. Pfaff, Eur. J. Inorg. Chem. 2013, 3784.
1
[
[
[
18] a) The catalyst loading could be lowered to 1 mol% but very
slow conversion is observed (entry 7, Table S3); b) Primary
phosphanes react with the cobalt complexes prior addition of
[24] The formation of a terminal oxo species of Group 9 by reaction
with N O is rare but not unprecedented. Caulton et al. recently
2
I
observed in situ the formation of a tetra-coordinated Rh
N O, to form an intractable mixture of products. See also: U.
Winterhalter, L. Zsolnai, P. Kircher, K. Heinze, G. Huttner, Eur.
J. Inorg. Chem. 2001, 89.
terminal oxo from N O, and calculated the mechanism of its
2
2
[
20b]
formation:
a) A. Y. Verat, H. Fan, M. Pink, Y.-S. Chen, K. G.
Caulton, Chem. Eur. J. 2008, 14, 7680; b) J. G. Andino, K. G.
Caulton, J. Am. Chem. Soc. 2011, 133, 12576.
[
[
19] a) H. P. Angstadt, J. Am. Chem. Soc. 1964, 86, 5040; b) T.
Achard, L. Giordano, A. Tenaglia, Y. Gimbert, G. Buono,
Organometallics 2010, 29, 3936; c) F. Dornhaus, S. Scholz, I.
Sꢅnger, M. Bolte, M. Wagner, H. W. Lerner, Z. Anorg. Allg.
Chem. 2009, 635, 2263.
20] After the first oxidation, the bisphosphane monoxide is less s-
donor and more p-acceptor than the bisphosphane precursor.
We believe that such electronic effects control the selectivity of
the transformation toward mono-oxidation, suggesting that
OAT occurs via a nucleophilic attack of the P atom to an O
atom-containing intermediate.
[25] A terminal Pt=O was reported to undergo stoichiometric oxo
transfer to phosphane: E. Poverenov, I. Efremenko, A. I.
Frenkel, Y. Ben-David, L. J. W. Shimon, G. Leitus, L. Konstan-
tinovski, J. M. L. Martin, D. Milstein, Nature 2008, 455, 1093.
[26] An optimization of the energy at the crossing points is beyond
the scope of this paper and not mandatory to explain the
observed reactivity because the MERPs are similar for both spin
states (see the SI). For a discussion of spin-crossovers in catalytic
system, see a) R. Poli, J. N. Harvey, Chem. Soc. Rev. 2003, 32, 1;
b) R. Poli, Comprehensive Inorganic Chemistry II, Vol. 9, 2nd ed.
(Eds.: J. Reedijk, K. Poeppelmeier), Elsevier, Amsterdam, 2013,
pp. 481 – 500.
[
[
21] None of the expected side products (anthracene, triphenyl
methane or triphenyl methanol) were detected by neither NMR
spectroscopy nor GC/MS analysis of the catalytic reaction.
22] Attempts to trap the cobalt oxo/oxyl intermediate using Lewis
acid (BPh , Sc(OTf) ), radical species (trityl radical), or via
Received: September 19, 2016
3
3
[2+2] reaction (styrenes, norbornene) were unsuccessful.
Published online: && &&, &&&&
6
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 7
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
N O Activation
2
T. L. Gianetti,* R. E. Rodrꢁguez-Lugo,
J. R. Harmer, M. Trincado, M. Vogt,
G. Santiso-Quinones,
H. Grꢂtzmacher*
&&&& — &&&&
Zero-Valent Amino-Olefin Cobalt
Complexes as Catalysts for Oxygen Atom
Transfer Reactions from Nitrous Oxide
Utilizing greenhouse gas: A series of zero
valent amino olefin cobalt complexes is
used to convert phosphanes to phos-
phane oxides with N O. N is the only
tion proved to be efficient and selective
for a large variety of substrates including
highly sensitive phosphanes such as
secondary, vinylic or diphosphanes with
PꢀP bonds.
2
2
side product. The catalytic transforma-
Angew. Chem. Int. Ed. 2016, 55, 1 – 7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7
These are not the final page numbers!