Formal loss of an H radical by a cobalt complex via metal-ligand cooperation

Article abstract of DOI:10.1039/c3cc39049g

A (PNP)Co(i)methyl diamagnetic complex formally loses an H atom from the pincer ligand, exhibiting a long-range metal-ligand cooperation in what may be considered as an unusual example of 'C-H cleavage'. Spectroscopic data indicate that the product is a neutral Co(i) complex with a radical delocalized in the ligand backbone. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3cc39049g

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Formal loss of an H radical by a cobalt complex via
metal–ligand cooperation†
Eugene Khaskin,a Yael Diskin-Posner,b Lev Weiner,b Gregory Leitusb and
David Milstein*a
Cite this: Chem. Commun., 2013,
49, 2771
Received 18th December 2012,
Accepted 15th February 2013
DOI: 10.1039/c3cc39049g
www.rsc.org/chemcomm
A (PNP)Co(I)methyl diamagnetic complex formally loses an H atom from
the pincer ligand, exhibiting a long-range metal–ligand cooperation in
what may be considered as an unusual example of ‘C–H cleavage’.
Spectroscopic data indicate that the product is a neutral Co(I) complex
with a radical delocalized in the ligand backbone.
Fig. 1 Ligand platforms relevant to the current study.
Pincer cobalt complexes have been extensively studied due to their
unique catalytic and electronic properties. Brookhart1 and Gibson2 CO2, as suggested by a recent DFT study,10 and we planned to
first noted the high catalytic activities of imine NNN pincer com- prepare such compounds and explore their properties.
plexes of iron and cobalt in ethylene polymerization. For the Co(II)
Reports by the groups of Fryzuk and Danopoulos,6a,8 using an
dichloride catalysts, it was later suggested that a Co(I) intermediate anionic PNP ligand to generate Co(II) alkyl complexes, as well as a
generated in situ was responsible for the polymerization activity.3 report by the groups of Erker and Bazan,11 suggested that Co(I)
The isolation of square planar halide and alkyl Co(I) intermediates alkyls might be generated from LCo(II)Cl2 by the addition of excess
engendered a study of their unique electronic properties by Grignard or alkyl lithium reagent. Indeed, reaction of the Co(II)
Budzelaar.4 This ligand set is similar to that used by Chirik’s group complex 1, obtained by reaction of CoCl2 with the iPr-PNP ligand
for cobalt complexes for polymerization and structural studies, often (see ESI†), with an excess of MeLi in diethylether generated the
differing only in the nature of the imine substituent.5
diamagnetic Co(I)Me species 2 with no deprotonation of the
PSiNSiP pincer cobalt complexes were extensively studied by benzylic protons being observed. The complex was isolated as a
1
Fryzuk and Caulton with the anionic ligand resulting in Co(II) dark brown solid and characterized by H NMR (Scheme 1).
species when generating mono-alkyl complexes.6 The three coordi-
Complex 2 was not stable in C6D6, transforming into dark brown
nate (PNP)Co(I) synthon reported by Caulton is a high-spin triplet.6b,c paramagnetic complex 3 (Scheme 1) whose 1H NMR exhibited a few
The POCOP ligand, reported to be stabilizing a Co dihydride broad peaks in the 40–60 ppm region, as well as two large, very broad
complex by Heinekey, presents a very similar electronic environment peaks from 12–16 ppm, probably ligand CH3 groups, and signals in
to the metal.7 Danoupolos reported a neutral Co(I) carbene pincer the upfield regions whose number is consistent with an unsymme-
complex without noting unusual reactivity.8 The aforementioned trical complex. Pure crystals of 3 were obtained from a pentane solu-
ligand systems are summarized in Fig. 1.
tion of 2. An X-ray diffraction study, in which all hydrogen atoms were
Curiously, neutral pincer phosphine donor ligands, such as the located, unambiguously shows a square planar, asymmetric complex
PNP platform used in this report, are not widely reported in cobalt that is missing one hydrogen atom on one of the arms (Fig. 2).
chemistry beyond studies on simple Co dihalide adducts.9 However,
The structure of 3 exhibits a short C6–C7 bond (1.415(4) Å) of
neutral Co(I) alkyls could have a role as catalysts in hydrogenation of the pyridine arm with only one hydrogen atom, which is typical
of a double bond. This suggests conjugation of the ring, which
a Department of Organic Chemistry, Weizmann Institute of Science, Rehovot, Israel.
E-mail: david.milstein@weizmann.ac.il; Fax: +972-8-934-4142;
Tel: +972-8-934-2599
b Department of Chemical Research Support, Weizmann Institute of Science,
Rehovot, Israel
† Electronic supplementary information (ESI) available: Synthesis of all com-
plexes, experimental details and DFT study. CCDC 916115–916117. For ESI and
crystallographic data in CIF or other electronic format see DOI: 10.1039/
Scheme 1 Synthesis of complexes 2–3 and 5–6.
c3cc39049g
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 2771--2773 2771

View Article Online
Communication
ChemComm
Scheme 3 Synthesis of 7 by 1eÀ oxidation of 5.
overnight reaction. Exposing the sample to air regenerated
the strong TEMPO triplet by presumably forming a hydroperoxy
radical (ESI† Fig. S20; see Scheme 2).12
Fig. 2 Selected bond lengths (Å) Co1-C20, 1.975(3); Co1-N1, 1.949(3); Co1-P1,
2.1732(9); Co1-P2, 2.1912(9); C1-C2, 1.499(4); C7-C6, 1.415(3). Selected angles
(degrees) P1-Co-1P2, 169.55(4); N1-Co1-C20, 179.27(12). One of the two inde-
pendent molecules in an asymmetric unit. ORTEP diagram of 3 (50% ellipsoid
probability level).
The tBu-PNP analogue of complex 1 was prepared by reaction of
t
the Bu-PNP ligand with CoCl2, resulting in the purple, crystalline
complex (tBu-PNP)CoCl2 4. Adding an excess of MeLi to 4 in benzene
or ether and stirring for several hours resulted in formation of the
diamagnetic, dark black complex 5 whose NMR characteristics
mirrored those of 2. Complex 5 was stable in the solid state or in
benzene, but it exhibited a slow interconversion to 6 in THF, and a
rapid interconversion to 6 in pentane (see Scheme 1). Unlike
complex 3, good-quality crystals of 6 could not be isolated. The
paramagnetic 1H NMR of 6 was closely analogous to that of 3, with
signals in the same areas and one signal missing in the far down-
field region, entirely consistent with a tBu for iPr substitution.
Reaction of complex 5 with an excess of MeI was slightly
different from that observed for related Co species.4d,5a Instead of
oxidative addition, a radical process resulting in oxidation of the
metal center to give the Co(II) complex 7a took place (Scheme 3 and
Fig. 3). The reaction is instantaneous at room temperature, resulting
in rapid color change from dark black to green. X-ray structural
determination of crystals obtained from a benzene solution exhibits
a square planar geometry with a non-coordinating iodide anion. We
believe that electron donation from 5 to MeI leads to the radical
remains planar and aromatic judging from the short bond lengths.
The plane made by C6–C7–P2 and the C7 hydrogen atom indicates
sp2 conjugation of the central carbon. The other arm with two
hydrogens has a C2–C1 bond length of 1.499 Å, considerably longer
than C6–C7. Evans magnetic measurements of 3 at several different
concentrations in C6D6 gave an average value of meff of B2.2. The
EPR signal of complex 3 by itself is reminiscent of an organic
radical’s spectrum that is visible at room temperature (g = 2.0046)
(see ESI† Fig. S20). The structural and spectroscopic evidence
suggests one unpaired electron for the complex that is mostly
delocalized on the ligand, although the nitrogen triplet remains
unresolved. The only structural difference between 2 and 3 is one
hydrogen atom (both are neutral species that are very soluble in non-
polar organic solvents). To the best of our knowledge, radical
formation at ambient temperatures using an organometallic species,
involving bond scission at a site remote from the metal center, has
not been reported before. Importantly, the yield of complex 3 was
above 80%, ruling out a disproportionation pathway.
anion [MeI]À followed by iodide expulsion and generation of a Me
ꢀ
radical. The analogous green complex 7b was obtained by electron
transfer to ferrocenium borate and was also characterized using
X-ray spectroscopy. Both 7b and 7a have exactly the same geometry
with fully protonated ligand arms. If cationic 7 is assumed to be
isostructural with complexes 2 and 5, comparing bond lengths
between the structure of 7 and 3 allows for further conclusions on
the radical nature of 3. As mentioned above, the aromaticity in the
central pyridine ring is maintained, however the bond lengths are
slightly elongated in 3 compared to complexes 7, consistent with
delocalization of an extra unpaired electron in an anti-bonding
orbital and suggesting that after the loss of an H atom, the lone
electron density is not confined to the ligand arm.
Intriguingly, complex 2 transforms into 3 also in the solid state,
albeit at a slower rate than in solution. When samples of solid 3
were intermittently measured using 1H NMR in C6D6 over a period
of several days, a steady transformation to 3 was observed. By the
end of three days, mostly 3 was present, and only a small amount of
2 remained. An experiment in the dark showed no qualitative
difference in the rate of the transformation.
The transformation of 2 to 3 was conveniently followed by
EPR in the presence of a TEMPO (2,2,6,6-tetramethylpiperidine-
1-oxy) radical using a 10-fold molar excess solution of 2 in THF
(Scheme 2). In the beginning a strong TEMPO signal was clearly
seen, and it slowly degraded over time as confirmed by a series
of five-minute measurements. Eventually, the signal was low
enough to allow observation of the signal of complex 3 as well,
and the TEMPO triplet completely disappeared after an
We turned our attention to determining the fate of the H atom
that is lost from the ligand arm. Other metal complexes, such as
Fig. 3 Selected bond lengths (Å) Co1-C24, 2.008(3); Co1-N1, 2.008(2); Co1-P1,
2.2274(8); Co1-P2, 2.2349(7); C1-C6, 1.503(4); C7-C5, 1.499(4). Selected angles
(degrees) P1-Co1-P2, 169.52(3); C24-Co1-N1, 178.59(11). ORTEP diagram of 7a
(50% ellipsoid probability level).
Scheme 2 (a) Transformation of TEMPO to an EPR-silent compound. (b) Regenera-
tion of TEMPO with oxygen.
c
2772 Chem. Commun., 2013, 49, 2771--2773
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
n-Bu3SnH, are known to be potent sources of H radicals. However,
The neutral complexes formed in the current report via
homolytic C–H cleavage to give an H radical at room temperature is homolytic cleavage of an arm CH bond can be compared to
unlikely (see the DFT section in ESI†). H2 was not detected by the heterolytic cleavage of pincer complexes of Ru and Rh
1H NMR when carrying out the syntheses of 3 and 6 in closed tubes, reported earlier by our group, where full dearomatization of
or by GC/TCD. In addition, the known H2 scavenger (PEt3)3IrCl13 the central pyridine ring in favor of a 6p conjugated system was
remained unchanged during the transformation, even when present observed.15 Crystal structure evidence points to an intermediate
in the same reaction mixture. Complexes 2 and 5 also decompose case for complex 3, where aromatization of the pyridine ring is
rapidly; 3 and 6 slowly, upon addition of H2 to undetermined still evident, but the unpaired electron is delocalized on the
paramagnetic species. Thus, the presence of H2 during a solution arm and throughout the ring.
reaction is unlikely. The products are obtained in high yield (>80%
This research was supported by the European Research
for complex 3), ruling out a disproportionation pathway but not an Council under the FP7 framework (ERC No. 246837), the
intermolecular one even in the solid state, both however being Minerva Foundation, and the Kimmel Center for Molecular
disfavored by sterics. In addition, the dramatically increased reaction Design. D.M. is the Israel Matz Professor of Organic Chemistry.
rate in aliphatic solvents argues for a significant solvent role in the
abstraction of the putative H radical.
It was of interest to abstract the (formally) H radical from the
complex by use of an appropriate substrate. The reaction with
Notes and references
1 B. L. Small, M. Brookhart and A. M. A. Bennett, J. Am. Chem. Soc.,
1998, 120, 4049–4050.
olefins and aliphatic alkynes was unsuccessful. However, when
diphenylacetylene (dpa) was added to samples of 2 and 5 in THFd8
or dioxaned8, stilbene products were obtained. Upon a slow, room
temperature transformation of 5 in the presence of an equivalent
amount of dpa, B50% conversion of dpa took place to yield a 1 : 1
mixture of cis-stilbene and trans-stilbene, in addition to a small
amount of a-methyl-stilbene4b (see ESI†). Half the dpa remained
unreacted after two days, with no further change after one week. The
presence of cis-stilbene may indicate a metal-complex mediated role
in H radical transfer, perhaps via coordination of a radical organic
intermediate. The dpa - stilbene experiment serves to confirm the
transformation and its stoichiometry (Scheme 4). In all reactions, the
stilbene products were not deuterated; hence the origin of H atoms
was the metal complex and not the solvent.
2 (a) G. J. P. Britovsek, V. C. Gibson, B. S. Kimberley, P. J. Maddox,
S. J. McTavish, G. A. Solan, A. J. P. White and D. J. Williams, Chem.
Commun., 1998, 849–850; (b) G. J. P. Britovsek, M. Bruce, V. C. Gibson,
B. S. Kimberley, P. J. Maddox, S. Mastoianni, S. J. McTavish, C. Redshaw,
G. A. Solan, S. Stroemberg, A. J. P. White and D. J. Williams, J. Am. Chem.
Soc., 1999, 121, 8728–8740; (c) V. C. Gibson and G. A. Solan, Top.
Organomet. Chem. ‘‘Iron-Based and Cobalt-Based Olefin Polymerisation
Catalysts’’, Springer-Verlag Berlin, Heidelberg, 2008.
3 (a) T. M. Kooistra, Q. Knijnenburg, J. M. M. Smits, A. D. Horton,
P. H. M. Budzelaar and A. W. Gal, Angew. Chem., Int. Ed., 2001, 40,
4719–4722; (b) K. P. Tellman, M. J. Humphries, H. S. Rzepa and
V. C. Gibson, Organometallics, 2004, 23, 5503–5513.
4 (a) Q. Knijnenburg, D. Hetterscheid, T. M. Kooistra and P. H. M.
Budzelaar, Eur. J. Inorg. Chem., 2004, 1204–1211; (b) Q. Knijnenburg,
S. Gambarotta and P. H. M. Budzelaar, Dalton Trans., 2006, 5442–5448;
(c) D. Zhu and P. H. M. Budzelaar, Organometallics, 2008, 27, 2699–2705;
(d) D. Zhu, I. Thapa, I. Korobkov, S. Gambarotta and P. H. M. Budzelaar,
Inorg. Chem., 2011, 50, 9879–9887; (e) D. Zhu, I. Korobkov and P. H. M.
Budzelaar, Organometallics, 2012, 31, 3958–3971.
In conclusion, we observed C–H bond cleavage at room tem-
perature at a site remote from the metal, mediated by a cobalt
center. This transformation should be considered in the chemistry
of cobalt based enzymes and complexes, where a remotely formed
radical can play a role in substrate transformations. A recent report
on a ytterbium complex that loses H atoms remotely shows that in
contrast, that reaction involves an intermediate dimerization and
the production of H2 with the organometallic product having a
metal centered radical.14
5 (a) A. C. Bowman, C. Milsmann, C. C. H. Atienza, E. Lobkovsky,
K. Wieghardt and P. J. Chirik, J. Am. Chem. Soc., 2010, 132, 1676–1684;
(b) C. C. H. Atienza, A. C. Bowman, E. Lobkovsky and P. J. Chirik, J. Am.
Chem. Soc., 2010, 132, 16343–16345; (c) A. C. Bowman, C. Milsmann,
E. Bill, E. Lobkovsky, T. Weyhermueller, K. Wieghardt and P. J. Chirik,
Inorg. Chem., 2010, 49, 6110–6123; (d) S. Monfette, Z. R. Turner,
S. P. Semproni and P. J. Chirik, J. Am. Chem. Soc., 2012, 134, 4561–4564.
6 (a) M. D. Fryzuk, D. B. Leznoff, R. C. Thompson and S. J. Rettig, J. Am.
Chem. Soc., 1998, 120, 10126–10135; (b) M. J. Ingleson, M. Pink,
H. Fan and K. G. Caulton, Inorg. Chem., 2007, 46, 10321–10334;
(c) M. J. Ingleson, M. Pink and K. G. Caulton, J. Am. Chem. Soc.,
2006, 128, 4248–4254; (d) M. J. Ingleson, M. Pink, H. Fan and
K. G. Caulton, J. Am. Chem. Soc., 2008, 130, 4262–4276.
7 T. J. Hebden, A. J. St. John, D. G. Gusev, W. Kaminsky, K. I. Goldberg
and D. M. Heinekey, Angew. Chem., Int. Ed., 2011, 50, 1873–1876.
8 A. A. Danopoulos, J. A. Wright, W. B. Motherwell and S. Ellwood,
Organometallics, 2004, 23, 4807–4810.
9 For examples see: (a) Q. Dong, M. J. Rose, W.-Y. Wong and
H. B. Gray, Inorg. Chem., 2011, 50, 10213–10224; (b) K. Abdur-
Rashid, G. Todd, C.-W. Tsang, C. Xuanhua, G. Rognwei, J. Wenli,
A. Dino and C. Sui-Seng, WO2008141439, 2008.
10 X. Yang, ACS Catal., 2011, 1, 849–854.
11 N. Kleigrewe, W. Steffen, T. Bloemker, G. Kehr, R. Froehlich,
B. Wibbeling, G. Erker, J.-C. Wasilke, G. Wu and G. C. Bazan,
J. Am. Chem. Soc., 2005, 127, 13955–13968.
12 G. I. Shchukin and I. A. Grigor’ev, in Imidazoline Nitroxides, ed.
L. B. Volodarsky, CRC Press, Boca Raton, FL, vol. I, Ch. 6, 1988.
13 O. Blum and D. Milstein, J. Am. Chem. Soc., 1995, 117, 4582–4594.
14 G. Nocton, C. H. Booth, L. Maron and R. A. Andersen, Organome-
tallics, 2013, ASAP.
15 For some examples see: (a) E. Khaskin, M. A. Iron, L. J. W. Shimon,
J. Zhang and D. Milstein, J. Am. Chem. Soc., 2010, 132, 8542–8543;
(b) M. Feller, Y. Diskin-Posner, L. J. W. Shimon, E. Ben-Ari and
D. Milstein, Organometallics, 2012, 31, 4083–4101.
Scheme 4 Capture of an H radical by dpa.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 2771--2773 2773