Halogenolysis of a nickelalactone complex produces β-halo-anhydrides

Article abstract of DOI:10.1002/ejic.201402952

Nickelalactone complex [(dppe)Ni{κ²-C,O-CH₂CH₂C(=O)O}] {dppe = 1,2-bis(diphenylphosphino)ethane} reacts with halogens to form 3-halo-propionic anhydrides, [(dppe)NiX₂], and [(dppeO₂)₃Ni][NiX₄] (X = Cl, Br, I). Studies of model complexes [(dppe)Ni(O₂CtBu)₂] and [(dppe)NiBr(O₂CtBu)] suggest that oxidation to Ni^III and P-O reductive elimination are key steps in this reaction.

Full text of DOI:10.1002/ejic.201402952

SHORT COMMUNICATION
DOI:10.1002/ejic.201402952
Halogenolysis of a Nickelalactone Complex Produces
β-Halo-Anhydrides
Ryan A. Zarkesh,[a][‡] Michael D. Hopkins,[a] and
Richard F. Jordan*[a]
Dedicated to the memory of Professor Gregory L. Hillhouse
Keywords: Nickel / Lactones / Anhydrides / Reductive elimination / Redox chemistry
Nickelalactone complex [(dppe)Ni{κ2-C,O-CH2CH2C(=O)O}]
{dppe = 1,2-bis(diphenylphosphino)ethane} reacts with hal-
ogens to form 3-halo-propionic anhydrides, [(dppe)NiX2],
and [(dppeO2)3Ni][NiX4] (X = Cl, Br, I). Studies of model
complexes [(dppe)Ni(O2CtBu)2] and [(dppe)NiBr(O2CtBu)]
suggest that oxidation to NiIII and P–O reductive elimination
are key steps in this reaction.
organic products derived from alkene/CO2 coupling,
though little is known about such reactions. The reaction
of nickelalactones with amine N-oxides or O2 followed by
acid hydrolysis yields β-hydroxy carboxylic acids.[1d,6a,9a]
Similarly, reaction with FeCl3 and acidic workup gives β-
chloro carboxylic acids.[9a] Here we report that prototypical
nickelalactone complex [(dppe)Ni{κ2-CH2CH2C(=O)O}]
{1, dppe = 1,2-bis(diphenylphosphino)ethane}[3b] reacts
with halogens to yield β-halo-propionic anhydrides. Model
reactions suggest that oxidatively induced P–O reductive eli-
mination is a key step in this process.
Introduction
Nickelalactone complexes, that is, cyclic nickel carboxyl-
ates of general structure [LnNi{κ2-C,O-CR2CR2C(=O)O}],
are formed by LnNi-mediated coupling of alkenes and
[1]
CO2 or other methods,[2–4] and they have long been of
interest for the conversion of CO2 to higher value prod-
ucts.[5] Nickelalactones can be cleaved under acidic condi-
tions to yield carboxylic acids,[1,2] or they can be elaborated
by insertion,[1c,1d,2a,6] C–C coupling with alkyl halides, or
other reactions,[2c,2d,3a,6a,7] and then hydrolyzed to yield
more complex products. Nickelalactones also undergo
transmetalation with ZnR2 reagents to yield C–C coupling
products,[8] Lewis acid-induced β-H elimination to form
ring-contracted nickelalactones or alkenes,[1d,9] and CO
insertion and reductive elimination to give an-
hydrides.[1a,1b,4,6b] Some nickelalactones react with
MeI to produce methyl acrylate.[10] Interestingly,
Results and Discussion
The reaction of 1 with 1.5 equiv. of Br2 in CH2Cl2 pro-
duces 3-bromo-propionic anhydride and [(dppe)NiBr2] in
high yield (Scheme 1 and Table 1).[13,14]
[(tBu2PCH2CH2PtBu2)Ni{κ2-CH2CH2C(=O)O}]
reacts
with NaOtBu to form sodium acrylate, enabling the Ni-cat-
alyzed (though step-wise) conversion of ethylene and CO2
to this commodity chemical.[11]
Hillhouse and co-workers showed that oxidation of
nickel oxa- and azametallacycles induces C,O- and C,N-re-
ductive elimination to give cyclic products.[12] Similarly, oxi-
dation of nickelalactones may provide a means of releasing
Scheme 1. Reaction of [(dppe)Ni{κ2-C,O-CH2CH2C(=O)O}] (1)
with halogens.
[a] Department of Chemistry
The University of Chicago
Table 1. Isolated yields of products in Scheme 1.
5735 South Ellis Ave., Chicago, IL 60637, USA
E-mail: rfjordan@uchicago.edu
Halogen [(dppe)NiX2] β-Halo-anhydride [(dppeO2)3Ni][NiX4]
http://jordan-group.uchicago.edu/
[‡] Current address: Sandia National Laboratories
P. O. Box 969 MS 9292, Livermore, CA 94551, USA
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201402952.
Cl2
Br2
I2
100%
100%
100%
86%
83%
30%
75%
40%
13%
Eur. J. Inorg. Chem. 2014, 5491–5494
5491
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
SHORT COMMUNICATION
A
minor product, [(dppeO2)3Ni][NiBr4],[15] is also
formed and was identified by IR spectroscopy and reaction
with dppe, which yields dppeO2 and [(dppe)2Ni]Br2 [Equa-
tion (1)].
(1)
Similarly, the reaction of 1 with Cl2 yields 3-chloro-pro-
pionic anhydride, [(dppe)NiCl2], and [(dppeO2)3Ni][NiCl4].
Compound 1 reacts with I2 in the same manner, but the
yield of 3-iodo-propionic anhydride is low. The stoichio-
metry in Scheme 1 assumes that the CH2CH2CO2 unit of 1
is fully converted into the anhydride and the remaining oxy-
Figure 1. Molecular structure of [(dppe)Ni(O2CtBu)2] (2).
Hydrogen atoms and solvent molecules are omitted. Bond lengths
[Å] and angles [°]: Ni1–P1 2.1150(10), Ni1–P2 2.1175(10), C1–O1
1.273(3), C1–O2 1.231(3), C6–O3 1.281(3), C6–O4 1.227(3); O1–
Ni1–O3 91.69(9), P1–Ni1–P2 87.22(5), P2–Ni1–O3 92.48(7), P1–
Ni1–O1 88.56(7).
gen is incorporated in [(dppeO2)3Ni]2+
.
The use of more than 1.5 equiv. of Br2 in Scheme 1 does
not affect the yields significantly but broadens the NMR
signals of [(dppe)NiBr2] in the product mixture, probably
because of the formation of [(dppe)NiBr3].[16] The use of
less than 1.5 equiv. of Br2 produces a mixture of 3-bromo-
propionic anhydride, [(dppe)NiBr2], and unreacted 1, but
the anhydride reacts further with 1 to give unidentified or-
ganic products.[17]
The incorporation of the halogens in the anhydride prod-
uct suggests that halogenolysis of the Ni–C bond of 1 oc-
curs in Scheme 1 to generate a [(dppe)NiX(O2CH2CH2X)]
species. Therefore, we investigated model species [(dppe)-
Ni(O2CR)2] and [(dppe)NiX(O2CR)]. The reaction of
[(dppe)NiBr2] with 2 equiv. of Tl(O2CtBu) in CH2Cl2 gives
[(dppe)Ni(O2CtBu)2] (2) [Equation (2)].
Attempts to prepare [(dppe)NiBr(O2CtBu)] by reaction
of [(dppe)NiBr2] with 1 equiv. of Tl(O2CtBu) in CH2Cl2 af-
forded mixtures of [(dppe)NiBr(O2CtBu)] (2), [(dppe)-
NiBr2], and [(dppe)2Ni]2+ (Scheme 2). The 31P{1H} NMR
spectrum of this mixture at 203 K contains resonances for
2 (s, δ = 51.2 ppm), [(dppe)NiBr2] (s, δ = 68.3 ppm), and
[(dppe)2Ni]2+ (s, δ = 49.5 ppm). The spectrum also contains
two equal-intensity doublets (δ = 55.1, 60.1 ppm; JPP
=
72 Hz) at chemical shifts that are intermediate between
those of 2 and [(dppe)NiBr2], which are assigned to
[(dppe)NiBr(O2CtBu)]. Importantly, the same mixture of
{(dppe)Ni} species is generated by reaction of [(dppe)-
NiBr2] and 2 (Scheme 2). These results show that [(dppe)-
NiBr(O2CtBu)] readily undergoes ligand redistribution.
(2)
In the solid state, 2 adopts a square-planar structure with
essentially κ1-pivalate ligands (Figure 1). The Ni1–O1
[1.892(2) Å] and Ni1–O3 [1.900(2) Å] distances are similar
to those in [{1,2-(PCy2)2-ethane}Ni(naphthyl)(O2CtBu)]
[1.919(4), 1.909(6) Å, respectively].[18] The Ni1–O2
[2.696(3) Å] and Ni1–O4 [2.685(2) Å] distances are much
longer [though still within the sum of the van der Waals
radii (3.15 Å)], which is consistent with only weak interac-
tions.
Scheme 2. Ligand redistribution of [(dppe)NiBr(O2CtBu)].
The H and 31P{1H} NMR spectra of 2 in CD2Cl2 at
The counterion of [(dppe)2Ni]2+ in Scheme 2 has not
1
203 K are sharp, which is consistent with a diamagnetic, been identified. The results of ESI-MS analysis shows
1
square-planar structure. However, the H NMR spectrum signals for [Ni(O2CtBu)Br2]–, [Ni(O2CtBu)2Br]–, [Ni-
broadens and shifts above 243 K, and at 298 K the 31P{1H} (O2CtBu)3]–, and [Ni2(O2CtBu)5]–. For comparison, the
spectrum is broadened into the baseline and a magnetic mo- ESI-MS spectra of [NiX4]2– (X = Cl, Br, I) salts contain
–
2–
ment is observed (μeff = 2.30 BM). These NMR spectral signals for NiX3 rather than NiX4 ions. Therefore,
changes are reversed when the temperature is lowered. [(dppe)2Ni][Ni(O2CtBu)yBr4–y], [(dppe)2Ni][Ni(O2CtBu)y-
These results indicate that 2 is in equilibrium with a para- Br3–y]Br, or [(dppe)2Ni][Ni(O2CtBu)yBr3–y][O2CtBu] are
magnetic form.
reasonable formulations for the [(dppe)2Ni]2+ product.
Eur. J. Inorg. Chem. 2014, 5491–5494
5492
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
SHORT COMMUNICATION
Bouwman, Mul, and co-workers reported that the ESI-MS ports of phosphine oxidation by Pd(OAc)2 and Ni-
spectrum of the [(o-MeO-dppe)2Ni]2+ salt formed by the (OAc)2,[19,20] suggest the mechanism in Scheme 4.
reaction of o-MeO-dppe and Ni(OAc)2·(H2O)4 contains sig-
nals for [Ni(OAc)3]–, [Ni2(OAc)5]–, and AcO–, and the prod-
uct was formulated as [(o-MeO-dppe)2Ni][Ni(OAc)3]-
[OAc].[19]
The observation that 1.5 equiv. of Br2 per equiv. of 1 is
required for full consumption of 1 in Scheme 1 suggests
that, in addition to Ni–C bond halogenolysis, oxidation to
NiIII is important for product formation. Therefore, we in-
vestigated the reactions of 2 and [(dppe)NiBr(O2CtBu)]
with Br2. Compound 2 reacts with 1 equiv. of Br2 in
CH2Cl2 to form pivalic anhydride, [(dppe)NiBr2], and
[(dppeO2)3Ni][NiBr4] (Scheme 3). Also, [(dppe)NiBr-
(O2CtBu)], that is, the mixture of [(dppe)NiBr(O2CtBu)]
(2), [(dppe)NiBr2], and [(dppe)2Ni][Ni(O2CtBu)y(Br)4–y
]
generated in Scheme 2, reacts with 0.5 equiv. of Br2 (vs. to-
tal Ni) to give the same products (Scheme 3). Finally, these
same products are formed by the reaction of the NiIII com-
plex [(dppe)NiBr3] with Na(O2CtBu). The product distribu-
tions from all three of these reactions parallel that in
Scheme 1. These results show that oxidation of the NiII spe-
cies is required for anhydride formation.
Scheme 4. Mechanism for anhydride formation.
In Scheme 4, Ni–C bond halogenolysis generates [(dppe)-
NiX(O2CCH2CH2X)] (A), which may undergo ligand redis-
tribution. Intermediate A or a ligand redistribution product
reacts with X2 to give NiIII species B, which undergoes P,O-
reductive elimination to afford acyloxyphosphonium NiI
species C. Nucleophilic attack at the acyloxyphosphonium
carbon of C by carboxylate furnishes the anhydride and
[(dppeO)NiIXY]– species D. Oxidation of D to NiII and
subsequent ligand exchange, oxidation, and P,O-reductive
elimination steps complete the reaction, though the obser-
vations reported here do not provide insight into the details
of these steps.
Scheme 3. Reactivity of {(dppe)Ni} pivalate complexes.
The fact that [(dppeO2)3Ni]2+ is the only detectable dppe
The formation of [(dppeO2)3Ni]2+ in Scheme 1 and oxidation product (no dppeO species are observed) likely
Scheme 3 suggests that P–O reductive elimination is a key reflects the lability of the dppeO ligand in D or in species
step in these reactions. Amatore, Jutand, and co-workers derived therefrom. Dissociation of the phosphine oxide
showed that (PPh3)2Pd(OAc)2 reacts under mild conditions would give a {(κ1-P-dppeO)Ni} species, which, because of
to yield Ph3P=O and Pd0 species.[20] This reaction is the absence of chelate-imposed constraints, may be more
thought to proceed by P,O-reductive elimination to form susceptible to P,O-reductive elimination. The selective for-
the acyloxyphosphonium species PPh3(OAc)+, which is hy- mation of [(dppeO2)3Ni][NiX4] is likely driven by the insol-
drolyzed to Ph3P=O and AcOH or reacts with OAc– to ubility of this salt.
form Ph3P=O and (AcO)2O;[21] the latter species is hy-
drolyzed to AcOH. Similarly, [Ni(OAc)2(H2O)4] reacts with volves direct reaction of free or κ1-coordinated dppe with
dppe in MeOH/CH2Cl2 at 50 °C to yield dppeO2.[19]
X2 and carboxylate. Control experiments show that dppe
An alternative mechanism that cannot be ruled out in-
While the mechanism of Scheme 1 is clearly complex, the reacts with Br2 or I2 and NaO2CtBu to give high yields of
observations that (1) C–X bonds are formed, (2) 1.5 equiv. pivalic anhydride, dppeO2, and NaBr or NaI. However, the
of Br2 is required for full reaction of 1, (3) [(dppe)- facts that (1) no free dppe or κ1-coordinared dppe species
NiBr(O2CtBu)] undergoes ligand redistribution reactions are observed in solutions of 1, 2, or [(dppe)NiBr(O2CtBu)],
–
that may involve tBuCO2 dissociation, and (4) oxidation and (2) only low yields of 3-iodo-propionic anhydride and
to NiIII is required for formation of anhydride and [(dppeO2)3Ni][NiI4] are formed in the reaction of 1 with I2,
[(dppeO2)3Ni]2+ in Scheme 3, along with the previous re- argue against this mechanism.
Eur. J. Inorg. Chem. 2014, 5491–5494
5493
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
SHORT COMMUNICATION
[4] T. Yamamoto, K. Igarashi, S. Komiya, A. Yamamoto, J. Am.
Chem. Soc. 1980, 102, 7448–7456.
Conclusions
[5] M. Cokoja, C. Bruckmeier, B. Rieger, W. A. Herrmann, F. E.
Kühn, Angew. Chem. Int. Ed. 2011, 50, 8510–8537; Angew.
Chem. 2011, 123, 8662.
The results presented show that halogenolysis of nickela-
lactones provides a means of synthetically elaborating or-
ganic fragments derived from alkene/CO2 coupling and trig- [6] a) A. M. Castaño, A. M. Echavarren, Organometallics 1994,
gering their release from the Ni center. These reactions pro-
ceed by halogen attack on the Ni–C bond to install Cl, Br,
13, 2262–2268; b) A. M. Echavarren, A. M. Castaño, Tetrahe-
dron 1995, 51, 2369–2378; c) A. M. Echavarren, A. M. Cas-
taño, Adv. Met.-Org. Chem. 1988, 6, 1–47.
or I groups at the C terminus of the fragment. Subsequent
[7] a) R. Fischer, B. Schönecker, D. Walther, Synthesis 1993, 1267–
oxidation of the nickel center induces reductive elimination,
providing the anhydride products. Whether this latter step
can be generalized to outer-sphere processes is a key ques-
tion in the development of catalytic systems involving
nickelalactones and is under investigation.
1270; b) J. Langer, M. Gärtner, H. Görls, D. Walther, Synthesis
2006, 2697–2706.
[8] E. M. O’Brien, E. A. Bercot, T. Rovis, J. Am. Chem. Soc. 2003,
125, 10498–10499.
[9] a) H. Hoberg, A. Ballesteros, J. Organomet. Chem. 1991, 411,
C11–C18; b) J. Dong, T. J. Schmeier, P. G. Williard, N. Hazari,
W. H. Bernskoetter, Organometallics 2013, 32, 2152–2159; c) D.
Jin, P. G. Williard, N. Hazari, W. H. Bernskoetter, Chem. Eur.
J. 2014, 20, 3205–3211.
[10] a) C. Bruckmeier, M. W. Lehenmeier, R. Reichardt, S. Vagin,
B. Rieger, Organometallics 2010, 29, 2199–2202; b) S. Y. T. Lee,
M. Cokoja, M. Drees, Y. Li, J. Mink, W. A. Herrmann, F. E.
Kühn, ChemSusChem 2011, 4, 1275–1279.
[11] M. L. Lejkowski, R. Lindner, T. Kageyama, G. É. Bódizs, P. N.
Plessow, I. B. Müller, A. Schäfer, F. Rominger, P. Hofmann, C.
Futter, S. A. Schunk, M. Limbach, Chem. Eur. J. 2012, 18,
14017–14025.
Experimental Section
Experimental details and spectroscopic data are in the Supporting
Information.
CCDC-1019252 [for 2·2(CH2Cl2)] contains the supplementary crys-
tallographic data for this paper. These data can be obtained free
of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[12] a) K. Koo, G. L. Hillhouse, A. L. Rheingold, Organometallics
1995, 14, 456–460; b) K. Koo, G. L. Hillhouse, Organometallics
1995, 14, 4421–4423.
Supporting Information (see footnote on the first page of this arti-
[13] P. S. Jarrett, P. J. Sadler, Inorg. Chem. 1991, 30, 2098–2104.
[14] While only 2.67 equiv. of X2 are required to form the products
in Scheme 1, it was observed experimentally that 3 equiv. are
required to completely consume 1. A small amount of unchar-
acterized insoluble white solid is also formed in this reaction,
which may explain this difference. It is also possible that some
[(dppe)NiBr3] is formed.
[15] a) S. S. Sandhu, R. S. Sandhu, Inorg. Chim. Acta 1972, 6, 383–
386; b) F. Mani, M. Bacci, Inorg. Chim. Acta 1972, 6, 487–490.
[16] L. R. Gray, S. J. Higgins, W. Levason, M. Webster, J. Chem.
Soc., Dalton Trans. 1984, 459–467.
cle): Experimental details and spectroscopic data.
Acknowledgments
This work was supported by the US National Science Foundation
(CHE-0911180 and CHE-1048528) and the US Department of En-
ergy (grant DE-FG02-07-ER15910 to M. D. H.).
[1] a) H. Hoberg, D. Schaefer, J. Organomet. Chem. 1983, 251,
C51–C53; b) H. Hoberg, D. Schaefer, G. Burkhart, C. Krüger,
M. J. Romäo, J. Organomet. Chem. 1984, 266, 203–224; c) H.
Hoberg, Y. Peres, C. Krüger, Y.-H. Tsay, Angew. Chem. Int. Ed.
Engl. 1987, 26, 771–773; Angew. Chem. 1987, 99, 799; d) H.
Hoberg, A. Ballesteros, A. Sigan, C. Jegat, A. Milchereit, Syn-
thesis 1991, 395–398.
[2] a) R. Fischer, B. Nestler, H. Schütz, Z. Anorg. Allg. Chem.
1989, 577, 111–114; b) A. M. Castaño, A. M. Echavarren, Tet-
rahedron Lett. 1990, 31, 4783–4786; c) R. Fischer, D. Walther,
G. Bräunlich, B. Undeutsch, W. Ludwig, J. Organomet. Chem.
1992, 427, 395–407; d) R. Fischer, D. Walther, R. Kempe, J.
Sieler, B. Schönecker, J. Organomet. Chem. 1993, 447, 131–136.
[3] a) J. Langer, R. Fischer, H. Görls, D. Walther, J. Organomet.
Chem. 2004, 689, 2952–2962; b) R. Fischer, J. Langer, A. Mal-
assa, D. Walther, H. Görls, G. Vaughan, Chem. Commun. 2006,
2510–2512.
[17] The organic products are likely derived from C–C coupling of
1 and the halo anhydride.
[18] K. Muto, J. Yamaguchi, A. Lei, K. Itami, J. Am. Chem. Soc.
2013, 135, 16384–16387.
[19] I. M. Angulo, E. Bouwman, M. Lutz, P. W. Mul, A. L. Spek,
Inorg. Chem. 2001, 40, 2073–2082.
[20] a) C. Amatore, A. Jutand, M. A. M’Barki, Organometallics
1992, 11, 3009–3013; b) C. Amatore, A. Jutand, A. Thuilliez,
Organometallics 2001, 20, 3241–3249; c) C. Bianchini, A. Meli,
W. Oberhauser, Organometallics 2003, 22, 4281–4285.
[21] a) D. L. Hughes, R. A. Reamer, J. Org. Chem. 1996, 61, 2967–
2971; b) P. J. Harvey, M. von Itzstein, I. D. Jenkins, Tetrahe-
dron 1997, 53, 3933–3942; c) J. McNulty, A. Capretta, V. Larit-
chev, J. Dyck, A. J. Robertson, Angew. Chem. Int. Ed. 2003, 42,
4051–4054; Angew. Chem. 2003, 115, 4185.
Received: October 6, 2014
Published Online: October 21, 2014
Eur. J. Inorg. Chem. 2014, 5491–5494
5494
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim