Nickel Fluorocarbene Metathesis with Fluoroalkenes

Article abstract of DOI:10.1002/anie.201802090

Alkene metathesis with directly fluorinated alkenes is challenging, limiting its application in the burgeoning field of fluoro-organic chemistry. A new nickel tris(phosphite) fluoro(trifluoromethyl)carbene complex ([P₃Ni]=CFCF₃) reacts with CF₂=CF₂ (TFE) or CF₂=CH₂ (VDF) to yield both metallacyclobutane and perfluorocarbene metathesis products, [P₃Ni]=CF₂ and CR₂=CFCF₃ (R=F, H). The reaction of [P₃Ni]=CFCF₃ with trifluoroethylene also yields metathesis products, [P₃Ni]=CF₂ and cis/trans-CFCF₃=CFH. However, unlike reactions with TFE and VDF, this reaction forms metallacyclopropanes and fluoronickel alkenyl species, resulting presumably from instability of the expected metallacyclobutanes. DFT calculations and experimental evidence established that the observed metallacyclobutanes are not intermediates in the formation of the observed metathesis products, thus highlighting a novel variant of the Chauvin mechanism enabled by the disparate four-coordinate transition states.

Full text of DOI:10.1002/anie.201802090

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Nickel Fluorocarbene Metathesis with Fluoroalkenes
Authors: Daniel J. Harrison, Alex L. Daniels, Jia Guan, Bulat M.
Gabidullin, Michael B. Hall, and R. Tom Baker
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201802090
Angew. Chem. 10.1002/ange.201802090
Link to VoR: http://dx.doi.org/10.1002/anie.201802090
http://dx.doi.org/10.1002/ange.201802090

10.1002/anie.201802090
Angewandte Chemie International Edition
Nickel Fluorocarbene Metathesis with Fluoroalkenes
Daniel J. Harrison,a,c Alex L. Daniels,a,c Jia Guan,b,c Bulat M. Gabidullin,a Michael B. Hall,b,* and R. Tom
Bakera,*
a Department of Chemistry and Biomolecular Sciences and Centre for Catalysis Research and Innovation, University of Ottawa, 30
Marie Curie, Ottawa, ON K1N 6N5 Canada
b Department of Chemistry and Laboratory for Molecular Simulation, Texas A&M University, College Station, TX 77843-3257 USA
c These authors contributed equally
Abstract: Alkene metathesis with directly fluorinated alkenes is
challenging, limiting its application in the burgeoning field of
fluoro-organic chemistry. A new nickel tris(phosphite) fluoro-
(trifluoromethyl)carbene complex ([P₃Ni]=CFCF₃ ) reacts with
CF₂=CF₂ (TFE) or CF₂=CH₂ (VDF) to yield both metallacyclobutane
and perfluorocarbene metathesis products, [P₃Ni]=CF₂ and
CR₂=CFCF3 (R = F, H). The reaction of [P₃Ni]=CFCF₃ with
trifluoroethylene also yields metathesis products, [P₃Ni]=CF₂
and cis/trans-CFCF₃=CFH. However, unlike reactions with TFE
and VDF, this reaction forms metallacyclopropanes and
fluoronickel alkenyl species, resulting presumably from instability of
the expected metallacyclobutanes. Density functional theory
calculations (DFT-M06) and experimental evidence establish
that the observed metallacyclobutanes are not intermediates to
the observed metathesis products, highlighting a novel variant of
the Chauvin mechanism enabled by the disparate four-coordinate
transition states.
d9 [Ni=CF2]+ may effect selected fluoroalkene metathesis
reactions depending on the thermodynamic stability of the
resulting fluoroalkene and Ni fluorocarbene partners.10,11 We
recently reported the first examples of d10 Ni fluorocarbenes and
their much faster (vs. Co) cycloaddition reactions with
tetrafluoroethylene (TFE).8,9 We now report the first Ni=CF(CF3)
carbene complexes and show their reactivity towards fluoro-
alkenes.
The Chauvin mechanism, involving a high-energy metalla-
cyclobutane intermediate, has been the mainstay for our
understanding of metathesis activity and selectivity using both
homogeneous and heterogeneous catalysts.1,3 In contrast, we
have demonstrated that the cycloaddition reaction of Co
fluorocarbene complexes with TFE involves a high-energy
diradical intermediate followed by formation of a very stable
metallacyclobutane (ca. 30 kcal/mol below starting materials).12
The difference in energy between [M]=CF2/TFE reactants (M =
Co, Ni) and the metallacyclic products is sufficiently large that
metathesis reactions following the Chauvin mechanism are likely
unfeasible for these systems. In this work we show that a similar
mechanism is in play for the Ni=CF(CF3) carbene and several
fluoroalkenes (CR2=CF2; R = H, F) to from stable metallacyclo-
butanes, whereas a novel, parallel reaction pathway affords the
metathesis products Ni=CF2 + R2C=CF(CF3). In addition, analo-
gous reactions of trifluoroethylene (TrFE) afford both metathesis
products and products derived presumably from unstable
metallacyclobutanes. The thermodynamics and mechanistic
pathways of these unprecedented transformations have been
examined by density functional theory (DFT) methods (M06).
The new fluoro(trifluoromethyl)carbene complex Ni[P(O-i-
Pr)3]3(=CFCF3) (1) was obtained from the reaction of Ni[P(O-i-
Pr)3]4 with Cd(CF2CF3)2∙DME (DME = 1,2-dimethoxyethane)13,14
(Scheme 2; ~40% isolated yield).15,16 Our synthetic approach is
similar to that developed by Roper and coworkers for making
precious metal difluorocarbenes [M0]=CF2; M = Ru, Os).10c,17
Analogous reactions with other NiL4 complexes {L = PPh3,
P(OMe)3, P(OPh)3, CN[2,6-Me2(C6H3)], CO} failed to give the
corresponding NiL3(=CFCF3) compounds in greater than trace
amounts.
Metal-catalyzed alkene metathesis is an immensely valuable
C–C bond-forming method in organic synthesis.1 This process
involving fluoroalkenes could be an appealing option for
obtaining new hydrofluoroalkenes which have recently been
identified as low global warming potential refrigerants and
blowing agents.2 Metathesis traditionally produces new alkenes,
polymers and organic compounds through the widely accepted
[2+2] cycloaddition/retro-cycloaddition pathway (Chauvin mech-
anism; Scheme 1).1 Computational studies suggest that this
mechanism would be problematic for fluoroalkenes using typical
Ru catalysts, as confirmed experimentally.3,4 Although recent
elegant work has shown that cross-metathesis of fluoroalkenes
and electron-rich alkenes can be accomplished with both Ru and
Group
6
catalysts,5,6 addition of metal fluorocarbenes to
electron-poor fluoroalkenes will likely require the former to be
strongly nucleophilic. In light of the strong M-CF bonds formed by
second- and third-row transition metals, we have targeted
Scheme 1
nucleophilic first-row metal fluorocarbenes and recently
demonstrated their [2+2] cycloaddition to fluoroalkenes.7,8,9 In
fact, early gas phase work by Beauchamp et al. suggested that
Scheme 2
This article is protected by copyright. All rights reserved.

10.1002/anie.201802090
Angewandte Chemie International Edition
COMMUNICATION
Complex 1 reacts with TFE (1.6 equiv, room temperature in
C6D6, < 20 min) to yield trifluoromethyl-substituted perfluoro-
nickelacyclobutane 2 (Scheme 3). Significantly, CFCF3/CF2
metathesis products are also observed in the same reaction
mixture; specifically, hexafluoropropene (HFP) and a new nickel
difluorocarbene (3). The ratio of metallacycle to [Ni]=CF2 is ~ 9:1
for this system. Additionally, the reaction is significantly retarded
upon addition of excess phosphite without affecting the product
ratio, suggesting that both products are formed via a dissociative
mechanism.
Scheme 4
were obtained (Scheme 4). Assuming initial formation of the two
regioisomeric metallacyclobutanes, we propose two routes for
their further conversion. For major isomer A a 2,1-fluoride
migration results in ring contraction to form the E/Z-isomers of
(4) in a similar reaction to that reported previously for the Co
perfluorinated variant using a Brønsted acid catalyst.7 The
Scheme 3
To assess the effect of hydrogen substitution on the alkene,
we showed that reaction of
1 with 1,1-difluoroethylene
(CF2=CH2, VDF) also gives metathesis and metallacyclobutane
products (Scheme 3; R = H).24 Note that the alkene generated
by the metathesis pathway, CH2=CFCF3, is the industrially
relevant refrigerant HFO-1234yf.18 Compared to the TFE
reaction, metathesis products now constitute a larger proportion
of the product mixture (2b:3 ≈ 5:1), illustrating the sensitivity of
the product distribution to the alkene substituents. For both
reactions, the mass balance is near-quantitative: the
metallacycle (2a or 2b) and nickel difluorocarbene (3) account
for >95% of 1. Further, the ratio of metallacycle to metathesis
products does not change with time, suggesting that the former
is not an intermediate en route to the latter.19 This observation
Scheme 5
driving force for these reactions is presumably formation of
stable CF3 groups (Scheme 5).21 In contrast, minor isomer B
undergoes C-F bond activation by the Ni center, forming the
E/Z-fluoronickel alkenyl isomers (5). Based on our previous
work involving fluoride abstraction from fluorometalla-
cyclobutanes7,9 and recent DFT studies on Ru carbenes22
a
points to
metathesis reactions (see below).
Interestingly, the cycloaddition reaction of
a non-traditional Chauvin mechanism for these
plausible mechanism for the formation of (5) is a β-fluoride
abstraction from metallacycle B forming a Ni-F allyl complex,
followed by a 1,3-fluoride shift (Figure S51). The detailed
reaction pathway for this transformation is still under
investigation. Although metallacycles A and B are not observed,
calculations confirm their intermediate thermodynamic stability
between those derived from TFE and VDF (see SI). In addition,
the absence of products like 4 and 5 from reactions of 1 with
either TFE or VDF23 suggests that the presence of a CHF moiety
in the metallacyclobutanes allows for a lower energy transition
state for the 2,1-F migration and C-F bond activation pathways.
Assuming intermediate formation of 4 and 5, the metathesis
products constitute almost 50% of the fluorine [(4 + 5):3 ≈ 2:1].
A related Ni fluorocarbene complex, Ni[P(OMe)3]3(=CFCF3)
(6), was synthesized through two-electron reduction8a,24 of a
mixture of Ni(PPh3)2(CF2CF3)Br and P(OMe)3 albeit with PPh3
and other minor contaminants (19F NMR yield: 17%; Scheme 6).
1 with VDF
produced a single regioisomer with the methylene fragment in
the β-position of the metallacycle. This may be due to steric
effects where the carbon with smaller substituents (CH2 vs. CF2)
can approach the large Ni=CFCF3 fragment more easily.
However, if the metallacycle is derived primarily through the
diradical mechanism of the cycloaddition reaction shown
previously for Co,12 it is also possible that the obtained
regioisomer results from added stabilization of the diradical
intermediate .20
We further probed the effect of fluoroalkene substituents
through reaction of 1 with 1,1,2-trifluoroethylene (TrFE) which
again produced the expected metathesis products Ni=CF2 and
cis/trans-CHF=CFCF3. However the expected metallacyclobu-
tane products were not observed by 19F, 31P and 1H NMR;
instead metallacyclopropanes and fluoronickel alkenyl species
2
This article is protected by copyright. All rights reserved.

10.1002/anie.201802090
Angewandte Chemie International Edition
COMMUNICATION
Figure 1: Reaction coordinate for reaction of 1 with TFE
Treatment of 6 with TFE afforded metathesis and metalla-
cyclobutane products in a 1:1.4 ratio, demonstrating a significant
ancillary ligand effect. In contrast, reaction of Ni(tripod)=CF(CF3)
complex (7)16 with TFE gave metallacycle with no metathesis
products ([tripod = MeC(CH2PPh2)3]; see SI).
The C2–C3 bond in TFE is then cleaved, with one –CF2
residue bound to the Ni, while the other –CF2 residue binds to
carbene carbon C1 to produce PI’ (ΔG = -13.5 kcal/mol). Finally,
metathesis products HFP and Ni[P(O-i-Pr)3]3(=CF2) (PI) are
formed by exchange between the ancillary ligand and
CF2=CFCF3, a process that is energetically favorable by 10.9
kcal/mol.
To further understand these observations, we investigated
these reactions in more detail using density functional theory
The unproductive metallacycle pathway (green in Figure 2),
commences with the formation of an open-shell singlet diradical
intermediate (osIM0) via one C atom of TFE weakly bonding with
the carbene C atom, without the direct involvement of the metal.
Formation of the C1–C2 σ bond (1.529 Å), through osTS1 (ΔG =
17.9 kcal/mol) followed by low-energy rotation through osTS2 (ΔG
= -5.8 kcal/mol) around the newly-formed C1–C2 bond affords
osIM2, another diradical intermediate that is slightly less stable
than os-1IM1. Finally, metallacycle product PII results from the
C3–Ni diradical of osTS3 (ΔG = -7.8 kcal/mol) with formation of
the C3–Ni bond in the same plane as the ancillary ligands (see
Table S1 for spin densities of the open-shell singlet species).
The metallacycle and metathesis products both exhibit
remarkably low energies (-44.4 kcal/mol and -24.4 kcal/mol,
respectively) relative to the starting materials. Thus, the path-
ways leading to metathesis products and metallacycle are too
exothermic to be reversible. Since the rate-controlling transition
states in both pathways, TS1 and os-1TS1, have comparable
barriers (19.0 vs 17.9 kcal/mol), the product ratio will be
determined by subtle differences in these transition states, as
shown with the observed increase of metathesis products with
ancillary ligand and upon replacement of C-F bonds (with C-H)
from the metathesis partner. Moreover, to eliminate the
possibility of this transformation progressing via a triplet state, it
was calculated that for the rate-determining barrier, TS1, the
open-shell singlet is significantly lower in energy than the
closed-shell singlet (17.9 kcal/mol vs 22.6 kcal/mol), which is
lower in turn than the triplet state (22.6 kcal/mol vs 24.1
kcal/mol). At the key intermediate IM1, the closed-shell singlet
state is very unstable, while the open-shell singlet is still more
stable than the triplet. After IM1, the open-shell singlet and triplet
are close in energy until the product, which is much more stable
Scheme 6
(DFT) approaches employing the M06 functional.21,25,26,27,28,29 In
the reaction between TFE and 1 both reactants have singlet
ground states, and the ancillary ligand P(O-i-Pr)3 has a low free
energy of dissociation of 0.5 kcal/mol from Ni[P(O-i-
Pr)3]3(=CFCF3). Hence we propose that the reaction initiates with
loss of P(O-i-Pr)3 and subsequent interaction of Ni[P(O-i-
Pr)3]2(=CFCF3) with TFE (Figure 2).30 The calculated results
reveal two independent pathways under experimental
conditions, which can lead to metathesis and metallacycle
products, respectively.
In the metathesis pathway (blue in Figure 2), TFE binds
weakly to the nickel center in IM0 through both C atoms leading
to transition state TS1, located 19.0 kcal/mol above the
reactants, and intermediate IM1, about 4.9 kcal/mol higher than
the reactants, in which the TFE C2–C3 bond is elongated by
back-bonding from the electron-rich Ni center. Further elon-
gation of the C2–C3 bond that accompanies formation of bicyclic
transition state TS2 (about 10.9 kcal/mol above the reactants)
occurs in a plane that is perpendicular to that of the ancillary
phosphite ligands. This unusual strained bicyclic intermediate is
high enough in energy to prevent the complex from going
through the highly downhill formation of the metallacycle, vide
infra (Scheme 7).
Scheme 7
3
This article is protected by copyright. All rights reserved.

10.1002/anie.201802090
Angewandte Chemie International Edition
COMMUNICATION
as the singlet, confirming that the diradical mechanism is the
most probable. (see experimental SI)
16
(
)
The [Ni]=CFCF3 sub-structure was confirmed by synthesizing
Ni(P3)(=CFCF3), 7 [P3 = MeC(CH2PPh2)3] from 1, by phosphite-substitution, to
To summarize, we have demonstrated the first examples of
metal fluorocarbene metathesis with fluoroalkenes, through a
novel variant of the Chauvin mechanism enabled by the
disparate four-coordinate transition states. These results extend
our understanding of metathesis mechanisms beyond the
original Chauvin postulate, in which the metallacycle is an
intermediate en route to metathesis products.1
Despite the important advances described herein, conside-
rable challenges still face metal-mediated polyfluoroalkene
metathesis. First, a more general source of nucleophilic metal
carbenes is needed.31 Second, ground state energy differences
between the Ni=CF2 and Ni=CFCF3 carbene complexes
introduce additional barriers to efficient metal-catalyzed
fluoroalkene metathesis that will require careful selection of the
carbene/fluoroalkene partners. Ongoing experimental and
computational efforts are focused on finding more systems that
favor the metathesis pathway.
obtain a crystalline derivative suitable for X-ray structural determination (see
SI).
(
17) G. R. Clark, S. V. Hoskins, T. C. Jones, W. R. Roper, J. Chem. Soc.,
Chem. Commun. 1983, 719-721.
(
18) M. Reisch, Chem. Eng. News 2010, 88, 23.
(
(
19) Complex 3 does not react with excess HFP (24 h, 50 oC).
20) J. Clayden, N. Greeves, S. Warren, P. Wothers. Radical stability. In
Organic Chemistry; Oxford: New York, 2004; pp. 1026-1028.
(
(
21) D. O’Hagan, Chem. Soc. Rev. 2008, 37, 308-319.
22) W. J. van Rensburg, P. J. Steynberg, W.H. Meyer, M. M. Kirk, G.S.
Forman, J. Am. Chem. Soc. 2004, 126, 14332-14333.
(
23) DFT calculations for the stability of the metallacyclobutanes/-propanes
formed from reactions of 1 with TFE and VDF are shown in the computational
SI.
(
24) (a) R. P. Hughes, R. B. Laritchev, J. Yuan, J. A. Golen, A. N. Rucker, A. L.
Rheingold, J. Am. Chem. Soc. 2005, 127, 15020-15021. (b) C. J. Bourgeois,
R. P. Hughes, J. Yuan, A. G. DiPasquale, A. L. Rheingold, Organometallics
2006, 25, 2908-2910. (c) J. Yuan, R. P. Hughes, A. L. Rheingold, Eur. J. Inorg.
Chem. 2007, 4723-4725.
(
25) (a) The density functional theory (DFT) calculations were performed with
the Gaussian 09 Suite of programs: Gaussian 09, revision D. 01, M. J. Frisch,
et al. Gaussian, Inc.: Wallingford, CT. 2013 (see SI for full reference). (b) The
M06 functional (see ref. 26) was employed with the 6-311+G* basis set for the
C, H, O, F, and P atoms, and the Stuttgart quasi-relativistic basis set and
effective core potential (see ref. 29) for Ni. Each species was optimized in the
gas phase on an ultrafine grid. Analytical frequency calculations were
performed on all optimized structures to ensure that either a minimum or a
first-order saddle point was achieved, and to obtain the thermal Gibbs free
energy corrections. Further single point calculations were carried out on the
basis of the optimized geometries with the continuum solvation model SMD
(see ref. 28) at the same level of theory to account for solvent effects, where
THF (tetrahydrofuran) was employed as the solvent. The reported energies
(ΔG) in this work include the electronic energy, DFT-D3 empirical dispersion
corrections proposed by Grimme et al. (see ref. 29), the gas-phase thermal
correction for the Gibbs free energy and SMD solvation corrections. The bond
energy of the ancillary ligand L (L = P(O-i-Pr)3) from the nickel carbene
complex was calculated from the following equation: NiL3(=CFCF3) 
NiL2(=CFCF3) + L.
Acknowledgements
We thank the NSERC and the Canada Research Chairs
program for generous financial support and the University of
Ottawa, Canada Foundation for Innovation and Ontario Ministry
of Economic Development and Innovation for essential
infrastructure. MBH thanks The Welch Foundation (A-0648) and
the National Science Foundation (CHE-1300787) for their
support. ALD thanks NSERC for a CGS-D scholarship.
Keywords: keyword
(fluoroalkenes)• keyword
1
(Alkene metathesis)
(nickel-carbene)
•
keyword
2
4
3
• keyword
(Chauvin Mechanism) • keyword 5 (metallacyclobutanes)
(1) (a) Olefin Metathesis: Theory and Practice; Grela, K., Ed.; John Wiley &
Sons: Hoboken, NJ, 2014. (b) Handbook of Metathesis; R. H. Grubbs, Ed.;
Wiley-VCH: Weinheim, Germany, 2003.
(
(
26) Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215-241.
27) (a) M. Dolg, U. Wedig, H. Stoll, H. Preuss, J. Chem. Phys. 1987, 86, 866-
(2) (a) S. F. Y. Motta, E. D. V. Bercerra, M. W. Spatz, Analysis of LGWP
Alternatives for Small Refrigeration (Plug-in) Applications; International
Refrigeration and Air Conditioning Conference; Purdue University: West
Lafayette, IN, 2010; Paper 1149. (b) A. Mota-babiloni, P. Makhnatch, R.
Khodabandeh, Intl. J. Refrig. 2017, 82, 288-301.
872. (b) J. M. L. Martin, A. Sundermann, J. Chem. Phys. 2001, 114, 3408-
3420.
(
28) A. V. Marenich, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. B 2009, 113,
6378-6396.
29) S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132,
154104-1-154104-19.
30) That the prior dissociation energy of a phosphine adds to this reaction
(
(3) (a) S. Fomine, M. A. Tlenkopatchev, Appl. Catal., A. 2009, 355, 148-155.
(b) M. Vasiliu, A. J. Arduengo, III, D. A. Dixon, J. Phys.Chem. C 2014, 118,
13563-13577.
(
barrier is confirmed through treatment of complex 7 with TFE which shows
little reactivity until the solution is heated to 60 oC (more details in SI).
(4) T. M. Trnka, M. W. Day, R. H. Grubbs, Angew. Chem. Int. Ed. 2001, 40,
3441-3444.
(
31) G. M. Lee, D. J. Harrison, I. Korobkov, R. T. Baker. Chem. Commun.,
2014, 50, 1128.
(5) Y. Takahira, Y. Morizawa, J. Am. Chem. Soc. 2015, 137, 7031-7034.
(6) (a) M. J. Koh, T. T. Nguyen, H. Zhang, R. R. Schrock, A. H. Hoveyda,
Nature 2016, 531, 459-465. (b) T. T. Nguyen, M. J. Koh, X. Shen, F. Romiti,
R. R. Schrock, A. H. Hoveyda, Science 2016, 352, 569-574. (c) M. J. Koh, T.
T. Nguyen, J. K. Lam, S. Torker, J. Hyvl, R. R. Schrock, A. H. Hoveyda,
Nature. 2017, 542, 80-86.
(7) D. J. Harrison, G. M. Lee, M. C. Leclerc, I. Korobkov, R. T. Baker, J. Am.
Chem. Soc. 2013, 135, 18296-18299.
(8) D. J. Harrison, S. I. Gorelsky, G. M. Lee, I. Korobkov, R. T. Baker,
Organometallics, 2013, 32, 12-15.
(9) D. J. Harrison, A. L. Daniels, I. Korobkov, R. T. Baker, Organometallics
2015, 34, 5683−5686.
(
10) L. F. Halle, P. B. Armentrout, J. L. Beauchamp, Organometallics. 1983, 2,
1829-1833.
(
11) For a unique example of apparent Ni-mediated alkene metathesis see: F.
G. N. Cloke, P. B. Hitchcock, M. F. Lappert, C. MacBeath, G. O. Mepsted, J.
Chem. Soc., Chem. Commun., 1995, 87-88.
(
12) J. T. Fuller, D. J. Harrison, M. C. Leclerc, R. T. Baker, D. H. Ess, R. P.
Hughes, Organometallics 2015, 34, 5210-5213.
(
(
(
13) S. D. Ittel, Inorg. Synth. 1990, 28, 98-104.
14) H. Lange, D. Naumann, J. Fluorine Chem. 1984, 26, 1-18.
15) Complex 1 was obtained in ~95% purity with the only NMR-visible
contaminant being the starting material, Ni[P(O-i-Pr)3]4. Sublimation gave
marginally improved purity but significantly diminished yield. See Supporting
Information.
4
This article is protected by copyright. All rights reserved.