Selective Copper Complex-Catalyzed Hydrodefluorination of Fluoroalkenes and Allyl Fluorides: A Tale of Two Mechanisms

Article abstract of DOI:10.1021/jacs.9b03101

The transition to more economically friendly small-chain fluorinated groups is leading to a resurgence in the synthesis and reactivity of fluoroalkenes. One versatile method to obtain a variety of commercially relevant hydrofluoroalkenes involves the catalytic hydrodefluorination (HDF) of fluoroalkenes using silanes. In this work it is shown that copper hydride complexes of tertiary phosphorus ligands (L) can be tuned to achieve selective multiple HDF of fluoroalkenes. In one example, HDF of the hexafluoropropene dimer affords a single isomer of heptafluoro-2-methylpentene in which five fluorines have been selectively replaced with hydrogens. DFT computational studies suggest a distinct HDF mechanisms for L₂CuH (bidentate or bulky monodentate phosphines) and L₃CuH (small cone angle monodentate phosphines) catalysts, allowing for stereocontrol of the HDF of trifluoroethylene.

Full text of DOI:10.1021/jacs.9b03101

Subscriber access provided by ALBRIGHT COLLEGE
Article
Selective Copper Complex-catalyzed Hydrodefluorination of
Fluoroalkenes and Allyl Fluorides: A Tale of Two Mechanisms
Nicholas O. Andrella, Nancy Xu, Bulat M. Gabidullin, Christian Ehm, and R. Tom Baker
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b03101 • Publication Date (Web): 17 Jun 2019
Downloaded from http://pubs.acs.org on June 17, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 19
Journal of the American Chemical Society
1
2
3
4
5
6
Selective Copper Complex-catalyzed Hydrodefluorination of
Fluoroalkenes and Allyl Fluorides: A Tale of Two Mechanisms.
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
1
*2
*1
Nicholas O. Andrella , Nancy Xu , Bulat M. Gabidullin , Christian Ehm , and R. Tom Baker
1
Department of Chemistry and Biomolecular Sciences and Centre for Catalysis Research and Innovation, 30 Marie Curie,
University of Ottawa, ON K1N 6N5 Canada
2
Università di Napoli Federico II, Dipartimento di Scienze Chimiche, Via Cintia 80126 Napoli, Italy
KEYWORDS. Hydrodefluorination catalysis, Fluoroalkene, Copper, Computational mechanistic study, Ligand-induced mechanism
changes
ABSTRACT: The transition to more economically friendly small-chain fluorinated groups is leading to a resurgence in the
synthesis and reactivity of fluoroalkenes. One versatile method to obtain a variety of commercially relevant hydrofluoroalkenes
involves the catalytic hydrodefluorination (HDF) of fluoroalkenes using silanes. In this work it is shown that copper hydride
complexes of tertiary phosphorus ligands (L) can be tuned to achieve selective multiple HDF of fluoroalkenes. In one example,
HDF of hexafluoropropene dimer affords a single isomer of heptafluoro-2-methylpentene in which five fluorines have been
2
selectively replaced with hydrogens. DFT computational studies suggest distinct HDF mechanisms for L CuH (bidentate or bulky
monodentate phosphines) and L
of trifluoroethylene.
3
CuH (small cone angle monodentate phosphines) catalysts, allowing for stereocontrol of the HDF
intimate ion pair (Scheme 1d). It can be expected that the
disparity between the activation energies of these processes
INTRODUCTION
The synthesis and chemistry of fluorinated alkenes [FAs] is
could lead to significantly different outcomes, especially when
considering product distribution in a multi HDF reaction.
1
,2
currently experiencing a renaissance due to the need for
3
smaller fluorinated chains in materials applications (cf.
Furthermore, selectivity issues with HDF of fluoroarenes
are less problematic because as fluorine is removed, HDF
becomes more difficult. On top of this, the use of directing
groups can increase selectivity. However, such strategies are
3
c,d
environmental persistence of long chain fluorocarbons) and
the development of economical late-stage fluoroalkylation
2
4
processes, including cross-metathesis. Moreover, selected
unsaturated hydrofluoroalkenes have been identified as low
8
not necessarily available when using FAs as substrates.
global warming potential refrigerants [e.g. H
2
C=CFCF
HC=CHCF
336mzz(Z)] when compared to their saturated counterparts
3
, R-
, R-
Product selectivity must arise from the inherent reactivity of
the catalyst or rely on thermodynamic distribution of isomers
with the former often leading to uncontrolled multi-HDF and
the latter leaving little control for obtaining the desired
product.
1
1
234yf] and blowing agents [e.g. (Z)-CF
3
3
[
i.e. hydrofluorocarbon refrigerants and blowing agents, such
5
as R-134a, R-401a, R-245fa, etc.]. Most currently used
synthetic routes to the latter employ well-established
technologies (i.e. halogen exchange [Swarts fluorination], HX
elimination [X
conditions, HF and expensive reactors.
= halogen]) that require harsh/caustic
5
Catalytic
6
hydrodefluorination [HDF] represents a potential alternative.
Selective and potentially consecutive C-F bond activation and
substitution by C-H bonds, could provide new routes to these
valuable compounds and additional previously unavailable
FAs.
Detailed studies of metal-catalyzed HDF have been focused
largely on fluoroarenes but these reaction conditions are not
easily transposed to FAs, as their mechanisms can be quite
different. For example, fluoroarene HDFs generally proceed
by attack of the hydride at the Cα-F carbon (Scheme 1a,b)
while FAs typically undergo hydride attack at the Cβ-CF
7
Scheme 1. HDF of fluoroarenes vs fluoroalkenes.
carbon (Scheme 1c,d). To further complicate matters, the
addition of M-H to the FA can proceed either via the more
traditional insertion mechanism (Scheme 1c) or, as is shown
herein, by nucleophilic addition of the hydride generating an
In studies of HDF of hexafluoropropene (HFP, 1a; Scheme
2
), Jones et al. reported the use of stoichiometric zirconocene
dihydride for consecutive substitution but selectivity could
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 19
only be achieved with careful control of the Zr/FA
stoichiometry. Similarly, using a stoichiometric ruthenium
Scheme 3. Copper complex-catalyzed HDF of fluoroalkenes.
[PMHS = polymethylhydrosiloxane]
9
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
hydride complex, Whittlesey and co-workers converted 1a to a
5:3 mixture of mono- and dihydrodefluorinated products.
Holland et al. demonstrated the iron complex-catalyzed HDF
1
0
As such, described herein is the reactivity of P-ligated [Cu]-
H with various FAs, development of tunable selectivity in Cu
complex-catalyzed HDF, and discovery of a new catalytic
HDF pathway. The latter allows for stereocontrol with some
FAs and for the HDF of less electrophilic FAs, like vinylidene
difluoride (2e) (Scheme 3, bottom).
of FAs although turnover numbers (ToNs) were limited and
high temperature was required (100 °C).11 In contrast, Lentz
and co-workers showed that Ti-based catalysts are active
under ambient conditions with ToN up to 125 although
1
2
dihydrodefluorination was only observed in trace amounts.
RESULTS AND DISCUSSION
Another contribution from the Lentz group showed that HDF
of FAs using Al or Ga hydrides can be catalyzed by N- and O-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Catalyzed HDF Reactions. Ligand Screening. Extending
our previous work to include catalytic HDF, excess
dimethylphenyl silane (vs 1a) was added to the reaction of a
1
3
donors with ToNs up to 87.
catalytic amount of [CuH(PPh
3 6
)] (10 mol % Cu) with 1a at 45
°
C for 8 h. Heating the reaction at 45 °C provided the optimal
rate, as higher temperatures were found to decompose the
catalyst (reactions turned brown) or lead to HFP
1
4c
oligomerization . As expected, this reaction led readily a
mixture of E- and Z-1,2,3,3,3-pentafluoropropene isomers
(
(
PFP, 1b,c) but, unexpectedly, also 2,3,3,3-tetrafluoropropene
1d), in a rare double HDF (Table 1, Entry 1). It is notable
that substitution of (E)-PFP, 1b, is significantly favored vs the
Z-isomer, 1c, such that 1b has been entirely consumed while
1
c still remains. No additional HDF was observed after
another 24 h. No differences in reactivity were observed when
employing less expensive tetramethyldisiloxane (TMDS)
(Table 1, Entry 2) or even moisture-tolerant waste product
polymethylhydrosiloxane (PMHS) (Table 1, Entry 3).
Similarly, the change in solvent had little impact on the
reaction outcome, except for dimethylformamide (DMF),
which gave the same product ratios independent of ligand,
followed by rapid catalyst deactivation (See SI Table S1).
However, while toluene and tetrahydrofuran (THF) both
worked well on NMR scale, they failed to match benzene for
consistent results when scaled to 1 L of HFP. Monitoring the
Scheme 2. Some examples of the HDF of hexafluoropropene
Recent studies have identified the potential of copper
1
4
complexes for catalyzed HDF of FAs (Scheme 3). For
example, the groups of Shi and of Ito have reported the HDF
1
9
of
gem-difluoroalkenes
to
E-
and
Z-terminal
conversion of 1a by F NMR spectroscopy showed that
monofluoroalkenes with high stereoselectivity using copper(I)
catalysts and base. Building on this body of knowledge, we
recently reported that isolated phosphine copper hydride
complexes readily insert HFP, followed by rapid ß-fluoride
elimination to yield Cu-F and 1,2,3,3,3-pentafluoropropene
addition of 3 equiv. or even excess PPh gave no change in the
3
HDF product slate (Table 1, Entries 4,5).
In contrast, the HDF product distribution showed a
pronounced ligand effect (
14c
(PFP, 1b,1c).
Using a similar methodology employing
Table 1). When employing bulky electron-rich ligands, like
silanes to regenerate Cu-H from Cu-F, a catalytic HDF could
be established (Scheme 3).14d
tBuXphos or PCy (Table 1, Entries 7,8), similar product
3
ratios to that with no added ligand were observed. Using the
bulky aromatic phosphine, P(o-tolyl) , however, the reaction
3
converted HFP to 1b/c isomers and then quickly ceased to be
functional (Table 1, Entry 6). This was based qualitatively on
a color change from orange to green and, quantitatively, upon
addition of more HFP that failed to react. For 1a, bidentate
ligands generated more efficient HDF catalysts.
Bis(phosphines) dppe, dppf and Xantphos (Table 1, Entries
9-11) all readily converted 1a to 1d and even 1,1-
difluoropropene, 1f, although the latter likely arises from HDF
of 1,1,2-trifluoropropene, 1e. While dppe provided the highest
selectivity for 1d, Xantphos produced 1f more slowly than
dppf. Interestingly, π-acidic triethylphosphite [P(OEt)
, Entry 12) gave selective tetra-substitution, yielding
primarily 1f in nearly the same amount as dppf. This increased
efficiency of the P(OEt) /CuH catalytic mixture is contrary to
the expected activity increase with hydricity using the more σ-
donating phosphines, dppf and Xantphos. Both P(OPh) and
P(O-o-tolyl) showed significantly reduced reactivity (Table
3
] (Table
1
3
3
3
2
ACS Paragon Plus Environment

Page 3 of 19
Journal of the American Chemical Society
, Entries 13 and 14), however, suggesting that ligand size is
1
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
also an important factor.
Table 1. Ratios of hexafluoropropene HDF products.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Entry
Ligand
1
b/1c
1e
1e
1d
1.7
1.7
1.6
1.5
1.5
1
1e
1f
Scheme 4. Selective and multiple HDF of various perfluorinated
substrates.
1
2
3
4
5
6
7
8
9
Nonea
None
Noneb
1
,2-difluorovinyl ether (4b) was observed but then eventually
consumed, presumably via β-OCF elimination (Scheme 5).
Isomer 2c is presumably formed preferentially due to the
1e
3
1e
PPh
3
1
5
gauche effect
as the cis isomer is also the most
c
1e
PPh
P(o-tolyl)
tBuXphos
PCy
3
thermodynamically stable, suggesting a late transition state.
3
9.5
1e
1.2
1.3
11
1
1e
3
dppe
dppf
1
1
0
3.0
11
Xantphos
1
2.4
1
1
1
2
3
4
P(OEt)
3
1
2.4
Scheme 5. Top: HDF of 4a leading to a 4:1 mixture of 2c,d via β-
OCF elimination. Bottom: The gauche effect increasing
3
selectivity for 2c.
d
P(OPh)
P(O-o-tolyl)
3
1
3
35
1
Ratios are based on 19F NMR integration of products vs
internal standard. All reactions gave 100% conversion of HFP
(1a) unless indicated otherwise. Following the reaction, the
headspace of the reaction was sampled to confirm solution
product distribution. a) PhMe
PMHS was used in place of TMDS; N.B. PMHS forms a biphasic
mixture with benzene. c) 60 mol % of PPh . d) complete
The HDF of perfluorocyclobutene (5a, Eq. 4.3) and HFP
dimer, perfluoro-2-methylpent-2-ene (6a, Eq. 4.5) proceeded
readily, yielding multi-HDF products. FA 5a was selectively
converted to 1,2-difluorocyclobutene (5c) via triple HDF
whereas 6a underwent quintuple HDF to generate the di-alkyl
substituted gem-difluoroalkene (6c), exclusively. The isomeric
HFP dimer, perfluoro-4-methylpent-2-ene (6b), also
underwent multiple HDF but in an unselective manner,
generating multiple products (see SI for details).
2
SiH was used in place of TMDS. b)
3
conversion of HFP not achieved after 8 h. e) only Z-isomer 1c was
observed.
Substrate Scope. Using these optimized conditions with
dppf or Xantphos, the scope was investigated for the HDF
reaction of various perfluorinated substrates (Scheme 4). Both
chlorotrifluoroethylene (3a, Eq. 4.1) and HFP analogue,
trifluoromethyl trifluorovinyl ether (4a, Eq. 4.2), yielded a 4:1
mixture of cis-:trans-1,2-difluoroethylene (2c,d), respectively.
During the first few hours of the latter reaction trifluoromethyl
3
Switching to P(OEt) , the HDF of 5a produced a mixture of
1,4,4-trifluorofluorocyclobutene (5b), 5c and 1-fluorocyclo-
butene (5d) in a 1:1:3 ratio before catalyst deactivation (Eq.
4.4). As with 1a, this ligand produced a significantly more
active
HDF
catalyst
than
the
electron-donating
bis(phosphines). FA 6a, as above, yielded only 6c,
demonstrating that this product is too electron-rich to undergo
further HDF.
The above electrophilic substrates (i.e. 1a, 5a, 6a and 6b)
did not provide an avenue to test the impact of stronger σ-
donating ligands as they react readily with electron-rich
phosphines
tricyclohexylphosphine
with
cone
(PCy
angles
).
smaller
Likewise,
than
the
1
g,16
3
decomposition of CuH complexes containing N-heterocyclic
1
7
carbenes and nitrogen chelates is well documented,
excluding these ligands from consideration. With this in mind,
the HDF of less electrophilic trifluoroethylene, 2b, was
performed. Using the optimized reaction conditions and the
Xantphos/CuH catalytic mixture, 2b was readily converted to
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 19
cis:trans (2c:2d) in a 7:1 ratio, reflecting the thermodynamic
preference, whereas use of P(OEt) gave no reaction (Scheme
6). Using three equivalents of basic phosphine PMePh
however, inverted the 2c:2d ratio to 1:3.7. As with HFP,
hydride addition always occurs regioselectively, suggesting
that Cu-H addition to 2b occurs with the FHC= fragment
oriented towards the copper.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
2
,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 6. Ligand-induced inversion of stereoselectivity in HDF
of trifluoroethylene.
Scheme 8. Synthesis of complexes 7a and 7b.
The 19F NMR spectrum of 7a in C
D displays two
6 6
In contrast to the above substrates, tetrafluoroethylene (2a)
resonances which can be easily differentiated based on their
JFH coupling constants: CF H = 50 Hz and CF R < 1 Hz
reacts with the copper hydride complexes to yield thermally
1
8
stable copper 1,1,2,2-tetrafluoroethyl complexes.
As
2
2
31
(
unresolved). The P NMR spectrum contains a broad singlet
previously reported, addition of a Lewis acid (or base) is
2j
at -20.0 ppm and the molecular structure was confirmed by
single crystal X-ray diffraction (Figure 1).
required to induce β-fluoride elimination, presumably via an
outer-sphere fluoride-elimination mechanism (Scheme ).
With three seemingly unrelated observations in hand - a)
Using P(OEt)
catalyst on par with dppf, b) Using a small cone-angle, strong
σ-donating phosphine [PMePh vs P(OEt) ] allows for HDF of
b; and c) Changing from Xantphos to PMePh in the HDF of
3
in the HDF of HFP (1a) generates an active
2
3
2
2
Scheme 7. HDF of 2a leading to a mixture of 2b, 2a’ and
incomplete conversion of 2a.
2b inverts isomer ratios of 2c:2d from 7:1 to 1:3.7 -
mechanistic studies and computations were carried out in an
attempt to consolidate them.
Cu-tetrafluoroethyl complexes. Interestingly, the Cu
2
tetrafluoroethyl complex with PMePh can be easily and safely
TM
prepared using TFE Safe Supply (generated by pyrolysis of
potassum perfluoropropionate in vacuo; see experimental).
Treatment of Stryker’s reagent, 3 equiv. PMePh
equivalents of tetramethyldisiloxane [TMDS] with 2a:CO
benzene at 70 °C afforded the Cu-CF CF H complex
(PMePh Cu(CF CF H)] (7a) in good yield (70%) and
excellent purity [>90%, Scheme , Eq. 8.1]. The use of this
methodology with Xantphos or P(OEt) did not yield the
2
and 1.5
2
in
2
2
[
2
)
3
2
2
3
desired product. The synthesis of the former complex was
instead carried out with limonene-inhibited 2a. In our first
2 2
attempt, formation of [(Xantphos)Cu(CF CF H)] (7b) was
accompanied by the [Cu](CF CF ) analog (7b’) derived from
2
3
insertion of 2a into [Cu]-F (Scheme 8, Eq. 8.2). Addition of
10 % TMDS was sufficient to convert [Cu]-F to [Cu]-H,
enabling isolation of pure 7b in good yield (Scheme 8, Eq.
Figure 1. ORTEP representation of the molecular structure of 7a.
Thermal ellipsoid probabilities are set to 35% and hydrogen atoms
omitted for clarity.
8
.3).
Mechanistic DFT Studies. To gain deeper insight into the
HDF mechanism(s) in these P-ligated copper-hydride systems,
we conducted solvent corrected (PCM: benzene) DFT studies
at the TPSSh(PCM)/TZ/TPSSTPSS-(PCM)/DZ level of
theory. For details and expected accuracy of the method see
computational details and supporting information. A model
phosphine ligand, L = PMe
possible reaction pathways in the LCu-H, L
3
, was chosen to fully analyze
Cu-H and L Cu-H
2
3
systems in the HDF of three prototypical FAs of varying
fluorine content, i.e. tetrafluoroethylene (2a), trifluoroethylene
(2b) and 1,1-difluoroethylene (2e). Relevant transition states
(TS) and resting states (RS) were then analyzed for
experimentally employed systems, i.e. L = PPh
3
, PMePh
2
and
4
ACS Paragon Plus Environment

Page 5 of 19
Journal of the American Chemical Society
TS(A-F)dimer
Hydride Addition
TS(C-D)
Insertion
TS(D-G)
F-Elimination
PMe
influence of the ligands on the HDF selectivity of 2b.
PMe model system. 2a. Spectroscopic studies of copper
hydrides have shown that the only observable species in
2
Ph, dppe and Xantphos, to understand the pronounced
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
L2
1
7a,b,19b-d
A
B
solution are dimers,
used . Kinetic studies have indicated that the active species
unless very bulky ligands are
1
9a
Cu-P 2.23(1)
Cu-P 2.28(1)
Cu-HA 1.69(1)
Cu-HB 1.75(2)
C=C 1.391
Cu-P 2.25(1)
2
0
Cu-F 2.093
C-F 1.938
C=C 1.407
Cu-C 2.129
in copper-bis(phosphine)-catalyzed hydrogenation
and
Cu-H 1.570
C-F 1.757
C=C 1.454
Cu-C 1.991
2
1
hydrosilylation reactions are mononuclear. We considered
HDF in the present systems to take place in monomeric or
TS(A-F)
Hydride Addition
TS(E-H)
TS(E-H) (2nd view)
F-Elimination
dimeric copper systems. For monomeric LCu, L
Cu system (L = PMe ), we considered a) an insertion-
elimination sequence, b) ‘S V’-like H-addition as for
2
Cu and/or
L
3
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
N
L3
example observed for early transition metal-catalyzed HDF
followed by Cu-C bond formation and subsequent F-
8a
23,24
elimination c) single electron transfer,
d) σ-bond
4
,7e,8a,25-30
Cu-P 2.26(1)
Cu-P 2.30(2)
Cu-F 2.390
C-F 1.977
C=C 1.384
Cu-C 2.413
PCuP 113(3)
metathesis
or e) ‘S
N
V’-like H-addition followed by
Cu-H 1.612
C=C 1.367
PCuP 115(2)
8
a,13c,28
immediate F-elimination.
The latter two mechanisms
could be quickly excluded; a σ-bond metathesis TS could not
be located and ion pairs are stable, showing no tendency for
immediate F-elimination. SET can be excluded based on the
Figure 2. Graphical depiction of important transition states in the
HDF system of 2a with L /L Cu-H. L = PMe . Bond distances in
Å, bond angles in deg. Color scheme: C grey, H white, P orange,
F lime. Level of theory TPSSTPSS(PCM)/DZ. Solvent = benzene.
7
e, 8a
known low electron affinity of (per)fluorinated alkenes.
3
2
3
8
a
An oxidative addition–reductive elimination pathway can be
excluded based on the experimental observation of 7a,b. Both
insertion and H-addition pathways for hydrometallation could
be found for monomeric copper systems and H-addition is also
The insertion barrier from the Cu-H(alkene) complex C is
very small (+3.2 kcal/mol) and overall, the coordination TS(B-
C) is rate-limiting in this process. Nucleophilic H-addition via
possible in a dimeric L
The (PMe Cu-(μ-H)
slightly more stable than (PMe
PMe ligands in (PMe Cu-H (A, 0.0 kcal/mol) are relatively
2
Cu-(μ-H)
2
-CuL
dimer (-3.7 kcal/mol) is
Cu-H (A). The first two
2
system.
3
)
2
2
-Cu(PMe
3
)
2
TS(B-D), although possible in the L
with a much higher barrier (+17.7 kcal/mol; see discussion of
the L system for further details).
Within the monomeric L hydrometallation pathway, alkene
system, is associated
2
3 3
)
3
3 3
)
3
weakly bound (ΔG323K = 7.8 + 5.1 kcal/mol). Nonetheless, this
penalty raises the barrier for HDF in the LCu-H system
2
coordination is endergonic, but irreversible, as insertion from
the alkene complex has a much lower barrier than dissociation
of the alkene. That insertion of the FA into the Cu-H bond is
associated with only a very small barrier is unsurprising;
TS(C-D) is geometrically very early compared to the Cu-
H(alkene) complex C (Cu-H bond length + 0.01 Å, see also
Figure 3), as coordination to the late TM copper has already
activated the C=C double bond and Cu-H bonds are rather weak
significantly over the corresponding barriers in the L
systems (see supporting information). Therefore, we will focus
the discussion on the competition between L and L HDF
2 3
and L
2
3
pathways. Figure 2 shows important transition states and
Figure 3 the potential energy surface (PES) for the two
competing mechanisms for the example of 2a.
Hydrometallation in the monomeric L
dissociation of one phosphine ligand from L
kcal/mol) to form L Cu-H (B). Thereafter, coordination of 2a
alkene coord. TS(B-C), +12.6 kcal/mol) forming Cu-olefin
complex C followed by alkene insertion into the Cu-H bond
HM TS(C-D), +9.1 kcal/mol) represents the lowest energy
pathway for HDF to form L Cu-CF CF H (D).
2
system starts with
3
0
3
Cu-H (A, +7.8
(~ 60 kcal/mol). Hydrometallation leads to D (-38.7 kcal/mol),
which can be further stabilized by coordination of another
2
(
phosphine forming L
HDF can also proceed without the release of one phosphine
ligand in the L system. Unlike in the monomeric L system,
where synchronous hydrometallation via insertion is favored,
insertion into the Cu-H bond in the L system proceeds via a
two-step, asynchronous, mechanism. Nucleophilic H-addition
in this system via TS(A-F) has a very low barrier (+7.4
kcal/mol) and proceeds to a contact ion pair (CIP F, -17.7
kcal/mol). The low barrier can be understood as a direct result
of the destabilization of a C=C double bond by multiple
fluorine substituents and the weak Cu-H bond.
Coordination of a third phosphine stabilizes the forming cation
in the TS with respect to the L system. Nucleophilic addition
along the Burgi-Dunitz trajectory via TS(A-F) is essentially
dagger-like’ and requires minimal space around Cu. Forming
a solvent-separated ion pair (SIP, +17.5 kcal/mol) from the
CIP is associated with a significant increase in energy (see
supporting information). However, full separation is not
necessary, as slight reorientation of the anion leads to
3 2 2
Cu-CF CF H (-50.4 kcal/mol, E).
(
3
2
2
2
2
3
3
1,32
2
‘
formation of a Cu-C bond, ultimately yielding L
E).
3
Cu-CF
2
CF
2
H
(
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 19
Hydrometallation in the dimeric system proceeds by
nucleophilic addition along the Burgi-Dunitz trajectory via
TSA-Fdimer, similar to the L system and leads to a similar CIP
from which Cu-C bond formation and dimer splitting can
proceed via simple reorientation. This pathway is preferred by
pathway is still associated with a significant barrier (36.5
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
kcal/mol); moreover, elimination is endergonic (+8.3
kcal/mol). In line with experiment, where 7b was isolated, E is
therefore predicted to be stable under experimental conditions
with respect to F-elimination and Lewis acids need to be
added to shift the equilibrium to the product side by removal
of fluoride via salt metathesis, enforcing HDF of 2a. The F-
3
4
.2 kcal/mol in the model system over the L
preference stems nearly entirely from the dimer stability (-3.7
kcal/mol) and can be understood, similarly to the L
environment, due to the stabilization of a cationic charge.
However, unlike the ‘dagger-like’ L system, the L Cu-(μ-H)
CuL system is much more susceptible to variation of steric
3
pathway. The
3
elimination TS(E-H) in the L
geometrically not late like the corresponding TS(D-G) in the
2
L system but rather central. On the one hand, the Cu-C bond
3
Cu system is rather unusual and
3
2
2
-
2
is much more elongated (D 2.016, E 1.979, TS(E-H) 2.413,
TS(D-G) 2.129 Å) and the forming C=C bond is shorter
(TS(E-H) 1.384, TS(D-G) 1.407 Å). On the other hand, the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
bulk in the substrate and the ligands as the phosphine
substituents point towards the approaching olefin (see Figure
3).
3
forming Cu-F bond is much longer in the L system (TS(E-H)
2.390, TS(D-G) 2.093 Å). Substantial charge separation
occurs in both TS, (see Table 2), with the natural population
Insertion of 2a into the Cu-H bond is irreversible, as the
reverse elimination barriers are > 58 kcal/mol. From the
resting state E (-50.4 kcal/mol), F-elimination can again
-
analysis (NPA) charge on the eliminating F exceeding 0.6 e ;
charge separation is somewhat more extensive in the L
H) and leads to a slightly higher dipole moment (5.6 D in L
vs. 6.1 D in L ).
3
TS(E-
proceed, after phosphine decoordination, via L
2
TS(D-G) (-
2
1
0.3 kcal/mol) or, interestingly, also directly from the L
3
3
species (-13.9 kcal/mol) via TS(E-H). The latter one is
preferred for formation of 2b. The lowest F-elimination
6
ACS Paragon Plus Environment

Page 7 of 19
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Potential energy surface (PES) for HDF of tetrafluoroethylene (2a) with L
in maroon, dimeric (L Cu-H) pathway in grey. CIP = contact ion pair. Level of theory TPSSh-D0(PCM)/TZ// TPSSTPSS(PCM)/DZ. T =
23 K. p = 0.1 bar. Solvent = benzene. Gibbs free energies in kcal/mol. PES after the high energy TS12-14 on the H-addition pathway in
3 2 3 2 3
/L Cu-H. L = PMe . L pathway in blue, L pathway
2
2
3
2
the L system not shown. Only one symbolic phosphine ligand shown for monomeric vs. dimeric competition for clarity.
The extensive charge delocalization in the TS explains the
experimentally observed rate dependence on the solvent.
Wiberg bond indices (WBI)33,34 support the notion that TS(E-
H) is central in the L
the Cu-F, Cu-C and F-C bonds are < 0.24, in all cases lower
than in the corresponding L
3
system and less covalent, as WBIs for
2
system TS(D-G). In contrast to
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 19
L
2
, where elimination proceeds to RS G with copper-
coordinated alkene, elimination in the L system proceeds
directly to L Cu-F (H) and the separated olefin 2b.
Decoordination of the olefin via TS(G-I) leads to L Cu-F I,
which can be further stabilized via coordination of a third
phosphine to form H. Overall, F-elimination in both the L and
system is reminiscent of the H-addition pathway in the L
system, which similarly shows extensive charge separation.
In conclusion, in the PMe
preferentially via a three-step nucleophilic H-addition/Cu-C
bond formation/F-elimination sequence either in an L ligand
environment along all steps or initially via a bimetallic L
environment. These pathways offer the possibility for lower
barriers compared to the traditional monomeric L insertion
3
/Cu system, HDF proceeds
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
3
3
2
2
2
2
L
3
3
mechanism. Interestingly, the mechanism described here for
FAs differs distinctively from the concerted SBM described by
the group of Zhang for fluorinated arenes using a similar
2
5
phosphine-stabilized copper hydride system. We could not
locate such TS for alkene systems. This presents a distinct
difference from early transition metal systems, as in the case
of Ti-catalyzed HDF, SBM TS could be found for both alkene
Table 2. NPA charges (q) on fragments and Wiberg bond indices
(WBI) in L
2
TS(D-G) and L
3
TS(E-H) Elimination TS.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
q (e-)
WBI
7
e,8a
and arene HDF systems,
relevant in the latter case.
although they are only kinetically
Fragment
Cu
L
2
L
3
Bond
Cu-F
Cu-C
F-C
L
2
L
3
-0.02
-0.02
-0.62
-0.24
-0.05
-0.67
0.28
0.34
0.37
1.38
0.15
0.24
0.23
1.5
Hydrometallation Pathway Competition. Substrate and
Phosphine Influence. The model system indicates that there
are (at least) three possible mechanistic pathways with low
barriers for hydrometallation in present phosphine copper
hydride systems and that the choice of phosphine ligand and
FA substrate determines which of these pathways will be
preferred.
F
2
C=CFH
F
C=C
The first coordination sphere of copper in the F-elimination
TS(E-H) can essentially be described as five-coordinate in the
system, consisting of three phosphines, a FA and a fluoride
Figure 5 shows trends in HM and F-Elimination barriers for
FAs 2a, 2b and 2e and L = PMe
relevant RS and TS). Barriers for hydrometallation via H-
addition/Cu-C bond formation (L pathway or dimeric
pathway) or insertion (L pathway) increase with decreasing
3
(Table S8-S10) shows all
L
3
1
0
ligand. However, Cu d does not possess the ability to engage
in sp d-hybridization required for trigonal-bipyramidal binding
of five ligands. The geometry of the P Cu fragment barely
3
changes from the resting state E; P-Cu-P angles are 113(3)°for
RS and TS, WBIs for the Cu-P bonds in TS(E-H) are with
3
3
2
fluorine content of the FA; i.e. no ‘runaway’ defluorination
can occur. Due to the stabilization of the corresponding
anions, hydrometallation in the bimetallic copper environment
via H-addition/Cu-C carbon bond formation is the preferred
mechanism for the highly fluorinated FAs 2a and 2b, and
likely others like 1a; hydrometallation via phosphine
decoordination/insertion is preferred for more electron-rich
FAs with lower fluorine content like 2e. The preference for H-
0
0
.72(1) even higher than in the L
.64(1). Therefore, it appears that the system is better
2
system TS(D-G) with
described as tetrahedral, with one coordination site being
simultaneously occupied by both a fluoride and the alkene in
the F-elimination TS (Figure 4). Both the HOMO and the
HOMO-1 are concentrated on the Cu, F and the alkene, which
supports this notion. The H-addition TS in the L system is
2
substantially higher in energy, as it only benefits from
stabilization of the partial cationic charge by two phosphines,
addition via the dimeric over the L
decreasing fluorine content.
3
pathway diminishes with
not three as in the L
3
system.
F
F
F
F
F
F
F
F
F
H
H
2a
2b
H
2e
HYDROMETALLATION1
L3
L2
7.4
10.9
13.4
12.6
(Coord. TS)
12.8
(Coord. TS)
>12.9
(Insertion)
L2-dimer
3.2
8.3
13.2
F-ELIMINATION2
19.8 (trans)
L3
L2
36.5
12.1
20.0 (cis)
40.1
25.1 (cis)
17.4
25.2 (trans)
Figure 4. Top: HOMO (left) and HOMO-1 (right) of the L
elimination TS(E-H). Bottom: tetrahedral vs. trigonal-
bipyramidal geometry in the L Cu(F, alkene) F-elimination TS. L
PMe
3
F-
Figure 5. Trends in hydrometallation and F-elimination barrier
heights for 2a, 2b and 2e in the L and L ligand environment. L =
from the isolated reactants L CuH and FA. from the
3
2
3
=
3
.
1
2
PMe
3
.
3
L
3
CuR resting state. Level of theory TPSSh-D0(PCM)/TZ//
8
ACS Paragon Plus Environment

Page 9 of 19
Journal of the American Chemical Society
TPSSTPSS(PCM)/DZ. T = 323 K. p = 0.1 bar. Solvent = benzene.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Gibbs free energies in kcal/mol.
Table 3. Barriers for FHC=CFH elimination in various L
x
Cu-
CFHCF H systems. Level of theory TPSSh-D0(PCM)/TZ//
2
DFT predicts that the copper hydride dimer is only favored
TPSSTPSS(PCM)/DZ. Solvent = benzene. T = 323K, p = 0.1 bar.
Energies in kcal/mol.
slightly (-0.3 kcal/mol) if L = PMePh
is preferred for PMePh /2a by 2 kcal/mol over the bimetallic
mechanism and 8.4 kcal/mol over the phosphine
decoordination/hydrometallation L pathway. If L = PPh , then
the H-addition L pathway for 2a is favored by 13.4 kcal/mol
over the L
for an L
2 3
and that the L pathway
2
‡
‡
cis Δ Gelim trans Δ Gelim ΔGElim (cis/trans)
2
3
3
L
2
systems
28.7
2
insertion pathway. Phosphine ligands that allow
environment can subsequently enable HDF in
dppe*
28.3
25.6
23.8
26.8
28.1
-9.6/-8.9
-9.2/-8.9
-14.4/-13.8
-8.9/-8.3
-3.1/-2.5
3
systems where ligands enforcing a traditional L
fail.
2
mechanism
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Xantphos*
26.8
PPh
PMePh
PMe Ph
3
26.1
F-elimination Pathway Competition. Substrate and Phosphine
Influence. F-elimination becomes easier the lower the fluorine
content of the forming FA is and is exergonic for both, 2b and
2
26.8
2
27.9
2
e (Figure 5). The L
all three FAs by 4-5 kcal/mol. In the case of L = PPh
were able to locate the L -F elimination TS for 2a (-1.8
kcal/mol vs. L ), however, all optimization attempts for 2b
3
elimination mechanism is preferred for
L
3
systems
20.8
3
, we
3
PMePh2
PMe Ph
22.8
20.4
-8.9/-8.3
-3.1/-2.5
2
2
19.0
lead to decoordination of one phosphine (see Supporting
Information for an example structure for one TS with
decoordinated phosphine). It appears that only phosphines
* Resting state is assumed to be L(PPh )Cu-CFHCF H, L =
3
2
3
6
dppe or Xantphos.
3
5
with a small cone angle (Tolman cone angles: PMe
PMePh 122°, PMe Ph 136°, PPh 145°) can allow for this TS
also in cases where the alkyl fragment is less stabilized.
Unlike the ‘dagger-like’ L H-addition TS, the F-elimination
3
118°,
Experimental trends, i.e. the preference for the cis isomer
2
2
3
2
c for ligands where F-elimination must proceed in the L
environment and for trans isomer 2d for systems where it can
proceed in the L environment are very well reproduced. The
2
3
3
TS needs significant space around the central metal to
accommodate both the FA product and the fluoride ligand.
model also correctly predicts that the F-elimination barrier is
smaller in bidentate systems with larger bite angle (Xantphos
Ligand induced HDF selectivity differences. One of the
striking observations in the present systems is the ability to
tune the cis/trans product selectivity in the HDF of 2b via the
choice of the ancillary phosphine ligand. Stryker has noted
similar observations in phosphine-stabilized copper(I) hydride
mediated carbonyl hydrogenation, pointing out that the
‘structure, reactivity and selectivity of copper(I) hydride
catalysts are a complex and subtle function of the ancillary
ligand’ and that those catalysts show an ‘exceptional
vs. dppe). Product 2c (cis) is preferred by 0.9 kcal/mol over 2d
37
(
trans), but clearly the ligand environment can modulate the
product distribution in the present copper hydride systems,
from enhanced cis preference all the way to a switch to trans
preference.
We considered that the pronounced shift in selectivity
could stem from short H-F contacts in the TS, as the trans TS
often has more short contacts of this type. This has been
postulated to lead to stabilizing TS interactions in other
2
0
sensitivity … to structural variation in the ancillary ligand’.
38,39
40,41
systems,
but NBO
does not locate any significant
We computationally tested the catalyst selectivity for several
phosphines (Table 3), i.e. for the monodentate phosphines
PPh , PMePh and PMe Ph, and two bidentate phosphines
3 2 2
with varying bite angle, dppe and Xantphos.
interactions between H-F in these systems (WBI < 0.002).
Steric interactions could conceivably be responsible for the
differences in the L
observations, a cis preference would be expected in this case.
The distribution of steric bulk in L systems as indicated by
2
systems and, in line with experimental
The experimentally observed selectivity for Xantphos
2
(
2c/2d, cis/trans 7:1 = ΔΔG‡50°C = 1.4 kcal/mol) is nicely
the topographic steric maps shown in Figure 6 leads to shorter
ligand-FA contacts and thus increased steric repulsion in the
trans TS (Table 4).
reproduced by DFT (1.2 kcal/mol), but the error for dppe is
somewhat larger (predicted cis preference 0.4 kcal/mol,
experimentally observed ΔΔG 50°C =1.4 kcal/mol, see SI). For
‡
3 3
In the more trans selective system L , pseudo C -symmetric
PMePh
pathway are possible. DFT predicts that the L
preferred by 6-9 kcal/mol. Moreover, while F-elimination in
the L system is predicted to yield a 1:1 ratio of cis and trans,
only the L system is in line with the experimentally observed
preference for trans elimination. The experimental selectivity
2
and PMe
2
Ph, both, the L
2
pathway and the L
3
distribution of the phosphines equally distributes the steric
bulk, offering no possibility for either of the two FHC=CFH
isomers to avoid it (Figure 6). Unfavorable dipole-dipole
3
pathway is
2
x
interactions between the FA and L CuF can only play a role in
3
the cis TSs (Figure 7) leading to 2c; this is due to the relative
orientation of the two dipoles in the cis TSs and the lack of a
dipole for 2d. Interestingly, the trans preference in the
phosphine systems follows qualitatively Tolman’s electronic
parameter (Table 4), which point to electronic effects being
responsible for the switch.
‡
for PMePh
kcal/mol) appears to be somewhat overestimated (DFT: 1.5-
.0 kcal/mol). However, repeated product re-insertion into the
2 2
and PMe Ph (cis/trans 1:4 = ΔΔG 50°C = 0.9
2
Cu-F bond could diminish the kinetic trans preference of these
catalysts somewhat, if regeneration of the catalyst is not fast
enough. The possibility for re-insertion of FAs in the Cu-F
bond could be successfully demonstrated by the experimental
observation of 7b’.
Although the unusual cis preference for the two isomers
of FHC=CFH has been the subject of considerable research
efforts in the past, it appears that no definitive conclusion on
3
9-46
its origin has been reached, yet.
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 19
the preference for the thermodynamic product; this is not
L3
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
observed for 2c/2d. Table 4 shows that the energy difference
between the FA fragments in the cis/trans TS favors the trans
isomer in all cases, indicating that the electronic stabilization
of cis-1,2-difluoroethene occurs very late and that initially
only the steric repulsion by the F atoms prevails. It appears
F
F
H
H
H
F
F
H
2
that within the L pathway, FA deformation and unfavorable
L2
X
dipole-dipole interactions in the cis TS favor the trans isomer,
but this is more than compensated for by unfavorable steric
interactions in the trans TS. L2 leads preferentially to
formation of the cis isomer 2c. An increase in the FA-Cu
F
H
F
X
H
F
F
L
F
F
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
H
H
3
distance in the L environment weakens unfavorable dipole-
H
L
dipole interactions in the cis TS and the symmetric distribution
of steric bulk lessens steric differences. Subsequently, the FA
deformation dominates, which leads to a preference for the
trans isomer 2d.
Figure 6. Symmetric distribution of steric bulk in L
origin of cis-preference in L systems (L = PMe
steric bulk generated with SambVca 2.0 (webtool for analyzing
3
systems and
3
) and map of
2
To the contrary, increased selectivity for the thermodynamic
product upon L /L switch is indeed observed experimentally
2 3
in the HDF of the bulky α-trifluoromethyl styrene (8a). Use of
the small cone angle phosphite P(OEt)3 increases the
selectivity for the thermodynamic product Z-β-fluoro-α-methyl
4
2
catalytic pockets) from L
3
Cu-H; sphere radius of 3.5 Å.
F F
H F
H
F
2 3
styrene (8c) from 2.8:1 (Xantphos/PMePh ) to 7.2:1 [P(OEt) ]
L
L
‡
F
3 2 3
(DFT, L = P(OMe) : ΔΔG L 1.3 kcal/mol, L 2.3 kcal/mol).
F
0
No HDF is observed for 2b with this ligand class. Phosphites
are electron-poor and DFT indicates that F-elimination in such
a system could lead to F/OR exchange on the organophosphite
H
L
H
L
5
3
ligand during HDF of 2a.
Figure 7. Unfavorable dipole-dipole interactions in L
to 2c (left) vs lack of dipole-dipole interactions in L
to 2d (right). L
3
TS leading
TS leading
Experimental Mechanistic Studies. To confirm these DFT
revelations, mechanistic studies were carried out. As the
L CuH systems like Xantphos presumably follow the
2
previously reported insertion-elimination mechanism, as
3
2
Cu-F fragment dipole orientation in maroon, FA
dipole orientation in green.
supported by DFT calculations (vide supra), mechanistic
Table 4. Ligand dependence of the preference for the cis
studies were focused on the L
PMePh
3
CuH HDF mechanism using
FHC=CFH isomer in different L
electronic parameter ν and differences between the shortest
ligand-FA contacts (lcis-trans) in the L ligand environment and
2 3
and L systems, Tolman’s
2
.
2
Control experiments for the HDF of trifluoroethylene, (2b),
confirmed that all three components of the reaction mixture
are required. The product ratio of the HDF of 2b using
deformation energy difference of the FA fragments in the TS
(EDef. cis-trans).
2
PMePh is altered (2c:2d, 1:1) when no silane is present; no
ΔΔG‡
ν
L
E
Def. cis-trans
cis-trans
such alteration is observed for Xanthphos. This experiment
suggests that re-insertion of 2c/2d into Cu-F occurs if no
silane is present as noted above for TFE (2a). Similarly,
TMDS does not effect HDF at RT without a source of copper.
Nor does the mixture of phosphine and TMDS, confirming
that the copper plays a crucial role in this HDF reaction.
However, when the reaction was attempted with varying
concentrations of silane; no rate dependence was observed. As
a significant rate increase with increasing solvent polarity was
observed, the silane substituents were varied to elaborate
further on its role. Interestingly, the reaction only progresses
L
2
L
3
Ph
3
P
-2.5
-1.2
-0.8
-0.2
0.0
2068.9
0.137
0.066
0.015
-2.2
-0.4
-1.2
Xantphos
dppe
Ph
PhMe
Me
2
MeP
2.0 2065.1 -0.023
1.5 2065.3 -0.025
0.4 2064.1 -0.048
-0.9/-1.0
-0.7/-0.3
-0.6/-1.9
2
P
3
P
0.1
smoothly at RT when either triphenylsilane (Ph
3
SiH) or
Nonetheless, most authors appear to agree that the
preference is of electronic origin.
TMDS are used. With triethoxysilane [(EtO) SiH] the reaction
3
4
0,47
In the absence of a
turns over once at RT and then ceases operation. If heated to
45 °C, the reaction progresses with some loss of selectivity.
This is reminiscent of Lentz’s observations in Ti(III) complex-
definitive consensus on the origin of the cis preference, we
deem a quantitative analysis of the trends here impossible.
However, tentatively the preference switch can be explained as
follows: The Hammond postulate assumes that electronic and
steric changes from the reactants to the TS and further to the
catalyzed HDF, where (EtO)
deactivation via formation of alkoxytitanium compounds.
3
SiH led to fast catalyst
2
9
The alkyl silanes either require heating (i.e. triethylsilane) or
are ineffective (i.e. triisopropylsilane). Interestingly,
diphenylsilane also does not work at room temperature, and
requires heating to 70 °C but the reaction mixture quickly
becomes black. The most productive silanes are thus those
with somewhat more electron-withdrawing functional groups.
4
8,49
products occur gradually.
central with respect to the overall TS geometry, they are later
than L type TS with respect to the forming olefin, as the
forming C=C bond is much shorter (see also Figure 2). A shift
from the L to the L environment should therefore increase
3
While TS of the L type are
2
2
3
1
0
ACS Paragon Plus Environment

Page 11 of 19
Journal of the American Chemical Society
Based
on
CuH)
the
stoichiometric
reactions
of
Xantphos ligands. It would be expected, however, that
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
PMePh /(PPh
/2b, a fast equilibrium between the
accessing analogues of TS(A-F), depends on a delicate
2
3
6
+
resulting products is suspected. In line with observations (see
above), this would lead to an increased 2c/2d ratio tending
towards the thermodynamic distribution. On this note, as the
rate of Cu-F to Cu-H conversion becomes slow, a secondary
reaction becomes competitive, leading eventually to complete
catalyst deactivation via formation of a copper(I)-alkenyl
complex by formal HF elimination (Scheme 9). Additionally,
formation of 1,1,1,2-tetrafluoroethane (2b’) suggests that the
generated Cu-alkyl reacts quickly with HF, explaining why it
is not observed over the reaction course.
balance of the stabilization of Cu and the carbanion (Alk). To
explore this idea, 1,1-difluoroethylene (2e) was used, as the
Xantphos catalyst cannot hydrodefluorinate this substrate
(Scheme 10, bottom). Interestingly, use of P(OEt
3
) also
afforded no HDF product from 2e. In contrast, the PMePh
2
Cu-H catalyst led to efficient HDF of 2e to a 3:1 mixture of
vinyl fluoride (2f) and ethylene (Scheme 10, Top) although
calculations show that this substrate could be hydrometallated
via both mechanisms (Figure 5). As such, a larger cone-angle,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
stronger σ-donating phosphine ligand, PCp , was also tested
with no HDF product detected. Again, because ligand size is
To probe the elementary steps further, the source of hydride
was confirmed using the Si-D reagent. When the reaction was
carried out with stoichiometric amounts of Cu-H and a slight
excess of triethylsilyl deuteride, the major product contained
mostly protons, confirming that Cu-H adds to the FA, and that
the silane’s role is likely restricted to Cu-H regeneration.
+
so important, PMePh
thus balancing the destabilized carbanion, H
CuH mechanism.
2
likely leads to greater Cu stabilization,
-
2
CR , in the
L
3
Scheme 10. Hydrodefluorination of 1,1-difluoroethylene (2e)
These results suggest that an electronic limit to the HDF of
FAs via the L
3
CuH mechanism exists (i.e., in cases where the
+
-
stabilization of the Cu is not great enough to offset Alk
Scheme 9. Reaction of copper hydride with 2b leading to
formation of a copper trifluorovinyl complex and R-134a (2b’).
destabilization). Therefore, alkyl-substituted FAs were tested
next (Scheme 11).
In this case, 1,1,1-trifluoropropene (9) was subjected to
HDF to generate 1f in situ (Eq. 11.1). As expected, no further
Effects controlling selectivity for the L
mechanism. With the DFT and experimental support for the
CuH mechanism, other FAs were examined to determine the
effects of FA substitution on the ability to access the L CuH
mechanism. For consistency with reactions using the Xantphos
catalyst, these reactions were carried out in benzene at 45 °C.
First, mono-substituted FAs were explored to determine if β-
fluoride elimination is crucial in accessing analogs of TS(E-
H). As such, FAs XFC=CF
were used. Unfortunately, when iodotrifluoroethylene, 3b, was
treated with stoichiometric copper hydride, HFC=CF , 2b, was
observed as the major product, independent of the choice of
ligand. It is suspected that this occurs via σ-bond metathesis of
the C-I bond. However, with chlorotrifluoroethylene (3a),
when using Xantphos, the observed product was a mixture of
cis-/trans-1,2,-difluoroethylene, 2c/2d. This result suggests
that the first HDF step produces 1-chloro-1,2-difluoroethylene
3a’) arising from addition of the hydride to the CF carbon.
2
This new alkene is considerably more reactive than 3a, such
that addition of Cu-H to 3a’ dominates. Consistent with this
2
result, PMePh yields the same product distribution as that
observed with Xantphos, indicating that β-fluoride elimination
is necessary to access analogues of TS(E-H). To confirm this,
the HDF of 4a was attempted with PMePh
2c,d were produced in the same ratio as the Xantphos reaction
(vide supra). β-X elimination (X = Cl or OCF ), excluding
fluoride, generally proceeds through the known analogous
transition states of TS(D-G).
2 3
CuH or L CuH
HDF was observed using the Xantphos or PMePh
2
system. On
addition of a stabilizing element to Alk , HDF through the
CuH mechanism should proceed. When α-trifluoromethyl
styrene (8a) was used to generate β-difluoro-α-methyl styrene
8b) in situ, Xantphos and PMePh yielded the same product
L
3
-
3
L
3
(
2
distribution of E/Z- β-fluoro-α-methyl styrene (8c,d) in a ratio
of 2.8:1 (Eq. 11.2). The steric congestion arising from
addition, with the =CPhMe carbon bonded to copper, now
2 3
[X = Cl (3a), I (3b) or OCF (4a)]
proves too much to accommodate the third PMePh
2
ligand
2
3 2
required for the L mechanism. As such, the L mechanism
dominates
(
2 3
and P(OEt) and
3
Scheme 11. Hydrodefluorination of allyl trifluoropropene (top)
and α-trifluoromethyl styrene (bottom)
On another note, the low activation energy for the L
hydride addition mechanism could explain why P(OEt)
competitive in efficiency in the HDF of 1a with the dppf or
3
CuH
3
is
1
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 19
for control of the product ratio in both cases. Nevertheless, if Dimethylformamide (DMF, Alfa Aesar Anhydrous 99.8%),
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
our hypothesis is correct, even the weaker σ-donating but
smaller ligand should function, due to the Alk benzylic
hexafluoropropene (1a, Synquest 98.5%), trifluoroethylene
(2b, Synquest 98%), chloro trifluoroethylene (3a, Synquest,
99%), iodo trifluoroethylene (3b, Synquest, 97%),
trifluoromethyl trifluorovinyl ether (4a, Synquest, 99%),
vinylidene difluoride (2e, Arkema Inc., 99%), alpha-
-
stabilization. Indeed when P(OEt)
b and on to 8c,d proceeded smoothly to provide an increased
product ratio of 7.2:1 (Eq. 11.3). Apparently, the L
3
was used, the HDF of 8a to
8
3
mechanism reinforces the E/Z ratio as 8c is both the least polar
trifluoromethyl
styrene
(8a,
Synquest,
97
%),
and most thermodynamically stable, as confirmed by DFT
hexafluorocyclobutene
(5a,
Synquest
98%),
3,3,3-
(vide supra).
trifluoropropene (9, Synquest, 99%), perfluoro(4-methyl-2-
pentene) (6b, Synquest, 95%), perfluoro(2-methyl-2-pentene)
(
6a, Synquest, 96%), triethylesilane (Sigma-Aldrich, 99%),
CONCLUSION
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
triethyl(silane-d) (Sigma-Aldrich, 97 atom D),
%
In summary, new P-ligated copper hydride complex-catalyzed
routes for the HDF of fluoroalkenes using silanes have been
developed. With our technology, hexafluoropropene (1a) can
be selectively converted to fourth-generation refrigerant
triisopropylsilane (Oakwood chemicals, 98%), triphenylsilane
(Oakwood chemicals, 97%), triethoxysilane (Sigma-Aldrich,
95%),
diphenylsilane
(Oakwood
chemicals,
97%),
tetramethyldisiloxane (TMDS, Sigma-Aldrich, 97 %),
poly(methyl-hydrosiloxane) (PMHS, Sigma-Aldrich, average
CH
dppe)Cu-H system stops at the double HDF, giving the
highest yield of 1d, small cone-angle, less electron-rich
P(OEt) allows for the quadruple HDF of 1a to 1f. On
evaluating the HDF substrate scope we found that use of a
smaller cone-angle, electron-rich ligand such as PMePh leads
to L Cu-H reactivity with electron-poor or bulky fluoroalkenes
not observed with the L CuH systems. DFT studies revealed
2 3
=CF(CF ) (1d) or to 1,1-difluoropropene (1f). While the
(
M
n
:
1,700-3,200), triphenylphosphine (PPh
chemicals, 99%), tri(o-tolyl)phosphine (P(o-tolyl)
Aesar, 98%), 2-di-tert-butylphosphino-2′,4′,6′-
3
,
Oakwood
3
,
Alfa
3
triisopropylbiphenyl (tbuXphos, Sigma-Aldrich, 97%), 1,2-
bis(diphenylphosphino) ethane (dppe, Strem chemicals, 99%),
2
3
1
,1’-bis(diphenylphosphino) ferrocene (dppf, Accela ChemBio
2
Inc., 99%), 4,5-bis(diphenylphosphino)-9,9-dimethylxanthene
(Xantphos, Accela ChemBio Inc., 99%), triethylphosphite
an exceptional, ligand- and substrate-dependent mechanistic
flexibility in these copper(I) hydride systems ranging from
monomeric insertion/elimination to bimetallic or monomeric
hydride transfer mechanisms. In the latter, F-elimination
involves a doubly-occupied coordination site at copper. Both
(P(OEt)
Alfa Aesar, 97%), tri-ortho-tolyl phosphite (P(O-o-tolyl)
Alfa Aesar), methyldiphenylphosphine (PMePh
Organics, 99%), dimethylphenylphosphine (PMe Ph, Acros
Organics, 97%). The following chemicals were synthesized as
3
, Sigma-Aldrich, 98%), triphenylphosphite (P(OPh)
3
,
3
,
2
, Acros
2
2 3
hydrometallation and F-elimination in the L as well as the L
system proceed via TSs showing significant charge separation,
explaining the experimentally observed rate dependence on
the solvent. DFT results are nicely in line with experimental
observations. We propose that the phosphine choice can
modulate the position of the F-elimination TS with respect to
the product and thereby modulate product selectivity, as
experimentally observed for 2b and 8a. It appears that in the
case of 8a, tuning the position of the F-elimination TS
reinforces the preference for the thermodynamic product. In
previously
diisopropylphenyl)imidazolidin-2-ylidene
reported:
1,3-bis(2,6-
(SIPr)55
and
56
[(PPh
3
)CuH]
6
.
Tetrafluoroethylene (2a) was prepared by
pyrolysis of polytetrafluoroethylene (Scientific Polymer
Products, powdered) under vacuum, using a slightly modified
literature procedure [10-20 mTorr, 650 °C, 30 g scale, product
stabilized with R(+)-limonene (Aldrich, 97%), giving 2a of ca.
57
9
7% purity] or by pyrolysis of KO
2
19
CCF
2
31
CF
3
under vacuum,
58 1
1
13
1
producing TFE Safe Supply®. H, F, P{ H}, and C{ H}
NMR spectra were recorded on a 300 MHz Bruker Avance
instrument at room-temperature (21-23 °C) unless stated
2 3
the case of 2b, the switch from L to L environment leads to a
switch from thermodynamic to kinetic product. We believe
that this effect can be traced to a breakdown of the assumption
that steric and electronic changes occur gradually from RS to
TS to product in this specific case.
1
otherwise. H NMR spectra were referenced to residual proton
6 6
peaks associated with the deuterated solvents (C D : 7.16
19
ppm). F NMR spectra were referenced to internal standard
With tetrafluoroethylene both (P-P)CuH and
P
3
CuH
α,α,α-trifluorotoluene (CF Ph) [unless stated otherwise]
3
complexes (7a,b) could be successfully isolated and are
(Sigma-Aldrich, 99%, deoxygenated by purging with nitrogen,
currently being assessed for the introduction of the 1,1,2,2-
stored over activated 4 Å molecular sieves), set to – 63.5 ppm.
P{ H} NMR data were referenced to external H PO (85 %
3 4
5
4
31
1
tetrafluoroethyl (-CF
2
CF
2
H) moiety to organic electrophiles.
aqueous solution), set to 0.0 pm. Electrospray ionization mass
spectral data were collected using an Applied Biosystem
API2000 triple quadrupole mass spectrometer. Elemental
analyses were performed by Ján Veizer Stable Isotope
Laboratory, University of Ottawa (Ottawa, Ontario, Canada).
Note that the NMR spectra ( H, F, F{ H}, and P{ H} for
the title compounds are displayed at the end of the Supporting
Information (Figures S4-44).
EXPERIMENTAL SECTION
General Procedures. Experiments were conducted under
nitrogen, using Schlenk techniques or an MBraun glove box.
All solvents were deoxygenated by purging with nitrogen.
Tetrahydrofuran (THF) and toluene were dried on columns of
activated alumina using a J. C. Meyer (formerly Glass
Contour®) solvent purification system. Chlorobenzene, m-
dimethoxybenzene, benzene, cyclopentyl methyl ether
1
19
19
1
31
1
General
Hydrodefluorination of Gaseous Fluoroalkenes, NMR
Scale. [(PPh )CuH] (5 mg, 0.02 mmol, 10 mol %) was placed
in a 7″ nmr tube and mixed with 400 μL of solvent. A ligand
31 mol % or 11 mol %) and silane (10 equiv) were added.
Experimental
Procedure
for
6 6 6
(CPME) and benzene-d (C D ) were dried by stirring over
activated alumina (ca. 10 wt. %) overnight, followed by
filtration. All solvents were stored over activated (heated at ca.
250 °C for >10 h under vacuum) 4 Å molecular sieves.
Glassware was oven-dried at 120 °C for >2 h. The following
chemicals were obtained commercially, as indicated:
3
6
(
The tube was capped with a rubber septum, removed from the
glovebox, further sealed by tightly wrapping the cap with a
1
2
ACS Paragon Plus Environment

Page 13 of 19
Journal of the American Chemical Society
strip of parafilm and shaken to ensure a homogeneous
solution. 3 mL of gaseous fluoroalkene were then added to the
reaction via air-tight syringe. The reaction was monitored by
removed from the glovebox, further sealed by tightly
wrapping the cap with a strip of parafilm and shaken to ensure
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
9
a homogeneous solution. The reaction was monitored by
F
1
9
F NMR over 8 hours at various temperatures. See Figure S4-
NMR over 8 h at various temperatures. See Figure S35-38 for
1
19
19
19
19
3
4 for H NMR and F NMR spectra and Table S1 for
F
F NMR spectra and Table S1 for F NMR data. Compounds
5
9
59
60
61
4
62
62
65
66
66
66
NMR data. Compounds 1b , 1c , 1d , 1e , 1f , 2c , 2d ,
6c , 8b, 8c and 8d were identified using available
6
2
63
64
64
64
2
f , 4b , 5b and 5c , 5d were identified using available
literature data.
Z-3,3,3,2,1-pentafluoropropene (1b). H NMR (300 MHz,
literature data.
1
1
,1,4,4,5,5,5-heptafluoro-2-methylpent-1-ene (6c). H NMR
1
3
(300 MHz, C
CH
6
D
6
): δ 2.13 (t, JHF = 19 Hz, CH
). F NMR (282 MHz, C ): -86.42 (s, CF ), -92.00 (d,
JFF = 44 Hz, 1F), -92.64 (d, JFF = 44 Hz, 1F), -117.07 ppm
2
), 1.22 (m,
2
3
4
19
C
6
D
6
): δ 6.13 (ddq, JHF = 68, JHF = 15, JHF = 1 Hz, =CHF).
3
6
2
D
6
3
1
9
3
4
2
F NMR (282 MHz, C
4JFH = 1 Hz, CF
), -155.07 (ddq, JFH = 68, JFF = 7, JFF = 6
Hz, =CFH), -158.97 ppm (ddq, JFH = 15, JFF = 7, 4JFF = 6 Hz,
6
D
6
): -72.59 (ddq, JFF = 14, JFF = 6,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
4
3
3
2
(tm, JFH = 19 Hz, CF ).
3
3
1
1
,1-difluoro-2-methyl-2-phenylethyl-1-ene (8a): H NMR
=
CF(CF
3
)).
4
4
19
(
300 MHz, C
NMR (282 MHz, C
92.30 ppm (dq, JFF = 44, JFH = 2 Hz).
6
D
6
): δ 1.59 (dd, JHF ≈
J
HF = 3 Hz, Me).
): -91.99 (dq, JFF = 44, JFH = 3 Hz), -
F
1
2
4
E-3,3,3,2,1-pentafluoropropene (1c): H NMR (300 MHz,
6
D
6
2
3
19
2
4
C
6
D
6
): δ 5.93 (dd, JHF = 68, JHF = 4 Hz, = CHF). F NMR
3
4
(282 MHz, C
6
2
D
6
): -70.09 (dd, JFF = 19, JFF = 12 Hz, CF
65.93 (ddq, JFH = 68, JFF = 137, JFF = 19 Hz, =CFH), -
180.12 ppm (ddq,
=CF(CF )).
,3,3,2-tetrafluoropropene (1d): H NMR (300 MHz, C
δ 4.36 (ddq, JHH = 5, JHF = 13, JHF = 1 Hz, =CH2cis), 4.49 (br
dd, JHH = 4, JHF = 52 Hz, =CH2trans). F NMR (282 MHz,
): -73.28 (d, JFF = 10 Hz, CF
2, JFH = 13 Hz, JFF = 10 Hz. =CF(CF
3
), -
1
(
E)-1-fluoro-2-methyl-2-phenylethyl-1-ene (8b): H NMR
3
4
1
2
4
(
2
(
6 6
300 MHz, C D ): δ 7.4 – 6.9 (Ar), 6.58 (dq, JHF = 85, JHH =
2
3
4
J
FH = 4
J
FF = 137,
J
FF = 12 Hz,
4
4
19
Hz, 1H(FHC=)), 1.80 (dd, JHF = 4, JHF = 2 Hz, Me). F NMR
3
2
4
282 MHz, C
6
D
6
): -131.66 ppm (dq, JFF = 85, JFH = 4 Hz).
1
3
6
D
6
):
1
(
Z)-1-fluoro-2-methyl-2-phenylethyl-1-ene (8c): H NMR (300
2
3
3
2
4
6 6
MHz, C D ): δ 7.4 – 6.9 (Ar), 6.22 (dq, JHF = 84, JHH = 2 Hz,
2
3
19
4
4
19
1H(FHC=)), 1.80 (dd, JHF = 5, JHF = 2 Hz, Me). F NMR (282
3
3
C
6
D
6
3
), -124.12 ppm (ddq, JFH
=
2
4
MHz, C
Synthesis of [(PMePh
complex [(PPh )CuH] (3.16 g, 9.69 mmol based on
monomeric unit), PPh Me (6.01 g, 30 mmol) and TMDS (1.4
6
D
6
): -131.66 ppm (dq, JFF = 84, JFH = 5 Hz).
3
3
5
3
)).
2
)
3
Cu(CF CF H)] (7a). The red
2
2
1,1,2-trifluoropropene (1e): 9F NMR (282 MHz, C
1
D
): -
6
6
3
6
1
(
(
06.32 (ddq, JFF = 93, JFF = 32, 4JFH = 5 Hz, =CF2cis), -126.15
ddq, JFF = 93, JFF = 115, JFH = 5 Hz, =CF2trans), -167.19 ppm
ddq, JFF = 115, JFF = 32, JFH = 17 Hz, =CF(CH
2
3
2
2
3
4
mL, 8 mmol or 16 mmol hydride equivalents) were placed in a
350 mL ampule and mixed with 30 mL of benzene. The
reaction vessel was attached via a three-way valve to a 2a:CO
2
canister with a regulator and a Schlenk line. The solution was
degassed using a regular freeze/pump/thaw method. The
3
3
3
3
)).
1
9
1
,1-difluoropropene (1f): F NMR (282 MHz, C
6
D
6
):
-
2
3
4
8
9.67 (ddq, JFF = 50, JFH ≈ JFH = 3 Hz, =CF2trans), -93.45
ppm (ddq, JFF = 50, JFH = 25, 4JFH = 2 Hz, =CF2cis).
2
3
1
2a:CO
set to 5 psi and the reaction mixture stirred at 70 °C. After 10
minutes, the regulator pressure dropped to 0 and more 2a:CO
2
was added to the degassed solution with the regulator
Z-1,2-difluoroethylene (2c): H NMR (300 MHz, C
6
D
6
): δ
1
9
5
6
.54 (m). F NMR (282 MHz, C
6
D
6
): δ -163.09 ppm (m).
1
2
E-1,2-difluoroethylkene (2d): H NMR (300 MHz, C D ): δ
6
6
2
was added. After ~2 h, the solution became clear with some
black precipitate. The solvent was removed in vacuo, leaving a
thick liquid. 10 mL of Et O was added, and the solution was
2
then filtered through a Celite-padded fritted funnel (30 mL
medium pore). The ampule was rinsed 2 more times with 10
2
mL of Et O. All volatiles were removed in vacuo and 50 mL
of methylcyclohexane were added to the solution. Upon
standing, the product crystallized from solution. The white
microcrystalline powder was collected (30 mL medium-pore
fritted funnel), triturated with pentane (2 x 20 mL), and dried
1
9
6 6
.68 (m). F NMR (282 MHz, C D ): -187.73 ppm (dd, JFH
3
=
49, JFH = 30 Hz).
1
Vinyl fluoride (2f): H NMR (300 MHz, C
6
6
D ): δ 6.14 (ddd,
2JHF = 85, JHH = 13, JHH = 5 Hz), 4.54 (ddd, JHF = 20, JHH
3
3
3
2
=
1
2, 3JHH = 3 Hz), 4.00 (ddd, JHF = 54, JHH = 5, JHH = 3 Hz).
3
2
3
1
9
2
3
F NMR (282 MHz, C
6
D
6
): -115.97 ppm (ddd, JFH = 85, JFH
3
=
54, JFH = 20 Hz).
Z-Trifluoromethyl-1,2-difluorovinyl ether (4b): 19F NMR
282 MHz, C ): -54.78 (m, OCF ), -118.28 (m, F), -122.22
ppm (dm, JFH = 59 Hz).
,4,4-trifluorofluorocyclobutene (5b): 19F NMR (282 MHz,
): δ -106.52 (m, 2F), -113.45 ppm (m, 1F).
,2-difluorofluorocyclobutene (5c): 1H NMR (300 MHz,
(
6
D
6
3
2
in vacuo to yield 5.17 g of 7a (6.8 mmol, 69 % based on
1
[(PPh )CuH] ). H NMR (300 MHz, C D ): δ 7.23 (m, 16H,
3
6
6
6
1
2
3
PMePh
2
2
), 6.89 (m, 25H, PMePh ), 6.75 (tt, 1H, JHF = 52, JHF
6 6
C D
19
=
5 Hz, -CF
2
H), 1.55 (br, 12H, PMePh
2
) ppm. F NMR (282
), -124.89 ppm
H). 31P{ H} NMR (121 MHz,
): −20.0 ppm (br, PMePh ). Anal. Calcd for
: C, 64.35, H, 5.27. Found: C, 60.49, H, 5.31.
1
C
3
3
MHz, C ): −96.68 (dt, JFF = JFH = 5 Hz, CF
6
D
6
2
1
9
6 6 6 6
D ): δ 1.67 (m, 2H) ppm. F NMR: (282 MHz, C D ) -
2
3
1
(dt, JFH = 52, JFF = 5 Hz, -CF
2
1
18.57 ppm (m, 4F).
C
C
6
D
6
2
1
-fluorocyclobutene (5d): 19F NMR (282 MHz, C
6
D
6
): -84.23
41
H
40CuF
4
P
3
ppm (m, 1F).
(These values reflect those expected for phosphine oxidation
General
Experimental
Procedure
for
prior to combustion. Anal. Calc. for C41 : C,
4 3 3
H40CuF O P
1
19
31
1
Hydrodefluorination of Non-Gaseous Fluoroalkenes, NMR
Scale.
60.56, H, 4.96.) See Figures S40−S42 for H, F, and P{ H}
NMR spectra.
[
(PPh
nmr tube and mixed with 400 μL of solvent. A ligand (31 mol
or 11 mol %), silane (10 equiv, 1.45 mmol) and
fluoroalkene (0.15 mmol) were added. The tube was capped,
3
)CuH]
6
(5 mg, 0.02 mmol, 10 mol %) was placed in a 7″
Synthesis of [(Xantphos)Cu(CF
2 2
CF H)] (7b). The red
complex [(PPh )CuH] (25 mg, 0.08 mmol based on
3
6
monomeric unit), Xantphos (40 mg, 0.08 mmol) and TMDS (2
μL, 0.01 mmol) were placed in a 7" NMR tube mixed with
%
1
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 19
4
00 μL of C
6
D
6
. The tube was capped with a rubber septum,
phosphine bond was included in the benchmark via the
complex Ni(CN) (PEt , in the solvents dichloroethane and
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
removed from the glovebox, further sealed by tightly
wrapping the cap with a strip of parafilm and shaken to ensure
a homogeneous solution. 3 mL of gaseous 2a was added to the
reaction via air-tight syringe. The reaction was monitored by
3 3
)
2
3
6
ethanol. For this set of 8 bonds, the protocol including
Grimme’s dispersion corrections with no damping produces an
MAD of 1.8 kcal/mol. Becke-Johnson damping leads to an
overestimation of bond dissociation energies and a higher
discrepancy (MAD 4.2 kcal/mol). Finally, in a limited test set,
we included dispersion corrections during optimization via the
1
9
F NMR over 8 hours at various temperatures. Compound 7b
1
9
was identified by comparing to 7a. F NMR (282 MHz,
): −103.9 (br m, CF
), -125.45 ppm (br dm, 2JFH = 52, -
H). P{ H} NMR (121 MHz, C ): −16.74 ppm (br m,
C
CF
6
D
2
6
2
3
1
1
89
6
D
6
use of the B97D functional, but this did not noticeably
1
9
31
1
Xantphos). See Figures S44−S45 for F and P{ H} NMR
spectra.
change the predictions.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ASSOCIATED CONTENT
Computational Details. All geometries were fully
67
Supporting Information. This material is available free of
charge via the Internet at http://pubs.acs.org.
optimized using the Gaussian 09 software package in
combination with an external optimizer (PQS, OPTIMIZE
routine of Baker ) and the BOpt software package.
Following the protocol proposed in ref. , all relevant minima
and transition states were fully optimized at the TPSSTPSS
level of theory employing correlation-consistent polarized
valence double-ζ Dunning (DZ) basis sets with cc-pVDZ
quality
68,69
70
Addition crystallographic information, DFT studies,
1
19
13
1
31
1
7
1
and H, F, C{ H} and P{ H} NMR Spectra.
PDF); optimized Cartesian coordinates in Å for
(
7
2
structures related to Figure 3 (XYZ), structures
related to Figure 5 (XYZ), structures related to Table
3 (XYZ) and method benchmark (XYZ). Animations
7
3,74
from the EMSL basis set exchange library, using a
7
5
of L TS (GIF).
small core pseudo-potential on Cu. The density fitting
approximation (Resolution of Identity, RI)
optimization stage and for single-point energy corrections.
3
7
6-79
was used at the
AUTHOR INFORMATION
Corresponding Author
Solvent effects (benzene, ε = 2.2706) were included with the
polarizable continuum model approach (PCM) at both stages.80
*
*
E-mail for R.T.B.: rbaker@uottawa.ca
E-mail for C.E.: christian.ehm@unina.it
All calculations were performed at the standard Gaussian 09
SCF convergence using an ultrafine grid [Scf=Tight and
Int(Grid=ultrafine)]. The nature of each stationary point was
checked with an analytical second-derivative calculation (no
imaginary frequency for minima, exactly one imaginary
frequency for transition states, corresponding to the reaction
coordinate). The accuracy of the TS was confirmed with an
IRC scan on preliminary gas phase calculations. Transition
states were located using a suitable guess and the Berny
algorithm (Opt=TS) or a relaxed potential energy scan to
arrive at a suitable transition-state guess, followed by a quasi-
Newton or eigenvector-following algorithm to complete the
optimization.
Author Contributions
The manuscript was written through contributions of all
authors. / All authors have given approval to the final version
of the manuscript.
ACKNOWLEDGMENT
8
1
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. NOA gratefully acknowledges support
from the province of Ontario, NSERC and the University of
Ottawa (OGS and CGS-M/D).
Final single-point energies were calculated at the TPSSh level
82
of theory employing triple-ζ Dunning (TZ) basis sets (cc-
pVTZ quality).73 Grimme dispersion corrections without
damping (keyword -zero) were added at this stage using the
REFERENCES
8
3,84
standalone dftd3 program.
Enthalpies and Gibbs free
(1)
New routes to fluoroalkenes: (a) Zheng, J.; Lin, J.-
energies were then obtained from TZ single-point energies and
thermal corrections from the TPSSTPSS(PCM)/cc-pVDZ-(PP)
vibrational analyses; entropy corrections were scaled by a
factor of 0.67 to account for decreased entropy in the
condensed phase.
SambVca 2.0 program. Maps of steric bulk were generated
using the same program. NBO 3.1 was used for NBO
H.; Yu, L.-Y.; Wei, Y.; Zheng, X.; Xiao, J.-C. Cross-Coupling
between Difluorocarbene and Carbene-Derived Intermediates
Generated from Diazocompounds for the Synthesis of gem-
Difluoroolefins. Org. Lett. 2015, 17, 6150-6153. (b) Zhang,
W.; Wang, Y. Recent Advances in Carbon-Difluoroalkylation
and -Difluoroolefination with Difluorocarbene. Tetrahedron
Lett. 2018, 59, 1301-1308. (c) Okoromoba, O. E.; Han, J.;
Hammond, G. B., Xu, B. C. Designer HF-Based Fluorination
Reagent: Highly Regioselective Synthesis of Fluoroalkenes
and gem-Difluoromethylene Compounds from Alkynes. J. Am.
Chem. Soc. 2014, 136, 14381-14384. (d) Vandamme, M.;
Paquin, J.-F. Eliminative Deoxofluorination Using XtalFluor-
8
5-87
Δ%VBur was calculated using the
4
3
88
analysis.
The correct prediction of copper-phosphine bond strengths is
crucial for the competition of mechanisms described in this
paper. Metal-phosphine bond strengths are known to be
challenging to predict and dispersion corrections are critically
needed for their accurate description. To our knowledge, no
accurate experimental data is available for Cu-P bonds. In the
absence of this data, we decided to benchmark the protocol
against available gas phase bond dissociation energies of
3
6
E:
Cyclohexanone Derivatives. Org. Lett. 2017, 19, 3604-3607.
e) Yang, M.-H.; Matikonda, S. S.; Altman, R. A. Preparation
A One-Step Synthesis of Monofluoroalkenes from
(
of Fluoroalkenes via the Shapiro Reaction: Direct Access to
Fluorinated Peptidomimetics. Org. Lett. 2013, 15, 3894-3897.
(f) Sakaguchi, H.; Uetake, Y.; Ohashi, M.; Niwa, T.; Ogoshi,
3
1
copper complexes with labile binding of ligands, i.e. Cu(0)-
NH
Cu (ethene)
9
0
+
91
st
nd
3
+
,
Cu CO and the 1 and 2 binding energies in
92
+
93
2
and Cu (acetonitrile)
4
.
A
labile metal
S.;
Hosoya,
T.
Copper-Catalyzed
Regioselective
1
4
ACS Paragon Plus Environment

Page 15 of 19
Journal of the American Chemical Society
Transition-Metal-Free Catalytic Hydrodefluorination of
Monodefluoroborylation of Polyfluoroalkenes en Route to
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Diverse Fluoroalkenes. J. Am. Chem. Soc. 2017, 139, 12855-
12862. (g) Leclerc, M. C.; Gorelsky, S. I.; Gabidullin, B. M.;
Korobkov, I.; Baker, R. T. Selective Activation of
Fluoroalkenes with N-Heterocyclic Carbenes: Synthesis of N-
Polyfluoroarenes by Concerted Nucloephilic Aromatic
Substitution with a Hydrosilicate. Angew. Chem. Int. Ed. 2017,
56, 16191-16196. (e) Krüger, J.; Leppkes, J.; Ehm, C.; Lentz,
D. Competition of Nucloephilic Aromatic Substitution, σ-bond
Metathesis, and syn Hydrometalation in Titanium (III)-
Catylzed Hydrodefluorination of Arenes. Chem. Asian J.
Heterocyclic
Imidazolium Salts. Chem. Eur. J. 2016, 22, 8063-8067.
2) Uses of fluoroalkenes in the generation of fluoro-
organics: (h) Andrella, N. O.; Sicard, A. J.; Gorelsky, S. I.;
Fluoroalkenes
and
Polyfluoroalkenyl
(
2
016, 11, 3062-3071. (f) Matsunami, A.; Kuwata, S.; Kayaki,
Y. Hydrodefluorination of Fluoroarenes Using Hydrogen
Transfer Catalysts with a Bifunctional Iridium/NH Moiety.
ACS Catal. 2016, 6, 5181-5185.
2
2 4
Korobkov, I.; Baker, R. T. A T-shaped Ni[κ -(CF ) -] NHC
F
complex: unusual Csp3-F and M-C bond functionalization
reactions. Chem. Sci. 2015, 6, 6392-6397. (i) Ritter, S. K
Tetrafluoroethylene Is Good for More Than Just Teflon.
C&EN 2015, 93(22), 28 (j) Ohashi, M.; Ogoshi, S. Transition
Metal Mediated Transformations of Tetrafluoroethylene into
Various Polyfluorinated Organic Compounds. J. Synth. Org.
Chem, Jpn. 2016, 74, 1047-1057. (k) Ritter, S. K. A new
edition of fluoroalkyl additions. C&EN 2017, 95(30), 9.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
8)
a) Krüger, J.; Ehm, C.; Lentz, D. Improving Selectivity in
Catalytic Hydrodefluorination by Limiting S V Reactivity.
Examples of hydrodefluorination of fluoroalkenes:
(
N
Dalton Trans. 2016, 45, 16789-16798. (b) Wu, J.; Xiao, J.;
Dai, W.; Cao, S. Synthesis of Monofluoroalkenes Through
Selective Hydrodefluorination of gem-Difluoroalkenes with
Red-Al®. RSC Adv. 2015, 5, 34498-34501.
(3)
(a) Ritter, S. K. Fluorochemicals go short. C&EN
(9)
Clot, E.; Mégret, C.; Kraft, B. M.; Eisenstein, O.;
2
010, 88(5), 12. (b) Ritter, S. K. The Shrinking Case for
Jones, W. D. Defluorination of Perfluoropropene Using
Fluorochemicals. C&EN 2015, 93(28), 27. (c) Buck, R.C.
Toxicology Data for Alternative “Short-Chain” Fluorinated
Substances. In Toxicological Effects of Perfluoroalkyl and
Polyfluoroalkyl Substances. Molecular and Integrative
Toxicology, DeWitt, J., Eds.; Humana Press, Switzerland,
2015; pp. 451.(d) Mudumbi, J. B. N.; Ntwampe, S. K. O.;
Matsha, T.; Mekuto, L.; Itoba-Tombo, E. F. Recent
Developments in Polyfluoroalkyl Compounds Research: A
Focus on Human/Environmental Health Impact, Suggested
Substitutes and Removal Strategies. Environ. Monit. Assess.
*
*
Cp
Joint Experimental-Theoretical Perspective. J. Am Chem. Soc.
004, 126, 5647-5653.
10) Kirkham, M. S.; Mahon, M. F.; Whittlesey, M. K. C-
2 2 2
ZrH and Cp ZrHF: A Mechanism Investigation from a
2
(
F Bond Activation of Perfluoroalkenes by Ruthenium
Phosphine Hydride Complexes: X-ray Crystal Structures of
cis-Ru(dmpe)
2
+
F(F--HF)
and
-
[Ru(dcpe) H] [(CF ) C=C(O)CF -CF ] . Chem. Commun,
2 3 2 2 3
2001, 0, 813-814. Also see: Noveski, D.; Braun. T.; Schulte,
M.; Neumann, B.; Stammler, H.-G. C-F Activation and
Hydrodefluorination of Fluorinated Alkenes at Rhodium.
Dalton Trans. 2003, 4075-4083.
2
017, 189, 402.
4) (a) Takahira, Y.; Morizawa, Y. Ruthenium-
(
Catalyzed Oelfin Cross-Metathesis with Tetrafluoroethylene
and Analogous Fluoroolefins. J. Am. Chem. Soc. 2015, 137,
(11) Vela, J.; Smith, M. J.; Yu, Y.; Ketterer, N. A.;
Falschenriem, C. J.; Lachicotte, R. J.; Holland, P. L. Synthesis
and Reactivity of Low-Coordinate Iron(II) Fluoride
7
031-7034.
(5) Myhre, G.; Shindell, D.; Bréon, F.-M.; Collins, W.;
Complexes
Hydrodefluorination of Fluorocarbons. J. Am. Chem. Soc.
005, 127, 7857-7870.
12) Kühnel, M. F.; Lentz, D. Titanium-Catalyzed C-F
and
Their
Use
in
the
Catalytic
Fuglestvedt, J.; Huang, J.; Koch, D.; Lamarque, J.-F.; Lee, D.;
Mendoza, B.; Nakajima, T.; Robock, A.; Stephens, G.;
Takemura, T.; Zhang, H. Anthropogenic and Natural
Radiative Forcing. In Climate Change 2013: The Physical
Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate
Change, Stocker, T.F.; Qin, D.; Plattner, G.-K.; Tignor, M.;
Allen, S.K.; Boschung, J.; Nauels, A.; Xia, Y.; Bex, V.;
2
(
Activation of Fluoroalkenes. Angew. Chem. Int. Ed. 2010, 49,
2933-2936.
(13) a) Jaeger, A. D.; Ehm, C.; Lentz, D. Organocatalytic
C-F Bond Activation with Alanes. Chem. Eur. J. 2018, 24,
6
769-6777. b): Schneider, H.; Hock, A.; Jaeger, A. D.; Lentz,
Midgley, P.M.,
Cambridge, United Kingdom and New York, NY, USA, 2013.
(6) (a) Banks, R. E.; Smart, B. E.; Tatlow, J. C.
Eds.; Cambridge University Press,
D.; Radius, U. NHC Alane Adducts as Hydride Sources in the
Hydrodefluorination of Fluoroaromatics and Fluoroolefins.
Eur. J. Inorg. Chem. 2018, 2018, 4031-4043. c) Jaeger, A. D.;
Walter, R.; Ehm, C.; Lentz, D., Gallium Hydrides and O/N-
Donors as Tunable Systems in C−F Bond Activation. Chem.
Asian J. 2018, 13, 2908-2915.
Organofluorine Chemistry: Principles and Commercial
Applications; Plenum: New York, 1994. (b) Chambers, R. D.
Fluorine Chemistry at the Millennium; Elsevier: Amsterdam,
2
000.
7)
(14) (a) Hu, J.; Han, X.; Yuan, Y.; Shi, Z. Stereoselective
(
Examples of hydrodefluorination of fluoroarenes: (a)
Synthesis of Z-fluoroalkenes Through Copper-Catalyzed
Hydrodefluorination of gem-Difluoroalkenes with Water.
Angew. Chem. Int. Ed. 2017, 56, 13342-13346. (b) Kojima, R.;
Kubota, K.; Ito, H. Stereodivergent Hydodefluorination of
gem-Difluoroalkenes: Selective Synthesis of (Z)- and (E)-
Monofluoroalkenes. Chem. Commun. 2017, 53, 10688-10691.
Whittlesey, M. K.; Peris, E. Catalytic Hydrodefluorination
with Late Transition Metal Complexes. ACS Catal. 2014, 4,
3
152-3159. (b) Lv, H.; Cai. Y.-B.; Zhang, J.-L. Copper-
catalyzed Hydrodefluorination of Fluoroarenes by Copper
Hydride Intermediates. Angew. Chem. Int. Ed. 2013, 52, 3203-
3207. (c) Cybulski, M. K.; McKay, D.; Macgregor, S. A.;
Mahon, M. F.; Whittlesey, M. K. Room Temperature
Regioselective Catalytic Hydrodefluorination of Fluoroarenes
(c) Andrella, N.O.; Liu, K.; Gabidullin, B.; Vasiliu, M.;
Dixon, D. A.; Baker, R. T. Metal Heptafluoroisopropyl (M-
hfip) Complexes for Use as hfip Transfer Agents.
Organometallics 2018, 37, 422-432. (d) Baker, R. T.,
Andrella, N. O. Process for Preparation of Hydrofluoroalkenes
with trans-[Ru(NHC)
Ru-H Attack Pathway. Angew. Chem. Int. Ed. 2017, 56, 1515-
519. (d) Kikushima, K.; Grellier, M.; Ohashi, M.; Ogoshi, S.
4 2
H ] through a Concerted Nucleophilic
1
1
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 19
by Selective Catalytic Consecutive Hydrodefluorination.
WO2018039794, March 8, 2018.
(15) Wilberg, K. B.; Murcko, M. A.; Laidig, K. E.;
MacDougall, P. J. Origin of the Gauche Effect in Substituted
Ethanes and Ethenes. J. Phys. Chem., 1990, 94, 6956-6959.
(24) Hintermann, S.; Pregosin, P. S.; Rüegger, H.; Clark,
H. C., Electron Transfer Reactions Involving trans-
[PtH (PCy ] and Fluorinated Benzonitriles. J. Organomet.
Chem. 1992, 435, 225-234.
(25) Kraft, B. M.; Lachicotte, R. J.; Jones, W. D.,
Aliphatic and Aromatic Carbon−Fluorine Bond Activation
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
3 2
)
(16) (a) Burton, D. J. Fluorinated Ylides and Related
Compounds. Chem. Rev., 1996, 96, 1641-1716. b) Leclerc, M.
C.; Da Gama, J. G.; Gabidullin, B. M.; Baker, R. T. A Closer
Look at the Reactivity Between N-Heterocyclic Carbenes and
Fluoroalkenes. J. Fluorine Chem., 2017, 203, 81-89.
2 2
with Cp* ZrH :ꢀ Mechanisms of Hydrodefluorination. J. Am.
Chem. Soc. 2001, 123, 10973-10979.
(26) Lu, H.; Cai, Y.-B.; Zhang, J.-L., Copper-Catalyzed
Hydrodefluorination of Fluoroarenes by Copper Hydride
Intermediates. Angew. Chem. Int. Ed. 2013, 52, 3203-3207
(27) Panetier, J. A.; Macgregor, S. A.; Whittlesey, M. K.,
Catalytic Hydrodefluorination of Pentafluorobenzene by
(17) a) Jordan, A. J.; Wyss, C. M.; Bacsa, J.; Sadighi, J. P.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Synthesis and Reactivity of New Copper(I) Hydride Dimers.
Organometallics, 2016, 35, 613-616. b) Frey, G. D.;
Donnadieu, B.; Solelhavoup, M.; Bertrand, G. Synthesis of a
Room-Temperature-Stable Dimeric Copper(I) Hydride. Chem.
Asian J., 2011, 6, 402-406. c) Jordan, A. J.; Lalic, G.; Sadighi,
J. P. Coinage Metal Hydrides: Synthesis, Characterization, and
Reactivity. Chem. Rev., 2016, 116, 8318-8372.
3 2 2
[Ru(NHC)(PPh ) (CO)H ]: A Nucleophilic Attack by a Metal-
Bound Hydride Ligand Explains an Unusual ortho-
Regioselectivity. Angew. Chem. Int. Ed. 2011, 50, 2783-2786
(28) D., J. A.; Christian, E.; Dieter, L., Organocatalytic
C−F Bond Activation with Alanes. Chem. Eur. J. 2018, 24,
6769-6777.
(29) Ehm, C.; Krüger, J.; Lentz, D., How a Thermally
Unstable Metal Hydrido Complex Can Yield High Catalytic
Activity Even at Elevated Temperatures. Chem. Eur. J. 2016,
22, 9305-9310
(30) Luo, Y. R., Comprehensive Handbook of Chemical
Bond Energies. CRC Press: Boca Rayton, FL, 2007.
(31) Ehm, C.; Lentz, D., Partially Fluorinated Butatrienes:
A Coupled Cluster Study. J. Phys. Chem. A 2010, 114, 3609-
3614.
(32) Ehm, C.; Lentz, D., Cyclic Dimers of
Tetrafluorobutatriene. Theor. Chem. Acc. 2011, 129, 507-515.
(33) Wiberg, K. B., Application of the Pople-Santry-Segal
CNDO Method to the Cyclopropylcarbinyl and Cyclobutyl
Cation and to Bicyclobutane. Tetrahedron 1968, 24, 1083-
1096.
(18) (a) Li, L.; Ni, C.; Xie, Q.; Hu, M.; Wang. F.; Hu, J.
TMSCF as Convenient Source of CF =CF for
3
a
2
2
Pentafluoroethylation, (Aryloxy)tetrafluoroethylation, and
Tetrafluoroethylation. Angew. Chem. Int. Ed. 2017, 56, 9971-
9
975. (b) Ohashi, M.; Ishida, N.; Ando, K.; Hashimoto, Y.;
I
Shigaki, A.; Kikushima, K.; Ogoshi, S. Cu - Catalyzed
Pentafluoroethylation of Aryl Iodides in the Presence of
Tetrafluoroethylene and Cesium Fluoride: Determining the
I
Route to the Key Pentafluoroethyl Cu Intermediate. Chem.
Eur. J. 2018, 24, 9794-9798.
(
19) (a) Romero, E. A.; Olsen, P. M.; Jazzar, R.;
Soleilhavoup, M.; Gembicky, M.; Bertrand, G., Spectroscopic
Evidence for Monomeric Copper(I) Hydride and
a
Crystallographic Characterization of a Monomeric Silver(I)
Hydride. Angew. Chem. Int. Ed. 2017, 56, 4024-4027. (b)
Goeden, G. V.; Huffman, J. C.; Caulton, K. G., A Cu-(μ-H)
Bond Can Be Stronger Than an Intramolecular P-C Bond.
(34) Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.;
Carpenter, J. E.; Bohmann, J. A.; Morales, C. M.; Weinhold,
F. NBO 5.0, Theoretical Chemistry Institute, University of
Wisconsin, Madison, WI, URL: http://nbo6.chem.wisc.edu/.
(35) Tolman, C. A., Steric Effects of Phosphorus Ligands
in Organometallic Chemistry and Homogeneous Catalysis.
Chem. Rev. 1977, 77, 313-348
2
Synthesis and Structure of Cu-(μ-H)
2
[η -CH
3
C(CH
2
PPh
2
)
3
]
2
.
Inorg. Chem. 1986, 25, 2484-2485. (c) Mankad, N. P.; Laitar,
D. S.; Sadighi, J. P., Synthesis, Structure, and Alkyne
Reactivity of
a
Dimeric (Carbene)copper(I) Hydride.
Organometallics 2004, 23, 3369-3371. (c) Zall, C. M.;
Linehan, J. C.; Appel, A. M., Triphosphine-Ligated Copper
Hydrides for CO
Thermodynamic Studies. J. Am. Chem. Soc. 2016, 138, 9968-
977. (d) Dhayal, R. S.: van Zyl, W. E.; Liu, C. W., Copper
2
Hydrogenation: Structure, Reactivity, and
(36) See for example: Walther, D.; Liesicke, S.; Böttcher,
L.; Fischer, R.; Görls, H.; Vaughan, G., Metal Complexes with
9
2,3-Bis(diphenylphosphino)-1,4-diazadiene
Ligands:
Hydride Clusters in Energy Storage and Conversion. Dalton
Trans. 2019, 48, 3531-3538.
Synthesis, Structures, and an Intramolecular Metal-Mediated
[4+2] Cycloaddtion Employing a Benzene Ring as a
Dienophile. Inorg. Chem. 2003, 42, 625-632.
(37) Craig, N. C.; Entemann, E. A., Thermodynamics of
cis-trans Isomerizations. The 1,2-Difluoroethylenes. J. Am.
Chem. Soc. 1961, 83, 3047-3050
(20) Chen, J.-X.; Daeuble, J. F.; Brestensky, D. M.;
Stryker, J. M., Highly Chemoselective Catalytic
Hydrogenation of Unsaturated Ketones and Aldehydes to
Unsaturated Alcohols Using Phosphine-Stabilized Copper(I)
Hydride Complexes. Tetrahedron 2000, 56, 2153-2166.
(38) Iwashita, A.; Chan, M. C. W.; Makio, H.; Fujita, T.,
Attractive Interactions in Olefin Polymerization Mediated by
(21) Issenhuth, J.-T.; Notter, F.-P.; Dagorne, S.; Dedieu,
A.; Bellemin-Laponnaz, S., Mechanistic Studies on the
Copper-Catalyzed Hydrosilylation of Ketones. Eur. J. Inorg.
Chem. 2010, 2010, 529-541.
Post-Metallocene
Catalysts
with
Fluorine-Containing
Ancillary Ligands. Catal. Sci. Technol. 2014, 4, 599-610.
(39) Talarico, G.; Busico, V.; Cavallo, L., “Living”
Propene Polymerization with Bis(phenoxyimine) Group 4
Metal Catalysts:ꢀ New Strategies and Old Concepts.
Organometallics 2004, 23, 5989-5993.
(40) Reed, A. E.; Curtiss, L. A.; Weinhold, F.,
Intermolecular Interactions From a Natural Bond Orbital,
Donor-Acceptor Viewpoint. Chem Rev. 1988, 88, 899-926.
(41) Minenkov, Y.; Occhipinti, G.; Jensen, V. R.,
Metal−Phosphine Bond Strengths of the Transition Metals: A
(22) Kuehnel, M. F.; Lentz, D.; Braun, T., Synthesis of
Fluorinated Building Blocks by Transition-Metal-Mediated
Hydrodefluorination Reactions. Angew. Chem. Int. Ed. 2013,
52, 3328-3348.
(23) Aizenberg, M.; Milstein, D., Homogeneous Rhodium
Complex-Catalyzed Hydrogenolysis of C-F bonds. J. Am.
Chem. Soc. 1995, 117, 8674-8675.
1
6
ACS Paragon Plus Environment

Page 17 of 19
Journal of the American Chemical Society
Challenge for DFT†. J. Phys. Chem. A 2009, 113, 11833-
11844.
(42) Falivene, L.; Credendino, R.; Poater, A.; Petta, A.;
Serra, L.; Oliva, R.; Scarano, V.; Cavallo, L., SambVca 2. A
Web Tool for Analyzing Catalytic Pockets with Topographic
Steric Maps. Organometallics 2016, 35, 2286-2293.
M.; Burns, E. G.; Matsnev, A. J.; Fernandez, R. E.; McMillen,
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
C. D.; Thrasher, J. S. Preparation of Tetrafluoroethylene from
the Pyrolysis of Pentafluoropropionate Salts. J. Fluorine
Chem., 2017, 196, 107-116.
(59) (a) Koroniak, H.; Palmer, K. W.; Dolbier, W. R.;
1
9
Zhang, H. F NMR Properties of Some Trifluorovinyl- and
Pentafluoropropenyl-Substituted Aromatics. Magn. Reson.
Chem. 1993, 31, 748-751. (b) Burton, D. J.; Spawn, T. W.;
Heinze, P. L.; Bailey, A. R.; Shin-Ya, S. Preparation of E-
1,2,3,3,3-Pentafluoropropene, Z-1,2,3,3,3-Pentafluoropropene
and E-1-Iodopentafluoropropene. J. Fluorine Chem. 1989, 44,
167-174.
(43) Banerjee, D.; Ghosh, A.; Chattopadhyay, S.; Ghosh,
P.; Chaudhuri, R. K., Revisiting the ‘Cis-Effect’ in 1,2-
difluoro Derivatives of Ethylene and Diazene Using Ab Initio
Multireference Methods. Mol. Phys. 2014, 112, 3206-3224.
(44) Bingham, R. C., The Stereochemical Consequences
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
of Electron Delocalization in Extended π- Systems. An
interpretation of the Cis Effect Exhibited by 1,2-Disubstituted
Ethylenes and Related Phenomena. J. Am. Chem. Soc. 1976,
98 (2), 535-540.
(60) Banks, R. E., Barlow, M. G., Nickkho-Amiry, M.
Preparation
of
2,3,3,3-Tetrafluoropropene
from
Trifluoroacetylacetone and Sulphur Tetrafluoride. J. Fluorine
Chem. 1997, 82, 171-174.
(45) Wiberg, K. B.; Hadad, C. M.; Breneman, C. M.;
Laidig, K. E.; Murcko, M. A.; Lepage, T. J., The Response of
Electrons to Structural Changes. Science 1991, 252, 1266.
(61) Meißner, G., Kretschmar, K., Braun, T., Kemnitz, E.
Consecutive Transformations of Tetrafluoropropenes:
Hydrogermylation and Catalytic C-F Activation Steps at a
Lewis Acidic Aluminium Fluoride. Angew. Chem. Int. Ed.
2017, 56, 16338-16341.
(46) Wiberg, K. B.; Murcko, M. A.; Laidig, K. E.;
MacDougall, P. J., Origin of the Gauche Effect in Substituted
Ethanes and Ethenes. J. Phys. Chem. 1990, 94, 6956-6959.
13
(
47) Ditchfield, R.; Ellis, P. D., C NMR Chemical Shifts
(62) Emsley, J. W., Phillips, L. Fluorine Chemical Shifts.
Prog. Nucl. Magn. Reson. Spectrosc. 1971, 7, 1-520.
for Fluoroethylenes and Fluoroacetylene. Chem. Phys. Lett.
19
13
1
972, 17, 342-344.
(63) Brey, W. S. F and C Spectra of Fluorinated and
Partially Fluorinated Vinyl Alkyl Ethers. J. Fluorine Chem.
2005, 126, 389-399.
(48) Yamamoto, T.; Tomoda, S., On the Origin of cis-
Effect in 1,2-Difluoroethene. Chem. Lett. 1997, 26, 1069-
070.
49) Cremer, D., The Role of Correlation in Calculations
1
(64) Kučnirová, K.; Šimůnek, O.; Rybáčkoá, M.; Kvíčala,
1
9
(
J. Structural Assignment of Fluorocyclobutenes by F NMR
1
9
on 1,2-Difluoroethylenes. The cis-trans Energy Difference.
Chem. Phys. Lett. 1981, 81, 481-485.
Spectroscopy – Comparison of Calculated F NMR Shielding
1
9
Constants with Experimental F NMR Shifts. Eur. J. Org.
Chem. 2018, 2018, 3867-3874.
(50) Feller, D.; Craig, N. C.; Groner, P.; McKean, D. C.,
Ab Initio Coupled Cluster Determination of the Equilibrium
Structures of cis- and trans-1,2-Difluoroethylene and 1,1-
Difluoroethylene. J. Phys. Chem. A. 2011, 115, 94-98.
(65) Snegirev, V. F.; Makarov, K. N.; Zabolotskikh, V.
F.; Sorokina, M. G.; Knunyants, I. L. Catalytic and Hydride
Reduction of Hexafluoropropylene Dimers. Russ. Chem. Bull.
1983, 32, 2489-2494.
(51) Farcasiu, D., The Use and Misuse of the Hammond
Postulate. J. Chem. Educ. 1975, 52, 76.
(52) Hammond, G. S., A Correlation of Reaction Rates. J.
Am. Chem. Soc. 1955, 77, 334-338.
(66) (a) Edwards, M. L.; Stemerick, D. M.; Jarvi, E. T.;
Matthews,
D.
P.;
McCarthy,
J.
R.
Difluoromethyldiphenylphosphine Oxide. A New Reagent for
Conversion of Carbonyl Compounds to 1,1-Difluoroolefins.
Tetrahedron Lett. 1990, 31, 5571-5574. (b) Tian, H.;
Shimakoshi, H.; Imamura, K.; Shiota, Y.; Yoshizawa, K.;
(53) IRC scans show that F-elimination proceeds first
towards the electron poor P, then to the Cu. In the presence of
silanes, an OR group could therefore be abstracted, which
would lead to catalyst decomposition.
Hisaeda, Y. Photocatalytic Alkene Reduction by a B12-TiO
2
(54) With the isolated complex preliminary reactivity was
Hybrid Catalyst Coupled with C-F Bond Cleavage for gem-
Difluoroolefin Synthesis. Chem. Commun. 2017, 53, 9478-
9481.
(67) Gaussian 09 Revision B.1; for the full citation see the
Supporting Information.
explored. Cross-coupling of the tetrafluoroethyl complex 7a
with iodobenzene (5 eq.) in the presence of 10 mol% of [(1,3-
bis(2,4,6-trimethylphenyl)-4,5-dihydroimidazol-2-ylidene)
copper(I) chloride] (80 % yield, determined by 1 F NMR)
could be achieved. This represents a new route for the
9
(68) J. Baker, 2.4 ed., Parallel Quantum Solutions,
Fayetteville, AR, 2001.
2 2
introduction of the CF CF H moiety to organic compounds.
The use of 7b under the same reaction conditions proved
ineffective. Further investigation of this reaction is beyond the
scope of this publication and will be reported in due course.
(69) Baker, J., An Algorithm for the Location of
Transition States. J. Comput. Chem. 1986, 7, 385-395.
(70) Budzelaar, P. H. M., Geometry Optimization Using
Generalized, Chemically Meaningful Constraints. J. Comput.
Chem. 2007, 28, 2226-2236.
(71) Ehm, C.; Budzelaar, P. H. M.; Busico, V.,
Calculating Accurate Barriers for Olefin Insertion and Related
Reactions. J. Organomet. Chem. 2015, 775, 39-49.
(72) Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G.
E., Climbing the Density Functional Ladder: Nonempirical
Meta-Generalized Gradient Approximation Designed for
Molecules and Solids. Phys. Rev. Lett. 2003, 91, 146401.
(73) Balabanov, N. B.; Peterson, K. A., Systematically
Convergent Basis Sets for Transition Metals. I. All-electron
(55) Bantreil, X.; Nolan, S. P. Synthesis of N-
Heterocyclic Carbene Ligands and Derived Ruthenium Olefin
Metathesis Catalysts. Nat. Protoc. 2011, 6, 69-77
(56) Lee, D.-W.; Yun, J. Direct Synthesis of Stryker’s
Reagent from a Cu(II) Salt. Tetrahedron Lett. 2005, 46, 2037-
2039.
(57) Hunadi, R. J.; Baum, K. Tetrafluoroethylene: A
Convenient Laboratory Preparation. Synthesis 1982, 39, 454.
(58) Hercules, D. A.; Parrish, C. A.; Sayler, T. S.; Tice, T.
K.; Williams, S. M.; Lowery, L. E.; Brady, M. E.; Coward, R.
B.; Murphy, J. A.; Hey, T. A.; Scavuzzo, A. R.; Rummler, L.
1
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 19
Correlation Consistent Basis Sets for the 3d Elements Sc–Zn.
J. Chem. Phys. 2005, 123, 064107.
(74) Balabanov, N. B.; Peterson, K. A., Basis Set Limit
Electronic Excitation Energies, Ionization Potentials, and
Electron Affinities for the 3d Transition Metal Atoms:
Coupled Cluster and Multireference Methods. J. Chem. Phys.
(92) a) Sievers, M. R.; Jarvis, L. M.; Armentrout, P. B.,
Transition-Metal Ethene Bonds: Thermochemistry of
M (C H ) (M = Ti-Cu, n = 1 and 2) Complexes. J. Am. Chem.
2 4 n
Soc. 1998, 120, 1891-1899. b) Hertwig, R. H.; Koch, W.;
Schröder, D.; Schwarz, H., A Comparative Computational
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
+
Study of Cationic Coinage Metal-Ethylene Complexes
+
2
006, 125, 074110.
75) Schuchardt, K. L.; Didier, B. T.; Elsethagen, T.; Sun,
(C
2
H
4
)M (M = Cu, Ag, and Au). J. Phys. Chem. 1996, 100,
(
12253-12260.
L. S.; Gurumoorthi, V.; Chase, J.; Li, J.; Windus, T. L., Basis
Set Exchange: A Community Database for Computational
Sciences. J. Chem. Inf. Model 2007, 47, 1045-1052
(93) Vitale, G.; Valina, A. B.; Huang, H.; Amunugama,
R.; Rodgeers, M. T., Solvation of Copper Ions by Acetonitrile.
+
Structures and Sequential Binding Energies of Cu (CH
3
CN)
x
,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(76) Whitten, J. L., Coulombic Potential Energy Integrals
x = 1-5, from Collision-Induced Dissociation and Theoretical
Studies. J. Phys. Chem. 2001, 105, 11351-11364.
and Approximations. J. Chem. Phys. 1973, 58, 4496.
(77) Baerends, E. J.; Ellis, D. E.; Ros, P., Self-Consistent
Molecular Hartree—Fock—Slater Calculations I. The
Computational Procedure. Chem. Phys. 1973, 2, 41-51
(78) Feyereisen, M.; Fitzgerald, G.; Komornicki, A., Use
of Approximate Integrals in Ab Initio Theory. An Application
in MP2 Energy Calculations. Chem. Phys. Lett. 1993, 208,
3
59-363.
79) Vahtras, O.; Almlöf, J.; Feyereisen, M. W., Integral
(
Approximations for LCAO-SCF Calculations. Chem. Phys.
Lett. 1993, 213, 514-518.
(80) Tomasi, J., Thirty Years of Continuum Solvation
Chemistry: A Review and Prospects for the Near Future.
Theor. Chem. Acc. 2004, 112, 184-203.
(
81) The Berny algorithm was never fully published, see
the Gaussian documentation for details.
82) Tao, J. M.; Perdew, J. P.; Staroverov, V. N.;
(
Scuseria, G. E., Climbing the Density Functional Ladder:
Nonempirical Meta-Generalized Gradient Approximation
Designed for Molecules and Solids. Phys. Rev. Lett. 2003, 91
(83) Grimme, S., Density Functional Theory with London
Dispersion Corrections. Wiley Interdiscip. Rev. Comput. Mol.
Sci. 2011, 1, 211-228.
(84) Grimme, S.; Ehrlich, S.; Goerigk, L., Effect of the
Damping Function in Dispersion Corrected Density Functional
Theory. J. Comput. Chem. 2011, 32, 1456-1465.
(85) Tobisch, S.; Ziegler, T., Catalytic Oligomerization of
Ethylene to Higher Linear α-Olefins Promoted by the Cationic
5
II
+
2 2
Group 4 [(η -Cp-(CMe -bridge)-Ph)M (ethylene) ] (M = Ti,
Zr, Hf) Active Catalysts:ꢀ A Density Functional Investigation
of the Influence of the Metal on the Catalytic Activity and
Selectivity. J. Am. Chem. Soc. 2004, 126, 9059-9071.
(86) Dunlop-Brière, A. F.; Budzelaar, P. H. M.; Baird, M.
C., α- and β-Agostic Alkyl–Titanocene Complexes.
Organometallics 2012, 31, 1591-1594.
(87) Raucoules, R.; de Bruin, T.; Raybaud, P.; Adamo,
C., Theoretical Unraveling of Selective 1-Butene
Oligomerization Catalyzed by Iron−Bis(arylimino)pyridine.
Organometallics 2009, 28, 5358-5367.
(88) Glendening, E. D.; Landis, C. R.; Weinhold, F.,
Natural Bond Orbital Methods. Wiley. Interdiscip. Rev.
Comput. Mol. Sci. 2012, 2, 1–42.
(89) Grimme, S., Semiempirical GGA-type Density
Functional Constructed with Long-Range Dispersion
a
Correction. J. Comput. Chem. 2006, 27, 1787-1799.
(90) Miyawaki, J.; Sugawara, K.-I., ZEKE Photoelectron
Spectroscopy of the Silver- and Copper-ammonia Complexes.
J. Chem. Phys. 2003, 119, 6539-6545.
(
91) Meyer, F.; Chen, Y.-M.; Armentrout, P. B.,
+
+
Sequential Bond Energies of Cu(CO)
x x
and Ag(CO) (x = 1-
4
). J. Am. Chem. Soc. 1995, 117, 4071-4081.
1
8
ACS Paragon Plus Environment

Page 19 of 19
1
Journal of the American Chemical Society
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
SYNOPSIS TOC (Word Style “SN_Synopsis_TOC”). If you are submitting your paper to a journal that requires a synopsis
graphic and/or synopsis paragraph, see the Instructions for Authors on the journal’s homepage for a description of what
needs to be provided and for the size requirements of the artwork.
Authors are required to submit a graphic entry for the Table of Contents (TOC) that, in conjunction with the manuscript title,
should give the reader a representative idea of one of the following: A key structure, reaction, equation, concept, or theorem,
etc., that is discussed in the manuscript. Consult the journal’s Instructions for Authors for TOC graphic specifications.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
9
ACS Paragon Plus Environment