Reductive Termination of Cyanoisopropyl Radicals by Copper(I) Complexes and Proton Donors: Organometallic Intermediates or Coupled Proton-Electron Transfer?

Article abstract of DOI:10.1021/acs.inorgchem.9b00660

Cyanoisopropyl radicals, generated thermally by the decomposition of azobis(isobutyronitrile) (AIBN), participate in reductive radical termination (RRT) under the combined effect of copper(I) complexes and proton donors (water, methanol, triethylammonium salts) in acetonitrile or benzene. The investigated copper complexes were formed in situ from [Cu^I(MeCN)₄]⁺BF₄^- in CD₃CN or Cu^IBr in C₆D₆ using tris[2-(dimethylamino)ethyl]amine (Me₆TREN), tris(2-pyridylmethyl)amine (TPMA), and 2,2′-bipyridine (BIPY) ligands. Upon keeping all other conditions constants, the impact of RRT is much greater for the Me₆TREN and TPMA systems than for the BIPY system. RRT scales with the proton donor acidity (Et₃NH⁺ a‰? H₂O > CH₃OH), it is reduced by deuteration (H₂O > D₂O and CH₃OH > CD₃OD), and it is more efficient in C₆D₆ than in CD₃CN. The collective evidence gathered in this study excludes the intervention of an outer-sphere proton-coupled electron transfer (OS-PCET), while an inner-sphere PCET (IS-PCET) cannot be excluded for coordinating proton donors (water and methanol). On the other hand, the strong impact of RRT for the noncoordinating Et₃NH⁺ in CD₃CN results from the formation of an intermediate Cu^I-radical adduct, suggested by DFT calculations to involve binding via the N atom to yield keteniminato [L/Cu-N=C=CMe₂]⁺ derivatives with only partial spin delocalization onto the Cu atom.

Full text of DOI:10.1021/acs.inorgchem.9b00660

Article
pubs.acs.org/IC
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
Reductive Termination of Cyanoisopropyl Radicals by Copper(I)
Complexes and Proton Donors: Organometallic Intermediates or
Coupled Proton−Electron Transfer?
Lucas Thevenin,† Christophe Fliedel,† Marco Fantin,‡ Thomas G. Ribelli,‡ Krzysztof Matyjaszewski,‡
and Rinaldo Poli*,†,§
†
́
CNRS, LCC (Laboratoire de Chimie de Coordination), Universite de Toulouse, UPS, INPT, 205 Route de Narbonne, BP 44099,
F-31077, Toulouse Cedex 4, France
‡Department of Chemistry, Carnegie Mellon University, 4400 Fifth Avenue, Pittsburgh, Pennsylvania 15213, United States
§Institut Universitaire de France, 1, rue Descartes, 75231 Paris Cedex 05, France
S
* Supporting Information
ABSTRACT: Cyanoisopropyl radicals, generated thermally by the decomposition of
azobis(isobutyronitrile) (AIBN), participate in reductive radical termination (RRT)
under the combined effect of copper(I) complexes and proton donors (water, methanol,
triethylammonium salts) in acetonitrile or benzene. The investigated copper complexes
were formed in situ from [CuI(MeCN)4]+BF4− in CD3CN or CuIBr in C6D6 using tris[2-
(dimethylamino)ethyl]amine (Me6TREN), tris(2-pyridylmethyl)amine (TPMA), and
2,2′-bipyridine (BIPY) ligands. Upon keeping all other conditions constants, the impact
of RRT is much greater for the Me6TREN and TPMA systems than for the BIPY system.
RRT scales with the proton donor acidity (Et3NH+ ≫ H2O > CH3OH), it is reduced by
deuteration (H2O > D2O and CH3OH > CD3OD), and it is more efficient in C6D6 than
in CD3CN. The collective evidence gathered in this study excludes the intervention of an
outer-sphere proton-coupled electron transfer (OS-PCET), while an inner-sphere PCET
(IS-PCET) cannot be excluded for coordinating proton donors (water and methanol).
On the other hand, the strong impact of RRT for the noncoordinating Et3NH+ in
CD3CN results from the formation of an intermediate CuI-radical adduct, suggested by DFT calculations to involve binding via
the N atom to yield keteniminato [L/Cu−NCCMe2]+ derivatives with only partial spin delocalization onto the Cu atom.
Scheme 2. Reaction between an Organic Radical and a
Metal Complex Leading to Metal−Carbon Bond Formation
1. INTRODUCTION
Organic radicals can interact with transition metal complexes
in a number of different ways.1 Of particular interest to us are
reversible reactions that yield dormant species, allowing radical
polymerization processes to become controlled via the
persistent radical effect.2 These types of polymerization
processes are known as “reversible deactivation radical
polymerizations” (RDRPs).3 Two such reactions are atom
transfer from a halogen-containing oxidized metal complex, Y−
Mtx+1/L, to yield the reduced complex Mtx/L and an alkyl
halide R−Y (Scheme 1; Y = halogen, Mt = metal, L =
coordinating ligand, R = organic radical, x and x + 1 = formal
oxidation states) and direct metal−carbon bond formation to
form an organometallic species, Scheme 2. The corresponding
RDRP strategies are respectively known as “atom transfer
radical polymerization” (ATRP)4−9 and “organometallic-
mediated radical polymerization” (OMRP).10−12 These two
mechanisms were shown to constructively interplay for certain
metal systems.13,14 One of the most versatile metals for ATRP
is copper, shuttling between the oxidation states +1 and +2,
whereas OMRP has so far been particularly successful using
CoII/CoIII systems and has not found successful implementa-
tions for copper complexes.15
In recent years, the use of copper complexes with greater
catalytic activity in ATRP, i.e., greater KATRP (Scheme 1),16,17
was shown to promote competitive radical trapping by the
CuI/L ATRP catalyst with formation of the organometallic
dormant species (i.e., sufficiently small KOMRP, Scheme 2).
However, rather than helping the system to gain better control,
Scheme 1. Reaction between an Organic Radical and a
Halogen-Containing Metal Complex Leading to Atom
Transfer
Received: March 6, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.9b00660
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
this phenomenon opened access to a metal-catalyzed radical
termination pathway, termed “catalytic radical termination”
(CRT, Scheme 3).18−22 This process consists of a reaction
CuI/CuII ratio. The observed trends and the accompanying
kinetic simulation studies confirmed that RT yields nearly
exclusively combination and suggested that the products with
the same molar mass distribution as the macroinitiator
originate from CRT.
Scheme 3. Reaction between an Organocopper(II) complex
and a Second Radical, Yielding Radical Termination and
Regenerating the Copper(I) Complex
A related investigation,28 however, revealed a new
phenomenon. In that study, a PMA−Br macroinitiator of
narrow molar mass distribution was again used and activated
by CuIBr/Me6TREN in the absence of monomer. Two
important differences relative to the investigations carried
out in our laboratories were the use of (undried) toluene as
solvent and metallic copper (Cu0), which was added to the
system rather than a CuII salt. The resulting polymer had again
a bimodal molecular weight distribution (MWD), with the two
contributing distributions having the same and double
molecular weights, in agreement with our study. However, a
detailed analysis of the recovered polymer by NMR and by
mass spectrometry showed the absence of unsaturated chain,
PMA(−H). Instead, the distribution with the same average
molar mass as the PMA−Br precursor contained only saturated
chain ends, PMA−H, thus excluding the possibility that it was
generated by radical disproportionation. Furthermore, an
additional experiment carried out in the presence of CH3OD
yielded the corresponding PMA−D while the double MWD
completely disappeared. The saturated chains were proposed
by the authors to originate from protonolysis (by adventitious
water in the medium or by the added methanol) of the
organometallic intermediate. The reaction scheme is summar-
ized in Scheme 5.
between the organometallic intermediate R−CuII/L and a
second radical to yield termination products and regenerate
the CuI/L catalyst, and it was found to be particularly prevalent
when R• is a growing polyacrylate chain.
Investigations aimed at learning about the intimate
mechanism of the CRT process have been recently conducted.
In an initial study,19 well-defined bromine-capped polystyrene
(PSt−Br) and poly(methyl acrylate) (PMA−Br) chains of
known average molecular mass and low dispersity, prepared by
ATRP, were activated by atom transfer to CuIBr/TPMA
(TPMA = tris(2-pyridylmethyl)amine, see Scheme 4), and the
Scheme 4. Ligands Used in This and Previous Studies
Scheme 5. Possible Events in the Copper-Mediated
Termination of Poly(methyl acrylate) Radicals in the
Presence of Methanol
produced radicals were allowed to terminate in the absence of
monomer. These studies were conducted in MeCN at 25 °C
with the activating system formed in situ from CuIBr and
TPMA, while CuIIBr2 was also present to moderate the
activation equilibrium (Scheme 1). The PSt−Br activation was
found to yield a double-molecular weight product, as expected
from radical−radical coupling (combination). Since CRT is
not efficient for these radicals,23 the latter mostly undergo
noncatalyzed bimolecular termination, which occurs predom-
inantly by combination.24 On the other hand, PMA−Br
activation yielded a product with the same average molecular
weight of the precursor, indicating the absence of combination.
Since polyacrylate radicals are also known to terminate mostly
by combination25,26 (although this view has recently been
challenged27), this result was attributed to the copper-catalyzed
termination (Scheme 3).
In a subsequent study, a PMA−Br sample was activated by
CuI systems made in situ from [Cu(MeCN)4]+OTf− and three
different ligands (TPMA*3, TPMA or Me6TREN, Scheme 4),
allowing termination reactions to occur in MeCN at 25 °C in
the absence of monomer.21 The recovered polymer had a
bimodal distribution, the two components having respectively
the same and a double average molecular weight relative to
that of the macroinitiator. A key feature of this study was that
the relative contribution of standard bimolecular (nonmetal-
mediated) radical termination (RT) and copper-catalyzed
radical termination (CRT) was tuned by a change of the initial
After activation of PMA−Br (a), the radicals can either
terminate spontaneously in a bimolecular fashion according to
path b or be trapped by [CuI(L)]+ according to path c, while
Cu0, if present together with additional L, would convert the
[CuIIBr(L)]+ product back to [CuI(L)]+ by comproportiona-
tion according to path d. Subsequently, the generated OMRP
dormant species [PMA−CuII(L)]+ may react with a second
B
DOI: 10.1021/acs.inorgchem.9b00660
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
radical leading to CRT (e). Yamago et al. also proposed an
alternative bimolecular CRT pathway involving interaction
between two [PMA−CuII(L)]+ complexes.28 Competing with
CRT, the [PMA−CuII(L)]+ species may react with a proton
donor to yield the observed PMA−H and a secondary CuII
complex, [CuII(OR)(L)]+ (f). We propose to name the latter
process “reductive radical termination” (RRT). It is a
stoichiometric process, not a catalytic one, because it does
not regenerate a CuI complex. Incidentally, a recent report has
also shown the ability of Cu0 to catalyzed acrylate radical
termination,29 which is not included in Scheme 5.
Scheme 6. Possible Pathways for Reductive Radical
Termination
While there is no doubt about the existence of RRT and on
the nature of the resulting product, the intimate mechanisms
and the nature of the bimolecular terminations (RT and CRT)
products are controversial. Yamago et al. have claimed that RT
results in disproportionation27 and, since the unsaturated
disproportionation product was not observed in the presence
of the copper complex (even in the absence of proton donor),
they proposed that RT (Scheme 5b) is totally suppressed by
the faster OMRP trapping (c) and subsequent decompositions
(e, f).28 Hence, for the experiment run in the absence of
deliberately added methanol, they deduced that the distribu-
tion with average molar mass equivalent to that of PMA−Br
(PMA−H) results from RRT, while the double MWD (PMA−
PMA) results from CRT, and proposed that this product is
generated by reductive elimination from a dialkylcopper
intermediate. However, our own above-mentioned study
carried out with variable CuII/CuI ratios, run under very
similar conditions to those of Yamago et al., gave essentially
the same bimodal distribution and clearly demonstrated that
the double molecular weight PMA−PMA product is generated
by RT.21 Our reinterpretation of the results from the study by
Yamago et al. is that the double MWD observed at low [ROH]
results from RT (Scheme 5b) and that an increasing
concentration of the proton donor skews the product
distribution in favor of RRT (Scheme 5, c and f). Complete
suppression of RT is expected at high [ROH] because the
(OS-PCET). Finally, the right-hand side shows the two-step
process involving initial coordination of the proton donor to
the [CuI(L)]+ complex, followed by proton-coupled electron
transfer, which we name inner-sphere PCET, or IS-PCET. The
involvement of metal complexes to mediate PCET is well
established30 and an IS-PCET process was found to be
kinetically competent in a recent mechanistic study of
Fe(acac)3-catalyzed radical alkene cross-coupling.31
It is also relevant to mention the seminal work of
Meyerstein, who has investigated the interaction between
organic radicals and transition metal complexes, including
those of CuI, carrying out their studies in water or aqueous
media.32−44 The resulting organocopper transients usually
decomposed by protonolysis.45 Those investigations, however,
did not pay particular attention to the competition between
trapping by CuI and spontaneous bimolecular termination and
have mostly dealt with relatively reactive (nonstabilized)
radicals, which have a strong tendency to generate
organocopper(II) species. Exceptions are •CH2COOH and
•CH2COO−,43 where the unpaired electron is stabilized by
delocalization onto the carboxylic function. The neutral
•CH2COOH radical gives the organocopper(II) adduct with
RRT rate (vRRT = KOMRPk
RRT[CuI(L)+][PMA•][ROH]) would
become dominant relative to that of RT (vt = kt[PMA•]2). At
Cu+ at a rate comparable to those of the more reactive
aq
•
•
radicals CH3 and CH2CH2COOH (>109 M−1 s−1), whereas
low [ROH], even though OMRP trapping (Scheme 5c,
da,OMRP[CuI(L)+][PMA•]) can compete with RT, the
the anionic CH2COO− radical adds 1 order of magnitude
•
k
slower than •CH3 to a [CuI(L)]+ complex (L = 2,5,8,11-
tetramethyl-2,5,8,11-tetraazadodecane), but still formed an
organocopper(II) transient. The decomposition of the [CuII−
reversibility of the former allows RT and CRT to remain in
competition.
One specific result of the contribution by Yamago et al.,28
however, requires further analysis. In the absence of any added
CH3OH(D), unsaturated chains expected from disproportio-
nation were still absent in the lower molecular weight fraction,
ruling out any pathway leading to disproportionation products.
While the double MWD (PMA−PMA) may result from either
RT (according to us)21 or from CRT (according to Yamago et
al.),28 the selective production of PMA−H in the absence of
ROH is puzzling. Yamago et al. attributed this phenomenon to
the efficient OMRP trapping (Scheme 5c) followed by RRT as
a consequence of the water impurities in the solvent. Another
interesting question is whether the RRT pathway occurs via an
organometallic species. Indeed, this termination process may
also occur by outer-sphere or inner-sphere electron transfer
processes, coupled with protonation, as shown in Scheme 6.
On the left, the organometallic pathway already considered in
Scheme 5 is represented, which we name OMRP because it
involves the dormant species of an organometallic-mediated
radical polymerization. In the middle, the scheme shows a
termolecular outer-sphere proton-coupled electron transfer
CH2COOH]+ and [(L)CuII−CH2COO] transients was
aq
found to occur by protonolysis to yield CH3COOH or
CH3COO− and Cu2+ or [CuII(L)]2+, respectively.
aq
In order to learn more about the RRT mechanism and about
its impact in an ATRP/OMRP process, we have embarked in a
detailed investigation of the nature of the termination products
for stabilized radicals interacting with a variety of CuI
complexes in nonaqueous solvents and in the presence of
controlled amounts of proton donors. The interaction between
carbon-based radicals and copper complexes in nonaqueous
solvents is a relatively unexplored area of investigations.45 The
guiding principle is to probe radicals with a greater or smaller
aptitude to generate an organometallic intermediate (thus
probing the OMRP pathway in Scheme 6 relative to the other
possible PCET pathways). In this report, we focus on the
stabilized cyanoisopropyl radical (Me2(CN)C•), with the idea
in mind that this radical may be unable to generate an
organocopper(II) intermediate because of an insufficiently
strong CuII−C bond.
C
DOI: 10.1021/acs.inorgchem.9b00660
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(see Supporting Information, section S1). All decomposition
experiments in CD3CN were conducted, unless otherwise
stated, at 80 °C (where the half-life is reported as ca. 70 min or
1.2 h) for 8 h, whereas all those in C6D6 were set at 75 °C (t1/2
= ca. 150 min or 2.5 h), also for 8 h. The observed fraction of
decomposed AIBN is consistent with the reported decom-
position rates in the presence of CuI complexes,53 with some
variability caused by the imperfect control of the heating bath
temperature and/or occasional induction periods because of
oxygen contamination. The product distributions reported in
all tables below are relative to the converted AIBN fraction.
Control experiments conducted in the absence of both copper
complex and proton donor (Table 1, entries 1 and 2) yield
2. EXPERIMENTAL SECTION
2.a. Material and Methods. Copper(I) bromide (CuBr, Sigma-
Aldrich, 99.8%), copper(0) powder (<425 μm, Sigma-Aldrich,
99.5%), tris[2-(dimethylamino)ethyl]amine (Me6TREN, Alfa Aesar,
99%), tris(2-pyridylmethyl)amine (TPMA, AmBeed or Koei Chem-
ical Co. Ltd., 98%), and 2,2′-bipyridine (BIPY, Sigma-Aldrich, 99%)
were used as received without any purification. 2,2′-Azobis-
(isobutyronitrile) (AIBN, Sigma-Aldrich, 98%) was recrystallized
from methylene chloride at −32 °C. Tetrakis(acetonitrile)copper(I)
tetrafluoroborate ([CuI(MeCN)4]BF4, Acros Organics, 99%) was
recrystallized from MeCN. Anhydrous deuterated acetonitrile
(CD3CN, Eurisotop, 99.8% D) was distilled over CaH2 under an
inert atmosphere and stored in a sealed flask. Anhydrous deuterated
benzene (C6D6, Eurisotop, 99.5% D) was distilled over NaK alloy
under an inert atmosphere and stored in a sealed flask. Methanol
(MeOH, Carlo Erba, analytical grade) and deuterated methanol
(CD3OD, Eurisotop, 99.9% D) were dried over activated 4 Å
molecular sieves and stored under an inert atmosphere in a sealed
flask. Triethylammonium tetrafluoroborate ([Et3NH]BF4 was synthe-
sized according to the literature.46 The 1H NMR spectra were
recorded on a Bruker Avance III 300 or 400 MHz spectrometer at
ambient temperature; chemical shifts (δ) are reported in ppm vs
Table 1. Results of the Thermal Decomposition of AIBN in
a
the Absence of Copper Complexes
b
c
entry
solvent
additive/AIBN
iBN/dimer
% D
1
2
3
4
5
6
7
8
CD3CN
C6D6
−
−
0.11
0.07
0.11
d
CD3CN
CD3CN
CD3CN
CD3CN
C6D6
H2O (200)
D2O (200)
CH3OH (40)
CD3OD (40)
CH3OH (200)
CD3OD (200)
1
SiMe4 and were determined by reference to the residual H solvent
peaks. The coupling constants are reported in hertz.
d
0.11
n.d.
n.d.
n.d.
0.12
0.13
0.08
0.09
2.b. General Procedure for the AIBN Termination in CD3CN.
A degassed solution of CD3CN containing the ligand (8.6 × 10−2
mmol, 2 equiv relative to Cu) and AIBN (3.6 mg, 2.1 × 10−2 mmol,
0.5 equiv relative to Cu) was transferred via cannula to a screw-
C6D6
−
capped NMR tube containing [Cu(MeCN)4]+BF4 (13.5 mg, 4.3 ×
a
Experiments in CD3CN: T = 80 °C, 8 h. Experiments in C6D6: T =
10−2 mmol). The solution was heated at 80 °C for 8 h and then
cooled to room temperature and analyzed by 1H NMR spectroscopy.
2.c. General Procedure for the AIBN Termination in C6D6. A
degassed solution of C6D6 containing the ligand (8.6 × 10−2 mmol,
1.7 equiv relative to CuI) and AIBN (8.2 mg, 5.1 × 10−2 mmol, 1
equiv relative to CuI) was transferred via cannula to a screw-capped
NMR tube containing CuBr (7.3 mg, 5.1 × 10−2 mmol) and Cu0 (12
mg, 20.4 × 10−2 mmol, 4 equiv relative to CuI). The solution was
heated at 75 °C for 8 h and then cooled to room temperature and
b
75 °C, 8 h. Ratio of the two Me2C resonances (see Supporting
Information), hence equivalent to the moral ratio of AIBN leading to
each product. Key: n.d. = not detectable (the integral of the two
c
peaks of the iBN(H) doublet are identical within experimental error).
d
A small amount of amide (Me2CHCONH2 or Me2CHCOND2, <
1%) was also detected.
mainly tetramethylsuccinonitrile, Me2C(CN)−C(CN)Me2
(indicated as “dimer”) and a minor amount of isobutyronitrile
(iBN) as secondary product (Scheme 7). This is in line with a
1
analyzed by H NMR spectroscopy.
2.d. Computational Details. The DFT calculations were carried
out using the Gaussian09 suite of programs.47 Geometry optimiza-
tions were performed for the isolated molecules without any
symmetry constraint using the BP86 functional48 and the 6-
311G(d,p) basis functions for all the light atoms (O, N, C, H) and
the LANL2DZ functions for Cu, to which was added an f polarization
function (α = 3.325 Cu).49 The unrestricted formulation was used for
open-shell spin doublets, yielding only minor spin contamination
(⟨S2⟩ at convergence was very close to the expected value of 0.75, the
maximum deviation being 0.758 for the free cyanoisopropyl radical,
(CH3)2(CN)C•). All final geometries were characterized as local
minima by verifying that all second derivatives of the energy were
positive. Thermochemical corrections were obtained at 298.15 K on
the basis of frequency calculations, using the standard approximations
(ideal gas, rigid rotor and harmonic oscillator). Corrections for
dispersion were carried out at the fixed BP86 optimized geometries
using Grimme’s D3 empirical method (BP86-D3), using S6 = 1, SR6
= 1.1390, and S8 = 1.6830.50 A solvation correction in acetonitrile (ε
= 35.688) was carried out using the SMD polarizable continuum
model.51 A further correction of 1.95 kcal/mol was applied to bring
the G values from the gas phase (1 atm) to the solution (1 mol/L)
standard state.52
Scheme 7. AIBN Decomposition in the Absence of CuI
Complex
previous investigation54 and with the known termination mode
in the polymerization of acrylonitrile and methacrylonitrile,
which occurs predominantly by combination.55 The observed
iBN probably derives from a small amount of disproportiona-
tion, but the expected methacylonitrile coproduct was not
observed in this specific control experiment, because it is
subsequently consumed by radical addition with formation of
oligomers as previously demonstrated.54 Small amounts of
3. RESULTS AND DISCUSSION
1
vinyl protons were indeed occasionally observed in the H
3.a. Control Experiments. The cyanoisopropyl radical
was conveniently generated by thermal decomposition of
azobis(isobutyronitrile) (AIBN; Me2C(CN)−NN−C(CN)-
Me2) in either CD3CN or C6D6 and the nature and amount of
the formed products were determined by 1H NMR integration
NMR spectra for other experiments during the investigation.
Since the putative oligomers could not be quantified, the
percent of coupling and saturated product reported in Table 1
cannot be taken as precise measures of the percent coupling
and disproportionation for the cyanoisopropyl radical. The
D
DOI: 10.1021/acs.inorgchem.9b00660
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Table 2. Results of the Thermal Decomposition of AIBN in the Presence of the CuI/Me6TREN System
b
c
entry
solvent
CuI complex
additive/AIBN
iBN/dimer
% D in iBN
−
9
10
11
CD3CN
C6D6
C6D6
[CuI(Me6TREN)]+BF4
CuBr/Me6TREN
−
−
−
0.16
0.09
0.06
d
CuBr/Me6TREN
e
[CuI(Me6TREN)]+BF4
[CuI(Me6TREN)]+BF4
[CuI(Me6TREN)]+BF4
[CuI(Me6TREN)]+BF4
[CuI(Me6TREN)]+BF4
[CuI(Me6TREN)]+BF4
[CuI(Me6TREN)]+BF4
[CuI(Me6TREN)]+BF4
CuBr/Me6TREN
CuBr/Me6TREN
CuBr/Me6TREN
CuBr/Me6TREN
CuBr/Me6TREN
H2O (200)
D2O (200)
CH3OH (2)
CH3OH (40)
CH3OH (200)
CD3OD (2)
CD3OD (40)
CD3OD (200)
CH3OH (20)
CH3OH (100)
CH3OH (100)
CD3OD (20)
CD3OD (100)
CD3OD (100)
4.28
−
−
−
−
−
−
−
−
12
13
14
15
16
17
18
19
20
21
CD3CN
CD3CN
CD3CN
CD3CN
CD3CN
CD3CN
CD3CN
CD3CN
C6D6
C6D6
C6D6
C6D6
C6D6
1.09
0.20
1.65
3.17
0.23
0.15
85.0
17.2
13.8
65.3
e
0.94
1.78
8.64
7.81
0.35
3.30
4.36
d
22
23
24
54.7
68.7
68.9
d
25
C6D6
CuBr/Me6TREN
a
Experiments in CD3CN: [CuI(MeCN)4]BF4/Me6TREN/AIBN = 1:2:0.5, T = 80 °C, 8 h. Experiments in C6D6: [CuBr]/Cu0/Me6TREN/AIBN
b
= 1:4:1.7:1, T = 75 °C, 8 h (unless otherwise stated). Ratio of the two Me2C resonances (see Supporting Information), hence equivalent to the
moral ratio of AIBN leading to each product. Fraction of Me2CD(CN) in the iBN product, determined by H NMR integration. Thermal
decomposition for 24 h in C6D6 that was dried and distilled over Na/K. A small amount of amide (Me2CHCONH2 or Me2CHCOND2, <1%) was
c
d
1
e
also detected.
observed dimer/iBN product ratio was about the same (ca.
10:1) in both CD3CN and C6D6. Additional control
experiments were carried out in the presence of proton
donors, still in the absence of any copper complex. Both water
and methanol (regular or deuterated) were used in CD3CN,
whereas only methanol could be used in C6D6 for miscibility
reasons. The results (entries 3−6 in CD3CN, entries 7 and 8 in
C6D6) do not greatly differ from those obtained in the absence
of these proton-donor additives. The only slight difference was
the detection of minor amounts of isobutyramide,
Me2CHONH2, resulting from hydrolysis of the isobutyroni-
trile, for the experiments carried out in the presence of water.
3.b. Decomposition in the Presence of CuI Com-
plexes with Me6TREN. The Me6TREN ligand was selected
for the initial studies in the presence of CuI (Table 2).
Termination experiments were first carried out without any
proton donor additive (entries 9 and 10 in CD3CN and C6D6,
respectively). The results, in both solvents, are again quite
similar to those obtained in the absence of copper complex,
showing that the presence of the complex alone does not
significantly affect the radical decomposition mode. A marginal
increase of the iBN production in CD3CN may result from the
presence of small amounts of water in the deuterated solvent
(which was found to be ca. 0.05 equiv relative to AIBN by 1H
NMR integration), but the difference is barely significant. The
was deuterated (Me2CD(CN)). The relative amount of this
product could be rather precisely determined by H NMR
1
integration of the Me proton resonances (1:1:1 triplet for the
deuterated product, vs doublet for the nondeuterated
compound; see section S1 in the Supporting Information).
The dimer/iBN difference for H2O and D2O may be attributed
to a kinetic isotope effect (KIE) in the reaction leading to the
iBN product, while the imperfect deuteration of iBN may be
attributed to the incorporation of the residual H in D2O, plus
the H2O originally present in the CD3CN solvent, in the
presence of the KIE bias.
Use of regular and deuterated methanol (entries 14−16 and
17−19, respectively) as a proton donor gave the same trends.
In this case, a finer study as a function of proton donor
concentration was carried out, showing that the iBN/dimer
ratio increases with the methanol/AIBN ratio. This ratio is
greater with CH3OH than with CD3OD under identical
conditions, again consistent with the presence of a KIE. It is
however smaller than with the same amount of water (cf.
entries 16 and 12, or 19 and 13). This shows that water is
slightly more effective than methanol for generating the
reduced iBN product. The observed deuterium incorporation
was smaller with CD3OD than with D2O, reflecting a greater
impact of the residual H in the deuterated reagent or in the
solvent.
The same trends are also found for the experiments run with
the CuIBr/Cu0/Me6TREN/methanol combination in C6D6
(entries 20−25 in Table 2). There is again a greater fraction
of iBN with an increased amount of proton donor, and the iBN
fraction is greater with CH3OH than with CD3OD. In order to
tell whether the residual H in iBN originates from the CD3OD
reagent or from the solvent, the experiments were repeated
with dried C6D6 (entries 22 and 25). The results, particularly
the percent D incorporation in iBN, are essentially identical to
those obtain in untreated C6D6, pointing to the CD3OD
reagent as the most probable H source.
1
deuterated benzene is much drierno detectable H2O by H
NMRand the results (entry 10) are essentially identical to
those of the control in entry 2 (Table 1). An additional control
experiment run in C6D6 that had been dried and distilled over
Na/K (entry 11) gave a marginally lower iBN/dimer ratio.
The deliberate addition of 200 equiv of water (relative to
AIBN, or 100 equiv relative to the Cu complex) in MeCN
dramatically changed the product distribution in favor of iBN
(Table 2, entry 12). Replacement of water with the same
amount of D2O (entry 13) yielded a lower amount of iBN
relative to entry 12, but it is still much greater than in the
control experiment of entry 9, and nearly 90% of this product
E
DOI: 10.1021/acs.inorgchem.9b00660
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Table 3. Results of the Thermal Decomposition of AIBN in the Presence of CuI/TPMA Systems
b
c
entry
solvent
CuI complex
additive/AIBN
iBN/dimer
% D in iBN
−
26
27
28
29
30
31
32
33
34
35
CD3CN
C6D6
[CuI(TPMA)]+BF4
CuBr/TPMA
−
−
0.09
0.10
0.10
0.14
3.6
0.11
0.12
2.35
4.66
2.85
−
−
−
−
−
−
CD3CN
CD3CN
CD3CN
CD3CN
CD3CN
CD3CN
C6D6
[CuI(TPMA)]+BF4
[CuI(TPMA)]+BF4
[CuI(TPMA)]+BF4
[CuI(TPMA)]+BF4
[CuI(TPMA)]+BF4
[CuI(TPMA)]+BF4
CuBr/TPMA
CH3OH (2)
CH3OH (40)
CH3OH (200)
CD3OD (2)
CD3OD (40)
CD3OD (200)
CH3OH (100)
CD3OD (100)
1.3
9.7
70.5
C6D6
CuBr/TPMA
87.4
a
Experiments in CD3CN: [CuI(MeCN)4]BF4/TPMA/AIBN = 1:2:0.5, T = 80 °C, 8 h. Experiments in C6D6: [CuBr]/Cu0/TPMA/AIBN =
b
1:4:1.7:1, T = 75 °C, 8 h. Ratio of the two Me2C resonances (see Supporting Information), hence equivalent to the moral ratio of AIBN leading to
each product. Fraction of Me2CD(CN) in the iBN product, determined by H NMR integration.
c
1
a
Table 4. Results of the Thermal Decomposition of AIBN in the Presence of CuI/BIPY Systems
b
c
entry
solvent
CuI complex
additive/AIBN
iBN/dimer
% D in iBN
−
36
37
38
39
40
41
42
43
44
45
CD3CN
C6D6
[CuI(BIPY)2]+BF4
CuBr/BIPY
−
−
0.11
0.07
0.10
0.08
0.11
0.11
0.12
0.15
0.13
0.13
−
−
−
−
−
−
CD3CN
CD3CN
CD3CN
CD3CN
CD3CN
CD3CN
C6D6
[CuI(BIPY)2]+BF4
[CuI(BIPY)2]+BF4
[CuI(BIPY)2]+BF4
[CuI(BIPY)2]+BF4
[CuI(BIPY)2]+BF4
[CuI(BIPY)2]+BF4
CuBr/BIPY
CH3OH (2)
CH3OH (40)
CH3OH (200)
CD3OD (2)
CD3OD (40)
CD3OD (200)
CH3OH (100)
CD3OD (100)
2.4
3.6
4.2
C6D6
CuBr/BIPY
38.2
a
Experiments in CD3CN: [CuI(MeCN)4]BF4/BIPY/AIBN = 1:4:0.5, T = 80 °C, 8 h. Experiments in C6D6: [CuBr]/Cu0/BIPY/AIBN = 1:4:3.4:1,
b
T = 75 °C, 8 h. Ratio of the two Me2C resonances (see Supporting Information), hence equivalent to the moral ratio of AIBN leading to each
product. Fraction of Me2CD(CN) in the iBN product, determined by H NMR integration.
c
1
adduct, [Cu(κ4-Me6TREN)(CD3CN)]+BF4 , like the related
crystallographically characterized [Cu(κ4-TPMA)-
−
A comparison of the results obtained in the two different
solvents shows that a greater iBN fraction was obtained in
C6D6 (for instance, cf. entries 21 and 16 or entries 24 and 19),
even though with a lower methanol/AIBN ratio. It is notable,
however, that small amounts of the radical−radical coupling
product (dimer) were always present, even when using 100
equiv of CH3OH. This result is in stark contrast with the
results reported by Yamago et al.,28 where the PMA• radicals
produced by PMA−Br activation were fully quenched to
PMA−H by CuI in the presence of only 10 equiv of CH3OH
or CD3OD, with no visible detection of coupling product.
If RRT takes place according to the OMRP pathway
(Scheme 6), the greater iBN fraction in C6D6 relative to
CD3CN may be attributed to either a more favorable
formation of the organometallic intermediate or to the greater
acidity of the proton donor in this solvent or to a contribution
of both factors. The first hypothesis requires that a
coordination site for the radical addition is more easily
available in C6D6. The CuI complex in this solvent is likely to
be pentacoordinated, [CuBr(κ4-Me6TREN)], like the crystallo-
graphically characterized related [CuBr(κ4-TPMA)],56 and the
resulting organocopper(II) adduct is also likely to adopt a 5-
coordinated structure. Thus, the radical addition requires
either dissociation of the bromide ion, [Cu{C(CN)Me2}(κ4-
Me6TREN)]+Br−, which is unlikely in a nonpolar solvent, or
dissociation of one of the Me6TREN N donor atoms,
[CuBr{C(CN)Me2}(κ3-Me6TREN)]. On the other hand, in
CD3CN the CuI complex may be either tetracoordinated,
57
(CH3CN)]+BPh4 . Hence, radical addition would require
in this case displacement of the solvent molecule. The relative
acidity (pKa) of methanol in C6D6 and CD3CN does not
appear to be known. Acid dissociation constants are available
in polar solvents but cannot be easily assessed in nonpolar ones
such as benzene.
−
If RRT follows the OS-PCET pathway, the reaction should
be facilitated in the solvent with higher dielectric constant
(acetonitrile), because of the better stabilization of charged
species (the transients [CuIIBr(Me6TREN)]+ and MeO−
would initially be produced, together with the product
PMA−H, from neutral [CuIBr(Me6TREN)], PMA•, and
MeOH), hence the observed greater fraction of iBN in C6D6
seems to exclude this pathway. Finally, if an IS-PCET pathway
is followed, the observed trends when changing solvent would
be consistent with more favorable methanol coordination to
[CuIBr(Me6TREN)] in C6D6 than to [CuI(Me6TREN)]+ in
CD3CN. In terms of the water/methanol comparison in
CD3CN, either a more favorable coordination of water than
methanol to [CuI(Me6TREN)]+ in CD3CN or a greater
aptitude of the coordinated water to undergo the PCET
process would explain the observed trend.
3.c. Decomposition in the Presence of CuI Complexes
with TPMA and BIPY. Additional experiments, using copper
complexes stabilized by the TPMA and BIPY ligands, were
restricted to methanol as the source of hydrogen or deuterium.
The results obtained in the presence of TPMA are collected in
−
[Cu(κ4-Me6TREN)]+BF4 , or pentacoordinated as a solvent
F
DOI: 10.1021/acs.inorgchem.9b00660
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 3. The control experiments run in the absence of proton
donor (entries 26 and 27) yielded again results identical to
those in the absence of copper complex. The addition of
increasing amount of CH3OH or CD3OD in CD3CN (entries
28−30 or 31−33, respectively) yields an increasing amount of
iBN, with the biggest jump between 40 and 200 equiv per
AIBN. The same trend holds for the percent of D
incorporation in the iBN product. The KIE induced a greater
fraction of iBN with CH3OH relative to CD3OD (cf. entries 30
and 33). Changing solvent from CD3CN to C6D6 shows the
same trends as for the Me6TREN system, with more iBN
generated in the C6D6 solution relative to the CD3CN solution
(e.g., cf. entries 34 and 30 or entries 35 and 33).
Table 4 reports all the results obtained using copper
complexes in combination with the BIPY ligand. In this case,
increasing the amount of methanol, whether regular or
deuterated, whether in CD3CN or in C6D6, did not
significantly alter the iBN/dimer ratio relative to the control
experiments run in the absence of copper complex. On the
other hand, there was a slight D incorporation when using
CD3OD. This D incorporation was much less significant in
CD3CN than in C6D6.
3.d. Probing OS-PCET. If RRT occurs via OS-PCET
(Scheme 6), the copper(I) complex is oxidized to [CuII(L)]2+
and the radical is reduced to anion, with concurrent or
subsequent proton transfer from ROH(D). Either way, the
aptitude to follow this decomposition pathway relative to RT
(or CRT) should scale with the standard reduction potentials
(E°) of the copper complexes. The values of E° for the
[CuII(L)]2+/[CuI(L)]+ redox couples with BIPY, TPMA, and
Me6TREN are 0.18, 0.02, and −0.07 V vs SCE in
acetonitrile.58 The iBN/dimer ratio, reflecting the aptitude to
undergo RRT for the three complexes and 200 equiv of
CH3OH or CD3OD in CD3CN is shown in Figure 1. The
charged species from neutral (or less charged) ones should be
facilitated in solvents with a greater dielectric constant. Finally,
a concerted OS-PCET process would be a termolecular
reaction with an inherently low probability.59
3.e. Investigation of Water and Methanol Coordina-
tion to [CuI(L)]+. The IS-PCET pathway requires coordina-
tion of the proton donor to the CuI/L activator. Thus, the
aptitude of the two proton donors (H2O and CH3OH in
CD3CN, CH3OH in C6D6) to yield equilibrium amounts of
1
CuI adducts was investigated by H NMR spectroscopy, using
the Me6TREN system. These investigations, however, could
not give unambiguous evidence in favor of an interaction. The
1H NMR spectra recorded for the [CuI(Me6TREN)]+/H2O
experiment in CD3CN are shown in the Supporting
Information, section S2. Before the addition of water, the
spectrum shows the three expected signals of the Me6TREN
ligand in a 3:1:1 ratio, although broad, plus an additional small
resonance at 2.23 ppm probably due to the water impurity that
was already present in the NMR solvent. The observation of
broad 1H NMR resonances is a common phenomenon for CuI
complexes, caused by the presence of paramagnetic CuII
impurities. These possibly originate from disproportionation
of the in situ-generated [CuI(Me6TREN)]+. However, this
equilibrium is not extensive in MeCN (Kdisp = 10−3).60
Addition of increasing amounts of H2O results in a continuous
downfield shift of the H2O resonance while the Me6TREN
resonances further broaden. A rapid exchange between free
and coordinated H2O that would only involve the diamagnetic
CuI complex would result instead in a signal sharpening and
displacement toward the typical chemical shift value of free
water in acetonitrile (2.13 ppm).61 Therefore, the observed
behavior rather suggests a rapid exchange with a paramagnetic
CuII(H2O) adduct, which would be generated in increasing
proportions by addition of water, because the latter favors
disproportionation of the [CuI(Me6TREN)]+ complex. Sim-
ilarly, electrochemical or spectrochemical titrations of Cu
complexes in the presence of various amounts of water did not
provide conclusive evidence on water binding to CuI (see
Supporting Information, section S3). It is therefore impossible
to obtain even qualitative evidence in favor of water binding to
CuI from this experiment.
The analogous addition of CH3OH to a [CuI(Me6TREN)]+
solution in CD3CN provides analogous results, see Supporting
Information, section S4. Notably, the Me6TREN ligand
resonances broaden upon addition of increasing methanol
amounts. At the same time, the OH resonance, while
remaining relatively sharp, diverges toward lower-field position
rather than converging toward the value of free CH3OH in
CD3CN. It is also notable that the coupling between the CH3
(doublet) and OH (quartet) protons for free CH3OH, which is
visible in the absence of the Cu complex (spectrum also shown
in the Supporting Information), is washed out in the presence
of the metal complex. The corresponding study of the addition
of CH3OH to [CuIBr(Me6TREN)] in C6D6 exhibits again the
same behavior. In this case, the methanol OH resonance is not
clearly visible in the absence of copper, in agreement with the
literature,61 but becomes observable as a weak and broad
resonance in the presence of the copper complex, once again
broadening and shifting downfield as the methanol amount is
increased (Supporting Information, section S5).
Figure 1. Observed isobutyronitrile(iBN)/tetramethylsuccinonitrile-
(dimer) ratio for the AIBN decomposition in the presence of 200
equiv of CH3OH or CD3OD and the [CuI(L)]+ complexes (L =
(BIPY)2, Me6TREN, TPMA) in CD3CN.
value of E° agrees with [CuI(BIPY)]+ yielding the lowest
degree of RRT, because it is the weakest reducing agent.
However, the trend of reduction potentials for [CuI(TPMA)]+
and [CuI(Me6TREN)]+ does not match the experimental
degrees of RRT, which is greater for the weaker reducing agent
[CuI(TPMA)]+. Therefore, the electrochemical properties of
the catalysts appear to rule out OS-PCET as the preferred
mechanism for this decomposition pathway. Further evidence
against the OS-PCET pathway is provided by the iBN/dimer
ratio (hence the impact of RRT), which is greater in benzene
than in MeCN, whereas reactions involving the generation of
3.f. DFT Calculations. Since no experimental methods are
available to assess the ability of the various copper systems to
bind proton donor molecules on one side and the
G
DOI: 10.1021/acs.inorgchem.9b00660
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 5. DFT-Estimated Gibbs’ Free Energies (ΔG298.15,MeCN,1M in kcal/mol) for the Formation of [CuI(L)n(CH3CN)]+,
[CuI(L)n(H2O)]+, and [CuII{CMe2(CN)}(L)n]+ from [CuI(L)n]+ plus CH3CN, H2O, or Me2C•(CN), Respectively (L =
Me6TREN and TPMA, n = 1; L = BIPY, n = 2)
L
[(L)nCuI−NCCH3]+
[(L)nCuI−OH2]+
[(L)nCuII−CMe2(CN)]+
[(L)nCuII−NCCMe2]+
Me6TREN, n = 1
TPMA, n = 1
BIPY, n = 2
−2.6
−0.5
+5.8
−1.1
−5.9
a
−6.6
−11.9
a
−12.9
−12.6
+2.9
a
A stable adduct could not be optimized (see text).
Table 6. Selected Structural Parameters (Distances in Å, Angles in deg) for the Optimized [CuII(L)n(CMe2CN)]+,
[CuII(L)n(NCCMe2)]+, [CuI(L)n((CH3CN)]+, and [CuI(L)n((H2O)]+ Structures
parameter
Me3TREN
TPMA
(BIPY)2
[CuII(L)n(CMe2CN)]+ systems
Cu−C
Cu−Nax
2.160
2.260
2.134
2.223
Cu−Neq
2.384, 2.409, 2.363
2.385
171.46
2.210, 2.211, 2.207
2.209
174.69
average Cu−Neq
C−Cu−Nax
C−Cu−Neq
99.26, 101.40, 100.96
102.36, 104.82, 102.78
[CuII(L)n(NCCMe2)]+ systems
Cu−N
1.983
2.008
2.040
Cu−Nax
Cu−Neq
average Cu−Neq
N−Cu−Nax
N−Cu−Neq
2.227
2.315, 2.321, 2.214
2.283
2.307
2.094, 2.150, 2.166
2.137
2.168, 2.220 (eq)
2.125, 2.164 (ax)
120.67, 137.39 (eq)
95.04, 89.44 (ax)
93.72, 96,21, 105.17
106.98, 100.83, 100.20
[CuI(L)n(CH3CN)]+ systems
Cu−N (MeCN)
Cu−Nax
Cu−Neq
N−Cu−Nex
N−Cu−Neq
1.976
2.332
2.038
2.438
1.984
2.108, 2.108 (eq)
2.325, 2.323 (ax)
125.25, 125.41 (eq)
93.75, 93.76 (ax)
2.336, 2.321, 2.342
179.29
100.79, 101.39, 100.53
2.142, 2.141, 2.143
179.93
105.08, 105.12, 105.04
[CuI(L)n(CH3CN)]+ systems
Cu−O
2.435
2.458
Cu−Nax
2.280
2.315
Cu−Neq
O−Cu−Nax
O−Cu−Neq
2.225, 2.224, 2.233
179.69
96.64, 97.16, 97.09
2.071, 2.078, 2.065
164.97
93.94, 92.30, 114.76
cyanoisopropyl radical on the other, we turned to computa-
tional studies. These were restricted to the cationic [CuI(L)n]+
complexes in acetonitrile solution and their interaction with
water and with the cyanoisopropyl radical. However, since the
complexes are likely to be solvated in acetonitrile, the 5-
coordinate [CuI(L)n(CH3CN)]+ system was also considered.
The calculations were carried out using a DFT approach (see
Computational Details), using the full molecular systems with
dispersion and solvation corrections in acetonitrile. The
calculated values are reported as Gibbs free energies at
298.15 K in the condensed phase 1 M standard state. On the
basis of our previous work, we believe that the chosen
functional/basis set combination gives relatively accurate
energy estimates,20 but we do not put a large amount of
faith on the quantitative results. However, we believe that the
calculated trends should accurately reproduce the experimental
reality. The interaction energies are shown in Table 5, and
selected optimized structural parameters are reported in Table
6, while views and Cartesian coordinates of all optimized
geometries are reported in the Supporting Information
(section S5).
The binding energies for CH3CN are very small and
negative (i.e., the interaction is favorable) only for the
Me6TREN and TPMA systems. The CH3CN adduct for the
BIPY system, although giving an optimized local energy
minimum, is endoergic relative to the sum of the separate
[CuI(BIPY)2]+ and MeCN. The lower binding aptitude of the
BIPY system may be associated to the greater stability of the
[CuI(BIPY)2]+ tetrahedral geometry, whereas the geometry of
the complexes with tripodal ligands, [CuI(Me6TREN)]+ and
[CuI(TPMA)]+, is strained. The trigonal bipyramidal structure
of the [CuI(L)n(CH3CN)]+ product, on the other hand, can
better adjust all ligands.
Water binding also appears weak according to the
calculations. For the Me6TREN system, binding water is less
favorable than acetonitrile, though still slightly exoergic relative
to the 4-coordinated unsaturated complex. For the TPMA
system, on the other hand, the calculations indicate that water
binds better than acetonitrile. For the BIPY system, a local
minimum for a CuI-bound water adduct could not even be
located, yielding instead a van der Waals [CuI(BIPY)2···OH2]+
adduct that features a loose interaction between the O atom
and a CH group of one of the BIPY aromatic rings.
H
DOI: 10.1021/acs.inorgchem.9b00660
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
The interaction with the cyanoisopropyl radical was found
possible in two different ways, leading to C-bonded and N-
bonded isomers. In the free radical, the spin density resides
mostly on the tertiary C atom (0.731), but a significant
amount (0.342) is also located on the cyano N atom. The most
relevant spin density values for all paramagnetic systems are
shown in Figure 2. Metal binding through the C atom, even
interaction, plus a comparatively reduced but nonzero spin
density on the cyanoisopropyl N atom; see Figure 2.
Strikingly, the alternative binding of the cyanopropyl radical
through the N atom, leading to the keteniminato [(L)n(CuII−
NCCMe2]+ isomer, is even more favorable. It is worth
noting that metal binding of a cyano-stabilized radical via the
N atom has been previously proposed for the addition of the
•C(CN)(Me)CH2C(OMe)Me2 radical generated from 2,2′-
azobis(4-methoxy-2,4-dimethylvaleronitrile) (V-70) to Co-
(acac)2 (acac = acetylacetonate), on the basis of both
experimental and computational evidence.62 Other ketenimi-
nato complexes of transition metals are also known, such as
[IrCp*(NCCPh2)(Ph)(PMe3)] (Cp* = pentamethylcy-
clopentadienyl)63 and [Ru(dmpe)2(H)(NCCHAr)] (Ar
= Ph, p-C6H4CF3; dmpe = 1,2-bis(dimethylphosphino)-
ethane),64,65 although they were not obtained by radical routes.
The energetic advantage for N-binding is greater for the
Me6TREN system (6.3 kcal/mol) than for the TPMA system,
for which the N-bonded complex is more stable than the C-
bonded isomer by only 0.7 kcal/mol. The reason for this
difference in relative stabilization can be appreciated from the
structural parameters in Table 6. The NMe2 groups of
Me6TREN generate a significantly greater steric repulsion
than the TPMA pyridine rings on the C-bonded tertiary
radical, which is greater than the steric repulsion exerted on the
isomeric N-bonded fragment. This can be judged by the
greater increase of the Cu−Neq distances from the N-bound to
the C-bound isomer (the average CuII−Neq distance increases
from 2.227 to 2.385 Å, a 0.158 Å increase, for the Me6TREN
system, vs an increase from 2.137 to 2.209 Å, only a 0.072 Å
increase, for the TPMA system). The N-bound isomer yields a
stable minimum also for the less reactive BIPY system, but this
is endoergic relative to the separate reagents. In these three
keteniminato CuII derivatives, the spin density is predom-
inantly located on the isopropyl C atom (0.472, 0.471, and
0.517 for Me6TREN, TPMA, and BIPY systems, respectively)
with the rest equally shared by the directly bonded Cu and N
atoms. In this respect, the assignment of the oxidation state II
to the copper atom in these radical adducts is just a formality,
because the majority of the spin density, particularly for the N-
bound keteniminato isomer, still resides on the cyanoisopropyl
group. It is of interest to note that the related cyanomethyl
group (CH2CN) was found bonded in a CuII complex as a
cyano-substituted alkyl ligand via the C atom, rather than as a
keteniminato ligand via the N atom.66
Figure 2. Views of all optimized paramagnetic structures with the
main Mulliken spin densities.
though the radical is stabilized and tertiary, is energetically
favored for the most reactive Me6TREN and TPMA systems.
On the other hand, no stable energy minimum could be
located for the less reactive BIPY system. Instead, the geometry
optimization for this system converged to a van der Waals
[CuI(BIPY)2···NC(CMe2)]+ adduct, similar to that already
described with water (loose interaction between the N atom
and a CH group of a BIPY aromatic rings). In this adduct,
which is located at +3.7 kcal/mol relative from the two
separate reagents, the spin density is entirely located on the
cyanoisopropyl group, with values on the C and N atoms very
close to those of the free radical, while the Cu atom has no
spin density. On the other hand, the organocopper(II)
products with the two tripodal ligands show an approximately
equal share of the spin density between the Cu atom and the
Cu-bonded C atom, in agreement with the weak bonding
In conclusion, the DFT calculations indicate that there is
substantial difference between the two tripodal ligands
Table 7. Results of the Thermal Decomposition of AIBN in the Presence of [Et3NH]+ Salts as Proton Donors and Different
a
CuI/L Systems
b
entry
anion
CuI complex
[Et3NH]+/AIBN
iBN/dimer
−
46
47
48
49
50
51
52
53
BF4
BF4
BF4
Br−
BF4
BF4
BF4
BF4
−
c
2
40
2
2
40
2
0.05
5.11
7.68
5.56
2.58
9.81
0.15
1.18
[CuI(Me6TREN)]+BF4
[CuI(Me6TREN)]+BF4
[CuI(Me6TREN)]+BF4
−
−
−
−
−
[CuI(TPMA)]+BF4
−
−
[CuI(TPMA)]+BF4
−
−
[CuI(BIPY)2]+BF4
−
−
[CuI(BIPY)2]+BF4
40
−
−
a
b
[CuI(MeCN)4]BF4/L/AIBN = 1:(2 for L = Me6TREN and TPMA; 4 for L = BIPY):0.5, T = 80 °C, 8 h. Ratio of the two Me2C resonances (see
c
Supporting Information), hence equivalent to the moral ratio of AIBN leading to each product. [Et3NH]+/AIBN = 2.
I
DOI: 10.1021/acs.inorgchem.9b00660
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Me6TREN and TPMA on one side and the bidentate BIPY
system on the other one. The two systems with the tripodal
ligands have greater aptitude than that with BIPY to bind both
water and the cyanoisopropyl radical. While [CuI(BIPY)2]+
shows no thermodynamic gain for binding either a neutral
molecule or the organic radical, the Me6TREN and TPMA
systems can be stabilized by addition of both a neutral donor
molecule (acetonitrile, water) and the cyanoisopropyl radical.
The stabilization provided by the radical is much greater and
appears more favorable when binding occurs via the N atom to
yield a less bulky CuII−NCCMe2 moiety than via the C
atom to yield an organometallic derivative. The experimental
observation of a much greater impact of [H2O] and [CH3OH]
on the product distribution for the Me6TREN and TPMA
system than for the BIPY system is thus consistent with both
the OMRP and the IS-PCET pathways in RRT (Scheme 6).
3.g. Protonolysis with Triethylammonium Salts. In
order to gain additional information on the RRT mechanism,
we have finally opted for the use of a proton donor that is
devoid of coordinating properties, namely a triethylammonium
salt. This proton donor cannot operate via the IS-PCET
pathway, whereas its greater acidity relative to water and
methanol should enhance the OMRP pathway. Two different
[Et3NH]+ salts, bromide and tetrafluoroborate, were used for
the AIBN decomposition experiments, which were conducted
only in CD3CN for solubility reasons. The results are collected
in Table 7. The blank experiment run in the absence of copper
complex (entry 46) yields a very close iBN/dimer ratio to that
of the decomposition without proton donor (Table 1, entry 1).
The results clearly show that the Et3NH+ salts enhance RRT
relative to water and methanol. With Me6TREN, an RRT/
dimerization ratio of 5.11 was obtained when using only 2
adduct with Me2C•(CN) provides greater stability than
coordination of water or acetonitrile. The stronger stabilization
predicted for the Me6TREN and TPMA system is in line with
the greater fraction of RRT product obtained for those
systems. Concerning the BIPY system, the DFT calculations
show a slightly endoergic adduct formation with the radical via
the N atom (e.g., the keteniminato form). However, the
relative energy of this derivative makes it a thermally accessible
intermediate for the RRT process. Therefore, it seems that the
agreement between the calculations and the experimental
result (for the RRT in the presence of Et3NH+) is good even at
the semiquantitative level. As a final note, we can conclude that
while the experiments run with Et3NH+ definitely prove the
intervention of a pathway involving an adduct between the
radical and the CuI/L complex, this does not exclude the
contribution (or even the exclusive action) of an IS-PCET
pathway in the presence of water or methanol.
4. CONCLUSION
We have analyzed the distribution of the cyanoisopropyl
radical termination products in the presence of proton donors
(water, methanol, and triethylammonium salts) and copper(I)
complexes stabilized by three different ligand systems
(Me6TREN, TPMA, and BIPY) in two different solvents
(CD3CN, C6D6). The simultaneous presence of copper and
proton donor provides access to a reductive radical termination
(RRT) with formation of isobutyronitrile by stoichiometric
transfer of an electron from the copper complex and by a
proton from the acidic additive. The extent of RRT is affected
by the nature of the Cu complex, of the proton donor and of
the solvent in the orders BIPY ≪ Me6TREN ∼ TPMA,
CD3OD < CH3OH, D2O < H2O, CH3OH(D) < H(D)2O ≪
Et3NH+, and CD3CN < C6D6.
equiv of Et3NH+BF4 per AIBN (entry 47), whereas the
−
equivalent amount of CH3OH gives a ratio of 0.20 (Table 2,
entry 14). This ratio further increased to 7.68 with 40 equiv
(entry 48), vs 1.65 with the same amount of CH3OH (Table 2,
entry 15). These ratios are even greater than that obtained
with 200 equiv of H2O (4.28, Table 2, entry 12). Using
Et3NH+Br− gave an essentially identical result (entry 49) to
that of the BF4− salt, in agreement with the weak association of
Br− to [CuI(Me6TREN)]+ in acetonitrile.67 For the TPMA
complex, the RRT is equally enhanced (entries 50−51, cf. with
the equivalent amounts of CH3OH in Table 3, entries 28 and
29). Finally, even the BIPY system, for which a low impact of
RRT was shown even in the presence of 200 equiv of CH3OH
(iBN/dimer = 0.11, Table 4, entry 40), exhibits an enhance-
ment of RRT when using the Et3NH+ salt (entries 52 and 53),
even though the impact of RRT remains largely inferior with
respect to the use of the more reducing complexes with
Me6TREN and TPMA.
The most surprising and unexpected result of this study, as
suggested by the DFT calculations, is the tendency of the most
reactive [Cu(L)n]+ complexes (i.e., those stabilized by the
Me6TREN and TPMA ligands, n = 1) to trap the
cyanoisopropyl radical, a stabilized radical for which we
considered trapping by a transition metal to be unlikely, and
yield stabilized adducts that may be formally described as
copper(II) complexes, though the spin density transfer from
the organic radical to the metal is quite partial. No evidence for
such an interaction was found for the less reactive system (L =
BIPY, n = 2). It is also relevant to note that CuI is not known
for forming stable organocopper(II) derivatives, even with
strong radicals.45 Even more surprising and unexpected is the
indication by the DFT calculations that the N-bound
keteniminato isomer is more stable than the C-bound
cyanoisopropyl organometallic derivative, particularly for the
more sterically demanding Me6TREN system.
Thus, it clearly appears that RRT occurs primarily by what
has been defined as the “OMRP route” in Scheme 6, although
the DFT result rather support the formation of an N-bound
CuII intermediate, rather than an organometallic complex. The
acidic additive (water, methanol, or Et3NH+) can then provide
a proton to cleave the Cu−N bond heterolytically and the
extent of this process is highly dependent on the proton donor
acidity. The observed isobutyronitrile product can result from
either direct protonation of the conjugated C atom, or cleavage
of the CuII−N bond with release of the ketenimine Me2C
CNH, which subsequently tautomerizes.68
Coordination of donor molecules (acetonitrile, water) to the
[Cu(L)n]+ complexes appears possible but is quite weak,
according to the DFT calculations, for the two most reactive
systems and not favorable at all for the BIPY system. An NMR
investigation has not provided positive evidence for the
presence of this interaction for water or methanol and the
Me6TREN system. Finally, the large extent of RRT for the
Et3NH+ proton donor, which is at the same time the strongest
acid of the series and has no coordinating properties, indicates
that RRT proceeds via an intermediate CuII adduct, followed
by protonolysis, rather than by an IS-PCET process. A
potential alternative OS-PCET is excluded by the solvent effect
and by the redox potentials trend. A possible contribution of
The experimental results are actually rather consistent with
the DFT predictions, because formation of the [CuI(L)n]+
J
DOI: 10.1021/acs.inorgchem.9b00660
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Notes
IS-PCET to RRT, however, cannot be excluded for proton
donors able to coordinate to the metal center.31
The authors declare no competing financial interest.
Overall, while this study is consistent with the proposition,
made in a previous contribution,28 that the saturated PMA−H
made by reductive termination of PMA• chains in the presence
of methanol in toluene solution may originate from the
protonolysis of a PMA−CuIIBr/Me6TREN intermediate, the
contribution (or dominance) of alternative PCET pathways
cannot yet be totally excluded. In addition, we have shown in
our study that, at least for the termination of the
cyanoisopropyl radical, the C6D6 solvent does not contain
sufficient amounts of water impurities to yield a significant
impact of RRT, even for the most active CuIBr/Me6TREN
system. Yet, the PMA• termination in the absence of methanol
was shown in the previous contribution28 to yield dominant
fraction of PMA−H product, not resulting from disproportio-
nation. This puzzling result merits further studies, currently
ongoing in our laboratories, which will be reported in a
forthcoming contribution. Another puzzling result is that,
although the occurrence of RRT (stoichiometrically oxidizing
the ATRP activator complex L/CuI to an inactive L/CuIIX
complex) is promoted by water and other proton donors, the
complexes investigated in the present study are quite robust
ATRP catalysts even in water, to give well-controlled halogen-
capped macromolecules with high chain-end fidelity.69,70 A full
understanding of how radicals interact with L/CuI complexes
under polymerization conditions is paramount not only to
rationalize the good performance of ATRP catalysts and
possibly further improve them but also for implementing
polymerizations successfully controlled by only OMRP, which
presents certain advantages in particular for less activated
monomers,71−74 without interference of metal-catalyzed or
-mediated terminations (CRT or RRT).
ACKNOWLEDGMENTS
■
We are grateful to the CNRS for funding this work via the
International Associated Laboratory entitled “Laboratory of
Coordination Chemistry for Controlled Radical Polymer-
ization (LCC-CRP)” and to the National Science Foundation
(Grant CHE 1707490). We also gratefully acknowledge
CALMIP (Calcul en Midi-Pyrenees) for granting access and
free CPU time on the University of Toulouse mesocenter and
́
́
̀
́
the Ministere de l’Enseignement Superieur et de la Recherche
Scientifique (MESRS) for a Ph.D. fellowship to L.T.
ABBREVIATIONS
■
ATRP, atom transfer radical polymerization; CRT, catalytic
radical termination; OMRP, organometallic-mediated radical
polymerization; RRT, reductive radical termination; AIBN,
azobis(isobutyronitrile); BIPY, 2,2′-bipyridine; Me6TREN,
tris[2-(dimethylamino)ethyl]amine; TPMA, tris(2-
pyridylmethyl)amine
REFERENCES
■
(1) Poli, R. Radical coordination chemistry and its relevance to
metal-mediated radical polymerization. Eur. J. Inorg. Chem. 2011,
2011, 1513−1530.
(2) Fischer, H. The Persistent Radical Effect: A Principle for
Selective Radical Reactions and Living Radical Polymerizations.
Chem. Rev. 2001, 101 (12), 3581−3610.
(3) Jenkins, A. D.; Jones, R. G.; Moad, G. Terminology for
reversible-deactivation radical polymerization previously called
“controlled” radical or “living” radical polymerization (IUPAC
Recommendations 2010). Pure Appl. Chem. 2009, 82 (2), 483−491.
(4) Wang, J.-S.; Matyjaszewski, K. Controlled/“living” radical
polymerization. Atom transfer radical polymerization in the presence
of transition-metal complexes. J. Am. Chem. Soc. 1995, 117, 5614−
5615.
(5) Kato, M.; Kamigaito, M.; Sawamoto, M.; Higashimura, T.
Polymerization of methyl methacrylate with the CCl4/(PPh3)2Cl2Ru-
(II)/MeAl(ODBP)2 iniating system: possibility of living radical
polymerization. Macromolecules 1995, 28, 1721−1723.
(6) Matyjaszewski, K.; Tsarevsky, N. V. Macromolecular Engineer-
ing by Atom Transfer Radical Polymerization. J. Am. Chem. Soc. 2014,
136 (18), 6513−6533.
(7) Ouchi, M.; Terashima, T.; Sawamoto, M. Transition Metal-
Catalyzed Living Radical Polymerization: Toward Perfection in
Catalysis and Precision Polymer Synthesis. Chem. Rev. 2009, 109
(11), 4963−5050.
(8) Matyjaszewski, K. Advanced Materials by Atom Transfer Radical
Polymerization. Adv. Mater. 2018, 30 (23), 1706441.
(9) Ribelli, T. G.; Lorandi, F.; Fantin, M.; Matyjaszewski, K. Atom
Transfer Radical Polymerization: Billion Times More Active Catalysts
and New Initiation Systems. Macromol. Rapid Commun. 2019, 40 (1),
1800616.
(10) Poli, R. Organometallic-Mediated Radical Polymerization.
Polymer Science: A Comprehensive Reference. 2012, 3, 351−375.
(11) Smith, K. M.; McNeil, W. S.; Abd-El-Aziz, A. S. Organo-
metallic-Mediated Radical Polymerization: Developing Well-Defined
Complexes for Reversible Transition Metal-Alkyl Bond Homolysis.
Macromol. Chem. Phys. 2010, 211 (1), 10−16.
(12) Allan, L. E. N.; Perry, M. R.; Shaver, M. P. Organometallic
mediated radical polymerization. Prog. Polym. Sci. 2012, 37 (1), 127−
156.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.9b00660.
Details of the NMR determination of the product
distribution, on the NMR investigation of the interaction
of proton donors and CuI complexes, and DFT results
(Cartesian coordinates, energies, spin densities, and
optimized geometrical parameters) (PDF)
AUTHOR INFORMATION
Corresponding Author
*(R.P.) E-mail: rinaldo.poli@lcc-toulouse.fr.
■
ORCID
Christophe Fliedel: 0000-0003-1442-0682
Rinaldo Poli: 0000-0002-5220-2515
Author Contributions
L.T. has carried out all the experimental work. R.P. has carried
out the DFT calculations. The manuscript was written through
contributions of all authors. All authors have given approval to
the final version of the manuscript.
Funding
Centre National de la Recherche Scientifique (CNRS): LIA
“Laboratory of Coordination Chemistry for Controlled Radical
Polymerization (LCC-CRP)”. National Science Foundation
(CHE 1707490).
(13) Le Grognec, E.; Claverie, J.; Poli, R. Radical Polymerization of
Styrene controlled by Half-Sandwich Mo(III)/Mo(IV) Couples: all
K
DOI: 10.1021/acs.inorgchem.9b00660
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Basic Mechanisms are Possible. J. Am. Chem. Soc. 2001, 123, 9513−
9524.
(32) Freiberg, M.; Mulac, W. A.; Schmidt, K. H.; Meyerstein, D.
Reations of aliphatic free radicals with copper cations in aqueous
solutions. 3. Reactions with cuprous ions - a pulse radiolysis study. J.
Chem. Soc., Faraday Trans. 1 1980, 76, 1838−1848.
(33) Sorek, Y.; Cohen, H.; Meyerstein, D. Kinetics od betah-hydroxy
elimination reactions from the protoporphyrin iron(III)-
CHRCH2OH complexes in aqueous solutions - a pulse radiolytic
study. J. Chem. Soc., Faraday Trans. 1 1986, 82, 3431−3438.
(34) Cohen, H.; Meyerstein, D. Kinetics of Formation and
Decomposition of the Methyl Copper(II) Complex in Aqueous-
Solutions-a Pulse-Radiolysis Study. Inorg. Chem. 1986, 25 (9), 1505−
1506.
(35) Cohen, H.; Meyerstein, D. Kinetics of the reaction of copper(I)
and copper(II) ions with 2,5-dioxacyclohexyl free radicals and
homolysis of the aque-copper(II)-2,5-dioxacyclohexyl complex in
aqueous solutions - a pulse radiolysis study. Inorg. Chem. 1987, 26
(14), 2342−2344.
(36) Cohen, H.; Meyerstein, D. Kinetics of beta-hydroxyl
elimination from (H2O)mCuIICH2C(CH3)2OH+ in aqueous solution
- a pulse radiolysis study. J. Chem. Soc., Faraday Trans. 1 1988, 84,
4157−4160.
(37) Goldstein, S.; Czapski, G.; Cohen, H.; Meyerstein, D.
Formation and decomposition of transient complexes with a
copper-carbon sigma bond in the reaction of copper(I) phenantroline
with aliphatic free radicals - a pusle radiolysis study. Inorg. Chem.
1988, 27 (23), 4130−4135.
(38) Masarwa, M.; Cohen, H.; Saar, J.; Meyerstein, D. Metal
induced decarboxylation of aliphatic free radicals. 1. Kinetics of the
reactions of copper(I) and copper(II) ions with the 2-methyl-2-
carboxylacid-propyl free radical in aqueous solution - a pusle radiolysis
study. Isr. J. Chem. 1990, 30 (4), 361−368.
(39) Masarwa, M.; Cohen, H.; Meyerstein, D. Mechanism of
oxidation of the 2-hydroxycyclohexyl radical to cyclopentanecarbal-
dehyde by copper ions in aqueous solutions. Inorg. Chem. 1991, 30
(8), 1849−1854.
(40) Goldstein, S.; Czapski, G.; Cohen, H.; Meyerstein, D.; Cho, J.
K.; Shaik, S. S. Deamination od 2-methyl-2-propanamine induced by
hydroxyl radicals and metal ions - a comparison between the rates of
betal-elimination of ammonia and water. Inorg. Chem. 1992, 31 (5),
798−803.
(41) Goldstein, S.; Czapski, G.; Cohen, H.; Meyerstein, D. Hydroxyl
radical induced decarboxylation and deamination of 2-methylaniline
catalyzed by copper ions. Inorg. Chem. 1992, 31 (12), 2439−2444.
(42) Goldstein, S.; Czapski, G.; Cohen, H.; Meyerstein, D.
Deamination of beta-alanine induced by hydroxyl radicals and
monovalent copper ions - a pulse radiolysis study. Inorg. Chim. Acta
1992, 192 (1), 87−93.
(43) Navon, N.; Golub, G.; Cohen, H.; Meyerstein, D. Kinetics and
Reaction-Mechanisms of Copper(I) Complexes with Aliphatic Free-
Radicals in Aqueous-Solutions - a Pulse-Radiolysis Study. Organo-
metallics 1995, 14 (12), 5670−5676.
(44) Szulc, A.; Meyerstein, D.; Cohen, H. Monovalent copper as a
potential catalyst for formation of acetaldehyde via the migration of
methyl radicals to the coordinated carbonyl in the complex (CO)Cu-
II-CH3+. Inorg. Chim. Acta 1998, 270 (1−2), 440−445.
(45) Ribelli, T. G.; Matyjaszewski, K.; Poli, R. The interaction of
carbon-centered radicals with copper(I) and copper(II) complexes. J.
Coord. Chem. 2018, 71, 1641−1668.
(46) Saba, S.; Hernandez, R.; Choy, C. C.; Carta, K.; Bennett, Y.;
Bondi, S.; Kolaj, S.; Bennett, C. A simple and efficient one-step
protocol for the preparation of alkyl-substituted ammonium
tetrafluoroborate and hexafluorophosphate salts. J. Fluorine Chem.
2013, 153, 168−171.
(14) Poli, R. Relationship between one-electron transition metal
reactivity and radical polymerization processes. Angew. Chem., Int. Ed.
2006, 45, 5058−5070.
́
̂
́
̂
(15) Debuigne, A.; Poli, R.; Jerome, C.; Jerome, R.; Detrembleur, C.
Overview of cobalt-mediated radical polymerization: roots, state of
the art and future prospects. Prog. Polym. Sci. 2009, 34, 211−239.
(16) Kaur, A.; Ribelli, T. G.; Schroeder, K.; Matyjaszewski, K.;
Pintauer, T. Properties and ATRP Activity of Copper Complexes with
Substituted Tris(2-pyridylmethyl)amine-Based Ligands. Inorg. Chem.
2015, 54 (4), 1474−1486.
(17) Ribelli, T. G.; Fantin, M.; Daran, J.-C.; Augustine, K. F.; Poli,
R.; Matyjaszewski, K. Synthesis and Characterization of the Most
Active Copper-based ATRP Catalyst based on tris[(4-
dimethylaminopyridyl)methyl]amine. J. Am. Chem. Soc. 2018, 140,
1525−1534.
̈
(18) Schroder, K.; Konkolewicz, D.; Poli, R.; Matyjaszewski, K.
Formation and Possible Reactions of an Organometallic Intermediate
with Active Copper(I) Catalysts in an ATRP. Organometallics 2012,
31, 7994−7999.
(19) Wang, Y.; Soerensen, N.; Zhong, M.; Schroeder, H.; Buback,
M.; Matyjaszewski, K. Improving the “Livingness” of ATRP by
Reducing Cu Catalyst Concentration. Macromolecules 2013, 46 (3),
683−691.
(20) Ribelli, T. G.; Wahidur Rahaman, S. M.; Daran, J.-C.; Krys, P.;
Matyjaszewski, K.; Poli, R. Effect of Ligand Structure on the CuII-R
OMRP Dormant Species and its Consequences for Catalytic Radical
Termination in ATRP. Macromolecules 2016, 49, 7749−7757.
(21) Ribelli, T. G.; Augustine, K. F.; Fantin, M.; Krys, P.; Poli, R.;
Matyjaszewski, K. Disproportionation or Combination? The Termi-
nation of Acrylate Radicals in ATRP. Macromolecules 2017, 50, 7920−
7929.
(22) Ribelli, T. G.; Rahaman, S. M. W.; Matyjaszewski, K.; Poli, R.
Catalyzed Radical Termination (CRT) in the Metal-Mediated
Polymerization of Acrylates: Experimental and Computational
Studies. ACS Symp. Ser. 2018, 1284, 135−159.
(23) Matyjaszewski, K. Radical nature of Cu-catalyzed controlled
radical polymerizations (atom transfer radical polymerization).
Macromolecules 1998, 31, 4710−4717.
(24) Bamford, C. H.; Jenkins, A. D. Termination reaction in vinyl
polymerization - Preparation of block copolymers. Nature 1955, 176
(4471), 78−78.
(25) Ayrey, G.; Humphrey, M. J.; Poller, R. C. Radiochemical
studies of free-radical vinyl polymerizations. 7. Polymerization of
methyl acrylate. Polymer 1977, 18 (8), 840−844.
(26) Szablan, Z.; Junkers, T.; Koo, S. P. S.; Lovestead, T. M.; Davis,
T. P.; Stenzel, M. H.; Barner-Kowollik, C. Mapping photolysis
product radical reactivities via soft ionization mass spectrometry in
acrylate, methacrylate, and itaconate systems. Macromolecules 2007,
40 (19), 6820−6833.
(27) Nakamura, Y.; Lee, R.; Coote, M. L.; Yamago, S. Termination
Mechanism of the Radical Polymerization of Acrylates. Macromol.
Rapid Commun. 2016, 37 (6), 506−513.
(28) Nakamura, Y.; Ogihara, T.; Yamago, S. Mechanism of Cu(I)/
Cu(0)-Mediated Reductive Coupling Reactions of Bromine-Termi-
nated Polyacrylates, Polymethacrylates, and Polystyrene. ACS Macro
Lett. 2016, 5 (2), 248−252.
(29) Krys, P.; Wang, Y.; Matyjaszewski, K.; Harrisson, S. Radical
Generation and Termination in SARA ATRP of Methyl Acryalte:
Effect of Solvent, Ligand and Chain Length. Macromolecules 2016, 49,
2977−2984.
(30) Warren, J. J.; Tronic, T. A.; Mayer, J. M. Thermochemistry of
Proton-Coupled Electron Transfer Reagents and its Implications.
Chem. Rev. 2010, 110 (12), 6961−7001.
(31) Kim, D.; Rahaman, S. M. W.; Mercado, B. Q.; Poli, R.; Holland,
P. L. The Roles of Iron Complexes in Catalytic Radical Alkene Cross-
Coupling J. Am. Chem. Soc., accepted for publication.
(47) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, K.; Robb,
M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.;
Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.;
Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.;
Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.;
L
DOI: 10.1021/acs.inorgchem.9b00660
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd,
J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.;
Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S.
S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J.
E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A.
Polymerization (CMRP) of Vinyl Acetate via Cobalt(III) Adducts as
Initiators. Chem. - Eur. J. 2008, 14, 4046−4059.
(63) Tellers, D. M.; Ritter, J. C. M.; Bergman, R. G. Synthesis of
iridium(III) carboxamides via the bimetallic reaction between
Cp*(PMe3)IrPh(OH) and)Cp*(PMe3)Ir(Ph)NCR)+. Inorg. Chem.
1999, 38, 4810−4818.
(64) Fulton, J. R.; Bouwkamp, M. W.; Bergman, R. G. Reactivity of a
parent amidoruthenium complex: A transition metal amide of
exceptionally high basicity. J. Am. Chem. Soc. 2000, 122 (36),
8799−8800.
(65) Fulton, J. R.; Sklenak, S.; Bouwkamp, T.; Bergman, R. G. A
comprehensive investigation of the chemistry and basicity of a parent
amidoruthenium complex. J. Am. Chem. Soc. 2002, 124 (17), 4722−
4737.
(66) Tehranchi, J.; Donoghue, P. J.; Cramer, C. J.; Tolman, W. B.
Reactivity of (Dicarboxamide) M-II-OH (M = Cu, Ni) Complexes -
Reaction with Acetonitrile to Yield M(II-)Cyanomethides. Eur. J.
Inorg. Chem. 2013, 2013 (22−23), 4077−4084.
(67) Bortolamei, N.; Isse, A. A.; Di Marco, V. B.; Gennaro, A.;
Matyjaszewski, K. Thermodynamic Properties of Copper Complexes
Used as Catalysts in Atom Transfer Radical Polymerization.
Macromolecules 2010, 43 (22), 9257−9267.
(68) Cuvigny, T.; Leborgne, J. F.; Larcheveque, M.; Normant, H.
Hyperbasic media. 1. Metalation of hydrazones - direct synthesis of
nitriles. Synthesis 1976, 1976 (4), 237−238.
̈
D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J.
Gaussian 09, Revision D.01; Gaussian, Inc.: Wallingford CT, 2009.
(48) Perdew, J. P. Density-Functional Approximation for the
Correlation-Energy of the Inhomogeneous Electron-Gas. Phys. Rev.
B: Condens. Matter Mater. Phys. 1986, 33 (12), 8822−8824.
̈
(49) Ehlers, A. W.; Bohme, M.; Dapprich, S.; Gobbi, A.; Hoellwarth,
A.; Jonas, V.; Koehler, K. F.; Stegmann, R.; Veldkamp, A.; Frenking,
G. A set of f-polarization functions for pseudopotential basis sets of
the transition metals Sc-Cu, Y-Ag and La-Au. Chem. Phys. Lett. 1993,
208 (1−2), 111−114.
(50) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A consistent and
accurate ab initio parametrization of density functional dispersion
correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 2010,
132 (15), 154104.
(51) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. Universal
Solvation Model Based on Solute Electron Density and on a
Continuum Model of the Solvent Defined by the Bulk Dielectric
Constant and Atomic Surface Tensions. J. Phys. Chem. B 2009, 113
(18), 6378−6396.
(69) Konkolewicz, D.; Magenau, A. J. D.; Averick, S. E.; Simakova,
A.; He, H. K.; Matyjaszewski, K. ICAR ATRP with ppm Cu Catalyst
in Water. Macromolecules 2012, 45 (11), 4461−4468.
(70) Simakova, A.; Averick, S. E.; Konkolewicz, D.; Matyjaszewski,
K. Aqueous ARGET ATRP. Macromolecules 2012, 45 (16), 6371−
6379.
(52) Bryantsev, V. S.; Diallo, M. S.; Goddard, W. A., III Calculation
of solvation free energies of charged solutes using mixed cluster/
continuum models. J. Phys. Chem. B 2008, 112 (32), 9709−9719.
(53) Braunecker, W. A.; Tsarevsky, N. V.; Pintauer, T.; Gil, R. R.;
Matyjaszewski, K. Quantifying vinyl monomer coordination to CuI in
solution and the effect of coordination on monomer reactivity in
radical copolymerization. Macromolecules 2005, 38 (10), 4081−4088.
(54) (a) Krstina, J.; Moad, G.; Willing, R. I.; Danek, S. K.; Kelly, D.
P.; Jones, S. L.; Solomon, D. Further studies on the thermal
decomposition of AIBN - Implications concerning the mechanism of
termination in methacrylonitrile polymerization. Eur. Polym. J. 1993,
29 (2−3), 379−388. (b) Moad, G. A Critical Assessment of the
Kinetics and Mechanism of Initiation of Radical Polymerization with
Commercially Available Dialkyldiazene Initiators. Prog. Polym. Sci.
2019, 88, 130−188.
́
̂
(71) Morin, A. N.; Detrembleur, C.; Jerome, C.; De Tullio, P.; Poli,
R.; Debuigne, A. Effect of head-to-head addition in vinyl acetate
reversible deactivation radical polymerization: why is Co(acac)2-
mediated polymerization so much better? Macromolecules 2013, 46,
4303−4312.
(72) Poli, R. New phenomena in organometallic-mediated radical
polymerization (OMRP) and perspectives for control of less active
monomers. Chem. - Eur. J. 2015, 21, 6988−7001.
(73) Banerjee, S.; Ladmiral, V.; Debuigne, A.; Detrembleur, C.; Poli,
́
R.; Ameduri, B. Organometallic Mediated Radical Polymerization of
Vinylidene Fluoride. Angew. Chem., Int. Ed. 2018, 57, 2934−2937.
(74) Fliedel, C.; Poli, R. Homolytically weak metal-carbon bonds
make robust controlled radical polymerizations systems for “less-
activated monomers. J. Organomet. Chem. 2019, 880, 241−252.
(55) Bamford, C. H.; Dyson, R. W.; Eastmond, G. C. Network
formation. 4. Nature of termination reaction in free-radical polymer-
ization. Polymer 1969, 10 (11), 885−899.
(56) Eckenhoff, W. T.; Garrity, S. T.; Pintauer, T. Highly efficient
copper-mediated atom-transfer radical addition (ATRA) in the
presence of reducing agent. Eur. J. Inorg. Chem. 2008, 2008 (4),
563−571.
(57) Eckenhoff, W. T.; Pintauer, T. Structural Comparison of
Copper(I) and Copper(II) Complexes with Tris(2-pyridylmethyl)-
amine Ligand. Inorg. Chem. 2010, 49 (22), 10617−10626.
(58) Braunecker, W. A.; Tsarevsky, N. V.; Gennaro, A.;
Matyjaszewski, K. Thermodynamic Components of the Atom
Transfer Radical Polymerization Equilibrium: Quantifying Solvent
Effects. Macromolecules 2009, 42 (17), 6348−6360.
(59) Moore, J. W.; Pearson, R. G. Kinetics and Mechanisms.. 3rd ed.;
John Wiley & Sons: New York, 1981.
(60) Ribelli, T. G.; Krys, P.; Cong, Y. D.; Matyjaszewski, K. Model
Studies of Alkyl Halide Activation and Comproportionation Relevant
to RDRP in the Presence of Cu-0. Macromolecules 2015, 48 (23),
8428−8436.
(61) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.;
Nudelman, A.; Stoltz, B. M.; Bercaw, M.; Goldberg, K. I. NMR
Chemical Shifts of Trace Impurities: Common Laboratory Solvents,
Organics, and Gases in Deuterated Solvents Relevant to the
Organometallic Chemist. Organometallics 2010, 29 (9), 2176−2179.
́
̂
(62) Debuigne, A.; Champouret, Y.; Jerome, R.; Poli, R.;
Detrembleur, C. Mechanistic Insights into Cobalt Mediated Radical
M
DOI: 10.1021/acs.inorgchem.9b00660
Inorg. Chem. XXXX, XXX, XXX−XXX