POCOP-Type Pincer Complexes of Nickel: Synthesis, Characterization, and Ligand Exchange Reactivities of New Cationic Acetonitrile Adducts

Article abstract of DOI:10.1021/acs.organomet.5b00272

This report describes the synthesis, characterization, and ligand exchange studies of a family of cationic acetonitrile adducts of nickel featuring resorcinol-based, pincer-type POCOP ligands. The compounds [(R-POCOP^R′)Ni(NCMe)][OSO₂CF₃] (R-POCOP^R′ = 2,6-(R′₂PO)₂(R_nC₆H_3-n); R′ = i-Pr: R = H (1), p-Me (2), p-OMe (3), p-CO₂Me (4), p-Br (5), m,m-t-Bu₂ (6), m-OMe (7), m-CO₂Me (8); R′ = t-Bu: R = H (9), p-CO₂Me (10)) were prepared in 80-93% yields by reacting the corresponding charge-neutral bromo derivatives with Ag(OSO₂CF₃) in acetonitrile. The impact of the R- and R′-substituents on electronics and structures of 1-10 have been probed by NMR, UV-vis, and IR spectra, X-ray crystallography, and cyclic voltammetry measurements. The observed ν(C≡N) values were found to increase with the increasing electron-withdrawing nature of R, i.e., in the order 7 < 3 ～ 2 ～ 6 < 1 < 5 ～ 8 < 4 and 9 < 10. This trend is consistent with the anticipation that enhanced electrophilicity of the nickel center should result in an increase in net MeCN→Ni σ-donation. That this transfer of electron density from acetonitrile to the nickel center does not adequately counteract the impact of electron-withdrawing substituents was evident from the measured redox potentials: the MeO₂C-substituted cations showed the highest oxidation potentials. Moreover, all cationic adducts showed greater oxidation potentials compared with their corresponding charge-neutral bromo precursors. Equilibrium studies conducted with selected [(R-POCOP^R′)Ni(NCMe)][OSO₂CF₃] and (R-POCOP^R′)NiBr (R′ = i-Pr) have confirmed facile MeCN/Br exchange between these derivatives and show that the cationic adducts are stabilized with MeO-POCOP, whereas the charge-neutral bromo species are stabilized with MeO₂C-POCOP. The potential implications of these findings for the catalytic reactivities of the title cationic complexes have been discussed. (Chemical Equation Presented).

Full text of DOI:10.1021/acs.organomet.5b00272

Article
pubs.acs.org/Organometallics
POCOP-Type Pincer Complexes of Nickel: Synthesis,
Characterization, and Ligand Exchange Reactivities of New Cationic
Acetonitrile Adducts
Sebastien Lapointe, Boris Vabre,† and Davit Zargarian*
́
́ ́ ́ ́ ́
Departement de Chimie, Universite de Montreal, Montreal, Quebec, Canada H3C 3J7
S
* Supporting Information
ABSTRACT: This report describes the synthesis, characterization, and ligand exchange studies
of a family of cationic acetonitrile adducts of nickel featuring resorcinol-based, pincer-type
POCOP ligands. The compounds [(R-POCOPR′)Ni(NCMe)][OSO2CF3] (R-POCOPR′ = 2,6-
(R′2PO)2(RnC6H3−n); R′ = i-Pr: R = H (1), p-Me (2), p-OMe (3), p-CO2Me (4), p-Br (5), m,m-
t-Bu2 (6), m-OMe (7), m-CO2Me (8); R′ = t-Bu: R = H (9), p-CO2Me (10)) were prepared in
80−93% yields by reacting the corresponding charge-neutral bromo derivatives with
Ag(OSO2CF3) in acetonitrile. The impact of the R- and R′-substituents on electronics and
structures of 1−10 have been probed by NMR, UV−vis, and IR spectra, X-ray crystallography,
and cyclic voltammetry measurements. The observed ν(CN) values were found to increase
with the increasing electron-withdrawing nature of R, i.e., in the order 7 < 3 ∼ 2 ∼ 6 < 1 < 5 ∼ 8
< 4 and 9 < 10. This trend is consistent with the anticipation that enhanced electrophilicity of the
nickel center should result in an increase in net MeCN→Ni σ-donation. That this transfer of
electron density from acetonitrile to the nickel center does not adequately counteract the impact
of electron-withdrawing substituents was evident from the measured redox potentials: the MeO2C-substituted cations showed
the highest oxidation potentials. Moreover, all cationic adducts showed greater oxidation potentials compared with their
corresponding charge-neutral bromo precursors. Equilibrium studies conducted with selected [(R-POCOPR′)Ni(NCMe)]-
[OSO2CF3] and (R-POCOPR′)NiBr (R′ = i-Pr) have confirmed facile MeCN/Br exchange between these derivatives and show
that the cationic adducts are stabilized with MeO-POCOP, whereas the charge-neutral bromo species are stabilized with MeO2C-
POCOP. The potential implications of these findings for the catalytic reactivities of the title cationic complexes have been
discussed.
We have initiated a systematic study aimed at mapping out
the impact of differently substituted, resorcinol-derived
POCOP ligands on structures and electronic properties of
their cationic Ni-acetonitrile adducts. Thus, we have reported
on the influence of P- and aromatic ring substituents on C−H
nickelation rates, structures, and redox potentials of charge-
neutral complexes (R-POCOPR′)NiBr.15 Analogous studies
have been reported by other groups on POCOP complexes of
Ir and Pt.16 These reports have inspired us to study the impact
of P- and aromatic ring substituents on the structures and
electronic properties of cationic adducts featuring POCOP
ligands. Reported herein are the syntheses, and spectroscopic
and structural studies, redox potential measurements, and
ligand exchange reactivities of the cationic acetonitrile adducts
INTRODUCTION
■
The pioneering studies begun in the 1970s by groups of Shaw1
and van Koten2 laid the foundation for what later came to be
known as pincer chemistry.3 Over the nearly four decades since,
researchers have combined old and new ligand types with
various d- and p-block elements to synthesize numerous pincer
complexes and explore their reactivities and physical properties.
These developments have been captured in a number of
authoritative review articles.4
Nickel complexes were among the early pincer derivatives
introduced by Shaw and van Koten, but it was only over the
past decade that organonickel pincer chemistry experienced
major developments. The introduction of many different
families of pincer nickel complexes (PCP,5 POCsp2OP,6
POCsp3OP,7 POCN,8 PNP,9 NNN,10 etc.11) has led to exciting
developments that have been reviewed recently.4e,12 Among
these, resorcinol-based POCOP-Ni systems have proven
particularly popular due to the facile synthesis of variously
substituted ligands and their complexes (via C−H nickel-
ation),6b,d,7a,13 as well as the interesting catalytic trans-
formations these compounds promote.14
[(R-POCOPR′)Ni(NCMe)][OSO2CF3], where R and R′
represent, respectively, various substituents on the resorcinol
aromatic ring (H, p-Me, p-OMe, p-CO2Me, p-Br, m-OMe, m-
CO2Me, m,m-t-Bu2) and P-substituents (i-Pr, t-Bu). Given the
importance of this family of complexes in a number of
Received: April 6, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.5b00272
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
interesting catalytic reactivities,7a,14c,17 we hope that the results
of our study will provide a basis for tuning the reactivities of
be noted that all but one of these precursors are known species
that have been reported previously;15a the only new bromide
[(R-POCOPR′)Ni(L)][OSO2CF3] for different catalytic appli-
cations.
derivative, (p-Br-POCOPR′)NiBr (R′ = iPr, 5′), was prepared
in the same manner as its counterparts 1′−4′ using the new
preligand p-Br-POCHOPi-Pr. The latter was prepared by
phosphorylation of 5-bromoresorcinol (Scheme 1), which was
itself obtained from acidic hydrolysis of 3,5-dimethoxy-
bromobenzene using a modified literature procedure.18
RESULTS AND DISCUSSION
Synthesis of [(R-POCOPR′)Ni(NCMe)][OSO2CF3]. The
■
target cationic complexes were prepared by treating the
The workup procedure for the cationic complexes consisted
of cannula filtration of the final reaction mixture, followed by a
second filtration through a short Celite plug to remove AgBr
and furnish the target complexes as yellow powders in 80−93%
yields. These cationic adducts proved to be less stable than
their bromo precursors: storing analytically pure samples in the
drybox (without protection from ambient light) led to a gradual
color change from yellow to dark green. Eluting a CH2Cl2
suspension of these postdecomposition green solids through a
short Celite column left a black residue on top of the column
and furnished the desired cationic complexes in analytically
pure form. Storing the purified samples under inert atmosphere
inside a −37 °C freezer circumvented further decomposition.
The title cationic adducts as well as the new bromo derivative
5′ were characterized by spectroscopy (NMR, IR, and UV−vis)
and cyclic voltammetry. Single-crystal X-ray diffraction studies
were also carried out for complexes 5′, 2−6, 9, and 10; single
crystals suitable for diffraction studies could not be obtained for
complexes 7 and 8, whereas the solid-state structure of the
parent cation 1 has been reported previously.13 The solid-state
structures are discussed below, followed by the results of the
spectroscopic studies and electrochemical analyses.
charge-neutral bromo derivatives (R-POCOPR′)NiBr with 1.2
equiv of AgOTf in dry acetonitrile under nitrogen and in the
absence of light (Scheme 1). The charge-neutral bromide
Scheme 1. General Synthetic Scheme for Complexes 1−10
Solid-State Structures. Table 1 lists the most pertinent
structural parameters for the complexes that were subjected to
X-ray diffraction studies, and Figures 1−3 show a side view of
the molecular diagrams of adducts 4, 6, and 10; the molecular
diagrams for the remaining four complexes, the front view of
the molecular diagrams of complexes 4, 6, and 10, and the
details of the diffraction studies are given in Figures S1−S8 and
Tables S1 and S2 (see the Supporting Information, SI).
The nickel center in all complexes adopts a square planar
geometry that displays slight distortions due, primarily, to the
small bite angle of the POCOP ligands: P−Ni−P ≈ 163−165°.
precursors used in these syntheses were, in turn, obtained from
the rt reaction of the preligands R-POCHOP with
NiBr2(NCMe)2 in the presence of 1.2 equiv of NEt3. It should
Table 1. Selected Bond Distances (Å) and Angles (deg) for Cationic Adducts 1−10 and Charge-Neutral Complexes 1′−10′
complex
Ni−C
Ni−N/Br
NC
Ni−P1
Ni−P2
P1−Ni−P2
C−Ni−N/Br
a
1
1.881(2)
1.879(3)
1.884(2)
1.880(2)
1.879(3)
1.890(2)
1.885(2)
1.881(2)
1.885(3)
1.882(3)
1.877(2)
1.872(2)
1.877(2)
1.892(4)
1.887(2)
1.877(2)
1.874(2)
1.875(2)
1.881(2)
1.879(2)
1.871(3)
1.875(2)
1.879(2)
1.876(2)
2.3231(5)
2.3305(5)
2.319(1)
2.312(1)
2.3211(5)
2.320(1)
2.338(2)
2.321(1)
1.140(3)
1.142(4)
1.135(3)
1.138(2)
1.145(4)
1.140(2)
1.140(2)
1.146(2)
2.1683(7)
2.1693(8)
2.1784(5)
2.1698(4)
2.1685(9)
2.1785(5)
2.1944(5)
2.1933(4)
2.1534(8)
2.1584(4)
2.155(1)
2.157(1)
2.1500(7)
2.139(1)
2.193(1)
2.189(1)
2.1704(7)
2.1687(7)
2.1747(5)
2.1748(4)
2.171(2)
2.1640(4)
2.1946(5)
2.1922(4)
2.1422(8)
2.1584(4)
2.152(1)
2.159(1)
2.1467(8)
2.143(1)
2.189(1)
2.194(1)
164.38(3)
163.52(3)
163.16(2)
164.45(1)
164.44(4)
163.79(2)
163.57(2)
164.025(1)
164.92(4)
164.21(3)
164.65(2)
165.26(2)
163.89(3)
165.06(5)
164.13(3)
164.58(3)
175.8(1)
171.7(1)
176.84(7)
178.43(6)
176.6(2)
170.35(6)
179.02(6)
178.51(6)
178.10(8)
180.0(1)
178.25(6)
179.3(1)
178.74(8)
179.08(12)
179.7(1)
178.25(7)
2
3
4
5
6
9
10
a
1′
2′
3′
4′
a
a
a
5′
6′
9′
a
a
a
10′
a
Previously reported complexes.13,15a
B
DOI: 10.1021/acs.organomet.5b00272
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
2.139−2.158 Å), whereas this impact is more muted in the
cationic adducts (range = 2.168−2.179 Å). The smallest Ni−P
distances (ca. 2.14 Å) were found in the bromo complex 6′,
featuring the m,m-(t-Bu)2 substituents; this observation was
made previously and has been interpreted elsewhere.15a
Additional bond distances and angles for cationic adducts
reported in Table 1 can be found in Table S3 in the SI.
Most of the remaining Ni−X distances appear to be
insensitive to the nature of the ring substituents. Thus, the
Ni−Cipso distances are essentially equal in all structures,
whether cationic adducts or charge-neutral bromides (1.87−
1.89 Å). Quite uniform distances were also observed for Ni−N
(1.87−1.88 Å) and Ni−Br (2.31−2.34 Å) bonds. The observed
Ni−N and NC distances fall closer to the lower values in the
range of distances available in the Cambridge Structural
Database: ca. 1.88 vs 1.71−2.51 Å for Ni−N and 1.14 vs
0.91−1.69 Å for NC. The Ni−O distances of 3.9−5.4 Å are
very long compared to the corresponding distances in
complexes wherein the triflate anion is directly bound to the
Ni(II) center (average of 2.05 Å);19 we conclude that little or
no covalent Ni−OSO2CF3 interactions are present in our
cationic acetonitrile adducts.
Figure 1. Molecular diagram for complex 4. Thermal ellipsoids are
shown at the 50% probability level. Hydrogens and the P-substituents
have been removed from this diagram for clarity.
IR Analyses. The IR spectra recorded using solid samples of
the cationic complexes showed the characteristic bands for
ν(CN) (ca. 2329−2293 cm−1), ν(SO) (ca. 1270, 1030, and
636 cm−1), and ν(CF) (ca. 1135 cm−1) (see Table S4 in the
SI). Comparing the ν(CN) values in various complexes
shows a fairly clear correlation between the electronic nature of
ring substituents R and the C−N bond strength (Table 2). For
instance, replacing the H at the para position of the central ring
(with respect to the metalated i-C) by the electron-withdrawing
substituent CO2Me increases the ν(CN) values by 32 cm−1
(1 vs 4) or 22 cm−1 (9 vs 10). The frequency shift, Δν(CN),
observed on going from 1 to 5 was +5 cm−1, which shows that
the impact of Br is smaller in magnitude but in the same
direction as that of CO2Me.
Figure 2. Molecular diagram for complex 6. Thermal ellipsoids are
shown at the 50% probability level. Hydrogens and the P-substituents
have been removed from this diagram for clarity.
In the case of adducts bearing the electron-releasing
substituents p-Me and p-OMe, the impact on ν(CN) was
similarly small in magnitude but in the opposite direction:
Δν(CN) = −3 cm−1 in 2 and −4 cm−1 in 3. Moreover, the
same types of frequency shifts result from substitution at the m-
position, but in this case the electron-releasing substituent OMe
leads to a larger impact than its electron-withdrawing
counterpart CO2Me: Δν(CN) = −13 cm−1 in 7 and +6
cm−1 in 8. Finally, the fairly small impact of the two t-Bu
substituents in adduct 6 (Δν(CN) = −3 cm−1) should be
interpreted more cautiously, because it likely is a composite of
the steric and electronic effects of these substituents.20
The observed correlation between ν(CN) values and the
POCOP ring substituents R can be rationalized in terms of the
MO description of bonding in acetonitrile, as follows. Similarly
to CO, acetonitrile’s HOMO possesses a certain degree of
antibonding character with respect to the CN bond, such
that σ-donation from this HOMO to a Ni-based acceptor
orbital should reinforce the C−N bond and increase the value
of ν(CN).22 This phenomenon is underlined by the
observation that in all the acetonitrile adducts the CN
stretching frequency is much higher than that of free
acetonitrile (2252 cm−1), consistent with a net transfer of
charge from acetonitrile to the Ni center.
Figure 3. Molecular diagram for complex 10. Thermal ellipsoids are
shown at the 50% probability level. Hydrogens and the P-substituents
have been removed from this diagram for clarity.
The displacement of the nickel atom out of the coordination
plane (P1−N−P2−Cipso) is either negligible (<0.01 Å) or very
minor (0.02−0.05 Å). Another source of notable structural
distortion is the Cipso−Ni−N angle, which is more acute in
some cases due to the lifting of the acetonitrile moiety out of
the coordination plane. For instance, Cipso−Ni−N angles of ca.
170−172° were found in complexes 2 (R = p-Me) and 6 (R =
m,m-t-Bu2), in contrast to angles of 175−180° in the remainder
of the complexes; thus, it is difficult to establish with confidence
whether these deviations are due to steric or electronic factors.
The Ni−P distances show some variations in these
compounds. First, the t-Bu2P−Ni distances are much longer
than their i-Pr2P−Ni counterparts, particularly for the charge-
neutral bromo derivatives: average of 2.19 Å for t-Bu2P
complexes vs 2.15 Å for i-Pr2P complexes. Moreover, the
cationic adducts display consistently longer Ni−P distances
relative to their charge-neutral bromo counterparts, particularly
among the i-Pr2P derivatives: average of 2.17 vs 2.15 Å,
respectively. Finally, the aromatic ring substituent(s) R appear
to have some impact on the i-Pr2P−Ni distances (range =
It follows then that ν(CN) values in the adducts examined
here should reflect the approximate electron-donating character
of R-POCOP ligands, those with electron-withdrawing
C
DOI: 10.1021/acs.organomet.5b00272
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 2. ν(CN) Data for 1−10, Related Complexes, and Free MeCN
c
complex
ν(CN) (cm−1
)
Δν(CN) (cm−1
)
a
b
b
[(H-POCOPi‑Pr)Ni(NCMe)][OSO2CF3], 1
2297 (2292)
2294
2293
2329
2302
2294
2284
2303
2293
2315
2284
2282
2274
2297
2257
2252
45 (40)
[(p-Me-POCOPi‑Pr)Ni(NCMe)][OSO2CF3], 2
[(p-OMe-POCOPi‑Pr)Ni(NCMe)][OSO2CF3], 3
[(p-CO2Me-POCOPi‑Pr)Ni(NCMe)][OSO2CF3], 4
[(p-Br-POCOPi‑Pr)Ni(NCMe)][OSO2CF3], 5
[(m,m-t-Bu2-POCOPi‑Pr)Ni(NCMe)][OSO2CF3], 6
[(m-OMe-POCOPi‑Pr)Ni(NCMe)][OSO2CF3], 7
[(m-CO2Me-POCOPi‑Pr)Ni(NCMe)][OSO2CF3], 8
[(H-POCOPt‑Bu)Ni(NCMe)][OSO2CF3], 9
42
41
77
50
42
32
51
41
63
32
30
22
45
[(p-CO2Me-POCOPt‑Bu)Ni(NCMe)][OSO2CF3], 10
a
[(H-POCsp3OPi-Pr)Ni(NCMe)][OSO2CF3]
a
[(H-PCPi-Pr)Ni(NCMe)][BPh4]
a
i-Pr
[(H-PC
P
)Ni(NCMe)][BPh4]
[(H-POCOPPh)Ni(NCMe)][OSO2CF3]
sp3
a
a
[(H-POCOPi-Pr)Ni(NCCHCH2)][OSO2CF3]
a
[(H-POCsp3OPi-Pr)Ni(NCCHCH2)][OSO2CF3]
Previously reported complexes.13,14b,c,21b For complex 1, the ν(CN) values in parentheses were measured using KBr pellets to provide a
a
b
c
comparison to the solid-state ATR measurements used in the discussion. Δν(CN) is relative to the free MeCN stretching frequency (2252 cm−1)
measured under our experimental conditions.
Figure 4. UV−vis spectra of cationic complexes 1−10 in full, dashed, and “dash dot dot” lines and of complexes 1′, 3′, and 4′ in dashed lines for
comparison. The spectra were recorded in air, using 10−4 M CH2Cl2 solutions.
substituents allowing greater MeCN→Ni donation and
showing higher ν(CN) values, and vice versa. This is
reflected in plots of the ν(CN) stretching frequencies for
complexes 1−10 and the σm or σp Hammett coefficients for the
corresponding ring substituents R (Figures S9−S11 in the SI).
Moreover, it stands to reason that the degree of electron
donation from RCN to the electrophilic Ni center should be
fairly proportional to its degree of activation toward
nucleophiles. Hence, ν(CN) values can help estimate the
degree of nitrile activation toward outer-sphere nucleophilic
attacks either at the nitrile carbon, as in amidination,17,23 or at
the R moiety, as in Michael-type hydroamination of
acrylonitrile and its substituted derivatives.13,14b,c,24
The above arguments can be extended to other structural
parameters in [(pincer)Ni(NCR)]+ and their impact on ν(C
N) values. For instance, changing the ligand backbone from an
aromatic to an aliphatic skeleton has a significant effect on
ν(CN), as seen from the values for 1 (2297 cm−1) and its
P O C s p 3 O P c o u n t e r p a r t [ ( H - P O C s p 3 O P i P r ) N i-
(NCMe)]−[OSO2CF3] (2284 cm−1).13 A similar influence is
exerted by the nature of the PR′2 moiety: replacing the
phosphinite moiety by a better net donor phosphine moiety
(P−O vs P−CH2 connections) leads to a significant decrease in
ν(CN) values from 2297 cm−1 in 1 to 2282 cm−1 in its PCP
analogue [(H-PCPiPr)Ni(NCMe)][BPh4].14c A further de-
crease in the ν(CN) value to 2274 cm−1 is observed in the
D
DOI: 10.1021/acs.organomet.5b00272
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
aliphatic PCsp3P analogue [(H-PCsp3P)Ni(NCMe)][BPh4].14c
These observations are consistent with lower net donation from
MeCN to Ni ligated by strong donor ligands.
the excited state better than would electron-donating groups,
thus lowering the π→π* transition energy and leading to a red
shift.15a In other words, ligands bearing the most effective
electron-accepting ring substituent should exhibit lower energy
π→π* transitions. A similar phenomenon is at work for the
MLCT transition in the complexes. As shown in Figure 4 and
Table S6 in the SI, the MLCT transitions of the charge-neutral
bromide complexes feature molar absorptivities and wave-
lengths that are very similar to those of cationic acetonitrile
adducts.
Electrochemical Studies. For the cationic adducts under
study, another indication of how POCOP ring or P-substituents
affect the electron density of the Ni center can be extrapolated
from their redox potentials. We have conducted cyclic
The influence of P-substituents R′ on ν(CN) values can be
discerned from a comparison among the following two series of
complexes: [(H-POCOPPh)Ni(NCMe)][OSO2CF3]14b vs [(H-
POCOPi‑Pr)Ni(NCMe)][OSO2CF3] (1) vs [(H-POCOPt‑Bu)-
Ni(NCMe)][OSO2CF3] (9); [(p-CO2Me-POCOPi‑Pr)Ni-
(NCMe)][OCO2CF3] (4) vs [(p-CO2Me-POCOPt‑Bu)Ni-
(NCMe)][OSO2CF3] (10). This comparison confirms that
for both of these series the observed correlation is in line with
the anticipated donor strength of the different P-substituents,
i.e., Ph < i-Pr < t-Bu for the first series and i-Pr < t-Bu for the
second.25 On the other hand, the t-Bu2P complexes display the
longest Ni−P distances (2.195(1) Å in 9 vs 2.173(1) Å in 1;
2.1928(8) Å in 10 vs 2.1723(8) Å in 4), which should normally
lead to poorer orbital overlap between the Ni center and the
bulky t-Bu2P moiety, less efficient P→Ni donation, and hence
more electrophilic Ni centers in 9 and 10. However, the
opposite scenario is implied by the IR data. We have further
probed this issue through cyclic voltammetry measurements,
and, as will be discussed below, the redox potentials of these
complexes tend to contradict the implications of ν(CN)
values for adducts 9 and 10.
a
Table 3. Redox Potentials of 1−10 and 1′−10′
complex
Eox or E1/2 (mV)
complex
Eox or E1/2 (mV)
b
b
1
1168
1′
2′
3′
4′
817
b
2
1151
692
b
3
933 (922)
1269
560 (572)
776
b
4
5
1159
5′
6′
7′
8′
9′
10′
738
b
6
1012 (997)
977 (967)
1262
567
b
NMR Studies. Table S5 in the SI lists the pertinent 31P, 1H,
and 13C NMR data for complexes 1−10, and the relevant
spectra are shown in Figures S13−S51 in the SI. The 31P NMR
spectra show a rather narrow chemical shift range, ca. 192−196
ppm for 1−8 and ca. 198−200 ppm for 9 and 10, indicating
that the ring substituents do not influence 31P δ values
significantly. To be sure, the electron-withdrawing substituents
CO2Me and Br seem to lead to downfield 31P chemical shifts
relative to the parent complex 1, but no clear trend is evident
with electron-releasing substituents in any of the spectra. The
parent compound 1 and its symmetrically substituted analogues
2−6, 9, and 10 show a singlet, whereas the unsymmetrical
complex 7 shows the anticipated doublet of doublets for the
two chemically inequivalent 31P nuclei (see Figure S37, SI).
Complex 8, the only other compound in the series bearing an
unsymmetrically substituted POCOP ligand, shows a 31P
singlet instead of the anticipated AB signals (see Figure S43
in the SI); this is presumably because the difference in the
chemical shifts of the two inequivalent 31P nuclei is smaller than
the resolution of the spectrum. The 13C NMR spectra of 7 and
8 showed ABX signals for the methyne carbon nuclei, giving an
apparent dt for 7 and a partially resolved dd for 8. (See Figures
S38 and S42 in the SI.)
7
761 (715)
900
b
8
b
9
1298
800 (750)
920 (860)
b
10
1457
a
E1/2 values are given for quasi-reversible redox couples. See caption of
Figure 5 for measurement details. Previously reported complexes13,15a
for which the redox values were remeasured in the current
experimental conditions.
b
voltammetry measurements on the charge-neutral bromo
complexes 1′−10′ and the cationic acetonitrile adducts 1−10,
and the results are reported in Table 3; the cationic adducts are
presented in Figures S52−S63 in the SI. Most complexes
displayed irreversible oxidation, whereas complexes 3, 6, and 7
displayed quasi-reversible behavior, presumably because the
substituents m,m-t-Bu2 and p- and m-OMe stabilize the Ni
center sufficiently to prevent decomposition during the NiII→
NiIII oxidation event. The redox potentials obtained from these
measurements (calibrated against the Cp2Fe/Cp2Fe+ redox
couple) are listed in Table 3 and plotted against Hammett
coefficients for the para and meta substituents (Figure S64, SI).
The electrochemical measurements reveal the following
order of oxidation potential for [(R-POCOPR′)Ni(NCMe)]-
[OSO2CF3] as a function of both ring substituent R and P-
substituent R′: p-CO2Me [t-Bu2P] (10) > H [t-Bu2P] (9) > p-
CO2Me (4) > m-CO2Me (8) > p-Br (5) > H (1) > p-Me (2) >
m,m-t-Bu2 (6) > m-OMe (7) > p-OMe (3). If we focus on the i-
Pr2P complexes, the observed order of oxidation potentials is
consistent with the anticipated electronic impact of the
aromatic-ring substituents. This trend shows some similarities
to the corresponding trend in ν(CN) stretching frequencies
(4 > 8 > 5 > 1 > 2), but there are also significant differences.
The most striking differences between these two trends lie
primarily in the positions of the t-Bu2P-containing complexes 9
and 10. It is conceivable that the steric bulk of the t-Bu2P
moieties constrains the acetonitrile moiety in a configuration
that would reduce orbital overlap of the nitrogen atom with the
nickel center, thus lowering the ν(CN) values compared to
their i-Pr2P analogues.
Absorption Spectra. The UV−vis spectra of complexes
1′−10′ and 1−10 were recorded in dry CH2Cl2 (ca. 10−4
M
solutions). The wavelengths and molar absorptivity of all
complexes are tabulated in Table S6 in the SI, and the spectra
of complexes 1−10, 1′, 3′, and 4′ are shown in Figure 4. The
low-energy bands (370−420 nm) display low intensities and
appear fairly indifferent to the nature of ring substituents,
implying that they are likely spin-allowed but Laporte-
forbidden d−d transitions. Conversely, the bands in the 300−
350 nm region that are most affected by the nature of the ring
substituent likely represent π→π* transitions. For instance, the
λmax for complex 4, bearing the p-CO2Me substituent, is at 341
nm, 12−16 nm higher than those of the unsubstituted complex
1 and the complexes 2 and 3, bearing electron-releasing
substituents. This observation can be rationalized by the fact
that electron-withdrawing groups stabilize the charge buildup in
E
DOI: 10.1021/acs.organomet.5b00272
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 5. Cyclic voltammograms of 3−7, 3′, and 4′. The measurements were carried out at 298 K using dry CH2Cl2 solutions containing equimolar
quantities of the given complex and [Bu4N][PF6] as electrolyte (10−4 M). Prior to beginning the measurements, the samples were purged for 2 min
by bubbling a stream of N2, and a nitrogen atmosphere was maintained over the samples throughout the measurements. A scan rate of 100 mV s−1
was used, and the potentials were referenced to the Cp2Fe/Cp2Fe+ redox couple.
+
Another insight provided by the oxidation potentials shown
in Table 3 pertains to the impact of P-substituents on the
electronic density of the metal center: the t-Bu2P analogues 9
and 10 have higher oxidation potentials than their i-Pr2P
counterparts, implying that among the cationic acetonitrile
adducts the supposedly better donor t-Bu substituents seem to
decrease instead the electron density on the Ni center. A similar
phenomenon has been observed in previous studies and has
been attributed to the longer Ni−P distances with the sterically
larger t-Bu2P moieties.5h Inspection of the Ni−P distances in
our complexes supports this assertion: the Ni−P(t-Bu)2 bond
distance is significantly longer than Ni−P(i-Pr)2 in every case
for both the cationic adducts and the charge-neutral bromo
species. On the other hand, this phenomenon appears to be
more complex, because the charge-neutral bromo complexes
represent a less clear-cut scenario: complexes 10′ and 4′ show
the oxidation potential order observed for their cationic
analogues, whereas 9′ and 1′ show the opposite order.
Evidently, more studies are required to develop a better
understanding of this phenomenon.
Finally, the overall charge of the complexes seems to be a
significant factor in the oxidation potential of the complexes
studied here: all cationic adducts are more difficult to oxidize
than their neutral bromo counterparts. We note also that
oxidation potential trends are maintained regardless of the
overall charge. In the i-Pr2P series, for instance, the same
oxidation potential trend is seen with the charge-neutral and
cationic complexes bearing the following ring substituents: p-
CO2Me > m-CO2Me > H > p-Me > m-OMe > p-OMe. The
same is also true for the two remaining cases, p-Br > m,m-t-Bu2,
as well as for the two t-Bu2P complexes (p-CO2Me > H).
Oxidation Studies with FeCp2. As alluded to above, the
values of redox potentials were calibrated against the redox
couple for ferrocene. This was done at the end of the CV
measurements by adding FeCp2 to the sample solution and
measuring the FeCp2/FeCp2 redox couple under the
conditions of the measurement in question. Cyclic voltammo-
grams of complexes 2 and 4 with added FeCp2 are shown in
Figures S53 and S56 in the SI. We were surprised to note a
dramatic color change, from pale yellow to black/dark green,
upon addition of ferrocene to some of our cationic adducts.
The rate of this color change varied from almost instantaneous
to over minutes on ca. 10−4 M solutions. In addition, the CV
traces of samples containing ferrocene also showed multiple
redox features (see for instance Figures S53 and S56 in the SI).
These observations prompted us to investigate the chemical
reaction at its origin.
We began our investigation by examining those samples that
did not appear to display this color change. Tests showed in
fact that the same color change took place instantaneously
when FeCp2 was added to more concentrated solutions of these
complexes. For instance, addition of 1.0 equiv of FeCp2 to
yellow (20−30) × 10−4 M CDCl3 solutions of all cationic
adducts (i.e., >20 times more concentrated than the samples
used for CV measurements) led to immediate formation of a
1
dark solution. Analysis of the resulting solution by H NMR
showed that the characteristic singlet resonance for the Cp
protons of FeCp2 (at 4.17 ppm) was absent from the spectrum.
In addition, the spectrum displayed a number of ligand signals
as well as a new signal that was very broad (LW1/2 ∼500 Hz)
and showed a variable intensity corresponding to between 4
and 7 protons depending on sample. The chemical shift of this
new signal was also quite variable from one sample to another
(10−6 ppm region), hinting that it might represent a
paramagnetic species. Significantly, no reaction was apparent
between the neutral bromide complexes 1′−6′ and one or more
equivalents of FeCp2, confirming that the observed reaction
takes place with the more electrophilic cationic adducts.
Another piece of evidence supporting this proposal was
furnished by the isolation of the oxidized species from a
F
DOI: 10.1021/acs.organomet.5b00272
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
NMR sample, as follows: slow evaporation of a CDCl3 solution
containing the cationic adduct 3 and 2 equiv of ferrocene gave
an amorphous green powder containing orange and dark-red
crystals that were identified by X-ray crystallography as Cp2Fe
and [Cp2Fe][OTf], respectively.
The observed oxidation of ferrocene by our cationic Ni(II)
complexes requires the concomitant reduction of the latter into
charge-neutral, monovalent species; unfortunately, however, we
have not succeeded in identifying the monovalent Ni product
of the proposed redox reaction. Indeed, the following
observation has indicated that the Ni-containing species
might in fact be diamagnetic: the amorphous dark green solid
obtained from the mixture of 8 and ferrocene showed a new 31P
NMR signal (LW1/2 = 217 Hz) ca. 1−2 ppm upfield of the
signal for the starting material.26 This would rule out the
formation of paramagnetic monomers of the type (R-
POCOPR′)Ni(NCMe). In conclusion, the fate of the Ni-
containing species remains obscure at this point, but
investigations are in progress to shed more light on this
unexpected outcome.
The predicted observation of near-unity Keq values obtained
for the exchange of precursors A and B′ bearing the same
substituents R (Table 4: entries 1, 2, and 3) served to confirm
Ligand-Exchange Studies. Previous reports in the context
of Michael-type hydroamination of acrylonitrile promoted by
pincer-Ni species have shown that the Keq for the equilibrium
[(PCPiPr)Ni(NCMe)][BPh4] ⇆ [(PCPiPr)Ni(NCCHCH2)]-
[BPh4] is near unity.5a This is an important consideration
when discussing the mechanism of this catalytic reaction,
because both the substrates and products feature nitrile
moieties and their competitive binding can have an impact
on catalytic turnover rates. On the other hand, in the context of
Ni-catalyzed nitrile amidinations, the in situ generated amidine
products are much more strongly donating ligands relative to
nitrile substrates, a phenomenon that can lead to product
inhibition, especially during the later phases of the catalytic
process. Thus, in cationic adducts [(pincer)NiL][OSO2CF3]
the kinetic lability of the Ni−L moiety can influence catalytic
turnover rates.
In contrast to the above cases involving the substitution of
neutral donors L in [(pincer)NiL][OSO2CF3] by other neutral
donors L′, little is known about the substitutional lability of the
corresponding Ni−X bond in the charge-neutral species
(pincer)NiX (X = halides, OR, NR2, etc.). We know, of
course, that halide exchange by other halides or pseudohalides
is possible in the presence of excess salts MX′ and that this type
of exchange is usually more facile in polar solvents. We also
know that halides can be displaced by charge-neutral
nucleophiles L, but in most cases this is a less facile exchange
that normally requires an abstracting agent such as Ag+.
The question arose, what about halide/L exchange involving
two complexes, a charge-neutral Ni−halide species and its
cationic Ni−L analogue, L1NiX + [L2NiL]+ ⇆ [L1NiL]+ +
L2NiX? Although such ligand X/L exchanges are rarely
considered, they can play a potentially important role in
ionization of M−X moieties during catalytic reactions, thus
affecting catalytic efficacy by creating resting states or dormant
species. To shed some light on this issue, we have examined the
ionization of the charge-neutral bromo species (R-POCOPi‑Pr)-
NiBr in the presence of their cationic acetonitrile adducts [(R-
POCOPi‑Pr)Ni(NCMe)][OSO2CF3], as depicted in eq 1. An
interesting question to probe is whether the lability of the Ni−
Br bond can be modulated by the nature of the POCOP ring
substituent R.
a
Table 4. Equilibrium Ligand Exchange as Per Eq 1
entry
A (R1)
1 (H)
B′ (R2)
1′ (H)
4′ (p-CO2Me)
3′ (p-OMe)
4′ (p-CO2Me)
3′ (p-OMe)
1′ (H)
4′ (p-CO2Me)
3′ (p-OMe)
1′ (H)
Keq
1
2
3
4
5
6
7
8
9
1.07
1.06
1.09
0.32
2.82
1.87
0.60
1.50
0.80
4 (p-CO2Me)
3 (p-OMe)
3 (p-OMe)
4 (p-CO2Me)
4 (p-CO2Me)
1 (H)
1 (H)
3 (p-OMe)
a
1
Experimental conditions: H NMR spectra were recorded for two
equimolar C6D6 solutions of A(R1) and B′(R2) each containing 2
equiv of dodecane as an internal standard (total volume of each
sample: 750 μL). The samples were then combined in one NMR tube
1
and subjected to repeated H NMR measurements over a 25 min
interval. The Keq values were determined based on the integration of
the aromatic region signals with respect to the dodecane signals.
the reliability of the methodology. Inspection of the remaining
data shows that the smallest Keq value is obtained for the
exchange of 4′ and 3 (entry 4), i.e., when the bromo species
bears an electron-withdrawing substituent R1 and the cationic
adduct bears an electron-donating substituent R2. Conversely,
the largest Keq value is obtained from the exchange of 4 and 3′
(entry 5), i.e., when the cationic adduct bears an electron-
withdrawing substituent R1 and the bromo species bears an
electron-donating substituent R2. These observations can be
rationalized by considering that the more labile acetonitrile
moiety should be found in the cationic complex bearing an
electron-withdrawing substituent, and vice versa. This explan-
ation is also consistent with the observations that the second
largest Keq value involves the exchange of 4 (entry 6), whereas
the second smallest Keq value involves the exchange of 4′ (entry
7). Finally, that the exchange of 3′ and 1 (entry 8) results in a
greater Keq than that of 1′ and 3 (entry 9) further supports the
above rationale.
We put the above rationale to test by studying the exchange
of one of our electron-poor POCOP-based cationic adducts
G
DOI: 10.1021/acs.organomet.5b00272
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 2. Ligand Exchange Reaction between 8 and (PCsp3P)Ni−Br
against a much more electron-rich charge-neutral bromo
species, namely, van Koten’s (NCN)Ni−Br (Scheme 2).27
Exchange of the latter with complex 8 (R2 = m-CO2Me) led to
an instantaneous reaction and gave a white precipitate. Analysis
of the reaction mixture by 31P NMR and 1H NMR showed that
the main product (∼82%) was the charge-neutral bromide 8′
(two doublets at 190.43 and 192.12 ppm; JPP = 323 Hz),
confirming the facile ligand exchange between the starting
materials (see Figures S66 and S67 in the SI). The reaction also
generated side-products that displayed two apparent doublets;
we believe that the latter represent the two central lines of
poorly resolved AB signals, as follows. The first of these side-
products displays two poorly resolved signals at 190.3 and
190.2 ppm, similar to the two peaks at 191.137 and 191.225
ppm in Figure S66 in the SI; this species (∼18% relative
intensity) was later identified as the charge-neutral triflate
species (m-CO2Me-POCOP)Ni-OTf (see below). The other
side-product displays two poorly resolved signals at 187.9 and
187.8 ppm (∼2% relative intensity) and remains unidentified.
The observed white precipitate was identified as the cationic
complex [(NCN)Ni(NCMe)][OSO2CF3] based on its charac-
teristic 1H NMR signal at 0.66 ppm for the coordinated
acetonitrile.28
Next we studied another ligand exchange equilibrium
involving the bromo complex (PCsp3Pi‑Pr)Ni−Br, a species for
which the oxidation potential is lower than the values recorded
for the POCOP systems 1−10, but higher than that of van
Koten’s (NCN)Ni−Br. 31P NMR analysis of a 1:1 mixture of
(PCsp3Pi‑Pr)Ni−Br and the cationic adduct 8 showed the rapid
establishment of an equilibrium mixture of five species, as
shown in Scheme 2 (see Figure S68 in the SI for the spectrum).
The assignment of the 31P signals for (m-CO2Me-POCOP)Ni−
OTf was confirmed by adding to the reaction mixture an in situ
generated sample from 8′ and AgOTf, as well as by adding a
substoichiometric amount of AgOTf to the reaction mixture,
which led to an increase in intensity of two signals for the
triflate complexes (see Figure S69 in the SI for the spectrum
with added AgOTf).28
[(m-CO2Me-POCOPi‑Pr)Ni(NCMe)][OSO2CF3] (8) and a
very electron-rich complex such as van Koten’s (NCN)NiBr.
A less electron-rich system also gives the ligand exchange, with
two of the species being charge-neutral triflate derivatives.
CONCLUSIONS
■
This study has resulted in a number of interesting observations
regarding the synthesis and stabilities of [(R-POCOPR′)Ni-
(NCMe)][OSO2CF3] and has allowed us to study the impact
of substituents R and R′ on the electrophilicity of the Ni center
as reflected in ν(CN) values, redox potentials, and Keq values
for ligand exchange reactions with charge-neutral bromo
complexes. The IR and CV measurements have confirmed
the important influence of both P- and ring substituents on the
electrophilicity of the Ni center in 1−10. Both parameters
show, for instance, that the most electrophilic adducts feature
the p-CO2Me substituent, but the redox potentials indicate that
the Ni center ligated by m-CO2Me-POCOPi‑Pr is nearly as
electrophilic. Indeed, the two sets of data lead to different
conclusions regarding the electron density on Ni in the case of
t-Bu2P-based adducts: according to ν(CN) values, the Ni
center is less electrophilic in [(p-CO2Me-POCOPt‑Bu)Ni-
(NCMe)][OSO2CF3] (10) compared to its i-Pr2P analogue;
that is, the t-Bu2P moiety is a stronger net donor, whereas the
opposite conclusion is suggested by the redox potentials. Of
course, the ν(CN) values were measured on solid samples,
whereas the redox potentials are obtained from solutions;
moreover, the fact that the redox events are irreversible in most
cases (and certainly for the adducts bearing CO2Me
substituents) require us to treat the implications of the
electrochemical data with caution.
The ligand exchange equilibrium studies demonstrated that
coordinated acetonitrile ligands can be displaced by bromide
and (to a lesser degree) triflate anions, and this substitution is
more facile for the more electrophilic Ni center. This result
indicates that the catalytic activities of cationic RCN adducts
(e.g., acrylonitrile in hydroamination reactions or nitriles in
amidination reactions) might be compromised in the presence
of halides (used as “additives” in some settings) or other in situ
generated anionic species (e.g., deprotonated amines or
alcohols).
The above observations establish that MeCN/Br− ligand
exchange between (R-POCOPi‑Pr)NiBr and [(R′-
POCOPi‑Pr)Ni(NCMe)][OSO2CF3] takes place readily when
the difference in the electronic proprieties of the aromatic ring
substituents R and R′ is relatively small (entry 8 or 9). It is also
noteworthy that the ligand exchange occurs in preference over
other possible reactions. For instance, no redox reaction took
place when we combined a very electron-poor complex such as
Finally, the present study revealed an unexpected side
reaction between the cationic acetonitrile adducts and FeCp2,
presumably a redox reaction giving what we believe is a
trivalent iron product. This intriguing possibility will be the
H
DOI: 10.1021/acs.organomet.5b00272
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(CArHmeta)2). 31P{1H} NMR (121 MHz, CDCl3): δ 191.6 (s). 19F{1H}
NMR (CDCl3, 282 MHz): δ −77.94 (s). 13C{1H} NMR (125 MHz,
CDCl3): δ 3.98 (s(br), 1C, NCCH3), 16.82 (s, 4C, P(CH(CH3)2)4),
subject of future investigations. We will also probe the catalytic
reactivities of the cationic adducts 1−10 for hydroamination of
acrylonitrile and its substituted derivatives, as well as for related
transformations involving outer-sphere nucleophilic attack on
the CN moiety (e.g., amidination and amidation).
v
17.61 (s(br), 4C, P(CH(CH3)2)2), 28.52 (vt, JPvC = 12, 4C,
P(CH(CH3)2)4), 55.67 (s, 1C, OCH3), 93.22 (vt, JPC = 7, 2C,
CArH), 163.47 (s, 1C, CArOCH3), 169.31 (vt, vJPC = 9, 2C, (CArOP)2).
IR (solid state, cm−1): 636 (SO), 1030 (SO3), 1140 (CF3), 1266
(SO3), 1462 (CCAr), 1561 (CCAr), 2293 (CN). UV−vis
((CH2Cl2, [1 × 10−4 M]), λmax, nm (ε, L mol−1 cm−1): 243 (12 834),
299 (1858), 329 (4214), 382 (912). E-chem (NBu4PF6, 10−4 M in dry
CH2Cl2, Eox vs FeCp2 [E1/2 vs FeCp2]): 933 mV [922 mV]. Anal.
Calcd for C22H36F3NNiO6P2S (620.22): C, 42.60; H, 5.85; N, 2.26; S,
5.17. Found: C, 42.83; H, 5.85; N, 1.97; S, 5.20.
EXPERIMENTAL SECTION
■
General Procedures. Unless otherwise indicated, all manipu-
lations were carried out under a nitrogen atmosphere using standard
Schlenk and glovebox techniques. The solvents were dried by passage
over activated alumina contained in MBRAUN-SPS systems and
analyzed by a Coulorimetric Karl Fischer titrator to acceptable water
content. Triethylamine was dried by distillation over CaH2. The
following reagents and NMR solvents were purchased from Sigma-
Aldrich and used without further purification: nickel powder, bromine,
ClP(i-Pr)2, ClP(t-Bu)2, resorcinol, orcinol, 5-methyl-1,3-benzenediol,
methyl 3,5-dihydroxybenzoate, methyl 2,4-dihydroxybenzoate, 4,6-di-
tert-butylresorcinol, ferrocene, silver trifluoromethanesulfonate, C6D6,
and CDCl3. 5-Methoxyresorcinol was purchased from Chemsavers and
used as received. 4-Methoxyresorcinol has been synthesized following
a published procedure.29 The precursor bromo complexes 1′−4′ and
6′−10′ were prepared following a previously reported procedure.15a
NMR spectra were recorded using the following Bruker
spectrometers: AV300, AVII400, and AV500. Chemical shift values
are reported in ppm (δ) and referenced internally to the residual
solvent signals (1H and 13C: 7.26 and 77.16 ppm for CDCl3; 7.16 and
128.06 ppm for C6D6) or externally (31P, H3PO4 in D2O, δ = 0).
Coupling constants are reported in hertz. UV/vis spectra were
recorded on a Varian Cary 500i. The IR spectra were recorded on a
Bruker Alpha-P FTIR (4000−400 cm−1). The elemental analyses were
[{2,6-(i-Pr2PO)2-4-(CO2Me)C6H2}Ni(II)(NCMe)][OSO2CF3], 4.
The procedure described above for the preparation of 2 was used
for this synthesis using 4′ (204 mg, 0.379 mmol, 1.00 equiv). The
desired product was obtained as a yellow solid (189 mg, 81%). Single
crystals suitable for X-ray diffraction were obtained by slow
evaporation in air of a dichloromethane solution layered with hexanes.
1H NMR (300 MHz, CDCl3): δ 1.38 (m, 24H, P(CHC(CH3)2)4),
2.52 (s, 3H, NCCH3), 2.57 (m, 4H, P(CHC(CH3)2)4), 3.85 (s, 3H,
C(O)OCH3), 7.10 (s, 2H, CArH). 31P{1H} NMR (161 MHz, CDCl3):
δ 196.4 (s). 19F{1H} NMR (CDCl3, 282 MHz): δ −78.17 (s).
13C{1H} NMR (75 MHz, CDCl3): δ 4.18 (s(br), 1C, NCCH3), 16.80
v
(s, 4C, P(CH(CH3)2)2), 17.59 (vt, JPC = 3, 4C, P(CH(CH3)2)2),
v
28.62 (vt, JPC = 11, 4C, P(CH(CH3)2)4), 52.41 (s, 1C, OCH3),
106.24 (t, JPC = 7, 1C, CipsoNi), 107.12 (vt, vJPC = 6, 2C, CArH), 133.30
(s, 1C, CArC(O)OCH3), 166.22 (s, 1C, CArC(O)OCH3), 168.85 (vt,
vJPC = 9, 2C, (CArOP)2). IR (solid state, cm−1): 635 (SO), 1029 (SO3),
1140 (CF3), 1277 (SO3), 1397 (CCAr), 1550 (CCAr), 1716 (C
O), 2329 (CN). UV−vis ((CH2Cl2, [1 × 10−4 M]), λmax, nm (ε, L
mol−1 cm−1): 248 (11 577), 329 (4791), 341 (5111, 390 (1697). E-
chem (NBu4PF6, 10−4 M in dry CH2Cl2, Eox vs FeCp2): 1269 mV.
Anal. Calcd for C23H36F3NNiO7P2S (648.24): C, 42.62; H, 5.60; N,
2.16; S, 4.95. Found: C, 43.10; H, 5.71; N, 1.98; S, 4.65.
́
performed by the Laboratoire d’Analyse Elem
́ ́
entaire, Departement de
Chimie, Universite de Montreal.
́
́
[{2,6-(i-Pr2PO)2-4-(Me)C6H2}Ni(NCMe)][OSO2CF3], 2. To an
aluminum-foil-covered Schlenk flask containing 2′ (400 mg, 0.810
mmol, 1.00 equiv) in dry acetonitrile (20.0 mL) was added silver
triflate (250 mg, 0.972 mmol, 1.20 equiv) at rt. The solution was then
agitated for 3 h, filtered to remove the insoluble silver salts, and
evaporated, and the resulting solids were extracted with dichloro-
methane (10.0 mL) and passed through a short Celite pad to remove
the remaining traces of silver salts. Evaporation of the filtrate gave the
desired product as a yellow solid (440 mg, 93%). Single crystals
suitable for X-ray diffraction were obtained by slow evaporation in air
by layering a CDCl3 solution with hexanes. 1H NMR (400 MHz,
CDCl3): δ 1.34 (appq*, 3J = 7, 12H, P(CHC(CH3)2)2), 1.41 (appq, vJ
= 8, 12H, P(CHC(CH3)2)2), 2.23 (s, 3H, CArCH3), 2.42 (s(br), 3H,
[{2,6-(i-Pr2PO)2-4-(Br)C6H2}Ni(NCMe)][OSO2CF3], 5. The proce-
dure described above for the preparation of 2 was used for this
synthesis using 5′ (400 mg, 0.716 mmol, 1.00 equiv). The desired
product was obtained as a yellow solid (412 mg, 86%). Crystals
suitable for diffraction studies were obtained by slow evaporation in air
1
of a dichloromethane solution layered with hexanes. H NMR (500
v
MHz, CDCl3): δ 1.32 (appq, J = 9, 12H, P(CH(CH3)2)2), 1.37
v
(appq, J = 8, 12H, P(CH(CH3)2)2), 2.37 (s(br), 3H, NCCH3), 2.55
3
(sept, JHH = 7, 4H, P(CH(CH3)2)4), 6.64 (s, 2H, CArH). 31P{1H}
NMR (161 MHz, CDCl3): δ 195.7 (s). 19F{1H} NMR (282 MHz,
CDCl3): δ −77.79 (s). 13C{1H} NMR (75 MHz, CDCl3): δ 3.84
3
NCCH3), 2.52 (sept, JHH = 7, 4H, P(CHC(CH3)2)4), 6.28 (s, 2H,
v
(s(br), 1C, NCCH3), 16.76 (s, 4C, P(CH(CH3)2)2), 17.57 (vt, JPC
=
(CArHmeta)2). 31P{1H} NMR (202 MHz, CDCl3) δ: 193.4 (s). 19F{1H}
NMR (282 MHz, CDCl3): δ −78.06 (s). 13C{1H} NMR (75 MHz,
CDCl3): δ 3.91 (s(br), 1C, NCCH3), 16.87 (s, 4C, P(CH(CH3)2)2),
17.68 (vt, vJPC = 3, 4C, P(CH(CH3)2)2), 21.66 (s, 1C, CArCH3), 28.52
v
3, 4C, P(CH(CH3)2)2), 28.58 (vt, JPC = 12, 4C, P(CH(CH3)2)4),
110.01 (vt, JPC = 6, 2C, CArHmeta), 123.54 (s, 1C, CArBr), 168.93 (vt,
v
vJPC = 9, 2C, (CArOP)2). IR (solid state, cm−1): 637 (SO), 1032 (SO3),
1142 (CF3), 1264 (SO3), 1466 (CCAr), 1554 (CCAr), 2302 (C
N). UV−vis ((CH2Cl2, [1 × 10−4 M]), λmax, nm (ε, L mol−1 cm−1):
243 (15 984), 303 (3547), 328 (6245), 386 (1221). E-chem
(NBu4PF6, 10−4 M in dry CH2Cl2, Eox vs FeCp2): 1159 mV. Anal.
Calcd for C24H42BrNNiO5P2S (655.08): C, 37.70; H, 4.97; N, 2.08; S,
4.79. Found: C, 37.79; H, 4.93; N, 2.04; S, 4.71.
v
v
(vt, JPC = 12, 4C, P(CH(CH3)2)4), 107.18 (vt, JPC = 6, 2C, CArH),
142.55 (s, 1C, CArCH3), 168.92 (vt, vJPC = 9, 2C, (CArOP)2). IR (solid
state, cm−1): 636 (SO), 1030 (SO3), 1141 (CF3), 1262 (SO3), 1464
(CCAr), 1555 (CCAr), 2294 (CN). UV−vis ((CH2Cl2, [1 ×
10−4 M]), λmax, nm (ε, L mol−1 cm−1): 242 (11 878), 300 (2302), 327
(2691), 386 (711). E-chem (NBu4PF6, 10−4 M in dry CH2Cl2, Eox vs
FeCp2): 1151 mV. Elemental analysis was not satisfactory for this
complex, because it proved difficult to remove all traces of solvents.
*appq refers to an apparent quartet signal resulting from two
overlapping virtual triplets.
{2,6-(i-Pr2PO)2-4-(Br)C6H2}NiBr, 5′. This compound was prepared
by adapting a literature procedure.18 In a 500 mL round-bottom flask
containing 1-bromo-3,5-dimethoxybenzene (6.40 g, 29.5 mmol) was
added HBr (100 mL), to which was attached a condenser under
nitrogen. The heterogeneous solution was stirred and heated for 24 h
at 120 °C. The solution was then cooled and treated with anhydrous
KHCO3 until the evolution of CO2 ceased and a precipitate formed.
The precipitate was filtered, washed with water, and dried under
vacuum at 80 °C, yielding the known 5-bromoresorcinol as a pale pink
solid (4.77 g, 85%). A dry Schlenk flask charged in the glovebox with
5-bromoresorcinol (1.00 g, 5.20 mmol in 50 mL) and chlorodiiso-
propylphosphine (1.77 mL, 11.1 mmol) was taken out of the box, and
to the contents were added dry THF (50.0 mL) and triethylamine
[{2,6-(i-Pr2PO)2-4-(OMe)C6H2}Ni(NCMe)][OSO2CF3], 3. The
procedure described above for the preparation of 2 was used for this
synthesis using 3′ (415 mg, 0.812 mmol, 1.00 equiv). The desired
product was obtained as a yellow solid (415 mg, 84%). Single crystals
suitable for X-ray diffraction were obtained by slow evaporation in air
1
of an acetone solution layered with hexanes. H NMR (400 MHz,
CDCl3): δ 1.35 (appq, vJ = 8, 12H, P(CHC(CH3)2)2), 1.41 (appq, vJ =
8, 12H, P(CHC(CH3)2)2), 2.39 (s(br), 3H, NCCH3), 2.52 (sept, 3JHH
= 8, 4H, P(CHC(CH3)2)4), 3.73 (s, 3H, OCH3), 6.08 (s, 2H,
I
DOI: 10.1021/acs.organomet.5b00272
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(1.57 mL, 11.6 mmol). The mixture was stirred for 1 h at rt, while the
ammonium salt precipitated as a white solid. Evaporation, followed by
extraction with dry hexane (50.0 mL) and evaporation, gave the crude
product as a pale yellow oil (1.52 g, ca. 68%). The 31P NMR spectrum
of this material showed it to contain the desired product as the major
product, in addition to minor quantities of oxidized phosphines.
Reaction of the crude sample of the ligand (1.52 g, ca. 3.61 mmol)
with the nickel precursor NiBr2(NCCH3)2 (1.30 g, 4.32 mmol) in dry
THF (50 mL) and in the presence of dry triethylamine (0.630 mL,
4.69 mmol) gave a dark green suspension. Stirring for 1 h at rt,
followed by evaporation and extraction with dry hexane (50.0 mL),
gave brown crystals of the desired product (1.21 g, 60%) after
crystallization in air. 1H NMR (400 MHz, C6D6): δ 1.08 (appq, vJ = 7,
12H, P(CH(CH3)2)2), 1.32 (appq, vJ = 8, 12H, P(CH(CH3)2)2), 2.16
[{2,6-(i-Pr2PO)2-3-(CO2Me)C6H2}Ni(NCMe)][OSO2CF3], 8. The
procedure described above for the preparation of 2 was used for this
synthesis using 8′ (405 mg, 0.752 mmol, 1.00 equiv). The desired
product was obtained as an orange solid that was in turn obtained by
1
evaporation in air from an orange oil (360 mg, 89%). H NMR (500
MHz, CDCl3): δ 1.31−1.41 (m, 24H, P(CHC(CH3)2)4), 2.46 (s, 3H,
3
NCCH3), 2.59 (sept, JHH = 4, 4H, P(CHC(CH3)2)4), 3.82 (s, 3H,
3
3
C(O)OCH3), 6.52 (d, JHH = 8, 1H, CArHmeta), 7.72 (d, JHH = 8,
CArHpara). 31P{1H} NMR (202 MHz, CDCl3): δ 195.3 (s). 19F{1H}
NMR (CDCl3, 282 MHz): δ −77.80 (s). 13C{1H} NMR (125 MHz,
CDCl3): δ 4.26 (s(br), 1C, NCCH3), 8.90 (s, 1C, NCCH3), 16.79 (s,
v
2C, P(CH(CH3)2)), 16.90 (s, 2C, P(CH(CH3)2)), 17.58 (vt JPC = 2,
v
2C, P(CH(CH3)2)), 17.69 (vt, JPC = 2, 2C, P(CH(CH3)2)), 28.59
1
1
(ABX, JPC + 3JPC = 33, 2C, P(CH(CH3)2)2), 28.67 (ABX, JPC + 3JPC
3
(sept, JHH = 7, 4H, P(CH(CH3)2)4), 6.80 (s, 2H, CArH). 31P{1H}
NMR (161 MHz, C6D6): δ 190.8 (s). 13C{1H} NMR (75 MHz,
C6D6): δ 16.61 (s, 4C, P(CH(CH3)2)2), 17.69 (s, 4C, P(CH-
= 32, 2C, P(CH(CH3)2)2), 51.96 (s, 1C, C(O)OCH3), 106.79 (dd,
5
3
5
3JPC= 7, JPC = 4; 1C, CArHmeta), 110.64 (dd, JPC= 7, JPC = 5; 1C,
v
v
CArHpara), 134.38 (s, 1C, CArC(O)OCH3)), 165.12 (s, 1C, CArC(O)-
(CH3)2)2), 28.24 (vt, JPC = 11, 4C, P(CH(CH3)2)4), 109.40 (vt, JPC
v
v
v
OCH3), 168.30 (vt, JPC = 9, 1C, CAr (OP)), 171.91 (vt, JPC = 8, 1C,
CArOP). IR (solid state, cm−1): 637 (SO), 1032 (SO3), 1149 (CF3),
1272 (SO3), 1383 (CCAr), 1578 (CCAr), 1712 (CO), 2303
(CN). E-chem (NBu4PF6, 10−4 M in dry CH2Cl2, Eox vs FeCp2):
1262 mV. UV−vis (CH2Cl2,[1 × 10−4 M], λmax, nm (ε, L mol−1
cm−1)): 249 (30 114), 327 (12 630), 346 (2461). Elemental analysis
was not satisfactory for this complex, because it proved difficult to
remove all traces of solvents.
= 6, 1C, CArHmeta), 121.42 (s, 1C, CArBr), 169.20 (vt, JPC = 10, 2C,
(CArOP)2). UV−vis ((CH2Cl2, [1 × 10−4 M]), λmax, nm (ε, mol−1
cm−2): 266 (6360), 306 (2900), 341 (13 600), 391 (2070). E-chem
(NBu4PF6, 10−4 M in dry CH2Cl2, Eox vs FeCp2): 738 mV. Anal. Calcd
for C18H30Br2NiO2P2 (558.88 g/mol): C, 38.68; H, 5.41. Found: C,
39.30; H, 5.59.
[{2,6-(i-Pr2PO)2-3,5-(t-Bu)2C6H}Ni(II)(NCMe)][OSO2CF3], 6. The
procedure described above for the preparation of 2 was used for this
synthesis using 6′ (407 mg, 0.686 mmol, 1.00 equiv). The desired
product was obtained as a yellow solid (410 mg, 87%). The crystals
were obtained by slow evaporation in air of the solid dissolved in
acetonitrile. 1H NMR (300 MHz, CDCl3): δ 1.29 (s, 18H,
(CAr(CH3)3)2), 1.31−1.42 (m, 24H, P(CHC(CH3)2)4), 2.38 (s(br),
3H, NCCH3), 2.55 (m(br), 4H, P(CHC(CH3)2)4), 7.00 (s, 1H,
CArH). 31P{1H} NMR (161 MHz, CDCl3): δ 192.9 (s). 19F{1H} NMR
(CDCl3, 282 MHz): δ −78.04 (s). 13C{1H} NMR (75 MHz, CDCl3):
δ 4.22 (s, 1C, NCCH3), 8.94 (s, 1C, NCCH3), 16.97 (s, 4C,
[{2,6-(tBu2PO)2C6H3}Ni(NCMe)][OSO2CF3], 9. The procedure
described above for the preparation of 2 was used for this synthesis
using 9′ (157 mg, 0.293 mmol, 1.00 equiv). The desired product was
obtained as a yellow solid (178 mg, 94%). Single crystals suitable for
X-ray diffraction were obtained by slow evaporation in air from a
1
dichloromethane solution layered with hexanes. H NMR (500 MHz,
v
CDCl3): δ 1.44 (vt, JPH = 5, 36H, P(C(CH3)3)4), 2.67 (s(br), 3H,
3
3
NCCH3), 6.47 (d, JHH = 10, 2H, (CArHmeta)2), 7.03 (t, JHH = 8, 1H,
CArHpara). 31P{1H} NMR (121 MHz, CDCl3): δ 197.8 (s). 19F{1H}
NMR (282 MHz, CDCl3): δ −77.64(s). 13C{1H} NMR (125 MHz,
CDCl3): δ 4.51 (s(br), 1C, NCCH3), 27.87 (s(br), 12C, P(C-
(CH3)3)4), 40.02 (vt, vJPC = 8, 4C, P(C(CH3)3)4), 106.21 (vt, vJPC = 6,
v
P(CH(CH3)2)2), 17.62 (vt, JPC = 3, 4C, P(CH(CH3)2)2), 28.43 (vt,
vJPC = 12, 4C, P(CH(CH3)2)4), 29.82 (s, 6C, (CAr(C(CH3)3)2), 34.42
v
(s, 4C, (CAr(C(CH3)3)4), 126.31 (s, 1C, CArH), 127.66 (vt, JPC = 5,
2C, (CAr(C(CH3)3))2) 164.36 (vt, vJPC = 11, 2C, (CArOP)2). IR (solid
state, cm−1): 636 (SO), 1028 (SO3), 1143 (CF3), 1251 (SO3), 1459
(CCAr), 1552 (CCAr), 2294 (CN). UV−vis (CH2Cl2, [1 ×
10−4 M], λmax, nm (ε, L mol−1 cm−1)): 299 (2432), 327 (2691), 386
(711). E-chem (NBu4PF6, 10−4 M in dry CH2Cl2, Eox vs FeCp2 [E1/2
vs FeCp2]): 1012 mV [997 mV]. Anal. Calcd for C29H50F3NNiO5P2S
(702.41): C, 49.59; H, 7.17; N, 1.99; S, 4.56. Found: C, 49.30; H, 7.23;
N, 1.95; S, 4.77.
2
2C, (CArHmeta)2), 122.53 (t, JPC = 19, 1C, CipsoNi), 131.10 (s, 1C,
v
CArHpara), 135.89 (s, 1C, OSO2CF3), 169.70 (vt, JPC = 8, 2C,
(CArOP)2). IR (solid state, cm−1): 635 (SO), 1027 (SO3), 1144 (CF3),
1263 (SO3), 1477 (CCAr), 1558 (CCAr), 2293 (CN). E-chem
(NBu4PF6, 10−4 M in dry CH2Cl2, Eox vs FeCp2): 1298 mV. UV−vis
((CH2Cl2, [1 × 10−4 M]), λmax, nm (ε, L mol−1 cm−1)): 258 (15464),
321 (9449), 338 (sh, 2378), 401 (305). Elemental analysis was not
satisfactory for this complex, because it proved difficult to remove all
traces of solvents.
[{2,6-(i-Pr2PO)2-3-(OMe)C6H2}Ni(NCMe)][OSO2CF3], 7. The
procedure described above for the preparation of 2 was used for this
synthesis using 7′ (201 mg, 0.395 mmol, 1.00 equiv). The desired
product was obtained as a yellow solid that was evaporated from a
yellow oil (188 mg, 81%). 1H NMR (400 MHz, CDCl3): δ 1.25−1.37
(m, 24H, P(CHC(CH3)2)4), 2.38 (s(br), 3H, NCCH3), 2.61−2.43 (m,
[{2,6-(t-Bu2PO)2-4-(CO2Me)C6H2}Ni(NCMe)][OSO2CF3], 10. The
procedure described above for the preparation of 2 was used for this
synthesis using 10′ (270 mg, 0.455 mmol, 1.00 equiv). The desired
product was obtained as a yellow solid (256 mg, 80%). Single crystals
suitable for X-ray diffraction were obtained by slow evaporation in air
3
4H, P(CH(CH3)2)4), 3.72 (s, 3H, OCH3), 6.35 (d, JHH = 9, 1H,
3
CArHmeta), 6.64 (d, JHH = 9, 1H, CArHpara). 19F NMR (282 MHz,
1
from a solution of CDCl3 layered with hexanes. H NMR (500 MHz,
CDCl3): δ −78.20 (s). 31P{1H} NMR (161 MHz, CDCl3): δ 195.7 (d,
JPP = 258, 1P), 191.39 (d, JPP = 258, 1P). 13C NMR (101 MHz,
CDCl3): δ 3.53 (s, 1C, NCCH3), 16.65 (s, 2C, P(CH(CH3)2), 16.78
(s, 2C, P(CH(CH3)2)) 17.46 (s, 2C, P(CH(CH3)2)), 17.52 (s, 2C,
v
CDCl3): δ 1.45 (vt, JPH = 8, 36H, P(C(CH3)3)4), 2.70 (s, 3H,
NCCH3), 3.87 (s, 3H, CArCOOCH3), 7.14 (s, 2H, CArHmeta). 31P{1H}
NMR (202 MHz, CDCl3): δ 199.3 (s). 19F{1H} NMR (470 MHz,
CDCl3): δ −77.96 (s). 13C{1H} NMR (125 MHz, CDCl3): δ 4.66 (s,
1
3
P(CH(CH3)2)), 28.57 (ABX, JPC + JPC = 22, 2C, P(CH(CH3)2)2),
v
1C, NCCH3), 27.84 (s, 12C, P(C(CH3)3)4), 40.28 (vt, JPC = 8, 4C,
1
3
28.37 (ABX, JPC + JPC = 22, 2C, P(CH(CH3)2)2), 56.98 (s, 1C,
v
3
P(C(CH3)3)4), 52.47 (s, 1C, OCH3), 107.14 (vt, JPC = 6, 2C,
OCH3), 105.03 (d, JPC = 13, 1C, CArHmeta), 115.11 (s, 1C, CArHpara),
2
2
3
(CArHmeta)2), 129.81 (t, JPC = 19, 1C, CipsoNi), 133.28 (s, 1C,
123.82 (t, JPC = 20, NiCipso), 140.98 (d, JPC = 15, 1C, CArOCH3),
2
4
2
CArCO2CH3), 136.70 (s, 1C, OSO2CF3), 166.19 (s, 1C, CArCO2CH3),
169.41 (vt, vJPC = 8, 2C, (CArOP)2). IR (solid state, cm−1): 635 (SO),
1032 (SO3), 1135 (CF3), 1270 (SO3), 1397 (CCAr), 1552 (C
CAr), 1715 (CO), 2315 (CN). E-chem (NBu4PF6, 10−4 M in dry
CH2Cl2, Eox vs FeCp2): 1457 mV. UV−vis ((CH2Cl2, [1 × 10−4 M]),
λmax, nm (ε, L mol−1 cm−1): 257 (19 757), 335 (18 544), 400 (718).
Elemental analysis was not satisfactory for this complex, because it
proved difficult to remove all traces of solvents.
157.21 (dd, JPC = 13; JPC = 6, 1C, CArOP), 162.52 (dd, JPC = 11;
4JPC =7, 1C, CArOP). IR (solid state, cm−1): 634 (SO), 1027 (SO3),
1144 (CF3), 1259 (SO3), 1461 (CCAr), 1570 (CCAr), 2284 (C
N). UV−vis ((CH2Cl2, [1 × 10−4 M]), λmax, nm (ε, L mol−1 cm−1):
279 (17 577), 323 (11 373), 384 (1881) . E-chem (NBu4PF6, 10−4
M
in dry CH2Cl2, Eox vs FeCp2 [E1/2 vs FeCp2]): 938 mV [967 mV].
Anal. Calcd for C22H36F3NNiO6P2S (620.23): C, 42.60; H, 5.85; N,
2.26; S, 5.17. Found: C, 42.36; H 6.00; N, 2.11; S, 5.29.
J
DOI: 10.1021/acs.organomet.5b00272
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Organometallic Chemistry, van Koten, G.; Milstein, D., Eds.; Springer:
Berlin, 2013; Vol. 40, pp 21−47. (e) Roddick, D. Tuning of PCP
Pincer Ligand Electronic and Steric Properties. In Topics in
Organometallic Chemistry; van Koten, G.; Milstein, D., Eds.; Springer:
Berlin, 2013; Vol. 40, pp 49−88.
ASSOCIATED CONTENT
* Supporting Information
■
S
Tables of crystal data and collection/refinement parameters;
table of additional distances and angles for complexes 1−10
and 5′; figures of molecular diagrams for complexes 2, 3, 5, 5′,
and 9; 1H, 31P, and 13C NMR spectra of complexes 1−10; table
with detailed IR frequency assignments for complexes 1−10;
additional information for IR, electrochemical, and ligand-
exchange studies. Complete details of the X-ray analyses
reported herein have been deposited at the Cambridge
Crystallographic Data Center (CCDC 1044410 (2), 942873
(3), 942874 (4), 942876 (5), 942875 (5′), 942877 (6),
1044411 (9), 1044412 (10)). These data can be obtained free
of charge via www.ccdc.cam.ac.uk/data_request/cif, or by e-
mailing data_request@ccdc.cam.ac.uk or by contacting the
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CD2 1EZ, UK; fax: +44 1223 336033. The
Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.5b00272.
(5) (a) Castonguay, A.; Sui-Seng, C.; Zargarian, D.; Beauchamp, A. L.
Organometallics 2006, 25 (3), 602−608. (b) Kennedy, A. R.; Cross, R.
J.; Muir, K. W. Inorg. Chim. Acta 1995, 231 (1−2), 195−200.
(c) Huck, W. T. S.; Snellink-Ruel, B.; van Veggel, F. C. J. M.;
̈
Reinhoudt, D. N. Organometallics 1997, 16 (20), 4287−4291.
(d) Kozhanov, K. A.; Bubnov, M. P.; Cherkasov, V. K.; Fukin, G.
K.; Abakumov, G. A. Chem. Commun. 2003, 20, 2610−2611.
(e) Kozhanov, K. A.; Bubnov, M. P.; Cherkasov, V. K.; Vavilina, N.
N.; Efremova, L. Y.; Artyushin, O. I.; Odinets, I. L.; Abakumov, G. A.
Dalton Trans. 2008, 21, 2849−2853. (f) Cam
́
pora, J.; Palma, P.; del
Río, D.; Alvarez, E. Organometallics 2004, 23 (8), 1652−1655.
́
́
́
(g) Campora, J.; Palma, P.; del Río, D.; Conejo, M. M.; Alvarez, E.
Organometallics 2004, 23 (24), 5653−5655. (h) Castonguay, A.;
Beauchamp, A. L.; Zargarian, D. Organometallics 2008, 27 (21), 5723−
5732. (i) Boro, B. J.; Duesler, E. N.; Goldberg, K. I.; Kemp, R. A. Inorg.
Chem. 2009, 48 (12), 5081−5087. (j) Groux, L. F.; Belanger-Gariepy,
F.; Zargarian, D. Can. J. Chem. 2005, 83 (6−7), 634−639.
(k) Schmeier, T. J.; Hazari, N.; Incarvito, C. D.; Raskatov, J. A.
Chem. Commun. 2011, 47 (6), 1824−1826. (l) Levina, V. A.; Rossin,
A.; Belkova, N. V.; Chierotti, M. R.; Epstein, L. M.; Filippov, O. A.;
AUTHOR INFORMATION
Corresponding Author
*E-mail: zargarian.davit@umontreal.ca.
■
́
Gobetto, R.; Gonsalvi, L.; Lledos, A.; Shubina, E. S.; Zanobini, F.;
Peruzzini, M. Angew. Chem., Int. Ed. 2011, 50 (6), 1367−1370.
(m) van der Boom, M. E.; Liou, S.-Y.; Shimon, L. J. W.; Ben-David, Y.;
Milstein, D. Inorg. Chim. Acta 2004, 357 (13), 4015−4023.
(n) Castonguay, A.; Beauchamp, A. L.; Zargarian, D. Inorg. Chem.
2009, 48 (7), 3177−3184.
Present Address
†Department of Chemistry, Texas A&M University, College
Station, Texas 77843-3255, United States.
Notes
́ ́
(6) (a) Gomez-Benítez, V.; Baldovino-Pantaleon, O.; Herrera-
́
The authors declare no competing financial interest.
Alvarez, C.; Toscano, R. A.; Morales-Morales, D. Tetrahedron Lett.
2006, 47 (29), 5059−5062. (b) Vabre, B.; Lindeperg, F.; Zargarian, D.
Green Chem. 2013, 15 (11), 3188−3194. (c) Vabre, B.; Petiot, P.;
Declercq, R.; Zargarian, D. Organometallics 2014, 33 (19), 5173−5184.
ACKNOWLEDGMENTS
The authors are grateful to NSERC of Canada for financial
support of this work (Discovery and RTI grants to D.Z.);
■
́
(d) Estudiante-Negrete, F.; Hernandez-Ortega, S.; Morales-Morales,
D. Inorg. Chim. Acta 2012, 387, 58−63. (e) Chakraborty, S.; Krause, J.
A.; Guan, H. Organometallics 2009, 28 (2), 582−586. (f) Zhang, J.;
Medley, C. M.; Krause, J. A.; Guan, H. Organometallics 2010, 29 (23),
6393−6401. (g) Chakraborty, S.; Patel, Y. J.; Krause, J. A.; Guan, H.
Polyhedron 2012, 32 (1), 30−34.
Universite
Michel Simard and Ms. Francine Bel
́
de Montrea
́
l (graduate fellowships to S.L.); Dr.
anger-Gariepy for their
́
́
valuable assistance with crystallography and many interesting
discussions; Ms. Elena Nadezhina for the elemental analyses;
Dr. D. Rochefort for valuable advice regarding the electro-
chemical measurements; and the reviewers of our manuscript,
who recommended that we measure the open-circuit (rest)
potentials of our complexes and offered many valuable
suggestions regarding the reporting of the NMR data. S.L. is
also grateful to Centre in Green Chemistry and Catalysis for a
travel award, to J.-P. Cloutier for the loan of complexes used in
the ligand exchange studies, and to all group members for many
valuable discussions and practical advice.
(7) (a) Pandarus, V.; Zargarian, D. Chem. Commun. 2007, 9, 978−
980. (b) Hao, J.; Vabre, B.; Mougang-Soume, B.; Zargarian, D. Chem. -
Eur. J. 2014, 20 (39), 12544−12552. (c) Hao, J.; Mougang-Soume, B.;
Vabre, B.; Zargarian, D. Angew. Chem., Int. Ed. 2014, 53 (12), 3218−
3222. (d) Pandarus, V.; Castonguay, A.; Zargarian, D. Dalton Trans.
2008, 35, 4756−4761.
(8) (a) Spasyuk, D. M.; Zargarian, D.; van der Est, A. Organometallics
2009, 28 (22), 6531−6540. (b) Spasyuk, D. M.; Zargarian, D. Inorg.
Chem. 2010, 49 (13), 6203−6213. (c) Zhang, B.-S.; Wang, W.; Shao,
D.-D.; Hao, X.-Q.; Gong, J.-F.; Song, M.-P. Organometallics 2010, 29
(11), 2579−2587. (d) Niu, J.-L.; Chen, Q.-T.; Hao, X.-Q.; Zhao, Q.-
X.; Gong, J.-F.; Song, M.-P. Organometallics 2010, 29 (9), 2148−2156.
(e) Spasyuk, D. M.; Gorelsky, S. I.; van der Est, A.; Zargarian, D. Inorg.
Chem. 2011, 50 (6), 2661−2674. (f) Yang, M.-J.; Liu, Y.-J.; Gong, J.-F.;
Song, M.-P. Organometallics 2011, 30 (14), 3793−3803. (g) Sanford,
J.; Dent, C.; Masuda, J. D.; Xia, A. Polyhedron 2011, 30 (6), 1091−
1094.
(9) (a) Fan, L.; Foxman, B. M.; Ozerov, O. V. Organometallics 2004,
23 (3), 326−328. (b) Ozerov, O. V.; Guo, C.; Fan, L.; Foxman, B. M.
Organometallics 2004, 23 (23), 5573−5580. (c) Liang, L.-C.; Chien,
P.-S.; Huang, Y.-L. J. Am. Chem. Soc. 2006, 128 (49), 15562−15563.
(d) Liang, L.-C.; Chien, P.-S.; Lin, J.-M.; Huang, M.-H.; Huang, Y.-L.;
Liao, J.-H. Organometallics 2006, 25 (6), 1399−1411. (e) Adhikari, D.;
Huffman, J. C.; Mindiola, D. J. Chem. Commun. 2007, 43, 4489−4491.
(f) Adhikari, D.; Pink, M.; Mindiola, D. J. Organometallics 2009, 28
(7), 2072−2077. (g) Fryzuk, M. D.; Montgomery, C. D. Coord. Chem.
Rev. 1989, 95, 1. (h) Ingleson, M. J.; Fullmer, B. C.; Buschhorn, D. T.;
REFERENCES
■
(1) Moulton, C. J.; Shaw, B. L. J. Chem. Soc., Dalton Trans. 1976, 11,
1020−1024.
(2) (a) van Koten, G.; Jastrzebski, J. T. B. H.; Noltes, J. G.; Spek, A.
L.; Schoone, J. C. J. Organomet. Chem. 1978, 148 (3), 233−245.
(b) van Koten, G.; Timmer, K.; Noltes, J. G.; Spek, A. L. J. Chem. Soc.,
Chem. Commun. 1978, 6, 250−252.
(3) Van Koten, G. Pure Appl. Chem. 1989, 61 (10), 1681−94.
(4) (a) van Koten, G. J. Organomet. Chem. 2013, 730, 156−164.
(b) van Koten, G. The Monoanionic ECE-Pincer Ligand: A Versatile
Privileged Ligand PlatformGeneral Considerations. In Topics in
Organometallic Chemistry; van Koten, G.; Milstein, D., Eds.; Springer:
Berlin, 2013; Vol. 40, pp 1−20. (c) Morales-Morales, D.; Jensen, C. G.
M. The Chemistry of Pincer Compounds; Elsevier Science, 2011.
(d) Poverenov, E.; Milstein, D. Noninnocent Behavior of PCP and
PCN Pincer Ligands of Late Metal Complexes. In Topics in
K
DOI: 10.1021/acs.organomet.5b00272
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Fan, H.; Pink, M.; Huffman, J. C.; Caulton, K. G. Inorg. Chem. 2008,
47 (2), 407−409. (i) Fryzuk, M. D.; MacNeil, P. A. J. Am. Chem. Soc.
1981, 103 (12), 3592−3593.
(10) (a) Vechorkin, O.; Proust, V.; Hu, X. J. Am. Chem. Soc. 2009,
131 (28), 9756−9766. (b) Madhira, V. N.; Ren, P.; Vechorkin, O.; Hu,
X.; Vicic, D. A. Dalton Trans. 2012, 41, 7915. (c) Breitenfeld, J.;
Scopelliti, R.; Hu, X. Organometallics 2012, 31 (6), 2128−2136.
(11) Vabre, B.; Canac, Y.; Duhayon, C.; Chauvin, R.; Zargarian, D.
Chem. Commun. 2012, 48 (84), 10446−10448.
(12) (a) Zargarian, D.; Castonguay, A.; Spasyuk, D., ECE-Type
Pincer Complexes of Nickel. In Topics in Organometallic Chemistry; van
Koten, G.; Milstein, D., Eds.; Springer: Berlin, 2013; Vol. 40, pp 131−
173. (b) Wang, Z.-X.; Liu, N. Eur. J. Inorg. Chem. 2012, 2012 (6),
901−911.
(13) Pandarus, V.; Zargarian, D. Organometallics 2007, 26 (17),
4321−4334.
(27) Grove, D. M.; van Koten, G.; Ubbels, H. J. C.; Zoet, R.; Spek, A.
L. Organometallics 1984, 3 (7), 1003−1009.
(28) To better understand the curious formation of the triflate
derivatives (NCN)Ni(OTf) and (PCsp3Pi‑Pr)Ni(OTf) in non-negligible
amounts, we set out to establish if they arise from the corresponding
acetonitrile adducts bearing triflate counter-anions. The following
experiment was conducted for this purpose: to a 0.04 or 0.08 mM
C6D6 solution of complex 1 was added 1−10 equiv of silver triflate, the
sample was agitated periodically, and the 31P NMR spectra were
recorded after 24 h. The original signal broadened slightly, but there
was no trace of the triflate derivative (singlet at 185 ppm) or any other
new species. This led us to believe that there might indeed be an
equilibrium where the triflate moiety would exchange with the
acetonitrile ligand, albeit shifted much more to the acetonitrile adduct.
(29) Foti, M. C.; Daquino, C.; Mackie, I. D.; DiLabio, G. A.; Ingold,
K. U. J. Org. Chem. 2008, 73, 9270−9282.
(14) (a) Chakraborty, S.; Zhang, J.; Krause, J. A.; Guan, H. J. Am.
Chem. Soc. 2010, 132 (26), 8872−8873. (b) Salah, A. B.; Offenstein,
C.; Zargarian, D. Organometallics 2011, 30 (20), 5352−5364.
(c) Castonguay, A.; Spasyuk, D. M.; Madern, N.; Beauchamp, A. L.;
Zargarian, D. Organometallics 2009, 28 (7), 2134−2141. (d) Knapen,
J. W. J.; van der Made, A. W.; de Wilde, J. C.; van Leeuwen, P. W. N.
M.; Wijkens, P.; Grove, D. M.; van Koten, G. Nature 1994, 372
(6507), 659−663.
(15) (a) Vabre, B.; Spasyuk, D. M.; Zargarian, D. Organometallics
2012, 31 (24), 8561−8570. (b) Vabre, B.; Lambert, M. L.; Petit, A.;
Ess, D. H.; Zargarian, D. Organometallics 2012, 31 (17), 6041−6053.
(c) Salah, A. B.; Zargarian, D. Dalton Trans. 2011, 40 (35), 8977−
8985.
(16) (a) Choi, J.; MacArthur, A. H. R.; Brookhart, M.; Goldman, A.
S. Chem. Rev. 2011, 111 (3), 1761−1779. (b) Zhu, K.; Achord, P. D.;
Zhang, X.; Krogh-Jespersen, K.; Goldman, A. S. J. Am. Chem. Soc.
2004, 126 (40), 13044−13053. (c) Krogh-Jespersen, K.; Czerw, M.;
Zhu, K.; Singh, B.; Kanzelberger, M.; Darji, N.; Achord, P. D.;
Renkema, K. B.; Goldman, A. S. J. Am. Chem. Soc. 2002, 124 (36),
10797−10809. (d) Gottker-Schnetmann, I.; White, P.; Brookhart, M.
̈
J. Am. Chem. Soc. 2004, 126 (6), 1804−1811. (e) Roddick, D. M.
Topics in Organometallic Chemistry; 2013; Vol. 40 (Organometallic
Pincer Chemistry), pp 49−88.
̀
(17) Lefevre, X.; Durieux, G.; Lesturgez, S.; Zargarian, D. J. Mol.
Catal. A: Chem. 2011, 335 (1−2), 1−7.
(18) Pratt, D. A.; Pesavento, R. P.; van der Donk, W. A. Org. Lett.
2005, 7 (13), 2735−2738.
(19) Based on two CCDC surveys, both conducted on November 4,
2014, of all structurally characterized nickel compounds containing a
covalently bonded triflate anion and of all complexes bearing at least
one Ni-NCR moiety.
(20) It has been shown that the presence of t-Bu substituents ortho to
the phosphinite moieties in (POCOP)NiBr generates significantly
shorter P−Ni bond distances. See ref 15a.
(21) (a) Chakraborty, S.; Patel, Y. J.; Krause, J. A.; Guan, H. Angew.
Chem., Int. Ed. 2013, 52 (29), 7523−7526. (b) Lefevre, X.; Spasyuk, D.
M.; Zargarian, D. J. Organomet. Chem. 2011, 696 (4), 864−870.
(22) Liang, S.; Wang, H.; Deb, T.; Petersen, J. L.; Yee, G. T.; Jensen,
M. P. Inorg. Chem. 2012, 51 (23), 12707−12719.
(23) Rozenel, S. S.; Kerr, J. B.; Arnold, J. Dalton Trans. 2011, 40 (40),
10397−10405.
(24) The relative ν(CN) values observed in these adducts is also
reflected in the greater electrophilicity of the double-bond moiety in
the acrylonitrile adduct of 1 in regard to its aliphatic counterpart (see
Table 2).
(25) It should be noted that this comparison was based on ν(CN)
values measured under identical experimental conditions, i.e., using
KBr disks for the comparison between [(H-POCOPPh)Ni(NCMe)]-
[OSO2CF3] and 1, and using solid samples in ATR-IR to compare 1 to
9.
(26) Note: Both of these signals are fairly broad, LW1/2 being ∼118
Hz for the new signal at ca. 193−194 ppm and 115 Hz for the cationic
adduct 8 signal at ca. 195 ppm.
L
DOI: 10.1021/acs.organomet.5b00272
Organometallics XXXX, XXX, XXX−XXX