Addition of amines and phenols to acrylonitrile derivatives catalyzed by the POCOP-type pincer complex [{κP,κC, κP-2,6-(i-Pr2PO)2C6H 3}Ni(NCMe)][OSO2CF3]

Article abstract of DOI:10.1016/j.molcata.2010.11.010

The pincer-type complex [{κ^P,κ^C, κ^P-2,6-(i-Pr₂PO)₂C₆H ₃}Ni(NCMe)][OSO₂CF₃] (1) can serve as a precatalyst for the regioselective, anti-Markovnikov addition of nucleophiles to activated olefins. The catalyzed additions of aliphatic amines to acrylonitrile, methacrylonitrile, and crotonitrile proceed at room temperature and give quantitative yields of products resulting from the formation of C-N bonds. On the other hand, aromatic amines or alcohols are completely inert toward methacrylonitrile and crotonitrile, and much less reactive toward acrylonitrile, requiring added base, heating, and extended reaction times to give good yields. The catalytic reactivities of 1 are thought to arise from the substitutional lability of the coordinated acetonitrile that allows competitive coordination of the nitrile moiety in the olefinic substrates; this binding enhances the electrophilicity of the CC moiety, rendering them more susceptible to attack by nucleophiles. In some cases, RCN → Ni binding results in double bond isomerization/migration (allyl cyanide) or attack of nucleophiles at the nitrile moiety (cinnamonitrile and 4-cyanostyrene). Reaction of morpholine with 1 at 60 °C led to formation of the amidine derivative 2 that has been characterized by X-ray crystallography.

Full text of DOI:10.1016/j.molcata.2010.11.010

Journal of Molecular Catalysis A: Chemical 335 (2011) 1–7
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Editor’s choice paper
Addition of amines and phenols to acrylonitrile derivatives catalyzed by the
POCOP-type pincer complex [{␬P,␬C,␬P-2,6-(i-Pr2PO)2C6H3}Ni(NCMe)][OSO2CF3]
Xavier Lefèvre, Guillaume Durieux, Stéphanie Lesturgez, Davit Zargarian∗
Département de Chimie, Université de Montréal, Montréal (Québec), Canada, H3C 3J7
a r t i c l e i n f o
a b s t r a c t
The pincer-type complex [{␬P,␬C,␬P-2,6-(i-Pr2PO)2C6H3}Ni(NCMe)][OSO2CF3] (1) can serve as a precata-
lyst for the regioselective, anti-Markovnikov addition of nucleophiles to activated olefins. The catalyzed
additions of aliphatic amines to acrylonitrile, methacrylonitrile, and crotonitrile proceed at room tem-
perature and give quantitative yields of products resulting from the formation of C–N bonds. On the other
hand, aromatic amines or alcohols are completely inert toward methacrylonitrile and crotonitrile, and
much less reactive toward acrylonitrile, requiring added base, heating, and extended reaction times to
give good yields. The catalytic reactivities of 1 are thought to arise from the substitutional lability of the
coordinated acetonitrile that allows competitive coordination of the nitrile moiety in the olefinic sub-
strates; this binding enhances the electrophilicity of the C C moiety, rendering them more susceptible to
attack by nucleophiles. In some cases, RCN → Ni binding results in double bond isomerization/migration
(allyl cyanide) or attack of nucleophiles at the nitrile moiety (cinnamonitrile and 4-cyanostyrene). Reac-
tion of morpholine with 1 at 60 ◦C led to formation of the amidine derivative 2 that has been characterized
by X-ray crystallography.
Article history:
Received 23 August 2010
Received in revised form
10 November 2010
Accepted 10 November 2010
Available online 18 November 2010
Keywords:
Hydroamination
Hydroalkoxylation
Hydroaryloxylation
Alcoholysis
Pincer complexes
Nickel–amidine complexes
© 2010 Published by Elsevier B.V.
1. Introduction
[17]. In previous reports, we have shown that cationic Ni(II) com-
plexes based on PCP- and POCOP-type pincer ligands serve as
Direct addition of N–H bonds of amines and O–H bonds of alco-
hols to olefins is an attractive, atom-efficient approach for the
preparation of substituted amines or ethers since no by-products
are formed [1]. Many reports have described olefin hydroamina-
tion processes catalyzed by complexes of lanthanides [2], group
4 metals [3], Rh [4], Ir [5], Ni [6], Pd [7], Pt [8], and Cu [9]. Inter-
estingly, a number of reports have also shown that homogeneous,
intermolecular hydroamination of styrene by electron-rich ani-
lines can be promoted by acids such as HOTf [10] or even HCl
[11], while PhNH3B(C6F5)4·Et2O promotes both the hydroami-
nation and hydroarylation of styrene and cyclic olefins such as
norbornene and cis-cyclooctene [12]. Similarly, there are multi-
ple different catalysts for hydroalkoxylation reactions, including
strong bases [13] or acids [14], nucleophilic phosphines in the pres-
ence of ␣,␤-unsaturated olefins [15], and metal-based catalysts
[9,16].
competent pre-catalysts for the addition of aliphatic amines and
aniline to acrylonitrile and its derivatives [17d,g–i]. For example,
the complex [POCOP)Ni(NCCH CH2)]+ (POCOP = ␬P,␬C,␬P-2,6-{(i-
Pr)2PO}2C6H3, Chart 1) promotes the anti-Markovnikov addition
of morpholine and cyclohexylamine to acrylonitrile, methacryloni-
trile and crotonitrile with turnover numbers (TON) of 80–2000
(room temperature, 5 min–3 h), but addition of aniline to acry-
lonitrile was less facile (TON = 100–150 at 115 ◦C over 4–24 h)
[17g,i]. The analogous PCP-type complexes (Chart 1) also pro-
mote the addition of aniline to acrylonitrile, but the reactivities of
these complexes are inferior to those of their POCOP counterparts
[17h,i].
As an extension to our previous studies, we have screened the
catalytic reactivities of a number of cationic adducts and neutral
precursors for the addition of amines and alcohols to activated
olefins. The complex [(POCOP)Ni(NCMe)][OSO2CF3], 1, was found
to be a practical precatalyst for the purposes of further studies
thanks to its ease of synthesis and stability toward air oxidation and
hydrolysis. This report describes the catalytic activities of 1 for the
addition of aliphatic amines, aniline and its substituted derivatives,
substituted phenols, and catechol to acrylonitrile, methacryloni-
trile and crotonitrile (Eq. (1)). Also presented is the solid state
structure of the amidine adduct 2 that was obtained from the ther-
mal reaction of 1 with morpholine.
Our group has been interested in development of synthetic
routes to pincer complexes of nickel and exploring their reactivities
as pre-catalysts for a variety of reactions, including hydroamina-
tion and hydroalkoxylation of activated olefins (Michael additions)
∗
Corresponding author. Tel.: +1 514 343 2247; fax: +1 514 343 7586.
E-mail address: zargarian.davit@umontreal.ca (D. Zargarian).
1381-1169/$ – see front matter © 2010 Published by Elsevier B.V.
doi:10.1016/j.molcata.2010.11.010

2
X. Lefèvre et al. / Journal of Molecular Catalysis A: Chemical 335 (2011) 1–7
Chart 1.
2. Results and discussion
alcohols. For instance, reacting acrylonitrile with aliphatic amines
for 24 h at room temperature and in the absence of a catalyst gave
poor %yields (or turnover numbers, TON) of the addition product
with Et2NH (25), i-PrNH2 (31), and N,Nꢀ-dibenzyl(ethylenediamine)
(5). By comparison, the corresponding Ni-catalyzed reactions pro-
ceeded with quantitative yields in 1 h or less (runs 1–3, Table 1;
turnover frequency, TOF ∼ 100–1000/h). It is worth noting that
double addition products were obtained quantitatively when two
equivalents of acrylonitrile were reacted with i-PrNH2 or N,Nꢀ-
dibenzyl(ethylenediamine) (Eq. (2)).
Catalyzed addition ofaromatic amines toacrylonitrile wasfound
to be much less efficient, but the presence in the catalytic mixture
of external base such as NEt3 led to substantial improvements. For
instance, addition of aniline and 3-methyl-aniline in the presence
of one equivalent of NEt3 proceeded to completion at room tem-
perature in 4 and 16 h, respectively (runs 4 and 5). As expected,
the use of sub-stoichiometric quantities of NEt3 (0.05–0.1 equiv.)
was equally effective in terms of yields, but the catalysis proceeded
more slowly. The presence of Me substituents at the ortho position
2.1. Catalytic studies
Various amines and alcohols were screened for their nucle-
ophilicity toward acrylonitrile derivatives and other nitrile
functionalized olefins. As expected, aliphatic amines were found to
be the most reactive nucleophiles, followed by substituted anilines
and phenols; curiously, aliphatic alcohols proved to be completely
inert under the reaction conditions employed. Among the olefins
examined, acrylonitrile was much more reactive than crotonitrile,
methacrylonitrile, and allyl cyanide, while cinnamonitrile and 4-
cyano-styrene did not undergo addition at the olefin moiety.
Most catalysis runs were carried out in C6D6 using 1% of 1 and
a 1:1 ratio of the substrates. The experiments were monitored by
NMR, and GC/MS analysis was used to identify the products and
to determine conversions/yields. Control experiments showed that
non-catalyzed Michael additions to acrylonitrile proceed sluggishly
with most aliphatic amines and not at all with aromatic amines or

X. Lefèvre et al. / Journal of Molecular Catalysis A: Chemical 335 (2011) 1–7
3
Table 1
Catalytic addition of nucleophiles to acrylonitrile catalyzed by 1.a
Run
Nucleophile
Et2NH
Temp.
Time
Yield (%) or TON
TOF (TON/h)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
r.t.
r.t.
r.t.
r.t.
r.t.
<5 min
<5 min
1 h
100
100
100
100
100
83
56
73
42
–
100
58
69
78
60
>1000
>1000
100
25
b
i-PrNH2
(PhCH2NH)2(CH2CH2)b
Anilinec
4 h
3-Methylanilinec
2,5-Dimethylanilinec
2,4,6-Trimethylanilinec
4-Chloroanilined
4-Nitroanilinec
16 h
24 h
24 h
24 h
24 h
24 h
4 h
24 h
24 h
24 h
24 h
24 h
∼6
∼3
∼2
∼3
1.5
–
60 ◦
60 ◦
60 ◦
60 ◦
60 ◦
r.t.
C
C
C
C
C
Ph2NHc
3-Methylphenolc
2-Methylphenolc
2,4,6-Trimethylphenolc
Catecholc,e
25
60 ◦
60 ◦
60 ◦
60 ◦
60 ◦
C
C
C
C
C
∼2
∼3
∼3
2.5
–
4-Phenylphenolc
Pentafluorophenolc
Trace
a
The catalytic tests were conducted in NMR tubes containing 0.5 mL of C6D6, 1 mol% of the precatalyst 1, 0.5 mmol each of acrylonitrile and the nucleophile, and (where
needed) one equivalent of NEt3. The yields are based on GC/MS analyses.
b
Use of two equivalents of acrylonitrile in this run gave the product of double addition.
One equivalent of NEt3 used in this run.
Traces of the homocoupling product, (4-Cl-C6H4)NH(4-NH2-C6H4), were also obtained from this run.
Only product of single addition formed in this run even though two equivalents of acrylonitrile were used.
c
d
e
on the ring led to reduced activities (runs 6 and 7). Lower yields
were also obtained when electron-withdrawing groups such as Cl
and NO2 were present, while the addition was inhibited altogether
with Ph2NH (runs 8–10); the reaction with p-chloroaniline gave
trace amounts of the homocoupling product (4-Cl-C6H4)NH(4-
NH2-C6H4) (MW: 208; Eq. (3)).
aniline (run 5) proceeded quite poorly even at 60 ◦C, giving 9% yield
with methacrylonitrile and no reaction at all with crotonitrile. The
alcohols 3-methylphenol and benzyl alcohol were also unreactive
(runs 6 and 7).
2.2. Comparison of catalytic reactivities to literature precedents
It was gratifying to find that the addition to acrylonitrile was
also possible with alcohols as nucleophiles. Initial tests showed
that only phenols were effective nucleophiles, aliphatic alcohols
being completely inactive for this reaction. As in the case of addi-
tions with anilines, NEt3 was essential for the hydroaryloxylation
of acrylonitrile. Methyl-substituted phenols as well as catechol and
4-phenylphenol underwent the addition reaction under the same
conditions as anilines (60 ◦C, 24 h; runs 11–15), whereas pentafluo-
rophenol was virtually unreactive. Interestingly, only one of the OH
moieties of catechol reacted to give 2-(OCH2CH2CN)-phenol even
when the reaction was run with two equivalents of acrylonitrile
(Eq. (4)).
Most literature reports on transition metal-catalyzed additions
of N–H and O–H bonds to activated olefins appear to be focused
on the reactivities of acrylates, acrylamides, and ␣,␤-unsaturated
ketones-substrates that are not reactive in our system (vide infra).
Moreover, there are relatively few reports on Ni-catalyzed addi-
tions to acrylonitrile and its derivatives [18]. Nevertheless, the
following comparisons to some of the more pertinent literature
precedents will help place the present results in the context of this
area of study.
2.2.1. Pd- and Ni-catalyzed hydroaminations
The crucial role of 1 in promoting Michael additions was con-
firmed for reactions involving methacrylonitrile or crotonitrile, for
which control experiments indicated no reactivity in the absence of
precatalyst even with aliphatic amines (24 h at room temperature).
Thus, the presence of 1% 1 led to quantitative yields for the addi-
tion of morpholine, Et2NH, cyclohexylamine, and i-PrNH2 to both
methacrylonitrile and crotonitrile in about 10 min at room tem-
perature (Table 2, runs 1–4); the regioselectivity of the addition
remains anti-Markovnikov with both methacrylonitrile and cro-
tonitrile (Eq. (5)). On the other hand, the catalyzed addition with
Trogler’s group has reported some of the earlier studies on the
addition of aniline to acrylonitrile [7a]. Using a system consist-
ing of 10 mol% of [PhNH3]BPh4 and 2 mol% of the PCP-type pincer
complex {R2P(CH2)2CH(CH2)2PR2}Pd-alkyl (R = t-Bu), these work-
ers obtained the product of anti-Markovnikov addition with up to
40–50 TON at 35 ◦C. It is intriguing to note that the presence of an
acid is crucial for this system, whereas in our system addition of ani-
lines was facilitated by an added base. Following Trogler’s report,
Hartwig’s group showed that different combinations of Pd(II) salts
(2%) and pincer ligands (2–10%) catalyze the addition of piperidine
Table 2
Addition of nucleophiles to methacrylonitrile and crotonitrile catalyzed by 1.a
Run
Nucleophile
Temp.
Time
Yield (%) or TON
TOF (TON/h)
1
2
3
4
5
6
7
Morpholine
r.t.
r.t.
r.t.
r.t.
10 min
10 min
10 min
10 min
24 h
100
100
100
100
9
∼600
∼600
∼600
∼600
∼0.4
–
Diethylamine
Cyclohexylamine
Isopropylamine
Anilineb,c
60 ◦
60 ◦
60 ◦
C
C
C
3-Methylphenolc
Benzyl alcoholc
24 h
24 h
–
–
–
a
The catalytic reactions were performed as indicated for the additions to acrylonitrile. Unless otherwise stated, the yields are the same for the reactions of methacrylonitrile
and crotonitrile.
b
The addition of aniline to methacrylonitrile gave a 9% yield, while no addition took place on crotonitrile.
Reaction yields for these runs were unaffected by the presence of NEt3.
c

4
X. Lefèvre et al. / Journal of Molecular Catalysis A: Chemical 335 (2011) 1–7
ever, that the nitrile binding is fairly labile and allows the formation
of minor quantities of the ␲-bound isomer, which is then attacked
by aniline to give the hydroamination product [7a]. Support for a
mechanism involving attack of nucleophiles on ␲-bound olefins
also comes from Abu-Omar’s studies on the Pd-catalyzed addition
of benzyl alcohol to methyl vinyl ketone (MVK) [16d]; interest-
ingly, acrylonitrile was unreactive in this system, because it binds
Pd center via its nitrile moiety. In contrast to the above reports, Yi
et al. have proposed a mechanism involving a bifunctional Ru(II)-
acetamido catalyst that promotes both a ␬N-nitrile binding to an
empty coordination site (Lewis acidity) and a heterolytic activa-
tion of the alcohol O–H bond moderated by the acetamido moiety
Scheme 1.
to crotonitrile, methacrylonitrile, and alkyl methacrylates and cro-
tonates at room temperature, whereas addition of aniline required
extended periods of heating at 100 ◦C [7e]. In comparison, our
POCOP–Ni system catalyzes the addition of aniline to acrylonitrile
at room temperature and shorter reaction times with TON of 100,
but is ineffective for the addition of anilines to crotonitrile and
methacrylonitrile. Togni’s group has studied the catalytic activities
of dicationic complexes of Ni(II) ligated by triphosphines for the
addition of morpholine, piperidine, aniline, and substituted anilines
to crotonitrile, methacrylonitrile, and alkyl acrylates and crotonates
[6b,c]. The addition of aniline to crotonitrile and methacrylonitrile
in this system proceeds over 24 h at room temperature with 70 and
35 catalytic turnover numbers (TON), respectively, whereas addi-
tion of substituted anilines gives lower TON (5–15). The advantages
of this system include room temperature hydroamination with aro-
matic amines and a wider scope of olefins, whereas addition of
aliphatic amines seems to be more facile in our system. The reac-
tivities of the above-discussed systems for the addition of alcohols
are not known.
(Lewis basicity); the in situ generated alkoxide then adds to the C
C
moiety of the substrate that has been activated by the RCN → Ru
interaction [21].
A number of observations suggest that hydroamination and
hydroaryloxylation reactions promoted by 1 involve attack of
nucleophiles on acrylonitrile (or its derivatives) coordinated to
the cationic Ni center in 1 through their nitrile moiety. First, we
have noted the facile formation of cationic pincer complexes fea-
turing RCN → Ni binding, and a number of these complexes have
been isolated and fully characterized [17g,h,j]. Moreover, NMR
monitoring of reaction mixtures containing 1, acrylonitrile, aniline
or m-cresol, and NEt3 showed that the only P-containing species
observed throughout the catalysis display signals in the same
region as authenticated nitrile adducts (i.e., ca. 193–194 ppm);
similar observations were made for the analogous reactions with
methacrylonitrile (193 ppm) and crotonitrile (194 ppm). In con-
trast, no reaction was detected between 1 and NEt3, morpholine,
aniline, m-cresol, or olefinic substrates that do not undergo
hydroamination or hydroaryloxylation in our system (methyl acry-
late, methyl methacrylate, 1-hexene, and styrene). We conclude,
therefore, that the coordinated acetonitrile moiety in 1 can only be
displaced by a substrate possessing a nitrile functionality.
2.2.2. Pd-, Ru-, and Cu-catalyzed hydroalkoxylations
Abu-Omar’s group has reported Pd(II)-based complexes that
promote the addition of a variety of alcohols and aniline to methyl
vinyl ketone (5–100 TON at r.t.), but no reactivity is observed
with acrylonitrile [16d]. Yi et al. have reported a Ru(II)-catalyzed
hydroalkoxylation of acrylonitrile and related derivatives includ-
ing crotonitrile, methacrylonitrile, 1-cyano-cyclohexene [21]. This
system allows high yield addition of aliphatic alcohols under mild
reaction conditions (r.t. to 40 ◦C; 1–24 h), but no examples are given
for the addition of phenols. Gunnoe’s group has reported that the
complexes LCuX (L = N-heterocyclic carbine; X = NHPh, OEt, OPh)
catalyze the addition of both N–H and O–H bonds to acryloni-
trile, crotonitrile, methyl acrylate, cyclohexenone, and methyl vinyl
ketone [9]. This system promotes the room temperature addition
of Et2NH, n-PrNH2, PhCH2NH2 and PhNH2 to acrylonitrile with
up to 20 TON, but addition to crotonitrile is much more sluggish
(80 ◦C, 40 h, TON ∼ 10). Interestingly, addition of PhOH to acryloni-
trile requires extended heating (80 ◦C, 40 h, TON ∼ 13), whereas
addition of EtOH proceeds at r.t. (TON ∼ 19 over 20 h). This system
is clearly superior to ours with respect to the reaction of aliphatic
alcohols, but our system is more efficient in the addition of aliphatic
amines, aniline, and phenols.
We have also considered but ruled out the viability of other
mechanistic scenarios. For instance, the involvement of ␲-bound
olefins was ruled out, because we found no spectroscopic evi-
dence in our system for the presence of even minor quantities of
such intermediates; in addition, olefinic substrates lacking a nitrile
moiety (e.g., unfunctionalized olefins and acrylates) are inactive
in our system. This fact, that in our system the scope of reactive
olefins is limited to those cyano olefins that coordinate readily to
[(POCOP)Ni]+, also argues against the involvement of a mechanism
not requiring olefin coordination. Such a mechanism, involving
an attack on uncoordinated olefins by nucleophilic M–NR2 and
M–OR species, has been proposed by Gunnoe for Cu-catalyzed
hydroamination and hydroalkoxylation of a range of olefins [9]. It is
reasonable to suppose that such a pathway should require the for-
mation of fairly nucleophilic M–NR2 and M–OR species, whereas we
have found little or no support for the formation of neutral species
(POCOP)Ni–X (X = NR2, OAr) under the conditions of the catalytic
reactions, as described below.
Monitoring the reaction of the neutral Ni–OSO2CF3 precursor
with excess morpholine led to partial replacement of the origi-
nal peak at 186 ppm by a broad new signal at ca. 185 ppm and a
sharper signal at ca. 179 ppm; the approximate ratio of these sig-
nals was 35:50:5 [19]. A similar experiment with excess aniline
led to the broadening of the 31P signal of the precursor and the
emergence of a new broad signal at 188 ppm; these two signals
were poorly resolved, making their integration unreliable, but the
ratio of the two peaks was approximately 1:1. Repeating the lat-
ter experiment in the presence of NEt3 led to the appearance of
a weak but sharp signal at 179 ppm (<1% by signal intensity). Our
efforts at driving these reactions to completion and isolating the
new species have not borne fruit yet, and so we cannot confirm
or rule out the formation of neutral anilido derivatives from the
2.3. Mechanistic considerations
The most frequently cited mechanistic proposal for late transi-
tion metal-catalyzed Michael-type additions involves outer-sphere
attack of an uncoordinated nucleophile on an olefin that is coordi-
nated to the electrophilic metal centre via the C C moiety or the
functional group (COOR, CN, etc.), followed by proton transfer to
generate the product (Scheme 1) [7a,16d]. For example, Trogler
showed that the major species observed in solution during the Pd-
catalyzed addition of aniline to acrylonitrile is the ␬N-acrylonitrile
adduct, [(PCP)Pd ← NC(CH CH2)]+; these authors suggest, how-

X. Lefèvre et al. / Journal of Molecular Catalysis A: Chemical 335 (2011) 1–7
5
Nuc
H
NucH
C
C
N
Ni
C
N
H2C
H
N
Ni
N
Ni
H
Nuc
Nuc
C
N
Ni
C
C
N
Scheme 2.
reaction of 1 with morpholine or aniline/NEt3. We propose, tenta-
tively, that these broad signals are due to Ni-amine adducts that
form hydrogen bonding type interactions with the triflate anion. It
is important to recall, however, that none of these interactions take
place in the presence of RCN, the nitrile–Ni interaction being more
strongly favourable; as mentioned above, the acetonitrile moiety
in 1 is not displaced by anilines or phenols, even in the presence
of NEt3. The combination of these observations and considerations
lead us to propose that the role of complex 1 in the hydroamination
and hydroaryloxylation of acrylonitrile derivatives is akin to that of
a Lewis acid in the Michael-type additions to activated olefins, the
dative binding of the nitrile moiety to the cationic Ni center enhanc-
ing the electrophilic character of the olefinic moiety, as shown in
Scheme 2.
Fig. 1. ORTEP diagram for complex 2. Thermal ellipsoids are shown at the
50% probability level. The triflate anion was severely disordered (over
at least
4 positions) such that no satisfactory solution was found for
definitive placement of the atoms involved. Selected bond distances (Å)
and angles (◦): Ni–C1 = 1.888(3); Ni–P1 = 2.1752(5); Ni–N1 = 1.949(4); N1=
C11 = 1.260(7); C11–N2 = 1.417(7); C11–C12 = 1.429(9); C1–Ni–P1 = 81.49(5);
C1–Ni–N1 = 173.64(19); P1–Ni–N1 = 97.6(3); Ni–N1–C11 = 130.3(5); N1–C11–N2
= 121.7(6).
The ORTEP diagram and the structural parameters of 2 (Fig. 1)
illustrate clearly that the square planar geometry of the Ni center is
more or less unchanged on going from 1 to 2. The Ni–P and Ni–C dis-
tances are also quite comparable in these complexes, whereas the
2.4. Nucleophilic attack at the nitrile moiety
Examination of the reactivities of cyano olefins other than
acrylonitrile and its derivatives has shown that 1 can promote ami-
nation of the nitrile moiety in some cases. For instance, heating
morpholine and cinnamonitrile at 60 ◦C for 24 h in the presence of
1 (1 mol%) gave 34% yield of the amidine product arising from the
addition of the N–H bond to the nitrile moiety (Scheme 3). Similarly,
reacting 4-cyanostyrene with morpholine at 40 ◦C and in the pres-
ence of 1% 1 led to the corresponding amidine derivative in about
25% yield, as confirmed by GC/MS analyses ([M−1]+ = 216). Neither
of these reactions produced products arising from hydroamination
of the olefin moiety.
˚
Ni–N bond distance is somewhat longer in 2 (1.89 vs. 1.87–1.88 A)
[17g]. The amidine proton could not be located, but the lengthened
˚
N1–C19 bond (1.27 A in 2 vs. the C≡N distance of 1.14 in 1) as well
as the angles Ni–N1–C19 (135◦) and N1–C19–N2 (125◦) confirm
the sp2 hybridization of N1 and C19 resulting from the conversion
of acetonitrile into an amidine. It is worth noting that heating free
acetonitrile and morpholine over extended periods of time in the
absence of 1 did not result in formation of the corresponding ami-
dine, thus confirming the important activating role of the cationic
Ni center in 1.
The above observations implied that when the olefinic moiety
is insufficiently electrophilic, the nucleophile can react with the
nitrile moiety. To confirm this conclusion, we reacted the catalyst
precursor, 1, with morpholine in order to determine if coordination
to the cationic Ni center of acetonitrile, which lacks an olefinic moi-
ety, would activate the nitrile moiety toward nucleophilic attack.
Addition of one equivalent of morpholine to a 1.25 M C6D6 solu-
tion of 1 and overnight heating at 60 ◦C formed a new derivative, as
Finally, we have also examined briefly the reaction of allyl
cyanide with morpholine in the presence of 1% 1. This reaction
gave quantitative yield of the product arising from hydroamina-
tion of crotonitrile. To understand this result, we reacted 1 with
one equivalent of allyl cyanide alone, which led over one hour to
the formation of the crotonitrile derivative; this implies that the
N-bound allyl cyanide isomerizes to crotonitrile at room tempera-
ture (Scheme 4). On the other hand, a tertiary amine can induce this
isomerization even more rapidly (in about 10 min), while the iso-
merization is virtually instantaneous with a combination of base
and 1. Similar observations have been reported by Yi et al. on
the Ru(II)-catalyzed isomerization of allyl cyanide to crotonitrile,
followed by nucleophilic attack on coordinated crotonitrile [21].
Moreover, Ni(0)-catalyzed isomerization of allyl cyanide to cis- and
1
indicated by the disappearance of the original 31P{ H} NMR signal
for 1 (at 193 ppm) and emergence of a new singlet at 183 ppm; the
1H NMR spectrum showed the presence of a new N–H resonance at
ca. 5.4 ppm. Yellow crystals obtained from the NMR sample were
subjected to an X-ray diffraction study that allowed us to identify
the new product as the anticipated Ni–amidine product, 2 [20].
Scheme 3.

6
X. Lefèvre et al. / Journal of Molecular Catalysis A: Chemical 335 (2011) 1–7
Scheme 4.
trans-crotonitrile has been reported previously by Jones et al. [22].
These authors propose a mechanism involving ␲-coordination of
allyl cyanide, followed by activation of the allylic C–H bond to give a
Ni(II)(allyl)hydride intermediate that eliminates ␲-bound crotoni-
trile. Since a similar mechanism operating in our system would
require the involvement of highly energetic Ni(IV) intermediates,
we postulate that the isomerization of allyl cyanide in our system
goes through a concerted H-shift, or a proton transfer facilitated by
the base present in the reaction medium.
complex 1 (1 mol%). A typical run was conducted as follows: A
NMR tube was charged with 0.5 mL of C6D6, a 200 ␮L aliquot of
a 25 mM C6D6 solution of 1 (0.005 mmol of pre-catalyst), 0.5 mmol
each of the substrate (for example, 32 ␮L of acrylonitrile and 44 ␮L
of morpholine), and 0.5 mmol of triethylamine (when required).
The tube was then capped and, when indicated, heated at the
desired temperature in an oil bath for the designated time. The
progress of the reaction was followed at regular intervals by NMR
until the end of the reaction, as signalled by the disappearance of
the vinylic protons of the olefinic substrate. Control reactions were
performed using the same protocol except that no catalyst and/or
triethylamine were introduced. The final reaction mixture was then
subjected to a flash filtration through silica gel (Et2O used as eluent)
in order to remove residual nickel particles, and the filtrate was ana-
lyzed by GC/MS to identify products and determine the conversion.
These analyses usually displayed two peaks, one corresponding to
the unreacted nucleophile, and the other to the product. (The MS
was kept closed during the first 2 min of the analysis in order to
protect the filament and avoid overwhelming the detector with
signals due to solvent peak. Thus, all volatile components of the
mixtures, including acrylonitrile, were not analyzed.) The MS frag-
mentation patterns allowed a reliable identification of the species
while integration of these two peaks gave the reported yields; the
conversions were verified against integration of the vinylic peaks
in the 1H NMR spectra.
3. Conclusion
The electrophilic nature of the nickel center in the cationic com-
plex 1 and the lability of the acetonitrile moiety in this complex
toward cyano olefins allow the promotion of nucleophilic additions
by amines and aromatic alcohols on the C C moiety of acryloni-
trile, methacrylonitrile, and crotonitrile. The requirement for the
coordination of a nitrile moiety to nickel limits the scope of these
reactions to olefinic substrates bearing a CN functionality. Future
studies will aim to develop precursors that can bind carbonyl and
nitro functionalities with a view to extending the hydroamination
and hydroaryloxylation reactions to a wider range of substrates. On
the other hand, in the case of 4-cyanostyrene and cinnamonitrile,
the addition takes place on the nitrile moiety to give the corre-
sponding amidines; a similar reaction took place with Ni-bound
acetonitrile to give the new amidine adduct 2. These observations
open up new avenues for exploration.
4.3. Crystal structure determination
The crystallographic data for complex 2 (Table 3) were col-
lected on a Bruker X8 Proteum system with Microstar-H generator,
Helios Optic, Kappa goniometer and Platinum-135 detector. Cell
refinement and data reduction were done using SAINT [23]. An
4. Experimental
4.1. General comments
Unless otherwise noted, reagents were used as received from
Sigma–Aldrich and were handled in ambient atmosphere. Com-
plex [{␬P,␬C,␬P-2,6-(i-Pr2PO)2C6H3}-Ni(NCMe)][OSO2CF3] (1) was
obtained following published preparation 1 was obtained by fol-
lowing a previously reported procedure [17d]. All NMR spectra
wererecorded at ambienttemperature onBrukerAV400 andAV300
instruments. The 1H and 13C NMR spectra were referenced to
solvent resonances, as follows: 7.15 and 128.06 ppm for C6D5H
Table 3
Details of X-ray diffraction studies on complex 2.
Formula: C25H43N2O6F3P2SNi
Mw (g/mol): 677.35
F(0 0 0): 2088
Crystal color and form: yellow
blocks
Crystal size (mm):
0.26 × 0.16 × 0.16
T (K): 150
Wavelength: 1.54178
Crystal system: hexagonal
Unit cell
Space group: P6522
V (Å3): 4768.90 (8)
Z: 3
dcalcd. (g/cm3): 1.413
ꢀ range (◦): 4.31–67.82
1
and C6D6, respectively. The 31P{ H} NMR spectra were referenced
to an external 85% H3PO4 sample (0 ppm). The GC/MS analyses
were done using a Agilent Technologies 6890 Network GC system
equipped with a HP-5MS capillary column and a 5973 MS selective
detector.
Completeness: 1.00
Collected reflections Rꢁ: 75,234; 0.0096
Unique reflections Rint: 2890; 0.0450
a (Å): 12.341
ꢂ (mm−1): 2.948
b (Å): 12.341
Abs. correction: multi-scan
R1(F); wR(F2) [I > 2ꢁ(I)]: 0.0366; 0.1056
R1(F); wR(F2) (all data): 0.367; 0.1056
GoF(F2): 1.096
4.2. Catalytic studies
c (Å): 36.1550 (4)
˛ (◦): 90
ˇ (◦): 90
All catalytic experiments were carried out in NMR tubes con-
taining C6D6 solutions of the substrates (0.5 mmol each) and
ꢃ (◦): 120
Residual electron density: 0.47

X. Lefèvre et al. / Journal of Molecular Catalysis A: Chemical 335 (2011) 1–7
7
empirical absorption correction, based on the multiple measure-
ments of equivalent reflections, was applied using the program
SADABS [24]. The space group was confirmed by XPREP routine [25]
in the program SHELXTL [26]. The structures were solved by direct-
methods and refined by full-matrix least squares and difference
Fourier techniques with SHELX-97 [27]. All non-hydrogen atoms
were refined with anisotropic displacement parameters. Hydrogen
atoms were set in calculated positions and refined as riding atoms
with a common thermal parameter.
[10] M. Beller, O.R. Thiel, H. Trauthwein, Synlett (1999) 243.
[11] H. Hart, J.R. Kosak, J. Org. Chem. 27 (1962) 116.
[12] L.L. Anderson, J. Arnold, R.G. Bergman, J. Am. Chem. Soc. 127 (2005)
14542.
[13] (a) J.L. Jensen, H. Hashtroudi, J. Org. Chem. 41 (1976) 3299;
(b) J.L. Duffy, J.A. Kurth, M.J. Kurth, Tetrahedron Lett. 34 (1993) 1259;
(c) E. Dumez, J. Rodriguez, J.-P. Dulcère, Chem. Commun. (1997) 1831.
[14] (a) D.S. Noyce, K.E. DeBruin, J. Am. Chem. Soc. 90 (1968) 372;
(b) L.R. Fedor, N.C. De, S.K. Gurware, J. Am. Chem. Soc. 95 (1973) 2905;
(c) J.L. Jensen, D.J. Carre, J. Org. Chem. 39 (1974) 2103.
[15] (a) I.C. Stewart, R.G. Bergman, F.D. Toste, J. Am. Chem. Soc. 125 (2004) 8696;
(b) P.B. Kisanga, P. Ilankumaran, B.M. Fetterly, J.G. Verkade, J. Org. Chem. 67
(2002) 3555.
[16] (a) For a few representative reports see: T. Hosokawa, T. Shinohara, Y. Ooka,
S.-I. Murahashi, Chem. Lett. (1989) 2001;
Acknowledgments
(b) S. Ganguly, D.M. Roundhill, Organometallics 12 (1993) 4825;
(c) A.V. Nikitin, S.N. Kholuiskaya, V.L. Rubailo, J. Chem. Biochem. Kinet. 3 (1997)
37;
(d) K.J. Miller, T.T. Kitagawa, M.M. Abu-Omar, Organometallics 20 (2001) 4403;
(e) H.L. van Lingen, W. Zhuang, T. Hansen, F.P.J.T. Rutjes, K.A. Jørgensen, Org.
Biomol. Chem. 1 (2003) 1953;
The authors gratefully acknowledge financial support of these
studies by NSERC of Canada (Discovery and Instrument grants to
DZ) and Université de Montréal (Bourse d’Excellence to X.L.). Mr.
Denis Spasyuk is thanked for his technical assistance with the X-ray
and GC/MS analyses, as well as many helpful discussions during the
preparation of this manuscript. Dr. M. Simard is thanked for help
with the resolution of the solid state structure for complex 2.
(f) M.V. Farnworth, M.J. Cross, J. Louie, Tetrahedron Lett. 45 (2004) 7441.
[17] (a) L.F. Groux, F. Bélanger-Gariépy, D. Zargarian, Can. J. Chem. 83 (2005) 634;
(b) A. Castonguay, F. Charbonneau, A.L. Beauchamp, D. Zargarian, Acta Crystal-
logr., E 61 (2005) m2240–m2241;
(c) A. Castonguay, C. Sui-Seng, D. Zargarian, A.L. Beauchamp, Organometallics
25 (2006) 602;
(d) C. Sui-Seng, A. Castonguay, Y. Chen, D. Gareau, L.F. Groux, D. Zargarian, Top.
Catal. 37 (2006) 81;
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molcata.2010.11.010.
(e) A. Castonguay, A.L. Beauchamp, D. Zargarian, Acta Crystallogr., E 63 (2007)
m196;
(f) V. Pandarus, D. Zargarian, Chem. Commun. (2007) 978;
(g) V. Pandarus, D. Zargarian, Organometallics 26 (2007) 4321;
(h) A. Castonguay, A.L. Beauchamp, D. Zargarian, Organometallics 27 (2008)
5723;
References
(i) A. Castonguay, D.M. Spasyuk, N. Madern, A.L. Beauchamp, D. Zargarian,
Organometallics 28 (2009) 2134;
[1] (a) For a few reviews see: J.J. Brunet, D. Neibecke, in: A. Togni, H. Grützmacher
(Eds.), Catalytic Heterofunctionalization, VCH, Weinheim, 2001, pp. 91–141;
(b) T. Müller, M. Beller, Chem. Rev. 98 (1998) 675;
(j) A. Castonguay, A.L. Beauchamp, D. Zargarian, Inorg. Chem. 48 (2009) 3177;
(k) D.M. Spasyuk, D. Zargarian, A. van der Est, Organometallics 28 (2009) 6531;
(l) D.M. Spasyuk, D. Zargarian, Inorg. Chem. 49 (2010) 6531.
[18] Indeed, a recent literature search indicated that complex 1 and a related dimeric
complex featuring a POCN-type pincer ligand 171 are the only reported exam-
ples of Ni complexes promoting addition of alcohols to olefins.
[19] These spectra also show minor signals associated with side-products arising
from oxidation of the ligand at the phosphinite moieties, but none of these
species has been isolated or identified.
[20] For a review on the addition of amines to coordinated nitriles see: V.Y.
Kukushkin, A.J.L. Pombeiro, Chem. Rev. 102 (2002) 1771.
[21] C.S. Yi, S.Y. Yun, Z. He, Organometallics 22 (2003) 3031.
[22] (a) N.M. Brunkan, W.D. Jones, J. Organomet. Chem. 683 (2003) 77;
(b) N.M. Brunkan, D.M. Brestensky, W.D. Jones, J. Am. Chem. Soc. 126 (2004)
3627.
[23] SAINT, Release 6.06; Integration Software for Single Crystal Data, Bruker AXS
Inc., Madison, WI, USA, 1999.
[24] G.M. Sheldrick, SADABS, Bruker Area Detector Absorption Corrections, Bruker
AXS Inc., Madison, WI, USA, 1999.
[25] XPREP, Release 5.10; X-ray data Preparation and Reciprocal space Exploration
Program, Bruker AXS Inc., Madison, WI, USA, 1997.
[26] SHELXTL, Release 5.10; The Complete Software Package for Single Crystal Struc-
ture Determination, Bruker AXS Inc., Madison, Wisconsin, USA, 1997.
[27] (a) G.M. Sheldrick, SHELXS97, Program for the Solution of Crystal Structures,
University of Gottingen, Germany, 1997;
(c) M. Beller, J. Seayad, A. Tillack, H. Jiao, Angew. Chem. Int. Ed. Engl. 43 (2004)
3368;
(d) M. Nobis, B. Drie␤en-Hölscher, Angew. Chem. Int. Ed. Engl. 40 (2001) 3983.
[2] S. Tian, V.M. Arredondo, C.L. Stern, T.J. Marks, Organometallics 18 (1999) 2568.
[3] (a) L. Ackermann, R.G. Bergman, Org. Lett. 4 (9) (2002) 1475;
(b) L. Ackermann, L.T. Kaspar, C. Gschrei, J. Org. Lett. 6 (15) (2004) 2515;
(c) J.A. Bexrud, J.D. Beard, D.C. Leitch, L.L. Schafer, Org. Lett. 7 (10) (2005) 1959;
(d) R.K. Thomson, J.A. Bexrud, L.L. Schafer, Organometallics 25 (2006) 4069;
(e) M.C. Wood, D.C. Leitch, C.S. Yeung, J.A. Kozak, L.L. Schafer, Angew. Chem.
Int. Ed. Engl. 46 (2007) 354.
[4] (a) J.J. Brunet, G. Commenges, D. Neibecker, K.J. Philippot, Organomet. Chem.
469 (1994) 221;
(b) M. Beller, H. Trauthwein, M. Eichberger, C. Breindl, T. Müller, Eur. J. Inorg.
Chem. (1999) 1121.
[5] (a) A.L. Casalnuovo, J.C. Calabrese, D. Milstein, J. Am. Chem. Soc. 110 (1988)
6738;
(b) R. Dorta, P. Egli, F. Zürcher, A. Togni, J. Am. Chem. Soc. 119 (1997) 10857.
[6] (a) J. Pawlas, Y. Nakao, M. Kawatsura, J.F. Hartwig, J. Am. Chem. Soc. 124 (2002)
3669;
(b) L. Fadini, A. Togni, Chem. Commun. (2003) 30;
(c) L. Fadini, A. Togni, Tetrahedron: Asymmetry 19 (2008) 2555.
[7] (a) A.L. Seligson, W.C. Trogler, Organometallics 12 (1993) 744;
(b) M. Kawatsura, J.F. Hartwig, J. Am. Chem. Soc. 122 (2000) 9546;
(c) O. Löber, M. Kawatsura, J.F. Hartwig, J. Am. Chem. Soc. 123 (2001) 4366;
(d) U. Nettekoven, J.F. Hartwig, J. Am. Chem. Soc. 124 (2002) 1166;
(e) M. Kawatsura, J.F. Hartwig, Organometallics 20 (2001) 1960.
[8] D. Karshtedt, A.T. Bell, T.D. Tilley, J. Am. Chem. Soc. 127 (2005) 12640.
[9] (a) C. Munro-Leighton, E.D. Blue, T.B. Gunnoe, J. Am. Chem. Soc. 128 (2006)
1446;
(b) G.M. Sheldrick, SHELXL97, Program for the Refinement of Crystal Structures,
University of Gottingen, Germany, 1997.
(b) C. Munro-Leighton, S.A. Delp, E.D. Blue, T.B. Gunnoe, Organometallics 26
(2007) 1483.