X-ray structure and fluxional behaviour of N,N′-di-p-fluorophenyltriazenido complex of nickel(II)

Article abstract of DOI:10.1016/S0022-328X(99)00684-1

The Ni(II) triazenide complex, trans-(o-Tol)Ni(PEt₃)₂N₃Ar₂ (Ar=p-FC₆H₄) (1), was synthesized by the reaction of trans-(o-Tol)Ni(PEt₃)₂Br with Ar₂N₃Na. The crystal structure as well as the ¹H-, ¹⁹F- and ³¹P-NMR spectra in toluene-d₈ at different temperatures are reported. It was found that the triazenido group acts as a monodentate ligand. Complex 1 in solid state has a S-cis-structure with Ni-N(3) σ-bond and exists as a 3:1 mixture of two isomers with cisoid or transoid orientation of the Me group of the o-tolyl ligand relative to the N(1)=N(2) bond. It was established that the N,N′-migration of (o-Tol)Ni(PEt₃)₂ group occurs in a solution of 1. The migration rate increases with the temperature increase in the range from -104 to -45°C. However, a further increase in the temperature slows down the process.

Full text of DOI:10.1016/S0022-328X(99)00684-1

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 597 (2000) 164–174
X-ray structure and fluxional behaviour of
N,N%-di-p-fluorophenyltriazenido complex of nickel(II)
A.S. Peregudov *, D.N. Kravtsov, G.I. Drogunova, Z.A. Starikova, A.I. Yanovsky
Institute of Organo-Element Compounds, RAS, 28 Va6ilo6 Street, Moscow, Russia
Received 20 April 1999; received in revised form 2 November 1999
Dedicated in honour of Professor Stanis*law Pasynkiewicz on the occasion of his 70th birthday.
Abstract
The Ni(II) triazenide complex, trans-(o-Tol)Ni(PEt3)2N3Ar2 (Ar=p-FC6H4) (1), was synthesized by the reaction of trans-(o-
Tol)Ni(PEt3)2Br with Ar2N3Na. The crystal structure as well as the 1H-, 19F- and 31P-NMR spectra in toluene-d8 at different
temperatures are reported. It was found that the triazenido group acts as a monodentate ligand. Complex 1 in solid state has a
S-cis-structure with NiꢀN(3) s-bond and exists as a 3:1 mixture of two isomers with cisoid or transoid orientation of the Me group
of the o-tolyl ligand relative to the N(1)ꢁN(2) bond. It was established that the N,N%-migration of (o-Tol)Ni(PEt3)2 group occurs
in a solution of 1. The migration rate increases with the temperature increase in the range from −104 to −45°C. However, a
further increase in the temperature slows down the process. © 2000 Elsevier Science S.A. All rights reserved.
Keywords: Nickel complexes; Triazenido complexes; X-ray structure; Molecular dynamics
Processes of this kind have been observed in Pd and
Pt complexes [13,16], as well as in related Hg and Au
1. Introduction
triazenides [17–21]. In these cases, relatively easily
available compounds with R=Ar and univalent metal-
containing groups MLn were most commonly used as
The variety of possible modes of metal coordination
with triazenido ligands in triazene complexes was the
focus of several studies of the structure of these com-
convenient model systems. For some of these complexes
pounds. However, in the available literature data, the
the X-ray data suggest the monodentate mode of coor-
triazenido group most frequently acts as either a bridg-
dination (III) [10,12,13]. However, up to now there
ing ligand between two metal centres (I) or as a biden-
have been practically no data on the structure of diaryl-
tate group (II) [1–9]. The monodentate form (III) has
triazenido nickel complexes. In fact, the structures of
until now eluded structural characterization [10–15]:
only two triazenido Ni complexes, Ni2L4 (L=PhN3Ph)
[5] and Ni2(OH)2(L%)2(L%%) (L%=PhꢀN3HꢀC6H4ꢀ
N3HꢀPh, L%%=PhꢀN3ꢀC6H4N3ꢀPh), have been studied
in solid state [8]. The structures of these bimetallic
complexes in solution have not been discussed.
At the same time, the structure (III) may be of
particular interest because in such tautomeric systems
the N(3)ꢀN(1) s,s-migration of MLn group is possible:
In a previous paper [22], we reported the study of the
structure of N,N%-di-p-fluorophenyltriazenido complex
of platinum(II), trans-(o-Tol)Pt(PEt3)2N3(C6H4F-p)2
(2), by dynamic NMR. It has been found that the
* Corresponding author. Fax: +7-095-1355085.
E-mail address: asp@ineos.ac.ru (A.S. Peregudov)
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(99)00684-1

A.S. Peregudo6 et al. / Journal of Organometallic Chemistry 597 (2000) 164–174
165
above-mentioned migration of the trans-(o-Tol)-
Pt(PEt3)2 group indeed occurs on the NMR time scale
in toluene-d8 solution:
methanol) and after 10 min of stirring, the solution of
1.1 mmol of 3 in the same solvent was added to the
reaction mixture. The reaction mixture was stirred for 3
h, the solvent was evaporated, the solid residue washed
with water, dried in vacuum and recrystallized from
methanol. Yellow solids of 1 (yield 40%) were obtained,
m.p. 135–138°C. Anal. Calc. for C31H45N3NiP2F2 (1):
% C, 60.14 (60.21); % H, 7.35 (7.33); % N, 6.82 (6.80).
2.2. X-ray crystal structure determination
Crystals of 1 are monoclinic; at −78°C, a=
,
10.499(4), b=35.24(2), c=9.143(3) A, i=108.76(3)°,
3
,
V=3203(2) A , space group P21/c, Z=4, Dcalc. =1.282
g cm−3, F(000)=1272.
The data for a greenish–blue crystal having approxi-
mate dimensions of 0.2×0.2×0.5 mm were collected
at −78°C on a Syntex P21 diffractometer using the
q/2q scan mode (q range 2.0–25.0°). A total of 4976
reflections were averaged to produce 4652 unique data.
The structure was solved by the direct method and
refined by the least-squares technique in the anisotropic
approximation for all non-hydrogen atoms, with the
exception of the carbon atoms of the disordered Ph ring
of the o-tolyl ligand (see Section 3). The H atoms apart
from those of the Me groups (at the C(4) and C(12)
atoms) of two Et substituents (which were placed geo-
metrically and included in the refinement in the riding
model approximation) were located in the difference
Fourier synthesis and refined isotropically; the H atoms
of the disordered Ph ring were not taken into account.
The refinement converged to wR2=0.2131 (on F2hkl for
all 4579 reflections, 498 variables) with R1=0.080 (on
Fhkl for 2190 reflections with I\2|(I)). The highest
peak in the final difference Fourier map had a height of
In the present paper, the related complex of Ni(II)
was synthesized and its crystal structure, as well as the
1H-, 19F- and 31P-NMR spectra at various tempera-
tures, were studied.
2. Experimental
2.1. Synthesis of trans-(o-Tol)Ni(PEt3)2N3(C6H4F-p)2
(1)
The
starting
N,N%-(di-p-fluorophenyl)triazene
(DFPT) and trans-(o-Tol)Ni(PEt3)2Br (3) were pre-
pared according to published procedures [19,23]. A
solution of DFPT in methanol was added to a
methanol solution of MeONa (1 mmol Na in 10 ml
0.66 e A−3. The maximum negative peak in the final
,
Table 1
−3
,
difference electron density synthesis was −0.43 e A
.
,
Selected bond lengths (A) and bond angles (°) for 1
All calculations were performed on IBM PC with the
help of the SHELXTL Plus 5 program [24]. Bond lengths
and angles are listed in Table 1.
Bond lengths
Ni(1)ꢀC(25)
Ni(1)ꢀN(3)
Ni(1)ꢀP(1)
Ni(1)ꢀP(2)
P(1)ꢀC(1)
P(1)ꢀC(5)
P(1)ꢀC(3)
P(2)ꢀC(9)
P(2)ꢀC(7)
P(2)ꢀC(11)
N(1)ꢀN(2)
1.924(6)
1.943(4)
2.220(2)
2.221(2)
1.829(7)
1.830(7)
1.847(7)
1.836(7)
1.843(7)
1.851(7)
1.298(7)
N(1)ꢀC(19)
N(2)ꢀN(3)
N(3)ꢀC(13)
F(1)ꢀC(22)
F(2)ꢀC(16)
C(1)ꢀC(2)
C(3)ꢀC(4)
C(5)ꢀC(6)
C(7)ꢀC(8)
C(9)ꢀC(10)
C(11)ꢀC(12)
1.403(8)
1.326(7)
1.421(8)
1.373(9)
1.377(9)
1.560(11)
1.542(11)
1.533(11)
1.529(11)
1.532(11)
1.486(11)
2.3. The NMR spectra
1
The H-, 19F- and 31P-NMR spectra were recorded
on a Bruker AMX-400 instrument. The operating fre-
quencies were 400.13, 376.46 and 161.98 MHz, respec-
tively. A 0.1 M solution in toluene-d8 has been used.
The fluorine chemical shifts (FCS) were measured rela-
tive to internal PhF; the plus sign corresponds to an
upfield shift of the fluorine signal. The proton chemical
shifts were determined relative to the residual signal of
Bond angles
C(25)ꢀNi(1)ꢀN(3) 178.5(3)
N(1)ꢀN(2)ꢀN(3)
N(2)ꢀN(3)ꢀC(13)
N(2)ꢀN(3)ꢀNi(1)
C(13)ꢀN(3)ꢀNi(1)
C(2)ꢀC(1)ꢀP(1)
C(4)ꢀC(3)ꢀP(1)
C(6)ꢀC(5)ꢀP(1)
110.1(4)
112.8(4)
120.5(4)
126.5(4)
115.6(5)
115.2(5)
116.3(5)
C(25)ꢀNi(1)ꢀP(1)
N(3)ꢀNi(1)ꢀP(1)
C(25)ꢀNi(1)ꢀP(2)
N(3)ꢀNi(1)ꢀP(2)
P(1)ꢀNi(1)ꢀP(2)
C(1)ꢀP(1)ꢀC(5)
86.9(2)
93.1(2)
89.7(2)
90.4(2)
174.80(7)
104.9(3)
1
the methyl group of toluene (2.30 ppm). An H ther-
mometer based on the temperature-dependent shift dif-
ference between CH- and OH-protons has been used
for temperature control. The method of measurement
involved replacing the sample tube with a tube contain-

166
A.S. Peregudo6 et al. / Journal of Organometallic Chemistry 597 (2000) 164–174
Table 2
The 19F-NMR data for the solution of trans-(o-Tol)Ni(PEt3)2N3(C6H4F-4)2 (1) in toluene-d8
Temperature (K)
FCS (ppm) a
DFCS (Hz) b
w1/2 (1) (Hz)
w1/2(2) (Hz)
w1/2(0) (Hz) c
~A×103 (s) d
160
165
171
181
186
191
196
8.412; 10.490
8.412; 10.483
8.429; 10.499
8.450; 10.499
8.457; 10.503
8.481; 10.491
8.530; 10.451
781
779
778
770
769
756
722
56
51
34
47
58
90
155
50
47
33
45
56
29
26
22
19
17
17
16
12
5.7
3.9
2.2
1.05
90
162
r=5.63
302
200
8.637; 10.330
628
282
15
0.69
r=2.52
r=1.93 15
226
r=2.39
32
207
213
8.865; 10.160
9.021; 9.982
478
368
227
39
15
14
223
9.101; 9.924
309
r=6.8
18
16
14
10
9
12
19
41
230
243
270
303
313
323
333
343
353
9.163; 9.901
9.197; 9.879
9.265; 9.850
9.350; 9.811
9.371; 9.801
9.391; 9.788
9.405; 9.772
9.407; 9.740
9.423; 9.700
277
256
220
173
162
149
138
125
104
18
16
14
10
11
13
25
53
r2=2.1
14
12
11
5
5
5
5
5
r1=4.0
a The fluorine chemical shift relative to internal PhF, positive sign corresponds to upfield shift.
b The difference between FCS of two signals.
c The line width: w1/2(1), left line; w1/2(2), right line; w1/2(0), PhF; r, ratio of main maximum to central minimum.
d The mean lifetime of fluorine in each site.
ing 4% methanol in CD3OD for low temperatures or
80% 1,2-ethanediol in DMSO-d6 for high temperatures.
tion of the ~A values in the temperature range from
−92 to −55°C. The errors in the thermodynamic
parameters were determined by the usual procedure [25]
taking into account the temperature range (29°C for
2.4. Kinetic studies by dynamic NMR method
1
19F, 37°C for 31P and 47°C for H) and are shown in
Table 4.
The mean lifetimes ~A of the indicator nuclei in each
site were calculated by using computer simulation for
two-site exchange with unequal population [22]. Due to
the fact that some temperature dependence of the dif-
ferences Dw0 between the frequencies of two sites in the
absence of exchange was observed (see Tables 2 and 3),
Dw0 values near and above collapse were interpolated
from low temperatures. In the case of 19F-NMR spec-
tra, the linewidth of internal PhF was chosen as the
linewidth w01/2 in the absence of exchange.
3. Results and discussion
3.1. X-ray data
The structure of molecule 1 is shown in Fig. 1. The
triazenide group in this complex acts as a monodentate
ligand coordinating the Ni atom by only one of its
nitrogen atoms. In fact, the Ni(1)···N(1) distance is as
long as 2.797 A and thus excludes the possibility of
bonding interaction. The Ni(1) atom has a square-pla-
The thermodynamic activation parameters were cal-
culated according to Arrhenius and Eyring equations.
For the the reasons discussed below, it is possible to
estimate correctly the ~A values from the 19F-NMR
data only in the temperature range from −102 to
−73°C. In fact only six points have been used for the
,
nar coordination geometry; the NiꢀP and NiꢀC
,
,
(Ni(1)ꢀP(1) 2.220(2) A, Ni(1)ꢀP(2) 2.221(2) A,
,
Ni(1)ꢀC(25) 1.924(6) A) bond lengths being close to
1
,
calculation of the ~A values. For H-NMR spectra, it is
their respective standard values (2.214 A for NiꢀPR3,
,
possible to estimate the ~A values (nine points) in the
temperature range from −92 to −45°C. For 31P-
NMR spectra, eight points have been used for calcula-
1.917 A for NiꢀC(Ar) according to Ref. [26]). The mean
planes of the o-tolyl and ArꢀN3ꢀAr ligands are approx-
imately normal to the Ni atom coordination plane

A.S. Peregudo6 et al. / Journal of Organometallic Chemistry 597 (2000) 164–174
167
Ni(1)P(1)P(2)N(3)C(25): the corresponding dihedral an-
gles are equal to 91.3 and 93.7°, respectively. The Et
groups of the PEt3 ligands are in a staggered conforma-
tion with respect to each other.
The triazenide p-FC6H4N3C6H4F-p ligand shows
only small deviations from planarity, the largest dis-
placement of the C(23) atom from the mean plane of
,
the whole ligand being equal to 0.27 A; the Ph ring
Table 3
The 1H- and 31P-NMR data for the solution of trans-(o-Tol)Ni(PEt3)2N3(C6H4F-4)2 (1) in toluene-d8
Temperature (K)
l1H (ppm) a
~A×103 (s) b
l
31P (ppm) c
~A×103 (s) d
160
165
171
176
181
3.004
(23)
3.018
(29)
3.012
(28)
3.026
(29)
3.037
(60)
r1=3.5
3.047
3.590
(18)
3.580
(26)
3.573
(23)
3.564
(22)
3.555
(25)
r2=20
3.543
(37)
8.86
(22)
8.84
(20)
8.81
(28)
8.74
(30)
8.73
9.70
(21)
9.64
(18)
9.58
(20)
9.50
(16)
4.1
2.8
9.43
(22)
r2=2.1
6.1
2.8
r1=13.4
9.34
(36)
186
r1=1.81
3.518
(54)
3.473
(73)
3.415
(68)
3.398
(33)
3.389
(22)
3.382
(17)
3.369
(11)
3.349
(9)
r2=12
191
196
200
207
213
218
228
243
292
1.65
1.0
9.22
(44)
9.09
(36)
9.02
(26)
8.95
(14)
8.89
(11)
8.82
(8.6)
8.69
(6.6)
8.48
(5.3)
7.99
(5.2)
1.8
1.4
0.57
0.315
0.21
0.175
0.115
0.77
0.45
0.34
0.295
3.314
(7)
a In parentheses, the line width; r1, the ratio of the right maximum to the central minimum; r2, the ratio of the left maximum to the central
minimum.
b The mean lifetime of o-methyl protons in each site.
c From external H3PO4 at r.t.
d The mean lifetime of 31P in each site.
Table 4
Thermodynamic parameters of dynamic processes for compounds 1 (present work) and 2 (from Ref. [22])
Compound
NMR
Ea
DHc
DSc
(e.u.)
Temperature of
coalescence (°C)
Temperature range of estimation
of ~A (°C)
(kcal mol−1
)
(kcal mol−1
)
a
1
19F
1H
6.890.7
6.790.3
6.390.4
6.490.7
6.590.3
6.090.4
−1293
−1191
−1392
−102 to −73
−92 to −45
−92 to −55
−84
−88
31P
2
19F
1H
16.490.4
15.790.4
492
−194
63
25
10
25 to 80
7 to 47
15.290.8
14.690.8
b
b
b
31P
a Coalescence does not take place.
b Thermodynamic parameters in Ref. [22] have not been estimated due to small Dl°.

168
A.S. Peregudo6 et al. / Journal of Organometallic Chemistry 597 (2000) 164–174
the positions of the corresponding atoms of the tolyl
ring within each of the components of the disorder.
That is why it was possible to resolve the disorder for
all phenylene ring carbon atoms with the exception of
the C(25) ipso-atom, the disordered positions of which
are still too close to each other and make the resolution
absolutely impossible. Moreover, the careful refinement
of the structure indicated that two components are not
equally represented in the crystal, one of the positions
(C(25)ꢀC(31)) having about three times larger occu-
pancy than its counterpart (C(1%)ꢀC(6%)). Thus, the ob-
served disorder allows us to treat the structure as a
superposition of complexes with cisoid and transoid
orientations of the 2-MeC6H4 group with respect to the
N(2)N(1)Ph fragment, the former compound appearing
to be three times more abundant than the latter.
Previously, the structures of only two complexes of
Ni(II) with the PhꢀN3ꢀPh fragments have been re-
ported. One of them has a lantern-shaped Ni2(L)4 (L=
PhN3Ph) (4) molecule with the square-planar
coordination of both Ni atoms and four bridging lig-
ands each of them coordinating different Ni atoms by
the N(1) and N(3) atoms [5]. The average NiꢀN and
Fig. 1. Molecular structure of Ni(PEt3)2(C6H4Me-2)(4-FC6H4N3-
C6H4F-4) (1).
,
NꢀN bond lengths (1.92(1) and 1.31(1) A, respectively)
are close to those found in the present structure. It is
noteworthy, in particular, that in spite of the different
functions of the ligands in the crystals of 1 and 4 and
the involvement of two nitrogen atoms of each of the
triazenide groups in metal coordination in the case of
complex 4, the N3 systems in both complexes show
almost the same degree of delocalization and are char-
acterized by very similar geometric parameters.
Another previously studied complex Ni2(OH)2-
(L%)2(L%%) (5) [8] shows two types of three-nitrogen
ligands (L%=PhꢀN3HꢀC6H4ꢀN3HꢀPh, L%%=PhꢀN3ꢀ
C6H4ꢀN3ꢀPh) coordinated in two different ways. Each
of the ꢀN1ꢁN2ꢀNHꢀ fragments in each of the L% ligands
coordinates the Ni atom by one nitrogen atom (in the
same way as in 1), whereas each of the deprotonated
ꢀN1N2N3ꢀ fragments in L%% is chelating the Ni atom via
its two nitrogens (N1 and N3) with the formation of the
four-membered NiN3 cycle:
planes C(13)ꢀC(18) and C(19)ꢀC(24) are rotated by
8.7(7) and 16.1(7)° away from the plane of the central
N3 fragment. The N(1)ꢀN(2) and N(2)ꢀN(3) bond
,
lengths (1.298(7) and 1.326(7) A, respectively) are equal
,
within 4|, their average value (1.312 A) being consider-
ably larger than the standard value typical of the NꢁN
,
bonds in azobenzene derivatives (1.255 A [26]) and
rather close to the delocalized NꢀN bond lengths (1.304
,
A) normally observed in the aromatic pyridazine sys-
tems [26]. Thus, the bonds in the NꢀNꢁN fragment are
delocalized and at least in part, involved in the conjuga-
tion with the Ph cycles. This is indicated by both the
coplanarity of the Ph rings and the N3 fragment and
some shortening of the NꢀC(Ph) bonds, their average
,
,
value 1.412 A being about 0.02 A shorter than the
,
standard C(Ar)ꢀN(ꢁN) bond (1.431 A as quoted in Ref.
[26]). The difference between the individual NꢀN and
CꢀN bonds (N(1)ꢀN(2) versus N(2)ꢀN(3) and
,
,
N(1)ꢀC(19) 1.403(8) A versus N(3)ꢀC(13) 1.421 A) may
be attributed to the influence of the metal atom which
is coordinated only by the N(3) atom, whereas the N(1)
atom is not involved in coordination (vide supra).
The detailed analysis of the difference Fourier maps
suggested the disorder of the tolyl ligand, which was
subsequently completely confirmed by the refinement of
the disordered model. It turned out that the o-tolyl
substituent in fact fills in two positions in crystal, which
differ en grosse by a 180° rotation about the NiꢀC
bond. The somewhat different steric environments of
the methyl group for each of the two orientations of the
substituent obviously give rise to slight differences in
The similar pattern of NꢀN bond-length distribution
is observed for all three ligands in 5 irrespective of the

A.S. Peregudo6 et al. / Journal of Organometallic Chemistry 597 (2000) 164–174
169
Fig. 2. The 19F-NMR spectra of 1 at different temperatures (solution in toluene-d8).

170
A.S. Peregudo6 et al. / Journal of Organometallic Chemistry 597 (2000) 164–174
Fig. 3. The 1H-NMR spectra of 1 at different temperatures (solution in toluene-d8).

A.S. Peregudo6 et al. / Journal of Organometallic Chemistry 597 (2000) 164–174
171
differences in their coordination modes. Specifically, one
coordination polyhedron of the nickel atom, manifested
also in the differences between the NiꢀN bond lengths.
half of each ligand (H1) has equal N1ꢀN2 and N2ꢀN3
,
bonds (the average values are 1.325(7) and 1.313(7) A,
respectively), whereas the other half (H2) shows signifi-
cant differences in these bonds (the corresponding aver-
3.2. NMR data
age values are 1.342(7) and 1.271(7) A). These two types
The 19F-NMR data for compound 1 at different
temperatures are given in Table 2. At −102°C there are
two somewhat broadened fluorine signals with the differ-
ence in FCS (DFCS) equals to 2.07 ppm. These signals
broaden upon increase of the temperature and almost
collapse at −65°C (Fig. 2). Simultaneously some lower-
ing of the DFCS value is observed. As the temperature
is further increased, the fluorine signals contrary to the
expectation do not collapse but at room temperature
again become sharp and DFCS value equals to 0.47 ppm.
On increasing the temperature above room temperature
these signals broaden again and the DFCS value equals
0.32 ppm at 80°C.
,
of NꢀN bond delocalization may be accounted for by the
difference in the strength of coordination with the nickel
atom. Indeed, all NiꢀN bonds involving the N atoms of
the H1 half are noticeably shorter (2.050(6), 2.064(5),
,
2.084(6) and 2.123(6) A) than the corresponding bonds
with the N atoms of the H2 moiety (2.062(5), 2.091(6),
,
2.106(6) and 2.210(5) A).
One may conclude that the distribution of the NꢀN
bond lengths in the Ni-coordinated N3ꢀligands depends
neither on the form of the ligand (protonated or depro-
tonated) nor on the type of coordination and is rather
determined by the redistribution of electron density in the
Fig. 4. The 31P-NMR spectra of 1 at different temperatures (solution in toluene-d8).

172
A.S. Peregudo6 et al. / Journal of Organometallic Chemistry 597 (2000) 164–174
The 1H- and 31P-NMR data for compound 1 at
DSc = −1293 e.u. The comparison of these thermo-
dynamic parameters with the corresponding values for
Pt complex 2 (Ea=16.490.4 kcal mol−1, DHc
1
different temperatures are given in Table 3. The H-
NMR spectrum of 1 at room temperature (Fig. 3)
reveals one relatively sharp signal of protons from the
o-methyl group. In the range 7.9–8.5 ppm, the signals
of five aromatic protons are observed. The doublet at
7.9 ppm is o-hydrogen from o-Tol group. Four other
signals correspond to the AA% part of AA%BB% pattern
from two p-FC6H4 groups. On lowering the tempera-
ture the coalescence of aromatic protons is observed
and at −97°C four broadened signals appear. The
aromatic multiplets in the range of 6.9–7.4 ppm repre-
sent the superposition of other aromatic protons of
compound 1, of residual protons from toluene-d8 and
PhF added as internal standard for 19F-NMR. The
signal of protons of the o-Me group broadens and
splits into two unequal (3:1) signals at −87°C. A
similar behaviour is observed in the 31P-NMR spectra
(Fig. 4). One sharp signal of 31P observed at room
temperature broadens and splits into two unequal sig-
nals with the same ratio of intensities (3:1) on lowering
the temperature below −92°C.
=
15.790.4 kcal mol−1, DEc =492 e.u.) allows us to
make the conclusion that the migration of MLn group
proceeds with considerably smaller activation barrier in
the case of M=Ni than for M=Pt. At the same time,
the negative change of activation entropy is observed
in the case of 1 and the small positive change of
activation entropy is observed in the case of 2.
The 19F-NMR spectra of 1 at the temperature as
high as −60°C suggest that the rate of the migration
decreases on increasing the temperature. It should be
emphasized that a strong decrease of DFCS value is
simultaneously observed. Moreover, in temperature
range −40 to +70°C a very good correlation of
DFCS value and T are observed (DFCS= −1.357T+
587; r=0.999; s=2.8). Taking into account that the
FCS value of 4-fluorophenyl group is the indicator of
electron density on the adjacent atom [27,28], the last
fact may indicate the decrease in the difference between
electron density on N(1) and N(3) atoms on increasing
the temperature. It may be suggested that at elevated
temperatures the conformation of S-trans type is
stabilized:
It should be noted that in all cases the spectral
behaviour does not depend on concentration (from
0.03 to 0.1 M) and is completely reversible with respect
to temperature changes.
Before discussing the results obtained for nickel com-
plex 1, it should be noted that in the case of the related
Pt complex 2 somewhat different spectral behaviour
has been found in 19F-NMR spectrum [22]. Thus in the
spectra of 2 two sets of sharp fluorine signals at 8 and
11 ppm are observed already at 0°C. The coalescence
of these signals takes place at 63°C and at higher
temperatures the broad singlet begins to narrow up to
80°C. In general the spectral pattern is typical for usual
dynamic spectra and was explained by the intramolec-
ular migration of (o-Tol)Pt(PEt3)2 group, which is ac-
celerated on increasing the temperature and leads to
the averaging of fluorine shielding. For Ni complex 1
the normal spectral pattern, which may be explained in
the context of the tautomeric process:
Such conformation is characterized by a higher de-
gree of coplanarity of NiꢀN(3) and N(1)ꢀN(2) bonds
and better conjugation between the N(3) lone electron
pairs and the NꢁN bond. At the same time, this con-
formation hinders the intramolecular migration of
organometallic group, which may be realized only in
the S-cis-conformation.
1
In discussion of the H- and 31P-NMR data obtained
at different temperatures, it should be noted that spec-
tra in Figs. 3 and 4 are also different from correspond-
ing spectra for Pt compound 2. In particular, the
splitting of proton signals of o-methyl group and of
phosphorus signals in two signals for Ni compound 1
(−84 and −88°C) takes place at substantially lower
temperatures than for Pt complex 2 (25 and 10°C).
Moreover, the ratio of integral intensities of signals for
1 (3:1) is higher than the corresponding ratio (1.4:1) for
2.
is observed at substantially lower temperature. At
−113°C the migration is slow (two separated fluorine
signals with large DFCS=2.08 ppm). On increasing the
temperature up to −60°C, the spectral pattern be-
comes consistent with the pattern that is expected for
acceleration of migration. The following thermody-
namic parameters were found for this process: Ea=
In the previous paper [22], we have suggested that one
of the reasons for the splitting of corresponding signals
1
in H- and 31P-NMR spectra of Pt complex 2 may be
6.890.7 kcal mol−1, DHc =6.490.7 kcal mol−1
,
the hindered rotation around the N(2)ꢀN(3) bond:

A.S. Peregudo6 et al. / Journal of Organometallic Chemistry 597 (2000) 164–174
173
syn- and anti-isomers in trans-(o-Tol)2Ni[P(Me)2Ph]2
[29]. In our situation magnetic non-equivalence may
also be observed, even in the case of the existence of the
only conformer A, and is not necessarily connected
with the existence of conformer B. However, in this
case the shielding of phosphorus nuclei of two PEt3
ligands in the coordination plane, which is orthogonal
on the one hand to the NiꢀN(3)ꢀN(2)ꢁN(1) plane and
on the other hand to the o-Tol ligand plane, should not
be different in the first approximation. This, however, is
not consistent with the experimental data. The mag-
netic non-equivalence of 31P nuclei observed for Pt
complex 2 was the basic evidence in favour of the
existence of two conformers A and B.
At the same time, according to X-ray data obtained
in the present paper, complex 1 in the crystalline state
exists as the 3:1 mixture of two isomeric complexes with
cisoid and transoid orientation of the o-methyl group
relative to the N(2)ꢁN(1) bond. It should be noted that
and the existence of two conformers A and B, in which
the o-methyl protons or 31P nuclei of PEt3 ligands may
be magnetically nonequivalent. The similar explanation
may be suggested for the interpretation of 1H- and
31P-NMR data for Ni complex 1:
1
in the H- and 31P-NMR spectra of compound 1 at low
temperatures, we observed the same ratio (3:1) of inten-
sities of corresponding signals of indicator nuclei.
Moreover, according to the X-ray structure of com-
pound 1, it is actually the CꢀNꢀNꢁNꢀC
rather than the NiꢀNꢀNꢁNꢀC chain which shows a
zig–zag conformation. Italian authors have observed a
similar structure for triazenido complexes of Pd and Pt
[10,12,13].
It should be emphasized that the tautomeric process
in the case of the migration of the (o-Tol)M(PEt3)2-
group implies transformation of two isomers C and D:
The increase in the ratio A/B in going from com-
pound 2 to compound 1 may be caused by the shorter
NiꢀN(3) bond in comparison with the PtꢀN(3) bond.
The thermodynamic parameters calculated from 1H-
and 31P-NMR spectra for possible AUB transforma-
tion in compound 1 as well as in Pt compound 2 are
given in Table 4. It follows from these data that the
possible AUB transformation as well as the migration
of MLn group have a considerably lower activation
barrier for Ni compound 1 than for Pt compound 2.
in which the shielding of o-Me protons must be differ-
ent. In the context of our discussion it is important to
take into account the fact that, in principle small devia-
tions from the orthogonality of planes NꢀNꢁN, o-Tol
and coordination plane of Ni may be sufficient for
causing the magnetic nonequivalence of 31P nuclei to be
observed in 31P-NMR spectra. The existence of such a
small deviation follows from our X-ray data. More-
1
It should be noted that in general H-NMR spectra
of Ni complex 1 may be complicated by the possible
existence of additional isomers for A as well as for B,
due to the hindered rotation around NiꢀN(3) and/or
NiꢀC bonds. For example, the hindered rotation
around NiꢀC bonds causes the existence of a mixture of

174
A.S. Peregudo6 et al. / Journal of Organometallic Chemistry 597 (2000) 164–174
over, it should be pointed out that the close values of
activation parameters obtained in the present paper from
dynamic 19F-, 31P- and 1H-NMR spectra for Ni
compound 1 as well as previously from 19F- and
1H-NMR spectra of Pt compound 2 (Table 4).
References
[1] F.P. Dwyer, J. Am. Chem. Soc. 63 (1941) 78.
[2] C.M. Harris, B.F. Hoskins, R.L. Martin, J. Chem. Soc. (1959)
3728.
[3] L. Toniolo, T. Boschi, G. Deganello, J. Organomet. Chem. 93
(1975) 495.
[4] M. Corbett, B.F. Hoskins, J. Am. Chem. Soc. 89 (1967) 1530.
[5] M. Corbett, B.F. Hoskins, N.J. McLeod, B.P. O’Day, Austr. J.
Chem. 28 (1975) 2377.
[6] W.H. Knoth, Inorg. Chem. 12 (1973) 38.
[7] M. Corbett, B.F. Hoskins, J. Chem. Soc. Chem. Commun.
(1968) 1602.
[8] M. Horner, H. Fenner, W. Hiller, J. Beck, Z. Naturforsh. Teil B
43 (1988) 1174.
[9] J.V. Cuevas, C. Garsia-Herbosa, M.A. Munoz, S.W. Hickman,
A.G. Orpen, N.G. Connelly, J. Chem. Soc. Dalton Trans.
(1995) 4127.
In this connection, it should be once more emphasized
that the dynamic 19F-NMR spectra of compounds 1 and
2 with a large difference in fluorine shielding of the
indicator 4-FC6H4 groups undoubtedly reflect the
migration of the MLn group. It should also be noted that
the close thermodynamic characteristics obtained from
1
dynamic H-, 31P- and 19F-NMR spectra may also be
explained if it is assumed that AUB transformation is the
rate-determining stage of the tautomeric process.
[10] G. Bombieri, A. Immirzi, L. Toniolo, Trans. Met. Chem. 1
(1976) 130.
[11] J. Bailey, V.J. Catalano, H.B. Gray, Acta Crystallogr., Sect. C
49 (1993) 1598.
[12] A. Immirzi, G. Bombieri, L. Toniolo, J. Organomet. Chem. 118
(1976) 355.
[13] L. Taniolo, A. Immirzi, U. Croatto, G. Bombieri, Inorg. Chim.
Acta 19 (1976) 209.
[14] K.R. Laing, S.D. Robinson, M.F. Uttley, J. Chem. Soc. Dalton
Trans. (1974) 1205.
[15] L.D. Brown, J.A. Ibers, Inorg. Chem. 15 (1976) 2794.
[16] C.J. Creswell, M. Queiros, S.D. Robinson, Inorg. Chim. Acta
60 (1982) 157.
[17] E.V. Borisov, A.S. Peregudov, S.A. Postovoi, E.I. Fedin, D.N.
Kravtsov, Izv. Akad. Nauk. Ser. Khim. (1986) 550.
[18] A.N. Nesmeyanov, E.V. Borisov, A.S. Peregudov, D.N.
Kravtsov, L.A. Fedorov, E.I. Fedin, S.A. Postovoi, Dokl. Acad.
Sci. USSR 247 (1979) 1154.
4. Conclusions
The dynamic NMR spectra of the triazenido complex
of nickel(II) (1) suggest that the migration of (o-
Tol)Ni(PEt3)2 group occurs at low temperatures in solu-
tion. This migration implies the interconversion of two
isomers with cisoid or transoid orientation of the o-Me
group relative to the N(2)ꢁN(1) bond. The interconver-
sion becomes slow at elevated temperatures due to the
rotation processes. In solid state, complex 1 exists as a
3:1 mixture of above isomers.
5. Supplementary material
[19] D.N. Kravtsov, A.N. Nesmeyanov, L.A. Fedorov, E.I. Fedin,
A.S. Peregudov, E.V. Borisov, P.O. Okulevich, S.A. Postovoi,
Dokl. Acad. Sci. USSR 242 (1978) 347.
[20] P. Peringer, Inorg. Nucl. Chem. Lett. 16 (1980) 461.
[21] A.S. Peregudov, Dissertation, 1993, Moscow.
[22] A.S. Peregudov, D.N. Kravtsov, G.I. Drogunova, I.A.
Godovikov, Inorg. Chim. Acta 280 (1998) 238.
[23] D.C. Morreli, J.K. Kochi, J. Am. Chem. Soc. 96 (1975)
7262.
[24] G.M. Sheldrik, SHELXTL Version 5.0, Software Reference Man-
ual, Siemens Industrial Automation, Madison, WI, 1994.
[25] J. Sandsrom, Dynamic NMR Spectroscopy, Academic, London,
1982.
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC no. 117228 for compound 1. Copies
of this information may be obtained free of charge from
The Director, CCDC, 12, Union Road, Cambridge CB2
1EZ, UK (Fax: +44-1223-336033 or e-mail: deposit@
ccdc.cam.ac.uk or http://www.ccdc.cam.ac.uk).
Acknowledgements
[26] H.-B. Bu¨rgi, J.D. Dunitz (Ed.), Structure Correlation, vol. 2,
VCH, Weinheim, New York, 1994, p. 767.
[27] N.A. Ogorodnikova, A.S. Peregudov, E.I. Fedin, D.N.
Kravtsov, Bull. Acad. Sci. 38 (1989) 2037.
[28] D.N. Kravtsov, A.S. Peregudov, O.B. Sherbakova, Yu. A.
Borisov, Russ. Chem. Bull. 44 (1995) 1841.
[29] K. Miki, M. Tanaka, N. Kasai, M. Wada, J. Organomet. Chem.
352 (1988) 385.
The authors are indebted to the Russian Foundation
for Basic Research for financial support (Project No.
97-03-33783) and for covering the License Fee for the use
of the Cambridge Structural Database, which was em-
ployed for the analysis of the structural results discussed
in the present paper (Project No. 96-07-89187).
.