Properties of some ferrocene macrocyclic dioxotetraamines: The roles of aromatic side-arms

Article abstract of DOI:10.1016/S0022-328X(01)00928-7

The coordination chemistry and electrochemical behavior of a series of ferrocene macrocyclic dioxotetraamines 1-4 each of which bears a pendant aromatic group and a malonic diamide fragment in their framework were investigated by electronic absorption spectroscopy and cyclic voltammetry. For the purpose of comparison, their model compounds 5-8 were also synthesized and studied. The stoichiometry of 1:1 for the complexes has been confirmed by FAB mass spectroscopy and UV-vis titration of Cu(II) ion with ligands 4 and 8. Stability constants for the equilibrium of the above complexes have been evaluated. The ferrocene ligands 1-4 can electrochemically recognize the transition metal ions (Co²⁺, Ni²⁺ and Cu²⁺), while those ligands with a complexible atom in the side-arm can stabilize selectively the uncommon state of Ni(III).

Full text of DOI:10.1016/S0022-328X(01)00928-7

Journal of Organometallic Chemistry 637–639 (2001) 327–334
www.elsevier.com/locate/jorganchem
Properties of some ferrocene macrocyclic dioxotetraamines: the
roles of aromatic side-arms
Peng Xue, Quan Yuan, Enqin Fu *, Chengtai Wu
Department of Chemistry, Wuhan Uni6ersity, Wuhan 430072, China
Received 5 January 2001; received in revised form 9 April 2001; accepted 10 April 2001
Abstract
The coordination chemistry and electrochemical behavior of a series of ferrocene macrocyclic dioxotetraamines 1–4 each of
which bears a pendant aromatic group and a malonic diamide fragment in their framework were investigated by electronic
absorption spectroscopy and cyclic voltammetry. For the purpose of comparison, their model compounds 5–8 were also
synthesized and studied. The stoichiometry of 1:1 for the complexes has been confirmed by FAB mass spectroscopy and UV–vis
titration of Cu(II) ion with ligands 4 and 8. Stability constants for the equilibrium of the above complexes have been evaluated.
The ferrocene ligands 1–4 can electrochemically recognize the transition metal ions (Co2+, Ni2+ and Cu2+), while those ligands
with a complexible atom in the side-arm can stabilize selectively the uncommon state of Ni(III). © 2001 Elsevier Science B.V. All
rights reserved.
Keywords: Ferrocene; Macrocyclic polyamine; Synthesis; Coordination; UV–vis spectrometry; Cyclic voltammetry
1. Introduction
the fact that this selectivity is ring-size-dependent [8].
In order to investigate the side-arm effect in this type
of ligands on stabilizing the trivalent state of these
cations, we have synthesized recently the ferrocene
macrocyclic dioxotetraamines which bear a pendant
pyridine, furan or ferrocene (1–4) [9]. For the purpose
In the past two decades, a large number of redox-ac-
tive host molecules have been synthesized with the hope
that they may be used as chemosensors [1,2], redox-
switchable ligands [3] or redox catalysts [4].
Among these compounds, ferrocene macrocyclic
polyamines have shown some interesting behavior due
to their redox activity and complexibility toward transi-
tion metal ions and anions [5]. On the other hand, the
macrocyclic dioxopolyamines possess the unique dual
structural features of saturated macrocyclic polyamines
and oligopeptides and show some interesting coordina-
tion behavior [6,7]. Recently, we have reported that the
ferrocene macrocyclic dioxopolyamines which bear
iminodiacetamide in the framework of their rings can
electrochemically recognize the cations such as Co2+
,
Ni2+ and Cu2+, and selectively stabilize the uncom-
mon state of Ni3+ or Cu3+. Particularly interesting is
* Corresponding author. Tel.: +86-27-87684117; fax: +86-27-
87647617.
E-mail address: jgqin@whu.edu.cn (E. Fu).
Scheme 1.
0022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(01)00928-7

328
P. Xue et al. / Journal of Organometallic Chemistry 637–639 (2001) 327–334
of comparison, model compounds 5–8, which were not
reported before, were also synthesized (Scheme 1).
Phenyl was chosen in the model compounds as the
substitute for the reason that it exhibits the aromaticity
similar to pyridine, furan and ferrocene but without the
complexible atom or redox activity. In this paper, the
synthesis of compounds 5–8 and some coordination
compounds of 1 and 2 are presented, and FABMS,
UV–vis spectroscopy, and cyclic voltammetry (CV) are
employed to investigate the coordination chemistry and
the electrochemical behavior of compounds 1–8.
EtOH, to which a solution of benzyl chloride (0.60 g,
5.0 mmol) in EtOH (40 ml) was added in drops under
reflux. After the addition, 0.40 g (2.9 mmol) of K2CO3
was added and the mixture was heated under reflux for
an additional hour. After the solvent was evaporated
under reduced pressure, the residue was taken up in
CHCl3, and washed with water. The organic layer was
separated, dried over anhydrous Na2SO4, filtered and
concentrated under reduced pressure. The residue was
separated by column chromatography on silica. Com-
pound 6 was eluted by a mixture of CHCl3 and MeOH
(100:3, v/v), and purified by crystallization from
Me2CO. Yield: 0.30
g (15%), m.p. 150–152 °C.
2. Experimental
FABMS (exact mass, 394.24, m/z, %): 395 [M++1,
77]; 394 [M+, 9]; 303 [M+−91, 42]; 91 [C7H+7 , 100].
1H-NMR (CDCl3, 90 MHz, l ppm): 2.24–2.54 (m, 8H,
CH2NCH2CH2NCH2); 3.26 (s, 2H, COCH2CO); 3.28–
3.40 (m, 4H, CONCH2); 3.50 (s, 4H, PhCH2N); 6.64–
6.94 (s, br, 2H, CONH); 7.08–7.40 (m, 10H, C6H5). IR
(KBr pellet, cm−1): 3303 (s, of CONH); 3121 (m, of
aryl); 2928, 2833 (m, of alkyl); 1668, 1634 (s, of
CONH); 1535, 1550 (s, of CONH); 734 (m, mono-sub-
stituted phenyl); 698 (m, mono-substituted phenyl).
The mixture of CHCl3 and MeOH (100:5–8, v/v)
eluted compound 5, which was purified by crystalliza-
tion from Me2CO. Yield: 0.91 g (60%), m.p. 204–
206 °C. FABMS (exact mass, 304.19, m/z, %): 305
[M++1, 100]; 304 [M+, 8]; 213 [M+−91, 5]; 91
[C7H+7 , 80]. 1H-NMR (CDCl3, 90 MHz l ppm): 2.20 (s,
1H, CNHC); 2.40–2.76 (m, 8H, CH2NCH2CH2NCH2);
3.22 (s, 2H, COCH2CO); 3.24–3.50 (m, 4H, CONCH2);
3.60 (s, 4H, PhCH2N); 7.04–7.16 (m, 6H, CONH,
C6115); 8.00–8.24 (br, 1H, CONH). IR (KBr pellet,
cm−1): 3311, 3257 (m, of CONH); 3171 (m, of CNHC);
3125 (m, of aryl); 2940, 2831 (m, of alkyl); 1686, 1658
(s, of CONH); 1649, 1462 (m, of aryl); 1547 (s, of
CONH); 769 (m, mono-substituted phenyl); 694 (m,
mono-substituted phenyl).
2.1. General
4-Ferrocenylmethyl-1,4,7,10-tetraaza-cyclotridecane-
11,13-dione (1), 4,7-bis(ferrocenylmethyl)-1,4,7,10-tetra-
aza-cyclotridecane-11,13-dione (2), 4-ferrocenylmethyl-
7 - furfuryl - 1,4,7,10 - tetraaza - cyclotridecane - 11,13-
dione (3), 4-ferrocenylmethyl-7-picolyl-1,4,7,10-tetra-
aza-cyclotridecane-11,13-dione (4) were prepared ac-
cording to the method in our previous report [9]. Pi-
colyl chloride was prepared according to a procedure in
the literature [10]. The solvents used for the reaction
were thoroughly dried analytical reagents. All melting
points (m.p.) are uncorrected. IR spectra were obtained
on a Nicolet 170SX FT-IR or Shimadzu FT-IR 8000
spectrophotometer. Proton magnetic resonance spectra
(300 or 90 MHz) were recorded on a Varian Mercury-
VX 300 or a JEOL FX90Q spectrometer with Me4Si as
internal reference. Mass spectra (FAB) were recorded
on a ZAB 3F-HF spectrometer.
CV was performed in a conventional three-electrode
cell at 2091 °C, using a PAR Model 173 potentiostat/
galvanostat (EG&G), which was described in our previ-
ous report [8] except that the solvent was made of 2:1
(v/v) EtOH–H2O.
UV–vis spectra were recorded on a Shimadzu UV-
160A or UV-3100 spectrometer at 2091 °C in the
mixture of 2:1 (v/v) EtOH–H2O. The concentration of
the ligands was always 1.0 mM for both UV–vis and
electrochemical studies. The transition metal salts were
used as in electrochemical experiments (the details can
be found in the literature [7]). After the addition of
metal ions, the solutions of the ligands were adjusted to
pH 6–7 with NaOH (0.01 M).
2.2.2. 4-Benzyl-7-ferrocenylmethyl-1,4,7,10-tetraaza-
cyclotridecane-11,13-dione (7)
To a mixture of compound 1 (0.20 g, 0.48 mmol) and
K2CO3 (0.55 g, 4.0 mmol) in MeCN (10 ml), a solution
of benzyl chloride (0.070 g, 0.55 mmol) in MeCN (5 ml)
was added under reflux with stirring, the reaction was
monitored by TLC. After being heated under reflux for
18 h, the reactant mixture was worked-up as described
above, and separated by column chromatography on
alumina (100:2, CHCl3–MeOH, v/v), crystallization
from MeCN afforded an orange powder. Yield: 0.15 g
(62%), m.p. 152–154 °C. FABMS (exact mass, 502.20,
m/z, %): 503 [M++1, 27]; 502 [M+, 35]; 359 [M+−
199+56, 5]; 303 [M+−199, 28]; 212 [M+−199−91,
3]; 199 [C5H5FeC5H4CH+2 , 100]; 91 [C7H+7 , 60]. 1H-
NMR (CDCl3, 300 MHz, l ppm): 2.31–2.49 (m, 8H,
CH2NCH2CH2NCH2); 3.26 (s); 3.35–3.37 (m); 3.52 (s)
2.2. Synthesis of ligands 5–8
2.2.1. 4-Benzyl-1,4,7,10-tetraaza-cyclotridecane-
11,13-dione (5) and 4,7-bis(benzyl)-1,4,7,10-tetraaza-
cyclotridecane-11,13-dione (6)
1,4,7,10 - Tetraaza - cyclotridecane - 11,13 - dione (L0)
(1.50 g, 7.0 mmol) was dissolved in 40 ml of absolute

P. Xue et al. / Journal of Organometallic Chemistry 637–639 (2001) 327–334
329
(10H, COCH2CO, CONCH2, FcCH2N, PhCH2N);
3.94–3.96, 4.05–4.11 (d, t, 9H, C5H5FeC5H4); 6.70,
7.03 (br, br, 2H, CONH); 7.28–7.34 (m, 5H, C6H5). IR
(KBr pellet, cm−1): 3429, 3315 (m-s, of CONH); 3128
(s, of phenyl); 2920, 2860 (m, of alkyl); 1674, 1647 (s, of
CONH); 1533 (s, of phenyl); 1549 (s, of CONH); 1101,
1000 (m, of ferrocenyl); 736, 710 (m, mono-substituted
phenyl).
2.3.3. Nickel(II) complex of 2, [NiH−2L(2)]
To the MeOH solution (6 ml) of 2 (0.305 g, 0.500
mmol) was added the aqueous solution (20 ml) of
Ni(OAc)2·4H2O (124 mg, 0.500 mmol). The mixture
was heated to 50–60 °C, and its pH value was adjusted
to ca. 9 with NaOH. After being heated for 20 min, the
mixture was filtered and cooled in a refrigerator, and
red–orange crystals were obtained. Yield: 0.11 g (33%).
FABMS (exact mass, 666.09, m/z, %): 667 [M++1,
22], 666 [M+, 7], 199 [C5H5FeC5H4CH+2 , 100]. IR (KBr
pellet, cm−1): 3078 (m, of ferrocenyl); 2923, 2868 (m, of
alkyl); 1585, 1566 (s, of CON−); 1104, 999 (m, of
ferrocenyl); 806 (m, of ferrocenyl). UV–vis: umax (nm)
(m, dm3 mol−1 cm−1): 450 (357), 323 (488).
2.2.3. 4-Benzyl-7-(2-picolyl)-1,4,7,10-tetraaza-
cyclotridecane-11,13-dione (8)
This compound was prepared following the same
procedure as described for 7 except that column chro-
matography was carried out on silica (100:5, CHCl3–
MeOH, v/v), the compound was obtained as a viscous
oil. Yield: 0.20 g (51%). FABMS (exact mass, 395.23,
m/z, %): 396 [M++1, 82]; 395 [M+, 17]; 304 [M+−
2.3.4. Cobalt(II) complex of 2, [CoH−2L(2)]
This complex was prepared following the same pro-
cedure as described for [NiH−2L (2)] except that the
pH value was not adjusted and MeOH was replaced by
MeCN, black–purple crystals were obtained. FABMS
(exact mass, 667.09, m/z, %): 668 [M++1, 4]; 667
[M+, 5]; 199 [C5H5FeC5H4CH+2 , 100]. IR (KBr pellet,
cm−1): 3070 (m, of ferrocenyl); 2926, 2870 (m, of
alkyl); 1588 (s, of CON−); 1104, 999 (m, of ferrocenyl);
806 (m, of ferrocenyl). UV–vis: umax (nm) (m,
dm3 mol−1 cm−1): 554 (264), 373 (850).
1
91, 12]; 91 [C7H+7 , 100]. H-NMR (CDCl3, 90 MHz, l
ppm): 2.28–2.60 (m, 8H, CH2NCH2CH2NCH2); 3.30,
3.12–3.46 (m, 6H, COCH2CO, CONCH2); 3.52 (s, 2H,
PhCH2N); 3.62 (s, 2H, NCH2-pyridyl); 6.68–6.88 (m,
1H, 5-H of pyridyl); 7.04–7.32 (d, 6H, C6H5, 3-H of
pyridyl); 7.48–7.72 (m, 1H, 4-H of pyridyl); 8.00–8.32
(m, 2H, CONH); 8.40–8.54 (m, 1H, 6-H of pyridyl). IR
(KBr pellet, cm−1): 3125 (s, of aryl); 2920, 2830 (m, of
alkyl); 1665, 1650 (s, of CONH); 1549 (m-s, of CONH);
735, 700 (m, mono-substituted phenyl).
3. Results and discussion
2.3. Preparation of the complexes
Compounds 1–8 were prepared following the proce-
dure shown in Scheme 1.
2.3.1. Copper(II) complex of 1, [CuH−2L(1)]
The EtOH solution (20 ml) of 1 (0.20 g, 0.49 mmol)
and Cu(OAc)2·H2O (0.10 g, 0.50 mmol) was heated to
reflux under N2 for 3 h. After the solution was concen-
trated under reduced pressure, crystallization from
EtOH yielded black–purple needles. Yield: 0.15 g
(65%). FABMS (exact mass, 473.08, m/z, %): 474 [M+
+1, 48]; 473 [M+, 15]; 413 [(1)++1, 5]; 199
[C5H5FeC5H4CH+2 , 100]. IR (KBr pellet, cm−1): 3070
(m, of ferrocenyl); 2922, 2876 (m, of alkyl); 1622, 1600
(s, of CON−); 1091, 1005 (m, of ferrocenyl); 805 (m, of
ferrocenyl). UV–vis: umax (nm) (m, dm3 mol−1 cm−1):
451 (147).
3.1. Coordination of compounds 1–8 with Ni2+, Co2+
and Cu2+ ions
IR, UV–vis spectra and FABMS data of all the
complexes prepared confirmed the coordination be-
tween the transition metal ions and the macrocyclic
ligands. Their IR spectra showed changes [8,11] in
stretching vibration of both CꢀO (i.e. from 1670–l630
to l620–l570 cm−1) and NꢁH (disappeared in the com-
plexes) in amide moieties. FABMS also showed the
molecular ion peaks of MH−2L.
The addition of Ni2+ ion to the solutions of those
ligands bearing ferrocene (i.e. 1–4, 7) usually did not
cause obvious effect on the shifts in umax of ferrocene,
but caused some changes in the extinction coefficient of
ferrocene (except in the case of 7 and Ni2+) (Table 1).
The addition of Cu2+ and Co2+ ions usually caused
the shifts in umax and an increase in absorbance of
ferrocene. However, in the case of those ligands bearing
no ferrocene (i.e. 5, 6, 8), the absorption peaks for all
the central ions Ni2+, Cu2+ and Co2+ in the com-
plexes showed distinctly (Fig. 1) that the extinction
coefficients of the central cations increased in compari-
2.3.2. Copper(II) complex of 2, [CuH−2L(2)]
This compound was prepared following the same
procedure as described for [CuH−2L (1)]. Recrystalliza-
tion from MeOH–H2O (1:2, v/v) afforded black–pur-
ple needles. FABMS (exact mass, 671.09, m/z, %): 672
[M++1, 3]; 199 [C5H5FeC5H4CH+2 , 100]. IR (KBr
pellet, cm−1): 3079 (m, of ferrocenyl); 2931, 2866 (m, of
alkyl); 1592, 1568 (s, of CON−); 1105, 1001 (m, of
ferrocenyl); 798 (m, of ferrocenyl). UV–vis: umax (nm)
(m, dm3 mol−1 cm−1): 441 (334).

330
P. Xue et al. / Journal of Organometallic Chemistry 637–639 (2001) 327–334
Table 1
The increase in absorbance at umax for the copper
complexes of ligands 4 and 8 upon incremental addition
of Cu2+ ion are given in Table 2, the data are illus-
trated in Fig. 2. The progressive increase in absorbance
intercepted the limiting ‘infinity’ absorbance both at
molar ratio 1.1:1 of Cu2+ to ligands 4 or 8. Therefore
the complexation stoichiometry for the complexes is
1:1. The equilibrium constants were evaluated by the
plot of 1/DA against 1/[M] based on the Benisi–Hilde-
brand equation [12] to obtain the values for bimolecu-
lar equilibrium constants of 161 (Cu2+ –4), 100
(Cu2+ –8) mol−1 dm3.
The values of umax (nm) and m (mol−1 dm3 cm−1) for ligands 1–8 and
their complexes
Ligand
In the presence of M2+
Ni2+
Cu2+
Co2+
1
2
3
4
u1 (m1)
u2 (m2)
436 (115)
325 (93)
430 (231)
315 (334)
412 (143) a 363 (887)
665 (22) 557 (279)
u1 (m1) 438, 0.115 b 450 (357) c 441 (334) c 373 (850) c
u2 (m2) 325, 0.082
323 (488)
433 (107)
324 (103)
429 (181)
315 (297)
387 (78)
554 (264)
433 (120)
589 (21) d
417 (389)
u1 (m1)
u2 (m2)
u1 (m1)
u2 (m2)
u (m)
u1 (m1)
u2 (m2)
u1 (m1)
u2 (m2)
u1 (m1)
u2 (m2)
435 (105)
323 (87)
432 (126)
320 (155)
448 (201)
524 (172)
402 (230)
595 (207)
522 (119)
528 (34)
3.2. Electrochemical beha6ior of compounds 1–8
5
6
585 (96)
590 (10)
634 (10)
428 (154)
311 (217) e
462 (599)
635 (89)
382 (65)
Some of the cyclic voltammograms for the ligands
before and after addition of Ni2+, Co2+ and Cu2+
metal ions were chosen and are shown in Figs. 3–5.
The peak potentials (E values vs. SCE) of each couple
for all the ligands and their complexes are summarized
in Tables 3 and 4.
7
8
434 (114)
322 (89)
435 (111)
324 (84)
462 (152)
413 (149)
657 (28)
601 (182)
a For the prepared complex, at 451 nm (m, 147).
b Absorbance of its saturated solution.
c Of the prepared complex.
The redox peaks separations (66–92 mV) of most
couples (Tables 3 and 4) indicated the quasi-reversible
redox process for these ligands and their complexes.
Some couples are reversible (e.g. an oxidation peak at
407 mV for the nickel(II) complex of 4, an oxidation
peak at 423 mV for ligand 3 at pH 7, etc.), and some
are irreversible (e.g. an oxidation peak at 415 mV for
the cobalt(II) complex of 1, 644 mV for the copper(II)
complex of 5, etc.). The wider potential peaks’ separa-
tion between the anodic and the cathodic peaks sug-
gests a stronger interaction between the metal ion and
the ferrocene subunit, because the electrostatic interac-
tion makes the oxidation of ferrocene less favorable
(which results in the anodic shift in the oxidation peak
potential) and is in favor of the reduction of ferroce-
nium (which leads to the cathodic shift in the reduction
peak potential).
d Another at 638 nm (m, 20).
e The others at 591 nm (m, 39); 639 nm (m, 39).
In order to investigate how the pH values of solu-
tions affect the redox peak potentials of ferrocene in the
ligands, the electrochemical behavior of these ligands
have been investigated under different pH values. It
was observed that a decrease in the pH value results in
an increase in the oxidation peak potential of ferrocene
(Table 3, Fig. 3). A lower pH value favors the protona-
tion of amines of macrocyclic polyamines, the electro-
static repulsion between protonated amine and
ferrocenium cation have made the oxidation of fer-
rocene more difficult; as a result, the oxidation poten-
tial of ferrocene is higher in the solution of lower pH
values.
Fig. 1. UV–vis spectra of ligands 3, 4, 8 before and after addition of
Cu2+ ion.
son with the hydrated ions. All these evidences showed
the occurrence of interaction between ligands 1–8 and
the metal ions mentioned above.
For model compounds 5–8, the extinction coeffi-
cients for each central metal ion have remarkable dif-
ferences. The approximate order 8\5\7$6 for
nearly all these metal ions may reflect the order of
coordination ability of these ligands. It may be con-
cluded that the coordination ability was enhanced by
the side-arm with a complexible atom while it reduced
for the side-arm without one.
Maybe there are different species in the solution of
2–4 at pH 3–5, 7, 7, respectively, and in the complex
solution of 1 (Ni(II), Co(II), Cu(II)), 3 (Cu(II)), 4
(Ni(II)), because more than two redox peaks were
observed (Tables 3 and 4).

P. Xue et al. / Journal of Organometallic Chemistry 637–639 (2001) 327–334
331
Table 2
Absorption data at complex umax for solution of ligands 4 (1 mM) and 8 (1 mM) upon incremental molar ratio addition of Cu2+
Cu2+–4
A
Cu2+–4
A
Cu2+–8
A
Cu2+–8
A
0
0.000
0.026
0.039
0.072
0.095
0.122
0.145
0.171
0.196
0.216
0.237
1.1
1.2
1.3
1.4
1.5
1.6
1.8
2.0
2.5
3.0
0.248
0.253
0.260
0.265
0.276
0.280
0.288
0.296
0.302
0.302
0
0.0028
0.0204
0.0415
0.0610
0.0868
0.1051
0.1250
0.1526
0.1725
0.1971
1.0
1.1
1.2
1.3
1.4
1.5
1.6
2.0
2.4
3.0
0.2086
0.2302
0.2347
0.2411
0.2517
0.2576
0.2515
0.2655
0.2611
0.2603
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.0
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
[13] aneN4 is just fit to this cation according to the
structure determined by X-ray crystallography [11]. But
the low-spin d7 Ni(III) (Ni(III)ꢁN bond length is equal
Compared with the free ligands 1, 2, 4 (at pH 7), the
addition (1:1, molar ratio) of each of the transition
metal ions Cu(II), Ni(II), Co(II) has caused significant
shifts in the oxidation peak potentials for the corre-
sponding Fc+/Fc couple to the more anodic direction.
The largest shifts for 1, 2 and 4 were caused by cop-
per(II) (DEpa=l47 mV), nickel(II) (DEpa=104 mV)
and cobalt(II) (DEpa=75 mV) ions, respectively. Fur-
ther addition (2:1, 3:1, molar ratio) of these metal ions
afforded almost the same results. The DEpa values show
the difference between the potentials of Fc+/Fc in the
complexed (Epa,c) and free (Epa,f) ligands (Table 4). No
obvious shift in potential was observed after the addi-
tion of these metal ions to the electrochemical solution
of the model compound 7. One of the reasons may be
short-time reaction with the metal ions; another is its
weaker coordination ability compared with ligand 1.
It should be noticed that ligands 3, 4 and 8 which
bear a side-arm with a complexible atom can stabilize
selectively the uncommon state of Ni(III) ion, which is
different from the reported results of stabilization of
both Ni(III) and Cu(III) ions [13]. While those bearing
a side-arm with no complexible atom such as ligands 1,
2, 6, 7 did not show such an ability. Ligand 5 can
stabilize selectively Cu(III) ion instead, although it is
irreversible (Table 4).
,
to 1.97 A [16]) would possess longer bond distance with
N atom than the low-spin d8 Ni(II). Moreover, the
low-spin d7 Ni(III) prefers an axial coordination; there-
fore, the uncommon Ni(III) has to rise from the plane
The selective stabilization of Cu(III) or Ni(III) may
be associated with the following factors [7]: (i) the ring
size of macrocycle, (ii) the ionic radii of the two metal
ions in different oxidation states, and (iii) the coordina-
tion configuration which depends on the ligand field
strength and the electronic states of metal cations.
Due to the strong co-planarity caused by the conju-
gated imide anions after coordination, the dioxo [13]
aneN4 (L0) yields the square-planar low-spin Ni(II)
complex [6,14]. The ideal MꢁN bond length for fitting
,
into the cavity of the coplanar 3, 4 or 8 (ꢀ1.92 A) is
not prohibitively short for the low-spin Ni(II)
,
(Ni(II)ꢁN average bond length is equal to 1.88 A [10],
Fig. 2. Plots of Amax vs. molar ratio of Cu2+ ion to ligands 4 and 8.
,
1.89 A [14,15]). Actually the N,N%-di-substituted dioxo

332
P. Xue et al. / Journal of Organometallic Chemistry 637–639 (2001) 327–334
ion to rise. So the oxidation of Ni(II) becomes easier
and the oxidation peak potential occurred at a lower
position (3, Epa=717 mV; 4, Epa=542 mV; 8, Epa=
540 mV) than that with no side-arm (L0, E°=0.92 V
[6]). To some extent, the reason for those without a
complexible atom in the side-arm exhibiting no evi-
dences of stabilization of Ni(III) in the potential
range tested is that they have no atom for axial coor-
dination.
On the other hand, the ring sizes of the coplanar
ligand 3, 4, 8 are short for the Cu(II)ꢁN bond length
,
,
(average value is 1.98–2.00 A [17], 2.03 A [6,14], so
that the metal ion Cu(II) has to stay above the copla-
nar N4. The change of Cu(II) to Cu(III) (d8, low-
,
spin) will drastically reduce (0.12–0.17 A) the ionic
radius of the metal [18]; as a result, Cu(III) tends to
fall into the cavity of the macrocycle without an axial
coordination. However, the axial coordination will
prevent the cation from falling while it is oxidized, so
the oxidation peak potential will shift toward the
more anodic direction, in our experiment it did not
appear in the scan range. Without these disadvan-
tages, the model compound 5 exhibits the oxidation
peak at 644 mV for Cu(III)H−2L/Cu(II)H−2L.
Model compounds 6 and 7 do not exhibit this prop-
erty as compound 5 does. This may due to its weak
coordination described in Section 3.1. The reason for
the ferrocene macrocycle 1 not to exhibit this prop-
erty may be the effect of ferrocenium adjacent to the
copper ion, and this will be discussed in another pa-
per.
Fig. 3. Cyclic voltammograms of ligand 1 in different pH values.
These results are in good agreement with that re-
ported in our previous paper [8] about ferrocene
macrocyclic polyamines bearing iminodiacetamides.
Kimura [13] has investigated the effect of the side-
arm attached to some dioxo-polyamines on stabilizing
Ni(III), Cu(III). The trend he found is in coincidence
with our work, but the results indicated that our
compounds exhibited higher selectivity under our ex-
perimental conditions.
Fig. 4. Cyclic voltammograms of ligand 3 in the presence of Ni2+
ion.
4. Conclusions
Cyclic voltammetry revealed that a decrease in the
pH value of the solution results in an increase in the
oxidation peak potential of ferrocene in ligands 1–4.
The ferrocene ligands can electrochemically recognize
the transition metal ions, Co(II), Ni(II) and Cu(II),
while an uncommon state of Ni(III) can be stabilized
selectively by ligands 3, 4 and 8 which bear a side-
arm with a complexible atom. Electronic absorption
spectroscopy identified that the interaction between
the ligands and the transition metal ions has an effect
on d–d transition of ferrocene and the guest cation.
The complexible atom in the side-arm plays an im-
Fig. 5. Cyclic voltammograms of ligand 4 in the presence or absence
of Ni2+ ion.
of the four N’s ring while Ni(II) is oxidized. The
axial coordination by the nitrogen or the oxygen of
pyridine or furan in the side-arm will help the metal

P. Xue et al. / Journal of Organometallic Chemistry 637–639 (2001) 327–334
333
complexes were verified by FABMS of the complexes
and UV–vis titration. The equilibrium constants for the
copper complexes of 4 and 8 were evaluated.
portant role in selective stabilization of the trivalent
state of Ni(III) and in the complexibility of a ligand.
The complexation stoichiometry 1:1 (M2+ –L) for the
Table 3
The E values for ligands 1–4, 7 in the solutions of different pH values
pH
1
3
4
5
6
7
8
9
Epa (mV)
Epc (mV)
DE (mV) a
Epa (mV)
Epc (mV)
DE (mV)
Epa (mV)
Epc (mV)
DE (mV)
Epa (mV)
Epc (mV)
DE (mV)
Epa (mV)
Epc (mV)
DE (mV)
400
328
72
382
315
67
373
299
74
328
254
74
345
293
52
308
231
77
328
260
68
308
229
79
2
3
4
7
388/502 b,c
324/440
64/62
384/504 c
317/438
67/66
435 d
361
74
336/423 c
270/375
66/48
345/407 c
273/375
72/32
340
328 e
247
81
485
419
66
457 f
387
70
337 e
255
82
328 g
255
73
408
342
66
332 e
257
75
269
71
a Redox peaks separation DE=Epa−Epc, error of E values, 95 mV.
b pH 3–4.
c Relative to Fc+/Fc of protonated species.
d pH 4–5.
e pH 8–9.
f pH 5–6.
g pH 9–10.
Table 4
The E values for ligands 1–5, 7, 8 before and after addition of metal ions a
Complex/ligand
Nil
Ni2+
Cu2+
Co2+
1
Epa (mV)
Epc (mV)
DEpa (mV) b
Epa (mV)
Epc (mV)
DEpa (mV)
Epa (mV)
Epc (mV)
DEpa (mV)
Epa (mV)
Epc (mV)
DEpa (mV)
Epa (mV)
Epc (mV)
Epa (mV)
Epc (mV)
DE (mV)
Epa (mV)
Epc (mV)
328
254
312/424
233
−16/96
449
375/475
243/338
47/147
400
313/415
232
−15/87
412
345
67
2
3
4
345
293
383
341
55
104
336/423
270/375
328 c/717 d
335 e/392 f
270
331
256
258
g
g
57
409
330
345/407
273/375
337 c/427 c/542 d
278/356/468
g/20
420
356/275
75
64
5
7
644 h
340
269
339
332
332
264
258
262
g
g
g
8
540 d
454
a All the experiments were carried out at the pH value of ca. 7 (no buffer).
b DEpa=Epa,c−Epa,f, error of E values, 95 mV.
c Relative to Fc+/Fc.
d Relative to Ni(III)H−2L/Ni(II)H−2L couple.
e Relative to Fc+/Fc of uncomplexed species.
f Relative to Fc+/Fc of complex.
g The absolute values are not more than 10 mV.
h Relative to Cu(III)H−2L/Cu(II)H−2L couple, irreversible.

334
P. Xue et al. / Journal of Organometallic Chemistry 637–639 (2001) 327–334
[4] M.J.L. Tendero, A. Benito, J. Cano, J.M. Lloris, R. Martinez-
Molecular systems capable of controlling redox prop-
Manez, J. Soto, A.J. Edwards, P.R. Raithby, A. Rennie, J.
Chem. Soc. Chem. Commun. (1995) 1643.
[5] P.D. Beer, Z. Chen, M.G.B. Drew, J. Kingston, M.
Ogden, P. Spencer, J. Chem. Soc. Chem. Commun. (1993)
1046.
[6] E. Kimura, in: S.R. Cooper (Ed.), Crown Compounds: Toward
Future Application, VCH, Weinheim, 1992, p. 81.
[7] E. Kimura, J. Coord. Chem. 15 (1986) 1.
[8] P. Xue, E. Fu, M. Fang, C. Gao, C. Wu, J. Organomet. Chem.
598 (2000) 42.
erties of metal ions are of interest in such fields as
catalysis and electrocatalysis [19], in which some appli-
cation may be found on the basis of the results of the
electrochemical study of these compounds. Besides, the
ferrocene ligands 1–4 may find application in electro-
chemical sensors.
[9] P. Xue, E. Fu, G. Wang, C. Wu, Synth. Commun. 29 (1999)
1585.
Acknowledgements
[10] J.F. Vozza, J. Org. Chem. 27 (1962) 3856.
[11] X. Bu, D. An, Z. Zhu, Y. Chen, M. Shionoya, E. Kimura,
Polyhedron 16 (1997) 179.
[12] H.A. Benisi, J.H. Hildebrand, J. Am. Chem. Soc. 71 (1949)
2703.
[13] E. Kimura, Pure Appl. Chem. 61 (1989) 823.
[14] V.J. Thom, J.C.A. Boeyens, G.J. Mcdougall, R.D. Hancock, J.
Am. Chem. Soc. 106 (1984) 3198.
[15] M.F. Richardson, R.E. Sievers, J. Am. Chem. Soc. 94 (1972)
4134.
The financial support from the National Natural
Science Foundation of China and the kind supply of
electrochemical instruments from Professor Fu-xing
Gan of this University are gratefully acknowledged.
References
[1] C.D. Hall, in: A. Togni, T. Hayashi (Eds.), Ferrocenes, VCH,
Weinheim, 1995, p. 279 (and references therein).
[2] P.D. Beer, Adv. Inorg. Chem. 39 (1992) 79 (and references
therein).
[3] A.E. Kaifer, S. Mendoza, in: G.W. Gokel (Ed.), Comprehensive
Supramolecular Chemistry, vol. 1, Pergamon, Oxford, 1996, p.
701.
[16] T. Ito, K. Sugimoto, H. Ito, Chem. Lett. (1981) 1477.
[17] X. Bu, Z. Zhang, D. An, Y. Chen, M. Shionoya, E. Kimura,
Inorg. Chim. Acta 249 (1996) 125.
[18] L.L. Diaddari, W.R. Robinson, D.W. Margerum, Inorg. Chem.
22 (1983) 1021.
[19] M.F. Richardson, R.E. Sievers, J. Am. Chem. Soc. 94 (1972)
4134.