Insertion of unsaturations into the 1,4,7-triazacyclononane skeleton. Photochemical reaction between benzil and diethylenetriamine: Synthesis and properties of 2,3-diphenyl-1,4,7-triazacyclonona-1,3-diene

Article abstract of DOI:10.1039/b003967p

Photochemical reaction between benzil and diethylenetriamine (dien) in anhydrous methanol in the presence of oxygen yields a novel triaza macrocycle 2,3-diphenyl-1,4,7-triazacyclonona-1,3-diene (L). It was characterised by FTIR, EI mass and NMR (¹

Full text of DOI:10.1039/b003967p

View Article Online / Journal Homepage / Table of Contents for this issue
Insertion of unsaturations into the 1,4,7-triazacyclononane skeleton.
Photochemical reaction between benzil and diethylenetriamine:
synthesis and properties of
2,3-diphenyl-1,4,7-triazacyclonona-1,3-diene¤
Prakriti Ranjan Bangal, Goutam Kumar Patra and Dipankar Datta*
Department of Inorganic Chemistry, Indian Association for the Cultivation of Science,
Calcutta 700 032, India. E-mail: icdd=mahendra.iacs.res.in
Received (in Montpellier, France) 17th May 2000, Accepted 3rd July 2000
Published on the Web 22nd August 2000
Photochemical reaction between benzil and diethylenetriamine (dien) in anhydrous methanol in the presence of
oxygen yields a novel triaza macrocycle 2,3-diphenyl-1,4,7-triazacyclonona-1,3-diene (L). It was characterised by
FTIR, EI mass and NMR (1H and 13C) spectra. A comparative study between the various physicochemical
properties of L and the product (L ; reported earlier by others) of the thermal reaction between benzil and dien is
1
made in detail. L and L can be regarded as isomers. While L cannot be converted to L by UV radiation, L is
1
1
transformed into L upon reÑuxing in anhydrous methanol. The macrocycle L a†ords cationic copper(I) complexes
1
of the type [CuL]X (X \ ClO ~ and PF ~). Cyclic voltammetry and coulometry at platinum electrodes in
4
6
dimethylformamide under an N atmosphere show that the CuII@I potential in [CuL]X is 0.25 V vs. SCE. This
2
indicates that L is a weak p-acid. L is photoluminescent in methanol at room temperature (emission maxima
j
\ 380 nm; quantum yield r \ 1.96 ] 10~2). Its emission when in the form of [CuL]X in methanol is quenched
em
considerably (j \ 370 nm; r \ 6.2È6.7 ] 10~3).
em
1,4,7-Triazacyclononane (tacn) is the simplest possible triaza
macrocyclic ligand. It was Ðrst synthesised by Koyama and
Yoshino in 1972.1 Later many workers, especially Chaudhuri
and Wieghardt, have developed its transition metal chemistry,
which has proved to be as diverse as interesting.2,3 The situ-
ation can be summarised by saying that it coordinates
““readily to metal ions producing stable complexes, many of
which have unusual chemical and physical properties.ÏÏ4 Sub-
sequently, a number of derivatives of tacn have been synthe-
sised to increase its dentition or to manipulate other
properties of it. Their chemistry has been more fascinating; for
some examples, see refs. 5È9. However, so far, no one has
reported on the introduction of unsaturation(s) into the basic
skeleton of tacn. With the expectation of generating some
interesting chemistry, we have tried to introduce two imino N
atoms in the tacn skeleton by reacting benzil with
diethylenetriamine (dien). The results of the Ðrst phase of our
experiments are reported here.
diene (L).10 This was veriÐed later by X-ray crystallography.11
Our AM1 calculations12 using standard MOPAC package
(version 1.10) indicate that L is energetically lower than L in
1
the gas phase by only 0.81 kcal mol~1. Because of this small
di†erence in the heats of formation of the two isomeric
species, we decided to investigate the photochemical
reaction between benzil and dien. We have found that when a
moderately dilute solution of benzil and dien, taken in equi-
molar proportions, in anhydrous methanol is irradiated with
UV light in the presence of air, L is obtained [eqn. (1)]. The
(1)
temperature at which the photochemical reaction is carried
out is very crucial. It should be kept as low as possible; with
rising temperature, the product is more and more contami-
nated with L and above 50 ¡C, only L is obtained. In the
Results and discussion
In 1992, Okawara et al. reported that reaction of benzil and
1
1
dien in equimolar proportion in ethanol yields 8,8a-diphenyl-
absence of oxygen, such as when reaction (1) is performed in
an inert atmosphere, L is not formed at all irrespective of the
1,2,3,5,6,8a-hexahydroimidazo[1,2-a]pyrazine (L ) and not its
1
macrocyclic isomer 2,3-diphenyl-1,4,7-triazacyclonona-1,3-
temperature; the product is solely L . Furthermore, when a
1
solution of L in anhydrous methanol is irradiated with UV
1
light, one gets back L only, that is L is not converted to L
1
1
upon irradiation. However, when L is reÑuxed in anhydrous
methanol (the reaction condition used to synthesise L ),13 L
1
1
is generated in moderate yield. These Ðndings are summarised
in Scheme 1. The role of oxygen in reaction (1) is, at present,
not understood; some radicals seem to be involved in the
¤ Electronic supplementary information (ESI) available: the
minimum energy structure of L as obtained by AM1 calculations
together with some metric parameters and atomic charges. The DEPT
process with the generation of 1O . On the other hand, the
2
thermal conversion of L to L appears to be a typical trans-
135 spectra of L É 1/3(CHCl ) and
L
in CDCl . See http://
1
3
1
3
www.rsc.org/suppdata/nj/b0/b003967p/
annular cyclisation14 catalysed by a proton (Scheme 2), which
DOI: 10.1039/b003967p
New J. Chem., 2000, 24, 719È723
719
This journal is ( The Royal Society of Chemistry and the Centre National de la Recherche ScientiÐque 2000

View Article Online
at least in solution as evidenced by the 13C NMR spectrum of
L in CDCl [Fig. 1(a)]. For comparison, the 13C NMR spec-
3
trum of L in CDCl is also reproduced in Fig. 1(b). The most
1
3
notable di†erence is that a quaternary alkyl C present in Fig.
1(b) at 82.08 ppm is absent in Fig. 1(a) as expected. Four
methylene C atoms are observed (43.47, 43.82, 48.03 and 51.00
Scheme 1
ppm) for L in Fig. 1(b), while for L only three such reso-
nances are observed (39.51, 45.75 and 48.40 ppm; two of the
1
comes from the protic solvent used. In our synthetic pro-
is actually isolated as L É 1/3(CHCl ) (see
Experimental). The presence of CHCl has been conÐrmed by
a qualitative test for chlorine, CHN analyses and 13C NMR
four methylene C atoms in L, possibly those bonded to the
amino N, are found to be magnetically equivalent). The two
imino C atoms of L resonate at 167.94 ppm, an expected posi-
tion;17 however, the two quaternary aromatic C atoms are
not magnetically equivalentÈthese appear separately at
131.31 and 142.07 ppm. In our AM1 calculations, the net
charges on these two quaternary aromatic C atoms of L are
not the same (viz., [0.090 and [0.049). In the DEPT
(dynamic enhancement by polarisation transfer) 135 spectrum
cedure,
L
3
3
(vide infra).
It is worth noting here that when dien is reacted with m-
phthaldehyde in CH CN at room temperature in equimolar
3
proportion, it yields a hexaaza macrocycle15 that has a com-
plicated structure in the solid state; this macrocycle reacts
with [Cu(CH CN) ]ClO in CH Cl to a†ord an orange, air-
[Fig. S2(a) of ESI], the presence of CHCl (at 77.47 ppm) can
3
4
4
2 2
3
sensitive dinuclear copper(I) complex that has not been char-
acterised structurally.15
be discerned very clearly.
There are two very signiÐcant di†erences in the chemical
In the IR spectra of L , the C2N stretching frequency
properties of L and L . (i) While L cannot be hydrogenated by
1
1
appears at 1629 cm~1 whereas this frequency in L is found to
NaBH in aqueous methanol at room temperature, under
4
be slightly higher in energy at 1643 cm~1. The C2N stretching
frequency in R ÈN2C(R )ÈC(R )2NÈR occurs typically16 at
the same experimental conditions L yields 2-(2,3-diphenyl-
1
piperazin-1-yl)ethylamine.13 (ii) L disintegrates into benzil
1
1
2
2
1
1640 cm~1. Thus the IR spectra of L, especially when com-
and dien when treated with an equimolar amount of H`
pared with that of L , clearly indicates the presence of a 1,4-
(added in the form of perchloric acid) in methanol,18 whereas
under the same conditions L remains intact.
1
diaza-1,3-diene function in L.
Our AM1 calculations predict that L is not symmetric in
the gas phase (Fig. S1 of ESI). This lack of symmetry prevails
Because of the presence of a 1,4-diaza-1,3-diene function,
the macrocycle L is expected to be a p-acid and hence capable
of preferentially stabilising the lower oxidation states of a
transition metal. To assess this, we have initially tried to syn-
thesise its copper(I) complexes. It is found that when L is
reacted with [Cu(CH CN) ]X (X \ ClO ~ and PF ~) in
3
4
4
6
equimolar proportions in anhydrous methanol under dry N
2
atmosphere, it a†ords brownish yellow copper(I) complexes of
the formulation [CuL]X (X \ ClO ~ and PF ~). These com-
Scheme 2
4
6
plexes are quite stable in air. In the solid state, they are stable
for about a week, after which they slowly become grey. They
are moderately soluble in methanol but dissolve well in
dimethylformamide (DMF) giving rise to a brownish yellow
colour; in solution these are stable for about an hour after
which they gradually turn light green. It is noted that reaction
of 1,4,7-trialkyl-tacn (tatacn) with [Cu(CH CN) ]X (X \
3
4
ClO ~ and PF ~) in tetrahydrofuran yields extremely air sen-
sitive copper(I) complexes of the type [Cu(tatacn)(CH CN)]X
where acetonitrile is coordinated to the metal giving rise to
4
6
3
CuIN chromophores.19 Evidently our macrocycle L stabilises
4
copper(I) much more than tacn against aerial oxidation.
The l(C2N) in [CuL]X appears at 1632È1633 cm~1, which
is lower in energy than that in the free ligand by 10È11 cm~1.
This lowering is presumably due to p-back bonding between
copper(I) and the 1,4-diaza-1,3-diene moiety in L. The l(NÈH)
in [CuL]X is also quite di†erent (3406È3437 cm~1) from that
(3311 cm~1) in the free macrocycle. These observations
support interaction of the metal with all three N atoms of L in
[CuL]X. In solution, the complexes [CuL]X behave as a 1 : 1
electrolytes, thereby excluding the possibility of anion coordi-
nation to the metal. Thus, a CuIN chromophore is deemed to
3
be present in [CuL]X (X \ ClO ~ and PF ~). This CuIN
4
6
3
chromophore is quite inert towards addition reactions. It does
not react with NO ~ or PPh in equimolar proportions in
2
3
anhydrous methanol under an N atmosphere to form
2
adducts. On the other hand, NO ~ and PPh react with
[Cu(tatacn)(CH CN)]X (X \ ClO ~ and PF ~) resulting in
2
3
3
4
6
substitution of the ligated acetonitrile moiety.19 It may be
noted that though we have not been able to isolate any four-
coordinate adduct of [CuL]X in the solid state, coordination
of solvents like acetonitrile, methanol or DMF in solution
cannot be ruled out as the cuprous ion in [CuL]` moiety is
quite exposed.
Fig. 1 300 MHz 13C NMR spectrum of (a) L É 1/3(CHCl ) and (b) L
3
1
in CDCl (reference, TMS). The resonances around 77 ppm are due to
3
the solvent. For the possible assignments of the various other reso-
nances, see text.
720
New J. Chem., 2000, 24, 719È723

View Article Online
potential in copper complexes is the ligand p-acidity.21h23
Incidentally, no redox potential data for the complexes of the
type [Cu(tatacn)(CH CN)]X have been reported;19 however,
3
considering their high sensitivity towards oxygen, the CuII@I
couple in these complexes is expected to occur at fairly nega-
tive potentials. L , the non-macrocyclic isomer of L, also gives
1
a cationic copper(I) complex (white in colour) of the formula-
tion [CuL ]ClO ,18 which at a platinum electrode in DMF
1
4
displays cyclic voltammograms characteristic of an adsorbed
species (E \ [0.02 V vs. SCE).24
1@2
Until now we have not been able to isolate any deÐnite
copper(II) complex of L. When L is reacted with Cu(ClO ) É
4 2
6H O in 2 : 1 molar proportion in methanol, a greenish para-
2
magnetic complex containing metal and with the ligand
moiety intact is obtained, but this could not be characterised.
Reaction of
is not a clean process. On the other hand, while reaction
of with Cu(ClO ) É 6H O is not clean,
L with Cu(CH COO) É H O in methanol
Fig. 2 Cyclic voltammogram of [CuL]PF in DMF at a platinum
3
2
2
6
electrode. Solute concentration, 1.23 mmol dm~3; supporting electro-
lyte, 0.1 mol dm~3 tetraethylammonium perchlorate; scan rate, 0.050
V s~1.
L
L
1
4 2
2
1
reacts Cu(CH COO) É H O
with
to
a†ord
blue
3
2
2
[Cu(dien)(CH COO)] (ClO ) .24 In contrast, the oxidised
3
n
4 n
species in eqn. (2) has a bright yellow colour in DMF with
absorption bands at 429 and 268 nm with e values (calculated
for [CuIIL]2`) of D700 and D8200 dm3 mol~1 cm~1, respec-
tively.
The electrochemical behaviour of [CuL]X (X \ ClO ~ and
4
PF ~) has been examined by cyclic voltammetry in DMF
6
under N atmosphere at a platinum electrode in the voltage
2
range [0.4 to 0.8 V vs. SCE (saturated calomel electrode). For
both the complexes, more or less identical (within experimen-
tal errors) quasi-reversible cyclic responses are obtained on
The macrocycle L shows a charge transfer band around 240
nm (e \ 2600 dm3 mol~1 cm~1) in its absorption spectrum in
methanol. This band is shifted to a slightly longer wavelength
in its copper(I) complexes; 250 nm (e \ 14 400 dm3 mol~1
the positive side of SCE (Fig. 2) with an average E of 0.25 V
1@2
(vs. SCE) and somewhat large peak-to-peak (*E ) separations;
p
*E increases from 0.2 V to 0.35 V as the scan rate is
p
cm~1) in [CuL]ClO and 246 nm (e \ 10 200 dm3 mol~1
4
cm~1) in [CuL]PF . Upon excitation at 300 nm (within the
increased from 0.05 V s~1 to 1 V s~1. Our coulometry at 0.6
6
charge transfer envelope) in methanol, the macrocycle L
V vs. SCE at a platinum wire gauge electrode in DMF under
shows a single broad structureless emission around 380 nm
(Fig. 3) with a quantum yield (r) of 1.96 ] 10~2 (vs. quinine
sulfate25 in 0.05 M H SO ). This emission is strongly
dry N atmosphere establishes that the electrode process is
2
oxidative and involves only one electron.” This implies that
the couple observed at 0.25 V vs. SCE is CuII@I [eqn. (2)]. So
2
4
quenched in [CuL]X in methanol with the emission maxima
slightly blue shifted to 370 nm; changing the anion has mar-
ginal e†ects on the Ñuorescence quantum yield [r for
[CuL]ClO is 6.7 ] 10~3 (Fig. 3) and that for [CuL]PF is
[CuIIL]2`e~ ¢ [CuIL]`
(2)
far, to our knowledge, there is only one report (by Karlin et
4
6
6.2 ] 10~3]. Interestingly, in [CuL ]ClO , the copper(I)
al.)20 of the observation of a CuII@I couple in a CuIN chromo-
1
4
3
complex of L , the Ñuorescence quantum yield of the Ñuoro-
phore L is increased eight-fold.18 For a complete compari-
son, it is noted that L has an emission maximum at 383 nm
phore, two of the three N atoms were pyridyl N and the other
1
a secondary amino N. In KarlinÏs complex, which is air-
sensitive, the CuII@I couple has a potential of [0.08 V vs. SCE
in DMF, which is signiÐcantly lower than that observed in
our copper(I) complexes. We attribute this di†erence to the
p-acidity of L, since at present it is quite well established that
one of the main factors responsible for increasing the CuII@I
1
1
in methanol with a r of 1.0 ] 10~3. While quenching of the
Ñuorescence of an organic Ñuorophore by transition metal
ions is a very common phenomenon as observed here,
[CuL ]ClO presents one of the rare cases18 where a tran-
1
4
sition metal ion induces enhancement of the Ñuorescence of an
organic Ñuorophore.
Concluding remarks
Here we have demonstrated that by reacting dien with benzil
in anhydrous methanol under UV radiation, two imino N
atoms can be introduced into the basic skeleton of 1,4,7-tri-
azacyclononane, resulting in a novel triaza macrocycle L. The
presence of oxygen is essential in the formation of L.
However, the mechanism by which oxygen brings about the
reaction is at present not clear. The physicochemical proper-
ties of L are very di†erent from those of L which is produced
1
by carrying out the reaction between benzil and dien ther-
mally. L and L can be regarded as isomers. The macrocycle
1
L can be thermally converted to the non-macrocyclic L . This
1
result is expected from our AM1 calculations, as they indicate
Fig. 3 Emission spectra of L (upper trace) and [CuL]ClO (lower
that L is energetically higher than L even if by only ca. 1 kcal
4
1
trace) in methanol at room temperature. Excitation wavelength, 300
mol~1.
nm. The absorbances of both the solutions at 300 nm are 0.58.
The magnitude of the CuII@I potential in [CuL]X (X \
ClO ~ and PF ~) shows that L stabilises copper(I) more than
4
6
copper(II) though not to any great extent. Thus, L can be
designated as a weak p-acid. It can be used to stabilise to
” [CuL]X (X \ ClO ~ and PF ~) samples (3È4 mg) were electro-
4
6
lysed. Average coulomb count corresponded to 1.07 electron.
New J. Chem., 2000, 24, 719È723
721

View Article Online
some extent low oxidation state(s) of a transition metal. The
p-acidity in L arises because of the presence of the 1,4-diaza-
1,3-diene function in L. The p-acid-like behaviour of L is also
apparent from the fact that the C2N stretching frequencies in
the copper(I) complexes are less than that in the free macro-
cycle by 10È11 cm~1. In any case, it is clear from our studies
that L stabilises copper(I) much more than tacn.
pound. Yield: 1.12 g (28%); mp 70 ¡C; found: C, 69.48; H,
6.15; N, 13.19. C
H
N Cl requires C, 69.40; H, 6.15;
18.333 19.333
3
N, 13.25%. EI MS: m/z 277 (L`, 10%). FTIR/cm~1 (KBr):
3311m (NÈH), 1643vs (C2N). 1H NMR (300 MHz, CDCl ,
3
TMS): 2.73È4.09 (10, ÈCH È 8H); 7.18È7.67 (several br, m
2
10.333H); NH not observed. UV-VIS
mol~1cm~1) (CH OH): 240 (2600).
j
/nm (e/dm3
max
3
It is appropriate to mention here that the thermal reaction
Conversion of L to L . L É 1/3(CHCl ) (1.0 g, 3.2 mmol) was
between dien and glyoxal gives rise to L , which is quite dif-
1
3
2
ferent from L in nature.26 Hence, our studies have been
1
reÑuxed in 10 ml of anhydrous methanol for 6 h. Then it was
left in air for 2 h. White crystals of L deposited, which were
restricted here to the reaction of dien with benzil only.
1
Ðltered o†, washed with 2 ml of cold methanol and dried in
However, at present we are engaged in investigating the reac-
tions of dien with phenanthrequinone.
vacuo over fused CaCl . Yield, 0.22 g (24%); mp 142 ¡C.
2
[CuL]ClO . Solid [Cu(CH CN) ]ClO (0.33 g, 1 mmol)
4
3
4
4
was added to 0.32 g (1 mmol) of L É 1/3 (CHCl ) dissolved in
3
25 ml of anhydrous, degassed methanol under a dry N atmo-
2
sphere and stirred for 3 h. The brownish yellow compound
precipitated was Ðltered o†, washed with 2 ml of methanol
and dried in vacuo over fused CaCl . Yield, 0.19 g (43%);
2
found: C, 49.04; H, 4.41; N, 9.49; Cu, 14.42.
Experimental
C
H
N CuClO requires C, 49.07; H, 4.35; N, 9.54; Cu,
18 19
3
4
14.43%. FTIR/cm~1 (KBr): 3406br (NÈH), 1633s (C2N),
General
1098vs, 623m (ClO ). K (CH OH): 92 )~1 cm2 mol~1 (1 : 1
4
M
3
[Cu(CH CN) ]ClO and [Cu(CH CN) ]PF were synthe-
electrolyte). UV-VIS j /nm (e/dm3 mol~1 cm~1) (CH OH):
3
4
4
3
4
6
max
3
max
sised by literature methods.27,28 Fresh analytical reagent
grade DMF (purchased from S. D. Fine-Chem Ltd., India)
was used directly for electrochemistry. All other reagents were
procured commercially. Copper was estimated gravimetrically
as CuSCN. Microanalyses were performed using a PerkinÈ
Elmer 2400II elemental analyser. Molar conductance was
measured by a Systronics (India) direct reading conductivity
meter (model 304). Melting points were determined by a
melting point apparatus procured from CBC Power System
(Calcutta, India) and are uncorrected. IR spectra (KBr disc)
were recorded on a Nicolet Magna-IR spectrophotometer
(Series II), UV-VIS spectra on a Shimadzu UV-160A spectro-
photometer, 1H and 13C NMR spectra by a Brucker DPX300
spectrometer and EI (electron impact) mass spectrum on an
AEI MS 30 instrument. All the photoluminescnce studies were
carried out in air using a Spex Fluorolog spectroÑuorimeter.
Cyclic voltammetry and coulometry were performed using
EG&G PARC electrochemical analysis system (model 250/5/
0) in DMF under a dry nitrogen atmosphere in conventional
three-electrode conÐgurations with tetraethylammonium per-
chlorate as the supporting electrolyte. An ECDA-Pt02 plati-
num disk electrode procured from Con-Serv Enterprises, India
was used as the working electrode in cyclic voltammetry.
Under the experimental conditions employed here, the
ferrocene-ferrocenium couple appears at 0.402 V vs. SCE with
355 sh (2050), 250 (14 400) and 218 (21 600). UV-VIS j /nm
(Nujol mull): 350 and 285.
[CuL]PF . This was synthesised by a procedure exactly
similar to that described for [CuL]ClO by starting with 0.37
g (1 mmol) of solid [Cu(CH CN) ]PF . Yield, 0.17 g (35%);
found: C, 44.51; H, 3.90; N, 8.68; Cu, 13.09. C
requires C, 44.48; H, 3.95; N, 8.65; Cu, 13.08%. FTIR/cm~1
(KBr): 3437br (NÈH), 1632s (C2N), 841vs (PF ). K -
6
4
6
3
4
H
N CuPF
18 19
3
6
6
M
(CH OH): 110 )~1 cm2 mol~1 (1 : 1 electrolyte). UV-VIS
3
j
/nm (e/dm3 mol~1 cm~1) (CH OH): 350sh (1200), 246
max
3
(10 200) and 211 (21 000). UV-VIS j /nm (Nujol mull): 352,
max
290, 255.
Caution! Although we have not encountered any problems in
handling the perchlorates, it is noted that perchlorate salts of
metal complexes with organic ligands are potentially explosive
and should be handled only in small quantities with appropri-
ate precautions.
Acknowledgements
Financial support received from the Department of Science
and Technology, New Delhi, India is gratefully acknowledged.
Thanks are due to Prof. S. Lahiri of the Department of
Organic Chemistry, IACS for allowing us to use the photo-
chemical reactor.
an *E of 95 mV at a scan rate of 0.050 V s~1.
p
Syntheses
References
1
2
H. Koyama and T. Yoshino, Bull. Chem. Soc. Jpn., 1972, 45, 481.
P. Chaudhuri and K. Wieghardt, Prog. Inorg. Chem., 1987, 35,
329 and refs. therein.
L Æ 1/3(CHCl ). A solution of 2.68 g (12.8 mmol) of benzil
3
and 1.4 ml (12.8 mmol) of diethylenetriamine in 110 ml of
anhydrous methanol was irradiated in a thermostated Pyrex
vessel using a Hanovia 450W mercury lamp for 6 h (the tem-
perature of the reaction mixture was below 18 ¡C). Then the
solvent was removed under reduced pressure to obtain a red
viscous liquid, which was loaded on a silica gel (60È120 mesh)
column (30 ] 2.5 cm) and eluted successively with chloroform
and methanol. The chloroform fraction, after work-up, a†ord-
3
4
S. W. Golding, T. W. Hambley, G. A. Lawrance, S. M. Luther,
M. Maeder and P. Turner, J. Chem. Soc., Dalton T rans., 1999,
1975 and refs. therein.
G. A. Melson, Coordination Chemistry of Macrocyclic Com-
pounds, Plenum Press, New York, 1979, section 2.2.
D. Zhang and D. H. Busch, Inorg. Chem., 1994, 33, 5138.
W. B. Tolman, Acc. Chem. Res., 1997, 30, 227.
J. Muller, T. Weyhermuller, E. Bill, P. Hildebrandt, L. Ould-
Moussa, T. Glaser and K. Wieghardt, Angew. Chem., Int. Ed.,
1998, 37, 616.
C. Krebs, T. Glaser, E. Bill, T. Weyhermuller, W. MeyerÈKlaucke
and K. Wieghardt, Angew. Chem., Int. Ed., 1999, 38, 359.
D. Burdinski, K. Wieghardt and S. Steenken, J. Am. Chem. Soc.,
1999, 121, 10781.
5
6
7
ed white crystals of L (1.16 g, 33%). From the methanol frac-
1
tion, evaporation of the solvent under reduced pressure at
8
9
room temperature yielded a thick red liquid, which was dis-
solved in 25 ml of chloroform and washed thoroughly with
distilled water (75 ml ] 3). The reddish yellow aqueous layer
was discarded. The yellow chloroform layer was evaporated at
room temperature under reduced pressure to D5 ml and then
10 T. Okawara, K. Uchiyama, Y. Okamoto, T. Yamasaki and M.
Furukawa, J. Chem. Res. (S), 1992, 264; J. Chem. Res. (M), 1992,
2035.
kept in vacuo over fused CaCl to obtain a Ñu†y yellow com-
2
722
New J. Chem., 2000, 24, 719È723

View Article Online
11 D. S. Parihar, R. Bohra, R. N. Prasad and P. N. Nagar, Acta
Crystallogr., Sect. C, 1995, 51, 482.
19 J. A. Halfen, S. Mahapatra, E. C. Wilkinson, A. J. Gengenbach,
V. G. Young, Jr., L. Que, Jr. and W. B. Tolman, J. Am. Chem.
Soc., 1996, 118, 763.
12 M. J. S. Dewar, E. G. Zoebisch, E. F. Healy and J. J. P. Stewart,
J. Am. Chem. Soc., 1985, 107, 3902.
20 K. D. Karlin, Y. Gultneh, J. P. Hutchinson and J. Zubieta, J. Am.
Chem. Soc., 1982, 104, 5240.
13 S. B. Majumdar, M. Mukherjee, G. K. Patra, D. Datta and M.
Helliwell, Acta Crystallogr., Sect. C, 1999, 55, 668.
14 D. C. Harrowven and G. Pattinden, in Comprehensive Organic
Synthesis, ed. B. M. Trost, Pergamon, Oxford, 1991, vol. 3, ch.
1.10.
21 S. Chowdhury, G. K. Patra, M. G. B. Drew, N. Chattopadhyay
and D. Datta, J. Chem. Soc., Dalton T rans., 2000, 235 and refs.
therein.
22 J. P. Naskar and D. Datta, Indian J. Chem., Sect. A, 2000, 39,
15 R. Menif, A. E. Martell, P. J. Squattrito and A. ClearÐeld, Inorg.
378.
Chem., 1990, 29, 4723.
23 G. K. Patra, S. Samajdar and D. Datta, J. Chem. Soc., Dalton
T rans., 2000, 1555.
16 H. t. Dieck and J. Dietrich, Chem. Ber., 1984, 117, 694 and refs.
therein.
24 G. K. Patra and D. Datta, unpublished results.
25 J. N. Demas and G. A. Crosby, J. Phys. Chem., 1971, 75, 991.
26 C. A. Mirkin, J. E. Premecz, M. E. Ford and T. A. Johnson, J.
Heterocycl. Chem., 1993, 30, 839.
27 P. Hemmerich and C. Sigwart, Experientia, 1963, 19, 488.
28 G. J. Kubas, Inorg. Synth., 1979, 19, 90.
17 D. M. Haddleton, D. J. Duncalf, D. Kukulj, M. C. Crossman, S.
G. Jackson, S. A. F. Bon, A. J. Clark and A. J. Shooter, Eur. J.
Inorg. Chem., 1998, 1799.
18 P. Purkayastha, N. Chattopadhyay, G. K. Patra and D. Datta,
Indian J. Chem., Sect. A, 2000, 39, 375 and refs. therein.
New J. Chem., 2000, 24, 719È723
723