Preparation of the copper(n) complex of the binucleating hexaaza macrocycle 2,5,8,17,20,23-hexaaza[9.9]paracyclophane and its solution chemistry, hydrolytic activity and acid dissociation kinetics

Article abstract of DOI:10.1039/b006199i

The binucleating hexaaza macrocycle 2,5,8,17,20,23-hexaaza[9.9]paracyclophane (PEA) and its N-permethylated derivative 2,5,8,17,20,23-hexamethyl-2,5,8,17,20,23-hexaaza[9.9]paracyclophane (Me₆PEA) have been prepared. The copper(II) complex of PEA has been prepared and the stability constants of the copper complexes determined in aqueous solution by potentiometric and spectrophotometric methods. The protonation sites of PEA have been identified by NMR titration. The crystal structure of the complex [Cu₂(PEA)Cl₃]⁺Cl-·2.6MeCN ·2.9H₂O has been determined. The copper centres have an N3C12 donor set with a square pyramidal geometry. The Cu ... Cu distance is 7.02 A. The copper complex catalyses hydrolysis of the phosphotriester 2,4-dinitrophenyl diethyl phosphate (DNPDEP) and the [Cu₂(PEA)(OH)₂]² and [Cu₂(PEA)(OH)]³⁺ complexes have been identified as the active species. The acid catalysed dissociation of the copper complex has been studied by stopped-flow methods. The reaction is monophasic and displays saturation kinetics at high acidities. The Royal Society of Chemistry 2001.

Full text of DOI:10.1039/b006199i

View Article Online / Journal Homepage / Table of Contents for this issue
Preparation of the copper(II) complex of the binucleating hexaaza
macrocycle 2,5,8,17,20,23-hexaaza[9.9]paracyclophane and its
solution chemistry, hydrolytic activity and acid dissociation
kinetics‡
Thomas Clifford,a* Andrew M. Danby,a Philip Lightfoot,b David T. Richensb and
(the late) Robert W. Hayb †
a Department of Chemistry, University of Kansas, Malott Hall, Lawrence, Kansas 66045, USA
b School of Chemistry, University of St. Andrews, St. Andrews, Fife, UK KY16 9ST
Received 1st August 2000, Accepted 21st November 2000
First published as an Advance Article on the web 11th January 2001
The binucleating hexaaza macrocycle 2,5,8,17,20,23-hexaaza[9.9]paracyclophane (PEA) and its N-permethylated
derivative 2,5,8,17,20,23-hexamethyl-2,5,8,17,20,23-hexaaza[9.9]paracyclophane (Me6PEA) have been prepared.
The copper() complex of PEA has been prepared and the stability constants of the copper complexes determined
in aqueous solution by potentiometric and spectrophotometric methods. The protonation sites of PEA have been
identified by NMR titration. The crystal structure of the complex [Cu2(PEA)Cl3]ϩClϪؒ2.6MeCNؒ2.9H2O has been
determined. The copper centres have an N3Cl2 donor set with a square pyramidal geometry. The Cu ؒ ؒ ؒ Cu distance
is 7.02 Å. The copper complex catalyses hydrolysis of the phosphotriester 2,4-dinitrophenyl diethyl phosphate
(DNPDEP) and the [Cu2(PEA)(OH)2]2ϩ and [Cu2(PEA)(OH)]3ϩ complexes have been identified as the active species.
The acid catalysed dissociation of the copper complex has been studied by stopped-flow methods. The reaction is
monophasic and displays saturation kinetics at high acidities.
Introduction
Experimental
There has been considerable interest in the chemistry of bi-
nucleating macrocycles capable of incorporating two metal
centres and the topic has been the subject of several recent
reviews.1 In part, this interest has arisen due to the occurrence
of binuclear metal centres in enzymes such as red kidney
bean purple acid phosphatase (see Fig. 1)2 and urease.3 Metal
complexes of some binucleating hexaaza macrocycles form
host–guest complexes with a range of anions such as maleate,
pyrophosphate and triphosphate.4 In addition they can display
catalytic activity in the hydrolysis of phosphate derivatives5 and
peptides.6 For these reasons we have been interested in the
coordination chemistry of large ring macrocycles capable of
incorporating two metal centres. There have been many studies
of the acid catalysed dissociation of mononuclear complexes7
but very few with binucleating macrocycles. Fernández-Trujillo
et al.8 have recently studied the acid catalysed dissociation of
the binuclear copper() complex of the related m-xylyl bridged
ligand, MEA (see Fig. 1).
Characterisation
Infrared spectra were recorded on a Perkin-Elmer 1710 Infra-
red Fourier Transform Spectrometer as KBr pellets. Melting
points were obtained using a Gallenkamp melting point
apparatus. Conductivities were measured in a Vernier con-
ductivity probe and meter the method of calibration is dis-
cussed elsewhere.9 UV/Visible spectra were recorded on a
Lambda 14P Spectrometer, NMR spectra using a 200 MHz
Varian Gemini, in D2O (reference sodium 4,4-dimethyl-4-
silapentane-1-sulfonate) or CDCl3 (reference TMS).
Materials
High purity potassium chloride, 3-azapentane-1,5-diamine
(dien), copper salts, and sodium tetrahydroborate were
obtained from Aldrich and used without further purification.
Terephthalaldehyde was obtained from Avocado.
In the present paper we describe the synthesis of the ligand
PEA (Fig. 1) and its N-permethylated derivative and the
characterisation of the copper complex of PEA. We have also
determined formation constants for the copper() complexes by
potentiometric and spectrophotometric methods and identified
the major active species in the hydrolysis of 2,4-dinitrophenyl-
phosphate (DNPDEP, Fig. 1) as the dihydroxy binuclear copper
complex of PEA ([Cu2(PEA)(OH)2]2ϩ). The crystal structure of
the binuclear complex [Cu2(PEA)Cl3]ϩClϪؒ2.6MeCNؒ2.9H2O
has been determined. The kinetics of the acid catalysed dissoci-
ation has been studied by stopped flow techniques.
Synthesis
The ligand was prepared by a non-template [2ϩ2] condensation
of dien with terephthalaldehyde to form a tetramine followed
by a NaBH4 reduction.
2,5,8,17,20,23-Hexaaza[9.9]paracyclophane (PEA) I. Under
a nitrogen atmosphere, a filtered acetonitrile solution (50 cm3)
of terephthalaldehyde (3 g, 22.4 mmol) was added dropwise
(3 h) to a rapidly stirred acetonitrile solution (100 cm3) of dien
(2.93 g, 22.4 mmol). After approximately half of the dialdehyde
had been added a fine white precipitate formed. The mixture
was left stirring overnight before the solid was collected by
filtration and dried at reduced pressure at 50 ЊC (4.15 g, 80%).
The tetramine Schiff base could be recrystallised from toluene
or ethyl acetate (if the temperature did not exceed 70 ЊC) but the
crude product could be used directly in the reduction step.
† Deceased 8th January 1999 and submitted in his memory.
‡ Electronic supplementary information (ESI) available: 2-D NOESY
NMR spectra, experimental details of the potentiometric, spectro-
photometric and crystallographic studies. See http://www.rsc.org/
suppdata/dt/b0/b006199i/
240
J. Chem. Soc., Dalton Trans., 2001, 240–246
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b006199i

View Article Online
(10 cm3) then extracted with dichloromethane (3 × 20 cm3).
The combined extracts were dried over MgSO4. Removal of
the solvent at reduced pressure gave the crude product as a
pale yellow oil. This was dissolved in ethanol (ca. 20 cm3) and
concentrated HCl (37%) added until the solution became
turbid. The white precipitate formed was filtered off, washed
first with PriOH and then diethyl ether. The product was
recrystallised from aqueous methanol and dried at reduced
pressure (2.42 g, 91%) (Found: C, 48.80; H, 7.96; N, 11.26%.
Calc. for C30H59Cl6N6O1.5: C, 48.66; H, 8.03; N, 11.35%).
νmax/cmϪ1(KBr):3453s broad (O–H str.); 2975s (C–H str.); 2663,
2380s broad (Nϩ–H str.); 1664s (Nϩ–H bend); 1537, 1474, 1439,
1230 (C–N str.); 1113, 947, 791, 567 (KBr).
[Cu2(PEA)Cl4]ؒ3H2OؒMeOH 1. Copper() chloride dihy-
drate (0.832 g, 4.88 mmol) was dissolved in ethanol (20 cm3)
and added dropwise to a boiling solution ethanol solution (20
cm3) of compound I (1.0 g, 2.44 mmol). After half the copper
dichloride had been added to the rapidly stirred amine solution,
a fine pale blue-green precipitate formed. This was filtered off,
washed with ethanol (3 × 10 cm3) and diethyl ether (1 × 10 cm3)
and dried at reduced pressure at 50 ЊC. The complex was further
purified by dissolving it in aqueous methanol (ca. 50%) then
adding an equal volume of acetonitrile to give blue plate like
crystals which were washed with ethanol and air dried (1.65 g,
88%) (Found: C, 39.10; H, 5.52; N, 11.06%. Calc. for C25H48Cl4-
Cu2N6O4: C, 39.22; H, 6.32; N, 10.98%). νmax/cmϪ1: 3404s (O–H
str.), 3182s (N–H str.); 2879s (C–H str.); 1617m (N–H bend);
1520, 1451, 1065, 1019 (C–N str.); 871, 837, 803 (CH2 rock);
656, 478 (KBr). λmax/nm (εmax/dm3 molϪ1 cmϪ1) 617(238), water
ΛM /S cm2 molϪ1 391 (water).
Potentiometric titrations
Potentiometric titrations were carried out using a Radiometer
ABU9 Autoburette dispensing sodium hydroxide solution
(0.1 mol dmϪ3) into a ligand solution (ca. 1 × 10Ϫ3 mol dmϪ3
)
also containing appropriate amounts of copper() (approxi-
mately 1:1 or 1:2 ligand to copper ratios or in absence of
copper()) maintained at 25.0 ЊC. The ionic strength was
adjusted to 0.1 mol dmϪ3 with potassium chloride as the
protonated forms of the ligand gave insoluble nitrate salts.
The titrations were carried out under nitrogen throughout. The
pH was measured with an Orion Ross combination electrode
calibrated with borate and phthalate buffers. A value of 10Ϫ13.86
mol2 dmϪ6 for Kw was assumed. NaOH (prepared from dilution
of a Rhone Poulenc Volucon ampule) was standardised with
potassium hydrogenphthalate. Carbonate contamination of
the NaOH was checked by Gran’s method (> 0.5%). Hydro-
chloric acid solutions were standardised against standard
sodium hydroxide and the copper() nitrate stock solution by
EDTA complexometric titration. All sample solutions were
prepared with freshly boiled out water cooled in a nitrogen
stream. Plots of pH versus the volume of standard sodium
hydroxide added were collected on a Radiometer VIT90 and
passed to an Elonex 386 PC. The methods used to pick the
species model have been described elsewhere.10 Equilibrium
constants were calculated using HYPERQUAD.11 Eight curves
with a total of 327 points were used in calculation of the
protonation constants and six curves with a total of 205 points
in calculation of the copper complexation constants.
Fig. 1 2,5,8,17,20,23-hexaaza[9.9]paracyclophane (PEA) I, 2,4-di-
nitrophenyl diethyl phosphate (DNPDEP), red kidney bean purple
acid phosphatase.
The Schiff base (4.15 g, 8.96 mmol) was dissolved in ethanol
(100 cm3) and cooled in an ice bath. With rapid stirring, NaBH4
(6 g) was added in 0.5 g quantities. The mixture was left stirring
overnight, then acidified with concentrated hydrochloric acid
(until pH was less than 2) and ethanol added (30 cm3). The
white precipitate of crude PEA was filtered off and dried at
50 ЊC at reduced pressure. The solid was redissolved in water
(100 cm3) and basified with sodium hydroxide (a white precipi-
tate formed immediately). The mixture was extracted with
dichloromethane (4 × 20 cm3), the extracts were combined and
dried over anhydrous sodium sulfate. The solid residue, after
removal of solvent, was recrystallised from hot (70 ЊC) toluene
giving fine white needles of compound I (3.42 g, 82%) (Found:
C, 70.21; H, 9.35; N, 20.44%. Calc. for C12H19N3: C, 70.20;
H, 9.33; N, 20.47%). νmax/cmϪ1: 3340s, broad(O–H str.); 3293,
3227, 2942, 2901s(N–H str.); 2873, 2800, 2731s (C–H str);
1503, 1471(N–H bend); 1447, 1434, 1139, 1122, 1113, 1018(C–N
bend); 998, 969, 945, 926, 817, 800, 485(KBr).
NMR titrations
Deuteriated compounds were obtained from Aldrich. NMR
spectra were obtained in D2O, the required pH being adjusted
with NaOD (30% w/w in D2O) or DCl (37% in D2O). Quoted
13C chemical shifts are relative to sodium 4,4-dimethyl-4-
silapentane-1-sulfonate (Aldrich) dissolved in D2O and sealed
into a capillary. The ligand concentration was approximately
0.01 mol dmϪ3. Spectra were recorded on a Varian Gemini
2,5,8,17,20,23-Hexamethyl-2,5,8,17,20,23-hexaaza[9.9]para-
cyclophane hydrochloride hydrate (1/6/1.5) (Me6PEAؒ6HClؒ
1.5H2O) II. Formic acid (8 cm3, 98%), formaldehyde (7 cm3,
35% aqueous solution) and compound I (1.52 g, 3.21 mmol)
were refluxed under argon for 18 hours. The solution was
cooled and basified with sodium hydroxide (9 g) in water
J. Chem. Soc., Dalton Trans., 2001, 240–246
241

View Article Online
200 MHz Spectrometer. 2-D spectra (HMQC, heteronuclear
multiple quantum correlation, and NOESY) were obtained on
a Bruker 500 MHz spectrometer.
c = 25.32(3) Å, U = 16012(26) Å3, Z = 16, µ(Mo-Kα) = 13.68
cmϪ1, 5676 reflections measured, 3195 independent (Rint
0.0164), 2306 observed (I > 2σ[I]), wR2 = 0.3101, R1 = 0.1011.
=
Structure analysis and refinement. Structure solution and
refinement were carried out using SHELXTL.12 Cu, Cl and
N atoms were refined anisotropically, C and O isotropically.
Hydrogens were included but not refined. The rather large R
factor reflects the samples weak scattering of X-rays. Intensity
data were collected only to 38Њ in 2θ. Considerable difficulty
was also encountered in modeling the disordered solvent
present in the structure. Unfortunately the ease with which
solvent was lost from the crystals prevented determination
of the degree of solvent of crystallisation by microanalysis of
fresh samples of the complex. A freshly recrystallised sample
was found to give a stable weight after 15 min. Microanalysis
of this sample indicated the presence of at least three water
molecules. IR spectra of fresh samples showed no CN stretch
attributable to acetonitrile however acetonitrile had to be
included in the structure to give a reasonable fit. Three
acetonitrile and one water site were disordered and modeled
in two orientations with refined occupancies of 0.40(2) for
N(4S)–C(6S) and O(4) and 0.60(2) for N(4SЈ)–C(6SЈ) and
N(7S)–C(9S). Restraints on the displacement parameters of
the light atoms in the cation were required for the refinement to
achieve convergence.
Spectrophotometric titrations
Under approximately the same conditions as for the potentio-
metric titrations, a visible spectrum (wavelength range 850–400
nm) was taken after each aliquot of standard sodium hydroxide
(0.1 mol dmϪ3) was added. Spectra taken at 28 Hϩ concen-
trations (26 wavelengths) in duplicate, giving a total of 1456
data points, were used in the calculation. The data were
processed using the HYPERQUAD suite of programs. Details
of the computation of the copper() binding constants with
PEA were carried out by procedures described in detail
elsewhere.11
Kinetic measurements
Phosphate ester hydrolysis. The hydrolysis of the triester
DNPDEP was measured spectrophotometrically by following
the production of 2,4-dinitrophenolate at 360 nm with a
Lambda 14P Spectrometer. The temperature was maintained at
25 0.1 ЊC by a Perkin-Elmer Peltier System. Solutions at
various pH were prepared by making up solutions 4.65 × 10Ϫ3
mol dmϪ3 with respect to CuII and 2.52 × 10Ϫ3 mol dmϪ3 with
respect to PEA with the buffers MES (2-morpholinoethane-
sulfonic acid) (pH 5.0–6.0), HEPES (6.5–7.5), TAPS {3-[tris-
(hydroxymethyl)methylamino]propanesulfonic acid} (8.0–9.0),
CHES [2-(N-cyclohexylamino)ethanesulfonic acid] (9.5) and
CAPS [3-(cyclohexylamino-1-propanesulfonic acid] (10.5) at
0.05 mol dmϪ3 and the pH noted immediately before initiation
of the reaction using a glass electrode. Ionic strength was
adjusted to 0.1 mol dmϪ3 with KCl. The reaction was initiated
by the addition of a 2 µl acetonitrile stock solution of DNPDEP
to a 2 cm3 sample of the solution in a 1 cm path length plastic
cuvette. The final DNPDEP concentration was 2 × 10Ϫ4 mol
dmϪ3. For slow reactions (below pH 7) data were collected over
one half-life only, then the infinity value was obtained after two
weeks at room temperature. Above pH 7 data could be collected
up to three half-lives. The absorbance vs. time curves were fitted
by a first order rate equation using a linear least squares method
(quoted rates are the averages of two runs). Duplicate runs
showed that the rate constants could be replicated to within 5%.
CCDC reference number 186/2281.
See http://www.rsc.org/suppdata/dt/b0/b006199i/ for crystal-
lographic files in .cif format.
Results and discussion
Synthesis
Reaction of dien with tert-phthalaldehyde in acetonitrile gives
the tetraimine in reasonable yield and purity to allow NaBH4
reduction to the amine without further purification. Although
this [2ϩ2] addition reaction has found great utility in recent
years to prepare large macrocycles, Pietraszkiewicz and
Gasiorowski13 have reported that the reaction in THF results
in formation of polymeric material, indicating that these
reactions are very solvent dependant. The amine is simply
purified by recrystallisation from toluene to give analytically
pure crystalline material. The amine can also fully be methy-
lated by the standard Eschweiler–Clarke14 procedure which is a
distinct improvement, in terms of yield and ease of synthesis,
on the halogenoalkane [1 ϩ 1] cyclisation route of Paoletti and
co-workers.15
With copper(), PEA forms a binuclear complex, but even
after prolonged drying at reduced pressure the microanalytical
data showed the presence of solvent of crystallisation. The
presence of solvent has been noted in similar complexes.16 The
copper() complex has a molar conductivity greater than 300 S
cm2 molϪ1 which, although rather low for a 1:4 electrolyte,17
may be due to the bulky nature of the complex, as conductivity
will depend upon the ionic mobility. Low ionic mobility has
been used to explain the unusually low conductivities of tetra-
phenylborate salts.18 The visible spectrum of the complex
(absorption coefficient 119 dm3 molϪ1 cmϪ1 per copper centre)
is consistent with a d9 five-co-ordinate metal ion.
Acid hydrolysis of the [Cu2(PEA)Cl4]ؒ3H2OؒMeOH complex.
The acid catalysed dissociation of the copper() complex
of PEA occurred rapidly and was measured spectophoto-
metrically at 640 nm using a Hi-Tech SF-51 stopped-flow
spectophotometer. Hydrochloric acid solutions (Volucon
standard concentrate) adjusted to I = 0.1 mol dmϪ3 were used.
Pseudo first order conditions were maintained by using at least
a tenfold excess of acid. Values of kobs were obtained by fitting
the experimental data by a single exponential and quoted
kobs are the averages of at least 3 runs. Rapid scan spectro-
photometry was carried out using a Spectrascan rotating
monochromator accessory with the Hi-Tech SF-51 stopped
flow instrument. Temperature was maintained to 0.1 ЊC using
a Heto thermostatted bath of circulating water.
X-Ray determination of the crystal structure of [Cu2(PEA)Cl3]؉-
Cl؊ 2.6MeCNؒ2.9H2O
A single crystal of [Cu2(PEA)Cl3]ϩClϪؒ2.6MeCNؒ2.9H2O was
grown by diffusion of acetonitrile into an aqueous methanolic
solution of the complex to give well formed dark blue blocks.
The crystals were efflorescent so the experiment was run at low
temperature (120 K).
Description of the crystal structure of [Cu2(PEA)Cl3]؉Cl؊ؒ
2.6MeCNؒ2.9H2O
Assorted bond lengths and angles are collected in Table 1. The
copper centres have an N3Cl donor set, with square pyramidal
geometry. Three nitrogens from the macrocycle and one of the
chlorides make up the base of the pyramid with the copper
somewhat displaced above the plane towards the apical chloride
(as is commonly found with square pyramidal complexes). The
apical chloride is directed away from the macrocyclic cavity
Crystal data. C29.19H51.60Cl4Cu2N8.60O2.90, M = 838.37, ortho-
rhombic, space group Ibca (no. 73), a = 14.971(8), b = 42.24(4),
242
J. Chem. Soc., Dalton Trans., 2001, 240–246

View Article Online
Table 1 Assorted bond lengths (Å), angles (Њ) and Cu ؒ ؒ ؒ Cu (Å)
separation for [Cu2(PEA)Cl3]ϩClϪؒ2.6MeCNؒ2.9H2O
Cu(1)–Cl(2)
Cu(1)–Cl(1)
Cu(1)–N(1)
Cu(1)–N(2)
Cu(1)–N(3)
Cu(2)–Cl(4)
Cu(2)–Cl(3)
Cu(2)–N(4)
Cu(2)–N(5)
Cu(2)–N(6)
Cu(1) ؒ ؒ ؒ Cu(2)
2.731(5)
2.241(5)
2.03(2)
1.99(2)
2.02(2)
2.56(2)
2.270(8)
2.01(2)
2.00(2)
2.06(2)
7.02
Cl(1)–Cu(1)–Cl(2)
N(1)–Cu(1)–Cl(2)
N(2)–Cu(1)–Cl(2)
N(3)–Cu(1)–Cl(2)
N(1)–Cu(1)–N(2)
N(1)–Cu(1)–N(3)
Cl(3)–Cu(2)–Cl(4)
N(6)–Cu(2)–Cl(3)
N(5)–Cu(2)–Cl(3)
N(4)–Cu(2)–Cl(3)
N(6)–Cu(2)–N(5)
N(6)–Cu(2)–N(4)
96.6(2)
88.8(5)
90.7(5)
87.9(5)
85.5(7)
83.8(6)
104.2(3)
97.7(7)
171.7(6)
95.2(6)
83.9(9)
168.0(9)
Fig. 3 % speciation curves of metal complexes in the presence of
two equivalents of copper().
Fig. 2 AnORTEP19 diagramof[Cu2(PEA)Cl3]ϩClϪؒ2.6MeCNؒ2.9H2O
showing the atom labeling scheme and ball-and-stick diagram showing
chloro-bridging. Solvent molecules not shown. Thermal ellipsoids are
at 50% probability.
to form a bridge to a copper center in an adjacent macrocycle.
One of the bridging chlorides is located on a crystallographic
2-fold axis and the other just off this axis. The bridging occurs
throughout the structure to form chains of chloro-bridged
complexes running parallel to the b axis (Fig. 2). The copper to
nitrogen distances are not unusual (the mean distance found for
five-co-ordinate secondary amine complexes in the Daresbury
database is 2.029 Å compared to 2.01(2) Å for this complex).
The copper to chloride distance is normal for the basal chloride
but that to the apical chloride is a little longer than expected
and may be due to a small degree of electron withdrawal in
bridge formation. The Cu ؒ ؒ ؒ Cu distance is 7.02 Å.
Fig. 4 Variation of 13C NMR chemical shifts of PEA resonances with
pH, and assignments.
The complex has C2 symmetry with the reflection plane
bisecting the Cu ؒ ؒ ؒ Cu axis. The nitrogen donor ends of the
ligand are folded back-to-back giving a basket like structure
with the benzene rings forming the base of the basket and one
unbound chloride within the cavity.
2.5 to 11. Unambiguous assignment of the 13C resonances as
made possible by correlating 13C resonances to 1H signals using
the 2-D NMR technique HMQC. Proton spectra assignments
were made by identifying NOEs found in NOESY spectra (see
ESI).
In the acidic region (pH 2 to 5) the resonance of carbon D
(labeling scheme in Fig. 4) shows a large upfield shift (shielding)
on protonation, indicative of a carbon β to an amine proto-
nation site. This is consistent with protonation of H4L at ring
positions 5 and 20 to form H6L. Inspection of the basic region
(pH 8 to 11) confirms that amine nitrogens located at ring
positions 2, 8, 17 and 23 are involved in formation of the species
HL, H2L, H3L and H4L. Protonation of the ligand at these
sites results in an upfield movement (shielding) of the 13C
resonance of carbon E, consistent with its position β to the
amine protonation sites. The 13C resonance of carbon D
Potentiometric titration
A speciation plot is shown in Fig. 3 and the β constants are
collected in Table 2. As for the related ligand studied by Martell
and co-workers,20 the weakly basic nitrogens may be assigned
to the amino groups at ring positions 5 and 20, Coulombic
repulsion from the two adjacent protonated benzylic nitrogens
(positions 2 and 8, 17 and 23) favours deprotonation of the
nitrogens 5 and 20.
The protonation sites were confirmed by measurement of
the chemical shift of the 13C resonances over the pH range
J. Chem. Soc., Dalton Trans., 2001, 240–246
243

View Article Online
Table 2 Protonation constants and copper() stability constants of PEA
Ion
Hϩ
Reaction
log β
Symbol
L ϩ Hϩ
[HL]ϩ
9.48 0.01
18.355 0.009
26.58 0.01
33.95 0.01
36.87 0.02
40.44 0.01
25.24 0.03
16.96 0.02
8.13 0.06
β1
β2
β3
β4
β5
β6
L ϩ 2Hϩ
L ϩ 3Hϩ
L ϩ 4Hϩ
L ϩ 5Hϩ
L ϩ 6Hϩ
L ϩ 2Cu2ϩ
[H2L]2ϩ
[H3L]3ϩ
[H4L]4ϩ
[H5L]5ϩ
[H6L]6ϩ
Cu2ϩ
[M2L]4ϩ
βLM2
βLM2OH
βLM2(OH)2
βLMH2
L ϩ 2Cu2ϩ ϩ OHϪ
L ϩ 2Cu2ϩ ϩ 2OHϪ
L ϩ Cu2ϩ ϩ 2H
[M2L(OH)]3ϩ
[M2L(OH)2]2ϩ
[MLH2]4ϩ
29.75 0.05
Table 3 Copper() stability constants of PEA determined spectrophotometrically
Ion
Reaction
log β
Symbol
Cu2ϩ
L ϩ 2Cu2ϩ
[M2L]4ϩ
25.27 0.04
16.11 0.10
8.77 0.12
29.37 0.13
βLM2
L ϩ 2Cu2ϩ ϩ OHϪ
L ϩ 2Cu2ϩ ϩ 2OHϪ
L ϩ Cu2ϩ ϩ 2H
[M2L(OH)]3ϩ
[M2L(OH)2]2ϩ
βLM2OH
βLM2(OH)2
βLMH2
[MLH2]4ϩ
shows a downfield (deshielding) movement indicative of a 13C
resonance α to the amine protonation site. Such behaviour has
also been observed in protonation of amino acids,21 peptides22
and amines.23
Titration in the presence of copper indicated the formation
of binuclear complexes and of hydroxo species at higher
pH. Although previous potentiometric studies of similar bi-
nucleating ligands have demonstrated significant amounts of
mononuclear species, this ligand appears to be almost
exclusively binucleating. The initial model picked for the
input for SUPERQUAD included a number of hydroxo and
protonated mononuclear complexes. The first minimisation
with curves from titrations of PEA in the presence of one
equivalent of copper() gave reasonably low error values, but on
introducing stability constants for the binuclear species resulted
in rejection of hydroxo and neutral species in subsequent
runs; those mononuclear species left (the protonated copper
complexes) had poor standard deviations and their presence
did not significantly decrease sigma. All, apart from the
minor complex CuLH2, were removed from the model and the
final calculated equilibrium constants derived from these
calculations are presented in Table 2. Although the related
polyazacyclophane, MEA, investigated by Martell and co-
workers,20 gives a number of other mononuclear complexes,
the MH2L species have formation constants of comparable
magnitude (log βLMH2 is 29.75 and 29.86 for ligands PEA and
MEA respectively), however the formation constant for the
binuclear complex is rather larger for PEA (log βLM2 is 25.24
for PEA compared to 23.47 for MEA).
Fig. 5 Visible spectra of the species CuLH2, Cu2L, Cu2L(OH), and
Cu2L(OH)2 calculated from the spectrophotometric titration data.
plex a lower energy conformation is formed, resulting in extra
stabilisation of the binuclear complex. Large macrocycles with
much greater flexibility are able to form more stable mononu-
clear complexes by simultaneous co-ordination of both donor
sets to a single metal centre [OBISDIEN (1,4,7,13,16,19-
hexaaza-10,22-dioxacyclotetracosane) for example], a binding
mode prevented by the presence of the p-xylyl linkages.
After obtaining this interesting result we investigated the
system spectrophotometrically to check the validity of this
equilibrium model.
Comparison with MEA suggests that the p-xylyl linkages
have increased the stability of the binuclear complex relative
to that with m-xylyl linkages and substantially destabilised the
mononuclear complexes resulting in only a single protonated
mononuclear copper PEA complex [Cu(PEA)H2]4ϩ
.
Spectrophotometric titration of PEA
The extraction of absorption coefficients (Fig. 5) for the
species found in solution by HYPERQUAD shows that the
ε0 of the mononuclear species, per copper centre, is nearly the
same as that of the binuclear species, inferring that the co-
ordination geometries of the metal centres are approximately
Stability constants determined spectrophotometrically are
shown in Table 3. The species model for the copper PEA system
determined spectrophotometrically compares exceedingly well
with that of the potentiometric method and confirms the
hypothesis that the introduction of a rigid spacer has increased
the tendency of this molecule to form binuclear complexes. This
may be due to two factors: stabilisation of the binuclear
complexes and destabilisation of the mononuclear complexes.
The presence of the unco-ordinated end of the molecule
possibly makes the co-ordinated end more strained, resulting
in destabilisation of the mononuclear complexes. When both
sets of donors become co-ordinated to form a binuclear com-
the same, the magnitude suggesting
a five-co-ordinate
geometry. The hydroxo complexes are blue shifted as one might
expect for co-ordination of higher field ligands.
DNPDEP hydrolysis kinetics
The solvolytic and hydroxide contributions to the hydrolysis
of DNPDEP have been investigated in an earlier publication24
244
J. Chem. Soc., Dalton Trans., 2001, 240–246

View Article Online
Table 4 kobs for acid catalysed dissociation of [Cu2(PEA)]4ϩ
T/ЊC
103[Hϩ]/
mol dmϪ3
19.7
24.4
29.3
34.1
38.9
43.9
49.1
36.8
24.5
12.3
4.91
3.68
2.45
1.23
0.49
0.74
0.73
0.65
0.55
0.44
0.39
0.29
0.16
0.05
1.14
1.11
1.01
0.89
0.65
0.58
0.45
0.26
0.10
1.72
1.68
1.56
1.35
0.98
0.83
0.61
0.34
0.12
2.71
2.56
2.37
1.95
1.40
1.21
0.92
0.46
0.17
3.95
3.75
3.37
2.94
2.04
1.68
1.23
0.62
0.22
6.28
5.62
5.09
4.16
2.83
2.33
1.75
0.77
0.32
Fig. 6 pH Dependence of the Cu2(PEA) promoted hydrolysis of
DNPDEP (filled squares) superimposed over % species in aqueous
solution (solid lines).
Fig. 8 [Hϩ] Dependence of the acid dissociation rate (kobs) of [Cu2-
(PEA)]4ϩ
.
However the Cu ؒ ؒ ؒ Cu separation suggests that mechanism
2 might be unlikely.
Acid dissociation kinetics
The acid dissociation of [Cu2(PEA)Cl4]ؒ3H2OؒMeOH dis-
played simple pseudo first order kinetics and there was no
evidence of a biphasic reaction. Time lapse rapid scanning
spectrophotometric measurements show simple first order
decay of the d–d transition at ca. 640 nm and the changes in the
molar absorption coefficient indicate that the reaction being
observed is metal dissociation of the 2:1 copper:ligand
complex. This result suggests that both metal centres in the
complex act independently of each other and have identical
dissociation rates. Values of kobs over a range of temperatures
and acidities are summarised in Table 4 and Fig. 8 shows a
plot of kobs vs. [Hϩ] at 24.4 ЊC. At each temperature the reac-
tion demonstrates saturation kinetics with an initial linear
dependence of kobs on [Hϩ], but at higher concentrations of
[Hϩ] kobs becomes independent of hydrogen ion concentration.
The linear parts of the plots at low acidities all intercept the
origin indicating the absence of a solvolytic pathway resulting
in metal ion dissociation which is sometimes observed in
reactions of this type. The dissociation can be represented as
shown in Scheme 1. One nitrogen donor is dissociated from the
metal ion without its replacement by a solvent molecule and
remains within the inner co-ordination sphere (Step 1). This
activated species is susceptible to protonation after which the
complex rapidly dissociates (Step 2). Application of the steady
Fig. 7 Diagram showing the mechanism of Cu2PEA promoted
hydrolysis of DNPDEP.
and were shown to occur ca. 2 orders of magnitude slower
than the complex catalysed hydrolysis. The pseudo first order
rate constants for hydrolysis of DNPDEP are plotted over
a scaled diagram of the species distribution curves of PEA in
the presence of two equivalents of copper() in Fig. 6. The
low activity below pH 7 but substantial rate increases above
pH 7 are consistent with the hypothesis that the active species is
a hydroxo species. The midpoint of the sigmoidal curve is at
approximately 8.8 pH units, indicating that the active species
has a pKa at approximately this pH. The pKa of the dihydroxo
binuclear PEA species (8.83) matches the kinetic pKa the
closest, therefore this is likely to be the active species. The
monohydroxo dicopper complex may also be active but cannot
be the only active species as the observed rate is still seen to
increase above pH 8.5 even though the concentration of the
monohydroxo species is decreasing above pH 8.5. This suggests
that the attacking and co-ordination sites reside on the same
copper centre (Fig. 7) rather than on separate copper atoms.
Of course this does not preclude the possibility that the two
mechanisms are operating at similar rates simultaneously.
state assumption to species 2 leads to the rate equation kobs
=
k12k23[Hϩ]/(k21 ϩ k23[Hϩ]). Since K = k12/k21, kobs = k23K[Hϩ]/
(1 ϩ k23k21Ϫ1[Hϩ]). Values of k12, k23K and k23/k21 were deter-
mined for each temperature and are summarised in Table 5. The
temperature dependence of k12 gives the activation parameters
Ea = 68 1 kJ molϪ1, ∆H ‡ = 66 1 kJ molϪ1 and ∆S ‡ = Ϫ22
3 J KϪ1 molϪ1
.
J. Chem. Soc., Dalton Trans., 2001, 240–246
245

View Article Online
Table 5 Rate and equilibrium constants for the acid catalysed
dissociation of [Cu2(PEA)]4ϩ
References
1 D. E. Fenton, Adv. Inorg. Bioinorg. Mech., 1983, 2, 187; T. Sorrel,
Tetrahedron, 1989, 45, 3; D. E. Fenton and H. Okawa, in Perspectives
in Bioinorganic Chemistry, eds. R. W. Hay, J. R. Dilworth and
K. B. Nolan, JAI Press, Connecticut, 1993, vol. 2.
T/ЊC
k12/sϪ1
k23K/dm3 molϪ1 sϪ1
k23k21Ϫ1/dm3 molϪ1
19.7
24.4
29.3
34.1
38.9
43.9
0.79
1.23
1.90
2.97
4.38
6.97
182.6
281.2
381.0
513.4
711.0
892.0
231.2
229.1
200.6
173.0
162.7
128.2
2 T. Klabunde and B. Krebs, Struct. Bonding (Berlin), 1997, 89, 177;
D. Gani and J. Wilkie, Struct. Bonding (Berlin), 1997, 89, 133;
N. Sträter, W. N. Lipscomb, T. Klabunde and B. Krebs, Angew.
Chem., Int. Ed. Engl., 1996, 35, 2025.
3 N. E. Dixon, C. Gazzola, R. L. Blakeley and B. Zerner, J. Am.
Chem. Soc., 1975, 97, 4131; J. B. Sumner, J. Biol. Chem., 1926,
69, 435; E. Jabri, M. H. Lee, R. P. Hausinger and P. A. Karplus,
J. Mol. Biol., 1992, 227, 934.
4 T. F. Pauwels, W. Lippens, G. G. Herman and A. M. Goeminne,
Polyhedron, 1998, 17, 1715.
5 P. Jurek and A. E. Martell, Inorg. Chim. Acta, 1999, 287, 47;
D. A. Nation, Q. Lu and A. E. Martell, Inorg. Chim. Acta, 1997,
263, 209.
6 L. Bordignon, T. Marilde, B. Szpoganicz, M. Rizzoto, A. E. Martell
and M. G. Basallote, Inorg. Chim. Acta, 1997, 254, 345.
7 See, for example, W.-J. Lau and C.-S. Chung, J. Chem. Soc., Dalton
Trans., 1994, 191 and references therein.
8 M. J. Fernández-Trujillo, B. Szpoganicz, M. A. Máñez, L. T. Kist
and M. G. Basallotte, Polyhedron, 1996, 15, 3511; M. G. Basallote,
J. Durán, M. J. Fernández-Trujillo and A. Máñez, J. Chem. Soc.,
Dalton Trans., 1999, 3817.
9 R. J. Angelici, in Synthesis and Technique in Inorganic Chemistry,
W. B. Saunders Company, London, 1977.
10 A. E. Martell and R. J. Motekaitis, in Determination and Use
of Stability Constants, VCH, New York, 1988.
11 P. Gans, A. Sabatini and A. Vacca, Talanta, 1996, 43, 1739.
12 G. M. Sheldrick, SHELXTL Version 5, Reference Manual,
Bruker-AXS, Madison, WI, 1994.
13 M. Pietraszkiewicz and R. Gasiorowski, Chem. Ber., 1990, 123, 405.
14 W. Eschweiler, Chem. Ber., 1905, 38, 880.
15 C. Bazzicalupi, A. Bencini, A. Bianchi, V. Fusi, C. Giorgi,
P. Paoletti, A. Stefani and B. Valtancoli, Inorg. Chem., 1995, 34,
552.
16 C. J. McKenzie, H. Toftlund, M. Pietraszkiewicz, Zb. Stojec and
K. Slowinski, Inorg. Chim. Acta, 1993, 210, 143.
17 M. Sneed and J. Maynard, General Inorganic Chemistry, Van
Nostrand, New York, 1942, p. 813.
18 J. F. Coetzee and G. P. Cunningham, J. Am. Chem. Soc., 1969,
91, 568.
19 C. K. Johnson, ORTEP II, Report ORNL-5138, Oak Ridge
National Laboratory, Oak Ridge, TN, 1976.
20 R. Menif, A. E. Martell, P. J. Squattrito and A. Clearfield, Inorg.
Chem., 1990, 29, 4723.
21 A. R. Quirt, J. R. Lyerla, I. R. Peat, J. S. Cohen, W. F. Reynolds and
M. H. Freeman, J. Am. Chem. Soc., 1974, 96, 570.
22 M. Christl and J. D. Roberts, J. Am. Chem. Soc., 1972, 94, 4565.
23 J. G. Batchelor, J. Am. Chem. Soc., 1975, 97, 3410; M. Periasamy,
Heterocycles, 1982, 18, 127; J. G. Batchelor, J. Feeney and
G. C. K. Roberts, J. Magn. Reson., 1975, 20, 19; J. E. Sarneski,
H. L. Surprenant, F. K. Molen and C. N. K. Reilley, Anal.Chem.,
1975, 47, 211.
Scheme 1
Recently Fernández-Trujillo and co-workers8 have studied
the acid dissociation kinetics of the binuclear copper() com-
plex of MEA. This dissociation is also monophasic and clearly
exhibits a solvolytic reaction with k0 = 14 3 sϪ1 at 25 ЊC. In
this case the rate equation is kobs = k0 ϩ k23K [Hϩ]/(1 ϩ
k23k21Ϫ1[Hϩ]) which gives a value of k12 = 118 sϪ1. Interestingly
this value is some 100-fold greater than for PEA. Macrocyclic
complexes usually have k12 values in the range25 0.005 to 0.5
sϪ1 while open chain ligands have much faster rates.26 [Cu2-
(MEA)]4ϩ behaves more like an open-chain polyamine ligand
with respect to the lability of the Cu–N bond, whereas the
Cu–N bond of [Cu2(PEA)]4ϩ is much less labile having a k12
value closer to that determined for smaller polyamine macro-
cycles. The absence of a solvolytic dissociation pathway for
PEA is to be expected due to the bulk of the macrocyclic ligand
preventing a solvent water molecule from entering the co-
ordination sphere of the metal.
Acknowledgements
MOD is acknowledged for the funding of this project. Dr
N. Govan is thanked for guidance at CBDE, Porton Down,
and Dr Douglas R. Powell, University of Kansas, for a second
look at the crystallography data.
24 R. W. Hay and N. Govan, Polyhedron, 1996, 15, 2381.
25 R. W. Hay, M. P. Pujari and R. Bembi, Transition Met. Chem., 1986,
11, 261.
26 L. H. Chen and C. S. Chung, Inorg. Chem., 1989, 28, 1402.
246
J. Chem. Soc., Dalton Trans., 2001, 240–246