Discrete heterobinuckar Cu(I1) MIII) couple

CHAPTER -IV

IV. 1 Introduction

The unique properties of copper(I1)-gadolinium(II1) complexes have attracted

increasing interest due to their latent applications on the design of bimetallic catalysts,14b

novel molecular based magnetism and molecule devices."' The administration of

contrast agent in magnetic resonance imaging (MRI) has greatly improved the potentials

of this m ~ d a l i t ~ . ' ~ ' ~ ' ~ ' Lanthanide complexes of high stability could turn out to be

especially vital in two very different areas of research where inert complexes are

potentially useful, namely, for the separation of the lanthanides as a set of metals and for

the design of Gd(II1) contrast agents for NMR imaging.'s0 Many medicines require that

the complex be inert to metal ion relesse in water.149 The MRI contrasting agents have

an indirect mode of action. Since they contain paramagnetic metal ions, they influence

the signal intensity primarily by altering proton relaxation rates in tissue.

Gadolinium(lI1) is the most effective relaxation enhancer and almost all commercially

available MRI contrast agents contain gadolinium(II1) complexes; important issues in the

development of gadolinium(II1) containing MRI contrast agent are its low toxicity, low

osmolality, high thermodynamic and 1 or kinetic stability and the presence of at least one

water molecule in the inner coordination sphere of the Gd metal ion. The coordination

chemistry of lanthanides has become of increasing significance in the last few years due

to the wide variety of potential application of lanthanide complexes. To date, most of the

studies with magnetic properties of 4f-3d complexes have been limited to copper(I1)-

gadolinium(II1) system. The EPR studies have not been studied in detail so far. In this

chapter, we describe the synthesis (Scheme IV. 2. I), specnal, crystal structure and EPR

spectral studies of gadolinium(II1)-copper(I1) couple.

IV. 2.1 Experimental section

Details about reagents, chcmicaldsolvents and Physical measurements are given

in Chapter -11.

IV. 2 . 2 Preparation of rare earth nitrates

Gadolinium nitrate (Gd(N03)3) is prepared by adding corresponding oxide

(Gd203) to an excess of nitric acid. The mixture is heated to 80" C, filtered and

recrystallised from water. The final product is washed with sufficient quantity of ice-

cooled water to eliminate traces of nitric acid, which adhere to the crystals. Finally the

product has been washed with minimum quantity of cold diethyl ether and has kept for

drying in a dessicator.

IV. 2 . 3 Synthesis of metal precursor (Scheme IV. 2.1)

The ligand H2salbn [N,N'-1,4-butylethylenebis(salicylaldimi)] is obtained by

mixing warm ethanolic solution of salicylaldehyde (75 mL, 12.212 g, 0.1 mmol) and 1,4-

diaminobutane (4.4 g, 0.05 mmol) in 2:1 molar ratio and the whole solution is stirred at

room temperature for 10 min, and set aside until a yellow precipitate

After removal of the solvent, the yellow solid is recrystallised in chloroform and dried in

vacuo. 'H NMR spectra show that the product is pure. Yield 27.52 g (93%); 'H NMR

(400 MHz, CDC13) S 8.34 (s, 2H), 7.30 (dd, 2H), 7.23 (dd, 2H), 6.96 (d, 2H), 6.87 (t, 2H),

3.64 (t, 4H), 1.85 (m, 4H); "C NMR (100 MHz, CDC1,) 8 165.0, 161.3, 132.2, 131.3,

118.8, 117.0, 59.2, 28.6; El-MS, m/z (relative intensity) 296 (40, MC). Anal. Calcd for

C1aH~~N202: C, 72.94; H, 6.80; N, 9.45; Found: C, 71.68; H, 6.42; N, 8.70.

Cpfsnlbn): [salbn = [N,W-1,4-butylmebis(salicylideneaminato)]] (1) The

compound is synthesized by adding the methanolic solution of copper(I1) acetate (25 mL,

1.00 g, 0.5 mmol) to the methsaolic solution of salbn (50 rnL, 1.481 8 g, 0.5 mmol) under

constant stirring. The mixture is subsequently refluxed for 4 h. After cooling to room

temperature a green precipitate is collected by filh.ation, which is washed with methanol

and dried in air. Yield 1.61 g (90%); X- ( m ) , CHCl]: 624; IR (KBr, cm.'): 1638; FAB

MS rniz (relative intensity) 358 (100); Anal. Calcd for C18H18N202C~: C. 60.41; H, 5.07;

N, 7.83. Found: C, 60.36; H, 5.00; N, 7.78.

EtOH + H,N-NHl ----r

I D min, stimng d6~ Hrb

MeOH: CHCI, n

N I 0

Scheme IV. 2.1

Cu(salbn)Gd(NO,),.H# (2) is prepared by slowly adding a methanolic solution

of gadolinium nitrate (10 mL, 0.2256 g, 0.5 mrnol) to a solution of Cu(sa1bn) (100 mL,

0.1789 g, 0.5 mmol) dissolved in hot chloroform under constant stirring. It is then

refluxed for 4 h and the solution is concentrated to 50 mL. After slow evaporation, dark

green precipitate is obtained. The compound then dissolved in hot chloroform-methanol

(15110) mixture under diffusion method by diethyl ether at room temperature, when a

dark green single crystal is formed after 21 days. The compound has the molecular

formula C~(salbn)Gd(NO~)~.H~0. Yield 0.24 g (67%); hx ( m ) , CHCI3: 688; IR (KBr,

cm-'1: 1645, 1538, 1385; Anal. Calcd for C18H20NsO12CuGd: C, 30.06; H, 2.78; N, 9.74;

Found: C, 29.29; H, 2.65; N, 9.70.

IV. 2. 4 X-ray structure determination

The single crystal of (2) with appropriate dimensions 0.40 x 0.32 x 0.26 mm is

used for X-ray diffraction studies. The intensity data are collected at room temperature

using the Siemens SMART CCD area detector three-circle diffractometer equipped with

graphite monochromated MoKa (1 = 0.71073 A) radiation. The data collection

nominally covers over a full hemisphere of reciprocal space by a combination of three

sets of exposures, each set has a different 4 angle for the crystal and each exposure covers

0 .39n o. The crystal to detector distance is 5.89 cm. Coverage of the unique set is over

86% complete to at least 25.6" in 8. Crystal decay has been monitored by repeating the

initial frames at the end of the data collection and analyzing the duplicate reflections; it is

found to be negligible. The substantial redundancy in data allows empirical absorption

corrections to be applied using multiple measurements of equivalent reflections. Data

frames are collected for 10-30 s frames, depending on the intensity of the data, giving an

overall time for data collection of 7-18 h. The data frames are integrated using SAINT

and are merged to give a unique data set. The structure is solved by automated Patters011

methods and subsequent difference Fourier technique using DIRDIF 98.3 and the third

one is solved using SHELX-97 and are refined on F~ using full matrix least squares

techniques using SHELXL-97. All hydrogen atoms are included at calculated positions

using a riding model. The U,,, of H atoms of CH and CH2 groups and the methyl group

are taken as 1.2 U,, of their carrier atoms, except for the water hydrogen atoms, which

can not be identified. All non-hydrogen atoms are refined with anisotropic thermal

parameters. The final R-value is 0.063 for 4607 observed reflection with 1>2 a (I) and

0.089 for (6356) all data. Anomalous dispersion effects for all atoms are included in the

final calculations.

IV. 3 Results and discussion

IV. 3 .1 Synthesis and spectroscopic studies

The precursor (salbn) has been synthesized by Schiffs base reaction; the spectral

studies such as 'H NMR as well as "C NMR c o n f m the tetradentate nature of the ligand

(Fig. N. 1-IV. 2).

In addition to the analytical data, the most important information is afforded by

positive ion FAB mass. The main signal (I = 100%) occurs at 358, which corresponds to

the Cu(salbn) species (Fig. IV. 3). The electronic spectrum of the complex (1) exhibits a

higher energy band at 624 nm, which has been shifted to lower energy (Red shift) at 688

nrn for (2) seems to be due to a distortion of geometry occurring at the copper center

(Fig. N. 4). There is a great similarity between the i.r. spectra of the heterodinuclear

c ~ m ~ l e x e s . ' ~ ' ~ ' ~ ~ They are almost superimposible with the exclusion of the presence of a

vr. = 1645 cm-' in the spectrum of (2). This absorption appears at vc. l, = 1638 cm"

in the case of compound (1). In the compound (2), the energy discrepancy between

asymmetric and asymmetric stretching frequency for (NO;) is Av = 153 cm-I, which

indicates the nature of bidentate nitrate bridging.

IV. 3. 2 Cwstal structures of the complex (2)

A view of the dinuclear unit is represented in Fig. IV. 5. The unit cell contains

four distinct entities [LCUG~(NO~)~.H~O] (Fig. IV. 6). The crystal data, selected bond

angles and bond lengths are given in Tables IV. 1-4. The ORTEP plot of the molecule

has drawn at 50% probability displacement thermal ellipsoids with atomic numbering

scheme, The coordination geometry about the Cu ion consists of a chelate ring and a

distorted coordination plane composed of two imine N atoms and two phenol 0 atoms.

The copper(I1) ion completes its coordination sphere with two imine nitrogen atoms from

Fi. IV. 1 'H NMR Spectrum of the ligmd Hzsalbn. Top one corresponds to expanded version in the range 6-8 ppm

Fig. IV. 2 ' )c NMR Spectrum of the ligand Hlsalbn

Fig. lV. 3 FAB Mass spectrum of the complex Cu(salbn) (1)

Fig, N. 4 Electronic absorption spectrum of Cu(&)Gd(N03)3.H20 (2)

Table IV. 1 Crystallographic Data for the Complex Cu(salbn)Gd(NO1)3.H20

(2)

Chemical formula

Chemical fonnula weight (Wt)

Crystal system

Space WuP Unit cell dimensions (A)

Volume ( A3)

z Dc (calculated) ~ g i m '

Wavelength (1, A)

Absorption coefficient (p, cm")

Temperahue (T, K)

Crystal size (mm)

F (000)

CIIIHZONSOIZG~CU

719.015

Monoclinic

p2'/n

a = 9.025(1)

b = 22.912 (1)

c = 12.790(1)

2609.47(8)

4

1.8255

0.71073

34.00

294

0.40 x 0.32 x 0.26

1400

Data collection

Diffractometer Siemens SMART

CCD area detector

Index ranges -1 1< = h<=12, -3O<=k<=25, -1 60<=1<=12

e m u 28.3'

Reflection measured 20241

Independent reflections 6356

Reflections with I> 20 (I) 4607

Refinement

Refmment method on F2

Reflections used 6356

Refinement parameters 334

Goodness of-fit (S) 1.267

Final R indices [I>2u(I)] R1=0.064&WIU=0.192

Largest diff peak and hole (e.A')) 2-74 to-3.95

Table IV. 2 Fractional atomic coordinstcs and equivalent isotropic tamperatun

factors of non-H atoms with e.8.d.'~

Atom b u, Gdl 0.0347(5) 0.0948(2) 0.3357(3) 0.0343(1)

C U ~ -0.0822(1) 0.1235(4) 0.0855(8) 0.0352(3)

N1 -0.1820(8) 0.0868(3) -0.0410(5) 0.0359(2)

N2 -0.1301(8) 0.2044(3) 0.0472(6) 0.042q2)

N3 0.2569(10) 0.0016(4) 0.3615(6) 0.0513(3)

N4 0.2215(10) 0.1223(4) 0.5363(6) 0.0484(3)

NS -0.1485(9) 0.2004(4) 0.3151(7) 0.0502(3)

01 -0.0739(7) 0.0522(2) 0.1713(4) 0.0403(2)

02 0.0891(6) 0.1483(2) 0.1897(4) 0.0366(2)

03 0.1507(7) 0.0042(3) 0.4144(5) 0.0456(2)

0 4 0.2550(8) 0.0378(3) 0.2868(5) 0.0524(2)

05 0.3600(15) -0.0380(6) 0.3784(10) 0.1190(6)

06 0.0888(8) 0.1040(3) 0.5338(5) 0.0479(3)

07 0.2650(7) 0.1287(3) 0.4466(5) 0.0446(2)

08 0.3038(11) 0.1318(5) 0.6173(6) 0.0855(4)

09 -0.0252(7) 0.1947(3) 0.3775(6) 0.0554(2)

010 -0.1986(7) 0.1556(3) 0.2650(5) 0.0454(2)

01 1 -0.2094(10) 0.2469(4) 0.3039(9) 0.0927(4)

01 W -0.1714(7) 0.0513(3) 0.3965(5) 0.0524(3)

Cl 0.1748(9) 0.1929(4) 0.1732(6) 0.0340(2)

C2 0.3276(10) 0.1943(5) 0.2251(8) 0.0492(3)

C3 0.4254(12) 0.2395(5) 0.2050(10) 0.0695(5)

C4 0.3672(13) 0.2842(5) 0.1382(9) 0.0620(4)

CS 0.2202(12) 0.2851(5) 0.0902(9) 0.0580(4)

C6 0.1227(10) 0.2396(4) 0.105 l(7) 0.0406(3)

C7 -0.0296(10) 0.2439(4) 0.0538(8) 0.0484(3)

C8 -0.2921(10) 0.2255(5) 0.0090(9) 0.0533(4)

C9 -0.4017(9) 0.1772(4) -0.0250(8) 0.0484(3)

ClO -0.3907(12) 0.1489(5) -0.1265(9)

Cl 1 -0.2345(13) 0.1235(4) -0.1362(7)

Cl2 -0.2099(9) 0.0324(4) -0.051 6(6)

C13 -0.1810(9) -0.01 15(3) 0.0304(6)

C14 -0.2182(10) -0.0699(4) 0.0005(7)

Cl5 -0.2015(10) -0.1153(4) 0.071 l(8)

C16 -0.1424(11) -0.1038(3) 0.1741(8)

C17 -0.1034(10) -0.0486(3) 0.2076(7)

Cl8 -0.1163(8) -0.0011(3) 0.1360(6)

U,, is defined as the 1/3* of the trace of orthogonalised Uij tensors.

Table IV. 3 Anisotropic displacement thermal parameter for all non-H atoms with e.s.d's

Atom UI I u22 u33 u23 u13 u12

Gdl

Cul

N1

N2

N3

N4

N5

01

02

03

04

05

06

07

08

09

010

01 1

OIW

C 1

C2

C3

C4

C5

C6

C7

C8

C9

ClO 0.0584(6)

C l l 0.0854(7)

Cl2 O.W3(4)

C13 0.0397(4)

C14 0.0549(5)

Cl5 0.0576(5)

C16 0.0631(6)

C17 0.0600(5)

C18 0.0358(4)

The form of the anisotropic displacement parameter is: exp [-2z2{h2a2IJll + k2b2u22 +

I ' C ~ U ~ , + 2hkabU12 + 2hlacUil + 2 k l b ~ U ~ ~ } ] w h w a, b and c are reciprocal lattice

constants

Table IV. 4 Seleoted bond distance (A) and angle (*) for complex 2

Bond lengths

CU 1 -Gd I

Gdl-Ol

Gd 1-02

Gd 1-03

Gd 1-04

Gd 1-06

Gdl-07

Gdl-09

Gdl-010

Gdl-OlW

CU-0 l

CU-NI

CU-02

CU-N2 1 I I

Bond angles

3.269(1)

2.380(5)

2.353(6)

2.463(6)

2.541(7)

2.510(7)

2.445(6)

2.431 (7)

2.564(6)

2.350(7)

1.961(6)

1.915(7)

1.953(5)

1.948(7)

Gd-01-Cul

01-Gdl-02

01-Gdl-01W

NI-Cul-01

NI-Cul-N2

Gd-02-Cul

01-Gdl-04

02-Gdl-07

N2-Cul -02

01-Cul-02

97.3(2)

67.7(2)

83.4(2)

94.4(3)

98.7(3)

98.4(2)

78.4(2)

146.3(2)

90.8(3)

84.6(2)

'ig. IV. 5 ORTEP representation ofthe compound (2) showing the 50% probabii t h d ellipsoids.

Fig. IV. 6 Packing view of the compound (2) in the unit cell

Schiff base. The Cu-0 bond lengths are 1.%1(6) and 1.953(5) A for Cu-01 and Cu-02,

reqatively while the Cu-N bond lengths are 1.915(7) and 1.948(7) A for Cu-N1 and Cu-

N2 respectively, which are normal values for gadolinium(I1)-copprr(I1) Schiff base

comp~exes155~lJ6,1S7 and also agrees with previously reported The most

interesting comparative aspects of copper complexa with imine phenols involve the

steric influence of the alkyl backbone upon the molecular structure. The Cu ion is

coordinated by two imine nitrogens and two oxygen from Schiff base ligand. T h w four

atoms are deviating significantly from the distorted coordination plane CulOlOZNlNZ

and they are 0.451(7) and 0.031(1) respectively. The copper is pulled out from the plane

in spite of the absence of any apical ligand. The steric interaction of the propyl, butyl and

phenyl backbones affects the copper coordination gcomewy significantly in many

respects. In the five membered ring systems with a two C atom backbone, the Cu-N

distance is short (average 1.916 A) and the N-Cu-N angle (82.7O) and the dihedral angle

(5.3') are small. Addition of a third C atom to the backbone to make a six membered

chelate ring results in increased Cu-N lengths N-Cu-N angles and dihedral angles.

Further increase in the backbone size to give a seven membered ring makes it more

difficult to maintain the configuration without considerable puckering of the ring. It

seems that tuning of the Cu-N lengths, N-Cu-N angle and dihedral contributes to the

flexibility of the coordination of copper by tekadentate iminephenol ligands.

Examination of the gauche conformation of the butane bridge, which has often

been found to be unsymmetrical, can also provide some basis for comparison of the

extent of distortion of Cu(I1) imina phenol complexes. The butane bridging C atoms are

asymmetrically buckled and its torsion angles are -73(1) (NZ-C8-C9-ClO), 56(1) (C8-

C9.Cl0-C11) and 50(1)" (C9-ClO-C11-N1) respectively. These torsion angles are

comparable with similar type of copper(I1) coordination comp~exes.~" It is well known

that increasing the steric hindrance by elongation of the akyl bridge will result in a

change in the chelate pattern from planar to tetrahedral. The distortion of the inner

ooordination sphere can be recogaized by the magnitude of the dihsdral angle between

the two planes defined by Cu2N202. The dihedral angle between two planes is 30.9(2)O.

The gadolinium environment is a distorted square antiprism of oxygen atoms, two

belonging to salbn ligand and six oxygem belonging to the three bidentate nitrato ions

and one water molecule orientated in axial position. The ninth coordinated water oxygen

is at 2.350(7) I$ from the gadolinium atom and this value is comparable with metal Hz0

distance observed in other lanthanide complexes (Gd-0 = 2.39(1) A . 1 5 2 3 1 5 6 The distance

between gadolinium(II1)-copper(I1) is 3.269(1) A, which is close to 3.252(4) A found for

the complexes of Khan and co-worker~ '~~ but shorter than 3.4275(9)-3,5231(4) of the

complexes of Costes et al,I6' which is still greater than Sasaki et and Kahn et a ~ , ' ~ ~ .

The Cu-02-Gd bridge is asymmetric. The Cu-0 bond distances are 0.82 A shorter than

the Gd ones. The two Cu-0-Gd angles (97.3(2) and 98.4(2)") are almost equal to each

other within an error of 1 degree. The three bidentate nitrato ions are bound to the

gadolinium(II1) ions in a slightly aspmetric fashion. All Gd-0 and N-0 bond distance

are in good agreement with corresponding values in similar type of Gd(II1)-Ni(I1)

complex.'s5 The evaluation of dihedral angle and copper(I1)-gadolinium(lII1) distance

with previously reported complexes are given in Table IV. 5.

The bridging network GdOlO2Cu has a butterfly shape taking 0 1 0 2 as the

hinge, the Gd0102 and Cu0102 planes forming a dihedral angle of 33.7(2)" and the 01-

0 2 distance being equal to 2.635(8) A. These values agree with the previously reported

result^.'^' The 0-0 distance (01-02, 03-04, 06-07 and 09-010) fall in the range of

2.161(10) to 2.635(8) A. In accordance with the bidentate nature of NO3 ligands, we note

in every case that N-0 bond lengths are nearly equal.

Table IV. 5 Comparison of structural parameters a and b for dinuclcar Cu-Gd complexes

a is the dihedral angle between the 0(1)Cu0(2) and 0(1)Gd0(2) planes in deg and b is

the Cu-Gd separation in A , salen = N,N'-ethylenebis(salicylideaminato). MeIm = 1-

methylimidazole. Salbn is described in the text, 02COMe = monomethyl carbonate, hfac

= 1,1,1,5,5,5-hexafluoro-acetylacetone and salabza = N,N'-bis(salicy1idene)-2-amino-

benzylarnine.

IV. 3 . 3 Electron Paramagnetic Resonance

Ref

166

155

162

167

168

164

169

170

Compound

LCu(OzCOMe)Gd(thd)2

L C U ( O C M ~ ~ ) G ~ ( N O ~ ) ~

LCU(OM~I)G~(NOI)~

[ L C U C I ~ G ~ ( H ~ O ) ~ ] C I . ~ H ~ O

Salen(Meim)C~Gd(hfa)~

C~(salabza)Gd(hfac)~

[CuLGd(NOi)sl

[CuGd(erns)(XO3),.H10]

The EPR spectrum of the polycrystallme sample of complex (1) has been

recorded at 77 K, yielding parameters gil = 2.137, g~ = 2.082, 41 = 142 x lo4 cm-I.

This is a typical of tetragonally coordinated monomeric copper(I1) complex with the

unpaired electron in the dr2+ orbital.'" A Polycrystalline powder EPR spectrum of

complex (2) at room temperature is shown in Fig. IV. 7 (inset). It shows a strong unique

quasi-isotmpic broad signal centered at g = 2.27 but no clear characteristic peak in the g

= 2 region.'72 More information is not available fmm this spectrum. The spectrum of

Cu(salen)Gd(NO3)1.CH3OH

Cu(salbn)Gd(N03)3.H20

a

19.1(2)

16.6(2)

12.9(2)

1.7(2)

39.6

132.61

4.3

24.5

b

3.4727(4)

3.5231(4)

3.4275(9)

3.5 121(5)

3.252(4)

3.2481(8)

3.401

3.428(1)

147.4

33.7(2)

3.224(1) ( 165

3.269(1) / This work

complex 2 at 77 K (Fig. IV. 7) exhibits an anisotropic broad signal, with zero-field

resonance lines. At 300 K room temperature the spechum is less informative, but on

cooling at 77 K, the fine structure exhibits seven transitions around 99, 175, 241, 300,

31 1, 436 and 465 mT, due to zero field The electron spin energy level

diagram for Gd(III), is shown in Fig. IV. 8. The four Kramers' doublets, which are

degenerate in the absence of an external magnetic field, splits into eight energy levels and

transitions between them yield seven zero-field transitions. The seven transitions,

mentioned above correspond to these seven transitions. The calculated g and D values

are 2.51 and 0.047 cm'l. Using these parameters, the powder specturn has been

calculated, where the agreement is very good. This experimental EPR spectrum also

corresponds to the superposition of the resonance signals of the copper(I1) ion, on the

gadolinium(lI1) signal. The intense feature at 241 mT becomes wider on cooling, may be

due to spin-spin relaxation effect.'16 This summing up reveals that the copper(I1) (S = %)

gadolinium(II1) (S = 712) interaction is weak which gives rise to an S = 4 ground state

and S = 3 low lying excited state and S = 4 may be significantly populated at 77 K.

g. N. 7 EPR spectrum of the compound (2) in solid state at RT (inset), 9.39537 GHz and 77 K, hquency = 9.4023 GHz

Fig. IV. 8 Energy level diagram of ~ d ' * ion (2). D is the zero-f eld splitting

parameter.

N.4 Summary

In this chapter, we describe the synthesis of a discrete dinuclear complex with

increasing backbone chain in diamine arm results from the larger distortion of the

geometry around copper(l1) towards a tetrahedral structure. The steric interaction of the

ethylene, propyl and biphenyl backbones affects the copper coordination geometry

significantly in CU-N distance, N-Cu-N angle and the dihedral angle. Further increase in

the backbone size (butyl) to give a seven membered ring makes it more difficult to

maintain the configuration without considerable puckering of the ring, which is tuning

the Cu-N lengths, N-Cu-N angle and dihedral angle contributes to the flexibility of the

coordination of copper by tetradentate iminopbenol ligand. We observe 7 fine structure

lines at 77 K, which exhibits a weak interaction between Cu(l1)-Gd(II1) core. This piece

of the data deserves further investigations to authenticate this hypothesis and we are at

present trying to obtain J value by magnetic measurements from RT to 4 K and single

crystals EPR study, which might possibly gives a better view of interaction.