We report a new macrocyclic ligand, N,N'-bis[(6-carboxy-2-pyridil)methyl]-4,13-diaza-18-crown-6 (H(2)bp18c6), designed for complexation of lanthanide ions in aqueous solution. Potentiometric measurements evidence an unprecedented selectivity of bp18c6 for the large Ln(III) ions. Among the different Ln(III) ions, La(III) and Ce(III) show the highest log K(ML) values, with a dramatic drop of the stability observed from Ce(III) to Lu(III) as the ionic radius of the Ln(III) ions decreases (log K(CeL) - log K(LuL) = 6.9). The X-ray crystal structures of the Gd(III) and Yb(III) complexes show that the metal ion is directly bound to the 10 donor atoms of the bp18c6 ligand. The structure of the complexes in solution has been investigated by (1)H and (13)C NMR spectroscopy, as well as by theoretical calculations performed at the DFT (B3LYP) level. Our results indicate that a conformational change occurs around the middle of the lanthanide series: for the larger Ln(III) ions the most stable conformation is Delta(deltalambdadelta)(deltalambdadelta), while for the smallest Ln(III) ions (Gd-Lu) our calculations predict the Delta(lambdadeltalambda)(lambdadeltalambda) form being the most stable one. This structural change was confirmed by the analysis of the Ce(III)-, Pr(III)-, and Yb(III)-induced paramagnetic (1)H shifts. The selectivity that bp18c6 shows for the large Ln(III) ions can be attributed to a better fit between the light Ln(III) ions and the relatively large crown fragment of the ligand. Indeed, our DFT calculations indicate that the interaction between the Ln(III) ion and several donor atoms of the crown moiety is weakened as the ionic radius of the metal ion decreases.

A new pyridine-containing ligand, N,N'-bis(6-carboxy-2-pyridylmethyl)ethylenediamine-N,N'-diacetic acid (H(4)L), has been designed for the complexation of lanthanide ions. (1)H and (13)C NMR studies in D(2)O solutions show octadentate binding of the ligand to the Ln(III) ions through the nitrogen atoms of two amine groups, the oxygen atoms of four carboxylates, and the two nitrogen atoms of the pyridine rings. Luminescence measurements demonstrate that both Eu(III) and Tb(III) complexes are nine-coordinate, whereby a water molecule completes the Ln(III) coordination sphere. Ligand L can sensitize both the Eu(III) and Tb(III) luminescence; however, the quantum yields of the Eu(III)- and Tb(III)-centered luminescence remain modest. This is explained in terms of energy differences between the singlet and triplet states on the one hand, and between the 0-phonon transition of the triplet state and the excited metal ion states on the other. The anionic [Ln(L)(H2O)]- complexes (Ln=La, Pr, and Gd) were also characterized by theoretical calculations both in vacuo and in aqueous solution (PCM model) at the HF level by means of the 3-21G* basis set for the ligand atoms and a 46+4 f(n) effective core potential for the lanthanides. The structures obtained from these theoretical calculations are in very good agreement with the experimental solution structures, as demonstrated by paramagnetic NMR measurements (lanthanide-induced shifts and relaxation-rate enhancements). Data sets obtained from variable-temperature (17)O NMR at 7.05 T and variable-temperature (1)H nuclear magnetic relaxation dispersion (NMRD) on the Gd(III) complex were fitted simultaneously to give insight into the parameters that govern the water (1)H relaxivity. The water exchange rate (k(298)(ex)=5.0 x 10(6) s(-1)) is slightly faster than in [Gd(dota)(H2O)]- (DOTA=1,4,7,10-tetrakis(carboxymethyl)-1,4,7,10-tetraazacyclododecane). Fast rotation limits the relaxivity under the usual MRI conditions.

The stability trends across the lanthanide series of complexes with the polyaminocarboxylate ligands TETA(4-) (H4TETA=2,2',2'',2'''-(1,4,8,11-tetraazacyclotetradecane-1,4,8,11-tetrayl)tetraacetic acid), BCAED(4-) (H4BCAED=2,2',2'',2'''-{[(1,4-diazepane-1,4-diyl)bis(ethane-2,1-diyl)]bis(azanetriyl)}tetraacetic acid), and BP18C6(2-) (H2BP18C6=6,6'-[(1,4,10,13-tetraoxa-7,16-diazacyclooctadecane-7,16-diyl)bis(methylene)]dipicolinic acid) were investigated using DFT calculations. Geometry optimizations performed at the TPSSh/6-31G(d,p) level, and using a 46+4f(n) ECP for lanthanides, provide bond lengths of the metal coordination environments in good agreement with the experimental values observed in the X-ray structures. The contractions of the Ln(3+) coordination spheres follow quadratic trends, as observed previously for different isostructural series of complexes. We show here that the parameters obtained from the quantitative analysis of these data can be used to rationalize the observed stability trends across the 4f period. The stability trends along the lanthanide series were also evaluated by calculating the free energy for the reaction [La(L)](n+/-)(sol)+Ln(3+)(sol)→[Ln(L)](n+/-)(sol)+La(3+)(sol). A parameterization of the Ln(3+) radii was performed by minimizing the differences between experimental and calculated standard hydration free energies. The calculated stability trends are in good agreement with the experimental stability constants, which increase markedly across the series for BCAED(4-) complexes, increase smoothly for the TETA(4-) analogues, and decrease in the case of BP18C6(2-) complexes. The resulting stability trend is the result of a subtle balance between the increased binding energies of the ligand across the lanthanide series, which contribute to an increasing complex stability, and the increase in the absolute values of hydration energies along the 4f period.