New complexes with Gadolinium and Samarium

Complexation of the macrocyclic ligand Tetramethyldibenzotetraaza[14] annulene $[(TMTAA)H$₂] 1 with d-block metals leads to the formation of monomeric complexes or to the stabilisation of metal-metal multiple bonds, as with Cr and Mo. The ligand was choosen to try whether Ln-Ln bonding - frequently observed in the solid state - is also accessable in discrete complexes. So far, this ambitious goal could not be reached. Yet interesting polynuclear complexes resulted from systematic investigations on reactivity and coordination of rare earth halides with this ligand system.

Red crystals of a binuclear complex $[(TMTAA)$₃Gd₂] 3 are formed in the reaction of $[(TMTAA)K$₂] 2 with $GdCl$₃. As a result of single crystal X-ray investigations 3 could be described as a neutral tripeldecker complex (Fig. 1).

Figure 1: Structure of 3 in the solid (Gd-Gd=370pm)

The Gd-atoms are coordinated by slightly distorded cubes of the N-atoms of the TMTAA-ligands. There are two different ways of coordination of the Gd-atoms: Gd(1) is displaced to the same side of the outer ligand, to which the benzoid rings are tipped (Gd(1)-N=240pm). Additional bonds are formed to the four N atoms of the central ring ligand. However, the coordination occurs on the side of the ligand where the $C$₆H₄-fragments are not tipped to (Gd(1)-[N(1)-N(4)]-plane=170.2pm, Gd(1)-[N(4)-N(8)]-plane = 190.2pm).

The bonding situation of Gd(2) is different. Gd(2) is connected to each of the four N-atoms of two TMTAA-ligands on the same side of the ligand where the $C$₆H₄-fragments are tipped to.(Gd(2)-[N(4)-N(8)]-plane=180pm, Gd(2)-[N(9)-N(12)]-plane=130pm). This is made possible by a $90o$ rotation of the third TMTAA-ligand around the Gd-Gd axis. The N-Gd(2) distances vary distinctly, the shortest N-Gd(2) bonds being those to the outer TMTAA-ligand. As a result of the staggered arrangement of the ligands, the Gd ions are perfectly shielded. 3 is also formed by reaction of $[Gd(N(SiMe$₃)₂)₃] with $[(TMTAA)H$₂].

The reaction of $[(TMTAA)K$₂] 2 with $[SmI$₂(thf)₂] yields a completely different kind of complex. According to the single crystal structure an unusual $[Sm$₂(ligand)₂] core is built up, where each Sm ion is connected to a $N$₄-unit (Sm-N=220pm ) and to a $C$₆H₄-unit by eta⁶ coordination (Sm-C=300pm) (Fig. 2).

Figure 2: Structur of 4 in the crystal

The coordination to only one of the $C$₆H₄ fragments causes different angles between the $N$₄ plane and the two benzoid units in the heterocyclic ligand. This leads to the conclusion that there might be a pi-type interaction between the $C$₆H₄ fragment and the d orbitals of Sm$III$.

The formerly unsaturated $[TMTAA]2-$ ligand $([C$₂₂H₂₂N₄]^2-) is transformed to the saturated anion TMTAT = Tetramethyldibenzotetraazatetradecan $[C$₂₂H₂₈N₄]^4- by reaction with the strong reducing agent $Sm2+$. The imid groups are now amid groups and the $Sm2+$-ion must have been oxidized to $Sm3+$.

The dimeric $[Sm$₂(TMTAT)₂]^2- core in 4 is connected to two $[K(thf)$₃]⁺ fragments. The potassium ion has a coordination number of seven. The $[K(thf)$₃]⁺-unit and the Sm ion are displaced to the two opposite sides of the $N$₄ plane of the TMTAT ligand. (K-$N$₄=310pm)

The two compounds described are the first examples of rare earth complexes with this ligand system. The use of TMTAA-type ligands seems to be an easy route to the synthesis of many new compunds with unusual structural features. In further investigations this field should be expanded with other macrocyclic ligands to built up bigger polynuclear complexes. This might be a possible way to realise the aim of a Ln-Ln bonding mentioned at the beginning.

(Jörg Magull and Arndt Simon)

From the yearbook of the institute ("Wissenschaftlicher Tätigkeitsbericht") 1991

webmaster