Free ion magnetism

W.M. Reiff, A.M. LaPointe, E. Witten, Virtual free ion magnetism and the absenee of Jahn— Teller distortion in a linear two-coordinate eomplex of high-spin iron(II). J. Am. Chem. Soc. 126, 10206-10207 (2004)... 

As seen in Fig. 8 the experimentally determined magnetic moments at room temperature are in general much lower than the free ion values. To extract the contribution of orbital reduction, the influence of intermediate coupling, crystal field effects and j-j mixing must be considered. 

The L and S values are those from which the / value was formed via the vector coupling rule. These formulae strictly apply only for the magnetism of free-ion levels. They provide a good aproximation for the magnetism of lanthanide complexes, as we shall note in Chapter 10, but provide no useful account of the magnetic properties of d block compounds. 

Consider the orbital angular momentum of a free-ion term. Here L = 3 and the orbital degeneracy is 7. Application of Van Vleck s formula (5.8) predicts an effective magnetic moment. 

Concerning induced orbital moments of U-based intermetallic compounds, many PND experiments have been performed and have shown that the ratio iL/ -is can be used as a measure of the hybridisation [42-44] (in the light actinides, orbital and spin moments are oppositely directed and the neutron magnetic form factors are highly sensitive to the ratio uL/us). Indeed, this ratio is reduced as compared to the free ion expectations (Figure 4). 

Figure 1.1 Effect of interelectronic repulsion, spin-orbit coupling and magnetic field on the energy levels arising from a given 4fn configuration for a free-ion Ln3+. The magnetic field effect is estimated assuming a 1 T field.

The magnetic susceptibility of the free ion will follow the Curie law ... 

The lanthanides have electrons in partly filled 4/orbitals. Many lanthanides show colors due to electron transitions involving the 4/orbitals. However, there is a considerable difference between the lanthanides and the 3d transition-metal ions. The 4/ electrons in the lanthanides are well shielded beneath an outer electron configuration, (5.v2 5p6 6s2) and are little influenced by the crystal surroundings. Hence the important optical and magnetic properties attributed to the 4/ electrons on any particular lanthanide ion are rather unvarying and do not depend significantly upon the host structure. Moreover, the energy levels are sharper than those of transition-metal ions and the spectra resemble those of free ions. 

Magnetic susceptibility measurements have been made for the lanthanide complexes of 2,6-DMePyO (171), EPO, BPO, EAsO, BAsO (237, 238), and AAP (95). In all these cases, the measured magnetic susceptibilities agree with the free ion values. Dipole moment measurements have been carried out on complexes of the type Ln(TBP)3-(N03)3 (387, 388). ESR studies on Gd(AP)6 I3 (389) and thermal decomposition studies on the complexes of the lanthanides with DMSO (250), DPSO (250, 259), QNO (175), HOQNO (190), 18-crown-6 (132), 15-crown-5 (131), and gycine (390) have also been reported. 

Results obtained from the alkali iodides on the isomer shift, the NMR chemical shift and its pressure dependence, and dynamic quadrupole coupling are compared. These results are discussed in terms of shielding by the 5p electrons and of Lbwdins technique of symmetrical orthogonalization which takes into account the distortion of the free ion functions by overlap. The recoilless fractions for all the alkali iodides are approximately constant at 80°K. Recent results include hybridization effects inferred from the isomer shifts of the iodates and the periodates, magnetic and electric quadrupole hyperfine splittings, and results obtained from molecular iodine and other iodine compounds. The properties of the 57.6-k.e.v. transition of 1 and the 27.7-k.e.v. transition of 1 are compared. 

The compounds Ln(C5H5)2Cl also have been made only with the lanthanides above samarium (772). These compounds are stable in the absence of air and moisture, sublime near 200 °C, are insoluble in non-polar solvents, and exhibit room temperature magnetic moments near the free ion values (772, 113). The chloride ion may be replaced by a variety of anions including methoxide, phenoxide, amide and carboxylate. Some of these derivatives are considerably more air-stable than the chloride — the phenoxide is reported to be stable for days in dry air. Despite their apparent stability, little is known about the physical properties of these materials. The methyl-substituted cyclopentadiene complexes are much more soluble in non-polar solvents than the unsubstituted species. Ebulliometric measurements on the bis(methylcyclopentadienyl)lanthanide(III) chlorides indicated the complexes are dimeric in non-coordinating solvents (772). A structmre analysis of the ytterbium member of this series has been completed (714). The crystal and molecular parameters of this and related complexes are compared in Table 5. 

The room temperature magnetic susceptibihty is known for most of the complexes mentioned here, and in general is very near the free ion value. However little low temperature data is available. The dimeric complexes provide an opportunity for magnetic exchange, but preliminary results show Curie-Weiss behavior down to 4 K for the [Yb(C5H4(CH3))2Cl]2 complex 126). 

An approach other than steric hindrance has been used to overcome the previously mentioned instability of the actinide homoalkyls. It was found that the inclusion of jT-bonding ligands in the coordination sphere considerably enhanced the stability of the alkyl complex. Recently, the same line of reasoning has also yielded a new series of 7r-cyclopentadienyl lanthanide alkyls (C5H5)2LnR where Ln =Gd, Er, Yb and R = C=C, and CH3 120,121). The infrared data for these complexes are consistent with u-bonded structures and the room temperature magnetic susceptibilities are very close to the free ion values. The actinide complexes (75,... 

If the electron-cloud radius yrms were exactly equal to the structural radius r, Wasastjerna s criterion would be obviously true. But in fact, for ions r ce. 2 yrms (Table 3). Hence the criterion needs justification. It is obviously most probable for isoelectronic ions (cp. Eauling (/)), but the electron-cloud radii should refer to the ions in the crystals, not to the free ions. For, with a gross difference between crystal and free-ion electron-cloud radii for the hydride ion, there may be significant differences for others 40). For the crystals the electron-cloud radii could be obtained either from polarizeability or from magnetic susceptibility. The theory of polarizeability is less certain and there is a considerable correction to infinite wavelength. We therefore adopt the magnetic evidence. But this must be corrected for the inner shell contribution (Table 3). 

---