Multiplet Effects in X-ray Absorption - Inorganic Chemistry and ...

More documents

Recommendations

Info

58 F. de Groot / Coordination Chemistry Reviews 249 (2005) 31–63tinguished by the energy position and branching ratio analysis.In addition, it is clear that there is a significant variationwithin each group. These variations are due to differences inthe local symmetry and covalence. The 2p XAS spectra ofCODH indicate that most of the Ni in as-isolated Ct-CODHis low-spin Ni II . Upon CO treatment, a fraction of the nickelis converted either to high-spin Ni II and/or to Ni I .Niindithionite-reduced Rr-CODH also exhibits a clear low spinNi II component, again mixed with either high-spin Ni II orNi I . The spectrum of Rr-CODH shifts to higher energy uponoxidation, suggesting either that most of the high-spin Ni IIis converted to low-spin Ni II and/or that some Ni is oxidizedbetween these two forms.A number of 2p XAS studies on coordination compounds,including nickel dithiocarbamate complexes, vanadiumwith oxyoxime ligands and iron with bidentateN-donor ligands have been performed by the groups ofGarner and co-workers [64–66]. An interesting study ison nickel dithiocarbamate and xanthate complexes, whereit is shown that the 2p XAS experiments induce photoreductionof such nickel complexes, for example, modifying[Ni IV (S 2 CNEt 2 ) 3 ][BF 4 ]intoaNi II square-planar species.Similar phenomena were often observed for many X-raysensitive compounds that have been measured at the mostintense X-ray sources, with the implication that the measurementsalways have to be checked for sample damage.Because this is an important issue, some of the results onthe nickel dithiocarbamate complexes will be discussed.Fig. 27 shows the gradual change from an octahedral Ni IVcenter to a square planar Ni II center under irradiation, wherethe final spectrum was reached after 3 h. Similar effects werefound on a range of samples and it was found that the decayrate is dependent on photon flux, photon energy andligand set, but is independent of temperature [66]. In caseof metallo-enzymes, it is well known that they are unstableand for those systems measurements at low temperature(4 K) also seemed to be better concerning radiation damagecompared with, for example, 77 K. In Fig. 26, the changeswere gradual and slow. It can be expected that more sensitivesystems will even decay within seconds, which impliesthat even the first spectrum is already taken on a modifiedsample. In case one expects radiation effects, one should doits best to take a first spectrum in a short time, ideally in amatter of seconds, thereby strongly limiting the possibilitythat one measures a spectrum of an already damaged sample.Arrio and co-workers have analyzed a number of metalcomplexes in detail, using the charge transfer multipletmodel [67–70]. An important addition to the multipletmodel was the treatment of -(back)bonding, i.e. metalto ligand charge transfer (MLCT). The charge transfermultiplet model had been developed with respect tosolid state transition metal oxides that are dominatedby -bonding and by ligand to metal charge transfer(LMCT), i.e. the addition of 3d 9 L¯ configurations to a 3d 8ground state (for Ni II systems). Arrio and co-workers studiedmolecular-based magnets Cs[Ni II Cr III( CN) 6 ]–2H 2 O,Fig. 27. Ni L-edge spectra of (a) [Ni IV (S 2 CNEt 2 ) 3 ][BF 4 ] initial spectrum;(b) [Ni IV (S CNEt 2 ) 3 ][BF 4 ] successive spectra; (c) [Ni IV (S 2 CNEt 2 ) 3 ][BF 4 ]final spectrum; (d) [Ni II (S 2 CNEt 2 ) 2 ]; and (e) [PPh 4 ][Ni II (SCOEt) 3 ](reprinted with permission from [66], copyright 1998 Royal Society ofChemistry).Co II 3[Cr III (CN) 6 ] 2 –12H 2 O, Fe II 3[Cr III (CN) 6 ] 2 –18H 2 O,and Cs[Mn II Cr III (CN) 6 ]–2H 2 O. These systems are cubicwith the CN groups bridging two transition metal sites.The carbon is bonded to the Cr III and the nitrogen to thedivalent Ni, Co, Fe and Mn sites. The divalent octahedralnitrogen bonded sites are all high-spin and the 2p XASspectra of these sites can all be simulated nicely with thecharge transfer multiplet model using LMCT, for example,the Ni II site has a ground state of 90% 3d 8 and 10% 3d 9 L¯character. In contrast, the Cr III sites could not be described
F. de Groot / Coordination Chemistry Reviews 249 (2005) 31–63 59Fig. 28. (a) Experimental Cr III 2p XAS spectrum. (b) Theoretical Cr III 2pXAS spectrum calculated with the 3d 3 + 3d 2 L interaction configuration.(c) Theoretical Cr III 2p XAS spectrum calculated with 10Dq) 3.5 eVand the Slater integrals reduced to 50% and no configuration interaction(reprinted with permission from [67], copyright 1996 American ChemicalSociety).adequately with MLCT and the calculations using 3d 3 with3d 4 L character all failed to describe the 2p XAS spectrum.The problem is that, besides LMCT, there is also MLCT.In principle, one should include both effects in the calculations,but to limit the calculation, it was decided to includethe LMCT by reducing the Slater integrals and to includethe 3d 2 L configuration describing MLCT, explicitly.Fig. 28 shows that this model gave a good description ofthe spectral shape of the Cr III sites. In particular the structureat 583 eV is shown to originate from the MLCT tothe cyanide. This phenomenon is generally confirmed for-(back)bonding systems.A beautiful example of the application of the chargetransfer multiplet model are the 2p XAS and MCDmeasurements and simulations of two paramagnetichigh-spin molecules Cr{(CN)Ni(tetren)} 6 ](ClO 4 ) 9 and[Cr{(CN)Mn(TrispicMeen)} 6 ]–(ClO 4 ) 9 , 3THF (Fig. 29).The 2p XAS and X-MCD spectra were calculated in thecharge transfer multiplet model. The Mn II sites were againfound to be weakly covalent with some LMCT, while theCr III sites again are affected by MLCT and the 3d 3 + 3d 2 Lcalculation yields a good description of the spectrum. Inparticular the X-MCD spectral shape is near perfectly reproduced,indicating that the two-state model is adequateto describe the present system. Comparing the sign of theX-MCD with the calculations, one finds that the Mn II sitesare antiferromagnetically coupled to the Cr III sites.An often studied system is [Fe II (phen) 2 (NCS) 2 ] (phen= 1,10-phenanthroline), which is well known for its spin-flipbehavior under external conditions. The octahedral Fe II siteschange their spin, for example, as a function of temperature,from high-spin S = 2 to low-spin S = 0, i.e. two electronsflip their spin states. This system was often studiedwith X-ray spectroscopic techniques. Briois et al. used 1sXAS and 2p XAS. Ligand field multiplet analysis of the 2pXAS spectral shapes found large changes in the crystal fieldFig. 29. (a) Experimental Mn 2p XAS spectrum of[Cr{(CN)Mn(TrispicMeen)} 6 ](ClO 4 ) 9 ,3THF, paralel (thick line) andanti-parallel (thin line) (top). Experimental (dots) and theoretical (line)X-MCD signal normalized to 100% circular polarized light (bottom).(b) The Cr 2pXAS and X-MCD spectra (reprinted with permission from[69], copyright 1999 American Chemical Society).value that is reduced from 2.5 eV for low-spin to 0.5 eV forhigh-spin [68]. Collison and co-workers studied the samesystem and performed soft X-ray induced spin transitions bytrapping the excited spin state. They performed a detailedstudy of the 2p XAS spectra as a function of temperature andX-ray excitation energies, thereby creating second high-spinstate different from the original [66]. Recently, Vanko hasstudied the same system using 1s2p and 1s3p resonant XES[71].3.4. The differential orbital covalence derived from 2p XASThe charge transfer multiplet analysis of 2p XAS andother spectroscopies gives good simulations on the spectralshape, but on the other hand, it makes a direct understandingof the spectral shape difficult. Fig. 11 showed that itcan be useful to turn off the multiplet effects once a goodsimulation was obtained. By turning off the multiplet effectsone obtains the one-electron analog of the 2p XASspectrum. Wasinger et al. have followed a similar approachrecently [72]. They extended the charge transfer multiplettheory, to derive the differential orbital covalence (DOC)directly from the 2p XAS spectral shape analysis (Fig. 30).
Page 3 and 4: F. de Groot / Coordination Chemistr
Page 19 and 20: The formation of small satellites o
Page 27: F. de Groot / Coordination Chemistr
Page 33: F. de Groot / Coordination Chemistr

Multiplet Effects in X-ray Absorption - Inorganic Chemistry and ...

Create successful ePaper yourself

Delete template?

Save as template?