Notes For Experiment 4 (The Preparation and Properties of Organo-cobaloxime Complexes)

14 September 2001

Dr John Maher, School of Chemistry, Bristol University

Cobalt(II)

This is a d7 ion. With weak ligands such as H2O, Co(II) has octahedral co-ordinationwith the metal in the high spin configuration. [Co(H2O)6]2+ion is pink to rose red coloured in its salts and solution (p730).Aqueous Co(II) complexes are labile. (p1289).

Solvents with a lower dielectric constant than water, such as ethanol or acetone, cause the Co(II) to adopt a tetrahedral co-ordination instead, this has a strong blue colour (p730). (cf. the blue colour of dry silica gel - used in dessicators). The tetrahedral form is also favoured by ligands such as Cl- and CNS-. Co(II) forms more tetrahedral complexes than any other transition metal ion. Ligand field stabilisation disfavours tetrahedral over octahedral much less than for the other dnconfigurations. Because of the small stability difference between the octahedral and tetrahedral co-ordinations, even in water there are small quantities of the tetrahedral form.(p727). Calculate the ligand field stabilisation energy for high spin d7 in tetrahedral and octahedral co-ordination (in units of

In aqueous solutions with no complexing ligands Co(II) cannot easily be oxidised to Co(III) (Eo = 1.84V). (p726) In the presence of complexing ligands such as NH3, en (1,2-diaminoethane) , dmgH-(dimethylglyoxinate ion), or in basic solution (OH-) the potential drops sharply ( Eo <0.2 V) (p726) . This lowering of the redox potential is accompanied by a change to low spin d7 . The strong Jahn-Teller distortion of the low spin ion causes the co-ordination chemistry of this ion to become rather complicated(p731).

Cobalt(III)

This is a d6 ion. Complexes of Co(III) are very numerous, their ligand exchanges are slow, they are generally rather inert to substitution by other ligands. Nearly all examples are low spin. Because of their slow reactions inorganic kineticists and others have intensively studied them. To quote Cotton and Wilkinson, they have "from the days of Werner and Jorgensen (mid 1800s), been studied and a large fraction of our knowledge of the isomerism, modes of reaction, and general properties of octahedral complexes as a class is based on studies of Co(III) complexes." Co(III) has a particular affinity to nitrogen donors’(p732). Glyoxime ligands, Schiff bases and various macroscopic ligands form another important class. (p734)

Cobalt(I)

This is a d8 ion. This oxidation state is favoured by ligands such as phosphines, phosphites or isocyanides. (p738) It is also stabilised to some extent by a chelate ligand like dimethylglyoxinate, and by Schiff-base ligands. A disproportionation reaction is possible with phosphite ligand(p739):

2Co(II) --> Co(I) + Co(III)

It is likely that this reaction is also important in cobaloxime chemistry, but evidence is sparse. Co(I) cobaloximes are a blue, rather the same as the tetrahedral Co(II) ion. Borohydride ion can reduce Co(III) {and possibly Co(II)} to Co(I), since this relies on the substitution of H- into the otherwise rather inert Co(III) ion, it is likely to be favoured by strongly basic solutions

Co(III)-X + H- ---> Co(III)-H + X-, then Co(III)-H + OH- ---> Co(I) + H2O

Anyway it’s interesting to speculate about mechanisms like this!

See also Toscano and Marzilli (p116-121).

The Trans effect

This effect is not confined to square planar complexes (pp1299), but can also be observed in octahedral complexes. The alkyl group has quite a strong Trans effect in square planar complexes, and for the cobaloximes causes the pyridine and other Lewis bases co-ordinated to the cobalt to be labile. Ligand exchange rates for these are in the millisecond timescale. (Cotton and Wilkinson, p335) (Toscano and Marzilli pp145)

UV-visible spectra

You do not need to try to assign these transitions, but see Toscano and Marzilli p140. The two prominentabsorptions in the near ultraviolet giving the yellow colour of the complex are not simple d-d transitions. However the spectra are an indicator of the behaviour of the Co(III)-pyridine bond in the cobaloxime complexes in solution, compare the aqueous solution spectra with the spectrum at the end of the script. The latter was measured in dichloroethane, a solvent which will not co-ordinate to the cobalt, so that the spectrum shown on the script corresponds to that of the undissociated complex [CH3Co(dmgH)2(pyridine)].

Photolysis

The Co-C bond can be broken by visible or near-UV light. Naturally a homolytic cleavage of the Co-C bond leads to alkyl radicals. The formation of radicals can then be proved by spin-trapping the released alkyl radical with a spin trap such as N-tert-Butyl--phenylnitrone – PBN – then measuring the electron magnetic resonance (EMR) spectrum. There are other spin traps, nitrosobenzene is the simplest compound, but gives a much more complicated EMR spectrum due to hype4fine coupling to the aromatic ring hydrogens. Toscano and Marzilli pp147, p 166

(See:

Electron Magnetic Resonance Spectra

EMR (aka ESR, Electron Spin Resonance,and EPR, Electron Paramagnetic Resonance) is anoldercousin of NMR, having been discovered by Zavoisky in Russia over 50 years ago, and also having a much older ‘time-line’ ( than NMR. An EMR spectrum can usually be observed in materials with one or more unpaired electrons - free radicals, transition metal ions and solid defects. The formula for transitions between the electronic Zeeman levels is:

h0 = gB0

Where his Planck's constant;0 is the frequency, about 9.5 GHz, 9.5 x 109 Hz for a magnetic field B0 of 0.35 Tesla (called X-band); g is the g-factor;is the Bohr magneton. ge, the Zeeman splitting constant for the free electron is one of the most accurately determined of all physical constants, ge = 2.0023193004386(20). The free-electron ge is altered by interaction between the spin and the orbital momenta of the electron via spin-orbit coupling. In organic free radicals g is close to 2.00 since spin-orbit coupling for light nuclei is weak. However in transition metal complexes g can vary over a very large range due to increased spin-orbit coupling, but also from the presence of other unpaired electrons. An ion such as high-spin Fe(III) can have signals at g=2, g=4 and g=8. Thus g has some analogies with the chemical shift ,, in NMR spectroscopy, but the value can change over a much wider range.

Electron-nuclear hyperfine interactions are also observed, they are like the nuclear spin-spin coupling (J) in NMR spectroscopy, except that one partner is the electron. (Electron-electron spin-spin interactions are also possible in multiple electron systems). Here, AN and AH refer to the electron-nitrogen and electron-hydrogen hyperfine splitting constants respectively, these are measured in units of millitesla, mT, (The old unit is a Gauss, 1 Tesla = 104Gauss). For 14N I=1, and for 1H, I =½.

In EMR spectroscopy, the spectra are (traditionally) displayed as the first derivative of the absorption mode, this display mode is very sensitive to the width of the line. Many EMR spectra are characterised by large numbers of lines, for which this is an easier mode of display. The six lines displayed in the spectrum of the spin-trapped radical are all of equal intensity despite looking slightly different, the derivative display mode emphasisesvariation in line width in this fashion.

References

1) F.A.Cotton and G.Wilkinson, Advanced Inorganic Chemistry, Wiley Interscience, 5th Ed., 1988 . (See italic page numbers)

2) P.J. Toscano and L.G. Marzilli, "B12 and Related Organocobalt Chemistry", Progress in Inorganic Chemistry, 1984, 31, 104-204

3) W.Kaim and B Schwederski, "Bioinorganic Chemistry", Wiley 1994.

4) For more information about EMR see :

5) Experiment 4 notes are also at