Element 70 – Ytterbium

Ytterbium is the penultimate element of the lanthanoid series. It is thus one of the heavy lanthanoid elements and lies between thulium and lutetium. Its discovery is attributed to the Swiss chemist Jean Charles Galissard de Marignac in 1878. It was the fourth element (the others are yttrium, erbium, and terbium) to be isolated from the mineral ‘ytterbite’, later named gadolinite (Fig. 1), which was obtained from the feldspar mine in the village of Ytterby on Resarö Island in the vicinity of Vaxholm in Sweden. These days, it is usually sourced from monazite though it is present in higher amounts in the less common mineral xenotime, which is a heavy rare earth mineral, substantially YPO₄. The name ytterbium is the ultimate tribute to Ytterby.

Fig. 1.  A photo of a plaque erected at the Ytterby mine, Sweden, from which the four highlighted lanthanoids were mined and later isolated. The overlay shows the lanthanoid series in addition to the transition (d-block) elements scandium (Sc) and yttrium (Y), which have considerable chemical similarities.

Initially, ytterbium salts were separated from other lanthanoids by fractional crystallisation. Next, ion exchange was used, but currently counter-current solvent extraction is the method of choice.^[1–6a]

Following back-extraction into aqueous solution, Yb³⁺ can be precipitated as the hydroxide or the oxalate to provide Yb₂O₃, a major commercial product, by thermolysis, or can be converted into the hydrated chloride (Scheme 1).

Scheme 1.  Isolation of ytterbium as the oxide, halide or metal.

To obtain the free metal requires a powerful reductant as Yb metal is highly electropositive. The oxide can be reduced to Yb metal by metallic La, which is more electropositive than Yb or the oxide can be converted into the trifluoride with HF at 700°C under Ar, and the fluoride can be reduced by Ca.^[6a] The metal has a metallic sheen (Fig. 2) but tarnishes rapidly in the air owing to oxidation.

Fig. 2.  A photo of Yb metal (photo courtesy of Z. Guo, School of Chemistry, Monash University).

Ytterbium has a limited number of uses. It can be used to improve the mechanical properties of stainless steel, and has applications in lasers, fibre optic cables, and portable X-ray devices.^[1] The salts, particularly the triflate, have uses in organic synthesis, mainly as powerful Lewis acids.^[2] An exciting potential use is development of photosensitized NIR emission in biological analysis and sensing.^[6b]

The electronic configuration of the element is [Xe]4f¹⁴6s², and the two main oxidation states are III (4f¹³) and II (4f¹⁴). Thus, it is one of the three elements (with samarium and europium) long known to have an accessible +II state, though now this state has been achieved for all rare earth elements except Pm.^[7]

At this stage, Yb zero-valent compounds have not been achieved, though they are known for most lanthanoids.^[8] Formation of zero-valent compounds requires either a 4fⁿ⁻¹5d¹6s² configuration for the element or easy access to it, a situation not obtained for Eu or Yb, which have the most stable divalent states.

The redox potential is −1.05 V (cf. Eu³⁺ −0.54 V, Sm³⁺ −1.55V).^[6a] Thus, Yb²⁺ reduces water, , and Yb^II compounds must be prepared and handled in an inert atmosphere. The redox potential is affected by the ligands attached. Thus E⁰ −1.5 V (versus Fc⁺/Fc) indicates YbCp₃ is somewhat harder to reduce than Yb³⁺ ions.^[9] Yb^II is obviously favoured by a closed-shell configuration. It has been proposed that the redox transmetallation between Hg(C₆F₅)₂ and Yb metal,

proceeds through a univalent intermediate (C₆F₅)Yb–Hg(C₆F₅)^[10] and a recent theoretical study^[11] provides support for this hypothesis by showing that it is energetically favoured, but no Yb^I compounds have been isolated.

The reductive ability of Yb^II can be advantageous where the less reducing Eu^II is ineffective and the more powerful Sm^II leads to over-reduction. Thus [Yb^II(p-HC₆F₄N(CH₂)₂NR₂)₂(thf)₂] (R = Me, Et; thf = tetrahydrofuran), formed by metathesis from YbI₂, by cleavage of [Yb{N(SiMe₃)₂}₂] with p-HC₆F₄NH(CH₂)₂NR₂ and by redox transmetallation between Yb, Hg(C₆F₅)₂, and p-HC₆F₄NH(CH₂)₂NR₂ (Scheme 2) has an o-F positioned for defluorination, and underwent C–F activation on heating in toluene to form the Yb^III cages, [Yb₄(p-HC₆F₄N(CH₂)₂NR₂)₆F₆] with seven- and eight-coordinate Yb atoms (Scheme 2).^[12]

Scheme 2.  Formation of [Yb₄(p-HC₆F₄N(CH₂)₂NR₂)₆F₆].

However, the Eu^II analogue [Eu(p-HC₆F₄N(CH₂)₂NEt₂)₂(thf)₂] only underwent C–F activation on irradiation with light, giving [Eu₄(p-HC₆F₄N(CH₂)₂NEt₂)₆O₂F₂],^[13] whereas Sm metal with Hg(C₆F₅)₂ and (p-HC₆F₄NH(CH₂)₂NMe₂) gave a complex mixture probably containing much SmF₃ and from which a few crystals of [Sm(p-HC₆F₄N(CH₂)₂NMe₂)₂F]₃ were isolated with a hexanuclear (SmF)₃ core and all ligands showing N,N′,F tridentate coordination.^[13] Thus the redox-induced C–F activation is straightforward only for ytterbium.

As a result of the lanthanoid contraction, Yb²⁺ and Yb³⁺ are the second smallest in the lanthanoid series. The eight-coordinate ionic radii are 1.14 and 0.985 Å respectively with Yb³⁺ smaller than Y³⁺ (1.02 Å) of the previous period.^[6a] Yb³⁺ is a good proxy for the more expensive Lu³⁺ provided redox does not intervene. In the Ln(O₃SCF₃)₃ catalytic acylation of arenes by (MeCO)₂O in nitromethane, Yb(O₃SCF₃)₃ was the standout performer^[14] (see, however, later use of Sc(O₃SCF₃)₃^[15] and Yb(O₃SCF₃)₃/LiClO₄ < Sc(O₃SCF₃)₃/LiClO₄ in reactivity^[16]). More recently, a remarkable C–F activation reaction was achieved by YbI₃ converting unactivated alkyl fluorides into the corresponding iodides:^[17]

YbI₃ was the most effective reagent of all LnI₃ compounds examined (LnI₃: Ln = La, Sm, Dy, Yb) and in competitive reactions with the corresponding chlorides and bromides, there was 100 % conversion for the fluoride and virtually no reaction of the other halides.^[17]

Size is also relevant to coordination chemistry with the lanthanoid contraction affecting the behaviour of Yb³⁺ relative to other lanthanoids (see below). Both Yb²⁺ and Yb³⁺ are strong Lewis acids and hard acids; hence, they prefer first-row donors, notably oxy-donors. High coordination numbers of seven to ten are particularly preferred, with eight-coordinate [Yb(H₂O)₈]³⁺ the dominant species in aqueous solution, and [Yb(H₂O)₉]³⁺ is present in the hydrated triflate, bromate, and ethylsulfate.^[18] High coordination numbers of eight to ten are particularly favoured with small organohydroborate and tetrahydroborate ligands, which can act as tridentate donors.^[19] Low coordination numbers can be achieved with bulky donors.^[20] Thus for Yb^II, three-coordination is observed in [Yb(OAr)(μ-OAr)]₂ (OAr = OC₆H₂^tBu₂-2,6-Me-4)^[21] and in monomeric [Yb{N(Dip)(Mes)}₂(thf)] (Dip = C₆H₃ⁱPr₂-2,6; Mes = C₆H₂Me₃-2,4,6),^[22] whereas two-coordination is observed in [Yb{C(SiMe₃)₃}₂],^[23,24] but there are additional C–Yb interactions and C–Yb–C is 137.0(4)°.^[23] More recently, [Yb{N(SiiPr₃)₂}₂] was prepared and has a near-linear N–Yb–N bond (166.01(14)°) but there are six additional agostic C–Yb interactions.^[7h] In the trivalent state, [Yb{N(SiMe₃)₂}₃]^[25] and [Yb(OAr)₃]^[26] provide examples of three-coordination. As to the effect of the lanthanoid contraction (0.18 Å from La³⁺ to Yb³⁺) on coordination number, decreases are well known, e.g. from [LaCp₃] (coordination number (CN) = 11) to YbCp₃ (CN = 9)^[27] or [LaCl₃(thf)₂] (eight-coordinate polymer) to [YbCl₃(thf)₂]₂ (six-coordinate dimer).^[28] There are cases where no change occurs across the series particularly with use of suitable macrocyclic ligands to override the Ln contraction.^[29]

Like organolanthanoids generally, ytterbium organometallics were late on the scene, beginning with tri(cyclopentadienyl)ytterbium,^[30] a nine-coordinate monomer (see above). The divalent analogue [YbCp₂] followed.^[31a] The controversy as to whether the compound was red^[31a–c] or green^[31d] was resolved by determining the crystal structure of red [YbCp₂] and green [NaYbCp₃], the latter inferentially being green YbCp₃.^[32] Cyclopentadienyls and other π-complexes have continued to dominate organolanthanoid chemistry.^[33] The divalent state has extended to the C₅Me₅ analogue^[34] and, more recently, to the super bulky [Yb(Cp^BIG)₂] (Cp^BIG = C₅(C₆H₄Bu-4)₅)^[35] and [Yb(C₅Ph₅)₂].^[36] The last two have coplanar rings, slightly bent towards one another, by contrast to the open sandwich of [Yb(C₅Me₅)₂].^[37] Remarkably, the latter reacts with [Pt(Ph₃P)₂(C₂H₄)],^[38a] disubstituted acetylenes,^[38b] and CO.^[38c]

Although ytterbium complexes with mixed cyclopentadienyl and alkyl or aryl ligands are quite common,^[33] species in which alkyl or aryl groups are the sole anionic organic ligands are rarer. Ytterbium pseudo-Grignard compounds YbR(X) (e.g. R = Ph, Me; X = Br, I) were the first representatives^[39] and structurally characterised examples were achieved with bulky ligands,^[23,24,40] e.g. C₆H₃Ph₂-2,6. Although Yb(Ph)I cannot be isolated, it is an excellent source of pyrazolato- and formamidinato-ytterbium complexes by reaction with 3,5-diphenylpyrazole^[41a] and N,N′-diaryl-formamidines^[41b] at low temperatures. Well-characterised alkyl- and aryl-ytterbium compounds include [Li(tmed₃)][YbMe₆] (tmed = N,N′-tetramethylethane-1,2-diamine),^[42] [YbPh₃(thf)₃],^[43] the mixed oxidation state [Yb^IIIPh₂(thf)(π-Ph)₃Yb^II(thf)₃],^[44] [Yb(C₆F₅)₂(thf)₄],^[10,43b,45] and [Yb(CH₂^tBu)₃(thf)₂].^[46] Synthetic routes used to prepare organoytterbium compounds include metathesis,^[2,33] redox transmetallation with mercurials^{[10,11,43–45]} and organothallium compounds,^[47] protolysis for cyclopentadienyls,^[33] and redox transmetallation/protolysis for cyclopentadienyls.^[48]

In summary, ytterbium is one of the most versatile lanthanoid elements, with two readily accessible oxidation states, strong Lewis acidity in the trivalent state, versatile coordination chemistry in both oxidation states, and attractive organometallic chemistry. As the number of uses increases, interest in recycling will develop and it is possible that the existence of the Yb^II state may assist in separating it from other metals even though it has not been utilised in lanthanoid separation. Our choice of element 70 for this essay was also influenced by the recent 70th birthday of a great friend and colleague, Professor Dr Gerd Meyer, as celebrated by a special issue of Z. Anorg. Allg. Chem. 2019, 645, 870.

The authors declare no conflicts of interest.