The new +2 oxidation states for the rare-earth metals Y, La, Ce, Pr, Gd, Tb, Ho, Er, and Lu were recently discovered by reduction of Cp x 3 Ln (Cp x = C 5 H 4 SiMe 3 , C 5 H 3 (SiMe 3 ) 2 ; Ln = rare-earth metal) using alkali metal reductants Li, Na, K, and KC 8 (Fig. 1) (Hitchcock et al., 2008;MacDonald et al., 2013;Fieser et al., 2015;Evans, 2016;Palumbo et al., 2018). In each of these cases, 2.2.2-cryptand was added in these reactions to encapsulate the alkali metal. It was thought that chelating agents were necessary to sequester the alkali metal to prevent interactions with cyclopentadienide ligands and subsequent ligand dissociation leading to product decomposition. This idea was challenged by examining reduction reactions of Cp 00
3 M (Cp 00 = C 5 H 3 (SiMe 3 ) 2 ; M = La, Ce, U) with Li and Cs in the absence of chelating agents (Huh et al., 2018). The reaction resulted in the isolation of the first chelate-free synthesis of La II , Ce II , and U II complexes. The [Li(THF) 4 ] 1+ cation of the Li salts in these chelate-free M II complexes were wellseparated from the (Cp 00 3 M) 1À anion. However, the Cs reductions yielded polymeric complexes of general formula [Cp 00 M(-Cp 00 ) 2 Cs(THF) 2 ] n where the Cs cation has coordinated THF and cyclopentadienide ligands. Attempts to extend this chemistry to smaller rare-earth metals by reduction of Cp 0 3 Ln (Cp 0 = C 5 H 4 SiMe 3 ; Ln = Y, Tb, Dy) showed evidence of Ln II in solution; however, the reduction products were highly unstable and decomposed even at 238 K.
In this study, we were interested in examining the reduction of Cp 0 3 Yb III with Cs metal. Unlike Y II , Tb II , and Dy II ions, Yb II complexes are more easily obtainable, as reflected by their less negative reduction potentials (Morss, 1976). A crystal containing the oligomeric compound; [(THF)Cs(-5 : 5 -Cp 0 ) 3 Yb] n , 1 (Cp 0 = C 5 H 4 SiMe 3 ) was isolated by reduction of the Cp 0 3 Yb III complex (Fieser et al., 2015) in THF using Cs metal (Figs. 2 and3).
All three Cp 0 rings remain coordinated to the Yb metal center after reduction and are coordinated in a trigonal-planar fashion. The Yb atom is within 0.107 A ˚of the plane of the three ring centroids. Each ring bridges Yb to Cs, which also is surrounded by three cyclopentadienyl ligands as well as a coordinated molecule of THF. The three ring centroids and the oxygen of THF are arranged in a pseudo-tetrahedral geometry around Cs with a calculated four-coordinate Cs 0 4 value of 0.76 ( 0 4 = 1 for tetrahedral; 0 4 = 0 for square planar; Rosiak et al., 2018). The Cs metal center has a pseudo-tetrahedral geometry with Cp 0 (centroid)Á Á ÁCsÁ Á ÁCp 0 (centroid) angles of 109.0, 114.3, and 121.4 and Cp 0 (centroid)Á Á Á CsÁ Á ÁO(THF) angles of 88.8, 94.1, and 127.8 .
The bond distances and angles in 1 are summarized in Table 1. The range of 2.504 (1)-2.513 (2) A ˚Cp 0 (centroid)Á Á Á Yb bond distances in 1 is the same as that in the complex [K(crypt)][Cp 0 3 Yb II ] (crypt = 2.2.2-cryptand), which was
Figure 2 Synthesis of [(THF)Cs(-5 : 5 -Cp 0 ) 3 Yb II ] n , 1, by caesium metal reduction of the Cp 0 3 Yb III precursor.
Figure 3 ORTEP representation of an asymmetric unit of [(THF)Cs(-5 : 5 -Cp 0 ) 3 Yb] n , 1, with probability ellipsoids drawn at the 50% probability level. Hydrogen atoms were omitted for clarity.
Table 1 Selected bond distances and angles for [(THF)Cs(-5 : 5 -Cp 0 ) 3 Yb] n , 1. Table 2 Bond distance (A ˚) ranges for YbÁ Á ÁCp 0 (centroid) and bond angle ( ) ranges for Cp 0 (centroid)Á Á ÁYbÁ Á ÁCp 0 (centroid) in Cp 0 3 Yb (Fieser et al., 2015), [K(crypt)][Cp 0
3 Yb] (Fieser et al., 2015), and [(THF)Cs(-5 : 5 -Cp fully characterized as a 4f 14 Yb II complex, Table 2 and Fig. 4.
In Cp 3 Ln reduction chemistry, the difference in LnÁ Á ÁCp(centroid) distances between the Ln III and Ln II complexes provides important information on the electronic configuration of the lanthanide ion (Evans, 2016). Differences in LnÁ Á ÁCp(centroid) distances for reduction of 4f n Ln III ions to 4f n+1 Ln II ions range from 0.1 to 0.2 A ˚ (Fieser et al., 2015).
In this study, the difference of 0.14 A ˚in the LnÁ Á ÁCp(centroid) distance is characteristic of a 4f 13 Yb III reduction to a 4f 14 Yb II ion. In contrast, Ln II ions with 4f n 5d 1 configurations where the additional electron populates a d-orbital instead of the an f-orbital have differences of only 0.02-0.05 A ˚ (Evans, 2016).
In 1, all of the cyclopentadienyl ligands are bridging. The threefold symmetry of three bridging Cp 0 ligands on each metal generates a hexagonal pattern as shown in Fig. 5. The YbÁ Á ÁCp 0 (centroid)Á Á ÁCs angles are 172.5-176.7 such that each side of the hexagon is nearly linear. The 112.4-117.3 YbÁ Á ÁCsÁ Á ÁYb angles are smaller than the 120.8-125.6 CsÁ Á ÁYbÁ Á ÁCs angles, which makes the hexagon slightly irregular. This could be of interest to quantum scientists trying to make thin-film layers of magnetic materials since the hexagonal pattern could lead to spin frustration with a paramagnetic lanthanide.
The side view of these layers in Fig. 6 shows how the space in between them is filled with THF and Me 3 Si substituent groups. The 116.6-122.8 Cp 0 (centroid)Á Á ÁYbÁ Á ÁCp 0 (centroid) and 109.0-121.4 Cp 0 (centroid)Á Á ÁCsÁ Á ÁCp 0 (centroid) angles generate the undulation of the hexagons shown in Fig. 6. coordinated by only two cyclopentadienyl ligands to give a bent metallocene Cp 00 2 Cs(THF) 2 sub-structure with these larger rings.
A survey of the Cambridge Structural Database (CSD, version 5.41, March 2020; Groom et al., 2016) also revealed four oligomeric complexes containing Yb-Cp x moieties with various types of cyclopentadienyl rings (Cp x ): [Na(-5 : 5 -C 5 H 5 ) 3 Yb II ] n (Apostolidis et al., 1997), [Na(-5 : 5 -Cp 00 ) 2 Yb II 2 (-5 : 5 -Cp 00 ) 2 ] n (Voskoboynikov et al., 1997), [(C 5 Me 5 )Yb(-I)(-5 : 5 -C 5 Me 5 )Yb(C 5 Me 5 )] n (Evans et al., 2006) and [Yb(-5 : 5 -C 5 H 5 )(Ph 2 Pz)(THF)] n (Ph 2 Pz = 3,5diphenylpyrazolate) (Ali et al., 2018). The [Na(-5 : 5 -C 5 H 5 ) 3 Yb II ] n (Apostolidis et al., 1997) complex adopts a hexagonal net extended structure similar to that in 1 except the alkali metal does not have a coordinated solvent. The structure of [Na(-5 : 5 -Cp tBu ) 3 Sm II ] is similar (Bel'sky et al., 1990). Three oligomeric complexes containing Cs-cyclopentadienyl moieties have previously been reported: [(THF) 2 Cs][(-5 : 5 -Cp 00 ) 2 U II ( 5 -Cp 00 )] n (Huh et al., 2018), {[(Me 3 Si) 2 NCs] 2 [(C 5 H 5 ) 2 Fe)]Á0.5(C 6 H 5 Me)} n (Morris et al., 2007) and Moral et al., 2005). An oligomeric, base-free Li-Cp 0 compound was also previously reported in the literature, [(-5 : 5 -Cp 0 )Li] n (Evans et al., 1992).
In an argon-filled glovebox, addition of a red solution of Cp 0
3 Yb (50 mg, 0.085 mmol) in THF (2 mL) to excess Cs as a smear produced a green solution. This was stirred for 15 min at room temperature and then layered at the bottom of a vial below an Et 2 O (10 mL) layer for crystallization at À35 C. After 1 d, X-ray quality dark-green crystals of [(THF)Cs(-5 : 5 -Cp 0 ) 3 Yb II ] n were isolated. A small number of crystals were obtained and used for crystallographic analysis. Too little sample was available for other characterization.
Crystal data and structure refinement for [(THF)Cs(-5 : 5 -Cp 0 ) 3 Yb II ] n , 1 are summarized in Table 3. Hydrogen atoms were included using a riding model with U iso (H) values of 1.2U eq (C) for CH 2 and aromatic hydrogens and 1.5U eq (C) for CH 3 hydrogens with C-H distances of 0.99 (CH 2 ), 0.95 (aromatic), and 0.98 A ˚(CH 3 ). (Sheldrick, 2008); software used to prepare material for publication: SHELXTL (Sheldrick, 2008).
Crystal data
Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. Refinement. A green crystal of approximate dimensions 0.079 x 0.086 x 0.148 mm was mounted in a cryoloop and transferred to a Bruker SMART APEX II diffractometer. The APEX2 program package was used to determine the unitcell parameters and for data collection (90 sec/frame scan time for a sphere of diffraction data). The raw frame data was processed using SAINT and SADABS to yield the reflection data file. Subsequent calculations were carried out using the SHELXTL program. The diffraction symmetry was 2/m and the systematic absences were consistent with the monoclinic space group P21/n that was later determined to be correct. The structure was solved by dual space methods and refined on F2 by full-matrix least-squares techniques. The analytical scattering factors for neutral atoms were used throughout the analysis. Hydrogen atoms were included using a riding model. The structure is polymeric. Least-squares analysis yielded wR2 = 0.0562 and Goof = 1.017 for 316 variables refined against 8223 data (0.75 Å), R1 = 0.0315 for those 6580 data with I > 2.0sigma(I).
Computer programs: APEX2(Bruker, 2014), SAINT(Bruker, 2013), SHELXT2014/4 (Sheldrick, 2015a), SHELXL2014/7(Sheldrick, 2015b) and SHELXTL(Sheldrick, 2008).
Acta Cryst. (2020). E76, 1131-1135 Huh et al. [CsYb(C 8 H 13 Si) 3 (C 4 H 8 O)]