Lanthanide (Substituted-)Cyclopentadienyl Bis(phosphinimino)methanediide Complexes: Synthesis and Characterization Authors: Fan Li, Yin-Cong Kong, Zi Yang, Chen Wang Published: 2024 ABSTRACT Design and synthesis of high-performance single-molecule magnets (SMMs) have long been a research focus. Inspired by the best dysprosium(III) metallocene SMMs and dysprosium(III) bis(methanediide) SMMs, we assumed dysprosium SMMs, which had electrical neutrality by combining the two types of ligands. As the Dy3+ center is coordinated by one (substituted-)cyclopentadienyl (CpR) ligand and one methanediide ({C(PPh2NSiMe3)2}2–) ligand on the axial sites, this ideal structure with linear Ccarbene–Dy–Cpcent would strengthen the magnetic anisotropy and exhibit excellent SMM properties. However, it is not easy to synthesize it in this configuration. We for the first time obtained [Y{C(PPh2NSiMe3)2}(CpR)(THF)] (CpR = Cp(1), CptBu(2), Cp*(3), Cp = C5H5, CptBu = C5H4tBu, Cp* = C5Me5) through the reactions of NaCp/KCpR and [Y{C(PPh2NSiMe3)2}(I)(THF)2]. In the molecular structures of 1–3, except for the two expected ligands, one coordination tetrahydrofuran (THF) molecule was also found in each complex. The P–C–P values of 1–3 were 135.46(7), 136.421(8), and 131.43(10)°, respectively, which were far less than 180°. The Ccarbene–Y–Cpcent of 1–3 deviated significantly from the linear shape, which was 118.021, 129.459, and 118.331°, respectively. Such a coordination environment makes the dysprosium congener [Dy{C(PPh2NSiMe3)2}(Cp*)(THF)] (4) whose Ccarbene–Dy–Cpcent (118.295°) was too small to maintain axiality and which almost exhibited no SMM properties. Even though the first exploration of lanthanide (substituted-)cyclopentadienyl bis(phosphinimino)methanediide complexes did not live up to our expectation, it provided great experiences for the future success of high-performance lanthanide (substituted-)cyclopentadienyl methanediide SMMs. Introduction Single-molecule magnets (SMMs) have attracted much attention for their potential application in high-density information storage, molecular spintronics, and quantum computing.1−3 The first SMM (Mn12)4 and many early-stage SMMs5−8 are based on transition metals. However, in the recent two decades, there has been an exponential increase in the number of publications on lanthanide SMMs9−16 for their easy-to-get magnetic anisotropy, which is essential for improving SMM properties.17 SMM is also a coordination complex. How to regulate the magnetic anisotropy of lanthanide ions through different ligands has attracted organolanthanide chemistry researchers’ interest. Among the best-performing Ln-SMMs, a lot of them are dysprosium single-ion magnets (SIMs), which also were mononuclear lanthanide coordination compounds or organometallics, giving us a chance for understanding the magneto-structural correlations and designing high-performance SMMs.17 Substituted or unsubstituted cyclopentadienyl, here abbreviated as CpR, has been applied in many lanthanide complexes for its rigid ring and concentrated charge. Layfield and Mills used the Cpttt (Cpttt = C5H2tBu3) ligand and weakened the equatorial coordination to synthesize dysprosium(III) metallocene SMMs,18,19 which was a breakthrough in that time. Then, they extended the dysprosium(III) metallocene series20−24 by regulating the substituted group on CpR and made [(CpiPr5)Dy(Cp*)]+23 (CpiPr5 = C5iPr5, Cp* = C5Me5) the best mononuclear SMMs by now. Gao, Tong, and Zheng used the steric bulk monodentate N-ligand or O-ligand to reinforce the axiality for oblate DyIII,25−29 which were successful for building high-performance SMMs. Introducing Ln–N/C multiple bonds30−32 in the axial axis would be advantageous for improving SMM behaviors. Liddle, who has made much contribution in the field of Ln–C multiple bonds (mainly pincer-type lanthanide carbene complexes),32,33 brought in the methanediide dianion and constructed SMMs characterized with almost linear C = Dy = C units with short average Dy–C distances, supporting the hypothesis that a more linear axial ligand field with shorter M–L distances produces enhanced SMM properties.34,35 However, all of the reported Ln-SMMs are not good enough for practical use. There has been an impetus toward the search for a new coordination environment of Ln-SMMs. The above strategies inspired us to construct possible high-performance Ln-SMMs, which employed one CpR ligand and one methanediide ligand on the axial sites to strengthen the magnetic anisotropy. This will be a remarkable challenge for synthetic chemists to keep this geometry. Here, we report our progress on this hypothesis. The raw materials were replaced by semblable yttrium complexes to explore the synthesis route because Y3+ and Dy3+ show similar chemical properties and yttrium complexes are diamagnetic for monitoring the reaction. Through the reactions of alkali metal cyclopentadienide and yttrium iodide [Y{C(PPh2NSiMe3)2}(I)(THF)2],36 we successfully obtained [Y{C(PPh2NSiMe3)2}(CpR)(THF)] (CpR = Cp(1), CptBu(2), Cp*(3), Cp = C5H5, CptBu = C5H4tBu). Due to the coordination of tetrahydrofuran (THF), the Ccarbene–Y–Cpcent of 1–3 deviated significantly from the linear shape, which was 118.021, 129.459, and 118.295°, respectively. Only the dysprosium congener of 3, [Dy{C(PPh2NSiMe3)2}(Cp*)(THF)] (4), was isolated. The Ccarbene–Dy–Cpcent (118.295°) is too small to maintain axiality, which made 4 exhibit almost no SMM properties. Results and Discussion Synthesis and Spectroscopic Characterization We first investigated Cp, since it has less steric bulk and is easily obtained, to determine the feasibility of our methodology. The yttrium bis(phosphinimino)methanediide complex [Y{C(PPh2NSiMe3)2}(I)(THF)2] was treated with NaCp. 1H NMR and 31P{1H}NMR spectral monitoring of the reaction in C6D6 and a few drops of d8-THF showed the formation of new complexes 1 at room temperature. The scaled-up reaction in THF provided mononuclear yttrium complexes [Y{C(PPh2NSiMe3)2}(Cp)(THF)] (1) in 43% yields (Scheme 1). With our methodology established, we then investigated synthesizing the yttrium analogue with bulk-substituted cyclopentadienyl. CptBu and Cp* were selected and successfully adapted in this way. [Y{C(PPh2NSiMe3)2}(CptBu)(THF)] (2) was obtained in 46% yields with heating at 50 °C, while [Y{C(PPh2NSiMe3)2}(Cp*)(THF)] (3) was obtained in 28% yields at 75 °C (Scheme 1). The complexes 1–3 were characterized by NMR spectroscopy (Figures S1–S9) and elemental analysis (EA). The 1H NMR spectra of 1–3 each exhibit signals for CpR fragments (δH: 1, 6.75 (s, 5H, Cp); 2, 6.73 (t, 3JHH = 2.5 Hz, 2H, CptBu–CH), 6.22 (t, 3JHH = 2.7 Hz, 2H, CptBu–CH), 1.59 (s, 9H, CptBu-CMe3); 3, 2.23(Cp*-Me)). Their 13C{1H} NMR spectra contain a distinctive doublet of triplet signals for the methanediides due to 31P and 89Y coupling (δC: 1, 51.4, 1JCP = 131.2 Hz, 1JCY = 6.2 Hz; 2, 54.6, 1JCP = 137.5 Hz, 1JCY = 7.5 Hz; 3, 53.7, 1JCP = 123.8 Hz, 1JCY = 8.8 Hz) (Table 1), and 89Y coupling is observed in their 31P{1H} NMR spectra (δP: 1, 3.2, d, 2JPY = 12.0 Hz; 2, 5.8, d, 2JPY = 14.0 Hz; 2, 5.9, d, 2JPY = 14.0 Hz) (Table 1). These resonances are comparable to the corresponding data for [Y{C(PPh2NSiMe3)2}(I)(THF)2] (δC 60.28, 1JCP = 207 Hz, 1JCY = 5 Hz; δP 3.48, 2JPY = 13 Hz)36 and other complexes of the general formula [Y{C(PPh2NSiMe3)2}(X)(THF)] (X = anionic ligand).37−40 The four complexes were characterized by elemental analysis (EA), where EA values obtained are >0.5% out from expected values, maybe due to the uncertainty of residual toluene. Table 1Characteristic NMR Data of Complexes 1–3 in C6D6 at 25 °Ccomp.13C{1H}: Ln = C (ppm)31P{1H}: Ln = C–P (ppm)151.4 (td, 1JCP = 131.2 Hz, 1JCY = 6.2 Hz)3.2 (d, 2JPY = 12.0 Hz)254.6 (td, 1JCP = 137.5 Hz, 1JCY = 7.5 Hz)5.8 (d, 2JPY = 14.0 Hz)353.7 (td, 1JCP = 123.8 Hz, 1JCY = 8.8 Hz)5.9 (d, 2JPY = 14.0 Hz) Characteristic NMR Data of Complexes 1–3 in C6D6 at 25 °C Synthesis of Complexes 1–3 from the Compound [Y{C(PPh2NSiMe3)2}(I)(THF)2] We tried to expend these synthetic methods to the dysprosium congeners of the three yttrium (substituted-)cyclopentadienyl methanediide complexes, but only one complex [Dy{C(PPh2NSiMe3)2}(Cp*)(THF)] (4) was isolated in total yields of 27% (Scheme 2). One of the main reasons probably was that the proposed intermediates “[Dy(Bn)2(I)(THF)3]” (Bn = CH2C6H5) and “[Dy{C(PPh2NSiMe3)2}(I)(THF)2]”, which were unsuccessfully isolated,41 are too active. It may need more harsh synthetic conditions to obtain dysprosium congeners of 1–2. As for the synthesis of 4 (or 3), which needs a higher temperature (75 °C) than 1 and 2, the properties of Cp* may be another reason for the success of the isolation of 4. Complex 4 was characterized by elemental analysis. Synthesis of Complex 4 Structural Characterization Crystals of 1–4 were formed from a toluene solution and proved suitable for structural determination by single-crystal X-ray diffraction. Both complexes 1 and 2 were crystallized in the monoclinic space group P21/c, while 3 and 4 were crystallized in the monoclinic space group P21/n (Table S1). Molecular structures of 1–4 are shown in Figures 1–4. The coordination environment of each center ion (Y3+ in 1–3 or Dy3+ in 4) is composed of one η5-cyclopentadienyl ligand (Cp, CptBu, or Cp*), one tridentate bis(phosphinimino)methanediide ({C(PPh2NSiMe3)2}2–), and one coordinated monodentate THF molecule. The {C(PPh2NSiMe3)2}2– dianion coordinates to the lanthanide center through C and P atoms. In complexes 1–3, the Y–Ccarbene (or Y–C16) bond lengths are 2.4260(12), 2.4446(13), and 2.4381(16) Å, respectively (Table 2). The {C(PPh2NSiMe3)2}2– intraligand bond lengths and angles are consistent with previously reported Y(III) methanediide.36−40 It is noteworthy that the P–C–P angle (1, 135.46(7)°; 2, 136.42(8); 3, 131.43(10)°) exhibited a significant deviation from the planarity of the pincer ligand scaffold compared with [Y{C(PPh2NSiMe3)2}(I)(THF)2] (172.5(2)°)36 and [Y{C(PPh2NSiMe3)2}2]− (168.9(2), 169.4(2)),34 giving the diagnostic pseudoboat conformation. However, this phenomenon was also found in complexes [Y{C(PPh2NSiMe3)2}(X)(THF)] (135.5(3)–139.24(14)°).37−40 The Y–Ccarbene bond length of 1–3 is slightly longer in comparison with the distances reported for [Y{C(PPh2NSiMe3)2}(I)(THF)2] (2.356(3) Å).36 The distances of Y3+ to the cyclopentadienyl center (Y–Cpcent) in 1–3 are 2.391, 2.423, and 2.421 Å, respectively (Table 2), which is not quite affected with the substituted group on the ring. Compared with the reported yttrium (substituted-)cyclopentadienyl complexes, the Y–Cpcent distance in 1 is obviously shorter by ca. 0.3 Å.42 The Y–Cpcent distance in 3 was in accordance with [Cp*Y(NC6H3(CF3)2-3,5)(THF)2]2.43 No structure data about Y–CptBu was found in the literature. The bite angles Ccarbene–Y–Cpcent of 118.021, 129.459, and 118.331° are found for 1–3, respectively (Table 2). It is noteworthy that the increased steric bulk from Cp to Cp* only makes the Ccarbene–Y–Cpcent enlarged by 0.31°. While comparing CptBu with Cp, the angle of Ccarbene–Y–Cpcent is obviously increased by ca. 11.4°. This observation can be attributed to the one tBu group which makes the CptBu ring lean to the pincer methanediide (Ccarbene–Y–CCtBu = 104.65(4)°). Thus, the Ccarbene–Y–Cpcent angle is enlarged in turn. Table 2Selected Bond Lengths (Å) and Angles (deg) of Complexes 1–3comp.Ln–CcarbeneLn–CpcentCcarbene–Ln–Cpcent1(Y)2.4260(12)2.391118.0212(Y)2.4446(13)2.423129.4593(Y)2.4381(16)2.421118.3314(Dy)2.4336(17)2.429118.295 Selected Bond Lengths (Å) and Angles (deg) of Complexes 1–3 Molecular structure of 1 with thermal ellipsoids at a 30% probability. All hydrogen atoms are omitted for clarity. Selected bond lengths (Å) and angles (deg): Y1–N1 2.3443(10); Y1–N2 2.3573(10); Y1–C16 2.4260(12); Y1–CCp: 2.6450(16), 2.6811(16), 2.6635(17), 2.6889(16), 2.6337(17), respectively; Y1–O1 2.375(4); Y1–O1A 2.414(11); P1–N1 1.6340(11); P1–C16 1.6694(11); P2–N2 1.6304(10); P2–C16 1.6681(12); O1–Y1–C16 132.50(12); O1A–Y1–C16 131.3(3); N1–Y1–N2 121.54(4); N1–Y1–C16 68.25(4); N1–Y1–O1 93.0(4); N1–Y1–O1A 93.9(9); N2–Y1–C16 67.88(4); N2–Y1–O1 89.4(3); N2–Y1–O1A 87.4(9); P1–C16–P2 135.46(7); P1–C16–Y1 89.84(5); P2–C16–Y1 89.83(5). Molecular structure of 2 with thermal ellipsoids at a 30% probability. All hydrogen atoms are omitted for clarity. Selected bond lengths (Å) and angles (deg): Y1–O1 2.3856(11); Y1–N1 2.3368(11); Y1–N2 2.3618(12); Y1–C16 2.4446(13); Y1–CCp: 2.7023(14), 2.6626(14), 2.7809(14), 2.7290(14), 2.6490(14), respectively; Y1–O1 2.3856(11); P1–N1 1.6324(11); P1–C16 1.6734(13); P2–N2 1.6295(12); P2–C16 1.6765(13); O1–Y1–C16 124.20(4); N1–Y1–N2 122.50(4); N1–Y1–C16 68.15(4); N1–Y1–O1 88.03(4); N2–Y1–C16 67.46(4); N2–Y1–O1 88.09(4); P1–C16–P2 136.42(8); P1–C16–Y1 89.02(5); P2–C16–Y1 90.33(5). Molecular structure of 3 with thermal ellipsoids at a 30% probability. All hydrogen atoms are omitted for clarity. Selected bond lengths (Å) and angles (deg): Y1–N1 2.4024(14); Y1–N2 2.3912(13); Y1–C16 2.4381(16); Y1–CCp: 2.6782(17), 2.6983(17), 2.6776(16), 2.7458(17), 2.7245(17), respectively; Y1–O1 2.4034(12); P1–N1 1.6382(14); P1–C16 1.6751(16); P2–N2 1.6304(10); P2–C16 1.6739(16); O1–Y1–C16 129.54(5); N1–Y1–N2 116.71(5); N1–Y1–C16 67.81(5); N1–Y1–O1 89.50(5); N2–Y1–C16 67.30(5); N2–Y1–O1 86.59(5); P1–C16–P2 131.43(10); P1–C16–Y1 90.37(7); P2–C16–Y1 90.67(7). Molecular structure of 4 with thermal ellipsoids at a 30% probability. All hydrogen atoms are omitted for clarity. Selected bond lengths (Å) and angles (deg): Dy1–N1 2.4025(15); Dy1–N2 2.3928(14); Dy1–C16 2.4336(17); Dy1–CCp: 2.6836(17), 2.7510(17), 2.7082(18), 2.7308(18), 2.6865(18), respectively; Dy1–Cpcent 2.429; Dy1–O1 2.4257(13); P1–N1 1.6334(15); P1–C16 1.6756(17); P2–N2 1.6319(15); P2–C16 1.6750(17); Cpcent–Dy1–C16 118.295; O1–Dy1–C16 129.47(5); N1–Dy1–N2 116.64(5); N1–Dy1–C16 67.81(5); N1–Dy1–O1 67.60(5); N2–Dy1–C16 67.19(5); N2–Dy1–O1 86.63(5); P1–C16–P2 131.59(10); P1–C16–Dy1 90.68(7); P2–C16–Dy1 90.98(7). Complexes 3 and 4 share the same coordinate ligands. In 4, if we treat Cp* as a point charge in the ring center, the geometry of the Cp* center and other three coordinate atoms would be a distorted pyramid. The Dy–Ccarbene bond length is 2.4336(17) Å, which is slightly longer than that of [Dy{C(PPh2NSiMe3)2}{CH(PPh2NSiMe3)2}] (2.3670 Å) but comparable with that of [Dy{C(PPh2NSiMe3)2}2][K(18C6)(THF)2] (2.434(6) and 2.433(6) Å).34 The Dy–Cpcent distance is 2.429 Å (Table 2), which is slightly longer than that of [Dy(DAD)Cp*(THF)] (DAD: [2,6-iPr2C6H3N–CMe = CM–NC6H3iPr2-2,6] (2.391(2) Å)).44 The angle Ccarbene–Dy–Cpcent is 118.295°, showing poor axiality. The shortest distance between the neighboring Dy3+ ions is 11.304 Å. Magnetic Characterization To explore the magnetic behavior of 4, static susceptibility measurements were performed under a direct current (1 kOe, dc) magnetic field over the temperature range from 2 to 300 K (Figure 5). The product (χmT) value of temperature-dependent magnetic susceptibility and temperature at 300 K was 13.29 cm3 mol–1 K, close to those expected for one free Dy3+ ion with ground terms of 6H15/2 (g = 4/3, χmT = 14.17 cm3 mol–1 K). the χmT values declined slowly with a decrease in temperature initially, then exhibited a sudden drop below 3 K to a minimum of 5.88 cm3 mol–1 K at 2 K. The zero-field-cooled/field-cooled (ZFC/FC) magnetizations (Figure S10) showed no clear divergences below 10 K. No magnetic hysteresis loop was found in the range of −50 to 50 kOe at 2 K with a field sweep speed of 500 Oe/s (Figures 6 and S11), proving poor/no SMM magnetic properties. The M(H) at 2 K saturates at a value of 4.55 β mol–1 (Figure S12). This value is a little lower than the expected saturation value of 5.23 β, which is likely due to crystal-field effects and low-lying excited states. The unsaturated magnetization together with the superimposed M vs HT–1 curves (Figure 6) at varying temperatures indicates the poor magnetic anisotropy in the systems.45 Temperature dependence of χmT or χm–1 for 4 under a 1 kOe dc magnetic field. M vs HT–1 plots at different temperatures for 4. We performed dynamic susceptibility measurements at nearly zero dc field to investigate further the magnetic behaviors of complex 4. No peak was observed in the out-of-phase alternating current susceptibilities (χm″) vs T plots at different frequencies (Figure S13), meaning that complex 4 did not behave like SMMs (as the data at 100 Hz fluctuated along a line, the peaks at this frequency were meaningless; the data above 100 Hz were counted). An obvious increase of χm″ as temperature declined below 7 K was attributable to unsuppressed quantum tunneling magnetization (QTM). These results were also in accordance with those of the dc measurements. Compared with the Dy SMMs sharing the Dy = C bond, like [Dy{C(PPh2NSiMe3)2}{CH(PPh2NSiMe3)2}] (effective energy barrier: Ueff = 177 cm–1) and [Ln{C(PPh2NSiMe3)2}2][K(18C6)(THF)2] (Ueff = 501 and 565 cm–1),34 no Ueff was observed in 4. Dy SMM with one Cp* ligand such as [Dy(DAD)Cp*(THF)] (Ueff = 254 cm–1)44 is also better than 4. Ab initio calculation about complex 4 exhibited that the MJ states are highly mixed (Table S2). The magnetic relaxation through the first excited state could not be discovered for the high possibility (0.330) of quantum tunneling magnetization (Figure S14), which could explain why no Ueff was observed in the absence of a dc field. Conclusions In summary, we successfully developed salt elimination/metathesis protocols to obtain three different heteroleptic yttrium complexes [Y{C(PPh2NSiMe3)2}(CpR)(THF)] (CpR=Cp(1), CptBu(2), Cp*(3)) by using [Y{C(PPh2NSiMe3)2}(I)(THF)2] as precursors. This synthetic route was also extended to obtain [Dy{C(PPh2NSiMe3)2}(Cp*)(THF)] (4), as the combination of {C(PPh2NSiMe3)2}2– and CpR in 1–4 did not provide sufficient steric bulk to exclude THF from the coordination sphere of the lanthanide ions. The Ccarbene–Y–Cpcent of 1–4 is 118.021, 129.459, 118.331, and 118.295°, respectively, showing low axiality. As a result, the magnetic behaviors of 4 are far from SMMs. We have attempted to remove the solvent by heating a solid sample in vacuo for hours. But analysis of the NMR spectra of the resultant solid revealed that THF could not be removed without destroying the coordination unit “Y{C(PPh2NSiMe3)2}(CpR)”. Carefully regulating the appropriate steric effect and charge effect of the two ligands {C(PPh2NSiMe3)2}2– and CpR would be useful to enlarge the Ccarbene–Y–Cpcent angle, making them near the idea Y{C(PPh2NSiMe3)2}(CpR) unit with a Ccarbene–Y–Cpcent angle of 180°. Next, we would conduct further studies in this direction to produce high-performance SMMs. Experimental Section General All operations were carried out under an atmosphere of argon by using Schlenk techniques or in an argon-filled glovebox. Toluene, THF, hexane, and C6D6 were dried over the Na/K alloy, transferred under vacuum, and stored in the glovebox. The raw materials [Y{C(PPh2NSiMe3)2}(I)(THF)2],36 DyI3(THF)3.5,46 KBn,47 NaCp,48 KCp*,49 CH2(PPh2NSiMe3)2,50 and C5H5tBu51 were synthesized following the literature. KCptBu was synthesized by slowly adding a solution of KN(SiMe3)2 (1.63 g, 8.18 mmol) in THF (ca. 10 mL) to a solution of C5H5tBu (1.00 g, 8.18 mmol) in THF (ca. 10 mL). After 5 h, KCptBu was isolated by removing the volatiles under vacuum and repeatedly washed with hexane. Brown KCptBu (0.81 g, 62%) was dried in vacuum. 1H NMR (500 MHz, C6D6, 25 °C): δ = 5.70 (m, 2H, CptBu–CH), 5.67 (m, 2H, CptBu–CH), 1.37 (s, 9H, CptBu–CMe3). These compounds were stored in the glovebox. 1H, 13C{1H}, and 31P{1H} NMR spectra were recorded on a Bruker Avance III HD 500 MHz spectrometer. Chemical shifts were reported in δ units with references to the residual solvent resonance of the deuterated solvents for proton and carbon chemical shifts and to external H3PO4 (85%) for phosphorus chemical shifts. Fourier-transform infrared spectroscopy (FTIR) spectra were recorded on a Bruker α II spectrometer placed in a glovebox. EA values were obtained from an Elementar UNICUBE analyzer. Synthesis of [Y{C(PPh2NSiMe3)2}(Cp)(THF)] (1) [Y{C(PPh2NSiMe3)2}(I)(THF)2] (274 mg, 0.30 mmol) and NaCp·0.25THF (32 mg, 0.30 mmol) were mixed in a vial with 5 mL of THF. After stirring at room temperature overnight, the precipitate was separated by centrifugation. The volatiles were removed under vacuum. The solid was recrystallized from toluene to afford 1 (0.5Toluene) as yellow crystals. Yield: 106 mg, 43%. 1H NMR (500 MHz, C6D6, 25 °C): δ = 8.30 (dd, 3JPH = 13.0 Hz, 3JHH = 7.0 Hz, 4H, o-PPh2), 7.26 (t, 3JHH = 7.2 Hz 4H, m-PPh2), 7.19 (m, 2H, p-PPh2), 7.16–7.13 (m, toluene-Ph overlapped with the residual solvent resonance of the deuterated solvent), 7.10 (m, 4H, o-PPh2), 7.02 (m, 1H, toluene-Ph), 6.84 (t, 3JHH = 7.2 Hz, 2H, p-PPh2), 6.75 (s, 5H, Cp), 6.71 (t, 3JHH = 7.5 Hz, 4H, m-PPh2), 3.86 (m, 4H, THF-OCH2), 2.11 (s, 1.5H, toluene-Me),1.37 (m, 4H, THF–CH2), 0.00 (s, 18H, SiMe3). 13C{1H} NMR (125 MHz, C6D6, 25 °C): δ = 142.5 (t, 1JCP = 56.8 Hz, i-PPh2), 137.5 (toluene-Ph), 135.7 (t, 1JCP = 41.9 Hz, i-PPh2), 131.4 (t, 2JCP = 5.6 Hz, o-PPh2), 130.8 (t, 2JCP = 5.0 Hz, o-PPh2), 129.1 (p-PPh2), 129.0 (toluene-Ph), 128.21 (toluene-Ph), 128.19 (p-PPh2), 126.8 (t, 3JCP = 5.6 Hz, m-PPh2), 125.3 (toluene-Ph), 111.03 (s, Cp), 71.4 (THF-OCH2), 51.4 (td, YCP, 1JCP = 131.2 Hz, JYC = 6.2 Hz), 25.1 (THF-CH2), 21.1 (toluene-Me), 3.8 (s, NSiMe3). 31P{1H} NMR(202 MHz, C6D6, 25 °C): δ = 3.2 (d, 2JP–Y = 12.1 Hz). IR v/cm–1: 2185(w), 1559(w), 1481(w), 1435(m), 1358(w), 1270(s), 1240(s), 1157(w), 1109(m), 955(w), 916(w), 825(s), 800(m), 772(m), 737(s), 686(s), 631(m), 603(m), 573(m). Elemental analysis calcd (%) for C40H51N2OP2Si2Y· 0.5Toluene: C, 63.03; H, 6.69; N, 3.38; found C, 63.51; H, 6.09; N, 3.48. Synthesis of [Y{C(PPh2NSiMe3)2}(CptBu)(THF)] (2) [Y{C(PPh2NSiMe3)2}(I)(THF)2] (274 mg, 0.30 mmol) and KCptBu (48 mg, 0.30 mmol) were mixed in a vial with 5 mL of THF. After stirring at room temperature for 2 h and another 2 h at 50 °C, the precipitate was separated by centrifugation. The volatiles were removed under vacuum. The solid was recrystallized from toluene to afford 2 as colorless crystals. Yield: 114 mg, 46%. 1H NMR(500 MHz, C6D6, 25 °C): δ = 8.20 (m, 4H, o-PPh2), 7.27 (m, 4H, m-PPh2), 7.20 (m, 2H, p-PPh2), 6.97 (m, 4H, o-PPh2), 6.81(m, 2H, p-PPh2), 6.73 (t, 3JHH = 2.5 Hz, 2H, CptBu–CH), 6.64 (m, 4H, m-PPh2), 6.22 (t, 3JHH = 2.7 Hz, 2H, CptBu–CH), 3.92 (m, 4H, THF-OCH2), 1.59 (s, 9H, CptBu-CMe3), 1.44 (m, 4H, THF–CH2), 0.17 (s, 18H, NSiMe3). 13C{1H} NMR(125 MHz, C6D6, 25 °C): δ = 142.3 (t, 1JCP = 55.6 Hz, i-PPh2), 141.5(CptBu-CCMe3), 137.6 (t, 1JCP = 41.2 Hz, i-PPh2), 132.4(PPh2), 130.6(PPh2), 129.1(PPh2), 128.2(PPh2), 127.0(PPh2), 126.6(PPh2), 109.6(CptBu-CH), 108.9(CptBu-CH), 72.07 (br, THF-OCH2), 54.6 (td, YCP, 1JCP = 137.5 Hz, 1JCY = 7.5 Hz), 33.0(CptBu-CMe3), 32.4(CptBu-CMe3), 25.0(THF-CH2), 4.50(NSiMe3). 31P{1H} NMR(202 MHz, C6D6, 25 °C): δ = 5.8 (d, 2JPY = 14.1 Hz). IR v/cm–1: 2185(w), 1581(w), 1481(w), 1436(m), 1403(w), 1262(s), 1235(s), 1177(w), 1124(w), 1104(m), 1061(w), 1028(w), 996(w), 860(m), 827(s), 797(s), 775(m), 734(s), 709(m), 691(s), 651(m), 631(m), 614(m), 573(m), 531(s). Elemental analysis calcd (%) for C44H59N2OP2Si2Y: C, 62.99; H, 7.09; N, 3.34; found C, 62.37; H, 7.39; N, 3.25. Synthesis of [Y{C(PPh2NSiMe3)2}(Cp*)(THF)] (3) [Y{C(PPh2NSiMe3)2}(I)(THF)2] (274 mg, 0.30 mmol) and KCp* (53 mg, 0.30 mmol) were mixed in a vial with 5 mL of THF. After stirring at room temperature for 2 h and another 8 h at 75 °C, the precipitate was separated by centrifugation. The volatiles were removed under vacuum. The solid was recrystallized from toluene to afford 3 (0.5Toluene) as colorless crystals. Yield: 76 mg, 28%. 1H NMR (500 MHz, C6D6, 25 °C): δ = 8.38 (dd, 3JHP = 13.5 Hz, 3JHH = 7.0 Hz, 4H, o-PPh2), 7.33 (t, 3JHH = 7.5 Hz, 4H, m-PPh2), 7.22 (t, 2H, 3JHH = 7.5 Hz, p-PPh2), 7.13 (m, 1H, toluene-Ph), 7.02 (m, 1.5H, toluene-Ph), 6.94 (dd, 3JHP = 12.2 Hz, 3JHH = 7 Hz, 4H, o-PPh2), 6.79 (t, 3JHH = 7.5 Hz, 2H, p-PPh2), 6.61 (t, 3JHH = 7.5 Hz, 4H, m-PPh2), 4.14 (m, 4H, THF-OCH2), 2.23 (s, 15H, Cp*-Me), 2.11 (s, 1.5H, toluene-Me), 1.48 (m, 4H, THF–CH2), 0.12 (s, 18H, SiMe3). 13C{1H} NMR (125 MHz, C6D6, 25 °C): δ = 142.8 (t, 1JCP = 58.1 Hz, i-PPh2), 137.5 (toluene-Ph), 137.2 (t, 1JCP = 39.4 Hz, i-PPh2), 132.81 (t, 2JCP = 5.6 Hz, o-PPh2), 131.0 (t, 2JCP = 5.6 Hz, o-PPh2), 129.1 (p-PPh2), 129.0 (toluene-Ph), 128.20 (toluene-Ph), 128.18 (p-PPh2), 126.9 (t, 3JCP = 5.6 Hz, m-PPh2), 126.6 (t, 3JCP = 5.6 Hz, m-PPh2), 125.31 (toluene-Ph), 117.82 (Cp*-CMe), 71.13 (br, THF-OCH2), 53.7 (td, YCP, 1JCP = 123.8 Hz, 1JCY = 8.8 Hz), 25.2 (THF-CH2), 21.0 (toluene-Me), 12.65 (Cp*-Me), 5.48 (NSiMe3). 31P{1H} NMR (202 MHz, C6D6, 25 °C): δ = 5.9 (d, 2JPY = 14.1 Hz). IR v/cm–1: 2185(w), 1586(w), 1481(w), 1433(m), 1267(m), 1237(m), 1172(w), 1109(m), 1056(m), 1023(m), 996(m), 915(w), 825(s), 802(m), 775(m), 737(s), 709(m), 691(s), 651(m), 631(m), 589(m), 546(m). Elemental analysis calcd (%) for C45H61N2OP2Si2Y· 0.5Toluene: C, 64.79; H, 7.29; N, 3.12; found C, 64.21; H, 7.06; N, 3.54. Synthesis of [Dy{C(PPh2NSiMe3)2}(Cp*)(THF)] (4) Step 1: 10 mL of THF was added into a mixture of DyI3(THF)3.5 (462 mg, 0.61 mmol) and KBn (158 mg, 1.22 mmol) at 0 °C. After stirring at 0 °C for 4 h, the precipitate was separated by centrifugation and the volatiles were removed under vacuum. The left brown oil was supposed to be “Dy(Bn)2(I)(THF)3” and it was used in the next step without further characterization. Step 2: 5 mL of the toluene solution of CH2(PPh2NSiMe3)2 (323 mg, 0.58 mmol) was added into the 5 mL toluene solution of Dy(Bn)2(I)(THF)3 at −78 °C. After stirring at room temperature for 18 h, the precipitate was separated by centrifugation. The toluene solution was concentrated to a saturated solution and a yellow sediment was deposited. The sediments were isolated and dried under vacuum with a mass of 298 mg, which was supposed to be [Dy{C(PPh2NSiMe3)2}(I)(THF)2] and it was used in the next step without further characterization. Step 3: [Dy{C(PPh2NSiMe3)2}(I)(THF)2] and KCp* (53 mg, 0.30 mmol) were mixed in a vial with 5 mL of THF. After stirring at room temperature for 2 h and another 8 h at 75 °C, the precipitate was separated by centrifugation. The volatiles were removed under vacuum. The solid was recrystallized from toluene to afford 4 as yellow crystals. Yield: 153 mg, 27%. IR v/cm–1: 2182(w), 1583(w), 1481(w), 1433(m), 1262(m), 1239(m), 1207(m), 1171(w), 1101(m), 1059(m), 1018(m), 918(w), 822(s), 799(m), 767(m), 734(s), 709(m), 691(s), 651(m), 631(m), 601(m), 548(m). Elemental analysis calcd (%) for C45H61DyN2OP2Si2: C, 58.33; H, 6.64; N, 3.02; found C, 59.21; H, 6.53; N2.97.