Bis(amido) rare-earth complexes coordinated by tridentate amidinate ligand: synthesis, structure and catalytic activity in the polymerization of isoprene and rac-lactide

Abstract

A series of bis(amido) complexes of rare earth metals {2-[Ph₂P(O)]C₆H₄NC(tBu)N(2,6-Me₂C₆H₃)}Ln(N(SiMe₃)₂)₂, (Ln = Y (2), Nd (3), La (4)) was synthesized using the amine elimination reaction of Ln[N(SiMe₃)₂]₃ (Ln = Y, Nd, La) and parent amidine 1 in THF, and the products were isolated in 60 (2), 61 (3) and 72% (4) yields, respectively. The X-ray studies revealed that complexes 2–4 are solvent-free and feature intramolecular coordination of the Ph₂P=O group to the Ln ion. Complexes 2–4 were investigated as precatalysts for isoprene polymerization. The ternary systems 2–4/borate/AlR₃ (AlR₃ = AliBu₃, AliBu₂H; borate = [Ph₃C][B(C₆F₅)₄], [PhNHMe₂][B(C₆F₅)₄]) were found to be active in isoprene polymerization and enable complete conversion of 1000 equivalents of monomer into polymer at 25 °C within 1–24 h affording polyisoprenes with polydispersities M_w/M_n = 1.12–9.46. The effect of the organoaluminium component and [Ln]/[AlR₃] ratio on the catalytic activity and selectivity of the ternary catalytic systems was investigated. Complexes 2–4 proved to be efficient catalysts for the ring-opening polymerization of rac-lactide, which allow conversion of up to 500 equivalents of monomer into a polymer at room temperature within 30 min and afford atactic polylactides with high molecular weights and moderate molecular-weight distributions (1.29–2.12). Complexes 2–4 appear to be well-suited for achieving immortal polymerization of lactide, through the introduction of large amounts of isopropanol as a chain-transfer agent.

---

Introduction

Amido derivatives of the rare-earth metals have attracted a considerable deal of attention due to their high reactivity.¹ They have been exhaustively explored as precursors for the preparation of electronic and ceramic materials,² as well as catalysts for the synthesis of various valuable chemical products.³ Moreover they turned out to be efficient initiators for the polymerization of a large range of monomers such as ethylene,⁴ styrene,⁵ methyl methacrylate⁶ or cyclic esters.⁷ Formerly, in the pioneering work of Marks it was reported that the strength of the Ln–N bond is comparable to that of the Ln–C bond. By means of calorimetric titration of compounds Cp*₂SmX (X = NMe₂, CH(SiMe₃)₂) it was evidenced that the absolute bond disruption enthalpies (D) of Sm–N and Sm–C bonds have rather close values: X = NMe₂, D = 48.2 kcal mol⁻¹; X = CH(SiMe₃)₂, D = 47.0 kcal mol⁻¹.⁸ Surprisingly, unlike alkyl species amido derivatives were scarcely investigated as precursors for the cis-stereoselective polymerization of butadiene or isoprene.⁹

Recently we demonstrated that bis(alkyl) complexes {2-[Ph₂P(O)]C₆H₄NC(tBu)N(2,6-Me₂C₆H₃)}Ln(CH₂SiMe₃)₂(THF)_n (Ln = Y, n = 1, Ln = Er, n = 1, Ln = Lu, n = 0) containing tridentate amidinate ligand featuring intramolecular coordination of a side chain pendant Ph₂P=O group to metal ion are efficient components of ternary catalytic systems for isoprene polymerization.¹⁰ The systems Ln/borate/AliBu₃ (borate = [PhNHMe₂][B(C₆F₅)₄], [Ph₃C][B(C₆F₅)₄]) enable complete conversion of 1000–10 000 equivalents of isoprene into polymer at 20 °C within 0.5–2.5 h affording polyisoprenes with very high content of 1,4-cis units (up to 96.6%) and rather narrow polydispersities. However, a common disadvantage of bis(alkyl) species is their thermal sensibility and difficulties to handle and to store. In order to find new efficient catalytic systems based on less sensitive and more available compounds we focused on the synthesis of rare-earth bis(amido) complexes coordinated by tridentate amidinate ligand {2-[Ph₂P(O)]C₆H₄NC(tBu)N(2,6-Me₂C₆H₃)}⁻. A comparative study of catalytic systems based on LLnR₂ complexes (R = CH₂SiMe₃, N(SiMe₃)₂) can provide a deeper insight into the effect of leaving group in rare-earth complex on catalytic activity and selectivity of isoprene polymerization. Moreover application of bulky silylamido N(SiMe₃)₂ ligand makes possible to use derivatives of rare-earth metals having larger ion size (La, Nd) whose alkyl species are hardly accessible. At the same time derivatives of these metals are of special interest since normally provide higher 1,4-cis selectivity in diene polymerization. Herein we report on the synthesis, structure and catalytic activity of bis(amido) complexes {2-[Ph₂P(O)]C₆H₄NC(tBu)N(2,6-Me₂C₆H₃)}Ln(N(SiMe₃)₂)₂ (Ln = Y, Nd, La) and their catalytic activity in isoprene polymerization as well as in ring-opening polymerization of rac-lactide.

Results and discussion

To synthesize a series of bis(amido) rare-earth complexes coordinated by tridentate amidinate ligand bearing a pendant Lewis base Ph₂P=O group linked to a NCN fragment via a conformationally rigid o-phenylene linker the amine elimination approach was used. The equimolar amounts of Ln[N(SiMe₃)₂]₃ (Ln = Y, Nd, La)¹¹ and amidine 2-[Ph₂P(O)]PhNHC(tBu)=N(2,6-Me₂C₆H₃) (1)¹⁰ were reacted in THF solution at room temperature (Scheme 1). After removal of the volatiles in vacuum and recrystallization of the resulted solid from THF/hexane mixture complexes {2-[Ph₂P(O)]C₆H₄NC(tBu)N(2,6-Me₂C₆H₃)}Ln[N(SiMe₃)₂]₂ (Ln = Y (2), Nd (3), La (4)) were isolated as pale yellow (2 and 4) or green (3) crystals in 60, 61 and 72% yields respectively. Complexes 2–4 are air and moisture sensitive, well soluble in THF and toluene and sparingly soluble in hexane.

Scheme 1

The ¹H NMR (400 MHz, C₆D₆, 25 °C) spectra of diamagnetic complexes 2 and 4 exhibit singlets at 0.39 ppm (2) and 0.37 ppm (4) due to the protons of N(SiMe₃)₂ group. The amidinate ligand appears as the expected set of resonances. The ³¹P{¹H} (161.9 MHz, C₆D₆, 25 °C) spectrum of 2 displays a doublet at 41.2 ppm (²J_PY = 6.4 Hz) due to coupling with ⁸⁹Y nuclei while in that of complex 4 a singlet at 39.7 ppm was observed.

The molecular structures of complexes 2–4 were established by X-ray diffraction studies. Transparent crystals of complexes 2–4 suitable for X-ray diffraction studies were obtained by slow condensation of hexane into concentrated THF solution of complexes. The molecular structure of complexes 2–4 is shown in Fig. 1. Crystallographic data and structure refinement details are given in Table S1 (see ESI†). Complexes 2–4 are isomorphous and crystallize in the monoclinic (C2/c) space group with eight molecules per unit cell. The X-ray studies revealed that complexes 2–4 feature intramolecular coordination of Ph₂P=O group to metal ion. In 2–4 the lanthanide ions are coordinated by two nitrogen and one oxygen atoms of the tridentate amidinate ligand and by two amido N(SiMe₃)₂ groups, thus resulting in coordination number of five. Despite high oxophilicity of rare earth metals coordination sphere of complexes 2–4 does not contains coordinated THF molecules. The C–N bonds within the amidinate fragment are slightly different (2: 1.332(2) and 1.353(2) Å; 3: 1.331(2) and 1.353(2) Å; 4: 1.330(3) and 1.354(4) Å). The M–N_(amidine) bonds are non-equivalent: 2.461(2) and 2.416(2) Å (2); 2.565(2) and 2.508(2) Å (3); 2.614(3) and 2.563(2) Å (4). The Ln–N_(amidine) bonds in 2–4 are longer compared to the related five-coordinate amidinate and guanidinate complexes (for comparison see: [4-Me-PhC(N-2,6-iPr₂C₆H₃)₂]Y(NHSiMe₂)₂(THF) 2.412(3), 2.351(2) Å;¹² [(SiMe₃)₂NC(NCy)₂]Y[N(SiHMe₂)₂]₂(THF) 2.356(4), 2.328(3) Å;¹³ [PhC(N-2,6-Me₂C₆H₃)₂]₂NdN(SiMe₃)₂ 2.462(3)–2.484(3) Å, [CyC(N-2,6-Me₂C₆H₃)₂]₂NdN(SiMe₃)₂ 2.448(4)–2.472(4) Å;¹⁴ [2-NC₅H₄(CH₂)₂NC(p-MeC₆H₄)NPh]₂La[N(SiMe₃)₂] (2.491(4)–2.609(4) Å)¹⁵. Obviously this difference in bond lengths originates from the coordination of sterically demanding Ph₂P=O group to the metal ions. The M–N_(N(SiMe3)2) bond lengths are also somewhat different: 2.251(2) and 2.301(2) Å (2); 2.335(2) and 2.385(2) Å (3); 2.389(3) and 2.447(2) Å (4). The Ln–N(N(SiMe₃)₂) bond lengths in complexes 2–4 are in a good agreement with the values previously published for related five-coordinate amido species: [4-Me-PhC(N-2,6-iPr₂C₆H₃)₂]Y(NHSiMe₂)₂(THF), 2.250(3), 2.258(3) Å;¹² [(SiMe₃)₂NC(NCy)₂]Y[N(SiHMe₂)₂]₂(THF) 2.241(4), 2.278(4) Å;¹³ CyNC{[N,N-(2,5-Me₂C₄H₂N)CH₂CH₂]₂N}NCyLn[N(SiMe₃)₂]₂ (Ln = La, Nd)¹⁶ Nd–N 2.327(2), 2.341(2) Å, La–N 2.386(2), 2.400(2); La[2-NC₅H₄-(CH₂)₂NC(p-MeC₆H₄)NPh]₂[N(SiMe₃)₂]) 2.453(4) Å.¹⁵ The MNCN fragments in complexes 2–4 are not planar. It is noteworthy that the value of the dihedral angle between the MNM and NCN planes (171.8° (2), 172.7° (3) and 172.9° (4)) increases with increase of ionic radius of the central metal atom (Y: 0.900 Å, Nd: 0.983 Å, La: 1.032 Å).¹⁷ The length of coordination bonds between metallocenter and oxygen atom of Ph₂P=O group (2.253(1) Å (2), 2.367(1) Å (3), 2.418(2) Å (4)) are in a good agreement with the values previously published for related five-coordinate amido species: [[(Me₃Si)₂N]₂(Ph₃PO)Y}₂(μ-η²:η²-N₂) (Y–O 2.2545(11)–2.2437(11)),¹⁸, [La[(Me₃Si)₂N]₂(PPh₂)(Ph₂PO)₂] (La–O 2.472(8) Å)¹⁹].

Fig. 1 ORTEP diagram (30% probability thermal ellipsoids) of 2–4 (M = Y (2), Nd (3), La (4)) showing the atom numbering scheme. Hydrogen atoms and carbon atoms in Ph-fragments of Ph₂P=O group are omitted for clarity. Selected bond lengths [Å] and angles [°] in 2: Y(1)–O(1) 2.253(1), Y(1)–N(1) 2.461(2), Y(1)–N(2) 2.416(2), Y(1)–N(3) 2.251(2), Y(1)–N(4) 2.301(2), N(1)–C(1) 1.332(2), N(2)–C(1) 1.353(2), N(1)–C(1)–N(2) 109.5(2), N(2)–Y(1)–N(1) 53.43(5), N(3)–Y(1)–N(4) 112.89(6). Selected bond lengths [Å] and angles [°] in 3: Nd(1)–O(1) 2.367(1), Nd(1)–N(1) 2.565(2), Nd(1)–N(2) 2.508(2), Nd(1)–N(3) 2.335(2), Nd(1)–N(4) 2.385(2), N(1)–C(1) 1.331(2), N(2)–C(1) 1.353(2), N(1)–C(1)–N(2) 110.4(2), N(2)–Nd(1)–N(1) 51.48(5), N(3)–Nd(1)–N(4) 113.65(6). Selected bond lengths [Å] and angles [°] in 4: La(1)–O(1) 2.418(2), La(1)–N(1) 2.614(3), La(1)–N(2) 2.563(2), La(1)–N(3) 2.389(3), La(1)–N(4) 2.447(2), N(1)–C(1) 1.330(3), N(2)–C(1) 1.354(4), N(1)–C(1)–N(2) 110.6(3), N(2)–La(1)–N(1) 50.45(7), N(3)–La(1)–N(4) 113.86(9).

Isoprene polymerization initiated by complexes 2–4

Bis(alkyl) rare earth species established a reputation as suitable precursors for catalytic systems performing controlled diene polymerization with high activities and selectivities.^9d,20 In contrast, the application of rare earth metal amide complexes as precatalysts for isoprene polymerization remains scarce.⁹

The catalytic activity of complexes 2–4 in isoprene polymerization was explored in toluene at room temperature. The polymerization results are summarized in Table 1. It was established that neither complexes 2–4 nor binary systems 2–4/AliBu₃, 2–4/AliBu₂H do not perform catalytic activity. Unlike the related bisamides coordinated by bidentate amidinate ligand [PhC(N-2,6-R₂C₆H₃)₂]Ln[N(SiHMe₂)₂]₂ (Ln = Sc, Y; R = Me, iPr)^9c,9e complexes 2–4 activated by equimolar amounts of [Ph₃C][B(C₆F₅)₄] or [PhNHMe₂][B(C₆F₅)₄] turned out to be inactive in isoprene polymerization. However this is in line with the observation of Luo and co-workers^9d who reported that yttrium bis(amido) amidinates containing N(SiMe₃)₂ groups instead of N(SiHMe₂)₂ do not perform catalytic activity in isoprene polymerization in the presence of borates. Upon combination with borate ([Ph₃C][B(C₆F₅)₄], [PhNHMe₂][B(C₆F₅)₄]) and alkylaluminium compounds (AliBu₃, AliBu₂H) complexes 2–4 enable isoprene polymerization. However catalytic activity of the ternary systems was found to be dramatically influenced by the nature of the metal atom, borate and alkyl aluminium components. Thus complexes 2 and 3 enable isoprene polymerization at ambient temperature (molar ratio [IP]/[Ln] = 1000) with 45–100% conversion of monomer into polymer (Table 1) in 1 h. Surprisingly complex 4 containing the largest La³⁺ ion¹⁷ displayed the lowest activity. The systems based on 4 required 24 h to achieve 92–100% conversions. Complex 2 turned out to be indifferent to the nature of borate or alkyl aluminium components: for both AlR₃ and both borates the polyisoprene (PIP) yields were 92–100%, though complex 3 is somewhat more prone to the influence of the nature of AlR₃ and borate components. The complex 3 based systems containing AliBu₂H performed higher activities than those with AliBu₃ for both borates. Neodymium complex 3 in the presence of AliBu₃ was noticeably less active than 2 with both borates (64 vs. 92 for [Ph₃C][B(C₆F₅)₄] and 45 vs. 97 for [PhNHMe₂][B(C₆F₅)₄]). However when AliBu₂H was used catalytic activities of complexes 2 and 3 became similar.

Table 1 Catalytic tests in isoprene polymerization initiated by systems 2–4/(AliBu₃, AliBu₂H)/([Ph₃C][B(C₆F₅)₄], [PhNHMe₂][B(C₆F₅)₄])^a

	Comp.	Borate	AlR3	[IP]/[Ln]	t, h	Yield, %	Cis-1,4	Trans-1,4	3,4-	Mn^b (×10^−3)	Mncalc^c (×10^−3)	Mn/Mw
a Conditions: complex (10 μmol in toluene, [AlR3] : [Ln] : [borate] = 10/1/1, 1/1/1, T: 25 °C); HNB = [PhNHMe2][B(C6F5)4], TB = [Ph3C][B(C6F5)4].b Determined by GPC against polystyrene standard.c Mcalc = ([IP]/[Ln]) × 68.12 × (conversion), Y* = complex 5.^10
1	Y	TB	10AliBu3	1000	1	92	76.2	18.2	5.6	30.66	62.67	4.18
2	Y	HNB	10AliBu3	1000	1	97	78.5	17.2	4.3	42.58	66.07	2.25
3	Y	TB	10AliBu2H	1000	1	100	72.3	21.4	6.3	19.01	68.12	1.58
4	Y	HNB	10AliBu2H	1000	1	96	71.5	21.9	6.6	10.04	65.39	1.53
5	Y	TB	1AliBu2H	1000	1	54	76.7	15.0	8.3	62.55	36.78	1.70
6	Y	HNB	1AliBu2H	1000	1	78.8	79.4	15.8	4.8	55.50	53.67	1.47
7	Nd	TB	10AliBu3	1000	1	64	82.6	9.6	7.8	19.71	43.59	9.06
8	Nd	HNB	10AliBu3	1000	1	45	88.7	4.9	6.4	13.73	30.65	9.46
9	Nd	TB	10AliBu2H	1000	1	100	55.9	37.4	6.7	11.20	68.12	1.23
10	Nd	HNB	10AliBu2H	1000	1	95	55.9	34.7	9.4	14.65	64.71	1.21
11	Nd	TB	1AliBu2H	1000	1	84.3	68.6	23.0	8.4	59.35	57.42	1.22
12	Nd	HNB	1AliBu2H	1000	1	100	89.6	3.4	7.0	115.36	68.12	1.627
13	La	TB	10AliBu3	1000	24	94	45.2	46.8	8.0	13.48	64.03	2.76
14	La	HNB	10AliBu3	1000	24	100	61.3	32.0	6.7	57.68	68.12	1.69
15	La	TB	10AliBu2H	1000	24	96	38.6	52.0	9.4	8.45	65.39	2.21
16	La	HNB	10AliBu2H	1000	24	100	39.8	52.7	7.5	6.74	68.12	2.96
17	La	TB	1AliBu2H	1000	24	100	70.6	22.4	7.0	134.19	68.12	1.202
18	La	HNB	1AliBu2H	1000	24	92	78.9	12.0	9.1	123.91	62.67	1.122
19	Y*	TB	10AliBu3	1000	1	100	91.6	2.7	5.7	34.3	67.7	2.49
20	Y*	HNB	10AliBu3	1000	0.5	100	87.2	8.6	4.2	75.3	68.1	1.49

---

The polyisoprenes obtained by using 2–3 combined with AliBu₃ are characterized by rather broad molecular weight distribution (2.25–9.46), whereas application of AliBu₂H allowed to decrease polydispersities considerably and to reach values 1.21–1.58. This observation can originate from higher efficiency of AliBu₂H as a chain transfer agent²¹ and different geometry of the real catalytic species. Interestingly the La based systems are less influenced by the nature of alkyl aluminium component, with both AliBu₃ and AliBu₂H the PDI values remain in the region 1.69–2.96. At the same time the catalytic systems containing AliBu₂H perform polymerizations with lower cis-1,4 selectivity. Thus the PIPs obtained with AliBu₃ contain 45.2–88.7% cis-1,4 units, while the catalytic tests performed with AliBu₂H afforded the polymers with contents of cis-1,4 units 38.6–72.3%. Interestingly the drop of cis-1,4 selectivity was less pronounced for Y-containing systems, whereas in the case of Nd the loss of cis-1,4 selectivity was ∼30%. The effect of [Ln]/[AliBu₂H] molar ratio on molecular weight and microstructure of the obtained PIPs was also explored. A series of polymerizations initiated by the systems 2–4/AliBu₂H/borate ([Ln]/[AliBu₂H]/[borate] = 1 : 1 : 1; borate = [Ph₃C][B(C₆F₅)₄], [PhNHMe₂][B(C₆F₅)₄]) containing one equivalent of AliBu₂H was carried out (Table 1, entries 5, 6, 11, 12, 17, 18). It should be noted that decrease of [Ln]/[AliBu₂H] molar ratio only in the case of Y results in the noticeable drop of catalytic activity while for the systems based on Nd and La compounds their efficiencies remain comparable. Also it was found that the ternary systems, containing one equivalent of AliBu₂H display higher cis-1,4 selectivity compared to those with 10 equivalents of organoaluminium component. The influence of the [Ln]/[AliBu₂H] ratio was particularly pronounced for the systems based on the lanthanum complex 4 (compare; 38.6 vs. 70.6% and 39.7 vs. 78.9%).

Comparison of the results of isoprene polymerization tests initiated by the systems based on bis(amido) complex 2 and related bis(alkyl) complex coordinated by the same amidinate ligand {2-[Ph₂P(O)]C₆H₄NC(tBu)N(2,6-Me₂C₆H₃)}Y(CH₂SiMe₃)₂ (THF) (5)¹⁰ revealed dramatic influence of the “leaving group” in Ln compound on both activity and selectivity (Table 1, entries 19 and 20). Thus complex 5 in combination with [PhNHMe₂][B(C₆F₅)₄] provides total conversion of monomer into polymer in 0.5 h affording a polymer with contents of 87.2% of cis-1,4 units and rather narrow polydispersity (1.49). Application of [Ph₃C][B(C₆F₅)₄] somewhat slows down the reaction, the total conversion requires already 1 h, however higher selectivity was demonstrated. So one can conclude that catalytic systems containing bis(alkyl) yttrium complexes provide better efficiency and cis-1,4 selectivity in isoprene polymerization and allow for formation of PIPs with rather narrow polydispersities.

The nature of real active species that forms in the ternary catalyst systems for isoprene polymerization have attracted a great deal of attention and to date the reactions of their components are explored in details.^9a,c,e,24a Thus Luo and co-workers have shown that the reactions of related mono(amidinate) rare-earth-metal bis(silylamide) complexes [PhC(N-2,6-R₂C₆H₃)₂]Ln[N(SiHMe₂)₂]₂(THF)_y (R = Me, iPr; Ln = Sc, Y; y = 0, 1) with 1 equiv. of [Ph₃C][B(C₆F₅)₄] in toluene afford cationic amidinate rare-earth-metal amide complexes [{PhC(N-2,6-R₂C₆H₃)₂}LnNSiHMe₂}{SiMe₂N(SiHMe₂)₂}(THF)_n][B(C₆F₅)₄]. The treatment of [PhC(N-2,6-R₂C₆H₃)₂]Ln[N(SiHMe₂)₂]₂(THF)_y with excess AlMe₃ resulted the heterometallic Ln/Al methyl complexes [PhC(N-2,6-R₂C₆H₃)₂]Ln[(μ-Me)₂AlMe₂]₂.^9c,d The cationic complex [Nd{N(SiMe₃)₂}₂(THF)₂][B(C₆F₅)₄] was prepared by reaction of Nd[N(SiMe₃)₂]₃ and [HNMe₂Ph][B(C₆F₅)₄].^9a Recently we reported that the reaction of LLn[N(SiHMe₂)₂]₃(thf) (L = amidine-amidopyrpdinate ligand) with [Et₃NH]⁺[BPh₄]⁻ affords cationic monoamido complex.^24a The reaction of the cyclopentadienyl rare-earth metal bis(silylamide) complexes Cp*Ln[N(SiHMe₂)₂]₂ with AlMe₃ produced the heterometallic Ln/Al methyl complexes Cp*Ln[(μ-Me)₂AlMe₂]₂, which were formed via the amide–alkyl exchange.⁹ⁱ Taking into account that neither binary systems 2–4/AliBu₃, 2–4/AliBu₂H nor 2–4 activated by equimolar amounts of [Ph₃C][B(C₆F₅)₄] or [PhNHMe₂][B(C₆F₅)₄] do not perform catalytic activity one can conclude that the most likely active species in our ternary catalyst system should be the Ln/Al heterobimetallic cationic alkyl complexes.

We have demonstrated a dramatic effect of “leaving group” in Ln compound on activity, selectivity as well as on molecular weight distribution of the obtained PIPs. Alkyl yttrium complex {2-[Ph₂P(O)]C₆H₄NC(tBu)N(2,6-Me₂C₆H₃)}Y(CH₂SiMe₃)₂(THF)¹⁰ provides better reaction rate, cis-1,4 selectivity and narrower polydispersities compared to its amide congener.

Rac-Lactide polymerization initiated by complexes 2–4

There is considerable interest in the controlled ring-opening polymerization (ROP) of lactides (LAs) by well-defined metal initiators because of the biodegradable and biocompatible nature of polylactides (PLAs) and their potential commercial applications.²² We investigated the catalytic activity of new compounds 2–4 containing potentially initiating nucleophilic amido group in the ROP of racemic lactide (rac-LA) (Scheme 2). Representative results are summarized in Table 2. Complexes 2–4 proved to be efficient catalysts of ROP of rac-LA under mild conditions, allowing for the total conversion of 100–500 equiv. of monomer (toluene, 25 °C, [LA] = 1.0 mol L⁻¹) into polymer in 30 min. PLAs obtained in all catalytic tests have atactic structures as evidenced by NMR analyses (Pr = 0.50–0.55). This indicates a poorly effective chain-end control and suggests that polymerization of lactide mediated by 2–4 takes place in a non sterically congested environment.

Scheme 2 Catalytic ring opening polymerization of rac-lactide.

Table 2 Polymerization of rac-lactide with complexes 2–4^a

No.	Cat.	[LA]/[Ln]/ROH	Time (min)	Conv., %	Mn × 10^−3	Mncalc × 10^−3	Mw/Mn
a General conditions: toluene, [LA] = 1.0 mol L^−1, T = 25 °C. Reaction time was not necessarily optimized. Conversion of monomer M, as determined by ^1H NMR spectroscopy. Experimental (corrected; see Experimental section) Mn and Mw/Mn values determined by GPC in THF vs. polystyrene standards. Mn value calculated by assuming two polymer chain per metal center with the relationship: (144 × conversion × [M]/2[Ln]).
1	Y	100 : 1 : 0	30	100	37.7	7.2	1.47
2	Nd	100 : 1 : 0	30	100	28.2	7.2	1.48
3	La	100 : 1 : 0	30	100	32.0	7.2	1.34
4	Y	500 : 1 : 0	30	94	126.5	33.8	1.63
5	Nd	500 : 1 : 0	30	100	127.7 (82%), 1.9 (8%)	36.0	1.67, 1.48
6	La	500 : 1 : 0	30	100	231.8 (74%), 1.7 (26%)	36.0	1.90, 2.33
7	Y	100 : 1 : 1	30	92	22.5	6.6	1.61
8	Nd	100 : 1 : 1	30	100	22.4	7.2	1.44
9	La	100 : 1 : 1	30	100	24.4	7.2	1.40
10	Y	500 : 1 : 1	30	85	97.2 (92%), 1.5 (8%)	30.6	1.66, 1.44
11	Nd	500 : 1 : 1	30	100	77.5 (84%), 1.2 (16%)	36.0	1.53, 1.73
12	La	500 : 1 : 1	30	100	101.2 (81%), 1.2 (19%)	36.0	1.57, 1.83
13	Y	100 : 1 : 2	30	88	25.2	6.3	1.73
14	Nd	100 : 1 : 2	30	100	12.2	7.2	1.53
15	La	100 : 1 : 2	30	94	10.8	6.7	1.43
16	Y	100 : 1 : 3	30	94	7.4	6.8	1.32
17	Nd	100 : 1 : 3	30	100	9.7	7.2	1.85
18	La	100 : 1 : 3	30	100	9.9	7.2	1.91
19	Y	100 : 1 : 4	30	90	4.3	6.5	1.29
20	Nd	100 : 1 : 4	30	87	4.5	6.3	1.49
21	La	100 : 1 : 4	30	86	5.0	6.2	1.74
22	Y	100 : 1 : 10	30	89	1.56	6.4	1.51
23	Nd	100 : 1 : 10	30	89	2.4	6.4	1.42
24	La	100 : 1 : 10	30	88	2.4	6.3	1.53
25	Y	100 : 1 : 20	30	91	0.7	6.6	1.64
26	Nd	100 : 1 : 20	30	81	1.3	5.8	1.35
27	La	100 : 1 : 20	30	88	1.9	6.3	1.61

---

Catalytic tests did not reveal noticeable effect of the nature of the metal center on activities of the complexes: at the ratio [LA]/[Ln] = 100 all three initiators in 0.5 h allowed to reach total conversion. However when the [LA]/[Ln] ratio was increased to 500 the Y compound was slightly less active (0.5 h, 94% vs. 100%; entries 4–6). All the PLAs obtained with complexes 2–4 at the ratio [LA]/[Ln] = 100 showed monomodal GPC traces with rather narrow polydispersity values M_w/M_n = 1.34–1.48. At ratio [LA]/[Ln] = 500 the polymers synthesized with Nd and La complexes displayed bimodal molecular weights distribution, while for the polymer obtained with Y compound it remained monomodal (entries 4–6). Increase of [LA]/[Ln] ratio lead to slight broadening of PDI. It is noteworthy that corrected experimental number-average molecular weights of obtained PLAs as determined by GPC noticeably exceed the values calculated from the initial monomer-to-lanthanide ratio, conversion and assuming that both amido groups are active initiators (Table 2). It should be noted that this general trend has been often observed for the ROP of lactide initiated by rare-earth amido complexes due to slow initiation.^22b,23

It was associated with lower nucleophilicity of the amido group N(SiMe₃)₂, acting on the initiation stage as compared to the nucleophilic alkoxide group –OR (R – the growing polymer chain) resulted from opening of the lactide and the insertion of the resulting fragments into M–(N(SiMe₃)₂) bond [For examples of ROP of LAs initiated by rare-earth amido complexes see ref. 12, 14 and 24]. In some cases the GPC curves display bimodal character. That most likely originates from slow initiation and trans esterification reaction.

Possibility to provide control over polymerization reaction as well as to achieve immortal polymerization with these systems, that is, to generate several PLA chains per metal center by introducing several equivalents of a chain transfer agent, was explored with complexes 2–4 in the presence of iso-propanol which acts as a chain transfer agent (Table 2, entries 10–27). Addition of 1 equivalent of iso-propanol to the reaction mixture at both [LA]/[Ln] ratios results in some decrease of M_n but does not influence strongly PDI. Bimodal character of the GPC curves at [LA]/[Ln] = 500 ratio was retained. When 2 equivalents of iso-propanol were added ([LA]/[Ln] = 100) the experimental M_n still noticeably surpass the calculated ones. In the presence of 3–20 equivalents of iso-propanol polymerization rates and conversions remain high. The M_n values expectedly decrease with increase of concentration of iso-propanol however no match of experimental and calculated values was observed. The polydispersities fall into the region 1.30–1.90.

Interestingly two related yttrium bis(amide) complexes {2-[Ph₂P(O)]C₆H₄NC(tBu)N(2,6-Me₂C₆H₃)}Y(N(SiMe₃)₂)₂ and [p-MeC₆H₄NPh₂]Y(N(SiMe₃)₂)₂(THF)¹² coordinated by amidinate ligands despite of different steric bulk and denticity of the latter perform very similar catalytic activity in ROP of lactide. The fact that in both cases atactic PLAs were obtained indicates that introduction of 2-[Ph₂P(O)] group into the amidinate scaffold is insufficient for creation of sterically congested environment and providing control of polymer microstructure by the means of chain-end control.

Conclusions

The amine elimination reactions of equimolar amounts of Ln[N(SiMe₃)₂]₃ (Ln = Y, Nd, La) and amidine 2-[Ph₂P(O)]PhNHC(tBu)=N(2,6-Me₂C₆H₃) allowed for the synthesis of a series of the bis(amido) amidinate complexes {2-[Ph₂P(O)]C₆H₄NC(tBu)N(2,6-Me₂C₆H₃)}Ln[N(SiMe₃)₂]₂ in good yields. These complexes proved to be solvent free and featuring intramolecular coordination of Ph₂P(O) group to Ln³⁺ ions in the solid state.

We demonstrated that amido complexes 2–4 activated by borate ([Ph₃C][B(C₆F₅)₄], [PhNHMe₂][B(C₆F₅)₄]) and alkylaluminium compounds (AliBu₃, AliBu₂H) similarly to the related bis(alkyl) species {2-[Ph₂P(O)]C₆H₄NC(tBu)N(2,6-Me₂C₆H₃)}Ln(CH₂SiMe₃)₂(THF)_n (Ln = Y, n = 1, Ln = Er, n = 1, Ln = Lu, n = 0)¹⁰ enable isoprene polymerization. Catalytic activity of the ternary systems was found to be dramatically influenced by the nature of the metal atom, borate and alkyl aluminium components. Interestingly lanthanum complex 4 containing the largest La³⁺ ion displayed the lowest activity. Moreover comparison of the results of polymerization tests obtained with the systems based on complex 2 and related bis(alkyl) complex coordinated by the same amidinate ligand {2-[Ph₂P(O)]C₆H₄NC(tBu)N(2,6-Me₂C₆H₃)}Y(CH₂SiMe₃)₂(THF)¹⁰ revealed a dramatic effect of “leaving group” in Ln compound on activity, selectivity as well as on molecular weight distribution of the obtained PIPs. Alkyl yttrium complex provides better reaction rate, cis-1,4 selectivity and narrower polydispersities.

Complexes 2–4 initiate ROP of rac-LA under mild conditions providing high reaction rates and formation of PLAs having atactic structures and rather narrow polydispersities. These initiators appear to be well suited for achieving immortal polymerization of lactide through the introduction of large amounts of isopropanol as a chain transfer agent, thus enabling the conversion of large amounts of monomer and the production of numerous macromolecular chains per metal initiator.

Experimental section

General conditions

All experiments were performed in evacuated tubes by using standard Schlenk techniques, with rigorous exclusion of traces of moisture and air. After being dried over KOH, THF was purified by distillation from sodium/benzophenone ketyl; hexane and toluene were dried by distillation from sodium and benzophenone ketyl prior to use. C₆D₆ was dried with sodium and condensed in vacuum into NMR tubes prior to use. CDCl₃ was used without additional purification. Ln[N(SiMe₃)₂]₃, (Ln = Y, Nd, La)¹¹ and [2-Ph₂P(O)]C₆H₄NHC(tBu)(2,6-Me₂C₆H₃)¹⁰ were prepared according to literature procedures. Rac-Lactide was recrystallized two times from dry THF. All other commercially available chemicals were used after the appropriate purifications. NMR spectra were recorded with Bruker Avance DRX-400 and Bruker DRX-200 spectrometers in CDCl₃, C₆D₆ at 20 °C, unless otherwise stated. Chemical shifts for ¹H and ¹³C NMR spectra were referenced internally to the residual solvent resonances and are reported relative to TMS. IR spectra were recorded as Nujol mulls with a “Bruker-Vertex 70” instrument. Lanthanide metal analyses were carried out by complexometric titration.²⁵ The C, H, N elemental analyses were performed in the microanalytical laboratory of the G. A. Razuvaev Institute of Organometallic Chemistry. GPC was carried out using chromatograph “Knauer Smartline” with Phenogel Phenomenex Columns 5u (300 × 7.8 mm) 10⁴, 10⁵ and Security Guard Phenogel Column with RI and UV detectors (254 nm). Mobile phase was THF, flow rate was 2 mL min⁻¹. Columns was calibrated by Phenomenex Medium and High Molecular Weight Polystyrene Standard Kits with peak M_w from 2700 to 2 570 000 Da. The number average molecular masses (M_n) and polydispersity index (M_w/M_n) of the polymers were calculated with reference to an universal calibration versus polystyrene standards. M_n values of PLAs were corrected with a Mark–Houwink factor of 0.58 to account for the difference in hydrodynamic volumes between polystyrene and polylactide. Microstructures of PLAs were measured by homodecoupling ¹H NMR spectroscopy at 25 °C in CDCl₃ on a Bruker Avance DRX-400 spectroscopy instrument.

General procedure for synthesis of {2-[Ph₂P(O)]C₆H₄NC(tBu)(2,6-Me₂C₆H₃)}Y(N(SiMe₃)₂)₂ (2)

A solution of Y[N(SiMe₃)₂]₃ (0.342 g, 0.60 mmol) in 20 mL of THF was added to solution of 2-[Ph₂P(O)]PhNHC(tBu)(2,6-Me₂C₆H₃) (1) (0.288 g, 0.60 mmol) in 20 mL of THF. The reaction mixture was stirred at room temperature for 24 h and the volatiles were removed in vacuum. Recrystallization of the solid residue from THF/hexane mixture afforded yellow transparent crystals in 60% yield (0.320 g). Anal. calcd for C₄₃H₆₈N₄OPSi₄Y (889.25): C, 58.08; H, 7.71; N, 6.30; Y, 9.99. Found: C, 57.81; H, 7.38; N, 6.07; Y, 10.10. ¹H NMR (400 MHz, C₆D₆, 25 °C): δ 0.39 (s, 36H, N(Si(CH₃)₃)₂), 0.65 (s, 9H, C(CH₃)₃), 2.48 (s, 6H, C₆H₃(CH₃)₂), 6.49–6.54 (m, 1H, Aryl-H), 6.85–7.11 (m, 12H, Aryl-H), 7.61, 7.77 (br. s, both 2H, Aryl-H). ¹³C{¹H} NMR (100.62 MHz, C₆D₆, 25 °C): δ 5.6 (N(Si(CH₃)₃)₂), 19.9, 22.8 (C₆H₃(CH₃)₂), 30.0 (C(CH₃)₃), 41.7 (d, ³J_CY = 2.3 Hz, C(CH₃)₃), 118.1 (d, J_CP = 108.1 Hz), 119.2 (d, J_CP = 14.3 Hz), 123.7, 126.9 (d, J_CP = 7.5 Hz), 128.5 (d, J_CP = 13.0 Hz), 132.6, 133.3 (d, J_CP = 1.8 Hz), 133.6 (d, J_CP = 13.3 Hz), 147.1, 156.1 (d, J_CP = 3.2 Hz), (Aryl-C), 176.2 (d, ²J_CY = 1.8 Hz) (NCN). ³¹P NMR (161.9 MHz, C₆D₆, 25 °C): δ 41.2 (d, ²J_PY = 6.4 Hz). IR (Nujol, KBr; ν (cm⁻¹)): 1686 (m), 1556 (w), 1410 (s), 1243 (s), 1216 (w), 1179 (m), 1166 (w), 1586 (m), 1065 (m), 1029 (w), 958 (s), 882 (m), 863 (w), 833 (s), 778 (w), 758 (w), 746 (w), 707 (w), 694 (m), 664 (m), 607 (m).

General procedure for synthesis of {2-[Ph₂P(O)]C₆H₄NC(tBu)(2,6-Me₂C₆H₃)}Nd(N(SiMe₃)₂)₂ (3)

Analogous synthetic procedure was used. 1 (0.207 g, 0.43 mmol), Nd(N(SiMe₃)₂)₃ (0.240 g, 0.43 mmol). Green crystals of 3 were isolated in 61% yield (0.247 g). Anal. calcd for C₄₃H₆₈N₄NdOPSi₄ (944.58): C, 54.68; H, 7.25; N, 5.93; Nd, 15.27. Found: C, 54.39; H, 7.04; N, 5.77; Nd, 15.51. IR (Nujol, KBr; ν (cm⁻¹)): 1588 (m), 1554 (w), 1407 (w), 1244 (s), 1213 (w), 1175 (m), 1162 (w), 1126 (s), 1088 (m), 1066 (m), 1030 (w), 976 (s), 936 (m), 878 (m), 858 (w), 832 (s), 769 (w), 754 (w), 705 (m), 693 (m), 661 (m), 609 (m).

General procedure for synthesis of {2-[Ph₂P(O)]C₆H₄NC(tBu)(2,6-Me₂C₆H₃)}La(N(SiMe₃)₂)₂ (4)

Analogous synthetic procedure was used. 1 (0.257 g, 0.53 mmol), La(N(SiMe₃)₂)₃ (0.332 g, 0.53 mmol). Yellow crystals of 4 were isolated in 72% yield (0.358 g). Anal. calcd for C₄₃H₆₈LaN₄OPSi₄ (939.25): C, 54.99; H, 7.29; N, 5.97; La, 14.79. Found: C, 54.64; H, 7.17; N, 6.11; La, 14.84. ¹H NMR (400 MHz, C₆D₆, 25 °C): δ 0.37 (s, 36H, NSi((CH₃)₃)₂), 0.74 (s, 9H, C(CH₃)₃), 2.43, 2.46 (s, both 3H, C₆H₃(CH₃)₂), 6.43 (td, J_HH = 7.1, J_HH = 3.2 Hz, 1H, Aryl-H), 6.76–7.11 (m, 12H, Aryl-H), 7.64–7.69 (m, 2H, Aryl-H), 7.73–7.82 (m, 2H, Aryl-H). ¹³C{¹H} NMR (100.62 MHz, C₆D₆, 25 °C): δ 4.5, 4.8 (NSi((CH₃)₃)₂), 20.1, 22.1 (C₆H₃(CH₃)₂), 30.0 (C(CH₃)₃), 42.8 (C(CH₃)₃), 117.3 (d, J_CP = 109.8 Hz), 117.2 (d, J_CP = 14.4 Hz), 123.4, 124.9 (d, J_CP = 7.5 Hz), 128.2 (d, J_CP = 15.7 Hz), 128.4, 128.5, 128.5, 128.6, 132.2, 132.3, 132.3, 132.4, 133.3 (d, J_CP = 1.8 Hz), 133.9 (d, J_CP = 13.2 Hz), 144.7, 155.6 (d, J_CP = 3.5 Hz) (Aryl-C), 177.1 (NCN). ³¹P NMR (161.9 MHz, C₆D₆, 25 °C): δ 39.7. IR (Nujol, KBr; ν (cm⁻¹)): 1675 (m), 1584 (s), 1573 (s), 1549 (m), 1419 (s), 1276 (w), 1251 (m), 1214 (m), 1169 (m), 1161 (m), 1139 (m), 1126 (w), 1120 (w), 1094 (s), 1069 (s), 1038 (s), 1023 (w), 998 (w), 978 (s), 933 (s), 871 (s), 836 (m), 822 (m), 764 (s), 747 (s), 735 (s), 692 (s), 665 (m), 625 (w), 601 (w).

General procedure for rac-lactide polymerization

In a typical experiment (Table 2, entry 1), in a glovebox, a Schlenk flask was charged with a solution of 2 (8.9 mg, 10 μmol) in toluene (1.0 mL). To this solution, rac-lactide (0.144 g, 1.0 mmol, 100 equiv.) was added rapidly. The mixture was immediately stirred with a magnetic stir bar at 20 °C for 30 min. After an aliquot of the crude material was sampled by pipette for determining monomer conversion by ¹H NMR, the reaction was quenched by adding of 1.0 mL of a 10% H₂O solution in THF, and the polymer was precipitated from CH₂Cl₂/pentane (ca. 2 : 100 mL) five times. The polymer was dried in vacuo to a constant weight.

General procedure for isoprene polymerization

A typical polymerization procedure was carried out as following. Under a nitrogen atmosphere and room temperature borate (0.01 mmol) was added to a solution of (2–4) (0.01 mmol) in toluene (1 mL) and 10 equivalents of AliBu₃ (0.1 mmol). After which to the reaction mixture isoprene (1000 equiv., 1 mL) was added. Polymerization was initiated and carried out for 1–24 hours. The reaction mixture was quenched by the addition of ethanol, and then poured into a large amount of ethanol to precipitate the polymer, which was dried under vacuum at 60 °C to a constant weight. Then, 1,4- 3,4-regioselectivity was determined by ¹H and ¹³C{¹H} NMR spectroscopy. GPC of polyisoprenes was performed in THF at 20 °C. The number average molecular masses (M_n) and polydispersity indexes (M_w/M_n) of the polymers were calculated with reference to a universal calibration against polystyrene standards.

X-ray crystallography

The X-ray data for 2–4 were collected on Bruker Smart Apex (2) and Agilent Xcalibur (EOS) (3, 4) diffractometers (graphite-monochromated, MoK_α radiation, ω-scan technique, λ = 0.71073 Å, T = 100(2) K). SADABS²⁶ and ABSPACK²⁷ were used to perform area-detector scaling and absorption corrections. The structures were solved by direct methods and were refined on F² using SHELX.²⁸ All non-hydrogen atoms were found from Fourier syntheses of electron density and were refined anisotropically. All hydrogen atoms were placed in geometrically idealized positions and treated as riding with U_iso(H) = 1.2U_eq (U_iso(H) = 1.5U_eq for the hydrogen atoms in CH₃ groups) of their parent atoms. The CCDC files 1439443 (2), 1439444 (3), 1439445 (4) contain the supplementary crystallographic data for this paper.†

Acknowledgements

This work was financially supported by the Russian Science Foundation (Project 14-13-00742).

Notes and references

1. (a) M. F. Lappert, A. V. Protchenko, P. P. Power and A. L. Seeber, Amides of the Group 3 and Lanthanide Metals, in Metal Amide Chemistry, John Wiley & Sons Ltd, Chichester, UK, 2008; (b) Lanthanide amides: R. Anwander, Top. Curr. Chem., 1996, 179, 33–112.
2. (a) D. C. Bradley, Chem. Rev., 1989, 89, 1317; (b) L. G. Hubert-Pfalzgraf, New J. Chem., 1987, 11, 663; (c) L. G. Hubert-Pfalzgraf, Appl. Organomet. Chem., 1992, 6, 627.
3. (a) Y. Liang and R. Anwander, Dalton Trans., 2013, 42, 12521; (b) S. Hong and T. J. Marks, Adv. Chem. Res., 2004, 37, 673; (c) T. E. Müller, K. C. Hultzsch, M. Yus, F. Foubelo and M. Tada, Chem. Rev., 2008, 108, 3795; (d) K. C. Hultzsch, Adv. Synth. Catal., 2005, 347, 367; (e) M. R. Douglass, C. L. Stern and T. J. Marks, J. Am. Chem. Soc., 2001, 123, 10221; (f) I. V. Basalov, S. C. Roşca, D. M. Lyubov, A. N. Selikhov, G. K. Fukin, Y. Sarazin, J. F. Carpentier and A. A. Trifonov, Inorg. Chem., 2014, 53, 1654; (g) I. V. Basalov, A. N. Dorcet, G. K. Fukin, Y. Sarazin, J. F. Carpentier and A. A. Trifonov, Chem.–Eur. J., 2015, 21, 6033; (h) V. Rodriguez-Ruiz, R. Carlino, S. Bezzenine-Lafolée, R. Gil, D. Prim, E. Schulz and J. Hannedouche, Dalton Trans., 2015, 44, 12029; (i) C. Zeng, D. Yuan, B. Zhao and Y. Yao, Org. Lett., 2015, 17, 2242; (j) M. Rastätter, A. Zulys and P. W. Roesky, Chem.–Eur. J., 2007, 13, 3606; (k) C. Wang, L. Huang, M. Lu, B. Zhao, Y. Wang, Y. Zhang, Q. Shen and Y. Yao, RSC Adv., 2015, 5, 9476; (l) H. Cheng, Y. Xiao, C. Lu, B. Zhao, Y. Wang and Y. Yao, New J. Chem., 2015, 39, 7667; (m) Z. Li, M. Xue, H. Yao, H. Sun, Y. Zhang and Q. Shen, J. Organomet. Chem., 2012, 713, 27.
4. T. J. Woodman, Y. Sarazin, G. Fink, K. Hauschild and M. Bochmann, Macromolecules, 2005, 38, 3060.
5. Y. Luo, X. Feng, Y. Wang, S. Fan, J. Chen, Y. Lei and H. Liang, Organometallics, 2011, 30, 3270.
6. L. Zhou, Y. Yao and Q. Shen, J. Appl. Polym. Sci., 2009, 114, 2403.
7. (a) H. X. Li, Z. G. Ren, Y. Zhang, Z. H. Zhang, J. P. Lang and Q. Shen, J. Am. Chem. Soc., 2005, 127, 1122; (b) H. Ma and J. Okuda, Macromolecules, 2005, 38, 2665; (c) H. E. Dyer, S. Huijser, N. Susperregui, F. Bonnet, A. D. Schwarz, R. Duchateau, L. Maron and P. Mountford, Organometallics, 2010, 29, 3602; (d) R. H. Platel, L. M. Hodgson and C. K. Williams, Polymer, 2008, 48, 11.
8. S. P. Nolan, D. Stern and T. J. Marks, J. Am. Chem. Soc., 1989, 111, 7844.
9. (a) V. Monteil, R. Spitz and C. Boisson, Polym. Int., 2004, 53, 576; (b) O. Tardif and S. Kaita, Dalton Trans., 2008, 19, 2531; (c) Y. Luo, Y. Lei, S. Fan, Y. Wang and J. Chen, Dalton Trans., 2013, 42, 4040; (d) Y. Luo, S. Fan, J. Yang, J. Fang and P. Xu, Dalton Trans., 2011, 40, 3053; (e) F. Chen, S. Fan, Y. Wang, J. Chen and Y. Luo, Organometallics, 2012, 31, 3730; (f) V. Monteil, R. Spitz and C. Boisson, Polym. Int., 2004, 53, 576; (g) C. Boisson, F. Barbotin and R. Spitz, Macromol. Chem. Phys., 1999, 200, 1163; (h) N. Martins, F. Bonnet and M. Visseaux, Polymer, 2014, 55, 5013; (i) R. Anwander, M. G. Klimpel, H. M. Dietrich, D. J. Shorokhov and W. Scherer, Chem. Commun., 2003, 1008.
10. A. O. Tolpygin, T. A. Glukhova, A. V. Cherkasov, G. K. Fukin, D. V. Aleksanyan, D. Cui and A. A. Trifonov, Dalton Trans., 2015, 44, 16465.
11. D. C. Bradley, J. S. Ghotra and F. A. Hart, J. Chem. Soc., Dalton Trans., 1973, 10, 1021.
12. K. B. Aubrecht, K. Chang, M. A. Hillmyer and W. B. Tolman, J. Polym. Sci., Part A: Polym. Chem., 2001, 39, 284.
13. Y. Wang, Y. Luo, J. Chen, H. Xue and H. Liang, New J. Chem., 2012, 36, 933.
14. Y. Luo, P. Xu, Y. Lei, Y. Zhang and Y. Wang, Inorg. Chim. Acta, 2010, 363, 3597.
15. K. Kincaid, C. P. Gerlach, G. R. Giesbrecht, J. R. Hagadorn, G. D. Whitener, A. Shafir and J. Arnold, Organometallics, 1999, 18, 5360.
16. F. Wang, Y. Wei, S. Wang, X. Zhu, S. Zhou, G. Yang, X. Gu, G. Zhang and X. Mu, Organometallics, 2015, 34, 86.
17. R. D. Shannon, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr., 1976, 32, 751.
18. J. F. Corbey, J. H. Farnaby, J. E. Bates, J. W. Ziller, F. Furche and W. J. Evans, Inorg. Chem., 2012, 51, 7867.
19. H. C. Aspinall, S. R. Moore and A. K. Smith, J. Chem. Soc., Dalton Trans., 1992, 1, 153.
20. (a) M. Zimmermann, K. W. Törnroos and R. Anwander, Angew. Chem., Int. Ed., 2008, 47, 775; (b) D. Li, S. Li, D. Cui and X. Zhang, Organometallics, 2010, 29, 2186; (c) K. Lv and D. Cui, Organometallics, 2010, 29, 2987; (d) L. Wang, D. Cui, Z. Hou, W. Li and Y. Li, Organometallics, 2011, 30, 760; (e) D. Robert, E. Abinet, T. P. Spaniol and J. Okuda, Chem.–Eur. J., 2009, 15, 11937; (f) E. Le Roux, Y. Liang, K. W. Törnroos, F. Nief and R. Anwander, Organometallics, 2012, 31, 6526; (g) S. Li, D. Cui, D. Li and Z. Hou, Organometallics, 2009, 28, 4814; (h) G. Du, Y. Wei, L. Ai, Y. Chen, Q. Xu, X. Liu, S. Zhang, Z. Hou and X. Li, Organometallics, 2011, 30, 160.
21. L. Guo, X. Zhu, S. Zhou, X. Mu, Y. Wei, S. Wang, Z. Feng, G. Zhanga and B. Deng, Dalton Trans., 2014, 43, 6842.
22. (a) C. K. Ha and J. A. Gardella Jr, Chem. Rev., 2005, 105, 4205; (b) O. D. Cabaret, B.-M. Vaca and D. Bourissou, Chem. Rev., 2004, 104, 6147; (c) R. E. Drumright, P. R. Gruber and D. E. Henton, Adv. Mater., 2000, 12, 1841; (d) M. H. Chisholm and Z. Zhou, J. Mater. Chem., 2004, 14, 3081.
23. (a) M. J. Stanford and A. P. Dove, Chem. Soc. Rev., 2010, 39, 486; (b) C. M. Thomas, Chem. Soc. Rev., 2010, 39, 165; (c) M. V. Yakovenko, N. Y. Udilova, T. A. Glukhova, A. V. Cherkasov, G. K. Fukin and A. A. Trifonov, New J. Chem., 2015, 39, 1083.
24. (a) V. Rad'kov, T. Roisnel, A. Trifonov, J.-F. Carpentier and E. Kirillov, Eur. J. Inorg. Chem., 2014, 25, 4168; (b) S. Sun, K. Nie, Y. Tan, B. Zhao, Y. Zhang, Q. Shen and Y. Yao, Dalton Trans., 2013, 42, 2870; (c) J. Wang, T. Cai, Y. Yao, Y. Zhang and Q. Shen, Dalton Trans., 2007, 45, 5275; (d) Q. Wang, F. Zhang, H. Song and G. Zi, J. Organomet. Chem., 2011, 696, 2186; (e) Y. Wang, Y. Lei, S. Chi and Y. Luo, Dalton Trans., 2013, 42, 1862; (f) F. Zhang, J. Zhang, H. Song and G. Zi, Inorg. Chem. Commun., 2011, 14, 72.
25. S. J. Lyle and M. M. Rahman, Talanta, 1953, 10, 1177.
26. Bruker, SADABS, Bruker AXS Inc, Madison, Wisconsin, USA, 2001.
27. Agilent, CrysAlis PRO, Agilent Technologies Ltd, Yarnton, Oxfordshire, England, 2014.
28. G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Adv., 2015, 71, 3–8.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 1439443 (2), 1439444 (3) and 1439445 (4). For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c5ra27960g

---