Alkylamido lutetium complexes as prospective lutetium imido precursors: synthesis, characterization and ligand design

Abstract

Mixed alkylamido lutetium complexes, L^iPrLu(CH₂SiMe₃)(NHCPh₃) (7_CPh3) and L^iPrLu(CH₂SiMe₃)(NHDipp) (7_Dipp) (L^iPr = 2,5-[ⁱPr₂P = N(4-ⁱPrC₆H₄)]₂C₄H₂N⁻), were synthesized by addition of a bulky primary amine, NH₂R (R = CPh₃, Dipp) (Dipp = 2,6-ⁱPr₂C₆H₃) to the dialkyl complex L^iPrLu(CH₂SiMe₃)₂ (6). Unlike complexes supported by the related pincer ligand L^Ph (L^Ph = 2,5-[Ph₂P = N(4-ⁱPrC₆H₄)]₂C₄H₂N⁻) these species proved resistant to C–H cyclometalative processes. Attempts to access lutetium imdes via addition of 4-dimethylaminopyridine (DMAP) to 7_CPh3 and 7_Dipp promoted disproportionation, affording 0.5 equivalents of the corresponding bisamide complexes L^iPrLu(NHCPh₃)₂ (8_CPh3) and L^iPrLu(NHDipp)₂ (8_Dipp), respectively, as well as 0.5 equivalents of L^iPrLu(CH₂SiMe₃)₂, which decomposed in the presence of DMAP. Incorporation of internal Lewis bases was accomplished by replacing the N-aryl substituents in L^iPr with 4,6-dimethylpyrimidine groups (L^Pm, 11). The correspondng dialkyl lutetium complex L^PmLu(CH₂SiMe₃)₂ (12) was prepared, from which loss of SiMe₄ occured over a period of hours in benzene-d₆ solution.

---

Introduction

Complexes that feature multiple bonding between transition metals and carbon, nitrogen, or other main group elements have led to remarkable new stoichiometric and catalytic transformations that have found great utility across numerous disciplines.¹ Although such success prompted tremendous efforts to isolate comparable rare earth analogues, they remained largely elusive until the past two decades. In particular, rare earth complexes bearing a terminal imido functionality (RE = NR, RE = group III and lanthanide elements) were targeted with great enthusiasm, yet prior to 2010 only μ₂- and μ₃-bridging species,^2–4 and complexes wherein a transient RE = NR moiety activates a C–H bond, were reported.^5–7 Accordingly, the discovery by Chen and co-workers that addition of one equivalent of DMAP to the scandium alkylamido complex, L₁Sc(NHDipp)(CH₃) (L₁ = DippNC(CH₃)CHC(CH₃)NCH₂CH₂N(CH₃)₂, Dipp = 2,6-ⁱPr₂C₆H₃, DMAP = 4-(CH₃)₂NC₅H₄N), affords the terminal scandium imido L₁(DMAP)Sc = NDipp (1, Fig. 1) drew considerable attention.⁸ Notably, DMAP can be removed by reaction of complex 1 with 9-BBN (9-BBN = 9-borabicyclo(3.3.1)nonane), generating the unstable base-free imide L₁Sc = NDipp, which readily participates in [2 + 2] cycloaddition reactions and activation of C_sp²–H, C_sp²–F, and B–H bonds.⁹ The following year Chen demonstrated that methane elimination can be induced in the absence of DMAP if a tethered Lewis base is incorporated into the β-diketiminato ancillary ligand.¹⁰ The corresponding scandium imide, L₂Sc = NDipp (2, L₂ = DippNC(CH₃)CHC(CH₃)NCH₂CH₂N(CH₃)CH₂CH₂N(CH₃)₂, Fig. 1), has exhibited rich reaction chemistry with unsaturated organic molecules,¹¹ main group elements,¹⁰ transition metal coordination complexes,¹² and more.¹³ Ultimately, seminal contributions from Chen,¹³ Mindiola,^5,6 Piers⁷ and Cui¹⁴ demonstrated that given the appropriate system, introduction of either internal or external Lewis bases to a scandium alkylamido complex is an effective strategy for accessing terminal imido functionalities.

Fig. 1 Selected examples of structurally characterized terminal rare earth imido and alkylamido complexes.

Following Chen's approach, Anwander prepared DMAP-stabilized imide complexes of the larger rare earth metals, yttrium and lutetium (Tp^tBu,MeY = N-2,6-(CH₃)₂C₆H₃(DMAP) (3a, Fig. 1) and Tp^tBu,MeLu = N-3,5-(CF₃)₂C₆H₃(DMAP)) (3b, Fig. 1) (Tp^tBu,Me = HB[3-CH₃-5-^tBu-N₂C₃H]₃⁻).¹⁵ More recently, Schelter expanded the list of terminal rare earth imides to include the cerium species K[(TriNOx)Ce = N-3,5(CF₃)₂C₆H₃] (TriNOx = N(CH₂-2-C₆H₄N(^tBu)O)₃); Anwander added the dysprosium and holmium complexes Tp^tBu,MeRE(N-2,6-ⁱPr₂C₆H₃)(DMAP) (RE = Dy, Ho).^16–18 Notably, these examples utilize starting materials that include a mixed methyl/tetramethylgallato system, Tp^tBu,MeRE(Me)(GaMe₄) (RE = Y, Dy, Ho), that reacts with H₂NDipp to yield mixed methylamido species primed for imide-formation.

It is important to note that all reported rare earth complexes bearing a terminal imido functionality contain an aromatic group on the imide nitrogen.^{4–8,10,14–18} The apparent requirement for aryl substitution piqued our interest, particularly given that neither Tp^tBu,MeLu(CH₃)(NHAd) (4a, Fig. 1) nor Tp^tBu,MeLu(CH₃)(NH^tBu) (4b, Fig. 1), served as appropriate precursors for a lutetium imide.¹⁹ Our group previously synthesized L^PhLu(CH₂SiMe₃)(NHCPh₃), (L^Ph = 2,5-[Ph₂P = N(Pipp)]₂C₄H₂N⁻, Pipp = 4-ⁱPrC₆H₄) in hopes of observing similar reactivity;²⁰ however, exhaustive efforts (e.g. addition of Lewis or Brønsted bases, heating, etc.) failed to yield the targeted imido L^PhLu = NCPh₃. Instead, C_sp²–H activation of the CPh₃ substituent occurred.²⁰ Herein we utilize the more electron rich NNN-pincer ligand L^iPr (L^iPr = 2,5-[ⁱPr₂P = N(Pipp)]₂C₄H₂N⁻)²¹ to alleviate unwanted activations and probe the impact that the amine bound substituent has on rare earth imido synthesis. Furthermore, with the goal of expanding the library of Lewis bases that can promote alkane loss from rare earth alkylamido species, a second NNN-pincer that features 4,6-dimethyl pyrimidine groups, L^Pm = [ⁱPr₂P = N(4,6-(Me)₂C₄H₄N₂)]₂C₄H₂N⁻, was synthesized. Unfortunately, mixed alkylamido lutetium species were unable to be isolated with this ligand, due to spontaneous loss of SiMe₄.

Results and discussion

Synthesis of L^iPrSc(CH₂SiMe₃)₂ and L^iPrLu(CH₂SiMe₃)₂

The dialkyl complexes L^iPrRE(CH₂SiMe₃)₂ (L^iPr = 2,5-[ⁱPr₂P = N(Pipp)]₂C₄H₂N⁻), 5 (RE = Sc) and 6 (RE = Lu), were prepared by reaction of HL^iPr with RE(CH₂SiMe₃)₃THF₂ in toluene for 1 hour (Scheme 1). Production of L^iPrSc(CH₂SiMe₃)₂ and L^iPrLu(CH₂SiMe₃)₂ was supported by the consumption of ¹H and ³¹P NMR resonances attributed to HL^iPr, as well as the emergence of a resonance at δ 0.00 in the ¹H NMR spectra that integrated for 12H, attributed to eliminated SiMe₄. New single resonances in the ³¹P{¹H} NMR spectra (5δ 48.0, 6δ 49.3) similarly support the synthesis of discrete metal complexes in high purity. As previously observed,²¹ the P–(CH(CH₃)₂)₂ methyl groups are chemically inequivalent on the NMR timescale (5: δ 0.93, 0.88, 6: δ 0.96, 0.85). Like L^PhLu(CH₂SiMe₃)₂, no C_sp²–H bond activation was observed at ambient temperature in solution.²²

Scheme 1 Synthesis and reactivity of complexes 5 and 6.

X-Ray diffraction analysis of complexes 5 and 6

Single crystals suitable for diffraction analysis of both complexes 5 and 6 were obtained by slowly cooling warm (50 °C) heptane solutions of each to ambient temperature. Both 5 and 6 crystallized in the P2_1/n space group (Fig. 2) and the geometry about each metal centre is best described as distorted square pyramidal (τ⁵ = 0.37 (5) τ⁵ = 0.31 (6)). The solid-state structures revealed Lu–C35, Lu–C39, and Lu–N1 distances that were, on average, 0.117 Å longer than those in the scandium congener. As expected, the phosphinimine P–N bond lengths are similar in both complexes (5: 1.619(2) Å, 6: 1.611(3) Å) and elongated relative to HL^iPr.²¹ These contacts are also comparable to those observed in rare earth complexes supported by the related pincer ligand, L^Ph (1.606(4)–1.610(3) Å).^22,23

Fig. 2 Solid-state structure of complex 6 with thermal ellipsoids represented at 50% probability. hydrogen atoms omitted for clarity. Selected bond lengths (Å) and angles (°): Lu1–N1 2.305(2), Lu1–N2 2.333(3), Lu1–C35 2.360(4), P1–N2 1.611(3), C35–Lu1–C39 112.1(1), N2–Lu1–N3 143.79(9), N1–Lu1–C35 125.0(1), N1–Lu1–C39 122.9(1).

Mixed alkylamido lutetium complexes

In order to probe what impact swapping P–Ph₂ for P–ⁱPr₂ groups has on the ability of our ligands to support rare earth terminal imido precursors, complexes 5 and 6 were independently exposed to one equivalent of various primary amines (Ph₃CNH₂, DippNH₂, AdNH₂, or Mes*NH₂). Unfortunately, compound 5 did not react with Ph₃CNH₂, nor Mes*NH₂ (Mes* = 2,4,6-^tBu₃C₆H₂) even upon exposure to heat. Slow, cold addition of 1-adamantylamine (AdNH₂) to complex 5 promoted formation of an inextricable mixture of L^iPrSc(CH₂SiMe₃)₂ (5), L^iPrSc(CH₂SiMe₃)(NHAd), and L^iPrSc(NHAd)₂ as indicated by multinuclear NMR spectroscopy. With complex 5 not undergoing successful protonolysis with primary amines, focus was directed to L^iPrLu(CH₂SiMe₃)₂ (6).

Addition of Ph₃CNH₂ and DippNH₂ to compound 6 gave rise to complexes 7_CPh3, L^iPrLu(CH₂SiMe₃)(NHCPh₃), and 7_Dipp, L^iPrLu(CH₂SiMe₃)(NHDipp), respectively (Scheme 1). As expected, one equivalent of SiMe₄ (δ 0.00) was liberated and observed in both reaction mixtures via¹H NMR spectroscopy. Furthermore, a sharp singlet (assigned as the N–H resonance), integrating for one proton, which did not give rise to cross peaks in ¹H–¹H COSY or ¹H–¹³C HSQC experiments, emerged (7_CPh3δ 2.43, 7_Dippδ 4.14). Installation of the NHCPh₃/NHDipp ligand produced C_s-symmetric complexes, resulting in the appearance of two sets of P–(CH(CH₃)₂)₂ resonances in the ¹H NMR spectra of complexes 7_CPh3 (δ 2.16, 1.75) and 7_Dipp (δ 2.26, 1.93). Further evidence for the formation of these mixed alkylamido compounds was provided by the disappearance of the singlet corresponding to L^iPrLu(CH₂SiMe₃)₂ (δ 49.3) in tandem with the emergence of a new, resonance (7_CPh3δ 47.9, 7_Dippδ 48.7) in the ³¹P{¹H} spectra. Complexes 7_CPh3 and 7_Dipp were resistant to the intramolecular C–H bond activation reported for our previous generation lutetium alkylamido complex L^PhLu(CH₂SiMe₃)(NHCPh₃),²⁰ with no decomposition observed after heating to 70 °C for 72 hours.

X-ray diffraction analysis of 7_CPh3 and 7_Dipp

Minimal quantities of warm (50 °C) heptane were used to produce saturated solutions of compounds 7_CPh3 and 7_Dipp. The solutions were allowed to cool to ambient temperature before being placed in a –35 °C freezer for 18 hours, yielding X-ray quality crystalline material. The solid-state geometry about the lutetium centres in 7_CPh3 and 7_Dipp is best described as distorted square pyramidal (τ⁵ = 0.15 (7_CPh3) τ⁵ = 0.12 (7_Dipp)) with the four nitrogenous atoms forming the square base in each structure (Fig. 3 and 4). Two equivalents of 7_CPh3 were found within the asymmetric unit; there were no notable structural differences between them.

Fig. 3 Solid-state structure for 7_Dipp with thermal ellipsoids represented at 50% probability. Hydrogen atoms omitted for clarity. Selected bond lengths (Å) and angles (°) for 7_Dipp: Lu1–N1 2.283(5), Lu1–N2 2.352(4), Lu1–N3 2.344(5), Lu1–C35 2.313(5), P1–N2 1.625(5), Lu1–N4 2.195(5), C35–Lu1–N4 110.5(2), N3–Lu1–N2 145.3(2). N1–Lu1–N4 138.4(2).

Fig. 4 Solid-state structure for 7_CPh3 with thermal ellipsoids represented at 50% probability. Hydrogen atoms, Pipp isopropyl groups and one equivalent of 7_CPh3 have been omitted for clarity. Selected average bond lengths (Å) and angles (°) for 7_CPh3: Lu1–N1 2.323(2), Lu–N2 2.342(3), P1–N2 1.617(2), Lu1–C35 2.344(2), Lu1–N4 2.154(2), N3–Lu1–N2 142.9(1), C35–Lu1–N4 105.3(1), N1–Lu1–N4 135.0(1).

The Lu1–C35 bond lengths of complexes 7_CPh3 (2.342(3) Å) and 7_Dipp (2.313(5) Å) are slightly shorter than the Lu–C distances in L^iPrLu(CH₂SiMe₃)₂ (Lu1–C35: 2.360(4) Å, Lu1–C39: 2.366(4) Å). However, these distances lie at the low end of Lu–C bonds for published lutetium alkylamido complexes (2.301(4)–2.40(1) Å).^{14,15,19,24–29} The phosphinimine P–N bond lengths in both complexes 7_CPh3 and 7_Dipp are not significantly different than those in 6.

To the best of our knowledge, 4a (Tp^tBu,MeLu(CH₃)(NHAd)), 4b (Tp^tBu,MeLu(CH₃)(NH^tBu)), and L^PhLu(NHCPh₃)₂ represent rare examples of structurally characterized lutetium complexes which possess an amido ligand bearing a non-aromatic substituent.^19,20 Complex 7_CPh3 exhibits a marginally longer Lu–NHCPh₃ bond (2.154(2) Å) than those reported in the literature (2.126(2) Å to 2.144(2) Å for complexes with non-aromatic substituents on nitrogen). Conversely, numerous examples of complexes bearing a monodentate Lu–NHAr (Ar = aromatic) functionality are known, with Lu–NHAr bond lengths ranging from 2.144(3)–2.245(4) Å.^{14,15,24–28} The Lu1–N4 distance in complex 7_Dipp (2.195(5) Å) falls in the middle of this range. Notably, complex 7_CPh3 has a shorter Lu1–N4 length of 2.154(2) Å than 7_Dipp (2.195(5) Å), despite the large steric presence of the NHCPh₃ group (Table 1).

Table 1 Crystallographic table for complexes, 7_CPh3, 7_Dipp, 8_CPh3, and 8_Dipp with data for complexes 4b (Tp^tBu,MeLu(CH₃)(NH^tBu)) and L^PhLu(NHCPh₃)₂ provided for comparison (RE = Sc or Lu)

Parameter	7CPh3	7Dipp	4b	8CPh3	8Dipp	L^PhLu(NHCPh3)2
Crystal system	Triclinic	Monoclinic	Monoclinic	Triclinic	Monoclinic	Triclinic
Space group	P1̄	P21	P21/n	P1̄	P21/n	P1̄
R1 (I > 2σ(I))	0.0585	0.0298	0.0211	0.0349	0.0297	0.0311
Selected bond lengths (Å)
RE1–C35	2.344(2)	2.313(5)	2.362(3)
RE1–N1	2.323(2)	2.283(5)		2.327(2)	2.280(2)	2.293(4)
P1–N2	1.617(2)	1.625(5)		1.617(3)	1.615(2)	1.620(4)
RE1–N4	2.154(2)	2.195(5)	2.126(2)	2.153(2)	2.176(2)	2.144(2)
Selected bond angles (°)
N2–RE1–N3	142.9(1)	145.3(2)		143.59(8)	143.76(7)	143.7(1)
N1–RE1–C35	119.65(2)	111.1(2)
N4–RE1–C35	105.3(1)	110.5(2)	92.81(9)
N1–RE1–N4	135.0(1)	138.4(2)		132.20(8)	119.68(8)	107.0(1)

---

Reactivity with 4-dimethylaminopyridine (DMAP)

Reaction of L^iPrLu(CH₂SiMe₃)(NHCPh₃), 7_CPh3, with one equivalent of DMAP promoted slow conversion (12% of compound 7_CPh3 consumed over 68 hours) to a new product which exhibited a single resonance in the ³¹P{¹H} NMR spectrum (δ 45.3). Another 12 hours at ambient temperature did not significantly change the product distribution. Heating this sample to 60 °C for 93 hours produced a dark brown, cloudy solution. This mixture contained a myriad of resonances in the ³¹P{¹H} NMR spectrum, with the major product (δ 45.3) comprising 47% of the phosphorous-containing material.

Repeating the experiment at 45 °C resulted in formation of a similar mixture, but this time the resonance at δ 45.3 integrated to only 25% of the material. Increasing the ratio of complex 7_CPh3 and DMAP to 1 : 2 led to slower consumption of 7_CPh3 (40% consumed after 4 days), with minimal change to the product mixture.

Exposure of L^iPrLu(CH₂SiMe₃)(NHDipp), 7_Dipp, to one equivalent of DMAP at 45 °C for 22.5 hours resulted in 20% consumption of 7_Dipp and formation of one major product (δ 46.4) according to the ³¹P{¹H} NMR spectrum. Further heating (83.5 hours) resulted in a murky brown solution. In this case, the ³¹P{¹H} NMR spectrum revealed that 60% of compound 7_Dipp had been consumed and the new resonance appearing at δ 46.4 accounted for 25% of the phosphorous-containing material. In addition, a second substantial resonance (21% by integration) appeared (δ 44.2), in conjunction with a variety of by-products.

Complex 7_Dipp was also reacted with two equivalents of DMAP at 45 °C. After 18 hours the ³¹P{¹H} NMR spectrum revealed 33% of complex 7_Dipp had been consumed, along with concomitant formation of one major product (δ 46.4) accounting for 25% of the phosphorous-containing material. Heating this sample for an additional 145.5 hours resulted in almost complete consumption of 7_Dipp, and similar to the equimolar reaction, the resonance at δ 46.4 remained the dominant product (43% of all signals).

Many commonalities were noted when comparing the ¹H NMR spectra for the reactions of 7_Dipp and 7_CPh3 across the various conditions. The unaltered chemical shifts attributed to free DMAP (δ 8.45, 6.07, 2.21) suggested that the Lewis base did not coordinate to the metal centres. Approximately one equivalent of SiMe₄ was produced (δ 0.00) and there was no evidence for the retention of any –CH₂SiMe₃ functionalities previously attached to the metal centre (complex 6). The observation of only one P–(CH(CH₃)₂)₂ resonance, coupled with the lone signal in the ³¹P{¹H} spectra, implies the major product formed in these reactions possesses C_2v symmetry. Furthermore, a singlet (δ 4.38 and 2.26) with no cross-peaks in ¹H–¹H COSY experiments grew in each of the reaction mixtures. Diagnostic aromatic resonances pertaining to the NHCPh₃ ligand were retained throughout the reaction of complex 7_CPh3 and DMAP, eventually doubling in integration with respect to DMAP. Similarly, retention and growth of the isopropyl doublet and septet signals pertaining to the NHDipp ligand were observed in the reaction between complex 7_Dipp and DMAP. The combination of this evidence led us to believe that the major products of the reaction between complexes 7 and DMAP were the respective bisamide species, L^iPrLu(NHCPh₃)₂ (8_CPh3) and L^iPrLu(NHDipp)₂ (8_Dipp). While attempts at isolating these complexes directly from the crude reaction mixtures were unsuccessful, independent synthesis of both 8_CPh3 and 8_Dipp was achieved by addition of two equivalents of NH₂CPh₃ or NH₂Dipp, respectively, to stirring toluene solutions of L^iPrLu(CH₂SiMe₃)₂. Heat (70 °C), and additional time (69 hours) was required to prepare 8_CPh3, compared to 8_Dipp, which was generated in 24 hours at ambient temperature. Characterization of complexes 8 revealed ¹H and ³¹P{¹H} NMR spectra consistent with the major products formed in the reactions of DMAP with complexes 7_CPh3 and 7_Dipp. In order to confirm that 8_CPh3 and 8_Dipp were in fact stable products, rather than long lived intermediates prone to decomposition, a sample of L^iPrLu(NHDipp)₂ was exposed to 2 equivalents of DMAP at 60 °C for 48 hours, during which no reaction was observed.

In 2004 Piers reported the thermal decomposition of the scandium alkylamido complex LSc(NH^tBu)(CH₃) (L = DippNC(CH₃)CHC(CH₃)NDipp).³⁰ Upon being heated to 100 °C LSc(NH^tBu)(CH₃) underwent a disproportionation reaction yielding half an equivalent of both LSc(NH^tBu)₂ and LSc(CH₃)₂, the latter of which decomposed upon formation. A Lewis base induced disproportionation route is proposed as a logical pathway for our systems, with the decomposition of L^iPrLu(CH₂SiMe₃)₂ being supported by the elimination of SiMe₄. Furthermore, heating (60 °C) a pure sample of L^iPrLu(CH₂SiMe₃)₂ (6) and DMAP resulted in decomposition and production of myriad products in the ³¹P{¹H} NMR spectrum which closely matched the product distribution produced when DMAP was added to 7_CPh3 or 7_Dipp (Scheme 1).

Upon comparison of complexes 7_CPh3 and 7_Dipp with L₁Sc(NHDipp)(CH₃), the alkylamido precursor to Chen's imido 1 (vide supra), few substantial structural differences are immediately apparent. The planar arrangement of the three L₁ nitrogen donors matches that observed in complexes 7, as does the similar N_amido–RE–C_alkyl bond angle of 103.7(2)° (cf. 105.3(1)° and 110.5(2)° in 7_CPh3 and 7_Dipp, respectively).⁸ Likewise, both 7_Dipp and L₁Sc(NHDipp)(CH₃) bear a sterically demanding Dipp group on the amido nitrogen. While the geometries at the metal centres are similar, it should be noted that Chen's complex features the smaller rare earth element, scandium. In addition, the metal in L₁Sc(NHDipp)(CH₃) bears a –CH₃, rather than the larger –CH₂SiMe₃.

Although Anwander's Tp^tBu,Me-ligated system supports imidos of different rare earth metals, they too were generated by the elimination of methane. Unfortunately, the substantial differences between the facially capping Tp^tBu,Me ligand and our pincer precludes direct structural comparison.¹⁵ With that said, it is worth noting that while Anwander's complexes contain the less sterically demanding 2,6-(CH₃)₂C₆H₃ or 3,5-(CF₃)₂C₆H₃ groups on the imido nitrogen, aromatic groups were still required.

Perhaps the best comparison to our alkylamido complexes is the phosphazene-based L₂Sc(CH₂SiMe₃)(NHDipp) (L₂ = N(Ph₂P = NPh)₂), reported by Cui in 2013 (Scheme 2).¹⁴ In that case, ligand phosphinimine donors, a metal–CH₂SiMe₃ moiety and an NDipp group were common with 7_Dipp. Though the solid-state structure was not obtained for alkylamido L₂Sc(CH₂SiMe₃)(NHDipp), several striking reactivity patterns were observed. Like complexes 7_CPh3 and 7_Dipp, heat led to mixtures of the corresponding bisamido and dialkyl species, the latter of which decomposed with concomitant generation of the proteo ligand HL₂. Nonetheless, addition of DMAP at ambient temperature afforded the terminal scandium imido L₂Sc = NDipp, 9, though two equivalents of DMAP were deemed necessary. Of particular interest is the fact that when Cui and co-workers attempted the analogous chemistry with lutetium and yttrium alkylamido complexes, no conversion to the corresponding imido species was observed, even when excess DMAP was added.¹⁴ Instead, bisamide species, as well as a mixture of intractable products, formed.

Scheme 2 Terminal imido scandium complex 9 reported by Cui and coworkers.

X-ray diffraction analysis of 8_CPh3 and 8_Dipp

X-ray quality crystals of independently synthesized 8_CPh3 and 8_Dipp were grown at ambient temperature from saturated warm (50 °C) heptane solutions. Like complexes 7_CPh3 and 7_Dipp, the geometry at lutetium in complex 8_CPh3 (Fig. 5) is best described as distorted square pyramidal (δ⁵ = 0.19). The square pyramid in complex 8_Dipp (Fig. 6), however, is considerably more distorted (δ⁵ = 0.38). This discrepancy in geometry, and the need for heat when preparing 8_CPh3, is attributed to the large steric influence of the NHCPh₃ ligand. Complex 8_CPh3 possesses a Lu1–N5 bond length of 2.164(3) Å, which is amongst the longest reported for a Lu–NHR species (when R ≠ aromatic group).³¹ In line with what was observed for complexes 7_CPh3 and 7_Dipp, the average Lu–NHDipp bond length in complex 8_Dipp (2.191(2) Å) is longer than that in 8_CPh3 (2.159(2) Å), despite the steric pressure of the CPh₃ substituent (Table 1).

Fig. 5 Solid-state structure for complex 8_CPh3 with thermal ellipsoids represented at 50% probability.Hydrogen atoms and Pipp groups (except N–C_ipso carbon atoms) omitted for clarity. Selected bond lengths (Å) and angles (°) for 8_CPh3: Lu1–N1 2.327(2), Lu1–N2 2.375(2), Lu1–N4 2.153(2), Lu1–N5 2.164(3), P1–N2 1.617(3), N5–Lu1–N4 103.58(9), N3–Lu1–N2 143.59(8), N1–Lu1–N4 132.20(8), N1–Lu1–N5 124.01(9).

Fig. 6 Solid-state structure for complex 8_Dipp with thermal ellipsoids represented at 50% probability. Hydrogen atoms and Pipp groups (except N–C_ipso carbon atoms) omitted for clarity. Selected bond lengths (Å) and angles (°) for 8_Dipp: Lu1–N1 2.280(2), Lu1–N2 2.327(2), Lu1–N4 2.176(2), Lu1–N5 2.206(2), P1–N2 1.615(2), N2–Lu1–N3 143.76(7), N1–Lu1–N4 119.68(8), N1–Lu1–N5 119.43(7), N4–Lu1–N5 120.78(8).

Incorporation of internal Lewis bases

In 2011 Chen demonstrated that an internal Lewis base could stabilize a terminal scandium imido (vide supra, L₂Sc(NDipp), 2).¹⁰ Taking inspiration from those findings, we sought to incorporate internal Lewis bases into our pincer scaffold. Addition of three equivalents of NaN₃ to 4,6-dimethyl-2-methylsulfonylpyrimidine in DMF at 100 °C for 3 hours proved an efficient method for producing 2-azido-4,6-dimethylpyrimidine (10a, Scheme 3). The valence tautomerization observed in the ¹H NMR spectrum between compound 10a and the dominant tetrazole isomer, 5,7-dimethylpyrazolo[1,5a]pyrimidine (10b) is consistent with previous reports of this molecule.^32,33

Scheme 3 Synthesis of complex 12.

Formation of HL^Pm (11) was accomplished by reaction of two equivalents of 10b with 2,5-bis(diisopropylphosphino)-N–H-pyrrole in toluene (Scheme 3). Likely as a consequence of the equilibrium between 10a and 10b, this reaction required additional time compared to the formation of HL^iPr (18 hours vs. 1 hour). Removal of volatiles in vacuo afforded compound 11 as a dark beige powder. The ³¹P{¹H} NMR spectrum of the solid revealed one new resonance (δ 29.6). It is worth noting that the methyl substituents of the pyrimidine functionality appear as a singlet in the ¹H NMR spectrum (δ 2.35), indicating that free rotation around the N^P=N–C^Pm bond occurs on the NMR timescale. A downfield N–H signal (δ 13.2) in the ¹H NMR spectrum integrated for one proton and gave rise to no crosspeaks in ¹H–¹H COSY and ¹H–¹³C{¹H} HSQC experiments. This resonance appears substantially further downfield than previously reported N–H chemical shifts for our related protio ligands (δ 10.40–12.53),^{21,22,34–39} possibly due to hydrogen bonding between N–H and the N^Pm atoms.

X-ray diffraction analysis of compound 11

A sample of compound 11 was dissolved in a warm (60 °C) 5 : 3 mixture of heptane and toluene, respectively, and allowed to cool to ambient temperature before being placed in a −35 °C freezer for 18 hours, yielding X-ray quality crystalline material. The solid-state structure of compound 11 (Fig. 7), which crystallized in the monoclinic P2₁/n space group, exhibited pyrimidines that lie out of the plane of the pyrrole ring (C1–P1–N2–C23: −57.20°, C4–P2–N3–C17: 173.89°). The phosphinimine (P1–N2 and P2–N3) distances agree well with previously reported compounds.

Fig. 7 Solid-state structure for compound 11 with thermal ellipsoids represented at 50% probability. Hydrogen atoms (except H1) omitted for clarity. selected bond lengths (Å) and angles (°) for compound 11: N1–N4: 2.859(2), P1–N1: 1.588(1), P2–N3: 1.609(1), P1–N2–C23: 129.3(1), P2–N3–C17: 119.7(1), C23–N4–N1: 109.6(1).

Rare earth complexes of ligand 11

To probe the ability of ligand 11 to support well-defined rare earth complexes, a toluene solution containing one equivalent of Lu(CH₂SiMe₃)₃THF₂ was added dropwise to a stirring solution of the compound. Analogous to the preparation of L^iPrSc(CH₂SiMe₃)₂ and L^iPrLu(CH₂SiMe₃)₂ (5 and 6, respectively), the resulting solution was allowed to stir for one hour at ambient temperature whereupon all volatiles were removed in vacuo, leaving L^PmLu(CH₂SiMe₃)₂ (12) as a slightly impure yellow powder (Scheme 3).

Evidence for the formation of complex 12 includes the emergence of a new peak in the ³¹P{¹H} NMR spectrum (δ 49.8) along with concomitant disappearance of the phosphorous resonance attributed to 11 (δ 29.6). The pyrimidine-CH₃ groups are no longer chemically equivalent (δ 2.67–1.95) in the ¹H NMR spectrum suggesting slow rotation about the N^P=N–C^ipso bond on the ¹H NMR timescale. A single set of resonances attributed to the lutetium bound alkyl substituents was observed (CH₂Si(CH₃)₃: δ 0.11, CH₂Si(CH₃)₃: δ −1.04), indicative of a C_s or C_2v symmetric product.

Unfortunately, complex 12 spontaneously decomposed in solution at ambient temperature. Production of a small quantity of SiMe₄ (δ 0.00) was observed in the ¹H NMR spectrum after 5 minutes. After 24 hours complex 12 had been 35% consumed. Correspondingly, one third of an equivalent of SiMe₄ was observed in the ¹H NMR spectrum at that point. Previously we prepared a dialkyl lutetium complex supported by a pincer ligand bearing unsubstituted pyrimidine groups at nitrogen. That species decomposed by alkyl migration from lutetium to pyrimidine, dearomatizing the ring.³⁹ A similar pathway was reported by Kiplinger in 2006.⁴⁰ However, such alkyl migration does not liberate SiMe₄, thereby ruling out that mode of decomposition for complex 12. Unfortunately, repeated attempts to grow X-ray quality crystals of L^PmLu(CH₂SiMe₃)₂, even at low temperature, were unsuccessful. Similarly, the instability of complex 12 precluded both subsequent reaction chemistry and satisfactory combustion analysis.

In an effort to ascertain whether the 4,6-dimethylpyrimidine groups in ligand 11 are likely to stabilize a lutetium imido species by coordinating to the metal, we targeted a dichloride complex expecting a more thermally robust product. To this end, ligand 11 was treated with a stoichiometric quantity of NaH, following protocols established for generating NaL^iPr.²¹ Subsequent reaction between the sodiated ligand NaL^Pm and LuCl₃(THF)₃ in THF solution afforded the dichloride complex L^PmLuCl₂ (13) as a pale yellow solid. Complex 13 displays C_2v symmetry in solution at ambient temperature, as indicated by one PCH(CH₃)₂ methine resonance in the ¹H NMR spectrum and a single ³¹P peak (δ 52.2). The pyrimidine CH₃ substituents appear as two distinct, signals each integrating to 6H (δ 2.44, 2.05), implying pyrimidine coordination to the metal centre.

X-ray quality crystals of complex 13 were grown from a saturated toluene solution cooled to −35 °C. The solid-state structure determined from X-ray diffraction experiments confirmed that both pyrimidines coordinate to lutetium. Notably, all five nitrogen donors lie approximately in the same plane (P1–N2–C17–N4 torsion angle = 174.3(2)°, P2–N3–C23–N6 torsion angle = 171.46(2)°; Fig. 8), leading to distorted pentagonal bipyramidal geometry at Lu. The two chlorides, which occupy the apical sites, give rise to a Cl–Lu–Cl angle of 162.48(3)°, which, as expected, is substantially greater than the N_amido–Lu–C_alkyl angles in the mixed alkylamido complexes 7 (105.3(1), 110.5(2)°) and the N_amido–Lu–N_amido angles in bisamides 8 (124.01(9), 120.78(8)°).

Fig. 8 Solid-state structure for compound 13 with thermal ellipsoids represented at 50% probability. Hydrogen atoms omitted for clarity. Selected bond lengths (Å) and angles (°) for compound 13: Lu1–N1: 2.324(2), Lu1–N2: 2.318(2), Lu1–N4: 2.546(2), Lu1–Cl2: 2.5808(8). N2–Lu1–N4: 54.59(6), Cl1–Lu1–Cl2: 162.48(3).

Conclusions

A series of potential precursors for terminal lutetium imido complexes, L^iPrLu(CH₂SiMe₃)(NHCPh₃) (7_CPh3) and L^iPrLu(CH₂SiMe₃)(NHDipp) (7_Dipp) were prepared. Unlike our previously reported lutetium alkylamido complex L^PhLu(CH₂SiMe₃)(NHCPh₃), which bears phenyl substituents on the ligand's phosphorus atoms,²⁰ no spontaneous C–H bond activation was observed in solution. Upon addition of DMAP to either 7_CPh3 or 7_Dipp a disproportionation reaction was induced, yielding half an equivalent of L^iPrLu(CH₂SiMe₃)₂ and the corresponding bis-amide species L^iPrLu(NHCPh₃)₂ or L^iPrLu(NHDipp)₂, respectively. The generated L^iPrLu(CH₂SiMe₃)₂ rapidly decomposed in the presence of DMAP. Given the paucity of systems similar to ours it is difficult to draw meaningful conclusions regarding what factors dictate imido formation vs. ligand redistribution, though greater success has been realized with sterically bulky aromatic substituents on the imido nitrogen and with the smaller metal scandium.

Installation of 4,6-dimethylpyrimidine groups on the iminophosphorane nitrogen atoms of the pincer framework afforded HL^Pm, (11), a ligand with the potential to stabilize rare earth imido species by coordination of the pyrimidine groups to the metal centre. While addition of Lu(CH₂SiMe₃)₃THF₂ to compound 11 generated the expected dialkyl complex L^PmLu(CH₂SiMe₃)₂ (12), rapid decomposition at ambient temperature rendered it impossible to prepare desired alkylamido complexes of the form L^PmLu(CH₂SiMe₃)(NHR). However, preliminary experiments indicate that the rare earth complex L^PmLuCl₂ is accessible via a salt metathesis route, and notably, both pyrimidine groups coordinate to lutetium in the solution and solid states. The possibility of using the amide salts ([(THF)LiNHCPh₃]₂ or [(THF)LiNHDipp]₂), to ultimately access alkylamido species that may serve as imido precursors, will be the focus of future studies.

Experimental

General procedures

Manipulation of air- and moisture-sensitive materials and reagents was carried out under an argon atmosphere using vacuum line techniques or in an MBraun glove box. Solvents used for air-sensitive materials were purified using an MBraun solvent purification system (SPS), stored in PTFE-sealed glass vessels over “titanocene” (pentane, benzene, and toluene), and freshly distilled at the time of use. Benzene-d₆ was dried over sodium benzophenone ketyl, de-gassed via three freeze–pump–thaw cycles, distilled in vacuo and stored over 4 Å molecular sieves in glass bombs under argon. Unless noted, all NMR spectra were recorded at ambient temperature with a Bruker Avance III NMR spectrometer (700.44 MHz for ¹H, 176.13 MHz for ¹³C, and 283.54 MHz for ³¹P). Chemical shifts are reported in parts per million relative to the external standards SiMe₄ (¹H, ¹³C) and 85% H₃PO₄ (³¹P); residual H-containing species in C₆D₆ (δ 7.16 (¹H), δ 128.39 (¹³C)) were used as internal references. Assignments were aided by the use of ¹³C{¹H} DEPT-135, ¹H–¹³C{¹H} HSQC, ¹H–¹³C{¹H} HMBC and ¹H–¹H COSY experiments (s = singlet, d = doublet, t = triplet, q = quartet, sp = septet, m = multiplet, br = broad, ov = overlapping signals). Elemental analyses were performed using an Elementar Vario Micro-cube instrument. The reagents Lu(CH₂SiMe₃)₃(THF)₂,⁴¹ Sc(CH₂SiMe₃)₃(THF)₂,⁴² and 2,5-[ⁱPr₂P = N(4-ⁱPrC₆H₄)]₂NH(C₄H₂)(HL^iPr),²¹ were prepared according to literature methods. The compound 2,6-diisopropylanaline was purchased from Sigma Aldrich and distilled under reduced pressure into a bomb equipped with a Kontes valve before using. Compound 10 was synthesized according to literature procedures with additional details provided in the text above.^32,33 All other reagents were purchased from Sigma Aldrich and used without further purification. Unless otherwise specified, reported yields correspond to those obtained for analytically pure samples. When additional purification was required to generate analytically pure compounds for combustion analysis, both the crude and analytical yields are included. In such cases, the reported crude yield corresponds to material that was utilized for successive synthetic steps and was >98% pure as indicated by multinuclear NMR spectroscopy. (N.B. NMR spectra displayed in the ESI† were obtained using these “crude” samples.) It should also be noted that when reported, analytical yields are generally artificially low, as excess crystals grown for X-ray diffraction experiments are used for this purpose. Those recrystallizations were not carried out under conditions that maximize yield, but rather, were optimized for the growth of X-ray quality crystals.

X-Ray crystallography

Crystals grown for X-ray diffraction analysis were coated in Paratone oil and placed on a glass slide. Close visual inspection and selection of the crystals was aided by either a standard microscope or polarizing light microscope. The desired crystal chosen for diffraction analysis was placed on a MiTeGen Dual Thickness MicroMount attached to a goniometer head. The crystal was centred on a Rigaku SuperNova diffractometer equipped with a Dectris Pilatus 3R 200K-A detector, Oxford CryoStream 800 cooling system, molybdenum (NOVA) radiation source (Kα = 0.71073 Å), and copper (MOVA) radiation source (Kα = 1.5406 Å). Experiments were performed at 100 K to reduce thermal motion of the atoms. CrysAlisPro software was utilized to determine unit cell parameters and SHELXTL software was utilized using least squares methodology for refinement.

Synthesis of L^iPrSc(CH₂SiMe₃)₂ (5)

A 100 mL round-bottomed flask was charged with a sample of HL^iPr (0.0913 g, 0.161 mmol). Toluene (10 mL) was transferred into this flask via vacuum distillation, yielding a clear, light brown solution. Sc(CH₂SiMe₃)₃(THF)₂ (0.0724 g, 0.161 mmol), was added to a 50 mL round-bottomed flask, and dissolved in toluene (10 mL). The Sc(CH₂SiMe₃)₃(THF)₂ solution was added dropwise to the stirring HL^iPr solution at ambient temperature over 3 minutes. The resulting clear, light brown solution was left to stir at ambient temperature for 1 hour upon which all volatiles were removed in vacuo. The reaction contraption was brought into an inert atmosphere glove box whereupon heptane (10 mL) was added to the light brown solid. The mixture was stirred and heated (50 °C) leaving behind a small amount of a dark, oily solid. This mixture was filtered through a bed of Celite which produced a clear, yellow solution. Removal of the heptane in vacuo liberated the desired product as a light, pale yellow solid (0.113 g, 90.0% (crude)). Facile preparation of analytically pure material was achieved through recrystallization from a minimal amount of warm (50 °C) heptane (0.0521 g, 41.4%). ¹H NMR (benzene-d₆): δ 7.50 (dd, ³J_H–H = 8.4 Hz, ⁴J_H–P = 2.1, 4H, o-Pipp–H), 7.23 (d, ³J_H–H = 8.4 Hz, 4H, m-Pipp–H), 6.51 (d, ³J_H–P = 2.1 Hz, 2H, pyrrole–H), 2.80 (sp, ³J_H–H = 7.0 Hz, 2H, Pipp–(CH(CH₃)₂)), 2.09 (m, ³J_H–H = 6.9 Hz, ²J_H–P = 2.7 Hz, 4H, P–(CH(CH₃)₂))), 1.24 (d, ³J_H–H = 7.0 Hz, 12H, Pipp–(CH(CH₃)₂)), 0.93 (dd, ³J_H–P = 32.1 Hz, ³J_H–H = 6.9 Hz, 12H, P–(CH(CH₃)(CH₃))), 0.88 (dd, ³J_H–P = 31.4 Hz, ³J_H–H = 6.9 Hz, 12H, P–(CH(CH₃)(CH₃)), 0.31 (s, 4H, Sc–CH₂SiMe₃), 0.16 (s, 18H, CH₂Si(CH₃)₃). ¹³C{¹H} NMR (benzene-d₆): δ 145.54 (s, ipso-Pipp), 144.57 (s, ipso-Pipp), 129.67 (s, o-Pipp C–H), 128.41 (d, ¹J_C–P = 24.1, ipso-pyrrole) 127.73 (s, m-Pipp C–H), 115.90 (dd, ²J_C–P = 24.1 Hz, ³J_C–P = 10.1 Hz, pyrrole C–H), 40.00 (br s, Sc–CH₂), 34.38 (s, Pipp–CH(CH₃)₂), 26.30 (d, ¹J_C–P = 54.0 Hz, P–CH(CH₃)(CH₃)), 24.79 (s, Pipp–CH(CH₃)₂), 16.60 (s, P–(CH(CH₃)(CH₃))), 16.13 (s, P–(CH(CH₃)(CH₃))), 4.65 (s, SiMe₃). ³¹P{¹H} NMR (benzene-d₆): δ 48.0. Anal calcd (%) for C₄₂H₇₄ScN₃P₂Si₂ C, 64.33; H, 9.51; N, 5.36. Found: C, 64.00, H, 8.79; N, 5.68.

Synthesis of L^iPrLu(CH₂SiMe₃)₂ (6)

A 100 mL round-bottomed flask was charged with a sample of HL^iPr (0.1729 g, 0.3040 mmol). Toluene (10 mL) was transferred to this flask via vacuum distillation, yielding a clear, light brown solution. Lu(CH₂SiMe₃)₃(THF)₂ (0.1766 g, 0.3040 mmol), was added to a 50 mL round-bottomed flask, and was dissolved in toluene (10 mL). The Lu(CH₂SiMe₃)₃(THF)₂ solution was added dropwise to the stirring solution at ambient temperature over 3 minutes. The resulting clear, light brown solution was left to stir at ambient temperature for 1 hour upon which all volatiles were removed in vacuo. The reaction contraption was brought into an inert atmosphere glove box whereupon heptane (10 mL) was added to the light brown solid. The mixture was stirred and heated (50 °C) leaving behind a small amount of a dark, oily solid. This mixture was filtered through a bed of Celite which produced a clear, yellow solution. Removal of the heptane in vacuo liberated the desired product as a pale yellow solid (0.2557 g, 92.01%). Analytically pure material was obtained by recrystallization of the crude sample from a minimal amount of warm (50 °C) heptane (0.1218 g, 43.83%). ¹H NMR (benzene-d₆): δ 7.43 (dd, ³J_H–H = 8.1 Hz, ⁴J_H–P = 1.8 Hz, 4H, o-Pipp–H), 7.22 (d, ³J_H–H = 8.1 Hz, 4H, m-Pipp–H), 6.50 (d, ³J_H–P = 2.1 Hz, 2H, pyrrole–H), 2.79 (sp, ³J_H–H = 6.6 Hz, 2H, Pipp–(CH(CH₃)₂)), 2.04 (m, ³J_H–H = 7.0 Hz, ¹J_H–P = 2.4 Hz, 4H, P–(CH(CH₃)₂)), 1.22 (d, ³J_H–H = 6.6 Hz, 12H, Pipp–(CH(CH₃)₂)), 0.96 (dd, ³J_H–H = 7.0 Hz, ²J_H–P = 16.2 Hz, 12H, P–(CH(CH₃)(CH₃))), 0.85 (dd, ³J_H–H = 7.0 Hz, ²J_H–P = 15.5 Hz, 12H, P–(CH(CH₃)(CH₃))), 0.21 (s, 18H, SiMe₃), −0.42 (s, 4H, Lu–CH₂SiMe₃). ¹³C{¹H} NMR (benzene-d₆): δ 144.32 (dd, ¹J_C–H = 98.6 Hz, ⁴J_C–H = 24.3 Hz, ipso-pyrrole), 129.15 (d, ³J_C–P = 13.2 Hz, o-Pipp C–H), 129.11 (d, ²J_C–P = 26.4 Hz, ipso-Pipp), 127.89 (s, m-Pipp C–H), 127.40 (s, ipso-Pipp), 116.51 (dd, ²J_C–P = 98.6 Hz, ³J_C–P = 40.2 Hz, pyrrole C–H), 41.30 (s, Lu–CH₂), 34.32 (s, Pipp–(CH(CH₃)₂)), 26.42 (d, ¹J_C–P = 54.0 Hz, P–(CH(CH₃)₂)), 24.76 (s, Pipp–(CH(CH₃)₂)), 16.51 (s, P–(CH(CH₃)(CH₃))), 15.99 (s, P–(CH(CH₃)(CH₃))), 5.19 (s, SiMe₃) ³¹P{¹H} NMR (benzene-d₆): δ 49.3. Anal. calcd (%) for C₄₂H₇₄LuN₃P₂Si₂ C, 55.18; H, 8.16; N, 4.60. Found: C, 55.01, H, 8.26; N, 4.97.

Synthesis of L^iPrLu(CH₂SiMe₃)(NHCPh₃) (7_CPh3)

A sample of 6 (0.1033 g, 0.1130 mmol) was weighed into a 20 mL scintillation vial containing a stir bar. The powder was dissolved in 6 mL of toluene, yielding a clear yellow solution. One equivalent of NH₂CPh₃ (0.0286 g, 0.110 mmol) was added into a 5 mL disposable vial and was dissolved using 2 mL of toluene. The amine solution was added dropwise over 3 minutes to the stirring yellow solution. After stirring at ambient temperature for 1 hour the reaction mixture was filtered through a bed of Celite, eliminating insoluble contaminants. All volatiles were removed in vacuo yielding the desired complex as a pale, off-white solid powder (0.103 g, 86.4% yield). X-ray quality crystals were grown at ambient temperature from dissolving the crude powder in a minimal amount of warm heptane (50 °C) (0.0611 g, 51.04% yield). ¹H NMR (benzene-d₆): δ 7.53 (m, ³J_H–H = 3.9 Hz, 6H, o-Ph₃), 7.08 (ov m, ³J_H–H = 3.9 Hz, 13H, m/p-Ph₃, Pipp–H), 6.96 (d, ³J_H–H = 6.3 Hz, 4H, Pipp–H), 6.54 (d, ³J_H–P = 1.5 Hz, 2H, pyrrole–H), 2.85, (sp, ³J_H–H = 6.6 Hz, 2H, Pipp–(CH(CH₃)₂)), 2.43 (s, 1H, N–H), 2.16 (dsp, ¹J_H–P = 9.6 Hz, ³J_H–H = 6.9 Hz, 2H, P–(CH(CH₃)₂)), 1.75 (dsp, ¹J_H–P = 9.6 Hz, ³J_H–H = 6.9 Hz, 2H, P–(CH(CH₃)₂)), 1.32 (d, ³J_H–H = 6.6 Hz, 6H, Pipp–(CH(CH₃)(CH₃))), 1.30 (d, ³J_H–H = 6.6 Hz, 6H, Pipp–(CH(CH₃)(CH₃))), 0.98 (dd, ²J_P–H = 7.2 Hz, ³J_H–H = 6.9 Hz, 6H, P–(CH(CH₃)(CH₃))), 0.85 (dd, ³J_P–H = 7.2 Hz, ³J_H–H = 6.9 Hz, 6H, P–(CH(CH₃)(CH₃))), 0.77 (d, ²J_P–H = 6.9 Hz, ³J_H–H = 6.9 Hz, 6H, P–(CH(CH₃)(CH₃))), 0.61 (dd, ³J_P–H = 7.2 Hz, ³J_H–H = 6.9 Hz, 6H, P–(CH(CH₃)(CH₃))), 0.05 (s, 9H, SiMe₃), −0.37 (s, 2H, Lu–CH₂SiMe₃). ¹³C{¹H} NMR (benzene-d₆): δ 154.67 (s, ipso-CPh₃ C), 144.38 (d, ¹J_C–P = 88.5 Hz, ipso-pyrrole), 129.60 (s, ipso-Pipp) 129.51 (s, Pipp C–H), 128.68 (s, Pipp C–H) 128.24 (ipso-Pipp) 127.75 (s, CPh₃ C–H), 127.69 (s, CPh₃ C–H), 125.91 (s, CPh₃ C–H), 116.63 (dd, ²J_P–C = 24.5 Hz, ³J_P–C = 9.6 Hz, pyrrole C–H), 75.74 (s, CPh₃), 34.39 (s, Lu–CH₂), 32.60 (s, Pipp–(CH(CH₃)₂)), 26.83 (d, ¹J_P–C = 54.5 Hz, P–(CH(CH₃)₂)), 26.33 (d, ¹J_P–C = 54.5 Hz, P–(CH(CH₃)₂)), 25.02 (s, Pipp–(CH(CH₃)(CH₃))), 24.89 (s, Pipp–(CH(CH₃)(CH₃))), 16.50 (ov s, P–(CH(CH₃)(CH₃))) and P–(CH(CH₃)(CH₃)), 16.49 (s, P–(CH(CH₃)(CH₃))), 16.25 (s, P–(CH(CH₃)(CH₃))), 4.61 (s, SiMe₃). ³¹P{¹H} NMR (benzene-d₆): δ 47.8. Anal. calcd (%) for C₅₉H₇₉LuN₄P₂Si C, 63.08; H, 7.34; N, 5.16. Found: C, 62.86, H, 7.12; N, 5.25.

L^iPrLu(CH₂SiMe₃)(NHDipp) (7_Dipp)

A sample of 6 (0.1193 g, 0.1305 mmol) was weighed into a 20 mL scintillation vial containing a stir bar. The powder was dissolved in 6 mL of toluene, yielding a clear yellow solution. NH₂Dipp (0.0241 g, 0.136 mmol) was weighed into a syringe and added to a vial containing 5 mL of toluene. This amine solution was then added dropwise to the stirring solution of L^iPrLu(CH₂SiMe₃)₂ over 5 minutes. The resulting golden yellow solution was left to stir at ambient temperature for 18 hours whereupon all volatiles were removed in vacuo. The resulting pale yellow solid was reconstituted in a minimal amount of a warm (50 °C) 1 : 1 heptane : benzene mixture and recrystallized at −35 °C (0.0722 g, 55.1% yield). ¹H NMR (benzene-d₆): δ 7.33 (d, ³J_H–H = 7.5 Hz, 2H, m-Dipp C–H), 7.22 (dd, ³J_H–H = 8.1 Hz, ⁴J_H–P = 1.8 Hz, 4H, o-Pipp C–H), 7.01 (d, ³J_H–H = 8.1 Hz, 4H, m-Pipp C–H), 6.93 (t, ³J_H–H = 7.5 Hz, 1H, p-Dipp C–H), 6.52 (d, ³J_H–P = 2.1 Hz, 2H, pyrrole C–H), 4.14 (s, 1H, N–H), 3.39 (sp, ³J_H–H = 6.6 Hz, 2H, Dipp–(CH(CH₃)₂)₂), 2.73 (sp, ³J_H–H = 6.9 Hz, 2H, Pipp–(CH(CH₃)₂)), 2.26 (m, ³J_H–H = 6.9 Hz, ²J_H–P = 6.6 Hz, 2H, P–(CH(CH₃)₂)), 1.93 (m, ³J_H–H = 6.9 Hz, ²J_H–P = 6.6 Hz, 2H, P–(CH(CH₃)₂)), 1.35 (d, ³J_H–H = 6.6 Hz, 12H, Dipp–(CH(CH₃)₂)₂), 1.20 (ov d, ³J_H–H = 6.9 Hz, 6H, Pipp–(CH(CH₃)(CH₃))), 1.19 (ov d, ³J_H–H = 6.9 Hz, 6H, Pipp–(CH(CH₃)(CH₃))), 1.07 (dd, ³J_H–H = 6.9 Hz, ³J_H–P = 6.9 Hz, 6H, P–(CH(CH₃)(CH₃))), 0.94 (dd, ³J_H–P = 7.2 Hz, ³J_H–H = 6.9 Hz, 6H, P–(CH(CH₃)(CH₃))), 0.80 ((dd, ³J_H–H = 6.9 Hz, ³J_H–P = 6.9 Hz, 6H, P–(CH(CH₃)(CH₃)))), 0.74 (dd, ³J_H–P = 7.5 Hz, ³J_H–H = 6.9 Hz, 6H, P–(CH(CH₃)(CH₃))), 0.03 (s, 9H, SiMe₃), −0.44 (s, 2H, Lu–CH₂SiMe₃). ¹³C{¹H} NMR (benzene-d₆): δ 154.70 (s, o-ipso-Dipp), 144.56 (s, ipso-Pipp), 144.08 (s, ipso-Pipp), 133.68 (s, ipso-Dipp), 128.33 (dd, ¹J_C–P = 135.5 Hz, ³J_C–P = 14.3 Hz, ipso-pyrrole), 128.28 (s, Pipp C–H), 127.82 (s, Pipp C–H), 123.08 (s, m-Dipp C–H), 117.20, (dd, ²J_C–P = 15.6 Hz, ³J_C–P = 10.1 Hz, pyrrole C–H), 114.56 (s, p-Dipp C–H), 36.50 (s, Lu–CH₂), 34.27 (s, Pipp–(CH(CH₃)₂)), 29.29 (s, Dipp–(CH(CH₃)₂)), 27.58 (s, P–(CH(CH₃)₂)), 26.85 (s, P–(CH(CH₃)₂)), 24.77 (s, Dipp–(CH(CH₃)₂)), 24.70 (s, Pipp–(CH(CH₃)(CH₃))), 24.58 (s, Pipp–(CH(CH₃)(CH₃))), 16.84 (s, P–(CH(CH₃)(CH₃))), 16.59 (ov s, P–(CH(CH₃)(CH₃))) and P–(CH(CH₃)(CH₃)), 16.38 (s, P–(CH(CH₃)(CH₃))), 4.44 (s, SiMe₃). ³¹P{¹H} NMR (benzene-d₆): δ 48.7. Anal. calcd (%) for C₅₀H₈₁LuN₄P₂Si C, 59.86; H, 8.14; N, 5.58. Found: C, 59.57, H, 8.11; N, 5.68.

Synthesis of L^iPrLu(NHCPh₃)₂ (8_CPh3)

A 20 mL scintillation vial was charged with L^iPrLu(CH₂SiMe₃)₂ (0.0513 g, 0.0561 mmol) and a small stir bar. A second vial was charged with NH₂CPh₃ (0.0291 g, 0.112 mmol). Toluene (7 mL) was added to each flask to dissolve all materials. The NH₂CPh₃ solution was transferred into the L^iPrLu(CH₂SiMe₃)₂ solution dropwise via Pasteur pipette. The resulting light brown solution was allowed to stir at 70 °C for 69 hours whereupon all volatiles were removed in vacuo. The desired product remained as a light brown solid (0.0544 g, 88.7%). X-ray quality crystals were grown by dissolving the mixture in a minimal amount of a warm (50 °C) 1 : 1 mixture of toluene and heptane. The solution was allowed to cool to ambient temperature over 2 hours and was placed in a −35 °C freezer for 16 hours. The resulting crystals were isolated by filtration, washed with cold pentane and dried for 8 hours under vacuum (0.0197 g, 32.1%). ¹H NMR (benzene-d₆): δ 7.37 (d, ³J_H–H = 7.7 Hz, 12H, o-Ph₃), 7.08 (ov m, 18H, m- and p-Ph₃), 6.85 (d, ³J_H–H = 8.4 Hz, 4H, o-Pipp C–H), 6.50 (dd, ³J_H–H = 8.4 Hz, ⁴J_H–P = 1.4 Hz, 4H, m-Pipp C–H), 6.55 (d, ³J_H–P = 2.1 Hz, 2H, pyrrole C–H), 2.83 (sp, ³J_H–H = 6.6 Hz, 2H, Pipp–CH(CH₃)₂), 2.26 (s, 2H, N–H), 1.96 (dsp ²J_H–P = 6.6 Hz, ³J_H–H = 6.6 Hz, 4H, P–CH(CH₃)₂) 1.29 (d, ³J_H–H = 6.8 Hz, 12H, Pipp–CH(CH₃)₂), 0.81 (dd, ³J_H–P = 16.1 Hz, ³J_H–H = 6.6 Hz, 12H, P–CH(CH₃)(CH₃)), 0.79 (dd, ³J_H–P = 16.1 Hz, ³J_H–H = 6.6 Hz, 12H, P–CH(CH₃)(CH₃)). ¹³C{¹H} NMR (benzene-d₆): δ 154.38 (s, CPh₃ipso-C), 146.04 (s, ipso-Pipp), 142.74 (s, ipso-Pipp), 129.98 (s, o-CPh₃ C–H), 129.16 (d, ³J_C–P = 7.1 Hz, o-Pipp C–H), 128.39 (d, ¹J_C–P = 125.0 Hz, ipso-pyrrole), 127.82 (s, m-CPh₃ C–H), 127.36 (s, m-Pipp C–H), 125.60 (s, p-CPh₃ C–H), 116.88 (dd, ²J_C–P = 22.9 Hz, ³J_C–P = 10.6 Hz, pyrrole C–H), 75.41 (s, CPh₃), 34.22 (s, Pipp–CH(CH₃)₂), 27.03 (d, ¹J_C–P = 54.6 Hz, P–CH(CH₃)₂), 24.92 (s, Pipp–CH(CH₃)₂), 16.65 (s, P–CH(CH₃)(CH₃)), 16.63 (s, P–CH(CH₃)(CH₃)). ³¹P{¹H} NMR (benzene-d₆): 45.3. Anal. calcd (%) for C₇₂H₈₄LuN₅P₂ C, 68.83; H, 6.74; N, 5.57. Found: C, 68.60, H, 7.66; N, 6.31.

Synthesis of L^iPrLu(NHDipp)₂ (8_Dipp)

A round-bottomed flask was charged with L^iPrLu(CH₂SiMe₃)₂ (0.1835 g, 0.2007 mmol) and a small stir bar. A second round-bottomed flask was charged with NH₂Dipp (0.0712 g, 0.402 mmol). Toluene (50 mL) was transferred into each flask at −78 °C. The flasks were warmed to ambient temperature whereupon the NH₂Dipp solution was transferred into the L^iPrLu(CH₂SiMe₃)₂ solution dropwise via syringe. The resulting light brown solution was allowed to stir at ambient temperature for 24 hours whereupon all volatiles were removed in vacuo. The desired product remained as a light brown solid (0.184 g, 84.1%). X-ray quality crystals were grown by dissolving the solid in a minimal amount of a warm (50 °C) 1 : 1 mixture of toluene and heptane. The solution was allowed to cool to ambient temperature over 2 hours and was placed in a −35 °C freezer for 16 hours. The resulting crystals were isolated by filtration, washed with cold pentane and dried for 8 hours under vacuum (0.0927 g, 42.3%). ¹H NMR (benzene-d₆): δ 7.24 (d, ³J_H–H = 7.5 Hz, 4H, m-Dipp C–H), 7.06 (dd, ³J_H–H = 8.4 Hz, ⁴J_H–P = 2.1 Hz 4H, o-Pipp C–H), 6.97–6.88 (ov m, 6H, m-Pipp C–H and p-Dipp C–H), 6.56 (d, 2H, ³J_H–P = 2.4 Hz, pyrrole C–H), 4.38 (s, 2H, N–H), 3.20 (sp, ³J_H–H = 6.9 Hz, 4H, Dipp–CH(CH₃)₂), 2.68 (sp, 2H, ³J_H–H = 7.2 Hz, Pipp–CH(CH₃)₂), 2.09 (dd, ³J_H–H = 7.2 Hz, ²J_H–P = 1.8 Hz, 4H, P–CH(CH₃)₂), 1.28 (d, ³J_H–H = 6.9 Hz, 24H, Dipp–CH(CH₃)₂), 1.15 (d, ³J_H–H = 7.2 Hz, 12H, Pipp–CH(CH₃)₂), 0.91 (dd, ³J_H–P = 15.7 Hz, ³J_H–H = 7.2 Hz, 12H, P–CH(CH₃)(CH₃)), 0.79 (dd, ³J_H–P = 15.7 Hz, ³J_H–H = 7.2 Hz, 12H, P–CH(CH₃)(CH₃)). ¹³C{¹H} NMR (benzene-d₆): δ 154.42 (s, ipso-Dipp), 144.04 (ov s, ipso-Pipp and p-ipso-Pipp), 134.04 (s, o-ispo-Dipp), 128.53 (dd, ¹J_C–P = 134.7 Hz, ³J_C–P = 13.7 Hz, ipso-pyrrole), 128.45 (d, ³J_C–P = 6.4 Hz, o-Pipp C–H), 127.65 (s, m-Pipp C–H), 122.78 (s, m-Dipp C–H), 117.95 (dd, ²J_C–P = 17.5 ³J_C–P = 9.9 Hz, pyrrole C–H), 114.81 (s, p-Dipp C–H), 34.17 (s, Pipp–CH(CH₃)₂), 29.86 (s, Dipp–CH(CH₃)₂), 27.40 (d, ¹J_C–P = 53.6 Hz, P–CH(CH₃)₂), 24.63 (s, Pipp–CH(CH₃)₂), 24.17 (s, Dipp–CH(CH₃)₂), 16.98 (s, P–CH(CH₃)(CH₃)), 16.54 (s, P–CH(CH₃)(CH₃)). ³¹P{¹H} NMR (benzene-d₆): δ 46.4. Anal. calcd (%) for C₅₈H₈₈LuN₅P₂ C, 63.78; H, 8.12; N, 6.41. Found: C, 62.82, H, 7.69; N, 7.23.

Synthesis of HL^Pm (11)

A sample of 2,5-bis(diisopropylphosphino)-N–H-pyrrole (0.1090 g, 0.3641 mmol) was weighed into a round-bottomed flask and attached to a vacuum line. Two equivalents of 2-azido-4,6-dimethylpyrimidine (10) (0.1050 g, 0.7040 mmol) was weighed into a second round-bottomed flask and attached to a vacuum line. Toluene (25 mL) was transferred into each flask at −78 °C. Both solutions were allowed to warm to ambient temperature whereupon a cannula was used to add the solution of compound 10 to the 2,5-bis(diisopropylphosphino)-N–H-pyrrole solution. Bubbles formed shortly after completion of the addition. The solution was left to stir under a constant flow of argon for 18 hours whereupon all volatiles were removed in vacuo leaving behind the desired product as a brown solid (0.173 g, 88.3%). X-ray quality crystals were obtained by dissolving the crude product in a minimal amount of toluene in a 20 mL scintillation vial at ambient temperature. Pentane (4 mL) was added to a small, 5 mL vial which was placed inside the vial containing the saturated toluene solution of HL^Pm. The entire system was sealed and placed in a −35 °C freezer for 16 hours. The mother-liquor was removed via Pasteur pipette and cold (−35 °C) pentane (10 mL) was added to the vial to wash the crystals. The pentane was removed via Pasteur pipette and the crystals were dried under vacuum for 2 hours. ¹H NMR (benzene-d₆): δ 13.21 (s, 1H, N–H), 6.35 (s, 2H, pyrrole C–H), 6.08 (s, 2H, pyrimidine C–H), 2.39 (spd, 4H, ³J_H–H = 7.0 Hz, ²J_H–P = 2.6 Hz, P–CH(CH₃)₂), 2.39 (s, 12H, pyrimidine CH₃), 1.10 (dd, ³J_H–P = 16.1 Hz, ³J_H–H = 7.0 Hz, P–(CH(CH₃)(CH₃))), 1.00 (dd, ³J_H–P = 16.1 Hz, ³J_H–H = 7.0 Hz, P–(CH(CH₃)(CH₃))). ¹³C{¹H} NMR (benzene-d₆): δ 168.56 (s, m-ipso-pyrimidine), 167.51 (d, ²J_C–P = 1.5 Hz, ipso-pyrimidine), 124.8 (d, ¹J_C–P = 41.5 Hz, ipso-pyrrole), 116.76 (br s, pyrrole C–H), 109.15 (s, pyrimidine C–H), 26.52 (d, ¹J_C–P = 65.9 Hz, P–CH(CH₃)₂), 24.58 (s, pyrimidine CH₃), 17.22 (s, P–(CH(CH₃)(CH₃))), 16.85 (s P–(CH(CH₃)(CH₃))). ³¹P{¹H} NMR (benzene-d₆): δ 27.7. Anal. calcd (%) for C₂₈H₄₄N₇P₂ C, 62.09; H, 8.37; N, 18.10. Found: C, 61.36, H, 8.27; N, 17.88.

Synthesis of L^PmLu(CH₂SiMe₃)₂ (12)

A sample of HL^Pm (11) (0.0580 g, 0.0999 mmol) was weighed into a 20 mL scintillation vial. A disposable shell vial was charged with Lu(CH₂SiMe₃)₃THF₂ (0.0552 g, 0.102 mmol). Toluene (3 mL) was added to each vial to dissolve all material. The Lu(CH₂SiMe₃)₃THF₂ solution was added to the stirring solution of compound 11 dropwise over 1 minute via Pasteur pipette. The clear, pale yellow solution was left to stir at ambient temperature for 1 hour whereupon all volatiles were remove in vacuo. Pentane (5 mL) was added to the remaining waxy solid and the mixture was stirred for 15 minutes. The solvent was removed in vacuo affording a pale yellow solid (0.0737 g). As complex 12 decomposes slowly in solution, the solid material contained a mixture of predominantly 12 and several intractable impurities. ¹H NMR (toluene-d₈): δ 6.62 (d, ³J_H–P = 3.5 Hz, 2H, pyrrole C–H), 6.00 (s, 2H, pyrimidine C–H), 2.62–2.56 (ov m, 10H, pyrimidine CH₃ and P–CH(CH₃)₂), 2.03 (s, 6H, pyrimidine CH₃₎, 1.22 (dd, ³J_H–P = 16.9 Hz, ³J_H–H = 7.1 Hz, 12H, P–(CH(CH₃)(CH₃))), 1.02 (dd, ³J_H–P = 17.2 Hz, ³J_H–H = 7.1 Hz, 12H, P–(CH(CH₃)(CH₃))), 0.02 (s, 18 h, SiMe₃), −1.15 (s, 4H, Lu–CH₂SiMe₃). ¹³C{¹H} NMR (toluene-d₈): δ 167.51 (s, ipso-pyrimidine), 165.56 (s, ipso-pyrimidine), 164.13 (s, ipso-pyrimidine), 126.92 (d, ¹J_C–P = 15.2 Hz, ipso-pyrrole), 117.12 (dd, ²J_C–P = 26.7 Hz, ³J_C–P = 11.2 Hz, pyrrole C–H), 110.98 (s, pyrimidine C–H), 26.37 (s, Lu–CH₂), 25.85 (d, ¹J_C–P = 53.8 Hz, P–CH(CH₃)₂), 23.87 (br s, endo/exo pyrimidine CH₃), 17.00 (s, P–(CH(CH₃)(CH₃))), 16.08 (s, P–(CH(CH₃)(CH₃))), 5.08 (s, SiMe₃). ³¹P{¹H} NMR (benzene-d₆): δ 42.4.

Synthesis of L^PmLuCl₂ (13)

A sample of NaL^Pm (0.6903 g, 1.225 mmol), synthesized according to the procedure disclosed for preparing NaL^iPr,²¹ was added to a 100 mL two-necked round bottomed flask and attached to a swivel frit apparatus. LuCl₃(THF)₃ (0.3486 g, 1.239 mmol) was added to a separate 100 mL round bottom flask. THF (50 mL) was added into each flask via vacuum distillation at −78 °C. Each flask was allowed to warm to ambient temperature while stirring. The LuCl₃(THF)₃ solution was then added dropwise via syringe to the stirring NaL^Pm solution over 5 minutes at ambient temperature. The reaction mixture was left to stir at ambient temperature for 3 hours whereupon all volatiles were removed in vacuo leaving behind a pale yellow solid. Toluene (50 mL) was transferred into the round bottomed flask via vacuum distillation at −78 °C. The solution was allowed to warm to ambient temperature, at which point the NaCl by-product was removed by filtration, affording a clear yellow solution. All volatiles were removed in vacuo giving the desired product as a pale yellow solid (0.8207 g, 85.2%). ¹H NMR (benzene-d₆): δ 6.68 (d, ³J_H–P = 2.4 Hz, 2H, pyrrole C–H), 5.90 (s, 2H, aromatic pyrimidine C–H), 2.60 (dsp, ³J_H–H = 7.2 Hz, ²J_H–P = 2.4 Hz, 4H, P–CH(CH₃)₂), 2.44 (s, 6H, pyrimidine CH₃), 2.04 (s, 6H, pyrimidine CH₃), 1.29 (dd, ³J_H–P = 7.2 Hz, ³J_H–P = 17.0 Hz, 12H, P– CH(CH₃)(CH₃)), 1.07 (dd, ³J_H–P = 7.2 Hz, ³J_H–P = 17.4 Hz, 12H, P–CH(CH₃)(CH₃)). ¹³C{¹H} NMR (benzene-d₆): δ 168.70 (s, pyrimidine ipso-C), 166.05 (d, ²J_C–P = 5.28 Hz, pyrimidine ipso-C), 164.94 (s, pyrimidine ipso-C), 128.42 (m, pyrrole ipso-C), 117.75 (dd, ²J_C–P = 26.0 Hz, ³J_C–P = 10.5 Hz, pyrrole C–H), 111.64 (s, aromatic pyrimidine C–H), 26.16 (d, ¹J_C–P = 54.4 Hz, P–CH(CH₃)₂), 24.29 (ov s, pyrimidine CH₃), 17.09 (s, P–CH(CH₃)(CH₃)), 16.23 (s, P–CH(CH₃)(CH₃)). ³¹P{¹H} NMR (benzene-d₆): δ 52.2. Despite exhaustive efforts, satisfactory combustion data was not obtained.

Author contributions

J. P. K. contributed to investigation, methodology, data curation, formal analysis, writing – original draft and writing – reviewing and editing. S.-J. H. contributed to data curation, formal analysis, writing – reviewing and editing. P. G. H. contributed to conceptualization, funding acquisition, project administration, formal analysis, supervision, methodology, writing – original draft and writing – reviewing and editing.

Data availability

Original data can be obtained from the authors on request. Deposition Numbers 2417587 (for 5), 2417588 (for 6), 2417589 (for 7_CPh3), 2417590 (for 7_Dipp), 2417591 (for 8_CPh3), 2417592 (for 8_Dipp), 2423088 (for 11) and 2429452 (for 13)† contain the supplementary crystallographic data for this paper. These data are provided free of charge by the joint Cambridge Crystallographic Data Centre (CCDC) and Fachinformationszentrum Karlsruche Access Structures service.

Conflicts of interest

The authors have no conflicts to declare.

Acknowledgements

This research was financially supported by a Discovery Grant from the Natural Sciences and Engineering Research Council (NSERC) of Canada. The authors thank Mr Tony Montina and Mr Michael Opyr for expert technical assistance with NMR experiments. Prof. René T. Boeré is gratefully acknowledged for his continued support regarding X-ray diffraction analysis. P. G. H. thanks the University of Lethbridge for a Tier I Board of Governors Research Chair in Organometallic Chemistry.

References

1. O. T. Summerscales and J. C. Gordon, RSC Adv., 2013, 3, 6682.
2. D. J. Beetstra, A. Meetsma, B. Hessen and J. H. Teuben, Organometallics, 2003, 22, 4372.
3. G. R. Geisbrecht, D. L. Clark, P. J. Hay, D. W. Keogh, R. Poli, B. L. Scott, J. G. Watkin and J. C. Gordon, Organometallics, 2002, 21, 4726.
4. A. A. Trifonov, M. N. Bochkarev, H. Schumann and J. Loebel, Angew. Chem., Int. Ed. Engl., 1991, 30, 1149.
5. J. Scott, F. Basuli, A. R. Fout, J. C. Huffman and D. J. Mindiola, Angew. Chem., Int. Ed., 2008, 47, 8502.
6. B. F. Wicker, H. J. Fan, A. K. Hickey, M. G. Crestani, J. Scott, M. Pink and D. J. Mindiola, J. Am. Chem. Soc., 2012, 134, 20081.
7. T. Chu, W. E. Piers, J. L. Dutton and M. Parvez, Organometallics, 2013, 32, 1159.
8. E. L. Lu, Y. X. Li and Y. F. Chen, Chem. Commun., 2010, 46, 4469.
9. J. X. Chu, X. H. Han, C. E. Kefalidis, J. L. Zhou, L. Maron, X. B. Leng and Y. F. Chen, J. Am. Chem. Soc., 2014, 136, 10894.
10. E. L. Lu, J. X. Chu, Y. F. Chen, M. V. Borzov and G. Y. Li, Chem. Commun., 2011, 47, 743.
11. J. X. Chu, E. L. Lu, Z. X. Liu, Y. F. Chen, X. B. Leng and H. B. Song, Angew. Chem., Int. Ed., 2011, 50, 7677.
12. J. X. Chu, E. L. Lu, Y. F. Chen and X. B. Leng, Organometallics, 2013, 32, 1137.
13. E. Lu, J. Chu and Y. F. Chen, Acc. Chem. Res., 2018, 51, 557.
14. W. Rong, J. Cheng, Z. Mou, H. Xie and D. Cui, Organometallics, 2013, 32, 5523.
15. D. Schädle, M. Meermann-Zimmermann, C. Schädle, C. Maichle-Mössmer and R. Anwander, Eur. J. Inorg. Chem., 2015, 2015, 1334.
16. L. A. Solola, A. V. Zabula, W. L. Dorfner, B. C. Manor, P. J. Carroll and E. J. Schelter, J. Am. Chem. Soc., 2016, 138, 6928.
17. T. E. Rieser, D. Schädle, C. Maichle-Mössmer and R. Anwander, Chem. Sci., 2024, 15, 3562–3570.
18. T. E. Rieser, R. Thim-Spöring, D. Schädle, P. Sirsch, R. Litlabø, K. W. Törnroos, C. Maichle-Mössmer and R. Anwander, J. Am. Chem. Soc., 2022, 144, 4102–4113.
19. D. Schädle, C. Maichle-Mössmer, C. Schädle and R. Anwander, Chem. – Eur. J., 2015, 21, 662.
20. J. P. Knott, M. M. Hänninen, J. M. Rautiainen, H. M. Tuononen and P. G. Hayes, J. Organomet. Chem., 2017, 845, 135.
21. C. S. MacNeil, K. E. Glynn and P. G. Hayes, Organometallics, 2018, 37, 3248.
22. K. R. D. Johnson, M. A. Hannon, J. S. Ritch and P. G. Hayes, Dalton Trans., 2012, 41, 7873.
23. M. T. Zamora, K. R. D. Johnson, M. M. Hanninen and P. G. Hayes, Dalton Trans., 2014, 43, 10739.
24. K. R. D. Johnson and P. G. Hayes, Organometallics, 2011, 30, 58.
25. T. M. Cameron, J. C. Gordon and B. L. Scott, Organometallics, 2004, 23, 2995.
26. T. M. Cameron, J. C. Gordon, B. L. Scott and W. Tumas, Chem. Commun., 2004, 1398.
27. J. D. Masuda, K. C. Jantunen, B. L. Scott and J. L. Kiplinger, Organometallics, 2008, 27, 1299.
28. R. K. Thomson, M. J. Monreal, J. D. Masuda, B. L. Scott and J. L. Kiplinger, J. Organomet. Chem., 2011, 696, 3966.
29. S. Zhu, W. Wu, D. Hong, F. Chai, Z. Huang, X. Zhu, S. Zhou and S. Wang, Inorg. Chem., 2024, 63, 14860.
30. L. K. Knight, W. E. Piers, P. Fleurat-Lessard, M. Parvez and R. McDonald, Organometallics, 2004, 23, 2087.
31. Z.-J. Lv, M. Zhu, W. Liu, Z. Chai, J. Wei and W.-X. Zhang, Organometallics, 2021, 40, 3992.
32. C. Temple and J. A. Montgomery, J. Org. Chem., 1965, 30, 826.
33. C. Temple, M. C. Thorpe, W. C. Coburn and J. A. Montgomery, J. Org. Chem., 1966, 31, 935.
34. K. R. D. Johnson and P. G. Hayes, Organometallics, 2009, 28, 6352.
35. K. R. D. Johnson, B. L. Kamenz and P. G. Hayes, Can. J. Chem., 2015, 94, 330.
36. K. R. D. Johnson and P. G. Hayes, Inorg. Chim. Acta, 2014, 422, 209.
37. K. R. D. Johnson, B. L. Kamenz and P. G. Hayes, Organometallics, 2014, 33, 3005–3011.
38. K. R. D. Johnson and P. G. Hayes, Dalton Trans., 2014, 43, 2448.
39. K. R. D. Johnson and P. G. Hayes, Organometallics, 2013, 32, 4046.
40. K. C. Jantunen, B. L. Scott, P. J. Hay, J. C. Gordon and J. L. Kiplinger, J. Am. Chem. Soc., 2006, 128, 6322.
41. A. Arndt, P. Voth, T. P. Spaniol and J. Okuda, Organometallics, 2000, 19, 4690.
42. F. Estler, G. Eickerling, E. Herdtweck and R. Anwander, Organometallics, 2003, 22, 1212.

---

Footnote

† Electronic supplementary information (ESI) available: Including NMR spectra of key compounds and X-Ray Crystallographic Data. CCDC 2417587–2417592, 2423088 and 2429452 for compounds 5–8, 11 and 13. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d5dt00338e

---