Selective white phosphorus activation and functionalization with inorganic Grignard reagents

Abstract

We describe the targeted and selective functionalization of white phosphorus (P₄) using ‘inorganic Grignard reagents’. Reactions of the Fe–Mg complexes [(^DippBDI)Mg(THF)_xFp] (Fp = CpFe(CO)₂, 1: x = 0; 1-THF: x = 1, ^DippBDI = 2,6-diisopropylphenyl-1,3-diketiminate)) and [(^DippBDI)MgFp*] (2, Fp* = Cp*Fe(CO)₂) afford the compounds [(^DippBDI)Mg(THF)_xFp(µ-P₄)] (3: x = 0; 3-THF: x = 1) and [(^DippBDI)MgFp*(µ-P₄)] (4) featuring P₄²⁻ ligands with a ‘butterfly’ structure bridging Fe and Mg. The coordination of the β-diketiminate magnesium cations [(DippBDI)Mg]⁺ plays a key role in stabilizing these reactive ‘P₄-butterfly’ anions through non-covalent Mg–P bonds and dispersion forces. Thermolysis or photolysis of 3-THF and 4 afforded rare tetraphosphacyclopentadienolate complexes 7 and 8. Complex 4 exhibits unmitigated reactivity towards a wide range of main group element electrophiles (ⁱPrNCNⁱPr, Ph₂BCl, Ph₃SnCl, (Me₃Si)₃SiCl and Mes₂PCl), furnishing rare examples of stable, mixed-substituted tetraphosphanes 9–13 as well as an octaphosphane 14 featuring a Sn₂P₈ core.

---

Introduction

White phosphorus (P₄) is the only molecular allotrope of phosphorus stable at room temperature and serves as a crucial feedstock for phosphorus-based chemicals.¹ Because of its important role in the phosphorus industry and its multifaceted and often poorly predictable reactivity, the chemistry of P₄ is a subject of significant current interest. Although many procedures for P₄ activation using main-group elements and transition metals have been reported, the targeted, selective functionalization of the resulting polyphosphorus frameworks lags behind simple P₄ activation.²

The bicyclo[1.1.0]tetraphosphabutane motive (‘P₄-butterfly’) resulting from the cleavage of one P–P bond of the P₄ tetrahedron represents the first step in P₄ activation.^3,4 The synthesis of ‘P₄-butterfly’ compounds has been achieved by reaction of P₄ with main-group radicals and transition-metal-based radicals,³ as well as using carbenes and heavier carbene analogues.⁴ Reactions of this type result in the generation of symmetrical P₄-butterfly species of types A or B (Fig. 1). Due to the strong, covalent phosphorus-element bonds, these species typically exhibit attenuated reactivity and limited potential for further functionalization. In contrast, the reaction of P₄ with charged nucleophiles affords highly reactive anions [RP₄]⁻ (C), which tend to decompose quickly.^5,6 Nonetheless, alkali metal salts of type M[HP₄] have been characterised by ³¹P NMR spectroscopy at low temperature.^5,7 Lammertsma and coworkers have reported that [RP₄]⁻ anions (C) containing sterically demanding aryl groups or the Fp*⁻ anion (Fp* = Cp*Fe(CO)₂, Cp* = C₅Me₅) can be stabilized by the coordination of boranes such as BPh₃ and B(C₆F₅)₃ to one of the wingtip phosphorus atoms (D, Fig. 1a, R = bulky aryl or Fp*).^8–12 However, the reactivity of these species has been limited to protonation and alkylation reactions, a [3 + 1] fragmentation reaction induced by PhNCO and Lewis acid coordination with BH₃, BPh₃, W(CO)₅ and [Au(IPr)]⁺ [IPr = 1,3-bis(2,6-diisopropylphenyl)imidazolin-2-ylidene]. The P₄-butterfly species E resulting from the functionalization of D with R′X decomposes with R′ = H or Me, but can be isolated when sterically demanding alkyl groups such as Ph₃C are employed.^8,11,12

Fig. 1 (a) Bicyclo[1.1.0]tetraphosphabutanes and related anions A–E resulting from the reaction of P₄ with radicals, carbene analogues, and nucleophilic anions. (b) Reaction products of P₄ with BDI-stabilized Mg(II), Mg(I) and Ca(I) complexes. (c) Selective P₄ activation and functionalization enabled by Mg–Fe pairs (this work; Mes* = 2,4,6-tBu₃C₆H₂, Dmp = 2,6-Me₂-C₆H₃, Dipp = 2,6-diisopropylphenyl, Dipep = 2,6-diisopentylphenyl, Ar = 2,6-Dipp₂-C₆H₃).

Recently, we have shown that the combination of a redox-active 3d-metal cation with a main-group element, such as Ga, Si, Sn, or a redox non-innocent d-block element such as Zn, facilitates P₄ activation.¹³ In contrast, the use of transition-metal-magnesium complexes (‘inorganic Grignard reagents’) for P₄ activation has hardly been explored,^14–16 despite recent studies demonstrating the significant potential of low-valent magnesium and calcium complexes for P₄ storage and activation (Fig. 1, complexes F–H, Fig. 1).^17,18 Here, we demonstrate that the inorganic Grignard reagents [(^DippBDI)Mg(THF)_xFp] (1: x = 0; 1-THF: x = 1, Fp = CpFe(CO)₂, ^DippBDI = 2,6-diisopropylphenyl-1,3-diketiminate) and [(^DippBDI)MgFp*] (2) selectively afford P₄-butterfly complexes [(^DippBDI)Mg(THF)_xFp(µ-P₄)] (3: x = 0; 3-THF: x = 1) and [(^DippBDI)MgFp*(µ-P₄)] (4). The reversible nature and the mechanism of the P₄ activation reaction are investigated using variable-temperature ³¹P NMR spectroscopy, ¹H exchange NMR spectroscopy, and DFT calculations. An analysis of the non-covalent interactions highlights the essential role of the magnesium cation in stabilizing the complexes. Thermolysis or photolysis of 3-THF and 4 generate rare tetraphosphacyclopentadienolate transition metal complexes 7 and 8. Additionally, we show that ‘P₄-butterfly’ ligand in 4 is readily functionalized with various electrophiles. These reactions afford unprecedented mixed-substituted bicyclo[1.1.0]tetraphosphabutanes of type Fp*P₄R [9–13, R =C(NⁱPr)₂ (9), BPh₂ (10), SnPh₃ (11), Si(SiMe₃)₃ (12), PMes₂ (13)] as well as the octaphosphane [Ar₂Sn₂P₈Fp*₂] (14, Ar = 2,6-Dipp₂-C₆H₃). Our results demonstrate that inorganic Grignard reagents act as effective metal-based Lewis pairs, enabling the selective activation of P₄ and the targeted functionalization of the resulting P₄²⁻ ligand.

Results and discussion

Synthesis and characterization of P₄-butterfly complexes

Inspired by the work of Crimmin and Mountford on the use of the inorganic Grignard reagents 1 and 1-THF in C–F activation and heterocumulene insertion,^14,15 we investigated their reactivity with P₄. Initial ³¹P NMR spectroscopic studies showed that P₄ readily inserts into the Mg–Fe bond of 1-THF in THF-d₈ or C₆D₆ to generate 3-THF, which gives rise to an AMX₂ spin system. Likewise, the reaction of 1 with P₄ in toluene-d₈ afforded the THF-free analogue 3. Complexes 3 and 3-THF can be isolated as air-sensitive yellow-orange crystals in moderate yields (50% and 22%, respectively). The AMX₂ spin systems observed in the ³¹P NMR indicate the formation of a P₄-butterfly structure. The ²J_31P–31P coupling constants between the wingtip P atoms (47.0 and 36.8 Hz) indicate an endo,exo configuration of the butterfly-P₄ ligands,^10,12,19 suggesting that there is no through-space interaction between these P atoms.²⁰ The ¹H and ³¹P NMR spectra furthermore indicate that 3 and 3-THF are in an equilibrium with the starting materials 1 and 1-THF and P₄ in deuterated benzene and deuterated THF (vide infra).

Single-crystal X-ray diffraction (scXRD) studies on 3 and 3-THF confirmed that the P₄ molecule has formally inserted into the Mg–Fe bond (Fig. 2). Both complexes crystallize as the endo,exo isomer. One of the wing-tip phosphorus atoms (P1) coordinates to iron. In the structure of 3, the Mg atom bridges the two wing-tip P atoms, P1 and P4, with similar distances (Mg1–P1 2.6529(8) Å, Mg1–P4 2.5623(8) Å). In contrast, a THF molecule is coordinated to the magnesium atom in 3-THF, resulting in a more asymmetric coordination environment (Mg1–P1 2.9427(8) Å, Mg1–P4 2.6320(8) Å). The P–P bond lengths are characteristic of P₄-butterfly species, revealing a short P–P bond between the bridgehead P atoms (3: P2–P3 2.163(1) Å; 3-THF: P2–P3 2.1598(8) Å).^8,11,19

Fig. 2 Solid-state molecular structures of 3 (top) and 3-THF (bottom) measured at 123 K. Displacement ellipsoids are shown at the 30% probability level. Hydrogen atoms, non-coordinating solvent molecules and disordered groups are omitted for clarity. Selected bond lengths [Å] and angles [°] of 3: Fe1–P1 2.3023(6), Mg1–P4 2.5623(8), Mg1–P1 2.6529(8), P1–P2 2.2188(7), P1–P3 2.2247(7), P2–P3 2.163(1), P2–P4 2.1922(9), P3–P4 2.1956(8), C1–O1 1.144(3),C2–O2 1.143(3), P1–Mg1–P4 70.55(2), N1–Mg1–N2 92.64(7). 3-THF: Fe1–P1 2.3527(6), Mg1–P4 2.6320(8), Mg1–P1 2.9427(8), P1–P2 2.2044(7), P1–P3 2.2112(8), P2–P3 2.1598(8), P2–P4 2.2034(8), P3–P4 2.1909(8), C1–O1 1.156(3), C2–O2 1.147(3), P1–Mg1–P4 63.44(2), N1–Mg1–N2 92.16(7).

To assess the influence of substituents on the Cp ligand, we synthesized the novel [(^DippBDI)MgFp*] complex 2.† Unlike 1, complex 2 forms a dimer in the solid state via the coordination of the CO ligands to magnesium, as previously observed for related complexes^9,10.‡ ¹H NMR and DOSY NMR studies reveal an equilibrium between monomeric and dimeric species in C₆D₆ solution (see Fig. S71–S73). The reaction of 2 (0.5 equiv.) with P₄ in toluene or benzene cleanly yields 4 as a single endo,exo-configured product, which can be isolated in a high yield (72%) as a yellow-orange crystalline solid (Scheme 1). The molecular structure is very similar to that of 3, as determined by scXRD studies (Fig. 2). In contrast, complex 4 forms a mixture of endo,exo and exo,exo complexes when using THF. Both isomers decompose rapidly in that solvent to the symmetrical [{Fp*}₂(µ-P₄)] (Fig. S83 and S84, SI), illustrating the stabilizing nature of the ion-pairing between the [(^DippBDI)Mg]⁺ cations and the [Fp*(µ-P₄)] anion.

Scheme 1 Synthesis of complexes 3-THF, 3 and 4. The reaction is reversible with 1 and 1-THF and irreversible with 2.

Mechanism of the P₄ insertion reaction

To gain more insight into the temperature-dependent equilibrium between 1-THF, 3-THF and P₄, we performed variable-temperature (VT) ³¹P{¹H} NMR spectroscopic measurements of 3-THF in THF-d₈. A van 't Hoff analysis provided the thermodynamic parameters of the P₄ activation reaction (ΔG(298 K) = −3.9(±1.7) kJ mol⁻¹, ΔH = −37.8(±0.8) kJ mol⁻¹ and ΔS = −115(±3) J mol⁻¹ K⁻¹; see also Table S8 and Fig. S63). Exchange spectroscopy (EXSY) of the Cp signals in the ¹H NMR spectrum revealed a low activation barrier (ΔG^‡(298 K) = 73.1(±4.3) kJ mol⁻¹, ΔH^‡ = 16.9(±2.9) kJ mol⁻¹, ΔS^‡ (298 K) = −188.3(±10.3) J mol⁻¹ K⁻¹; see Table S11 and Fig. S69). 3-THF shows a similar behaviour in toluene-d₈ (ΔG(298 K) = −5.0(±6.4) kJ mol⁻¹, ΔH = −43.7(±3.1) kJ mol⁻¹ and ΔS = −129(±11) J mol⁻¹ K⁻¹; see Table S9 and Fig. S63, S64, S76). However, in this case, the van 't Hoff plot shows a nonlinear relationship between 193 K and 233 K, suggesting that the equilibrium was not reached at low temperatures.

The donor-free complex [(^DippBDI)MgFp] (3) shows a similar temperature-dependent equilibrium in non-coordinating solvents (C₆D₆ or toluene-d₈).¹⁵ In toluene-d₈, an additional P-containing species is observed at temperatures below 253 K besides P₄, which we assign to the exo,exo isomer of 3 (Fig. S79 and S80).

DFT calculations suggest that the mechanism of the P₄ activation reaction in THF-d₈ likely involves the heterolytic cleavage of the Mg–Fe bond in compound 1-THF (Fig. 3), which appears to be barrierless (see Fig. S128). The coordination of THF generates a solvent-separated ion pair consisting of the Fp⁻ anion and the [(^DippBDI)Mg(THF)₂]⁺ cation. Fp⁻ attacks P₄ to form the [FpP₄]⁻ anion, which is then trapped by [(^DippBDI)Mg(THF)₂]⁺. The Gibbs free energy (ΔG⁰ and ΔG^‡) profile calculated by density functional theory (DFT) for this mechanism shown in Fig. 3 qualitatively agrees with the experimental data. However, the calculations slightly overestimate ΔG⁰ and underestimate ΔG^‡. In non-coordinating solvents such as toluene, an alternative mechanism to that depicted in Fig. 3 is likely, in which the CO ligands coordinate to magnesium as previously observed in the molecular structures of magnesium-transition metal carbonylates.^9,10 A different reaction pathway, in which P₄ coordinates to Mg, but the [(^DippBDI)Mg]⁺ unit does not dissociate from the Fp⁻ anion, is also feasible and features only a slightly higher activation barrier as shown by DFT calculations (see Fig. S129 in the SI). The coordination of P₄ to magnesium(II) centers has already been reported.¹⁸

Fig. 3 Computed Gibbs free energy profile for P₄ activation with 1-THF. Relative Gibbs free energies are presented in kJ mol⁻¹. Level of theory: MN15/def2-TZVP/SMD(THF)//MN15/def2-SVP/SMD(THF).

It is noteworthy that the complexes [(^DepBDI)Mg(THF)Fp] (5-THF) and [(^MesBDI)Mg(THF)Fp] (6-THF), featuring smaller Mes and Dep substituents on the BDI ligand, gave only traces of the desired P₄-butterfly in C₆D₆ (5% and 3%, respectively), according to qualitative integration of the ³¹P{¹H} NMR spectra (Dep = 2,6-Et₂-C₆H₃, see Table S15 and Fig. S83). This indicates that the strength of the ion-pairing interaction between the Fp⁻ and the magnesium ion significantly influences the equilibrium. The DFT-calculated reaction energies are in full agreement with these findings, showing a greater thermodynamic preference for the formation of the P₄-butterfly derivatives with Ar = Dipp by up to −16.2 kJ mol⁻¹, compared to Ar = Dep or Mes (Table S21).

Non-covalent interaction analysis

To investigate the influence of the cation on the thermodynamic stability, we performed a non-covalent interaction (NCI) analysis on the complexes 3, 4 and the hypothetical complex [LiFp*(µ-P₄)]²¹.§ The resulting NCI and reduced density gradient (RDG) plots obtained from the wavefunction calculated at the MN15/def2-TZVP/SMD(THF)//MN15/def2-SVP/SMD(THF)²² level of theory are shown in Fig. 4 (see also Fig. S130–S132). Ionic interactions within compound 4 (see Fig. 4 left) and [LiFp*Fe(µ-P₄)] (see Fig. 4 right) are visualized in green to blue (sign(λ₂)ρ < −0.015 a.u.) in the NCI isosurfaces and RDG scatter plots. The Mg cation in 4 shows a stronger interaction with the wingtip phosphorus atoms compared to the Li cation in [LiFp*(µ-P₄)], as indicated by the dark blue color. Dispersion interactions between the [(^DippBDI)Mg]⁺ cation and [Fp*(µ-P₄)]⁻ additionally stabilize the complex (Fig. 4).

Fig. 4 3D NCI-plot and RDG scatter plot of 4 (left) and of [LiFp*(µ-P₄)] (right). RDG scatter plot and 3D NCI-plots for ionic (sign(λ₂)ρ range: −0.05 a.u. to −0.015 a.u.) and van der Waals region (sign(λ₂)ρ range: −0.015 a.u. to +0.015 a.u.) The repulsive interaction regions were omitted for clarity.

Thermolysis of 3-THF and photolysis of 4: synthesis of 7 and 8

Given the propensity of carbonyl complexes for the dissociation of CO,²³ we examined the thermal and photolytic stability of 3-THF. Upon heating of a solution in C₆D₆ to 60 °C for 5 d, 3-THF was converted to a new species 7 (Scheme 2) characterized by an AA′BB′ spin system in the ³¹P{¹H} NMR spectrum (δ = 44.3 ppm and 30.7 ppm, Fig. S15 and Table S1). It is noteworthy that the presence of the THF ligand in 3-THF is essential for the formation of 7, since the THF-free 3 appears to be completely stable in C₆D₆ upon heating to +60 °C and under irradiation with a neutral white-light LED (see Fig. S87). In contrast, 4 is also thermally stable but rearranges upon exposure to daylight, forming complex 8 over several days (Scheme 2).

Scheme 2 Synthesis of 7 and 8, r.t. = room temperature.

The ³¹P{¹H} NMR spectrum of 8 shows an AA′XX′ spin system similar to that of 7 (δ = 61.3 and 36.4 ppm, see Fig. 5). The scXRD analyses of 7 and 8 revealed the formation of ferrocene-like sandwich complexes with η⁵-coordinated tetraphosphacyclopentadienolate ligands. The planar P₄C(O) rings are eclipsed with respect to the co-planar Cp or Cp* moiety. The P–P and P–C bond lengths indicate a delocalized bonding situation (7: P–P2.1013(8) to 2.124(1) Å, P–C1.786(2) and 1.790(2) Å; 8: P–P2.100(2) to 2.135(2) Å, P–C1.781(5) and 1.785(6) Å). Complex 7 crystallizes as a monomer while the THF-free compound 8 forms a polymeric structure via coordination of the P2 atom to a [(^DippBDI)Mg]⁺ cation of an adjacent molecule (Mg1–P2 2.793(2) Å, Fig. S93). To our knowledge, 7 and 8 are the first complexes featuring η⁵-coordinated tetraphosphacyclopentadienolate P₄C(O)⁻ ligands. A similar CO insertion resulting in an η³-P₄C(O) ligand has been observed for the reaction of [Fe₂(CO)₉] with (Cp‴Fe)₂(µ-P₄) (Cp‴ = C₅H₂^tBu₃), which affords the oligonuclear complex [{Cp‴Fe(CO)₂}₂(µ₄,η^{3 : 1 : 1 : 1}-P₄CO){Fe(CO)₄}{Fe(CO)₃}].²⁴ Additionally, a few related tetraphospholide complexes are known which display similar P–P bond lengths,^25–29 e.g. [Cp^RFe(η⁵-P₄CR′)] (Cp^R = Cp, R′ = ^tBu, adamantyl, SiMe₃; Cp^R = Cp*, R′ = Mes, phenyl; Cp^R = Cp‴, R′ = ^tBu),^25,29 and [Cp*Fe(P₄SiL)] (L = PhC{N^tBu}₂).²⁶

Fig. 5 Experimental (upwards) and simulated (downwards) ³¹P{¹H} NMR spectrum of 7 (left) and its solid-state molecular structure measured at 123 K (right). The solid-state molecular structure and the ³¹P{¹H} NMR spectrum of 8 are similar and therefore displayed in the ESI. Displacement ellipsoids are shown at the 30% probability level. Hydrogen atoms are omitted, and the Dipp groups are shown in wireframe representation for clarity. Selected bond length [Å] and angles [°]: P3–P4 2.100(2), P1–P2 2.112(2), P2–P3 2.135(2), P1–C1 1.781(5), P4–C1 1.785(6), C1–O1 1.309(5), Fe–P 2.345(1)–2.361(2), Fe1–C1 2.231(4), C1–O1–Mg1 145.8(3), P₄C_centroid–Fe1–C_5-centroid 174.1.

Reactivity of 4 towards electrophiles

The high-yielding synthesis of complex 4 enabled us to investigate its reactivity towards electrophiles in detail. During these studies, we found that 4 reacts readily and selectively with numerous electrophiles, including main-group element halides and heterocumulenes (Scheme 3). The products exhibit characteristic AMX₂ spin systems in the ³¹P{¹H} NMR spectra, which provide evidence for P-functionalized P₄-butterfly compounds. Using bulky electrophiles such as ⁱPrNCNⁱPr, Ph₂BCl, Ph₃SnCl, (Me₃Si)₃SiCl, and Mes₂PCl we were able to isolate the new P₄-butterfly complexes 9–13. Complexes 9, 10 and 12 are stable in solution for several days, while 11 and 13 gradually decompose at ambient temperature (Fig. S54 and S84). It is noteworthy that 11 dimerises slowly to a tricyclic P₈ cluster 15, as revealed by ³¹P NMR data and a preliminary scXRD analysis (Fig. S48). Weigand and co-workers observed a similar dimerization reaction for the pentaphosphane P₅(C₆F₅)₂ generated from (^DippBDI)GaP₄ butterfly with (C₆F₅)₂PBr.³⁰

Scheme 3 Reactivity of 4 towards selected electrophiles.

ScXRD studies revealed the molecular structures of 9–13. Complex 9 results from an insertion of the carbodiimide ⁱPrNCNⁱPr into the Mg–P bond of 4. The two nitrogen atoms coordinate to magnesium, and the carbon atom is bound to phosphorus (Fig. 6). For 9, 12, and 13 (Fig. S1), the butterfly-P₄ unit is an exo,exo configuration with the Fp* and carbodiimide, (Me₃Si)₃Si- and Mes₂P-substituents pointing toward the bridgedhead P atoms. In contrast, complex 10 shows a symmetrically bridging Ph₂B moiety (P–B 2.016(2) Å and 2.055(2) Å). An endo,exo configuration is likewise observed for the Ph₃Sn-substituted complex 11.

Fig. 6 Solid-state molecular structures of 9, 10, 11 and 12 measured at 123 K. Displacement ellipsoids are shown at the 30% probability level. Hydrogen atoms, non-coordinating solvent molecules and disordered groups are omitted for clarity. The Dipp groups in 7 are depicted in wireframe representation for clarity. Selected bond length[Å] and angles [°] are for 9: Fe1–P1 2.3064(6), P1–P2 2.2169(6), P1–P3 2.2286(7), P2–P3 2.1795(7) P2–P4 2.2119(7), P3–P4 2.2083(7), P4–C3 1.889(2), C1–O1 1.148(2), C2–O2 1.147(2), N2–C3 1.332(2), N1–C3 1.335(2), N1–C3–N2 114.4(2), N3–Mg1–N4 94.24(7). 10: Fe1–P1 2.2434(4), P1–B1 2.016(2), P4–B1 2.055(2), P1–P2 2.2063(6), P1–P3 2.2035(5), P2–P3 2.1974(7), P3–P4 2.2100(6), P2–P4 2.2108(7), C1–O1 1.145(2), C2–O2 1.142(2), P1–B1–P4 83.67(7), Fe1–P1–B1 134.90(5). 11: Fe1–P4 2.2734(7), Sn1–P1 2.5675(7), P1–P2 2.2063(9), P1–P3 2.2080(9), P2–P3 2.179(1), P2–P4 2.2233(9), P3–P4 2.2060(9), C1–O1 1.147(3), C2–O2 1.149(3). Second molecule in asymmetric unit: Fe1–P4 2.2848(7), Sn1–P1 2.5549(7), P1–P2 2.206(1), P1–P3 2.208(1), P2–P4 2.2261(9), P3–P4 2.2084(9), C1–O1 1.145(3), C2–O2 1.145(3). 12: Fe1–P1 2.294(1), P4–Si 2.294(1), P1–P2 2.218(1), P1–P3 2.231(2), P3–P4 2.217(1), P2–P3 2.167(1), P2–P4 2.214(1), C1–O1 1.152(5), C2–O2 1.144(5), Si1–P4–P3 101.91(5), P1–P3–P4 81.98(5), Fe1–P1–P3 109.49(5).

The ³¹P{¹H} NMR data are consistent with these crystallographic findings. For complex 9, the ²J_31P–31P coupling (258 Hz) is larger than the ¹J_PP couplings (179 Hz, 185 Hz), indicating an exo,exo butterfly structure in solution. The ¹³C{¹H} NMR spectrum displays a doublet of doublet for the quaternary carbon atom of the carbodiimide unit due to C–P coupling (δ = 175.0 ppm, ¹J_PC = 103.9 Hz, ³J_PC = 23.3 Hz, Fig. S21). The exo,exo isomer is also present for 12 and 13, as evidenced by the high-field shifted P_M resonances of the AMX₂ spin systems and large ²J_PP couplings (²J_AM = 235 Hz and 259 Hz, respectively; see Tables S5 and S6), which result from through-space interactions of the lone pairs.¹⁰ Additionally, ²⁹Si satellites (¹J_SiP = 97.5 Hz) were detected for the P atom with the (Me₃Si)₃Si substituent. In contrast, the small ²J_PP coupling constant (²J_AM = 33 Hz) indicates that compound 10 retains the endo,exo configuration in solution. The resonances assigned to the wingtip phosphorus atoms P1 and P4 exhibit chemical shifts similar to compound 4, while the signal of the bridgehead phosphorus atoms P2 and P3 is shifted downfield by 50.4 ppm. The ¹¹B{¹H} NMR spectrum shows a broad signal (Δν_½ = 362 Hz) at 20.5 ppm with unresolved coupling to phosphorus. The ³¹P{¹H} NMR spectra of the Ph₃Sn-substituted complex 11 display two distinct AMX₂ spin systems in an approximate 1 : 1 integral ratio, indicating that the endo,exo and exo,exo isomers coexist in solution (Fig. S30). However, when a sample of isolated 11 was dissolved at −80 °C in toluene-d₈ and the ³¹P{¹H} NMR was recorded at −80 °C, only the signals of the endo,exo isomer were observed (²J_AM = 16 Hz, Table S3). Variable-temperature ³¹P{¹H} NMR spectroscopy of 11 shows a slow conversion to an equilibrium mixture of the endo,exo and exo,exo isomer at elevated temperature (Fig. S81). The ¹¹⁹Sn NMR spectrum of endo,exo 11 dissolved at −80 °C shows a doublet of doublets with a ¹J_119Sn–31P coupling of 973 Hz and a smaller coupling of ¹J_119Sn–31P = 299 Hz in agreement with the ^117/119Sn–P coupling constants derived from the ³¹P{¹H} NMR spectrum (see Table S3).

The reactivity of 4 towards electrophiles strongly depends on the nature and steric bulk of the electrophile. The substituents must provide sufficient steric protection to stabilize the resulting butterfly complex without suppressing reactivity. Thus, ³¹P{¹H} NMR spectroscopic monitoring studies indicate that 4 cleanly reacts with the moderately bulky electrophile Me₃SiBr to afford the desired P₄-butterfly compound Fp*P₄SiMe₃. However, this compound is unstable and decomposes overnight at room temperature. In contrast, the reaction of the bulkier ^tBuMe₂SiCl with 4 is slow, resulting in an incomplete substitution reaction. Oxidizing electrophiles such as (2,6-Mes₂-C₆H₃)PCl₂ (Mes = C₆H₂-2,4,6-Me₃) produce Fp*₂P₄ and P₄ among other decomposition products.

The reaction of 4 with [ArSnCl]₂ (0.5 equiv., Ar = 2,6-Dipp₂-C₆H₃, Dipp = 2,6-ⁱPr₂-C₆H₃) affords the polycyclic compound 14 (Fig. 7a), which formally is the dimerization product of the expected P₄-butterfly species Fp*P₄SnAr. The molecular structure of 14 consists of a pentacyclo[5.2.1.0^2,6.0^3,9.0^4,8]decane-like P₈Sn₂ core derived from a norbornane-like P₇ moiety with an exocyclic ArP unit and a Sn–Sn single bond (2.8591(2) Å).³¹ The P–P bond lengths (2.1747(6)–2.2483(6) Å) are in the range of P–P single bonds. Notably, the Sn1–P1 bond length 2.6726(4) Å results from a dative interaction involving the lone pair of P1.

Fig. 7 (a) Solid-state molecular structure of 14 measured at 123 K. Displacement ellipsoids are shown at the 30% probability level. Hydrogen atoms are omitted and Ar-groups are represented as wireframe for clarity. Selected bond length [Å] and angles [°]: Sn1–Sn2 2.8591(2), Sn1–P1 2.6726(4), Sn1–P8 2.5967(4), Sn2–P2 2.5968(4), Sn2–P6 2.5707(4), P1–P2 2.2644(6), P1–P5 2.2191(6), P2–P3 2.2015(6), P3–P4 2.1950(6), P3–P6 2.2483(6), P4–P5 2.1747(6), P5–P7 2.2054(6), P6–P7 2.2337(6), P7–P8 2.2008(6), Sn1–Sn2–P2 89.673(10)), P1–Sn1–Sn2 74.309(9), P6–Sn2–Sn1 106.62(1). (b) ¹¹⁹Sn{¹H} experimental NMR spectrum (224 MHz, 298 K, C₆D₆) (top) and simulation (bottom). An exponential window function (WDW EM) with a line broadening of 50 Hz has been used. The baseline has been corrected using a multipoint spline with the “.basl” and “sab” commands in Bruker TopSpin 4.20. (c) ³¹P{¹H} experimental spectrum (162 MHz, 298 K, C₆D₆) (top) and simulation bottom.

The remaining Sn–P distances are in a close range (2.5707(4)–2.5967(4) Å), consistent with Sn–P single bonds. The structural framework of 14 is reminiscent of bishomocubanes derived from dimethyl dicyclopentadienedicarboxylate (Thiele's ester).³² Additionally, some related carbocyclic polyphosphanes have been described, resulting from the dimerization of 1,2,4-triphosphacyclopenta-1,3-dienes or the reaction of R₃Sn stabilized 1,2,4-triphosphacyclopenta-1,3-dienes with tert-butylphosphaalkyne.³³ Notably, the dimerization of the As₄-butterfly complexes [{Cp‴Fe(CO)₂}₂(µ,η^1 : 1-As₄)] or [{Cp*Cr(CO)₃}₂(µ,η^1 : 1-As₄)] afford realgar-like (tricyclo[3.3.0.0^3,7]octane)-As₈ species.³⁴ Similar realgar-like P₈ units are also known.³⁵

The ¹¹⁹Sn NMR spectrum of 14 shows an apparent doublet of doublets at 380 ppm, and a broad doublet (Δν_½ > 700 Hz) with unresolved fine structure at −47 ppm. The assignment of the Sn1 and Sn2 signals is based on the calculated ¹J_119Sn–31P coupling constants of 1471 Hz for Sn1–P8, 970 Hz for Sn1–P1 and 823 Hz for Sn2–P6 (Fig. 7b; see SI, section 8.3 for details). These data compare well with the experimental values of 1540 Hz, 882 Hz and 830 Hz, respectively. The ¹¹⁷Sn and ¹¹⁹Sn satellites could be detected in the ¹¹⁹Sn NMR spectrum (¹J_{119Sn1–119Sn2} ≈ 6700 Hz; calculated −7663 Hz), confirming that the Sn–Sn bond remains intact in solution. For comparison, the ¹J_{119Sn−119Sn} coupling constants in Ar*SnSnPh₂Ar* and Ar*SnSn(Me₂)Ar* [Ar* = 2,6-(2,4,6-ⁱPr₃-C₆H₂)₂-C₆H₃] are 7237 Hz and 8330 Hz, respectively.³⁶

The ³¹P{¹H} NMR spectrum of 14 displays a set of eight multiplets between 112.4 ppm and −67.1 ppm for the eight distinct P-atoms. The coupling constants were determined by simulating the spectrum using an iterative fitting procedure (Fig. 7c and Table S7, SI). Compound 14 is unstable in solution at ambient temperature. Upon storing a solution at room temperature, the signals of 14 gradually disappear over the course of three weeks, and two new sets of signals appear in the ³¹P{¹H} NMR spectrum in the same chemical shift range as for 14, indicating that it rearranges slowly in solution to form a new complex 16 (see SI, Fig. S60). A preliminary scXRD analysis shows that 14 eliminates one CO ligand. As a result, one of the Fe atoms binds to two adjacent phosphorus atoms (Fig. S49). Based on the very similar chemical shifts and coupling constants in the ¹¹⁹Sn and ³¹P{¹H} NMR spectra (Fig. S57 and S60), we assume that two isomers of 16 coexist in solution.

Conclusions

We have shown that the inorganic Grignard reagents 1 and 2 effectively activate P₄ through furnishing P₄-butterfly dianions that are coordinated by Fe and the bulky β-diketiminato magnesium cation [(^DippBDI)Mg]⁺. Our theoretical analysis emphasizes the importance of non-covalent and dispersion interactions in stabilizing these complexes, which appear to be stronger for [(^DippBDI)Mg]⁺ than for Li⁺.§ The reversible dissociation of the Fe–Mg Lewis pairs 1 and 2 into solvent-separated ion pairs appears to be key. According to DFT calculations, the reactions proceed stepwise via nucleophilic attack by the Fp⁻ or Fp*⁻ anion on P₄ and subsequent coordination to [^DippBDIMg]⁺. However, unlike previous studies, these reactions do not require an external Lewis acid, simplifying reactivity investigations. Reactions of 4 with electrophiles generate a variety of new mixed substituted P₄-butterfly species 9–13, as well as the novel Sn₂P₈ cluster 14. Additionally, thermally or photochemically induced rearrangements led to rare tetraphosphole complexes 7 and 8.

Our results indicate that combining transition-metalate anions with the [^DippBDIMg]⁺ cation is a promising strategy for deliberate and selective P₄ activation and functionalization. When sufficient steric bulk is provided, the resulting P₄-butterfly complexes can be easily isolated, marking an ideal starting point for future reactivity studies. In other cases, dimerization to P₈ frameworks can be observed. Ongoing work in our group focuses on the reaction chemistry of complexes 9–14 and on the broader applicability of our approach using a wider range of metal-based Lewis pairs.

Author contributions

F. G. conceptualisation, investigation – synthesis and characterisation, writing – original draft; J. B. investigation – DFT studies of the mechanism and NCI analysis; W. M. S. investigation – DOSY and EXSY studies; F. W. investigation – Sn NMR studies and DFT calculations; R. W., G. B. conceptualisation, supervision; R.W., R. G., H. Z. supervision and funding acquisition. All authors contributed to manuscript review, editing and discussion.

Conflicts of interest

There are no conflicts to declare.

Data availability

CCDC 2529860–2529874 and 2541029 contain the supplementary crystallographic data for this paper.^38a–p

The data supporting the findings of this work is available in the main text or the electronic supplementary information (SI). Primary research data for this work is openly accessible on Radar4Chem (https://radar4chem.radar-service.eu/radar/de/home) under the DOI: https://doi.org/10.22000/7wa7f1kdq20s260j. Supplementary information: experimental procedures, details of NMR spectroscopic studies, scXRD data, cyclic voltammetry data, IR and UV-vis spectra, and computational data. See DOI: https://doi.org/10.1039/d6sc01229a.

Acknowledgements

We thank the Deutsche Forschungsgemeinschaft for funding (RTG 2620 Ion Pair Effects in Molecular Reactivity, project number 426795949). J. B. and H. Z. acknowledge the computational and data resources provided by the Leibniz Supercomputing Centre (https://www.lrz.de). F. G. thanks the Fonds der Chemischen Industrie for a Kekulé Fellowship.

Notes and references

1. (a) D. Corbridge, Phosphorus. Chemistry, Biochemistry and Technology, Sixth Edition, CRC Press, 2013; (b) M. A. de Boer, L. Wolzak and J. C. Slootweg in Phosphorus Recovery and Recycling, ed. H. Ohtake and S. Tsuneda, Springer Singapore, Singapore, 2019, pp. 75–100; (c) N. Weferling, S. M. Zhang and C. H. Chiang, Procedia Eng., 2016, 138, 291.
2. (a) S. Hauer, T. M. Horsley Downie, G. Balázs, K. Schwedtmann, J. J. Weigand and R. Wolf, Angew. Chem., Int. Ed., 2024, 63, e202317170; (b) F. Dielmann, M. Elsayed Moussa, C. Riesinger and M. Scheer, Z. Anorg. Allg. Chem., 2024, 650; (c) S. Reichl, E. Mädl, F. Riedlberger, M. Piesch, G. Balázs, M. Seidl and M. Scheer, Nat. Commun., 2021, 12, 5774; (d) S. Reichl, F. Riedlberger, M. Piesch, G. Balázs, M. Seidl and M. Scheer, Chem. Sci., 2023, 14, 7285; (e) M. Peruzzini, L. Gonsalvi and A. Romerosa, Chem. Soc. Rev., 2005, 34, 1038; (f) B. M. Cossairt, N. A. Piro and C. C. Cummins, Chem. Rev., 2010, 110, 4164; (g) M. Caporali, L. Gonsalvi, A. Rossin and M. Peruzzini, Chem. Rev., 2010, 110, 4178; (h) C. M. Hoidn, D. J. Scott and R. Wolf, Chem. Eur J., 2021, 27, 1886; (i) L. Giusti, V. R. Landaeta, M. Vanni, J. A. Kelly, R. Wolf and M. Caporali, Coord. Chem. Rev., 2021, 441, 213927; (j) K. Trabitsch, S. Hauer, K. Schwedtmann, P. Royla, J. J. Weigand and R. Wolf, Inorg. Chem. Front., 2025, 12, 2013; (k) M. Scheer, G. Balázs and A. Seitz, Chem. Rev., 2010, 110, 4236.
3. (a) O. J. Scherer, G. Schwarz and G. Wolmershäuser, Z. Anorg. Allg. Chem., 1996, 622, 951; (b) S. Pelties, D. Herrmann, B. de Bruin, F. Hartl and R. Wolf, Chem. Commun., 2014, 50, 7014; (c) D. W. Agnew, C. E. Moore, A. L. Rheingold and J. S. Figueroa, Angew. Chem., Int. Ed., 2015, 54, 12673; (d) M. J. Drance, S. Wang, M. Gembicky, A. L. Rheingold and J. S. Figueroa, Organometallics, 2020, 39, 3394; (e) M. Haimerl, C. Graßl, M. Seidl, M. Piesch and M. Scheer, Chem. Eur J., 2021, 27, 18129; (f) S. Heinl, S. Reisinger, C. Schwarzmaier, M. Bodensteiner and M. Scheer, Angew. Chem., Int. Ed., 2014, 53, 7639; (g) R. J. Schwamm, M. Lein, M. P. Coles and C. M. Fitchett, Angew. Chem., Int. Ed., 2016, 55, 14798; (h) J. Haberstroh, C. Taube, J. Fidelius, S. Schulz, N. Israel, E. Dmitrieva, R. M. Gomila, A. Frontera, R. Wolf, K. Schwedtmann and J. J. Weigand, Chem. Commun., 2024, 60, 8537; (i) I. Kovács, G. Baum, G. Fritz, D. Fenske, N. Wiberg, H. Schuster and K. Karaghiosoff, Z. Anorg. Allg. Chem., 1993, 619, 453; (j) S. Khan, R. Michel, J. M. Dieterich, R. A. Mata, H. W. Roesky, J.-P. Demers, A. Lange and D. Stalke, J. Am. Chem. Soc., 2011, 133, 17889.
4. (a) Y. Xiong, S. Yao, M. Brym and M. Driess, Angew. Chem., Int. Ed., 2007, 46, 4511; (b) J. W. Dube, C. M. E. Graham, C. L. B. Macdonald, Z. D. Brown, P. P. Power and P. J. Ragogna, Chem. Eur J., 2014, 20, 6739; (c) C. M. E. Graham, C. L. B. Macdonald, P. P. Power, Z. D. Brown and P. J. Ragogna, Inorg. Chem., 2017, 56, 9111; (d) G. Prabusankar, A. Doddi, C. Gemel, M. Winter and R. A. Fischer, Inorg. Chem., 2010, 49, 7976; (e) D. Reiter, P. Frisch, D. Wendel, F. M. Hörmann and S. Inoue, Dalton Trans., 2020, 49, 7060; (f) M. M. D. Roy, A. Heilmann, M. A. Ellwanger and S. Aldridge, Angew. Chem., Int. Ed., 2021, 60, 26550; (g) D. Sarkar, C. Weetman, D. Munz and S. Inoue, Angew. Chem., Int. Ed., 2021, 60, 3519; (h) P. Dabringhaus, T. Heizmann and I. Krossing, Chem. Eur J., 2023, 29, e202302212; (i) D. Holschumacher, T. Bannenberg, K. Ibrom, C. G. Daniliuc, P. G. Jones and M. Tamm, Dalton Trans., 2010, 39, 10590; (j) J. J. Weigand, M. Holthausen and R. Fröhlich, Angew. Chem., Int. Ed., 2009, 48, 295.
5. M. Baudler, C. Adamek, S. Opiela, H. Budzikiewicz and D. Ouzounis, Angew. Chem., Int. Ed., 1988, 27, 1059.
6. (a) M. M. Rauhut and A. M. Semsel, J. Org. Chem., 1963, 28, 471; (b) M. M. Rauhut and A. M. Semsel, J. Org. Chem., 1963, 28, 473; (c) M. M. Rauhut, R. Bernheimer and A. M. Semsel, J. Org. Chem., 1963, 28, 478.
7. K. X. Bhattacharyya, S. Dreyfuss, N. Saffon-Merceron and N. Mézailles, Chem. Commun., 2016, 52, 5179.
8. J. E. Borger, A. W. Ehlers, M. Lutz, J. C. Slootweg and K. Lammertsma, Angew. Chem., Int. Ed., 2014, 53, 12836.
9. J. E. Borger, M. S. Bakker, A. W. Ehlers, M. Lutz, J. C. Slootweg and K. Lammertsma, Chem. Commun., 2016, 52, 3284.
10. J. E. Borger, A. W. Ehlers, M. Lutz, J. C. Slootweg and K. Lammertsma, Angew. Chem., Int. Ed., 2017, 56, 285.
11. J. E. Borger, A. W. Ehlers, M. Lutz, J. C. Slootweg and K. Lammertsma, Angew. Chem., Int. Ed., 2016, 55, 613.
12. J. E. Borger, M. K. Jongkind, A. W. Ehlers, M. Lutz, J. C. Slootweg and K. Lammertsma, ChemistryOpen, 2017, 6, 350.
13. (a) C. G. P. Ziegler, T. M. Maier, S. Pelties, C. Taube, F. Hennersdorf, A. W. Ehlers, J. J. Weigand and R. Wolf, Chem. Sci., 2019, 10, 1302; (b) C. G. P. Ziegler, C. Taube, J. A. Kelly, G. Hierlmeier, M. Uttendorfer, J. J. Weigand and R. Wolf, Chem. Commun., 2020, 56, 14071; (c) Y. Liu, F. F. Westermair, I. Becker, S. Hauer, M. Bodensteiner, C. Hennig, G. Balázs, F. Meyer, R. M. Gschwind and R. Wolf, J. Am. Chem. Soc., 2025, 147, 7083; (d) M. D. Fritz, J. A. Kelly, G. Balázs and R. Wolf, ChemistryEurope, 2026, 4, e202500257.
14. M. P. Blake, N. Kaltsoyannis and P. Mountford, Chem. Commun., 2013, 49, 3315.
15. M. Garçon, C. Bakewell, A. J. P. White and M. R. Crimmin, Chem. Commun., 2019, 55, 1805.
16. (a) H. Felkin, P. J. Knowles, B. Meunier, A. Mitschler, L. Ricard and R. Weiss, J. Chem. Soc., Chem. Commun., 1974, 44; (b) J. A. Kelly, J. Gramüller, R. M. Gschwind and R. Wolf, Dalton Trans., 2021, 50, 13985; (c) M. Gawron, F. Gilch, D. Schmidhuber, J. A. Kelly, T. M. Horsley Downie, A. Jacobi von Wangelin, J. Rehbein and R. Wolf, Angew. Chem., Int. Ed., 2024, 63, e202315381; (d) V. R. Landaeta, T. M. Horsley Downie and R. Wolf, Chem. Rev., 2024, 124, 1323.
17. (a) M. Arrowsmith, M. S. Hill, A. L. Johnson, G. Kociok-Köhn and M. F. Mahon, Angew. Chem., Int. Ed., 2015, 54, 7882; (b) S. Thum, O. P. E. Townrow, J. Langer and S. Harder, Chem. Sci., 2025, 16, 4528.
18. M. J. Evans, D. T. Nguyen, J. M. Parr, J. Mullins, R. Mondal, T. Rajeshkumar, L. Maron and C. Jones, Chem, 2025, 102650.
19. J. Bresien, K. Faust, C. Hering-Junghans, J. Rothe, A. Schulz and A. Villinger, Dalton Trans., 2016, 45, 1998.
20. D. C. Meier, Á. García-Romero, D. González-Pinardo, N. H. Rees, A. Lovstedt, I. Fernández and J. M. Goicoechea, Chem. Sci., 2025, 17, 544.
21. J. Contreras-García, E. R. Johnson, S. Keinan, R. Chaudret, J.-P. Piquemal, D. N. Beratan and W. Yang, J. Chem. Theory Comput., 2011, 7, 625.
22. (a) H. S. Yu, X. He, S. L. Li and D. G. Truhlar, Chem. Sci., 2016, 7, 6278; (b) F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005, 7, 3297; (c) A. V. Marenich, C. J. Cramer and D. G. Truhlar, J. Phys. Chem. B, 2009, 113, 6378.
23. (a) J. J. Turner, M. W. George, M. Poliakoff and R. N. Perutz, Chem. Soc. Rev., 2022, 51, 5300; (b) O. J. Scherer, T. Hilt and G. Wolmershäuser, Organometallics, 1998, 17, 4110; (c) O. J. Scherer, T. Brück and G. Wolmershäuser, Chem. Ber., 1988, 121, 935; (d) M. Scheer and U. Becker, J. Organomet. Chem., 1997, 545–546, 451; (e) O. J. Scherer, R. Winter and G. Wolmershäuser, Z. Anorg. Allg. Chem., 1993, 619, 827; (f) C. Eichhorn, O. J. Scherer, T. Sögding and G. Wolmershäuser, Angew. Chem., Int. Ed., 2001, 40, 2859; (g) O. J. Scherer, J. Schwalb, G. Wolmershäuser, W. Kaim and R. Gross, Angew. Chem., Int. Ed., 1986, 25, 363.
24. R. Grünbauer, C. Schwarzmaier, M. Eberl, G. Balázs and M. Scheer, Inorg. Chim. Acta, 2021, 518, 120234.
25. M. Scheer, S. Deng, O. J. Scherer and M. Sierka, Angew. Chem., Int. Ed., 2005, 44, 3755.
26. R. Yadav, T. Simler, S. Reichl, B. Goswami, C. Schoo, R. Köppe, M. Scheer and P. W. Roesky, J. Am. Chem. Soc., 2020, 142, 1190.
27. J. A. Kelly, V. Streitferdt, M. Dimitrova, F. F. Westermair, R. M. Gschwind, R. J. F. Berger and R. Wolf, J. Am. Chem. Soc., 2022, 144, 20434.
28. A. S. Ionkin, W. J. Marshall, B. M. Fish, A. A. Marchione, L. A. Howe, F. Davidson and C. N. McEwen, Eur. J. Inorg. Chem., 2008, 2386.
29. M. J. Ernst, A. Petrov, M. Weber and C. Müller, ChemistryEurope, 2026, 4, e70252.
30. F. Hennersdorf and J. J. Weigand, Angew. Chem., Int. Ed., 2017, 56, 7858.
31. (a) H. Lei, J. C. Fettinger and P. P. Power, Organometallics, 2010, 29, 5585; (b) S. Wang, M. L. McCrea-Hendrick, C. M. Weinstein, C. A. Caputo, E. Hoppe, J. C. Fettinger, M. M. Olmstead and P. P. Power, J. Am. Chem. Soc., 2017, 139, 6586.
32. (a) G. L. Dunn and J. K. Donohue, Tetrahedron Lett., 1968, 9, 3485; (b) G. O. Schenck and R. Steinmetz, Chem. Ber., 1963, 96, 520.
33. (a) M. Hofmann, C. Höhn, F. W. Heinemann and U. Zenneck, Chem. Eur J., 2009, 15, 5998; (b) R. Bartsch, P. B. Hitchcock and J. F. Nixon, J. Chem. Soc., Chem. Commun., 1989, 1046.
34. C. Schwarzmaier, A. Y. Timoshkin, G. Balázs and M. Scheer, Angew. Chem., Int. Ed., 2014, 53, 9077.
35. (a) W. Huang and P. L. Diaconescu, Chem. Commun., 2012, 48, 2216; (b) S. N. Konchenko, N. A. Pushkarevsky, M. T. Gamer, R. Köppe, H. Schnöckel and P. W. Roesky, J. Am. Chem. Soc., 2009, 131, 5740; (c) M. Scheer, U. Becker and E. Matern, Chem. Ber., 1996, 129, 721; (d) M. E. Barr, B. R. Adams, R. R. Weller and L. F. Dahl, J. Am. Chem. Soc., 1991, 113, 3052.
36. A. D. Phillips, S. Hino and P. P. Power, J. Am. Chem. Soc., 2003, 125, 7520.
37. J. R. Pinkes and A. R. Cutler, Inorg. Chem., 1994, 33, 759.
38. (a) CCDC 2529860: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjdl; (b) CCDC 2529861: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjfm; (c) CCDC 2529862: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjgn; (d) CCDC 2529863: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjhp; (e) CCDC 2529864: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjjq; (f) CCDC 2529865: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjkr; (g) CCDC 2529866: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjls; (h) CCDC 2529867: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjmt; (i) CCDC 2529868: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjnv; (j) CCDC 2529869: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjpw; (k) CCDC 2529870: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjqx; (l) CCDC 2529871: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjry; (m) CCDC 2529872: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjsz; (n) CCDC 2529873: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjt0; (o) CCDC 2529874: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qxjv1; (p) CCDC 2541029: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2r94px.

---

Footnotes

† The Fp*⁻ anion in 2 is more electron-rich, which should favour the insertion product as demonstrated for the reaction of KFp* with CO₂.³⁷

‡ Complex 2 forms an Mg–OC–Fe bonded monomer upon crystallization from DME (Fig. S103) and a dimer with similar Mg–OC–Fe linkages upon crystallization from toluene (Fig. S101). Additionally, monomeric Mg–Fe-bonded structures have been obtained for 5-THF and 6-THF by crystallization from THF (Fig. S99 and S100). In line with previous results by Mountford and co-workers, these data indicate that the aggregation is reversible, depending on the solvent and the crystallization conditions.¹⁴

§ Lammertsma and co-workers have demonstrated that the reaction of LiFp* and P₄ in THF affords a mixture of iron polyphosphides.¹²

---