Skip to main content
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2015 May 26;54(27):7882–7885. doi: 10.1002/anie.201503065

Attenuated Organomagnesium Activation of White Phosphorus**

Merle Arrowsmith 1, Michael S Hill 1, Andrew L Johnson 1, Gabriele Kociok-Köhn 1, Mary F Mahon 1
PMCID: PMC4648029  PMID: 26014162

Abstract

Sequential reactions between a 2,6-diisopropylphenyl-substituted β-diketiminato magnesium n-butyl derivative and P4 allow the highly discriminating synthesis of unusual [nBu2P4]2− and [nBu2P8]2− cluster dianions.

Keywords: cluster compounds, magnesium, P4 activation, phosphorus, structure elucidation


White phosphorus is the starting point in the synthesis of many organophosphanes and phosphorus-containing molecules. As a consequence, the controlled activation of P4 has attracted considerable attention.[1] While notable metal-free examples of P4 activation are provided by Bertrand’s use of nucleophilic N-heterocyclic carbenes for sequential P–P activation of the tetrahedral molecule,[2] the majority of approaches have utilized either reactive transition-metal or main-group reagents. Typical transition-metal-based activation can induce the cleavage of bonds within P4 and also the aggregation of smaller fragments to form Pn (n>4) cluster anions. Cummins, for example, has shown that disproportionation of a niobium(IV) tris(enolate) in the presence of P4 provides the {P8} cluster compound 1,[3] while in a very recent report, Bergman and Arnold have described a trinuclear β-diketiminato μ-P12 niobium complex, which was isolated as one of multiple minor by-products during the reaction of the NbV complex (DippBDI)Nb(NtBu)Me2 (DippBDI=HC{C(Me)2N(2,6-iPr2C6H3)}2) with H2 in the presence of P4.[4] Notably, this latter reaction gave a compound containing a cyclo-P4 dianion as its predominant reaction product and, in some respects, the irrationality of the formation of both of these compounds is typical of the synthesis of high-nuclearity {Pn} clusters.[1f]

Inline graphic

Reductive chemistry to cyclo-P4 anions has also been described through the formal oxidative addition of P4 to low-oxidation-state Group 13 and 14 derivatives (for example, compounds 2 and 3),[1c, 5, 6] and there is precedent for the activation of P4 through its interaction with transition-metal hydrides and alkyls.[7] Well-defined reactions with anionic Group 14 derivatives of elements from Groups 1 and 2 are, however, uncommon and dominated by the use of highly sterically demanding carbon or silicon reagents. Wright and co-workers described the synthesis of the [R2P8]2− cluster dianion (4) derived from a {P7}3− Zintl ion core through the reaction of the very bulky hypersilyl potassium reagent with P4.[8] The disubstituted butterfly compound 5 (Mes*=2,4,6-tBu3C6H2) and the diphosphene Mes*P=PMes* were similarly isolated from the reaction of Mes*Li and P4 in the presence of Mes*Br.[9] Production of compound 5 presumably occurs through initial addition of Mes*Li across a P–P bond within the P4 unit and the intermediacy of the consequent organopolyphosphorus anion. The existence of species of this latter type has recently been confirmed by Lammertsma and co-workers, who reported that the B(C6F5)3-stabilized lithium bicyclo[1.1.0]tetraphosphabutane species 6 a and 6 b could be isolated from reactions of P4 and sterically demanding lithium aryls in the presence of the borane Lewis acid.[10] Although similar but indiscriminate reactivity was posited some 50 years ago during reactions with simple Grignard reagents,[11] the activation of white phosphorus by nucleophilic Group 2 reagents is even less widespread than that by the alkali metals. Our own research interests lie in the development of a defined bond activation and homogeneous catalytic chemistry for the earth-abundant and environmentally benign alkaline earth elements.[12] As a first step toward the functionalization of white phosphorus catalyzed by a Group 2 metal, we present in this contribution the results of our initial study, which indicate that remarkable kinetic control during P4 activation is achievable with organomagnesium reagents (Scheme 1). Key to the success of this chemistry is the application of the sterically demanding DippBDI supporting ligand, which was employed in the syntheses of Arnold and Bergman’s aforementioned niobium systems,[4] compounds 2 and 3, and very recently described binuclear cobalt(I) and copper(I) species containing a P4-derived tetraphospacyclobutadiene and an unactivated P4 ligand respectively.[13]

Scheme 1.

Scheme 1

Synthesis of the complexes described herein.

An initial reaction between an equimolar quantity of P4 and the β-diketiminato n-butylmagnesium complex, [(DippBDI)MgnBu] (7), resulted in the smooth consumption of the magnesium starting material to provide a single new species (8) characterized by two mutually coupled doublet resonances at δ −25.6 and −195.5 ppm (1JP-P=99.6 Hz) in the resultant 31P{1H} NMR spectrum. Notably, although compound 8 formed with only 50 % consumption of P4, some evidence of further reaction was apparent on continued monitoring (see below). Repetition of this reaction with the reagent quantities adjusted to reflect the apparent 2:1 reaction stoichiometry resulted in the complete consumption of both P4 and 7 and the sole production of compound 8, which also comprised a single set of temperature-invariant (from −90 to +90 °C) n-butyl and (DippBDI) ligand environments in the corresponding 1H and 13C NMR spectra. The origin of these observations was resolved through the isolation of colorless single crystals of compound 8 suitable for X-ray diffraction analysis from an n-hexane solution at low temperature, the results of which are displayed in Figure 1. The molecular structure of 8 comprises a dinuclear magnesium complex of an unprecedented nonplanar [nBu2P4]2− dianion, which may be considered as arising from the formal addition of two magnesium n-butyl fragments to, and the consequent activation of two P-P single bonds within, the P4 tetrahedron. The P–C bonds to the n-butyl chains are located on the adjacent P1 and P2 atoms, while the magnesium centers each coordinate to one of these alkylated phosphorus atoms and the phosphorus atom located at the opposite diagonal corners of the tetraphosphorus unit. The adoption of a square array is prevented by a distinct pyramidalization at each phosphorus center such that the P-P-P angles are compressed to about 80°. The P3–P4 bond length [2.2647(7) Å] is significantly elongated in comparison to the interatomic separation between the alkylated phosphorus atoms [P1–P2 2.1762(7) Å] but is comparable to the P–P distances observed in molecules containing the [P4]4− anion, such as compound 2 [mean 2.29 Å].[5] The Mg–P distances to P3 and P4 [Mg1–P3 2.5625(7), Mg2–P4 2.5592(7) Å] are also shorter than those to P1 and P2 [Mg1–P1 2.6336(7), Mg2–P2 2.6652(7) Å] and we interpret both of these structural features to indicate that the distribution of negative charge is polarized toward the non-alkylated P3 and P4 centers and that the [nBu2P4]2− unit as a whole may be considered as derived from the formal addition of two n-butyl carbenium ions to the [P4]4− constituent anion of compound 2.

Figure 1.

Figure 1

ORTEP representation (30 % probability ellipsoids) of compound 8.[17] Hydrogen atoms and isopropyl groups were removed for clarity. Selected bond lengths [Å] and angles [°]: Mg1–N1 2.0462(15), Mg1–N2 2.0465(15), Mg1–P3 2.5625(7), Mg1–P1 2.6336(7), Mg1–P4 3.1607(7), Mg2–N4 2.0424(15), Mg2–N3 2.0466(15), Mg2–P4 2.5592(7), Mg2–P2 2.6652(7), P1–C59 1.8584(19), P1–P2 2.1762(7), P1–P4 2.2127(7), P2–C63 1.8575(19), P2–P3 2.2108(7), P3–P4 2.2647(7); N1-Mg1-P3 126.57(5), N2-Mg1-P3 122.72(5), N1-Mg1-P1 127.76(5), N2-Mg1-P1 122.16(5), P3-Mg1-P1 67.14(2), N1-Mg1-P4 167.48(5), N2-Mg1-P4 99.06(5), P3-Mg1-P4 45.131(17), P1-Mg1-P4 43.731(16), N4-Mg2-N3 93.52(6), N4-Mg2-P4 126.28(5) N3-Mg2-P4 121.83(5), N4-Mg2-P2 121.40(5), N3-Mg2-P2 130.79(5), P4-Mg2-P2 66.130(19).

The previously noted onward reaction of compound 8 with additional P4 (see above) prompted us to carry out a further reaction of an equimolar quantity of P4 and compound 8. The reaction in benzene at room temperature provided the slow and simultaneous consumption of the starting materials and direct conversion to a single new compound (9), which is characterized by the emergence of eight mutually coupled signals in the 31P NMR spectrum. The corresponding 1H and 13C NMR spectra indicated that this new compound comprised two differentiated groups of resonances indicative of two separate n-butyl and (DippBDI) ligand environments. Compound 9 could also be synthesized directly through the slow (7 days) reaction of compound 7 with an equimolar amount of P4 at room temperature or by heating of a similar reaction at 60 °C for 2 days. In this latter case, compound 9 crystallized as colorless single crystals suitable for a further X-ray diffraction analysis, allowing the resolution of the solution NMR spectroscopic observations. The result of this analysis is shown in Figure 2 and highlights that the disposition of the phosphorus and magnesium centers within compound 9 is consistent with the coupling pattern observed in the solution-state 31P NMR spectrum and the multiplicity of signals arising in the 1H and 13C NMR data. The structure comprises two (DippBDI) magnesium units bonded to a [nBu2P8]2− cluster dianion derived from the formal reductive coupling of two neutral P4 units. Although the rationality of this {P8} cluster synthesis from the combination of two {P4} fragments appears to be unique, its structure may be considered as a heptaphosphanorbornane fused to an exo-oriented cyclotriphosphane ring and a constitutional isomer of the {P8} cage within the niobium(V) species [Ph2CP8Nb(OC[2Ad]Mes)3], which was derived through treatment of compound 1 with benzophenone.[14] The single tris(enolato) niobium dication and the {Ph2C} unit of this previously described species is replaced in the present case by two (DippBDI) magnesium monocations and two n-butyl substituents. The magnesium centers contact the {P8} cage structure in endo [Mg1] and exo [Mg2] configurations with respect to the P4 bridgehead and provide Mg–P distances of 2.575(3) Å [P4–Mg2], 2.625(3) Å [P8–Mg2], 2.594(3) Å [P2–Mg1] and 2.831(3) Å [P7–Mg1]. Although these latter values indicate some variability in the charge donation across the cage structure, it should be noted that the Mg–P contacts are all within the range expected [2.45 to 2.99 Å] from a search of the Cambridge structural database.[15] All of the P–P distances within this new cluster are also consistent with P–P single bonds and range between 2.096(3) Å [P3–P4] and 2.262(4) Å [P5–P6].

Figure 2.

Figure 2

ORTEP representation (30 % probability ellipsoids) of compound 9.[17] Hydrogen atoms, isopropyl methyl groups and n-butyl groups aside from the P-bonded C59 and C63 atoms were removed for clarity. Selected bond lengths [Å] and angles [°]: P1–P5 2.195(3), P2–P3 2.160(2), P2–Mg1 2.594(3), P3–P4 2.096(3), P3–P7 2.186(3), P4–P5 2.222(3), P4–Mg2 2.575(3), P5–P6 2.262(4), P6–P7 2.170(4), P6–P8 2.185(3), P7–P8 2.153(3), P7–Mg1 2.831(3), P8–Mg2 2.625(3) Mg1–N2 2.046(5); P2-P1-P5 105.23(12), P1-P2-P3 96.79(11), P1-P2-Mg1 90.77(11) P3-P2-Mg1 97.47(9), P4-P3-P7 113.74(12), P2-P3-P7 97.95(10), P3-P4-P5 88.12(10), P3-P4-Mg2 101.38(10), P5-P4-Mg2 91.12(11), P1-P5-P4 102.66(14), P1-P5-P6 92.00(12), P4-P5-P6, 105.80(12), P7-P6-P8 59.27(11), P7-P6-P5 106.69(12).

Subsequent preliminary investigations indicated that the kinetic control exerted during the syntheses of compounds 8 and 9 is an apparent consequence of the steric demands of both the supporting (DippBDI) platform and the reactive n-butyl co-ligand. Analysis by 31P NMR spectroscopy of a reaction between [(DippBDI)MgH]2 and half a molar equivalent of P4 provided a mixture of several phosphorus-containing species. Although a minor component comprised broadened resonances at δ −3.5 and −159 ppm reminiscent of those observed for compound 8, the major product of this reaction was characterized by a triplet signal observed at δ −284.8 ppm (1JP-H=45 Hz). This latter species was identified as the trimeric primary magnesium phosphide, compound 10, through a further X-ray analysis performed on single crystals produced by fractional crystallization of the reaction mixture (Figure 3 a). Analytically pure samples of compound 10, notable as the first primary phosphide of a Group 2 element to have been structurally characterized, were found to be unstable once redissolved in [D8]toluene. Monitoring by 1H and 31P NMR spectroscopy at room temperature evidenced the onset of a redistributive process to as yet unidentified magnesium-containing species and the production of PH3, which was observed as a binomial quartet at δ −245.0 ppm in the 31P NMR spectrum.

Figure 3.

Figure 3

ORTEP representations (30 % probability ellipsoids) of a) compound 10 with hydrogen atoms (except those attached to P1, P2, and P3) and isopropyl methyl groups removed for clarity, and b) compound 11 with hydrogen atoms and mesityl methyl groups removed for clarity.[17]

In a similar manner, monitoring by NMR spectroscopy of a reaction analogous to that employed in the synthesis of compound 8, but with replacement of the N-{2,6-iPr2C6H3} groups of the (DippBDI) ligand of 7 by less sterically demanding N-{2,4,6-Me3C6H2} substituents (MesBDI), indicated less discriminating behavior and reactivity, which again resulted in the formation of multiple products. A further reaction performed between [(MesBDI)MgnBu] and P4 in a 2:3 molar ratio also provided a complex mixture containing at least three new phosphorus-containing compounds. Fractional crystallization from this solution and a resultant single crystal X-ray diffraction analysis of compound 11 (Figure 3 b) provided some insight into the nature of these processes. Compound 11 may be viewed as a [P7]3− Zintl ion cage reminiscent of compound 4 but decorated with three β-diketiminato magnesium units. We suggest, therefore, that the formation of this phosphorus cage constitutes a thermodynamic sink during the solution comproportionation of magnesium polyphosphorus species of varying nuclearity.[16]

In conclusion we have shown that the controlled, sequential organomagnesium-mediated activation of white phosphorus may be achieved through selection of an appropriate combination of supporting and reactive co-ligands. We are continuing to explore the parameters of this reactivity and will report our findings in due course.

Supporting Information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

miscellaneous_information

anie0054-7882-sd1.pdf (533.3KB, pdf)

References

  • 1a.Scheer M, Balázs G, Seitz A. Chem. Rev. 2010;110:4236–4256. doi: 10.1021/cr100010e. For reviews, see. [DOI] [PubMed] [Google Scholar]
  • 1b.Cossairt BM, Piro NA, Cummins CC. Chem. Rev. 2010;110:4164–4177. doi: 10.1021/cr9003709. [DOI] [PubMed] [Google Scholar]
  • 1c.Khan S, Sen SS, Roesky HW. Chem. Commun. 2012;48:2169–2178. doi: 10.1039/c2cc17449a. [DOI] [PubMed] [Google Scholar]
  • 1d.Serrano-Ruiz M, Romerosa A, Lorenzo-Luis P. Eur. J. Inorg. Chem. 2014:1587–1598. [Google Scholar]
  • 1e.Caporali M, Gonsalvi L, Rossin A, Peruzzini M. Chem. Rev. 2010;110:4178–4235. doi: 10.1021/cr900349u. [DOI] [PubMed] [Google Scholar]
  • 1f.Ehses M, Romerosa A, Peruzzini M. Top. Curr. Chem. 2002;220:107–140. [Google Scholar]
  • 2a.Masuda JD, Schoeller WW, Donnadieu B, Bertrand G. Angew. Chem. Int. Ed. 2007;46:7052–7055. doi: 10.1002/anie.200703055. [DOI] [PubMed] [Google Scholar]
  • Angew. Chem. 2007;119 [Google Scholar]
  • 2b.Back O, Kuchenbeiser G, Donnadieu B, Bertrand G. Angew. Chem. Int. Ed. 2009;48:5530–5533. doi: 10.1002/anie.200902344. [DOI] [PubMed] [Google Scholar]
  • Angew. Chem. 2009;121 [Google Scholar]
  • 2c.Martin CD, Weinstein CM, Moore CE, Rheingold AL, Bertrand G. Chem. Commun. 2013;49:4486–4488. doi: 10.1039/c3cc42041h. [DOI] [PubMed] [Google Scholar]
  • 3a.Cossairt BM, Cummins CC. Angew. Chem. Int. Ed. 2008;47:169–172. doi: 10.1002/anie.200704354. [DOI] [PubMed] [Google Scholar]
  • Angew. Chem. 2008;120 [Google Scholar]
  • 3b.Cossairt BM, Cummins CC. Angew. Chem. Int. Ed. 2008;47:8863–8866. doi: 10.1002/anie.200803971. [DOI] [PubMed] [Google Scholar]
  • Angew. Chem. 2008;120 [Google Scholar]
  • 4.Camp C, Maron L, Bergman RG, Arnold J. J. Am. Chem. Soc. 2014;136:17652–17661. doi: 10.1021/ja5107282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Peng Y, Fan H, Zhu H, Roesky HW, Magull J, Hughes C. Angew. Chem. Int. Ed. 2004;43:3443–3445. doi: 10.1002/anie.200353406. [DOI] [PubMed] [Google Scholar]
  • Angew. Chem. 2004;116 [Google Scholar]
  • 6.Xiong Y, Yau S, Brym M, Driess M. Angew. Chem. Int. Ed. 2007;46:4511–4513. doi: 10.1002/anie.200701203. [DOI] [PubMed] [Google Scholar]
  • Angew. Chem. 2007;119 [Google Scholar]
  • 7.Barbaro P, Ienco A, Mealli C, Peruzzini M, Scherer OJ, Schmitt G, Vizza F, Wolmershäuser G. Chem. Eur. J. 2003;9:5195–5210. doi: 10.1002/chem.200305091. See, for example. [DOI] [PubMed] [Google Scholar]
  • 8.Chan WTK, García F, Hopkins AD, Martin LC, McPartlin M, Wright DS. Angew. Chem. Int. Ed. 2007;46:3084–3086. doi: 10.1002/anie.200604267. [DOI] [PubMed] [Google Scholar]
  • Angew. Chem. 2007;119 [Google Scholar]
  • 9.Riedel R, Hausen H-D, Fluck E. Angew. Chem. Int. Ed. Engl. 1985;24:1056–1057. [Google Scholar]
  • Angew. Chem. 1985;97 [Google Scholar]
  • 10.Borger JE, Ehlers AW, Lutz M, Slootweg JC, Lammertsma K. Angew. Chem. Int. Ed. 2014;53:12836–12839. doi: 10.1002/anie.201405879. [DOI] [PubMed] [Google Scholar]
  • Angew. Chem. 2014;126 [Google Scholar]
  • 11a.Rauhut MM, Semsel AM. J. Org. Chem. 1963;28:471–473. [Google Scholar]
  • 11b.Rauhut MM, Semsel AM. J. Org. Chem. 1963;28:473–478. [Google Scholar]
  • 12a.Arrowsmith M, Hill MS. In: Comprehensive Inorganic Chemistry II, Vol. 1. Chivers T, editor. Amsterdam: Elsevier; 2013. p. 1189. “Alkaline Earth Chemistry: Applications in Catalysis”: For reviews, see. [Google Scholar]
  • 12b.Barrett AGM, Crimmin MR, Hill MS, Procopiou PA. Proc. R. Soc. London Ser. A. 2010;466:927. [Google Scholar]
  • 12c.Crimmin MR, Hill MS. In: Topics in Organometallic Chemistry, Vol. 45. Harder S, editor. Berlin: Springer; 2013. p. 191. [Google Scholar]
  • 12d.Harder S. Chem. Rev. 2010;110:3852. doi: 10.1021/cr9003659. [DOI] [PubMed] [Google Scholar]
  • 13a.Yao S, Lindenmaier N, Xiong Y, Inoue S, Szilvási T, Adelhardt M, Sutter J, Meyer K, Driess M. Angew. Chem. Int. Ed. 2015;54:1250–1254. doi: 10.1002/anie.201409469. [DOI] [PubMed] [Google Scholar]
  • Angew. Chem. 2015;127 [Google Scholar]
  • 13b.Spitzer F, Sierka M, Latronico M, Virovets AV, Scheer M. Angew. Chem. Int. Ed. 2015;54:4392–4396. doi: 10.1002/anie.201411451. [DOI] [PubMed] [Google Scholar]
  • Angew. Chem. 2015;127 [Google Scholar]
  • 14.Cossairt BM, Cummins CC. Inorg. Chem. 2008;47:9363–9371. doi: 10.1021/ic8009282. [DOI] [PubMed] [Google Scholar]
  • 15. CSD search performed 31/3/15.
  • 16.Turbervill RSP, Goicoechea JM. Chem. Rev. 2014;114:10807–10828. doi: 10.1021/cr500387w. [DOI] [PubMed] [Google Scholar]
  • 17. CCDC 1057657 (8), 1057658 (9), 1057659 (10), and 1057660 (11) contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via http://www.ccdc.cam.ac.uk/data_request/cif.

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

miscellaneous_information

anie0054-7882-sd1.pdf (533.3KB, pdf)

Articles from Angewandte Chemie (International Ed. in English) are provided here courtesy of Wiley

RESOURCES