Diaminophosphinoboranes: effective reagents for phosphinoboration of CO₂

Abstract

The monomeric diaminophosphinoboranes readily react with CO₂ under mild conditions to cleanly form products of the general formula in the absence of a catalyst. The isolated products from the CO₂-phosphinoboration were fully characterized by NMR spectroscopy, IR spectroscopy, and X-ray diffraction. The mechanism of CO₂ phosphinoboration with diaminophosphinoboranes was elucidated by DFT calculations.

---

Introduction

Increasingly stringent environmental standards are prompting us to look for clean and efficient methods for the synthesis of organic compounds. In recent years, much attention has been focused on reactions employing simple, nonmetallic P,B-based^1–6 systems that can activate waste gases such as CO,^7,8 CO₂,^9–13 NO_x,^10,14 SO₂,^14,15 or even H₂.^5,9,16–19 These metal-free reactions not only involve easily accessible and inexpensive gaseous reagents as the feedstock but also enable the straightforward and efficient synthesis of organic compounds through the functionalization of small inorganic molecules.^20–26 The resulting products can be directly used as substrates for the synthesis of more complex systems that are difficult or impossible to obtain by other means.

The largest portion of reports are those devoted to CO₂ ^{11,12,17,22,27} and its incorporation into organic molecules under mild conditions. While highly nucleophilic phosphines form stable adducts with CO₂,¹¹ boranes capture CO₂ only in the presence of a Lewis base¹² that can act as either a stoichiometric^4,27 or catalytic²³ coreagent or as the site of the ambiphilic molecule. One of the most common systems of this kind is frustrated Lewis pairs (FLPs), involving inter- or intramolecular combinations of sterically encumbered Lewis acids and bases that cannot quench each other.¹⁸ In the reaction of P,B-based FLPs with CO₂, the synergistic interactions of the P-center with a carbon atom and the B-center with an oxygen atom lead to the formation of P–C and B–O bonds, respectively, giving zwitterionic products with the formula .²⁷ In the vast majority of FLPs, the presence of highly electron-withdrawing substituents to increase the acidity of the B atom is crucial for efficient CO₂ fixation. However, it was shown that geminal FLPs, in which the donor and acceptor sites are separated by one atom, can activate CO₂ despite the mild Lewis acidity of the boron center.^28,29

These reports inspired us to go one step further and investigate whether systems in which the B atom is directly bound to the phosphorus atom also activate CO₂. Hence, we focused on monomeric phosphinoboranes, which are another type of ambiphilic P,B-based molecules.^30,31 A recent very comprehensive review³¹ by Pringle et al. classifies these species containing a single P–B bond with a pyramidal P atom and the general formula R₂P–BR₂ as borylphosphines^30,32,33 and those with a double P=B bond and trigonal planar P atom and the general formula as phosphinoboranes.^34–36 Notably, we found that there are a limited number of reports on the reactivity of P–B bond systems towards small molecules. Phosphinoboranes R₂P=B(C₆F₅)₂ (R = tBu, Cy) exhibit FLP-like reactivity towards H₂.^34,37 The Westcott group explored the chemistry of phosphinoboronate ester Ph₂P–Bpin³⁸ (pin = 1,2-O₂C₂Me₄), which effectively functionalizes a wide variety of heterorganic systems, such as carbonyls,^38,39 N-heterocycles,⁴⁰ aldimines,^38,39 carbodiimides⁴¹ and isocyanates.⁴¹ Very recently Wescott and Stephan tested reactivity of R₂PBpin (R = Ph, tBu), R₂PBMes₂ (R = Ph, tBu), and R₂PBcat (R = Ph, tBu, Mes) towards CO₂, where the reactions with the first and the second group of phosphinoboranes gave R₂PCO₂Bpin and R₂PCO₂BMes respectively, whereas the reaction involving R₂PBcat yielded (R₂P)₂CO and O(Bcat)₂.⁴²

As a part of our research program on applying P–P bond systems to the activation of small molecules, we recently reported the first example of CO₂ diphosphination by unsymmetrical diphosphanes⁴³ in the presence of BPh₃. Herein, a weak Lewis acid catalyzes the insertion of CO₂ into the P–P bond of the Lewis basic component with the formation of P–C and P–O bonds in a reversible manner (Scheme 1).⁴⁴

Scheme 1 Diphosphination of CO₂ by unsymmetrical diphosphanes.⁴⁴

Our studies revealed that the presence of long and polarized P–P bonds is crucial for the activation of CO₂ by diphosphanes. Herein, we decided to apply this synthetic approach to trivalent phosphorus and boron compounds with direct P–B bonds.

Results and discussion

To this end, we designed and synthesized a series of new phosphinoboranes 1–3, which are structural analogs of unsymmetrical diphosphanes where one P atom has been replaced by a B atom (Scheme 2). The first step of the synthesis was the reaction of BBr₃ with a four-fold excess of iPr₂NH in petroleum ether with the elimination of an ammonium salt and formation of the bromo(diamino)borane. Then, the obtained product was used in the equimolar reaction with the corresponding lithium phosphide (RR′PLi) in toluene at −50 °C. Diaminophosphinoboranes 1–3 were isolated by removal of the LiBr followed by evaporation of the solvent, giving analytically pure products in high yields (83–88%).

Scheme 2 Synthesis of diaminophosphinoboranes 1–3.

The ³¹P{¹H} NMR spectra of 1, 2 and 3 show broad singlets at −8.0 ppm, −26.2 ppm and −36.1 ppm, respectively. The upfield resonances in the ³¹P{¹H} NMR spectra of 1–3 indicate a lack of significant P–B π-interaction within these molecules. The ¹¹B NMR spectra of 1–3 each exhibit only one broad singlet (39.5 ppm (1); 40.4 ppm (2); and 38.8 ppm (3)) at values typical for trivalent species.^30,31 The broadness of the signals in the ³¹P{¹H} and ¹¹B spectra can be explained by the quadrupolar nature of boron. Moreover, the ¹J_P–B coupling is not visible likely because of the small value of this coupling constant and mentioned broadness of the signals. The crystallization from petroleum ether at −20 °C resulted in colorless crystals of 2 and 3. The X-ray structures of 2 and 3 are presented in Fig. 1. The most characteristic structural features of 2 and 3 are the planar geometry of the N and B atoms with the sum of the angles being close to 360°, the pyramidal geometry of the P atom (sum 323.8° (2); 316.95 (3)) and the very long B–P distance (1.983(2) Å (2); 1.984(1) Å (3)). The B–P bond lengths in 2 and 3 are among the longest distances reported for phosphinoboranes,^30,31 and they are even slightly longer than the sum of the single covalent bond radii of B and P (1.96 Å).⁴⁵

Fig. 1 Molecular structures of 2 and 3.

Moreover, the shortening of the B–N bonds in 2 and 3 to B–N distances in the range of 1.425(1)–1.442(3) Å (sum of the single covalent bond radii of B and N: 1.56 Å; sum of the double covalent bond radii for B and N: 1.38 Å) was observed. ^45,46 This feature, together with the planar geometry of the N atom, indicates significant B–N π-bonding. NBO analysis of these species provided further insight into the electronic structures of 1–3. The strong interaction between the lone pairs on the N atoms and the formally empty p-orbital of the B atom resulting from the formation of B–N π-bonds is clearly visible. Otherwise, donor–acceptor interactions between the P and B centers are very weak. Hence, the lone pair on the P atom is accessible for reactions with electrophilic compounds. The nucleophilic and electrophilic properties of 1–3 were studied by analysis of the condensed Fukui functions. In general, 1–3 have strong nucleophilic character with the nucleophilicity centered at the P atom. The nucleophilicity of the diaminophosphinoboranes increases in the order 3 < 2 < 1, where the values of f_N for the P atoms are 0.175, 0.198 and 0.277, respectively. Interestingly, boron is neither a nucleophilic nor an electrophilic center; the values of both the f_N and f_E Fukui functions are close to zero. Taking into account the results of all the structural and NBO analyses, 1–3 can be described as nucleophilic borylphosphine-like species.

Next, we studied the reactivity of diaminophosphinoboranes 1–3 towards CO₂. The reactions of 1, 2 or 3 with CO₂ (1 atm) at room temperature in toluene resulted in the formation of phosphinoboration products 1a, 2a and 3a, respectively (Scheme 3).

Scheme 3 Phosphinoboration of CO₂.

The reaction progress was monitored by ³¹P{¹H} and ¹¹B NMR spectroscopy. The complete conversions of 1 into 1a and 2 into 2a took 3 hours. In contrast to experiments involving 1 and 2, the reaction of less nucleophilic 3 with CO₂ is much slower and was complete after 65 days.

The analytically pure 1a, 2a, and 3a were isolated in high yields (87–94%) by evaporation of the solvent as colorless oils solidified below −20 °C. The ³¹P{¹H} spectra of 1a, 2a and 3a exhibit sharp singlets at 52.4 ppm, 23.5 and −1.0 ppm, respectively. The signals are strongly downfield shifted in comparison to the corresponding resonance of the parent diaminophosphinoboranes, indicating a change in the coordination environment of the P atom. 1a, 2a, and 3a have very similar ¹¹B NMR spectra, which show one resonance at a shift of approximately 27 ppm. Furthermore, the ¹³C{¹H} spectra of the reaction products show a downfield doublet in the range of 180.2–177.9 ppm with ¹J_PC (30.9–7.3 Hz). Notably, the ³¹P{¹H} and ¹³C{¹H} NMR data of 3a are very similar to those reported for Ph₂P–C(O)–O–Bpin.⁴¹ Moreover, the IR data for 1a and 2a showed absorption bands at 1644 cm⁻¹ and 1664 cm⁻¹, respectively, which are very characteristic for carbonyl groups.

X-ray diffraction studies of 1a and 2a unambiguously confirmed that the CO₂ molecule is incorporated between the B and P atoms (Fig. 2). The X-ray structures of 1a and 2a are very similar and will be discussed collectively. The CO₂ moiety is bound to the phosphanyl group via a carbon atom, whereas one oxygen atom links this moiety with the boryl group. In comparison to parent compounds 1 and 2, the planar geometries of the B and N atoms and the pyramidal geometry of the P atom are retained. The geometry of C1 is also planar. The P1–C1 (1.863(1) Å (1a), 1.851(5) Å (2a)) and B1–O1 (1.459(2) Å (1a), 1.452(6) Å (2a)) bonds values are consistent with single covalent bonds.⁴⁵ The P1–C1 distances in 1a and 2a are very similar to corresponding bond distances in Ph₂PCO₂Bpin and Ph₂PCO₂BMes₂, however B1–O1 bond lengths in 1a and 2a are about 0.05 Å longer than in mentioned compounds.⁴² The C1–O1 (1.345(2) Å (1a), 1.351(5) Å (2a)) and C2=O2 (1.215(2) (1a), 1.218(6) Å (2a)) bond distances are very similar to those observed for carboxylic esters (∼1.33–1.41 Å and ∼1.19–1.20 Å, respectively).⁴⁷ The B–N bond lengths in 1a and 2a are in the range of 1.408(1)–1.438(2) Å and are slightly shorter than the corresponding distances in parent species 1 and 2. The structures of 1a and 2a differ significantly from known structures of CO₂ activation products; typically, inter- and intramolecular P,B-based frustrated Lewis pairs form zwitterionic adducts containing tetra-coordinated P and B centers.²⁷ Compounds 1a, 2a, and 3a are stable under argon or vacuum, and even heating these products up to 60 °C did not regenerate the parent diaminophosphinoboranes.

Fig. 2 Molecular structures of 1a and 2a.

To investigate the mechanism of CO₂ capture by the diaminophosphinoboranes and elucidate the differences in their reactivities, we carried out DFT calculations. According to the Gibbs energy profile, the insertion of carbon dioxide into the P–B bond of 1 proceeds via a simple two-step mechanism (Fig. 3).

Fig. 3 The Gibbs energy profile of the formation of 1a.

The reaction starts with a nucleophilic attack of P on the electrophilic CO₂ atom. Upon attachment of C to the PtBu₂ atom, the electron density shifts towards the PCO₂ moiety; the value of the B(iPr₂N)₂ electrophilic Fukui function, f_E, increases from 0.001 for 1 to 0.152 for I (Table S10†). Consequently, the B(iPr₂N)₂ atom becomes the electrophilic center of the molecule, facilitating the formation of the B–O bond, and the simultaneous cleavage of the P–B bond leads to the generation of 1a. Although the values of the free energies (ΔG₂₉₈) confirmed that in the reaction with CO₂, all three systems form stable products via exergonic processes, the kinetics of these reactions are notably different (Table 1).

Table 1 Free energies (ΔG) and activation free energies (ΔG^‡) for the capture of CO₂ by 1–3^a

Reaction	ΔG^‡ [kcal mol^−1]	ΔG [kcal mol^−1]
a ωB97XD/6-31G+(d,p).
1a	27.3	−20.1
2a	26.4	−17.8
3a	30.2	−10.8

---

In general, as the nucleophilicity of the phosphorus atom in 1–3 decreases, the energy barrier ΔG^‡ increases. A small aberration is observed for 2a; PtBuPh is less nucleophilic than PtBu₂ but is also less sterically hindered, and therefore, the value of ΔG^‡ is slightly lower for 2a (Fig. S41†). While 1 and 2 react with CO₂ in an analogous manner involving the generation of adduct I, a PES scan of 3a revealed that the corresponding intermediate is not formed along the reaction path (Fig. S42†). In this case, the activation of CO₂ proceeds via a single four-membered ring transition state. This path requires the simultaneous interaction of both reactive centers with CO₂, which in conjunction with the PPh₂ atom being the least nucleophilic, justifies the highest energy barrier and the extremely long reaction time for 3.

We found it very interesting to compare the reactivities of trivalent species containing single and double boron-phosphorus bonds (Scheme 4).

Scheme 4 Comparison of the reactivities of species possessing P–B or P=B bonds towards H₂ and CO₂.

Stephan and coworkers revealed that phosphinoboranes R₂P=B(C₆F₅)₂ (R = tBu, Cy) can activate dihydrogen^34,37 (Scheme 4, bottom left); however, we did not find any reports on the reactivity of such species towards CO₂.

For this reason, we reacted a representative species, tBu₂P=B(C₆F₅)₂ ^34,37 (A), with CO₂ under the same conditions as described for 1. The monitoring of the reaction mixture by ³¹P{¹H} and ¹¹B spectroscopy showed that compound A does not activate CO₂. Furthermore, in the reaction of 1 with H₂ conducted under the same conditions, the ³¹P{¹H} and ¹¹B NMR spectra revealed only signals of unreacted 1. The differences in the reactivity of 1 and tBu₂P=B(C₆F₅)₂ towards H₂ and CO₂ can be explained by their electronic structures. According to the mechanism of H₂ activation proposed by Stephan, the reaction starts with the addition of H₂ (H–H bond acts as a Lewis base) to the Lewis acidic B-center of tBu₂P=B(C₆F₅)₂.^34,37 In the case of 1, the Lewis acidity of boron is quenched due to the strong donation from both N atoms, which explains its lack of reactivity towards H₂. In regard to CO₂, our mechanistic study reveals that the presence of a strong nucleophilic P center with an accessible lone pair is crucial for the activation of electrophilic CO₂. While compound 1 meets these conditions, in the case of tBu₂P=B(C₆F₅)₂, the lone pair on the P atom is involved in a strong donor–acceptor interaction with the B atom, resulting in π-bonding.

Conclusions

We have synthesized a series of novel monomeric diaminophosphinoboranes which are not only effective species for functionalization of CO₂ but are also promising systems for activation of other small molecules. Our studies revealed which factors are crucial for the activation of small molecules by trivalent phosphorus and boron species with direct P–B bond and how to design effective systems of this kind. Studies on the reactivity of diaminophosphinoboranes towards a wide range of small, electrophilic molecules are currently in progress.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

R. G., N. S., and A. O. thank the National Science Centre NCN, Poland (Grant 2016/21/B/ST5/03088) for their financial support. The authors thank TASK Computational Center for access to computational resources.

Notes and references

1. F. G. Fontaine, M. A. Courtemanche, M. A. Légaré and É. Rochette, Coord. Chem. Rev., 2017, 334, 124–135.
2. D. J. Scott, M. J. Fuchter and A. E. Ashley, Chem. Soc. Rev., 2017, 46, 5689–5700.
3. L. Greb, P. Oña-Burgos, B. Schirmer, S. Grimme, D. W. Stephan and J. Paradies, Angew. Chem., Int. Ed., 2012, 51, 10164–10168.
4. D. W. Stephan, J. Am. Chem. Soc., 2015, 137, 10018–10032.
5. S. Mummadi, D. K. Unruh, J. Zhao, S. Li and C. Krempner, J. Am. Chem. Soc., 2016, 138, 3286–3289.
6. D. W. Stephan, Acc. Chem. Res., 2015, 48, 306–316.
7. E. Theuergarten, T. Bannenberg, M. D. Walter, D. Holschumacher, M. Freytag, C. G. Daniliuc, P. G. Jones and M. Tamm, Dalton Trans., 2014, 43, 1651–1662.
8. M. A. Dureen and D. W. Stephan, J. Am. Chem. Soc., 2010, 132, 13559–13568.
9. G. Ménard and D. W. Stephan, J. Am. Chem. Soc., 2010, 132, 1796–1797.
10. W. B. Tolman, Activation of Small Molecules, Wiley, 2006.
11. F. Buß, P. Mehlmann, C. Mück-Lichtenfeld, K. Bergander and F. Dielmann, J. Am. Chem. Soc., 2016, 138, 1840–1843.
12. S. Bontemps, Coord. Chem. Rev., 2016, 308, 117–130.
13. H. Sabet-Sarvestani, M. Izadyar and H. Eshghi, J. CO2 Util., 2017, 21, 459–466.
14. D. W. Stephan and G. Erker, Chem. Sci., 2014, 5, 2625–2641.
15. M. Sajid, A. Klose, B. Birkmann, L. Liang, B. Schirmer, T. Wiegand, H. Eckert, A. J. Lough, R. Fröhlich, C. G. Daniliuc, S. Grimme, D. W. Stephan, G. Kehr and G. Erker, Chem. Sci., 2013, 213–219.
16. P. A. Chase and D. W. Stephan, Angew. Chem., Int. Ed., 2008, 47, 7433–7437.
17. D. W. Stephan and G. Erker, Angew. Chem., Int. Ed., 2010, 49, 46–76.
18. G. C. Welch, R. R. S. Juan, J. D. Masuda and D. W. Stephan, Science, 2006, 314, 1124–1126.
19. T. A. Rokob, I. Bakó, A. Stirling, A. Hamza and I. Pápai, J. Am. Chem. Soc., 2013, 135, 4425–4437.
20. M. A. Courtemanche, M. A. Légaré, L. Maron and F. G. Fontaine, J. Am. Chem. Soc., 2014, 136, 10708–10717.
21. T. Mahdi and D. W. Stephan, J. Am. Chem. Soc., 2014, 136, 15809–15812.
22. R. Declercq, G. Bouhadir, D. Bourissou, M. A. Légaré, M. A. Courtemanche, K. S. Nahi, N. Bouchard, F. G. Fontaine and L. Maron, ACS Catal., 2015, 5, 2513–2520.
23. T. Wang and D. W. Stephan, Chem. Commun., 2014, 50, 7007–7010.
24. T. Privalov, Chem.–Eur. J., 2009, 15, 1825–1829.
25. W.-H. Wang, X. Feng and M. Bao, in SpringerBriefs in Molecular Science, 2018, pp. 7–42.
26. D. Chen, Y. Wang and J. Klankermayer, Angew. Chem., Int. Ed., 2010, 49, 9475–9478.
27. C. M. Mömming, E. Otten, G. Kehr, R. Fröhlich, S. Grimme, D. W. Stephan and G. Erker, Angew. Chem., Int. Ed., 2009, 48, 6643–6646.
28. F. Bertini, V. Lyaskovskyy, B. J. J. Timmer, F. J. J. de Kanter, M. Lutz, A. W. Ehlers, J. C. Slootweg and K. Lammertsma, J. Am. Chem. Soc., 2012, 134, 201–204.
29. Y. Wang, Z. H. Li and H. Wang, RSC Adv., 2018, 8, 26271–26276.
30. R. T. Paine and H. Nöth, Chem. Rev., 1995, 95, 343–379.
31. J. A. Bailey and P. G. Pringle, Coord. Chem. Rev., 2015, 297–298, 77–90.
32. J. A. Bailey, M. Ploeger and P. G. Pringle, Inorg. Chem., 2014, 53, 7763–7769.
33. M. Kaaz, C. Bäucker, M. Deimling, S. König, S. H. Schlindwein, J. Bender, M. Nieger and D. Gudat, Eur. J. Inorg. Chem., 2017, 2017, 4525–4532.
34. S. J. Geier, T. M. Gilbert and D. W. Stephan, Inorg. Chem., 2011, 50, 336–344.
35. P. P. Power, Angew. Chem., Int. Ed. Engl., 1990, 29, 449–460.
36. D. C. Pestana and P. P. Power, J. Am. Chem. Soc., 1991, 113, 8426–8437.
37. S. J. Geier, T. M. Gilbert and D. W. Stephan, J. Am. Chem. Soc., 2008, 130, 12632–12633.
38. E. N. Daley, C. M. Vogels, S. J. Geier, A. Decken, S. Doherty and S. A. Westcott, Angew. Chem., Int. Ed., 2015, 54, 2121–2125.
39. M. B. Kindervater, J. F. Binder, S. R. Baird, C. M. Vogels, S. J. Geier, C. L. B. Macdonald and S. A. Westcott, J. Organomet. Chem., 2019, 880, 378–385.
40. S. J. Geier, C. M. Vogels, N. R. Mellonie, E. N. Daley, A. Decken, S. Doherty and S. A. Westcott, Chem.–Eur. J., 2017, 23, 14485–14499.
41. S. J. Geier, J. H. W. Lafortune, D. Zhu, S. C. Kosnik, C. L. B. Macdonald, D. W. Stephan and S. A. Westcott, Dalton Trans., 2017, 46, 10876–10885.
42. J. Lafortune, Z.-W. Qu, K. Bamford, A. Trofimova, S. Westcott and D. W. Stephan, Chem.–A Eur. J., 2019 DOI:10.1002/chem.201903407.
43. N. Szynkiewicz, Ł. Ponikiewski and R. Grubba, Dalton Trans., 2018, 47, 16885–16894.
44. N. Szynkiewicz, Ł. Ponikiewski and R. Grubba, Chem. Commun., 2019, 55, 2928–2931.
45. P. Pyykkö and M. Atsumi, Chem.–Eur. J., 2009, 15, 186–197.
46. P. Pyykkö and M. Atsumi, Chem.–Eur. J., 2009, 15, 12770–12779.
47. F. H. Allen, D. G. Watson, L. Brammer, A. G. Orpen and R. Taylor, in International Tables for Crystallography, ed. T. Hahn, International Union of Crystallography, Chester, England, 2006, vol. A, pp. 790–811.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental, crystallographic, spectroscopic and computational details. CCDC 1906496 and 1906498–1906500. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c9ra06638a

---