Sustainable catalysts based on main-group elements have emerged as alternatives to expensive and environmentally unfriendly precious metal systems. Frustrated Lewis pairs (FLPs) are precluded from forming a classical Lewis adduct, and have displayed remarkably versatile reactivity in the fields of small-molecule activation and catalysis. The initial reaction of the acid, base and small molecule (e.g. H₂) is formally termolecular, but the viability of this reaction is rationalised by the pre-association of the acid and base in an encounter complex - we reviewed this topic revently (Dalton Trans. 2022, 51, 10681). However, there is no experimental methodology to study the active encounter complex, i.e. the pre-associated complex that is in the correct orientation for small-molecule activation.
We have very recently shown that the charge-transfer band between PMes₃ and B(C₆F₅)₃ can be analysed by supramolecular techniques to provide the key thermodynamic parameter, Kₐ, for the active encounter complex. We also demonstrated that a higher concentration of active encounter complex in solution leads to a faster activation of hydrogen. This method enables researchers to directly probe the complex that underpins FLP small-molecule activation and subsequent catalysis, and will aid the design of more active sustainable catalysts. This manuscript was published in Nature Communications: Nat. Commun. 16, 3666 (2025)
The key work that underpins this study is the fact that FLPs can absorb light to promote single-electron transfer to form frustrated radical pairs. The field of frustrated radical pairs has many exciting applications that are being explored around the world, but the role of light in these systems was unknown until this work. When Andy was working as a Veni laureate under the mentorship of Prof Chris Slootweg at the University of Amsterdam, we were able to unambiguously demonstrate that specific wavelengths of light are required to promote the single-electron transfer in these systems (Angew. Chem. Int. Ed. 2020, 59, 22210; Chem. Eur. J. 2020, 26, 9005). The figure below shows evidence from the PMes3/B(C6F5)3 system using UV-Vis spectroscopy, electron paramagnetic resonance (EPR) spectroscopy, and transient absorption spectroscopy.
The final example of the interactions of FLPs with light was a project Andy did during his Banting Postdoctoral Fellowship with Prof Stephan in Toronto. Typically, FLP reactions occur on addition of the substrate, but we designed the first (and so far, only) example of an FLP that is initiated by light to promote the activation of alkynes (J. Am. Chem. Soc. 2018, 140, 8119), which permitted enhanced temporal control over reactivity. Furthermore, the reaction could also be initiated by mild heating to give a different isomer. Ultimately, this is a "switch on" FLP system where the stereochemistry of the product can be controlled depending on whether light or heat is used as the stimulus.
The conceptual replacement of nitrogen with phosphorus in common organic functional groups unlocks new properties and reactivity. Azophosphines (ArN=N-PR₂) are the phosphorus-containing analogues of triazenes (ArN=N-NR₂), and are underexplored but offer great potential as flexible and small bite-angle ligands. We have recently explored the synthesis and characterisation of a family of air-stable azophosphine-borane complexes, and their subsequent deprotection to the free azophosphines (Chem. Eur. J. 2024, e202401358). These compounds were structurally characterised, both experimentally and computationally, and highlight the availability of the phosphorus lone pair for coordination. This is confirmed by demonstrating that neutral azophosphines can act as ligands in Ru complexes, and can coordinate as monodentate or bidentate ligands in a controlled manner, in contrast to their nitrogen analogues.
Heterocycles containing both phosphorus and nitrogen have seen increasing use in recent years in luminescent materials, coordination chemistry and building blocks for inorganic polymers, yet their chemistry is currently dominated by five- and six-memberd derivatives. Seven-membered P/N heterocycles are comparatively scarce and lack general, high yielding syntheses. We have explore the synthesis and characterisation of 1,2,5-diazaphosphepines from azophosphines. The mechanism has been probed in detail with both computational and experimental studies supporting a stepwise mechanism to form a five-membered ring, and subsequent ring expansion to the diazaphosphepine. Regioselective synthesis of five- and seven-membered rings is possible using asymmetric alkynes. The Lewis acidic borane B(C6F5)3 could either catalyse the formation of the seven-membered ring (iPr derivative) or trap out a key intermediate via a frustrated Lewis pair (FLP) mechanism (tBu derivative). This work is under review but the pre-print is available: ChemRxiv 2024, DOI: 10.26434/chemrxiv-2024-rv4w0
Andy's interest in phosphorus-containing analogues of organic molecules started during his PhD with Prof Jose Goicoechea. He worked on the synthesis and reactivity of the PCO– anion (Angew. Chem. Int. Ed. 2013, 52, 10064); the discovery of phosphinecarboxamide (J. Am. Chem. Soc. 2013, 135, 19131); and the first synthesis of P,P-dimethylformylphosphine (J. Am. Chem. Soc. 2018, 140, 12751). These molecues are P-containing analogues of the cyanate anion, urea, and the ubiquitous solvent DMF, respectively. Far from being solely academic curiosities, these new functional groups serve as ligands (Chem. Sci. 2015, 6, 6379; Chem. Commun. 2014, 50, 12281; Eur. J. Inorg. Chem. 2016, 5, 639), catalysts (Chem. Sci. 2015, 6, 4017), and air-stable substitutes for pyrophoric and toxic phosphine (PH3) gas in the synthesis of zinc phosphide thin films via chemical vapour deposition (Dalton Trans. 2018, 47, 9221). We are continuing to explore new P-containing architectures and their applications.
The alarming trend in rising atmospheric CO2 levels since the industrial revolution necessitates urgent action and a variety of strategies and technologies. One approach is the capture and utilisation of carbon dioxide as a sustainable C1 feedstock for useful chemicals. In our on-going collaboration with the Xu group (Tongji University, China) and the Stephan group (University of Toronto, Canada), we have shown that CO2 can be converted into useful urea derivatives using silylamines and an indium catalyst (Angew. Chem. Int. Ed. 2017, 56, 14277). The process was subsequently improved to remove the need for the indium catalyst, and the scope extended to incorporate 13C-labelled, chiral or macrocyclic derivatives of urea (Angew. Chem. Int. Ed. 2019, 58, 5707). Finally, FLP hydrogenation catalysis reactions of CO2 in the presence of silylhalides were developed to selectively reduce CO2 to silylacetal, methoxysilane, methyliodide and methane derivatives under mild conditions, where the judicious choice of silylhalide and solvent could control the selectivity (Angew. Chem. Int. Ed. 2021, 60, 25771).
Furthermore, we have been exploring the small-molecule activation and functionalisation of carbon monoxide, CO. The anion [PtBu2]− reacts with CO to generate a transient intermediate with carbene character (Angew. Chem. Int. Ed. 2019, 58, 3548). This species undergoes either 1,2-alkyl migration or dimerization to form acyl-phosphide anions, and further reactivity with a second equivalent of CO to give a diacyl-phosphide anion. This project was extended to explore the reactivity of benzyl potassium with CO, with various products including alkenes, five- and seven-membered rings (Chem. Asian J. 2021, 16, 3640). We recently showed that the anionic heterocumulene [N=C=C=C=O]− could be accessed in >90% yields on a multigram scale from the functionalisation of CO with a deprotonated cyano-substituted ylid derivative (Angew. Chem. Int. Ed. 2024, 63, e202402728). We showed that the carbene-stabilised neutral cyanoketene could be accessed and crystallographically characterised; the cyanoketene is actively being hunted in interstellar space as a precursor for life in the universe as it is a small molecule containing the four essential elements (C, H, N, O).
There is a pressing need to better manage the planet’s resources. We are interested in accessing commercially relevant nitrogen-containing products, such as dyes and pharmaceuticals, directly from nitrogen gas, N2. By harnessing the energy in light, we can access highly reactive intermediates that can react with a range of small molecules, and access commercially relevant compounds in a more sustainable manner. This section will be updated as we gather more results.
We are continuing the rich legacy of molecular machines at Birmingham by designing novel photo-switches that contain phosphine functionalities. This will enable us to explore a new family of switchable ligands, and their application in light-controlled catalysis. This section will be updated as we gather more results.