Open Access Article
This Open Access Article is licensed under a
Creative Commons Attribution 3.0 Unported Licence

Ethylene carbonate splitting and Claisen-type self-addition of γ-butyrolactone promoted by an oxygen-bridged Ga/P FLP

Julian Butha, Beate Neumanna, Jan-Hendrik Lamma, Hans-Georg Stammlera, Yury V. Vishnevskiyab and Norbert W. Mitzel*a
aLehrstuhl für Anorganische Chemie und Strukturchemie, Centrum für Molekulare Materialen CM2, Fakultät für Chemie, Universität Bielefeld, Universitätsstraße 25, Bielefeld 33615, Germany. E-mail: mitzel@uni-bielefeld.de
bDepartment of Chemistry, M. V. Lomonosov Moscow State University, Leninskie Gory 1–3, Moscow 119991, Russia

Received 25th February 2026 , Accepted 16th April 2026

First published on 20th April 2026


Abstract

The versatile reactivity of the geminal oxygen-bridged frustrated Lewis pair (FLP) Bis2Ga–O–PtBu2 (Bis = CH(SiMe3)2; GaOP) is presented towards a series of C[double bond, length as m-dash]O double bonds, strained oxygen-containing rings, alkynes plus alkenes and a C–Cl bond. By adding benzaldehyde, cyclopentanone (CP), γ-butyrolactone (GBL) or ethylene carbonate (EC), classical 1,2-adducts are formed. The reversibility of the GaOP·EC adduct stands out, enabling a splitting of EC after several days or when heated to 70 °C, resulting in FLP adducts of CO2 and ethylene oxide (EO). The latter consists of a six-membered ring and is obtained via a ring-opening reaction, as verified by a further reaction involving the similar propylene oxide (PO). The GaOP·GBL adduct is further converted, with two adduct molecules combining into a 2[thin space (1/6-em)]:[thin space (1/6-em)]1 FLP adduct in a Claisen-type addition and ring-opening. Tests towards various acetylene containing species revealed that the ratio of deprotonation to ring-closure product can be adjusted. While PhC[triple bond, length as m-dash]CH is almost exclusively bound as the deprotonation product, for Me3SiC[triple bond, length as m-dash]CH, the ring-closure product also forms. In the reaction with benzyl chloride, Bis2(Cl)Ga–O–P(Bn)tBu2 is formed by a C–Cl bond activation.


Introduction

Understanding the activation and transformation steps of small molecules is of crucial importance for many chemical applications. With the concept of frustrated Lewis pairs (FLPs), D. W. Stephan introduced a way to polarise substrates through the simultaneous action of Lewis acid (LA) and Lewis base (LB) functions.1 Over the last two decades, this field of research has generated enormous interest, particularly within main-group chemistry and in the pursuit of transition metal-free catalysis.2–5

While many transition-metal-based systems boast catalysts for a variety of transformations,6–13 the investigations of main-group-based FLP-mediated conversions have focused primarily on hydrogenation reactions, C–C couplings and CO2 reductions.14–19 Although FLPs are capable of activating many types of small molecules,20–25 the majority form irreversible adducts, making the development of catalytic reactions difficult. Only a few undergo further transformation due to a conditional reversibility, which is essential for the broader application of FLPs as catalysts.26,27

Multiple reactivity studies have shed light on the following adjustment screws for tuning intramolecular FLP reactivity: the combination of LA and LB elements, their respective substituents and the linker unit. The literature predominantly features boron-containing FLP systems that have to bear strongly electron-withdrawing groups in order to be reactive.28 Such groups are unnecessary for the heavier homologue aluminium, which is inherently more Lewis acidic.29 However, its strong oxophilicity is a major drawback, particularly when it comes to the release of a transformed product that contains oxygen atoms.

Instead, the group 13 element gallium has a similar Lewis acidity level but prefers softer Lewis bases according to the concept of hard and soft acids and bases (HSAB) by R. G. Pearson.30,31 W. Uhl et al. took advantage of this tendency with the benzylidene-bridged E/P FLPs (E = Al and Ga) in a reaction involving carbon dioxide: the aluminium system irreversibly captures CO2, while the gallium analogue binds it reversibly.29,32

Similarly, adjusting the linker unit, Y. Wang et al. have shown the reversibility of a CO2 adduct with an oxygen-bridged B/P FLP.33 Reducing the Lewis acidity using donating atoms like oxygen affects the reactivity. Recently, we demonstrated how to tune FLPs for specific reactivities with four chalcogen-bridged systems Bis2E–X–PtBu2 (Bis = CH(SiMe3)2, E = Al and Ga, X = O and S).34

As Bis2Al–O–PtBu2 (AlOP) imposes with H2 and CO2 activation, resulting in a reduction of carbon dioxide to the formate stage,35 Bis2Ga–O–PtBu2 (GaOP) forms an unusual temperature-dependent H2 equilibrium by cleavage into Bis2GaH and tBu2P(O)H.36 In particular, this difference in oxophilicity prompted us to test reactivity towards molecules with C[double bond, length as m-dash]O double bonds and small rings containing oxygen atoms.

In this work, we show impressive reactivities of Bis2Ga–O–PtBu2: the cleavage of ethylene carbonate (EC) to ethylene oxide (EO) and CO2, a Claisen addition of γ-butyrolactone (GBL), the ring opening of propylene oxide (PO) and deprotonation versus ring formation of phenyl- and trimethylsilylacetylene plus C–Cl bond activation of benzyl chloride.

Results and discussion

We probed the reactivity of the FLP Bis2Ga–O–PtBu2 (GaOP) with a series of molecules containing C[double bond, length as m-dash]O double bonds: benzil, benzophenone, benzaldehyde, cyclopentanone (CP), γ-butyrolactone (GBL) and ethylene carbonate (EC). Unlike the AlOP system, which forms a stable 1,2-addition product with benzil, the reaction of the GaOP system and benzil ends up in an unselective decomposition. As no reaction was observed with benzophenone, benzaldehyde with a less sterically demanding carbonyl unit was tested, affording the 1,2-adduct 1 (Scheme 1).
image file: d6qi00380j-s1.tif
Scheme 1 Reactions of GaOP with benzaldehyde, cyclopentanone (CP), γ-butyrolactone (GBL) and ethylene carbonate (EC) to five-membered ring adducts 1–4. Tests with benzil and benzophenone show decomposition or no reaction.

In the 1H NMR spectrum, the resonances of the Bis and tert-butyl groups are split due to the implemented stereocentre. The chemical shifts of the methine protons at the gallium atom with −0.26 and −0.48 ppm indicate a tetra-coordination. The aldehyde proton resonance at 5.85 ppm is significantly shifted towards higher field compared to “free” benzaldehyde (9.64 ppm). The presence of a five-membered ring is confirmed by 13C{1H} NMR spectroscopy with the carbon stereocentre leading to a doublet at 75.5 ppm (1JP,C = 40.4 Hz) and by 31P{1H} NMR with a signal at 70.9 ppm.

Tests towards CP, GBL and EC result in similar adducts 2–4 with five-membered ring motifs (Scheme 1). In all cases, the 1H NMR signals of the adducts are broadened, indicating weak and potentially reversible adduct formation. In particular, the signals of the CH2 groups show poor resolution and are low field-shifted in relation to their resonances in the free substrates.

The 1H NMR shifts of the methine protons range from −0.20 to −0.62 ppm, characteristic of quadruple coordinated gallium atoms (Table 1). Likewise, the 13C{1H} and 31P{1H} NMR shifts as well as the 1JP,C coupling constants are consistent with normal 1,2-addition adducts.

Table 1 1H, 13C{1H} and 31P{1H} NMR spectroscopic data of 1, 2, 3, 4, 7 and 8
Compound No. δ(1H) (GaCH)/ppm δ(1H) (tBu)/ppm δ(13C{1H}) (PCX3)/ppm δ(31P{1H})/ppm
GaOP·PhCHO 1 −0.48, −0.26 0.84/1.15 (3JP,H = 13.4/13.3 Hz) 75.5 (1JP,C = 40.4 Hz) 70.9
GaOP·CP 2 −0.62, −0.39 0.99/1.11 (3JP,H = 13.0/13.3 Hz) 87.0 (1JP,C = 40.0 Hz) 77.2
GaOP·GBL 3 −0.38, −0.20 br. 1.10 br. 109.9 (1JP,C = 73.2 Hz) 74.1
GaOP·EC 4 −0.30 br. 1.19 br. 122.7 (1JP,C = 120.4 Hz) 70.6
GaOP·EO 7 −0.40 br. 0.87 (3JP,H = 14.0 Hz) 23.7 (1JP,C = 54.5 Hz) 75.1
GaOP·PO 8 −0.49, −0.39 0.85/0.90 (3JP,H = 13.9/14.0 Hz) 30.0 (1JP,C = 54.5 Hz) 73.0


Further confirmation was provided by elemental analysis and the determination of the molecular structures in the crystalline state, which verified the proposed connectivity for 1, 2 and 3 (Fig. 1). Despite numerous crystallizations attempts, the ethylene carbonate adduct 4 could only be obtained in a resin-like form.


image file: d6qi00380j-f1.tif
Fig. 1 Molecular structures of 1, 2 and 3 in the solid state. Hydrogen atoms (except H(1) in 1) and minor occupied disordered parts are omitted for clarity. Ellipsoids are set at the 50% probability level. Selected distances, bond lengths [Å] and angles [°]: 1: Ga(1)⋯P(1) 2.989(1), Ga(1)–O(1) 2.004(2), Ga(1)–O(2A) 1.834(13), O(1)–P(1) 1.524(2), P(1)–C(23A) 1.891(16), O(2)–C(23) 1.44(2); Ga(1)–O(1)–P(1) 115.2(2), C(1A)–Ga(1)–C(8A) 126.9(8), O(1)–Ga(1)–O(2A) 86.4(4), O(2A)–Ga(1)–P(1)–C(23A) 25.0(9); 2: Ga(1)⋯P(1) 2.990(1), Ga(1)–O(1) 1.986(1), Ga(1)–O(2) 1.894(1), O(1)–P(1) 1.538(1), P(1)–C(23) 1.897(2), O(2)–C(23) 1.404(2); Ga(1)–O(1)–P(1) 115.5(1), C(1)–Ga(1)–C(8) 119.1(1), O(1)–Ga(1)–O(2) 87.0(1), O(2)–Ga(1)–P(1)–C(23) 18.9(1); 3: Ga(1)⋯P(1) 2.986(1), Ga(1)–O(1) 1.993(1), Ga(1)–O(2) 1.907(1), O(1)–P(1) 1.533(1), P(1)–C(23) 1.906(1), O(2)–C(23) 1.360(1); Ga(1)–O(1)–P(1) 115.2(1), C(1)–Ga(1)–C(8) 122.4(1), O(1)–Ga(1)–O(2) 86.9(1), O(2)–Ga(1)–P(1)–C(23) 18.4(1).

A comparison of the molecular structures shows similar five-membered rings: the Ga(1)–O(1)–P(1) angles (115.2(2)–115.5(1)°) and the Ga(1)⋯P(1) distances (2.986(1)–2.990(1) Å) are identical. The gallium atoms are, as expected, in distorted tetrahedral environments, with the O(1)–Ga(1)–O(2) angles ranging from 86.4(4) to 87.0(1)°. Only the O(2)–C(23) bond length varies significantly with 1.44(2) (for 1), 1.404(2) (for 2) and 1.360(1) Å (for 3). However, these values are higher than those of the carbon dioxide adduct (6: 1.285(3) Å), evidencing an oxygen–carbon single bond.

Strikingly, the 1,2-addition adducts 3 and 4 undergo further conversions in solution within two days at ambient temperature (partially) or when heated to 70 °C for six hours (fully). The GBL adduct 3 is subject to a rearrangement, forming the adduct type 5 in a Claisen addition and ring-opening reaction (Scheme 2). In this reaction, two molecules of the adduct 3 react to a Claisen addition product, which then forms the α,β-unsaturated ketone resulting from a ring-opening. This molecule is bound to two FLP units. The reaction to 5 is presumably triggered by the deprotonation of the α-proton, resulting in the formation of the P–H unit (Scheme 2 or SI Scheme S1).


image file: d6qi00380j-s2.tif
Scheme 2 The 1,2-adduct 3 is further converted in a Claisen-type addition, forming a 2[thin space (1/6-em)]:[thin space (1/6-em)]1 FLP Claisen addition adduct with subsequent ring-opening via the proposed mechanism and structure 5 (left). Heating 4 to 70 °C results in the splitting of EC, leading to the FLP CO2 and EO adducts (6/7) and one half of an equivalent of EC (right). In a ring-opening reaction, propylene oxide forms a six-membered ring adduct 8. Also exchange reactions of 4 and 6 with propylene oxide (PO) result in adduct 8.

Multinuclear NMR spectroscopy analysis approves the formation of 5. The 31P NMR spectrum shows a doublet of multiplets at 63.2 ppm (1JP,H = 422.9 Hz), confirming the generation of phosphane oxide. The broadened signal indicates its interaction with the Lewis acid moiety. The 1H NMR shifts of the methine protons at −0.30 and −0.48 ppm verify tetra-coordinated gallium atoms and suggest the presence of two different Bis2Ga-groups, which are also confirmed by 29Si NMR spectroscopy. The framework of the Claisen added and ring-opened product 5 is identified by 1H and 13C{1H} NMR spectroscopy, with five CH2 units observable in the 13C DEPT-135 NMR spectrum. The remaining quaternary carbon atoms show shifts of 188.0, 178.7 and 90.7 ppm, respectively. The 1H–1H COSY NMR spectrum validates the correlation of the methylene protons at 3.68 and 2.49 ppm and of those at 4.07, 2.49 and 2.10 ppm. All these assignments are consistent within the corresponding 2D NMR spectra (1H–13C HSQC and HMBC). As for the 1,2-adduct 4, multiple attempts to crystallize 5 failed and only yielded resin-like residues.

A different transformation is observed for 4. The added ethylene carbonate (EC) is cleaved, resulting in the formation of the FLP adducts of carbon dioxide (6) and of ethylene oxide (EO) (7) accompanied by the release of half an equivalent of EC (Scheme 2). This observation lends support to the assumed reversible nature of the adduct formation with EC.

There are two main pathways for the splitting of EC (see SI Scheme S2): the activation of the carbonyl unit followed by ring-opening (path A) or ring-opening, where the carbonate unit remains intact followed by an extrusion (retro-insertion) of CO2 leading to the EO adduct 7 (path B). Both pathways have been reported in the literature,37,38 but the experimental outcome (formation of more CO2 adduct 6) matches only path A, as 7 is stable in the presence of CO2.

In the 1H NMR spectrum one set of signals corresponds to the literature values of 6.34 Characteristic of the EO adduct 7 is the doublet of triplets of the oxygen-bound CH2 group at 4.10 ppm (3JP,H = 18.5 Hz, 3JH,H = 5.6 Hz). Due to overlapping resonances, it was not possible to localise a resonance for the second CH2 unit exactly in the 1H NMR spectrum, even when using 2D NMR. In contrast, the 13C{1H} NMR spectrum contains a resonance in this regard as a doublet at 23.7 ppm (1JP,C = 54.5 Hz).

To further verify the formation of a six-membered ring, we reacted GaOP with the similar propylene oxide (PO). In a ring-opening reaction the adduct 8 was formed. Both 7 and 8 have comparable chemical shifts. For example, the 31P{1H} NMR shifts differ slightly (7: 75.1 ppm; 8: 73.0 ppm). In the 13C{1H} NMR spectrum the carbon atom attached to the oxygen atom is detected at 60.5 (7) and 65.6 ppm (8), respectively. The propylene unit of 8 induces signals at 4.35, 1.63, 1.36 and 1.32 ppm in the 1H NMR spectrum, which are similar to those of 7. Due to the diastereotopic nature of the P–CH2 protons, both resonances are split into doublets of multiplets with a characteristic 3JP,H coupling constant of 14.8 Hz (see SI Fig. S38).

Crystal structure determinations of both the EO and PO adducts 7 and 8, confirm the six-membered ring motif (Fig. 2). Since they differ only by an exocyclic methyl group, the structural parameters are nearly the same. The Ga(1)⋯P(1) distances of 3.191–3.198(1) Å are longer than in free FLP GaOP (3.082(1) Å).36 The Ga(1)–O(1)–P(1) angles in 7 (129.1(1)°) or 8 (129.3(1)°) are slightly wider than that in GaOP of 126.2(1)°.


image file: d6qi00380j-f2.tif
Fig. 2 Molecular structures of 7 and 8 in the solid state. Hydrogen atoms and minor occupied parts are omitted for clarity. Ellipsoids are set at the 50% probability level. Selected distances, bond lengths [Å] and angles [°]: 7: Ga(1)⋯P(1) 3.198(1), Ga(1)–O(1) 2.007(1), Ga(1)–O(2) 1.866(1), O(1)–P(1) 1.527(1), P(1)–C(23) 1.826(1), C(23)–C(24) 1.535(2), O(2)–C(24) 1.401(1); Ga(1)–O(1)–P(1) 129.1(1), O(1)–Ga(1)–O(2) 92.3(1), O(2)–Ga(1)–P(1)–C(23) 15.2(1); 8: Ga(1)⋯P(1) 3.191(1), Ga(1)–O(1) 2.001(2), Ga(1)–O(2) 1.854(3), O(1)–P(1) 1.523(2), P(1)–C(23) 1.847(4), C(23)–C(24) 1.520(6), O(2)–C(24) 1.393(5); Ga(1)–O(1)–P(1) 129.3(1), O(1)–Ga(1)–O(2) 93.1(1), O(2)–Ga(1)–P(1)–C(23) 14.3(3).

In an NMR spectroscopic study, we tested the exchange reaction of the ethylene carbonate adduct 4 and the CO2 adduct 6 with an excess of propylene oxide (PO). When 4 is reacted with PO, the formation of adduct 8 and free EC is observed at ambient temperatures. For 6, heating to 70 °C is necessary to exchange the carbon dioxide with PO. These findings gave further insights and strengthened the proposed mechanism (path A) of the ethylene carbonate splitting (see SI Scheme S2).

Relating to previous investigations of slightly polarized alkynes with the AlOP system,39 we tested the reactivity of phenyl- and trimethylsilylacetylene with the GaOP FLP. Interestingly, there is a clear difference in the product ratio of deprotonation vs. ring-closure. Whereas deprotonation predominates in the case of phenylacetylene (>96%), and crystallization gave the deprotonation product 9a in pure form, only mixtures with deprotonation/ring-closure ratios of approximately 2[thin space (1/6-em)]:[thin space (1/6-em)]1 (10a[thin space (1/6-em)]:[thin space (1/6-em)]10b) were obtained for trimethylsilylacetylene (Scheme 3).


image file: d6qi00380j-s3.tif
Scheme 3 Reactions of GaOP with phenylacetylene and trimethylsilylacetylene result in deprotonation (9a/10a) or ring-closure (9b/10b) products. Using benzyl chloride leads to Bis2(Cl)Ga–O–P(H)tBu2 (11).

Heating the reaction to 70 °C for 24 h did not result in formation of the ring-closure products 9b and 10b. The same behaviour, along with a similar deprotonation/ring-closure ratio, was previously observed for AlOP·PhC[triple bond, length as m-dash]CH (85[thin space (1/6-em)]:[thin space (1/6-em)]15).39 Although the experimental outcome is consistent, quantum chemical calculations suggest that the ring-closure products 9b/10b are thermodynamically favoured (see SI). To explain these contradictions, we assume that the deprotonation products 9a/10a are kinetically preferred and the activation barriers of these meta-stable states to the ring-closure products 9b/10b are too high. Furthermore, a partial separation of 9a and 10a into di-tert-butylphosphane oxide and Bis2GaC[triple bond, length as m-dash]CR (R = Ph and SiMe3) at higher temperatures, as we reported it for the H2 adduct of the GaOP system,36 would prevent a rearrangement to ring-closure.

While the ring-closure product 10b was identified by 1H NMR spectroscopy by its characteristic doublet at 9.25 ppm (3JP,H = 56.0 Hz), the deprotonation product 10a has a typical doublet at 5.97 ppm with a 1JP,H coupling constant of 442.2 Hz. The 31P NMR spectrum displays a doublet of multiplets with the same large coupling constant at 66.9 ppm (10a) and a single multiplet at 84.8 ppm (10b), confirming the identity of these two species. However, the 1H NMR integral ratios of the deprotonation product 10b do not match exactly; there is a surplus of phosphane oxide present. Besides hydrolysis, the aforementioned partial separation into the phosphane oxide species and Bis2GaC[triple bond, length as m-dash]CSiMe3 is probably another plausible explanation for these mismatching integral ratios. The excess of phosphane oxide undergoes a dynamic exchange with the deprotonation product 10a, meaning no “free” phosphane oxide being detected in the 31P NMR spectrum. Elemental analyses confirm this hypothesis and the compositions of 9 and 10.

We were able to determine the solid-state molecule structures of the deprotonation products 9a and 10a (Fig. 3). Contrary to the structure of AlOP·PhC[triple bond, length as m-dash]CH in both forms,39 and although the formation of the ring-closure product 10b is evident by NMR spectroscopy, the crystalline sample showed no evidence for different types of crystals. The determined bond lengths of C(23)–C(24) in the deprotonation products fall over the expected range for triple bonds (9a: 1.186(3) Å; 10a: 1.206(2) Å). The angles ∠(Ga–C[triple bond, length as m-dash]C) (9a: 175.3(2)°; 10a: 177.5(2)°) and ∠(C[triple bond, length as m-dash]C–R) (9a: 178.5(2)°; 10a: 178.0(2)°) in the acetylide units are similar, nearly matching an idealized bond angle of 180°. In contrast, both structures differ in their Ga(1)–O(1)–P(1) angles (9a: 155.3(1)°; 10a: 142.4(1)°) and their torsion angles τ(C(23)–Ga(1)–P(1)–H(1)) with 57.0(8)° for 9a and 41.1(9)° for 10a. These values for the heavier gallium homologue 9a are significantly smaller than those of its aluminium analogue AlOP·PhC[triple bond, length as m-dash]CH (168.8(2)°; 95.9(5)°).39


image file: d6qi00380j-f3.tif
Fig. 3 Molecular structures of 9a, 10a and 11 in the solid state. Hydrogen atoms except H(1) and minor occupied disordered parts are omitted for clarity. Ellipsoids are set at the 50% probability level. Selected distances, bond lengths [Å] and angles [°]: 9a: Ga(1)⋯P(1) 3.469(1), Ga(1)–O(1) 2.037(1), O(1)–P(1) 1.512(1), Ga(1)–C(23) 2.004(1), C(23)–C(24) 1.187(2); Ga(1)–O(1)–P(1) 155.3(1), C(1)–Ga(1)–C(8) 117.9(1), C(23)–Ga(1)–P(1)–H(1) 57.0(8); 10a: Ga(1)⋯P(1) 3.338(1), Ga(1)–O(1) 2.008(1), O(1)–P(1) 1.514(1), Ga(1)–C(23) 1.978(2), C(23)–C(24) 1.206(2); Ga(1)–O(1)–P(1) 142.4(1), C(1)–Ga(1)–C(8) 121.6(1), C(23)–Ga(1)–P(1)–H(1) 41.1(9). 11: Ga(1)⋯P(1) 3.371(1), Ga(1)–O(1) 2.014(2), O(1)–P(1) 1.528(2), Ga(1)–Cl(1A) 2.257(3), P(1)–C(23) 1.827(3); Ga(1)–O(1)–P(1) 143.8(1), C(1)–Ga(1)–C(8) 122.8(1), Cl(1A)–Ga(1)–P(1)–C(23) 45.9(1).

Activation attempts with the non-polar diphenylacetylene and the alkene species styrene, stilbene, 1,2-diphenylethene and 1-phenyl-1-(trimethylsiloxyl)ethylene remained unsuccessful. However, we demonstrated the versatility of the GaOP FLP by reacting it with benzyl chloride. After heating to 70 °C, the adduct Bis2(Cl)Ga–O–P(Bn)tBu2 (11) was isolated and fully characterized by means of NMR spectroscopy, elemental analysis and single crystal X-ray diffraction experiments (Fig. 3). Cleavage of the C–Cl bond resulted in the chloride ion bound to the Lewis acidic gallium and a benzyl group attached to the phosphorus atom.

In the 1H NMR spectrum the formation of 11 is proven by a doublet of the CH2 group at 3.40 ppm (3JP,H = 14.0 Hz). The broad methine proton singlet at −0.06 ppm verifies the tetra-coordinated gallium atom. Besides the signal at 63.6 ppm, another species containing a P–H unit was detected by a doublet 31P NMR resonance at 69.7 ppm (1JP,H = 451.5 Hz). This minor impurity was identified as the formal hydrogen chloride adduct Bis2(Cl)Ga–O–P(H)tBu2, likely originating from the partial hydrolysis of benzyl chloride. The chemical shifts are comparable with those of the recently published HBr adduct, GaOP·HBr.36

The molecular structure of 11 shows a tetrahedral environment at the gallium atom and the Ga(1)–Cl(1A) bond length of 2.257(3) Å is slightly longer than in the free Bis2GaCl (2.193(1) Å).40 Compared with GaOP·HBr, the Ga(1)–O(1)–P(1) angles are almost identical (GaOP·HBr: 143.1(1)°; 11: 143.8(1)°). As the value of τ(C(23)–Ga(1)–P(1)–H(1)) = 45.9(1)° indicates, the chloride ion and benzyl group are twisted, in a similar manner to the proton and acetylide unit in the deprotonation adducts 9a and 10a.

Conclusions

Bis2Ga–O–PtBu2 (GaOP) exhibits a remarkably broad and tuneable reactivity profile towards a wide variety of substrates, ranging from carbonyl compounds to strained oxygen heterocycles, alkynes and polar σ-bonds. In classical 1,2-additions to the C[double bond, length as m-dash]O double bond of benzaldehyde, cyclopentanone (CP), γ-butyrolactone (GBL) and ethylene carbonate (EC), FLP adducts with five-membered rings are observed. Beyond simple adduct formation, the system reveals unexpected secondary transformations such as the Claisen-type reaction of two GBL adducts and subsequent trapping of the resulting product as a 2[thin space (1/6-em)]:[thin space (1/6-em)]1 FLP adduct. The reversible activation of EC is particularly notable, as it enables a fragmentation into CO2 and ethylene oxide (EO), both of which were detected as the corresponding FLP adducts. This ring-opening reaction to a six-membered ring was proven using propylene oxide (PO), which results in the formation of a similar adduct. Reactions involving phenyl- and trimethylsilylacetylene further demonstrate that subtle changes in the substrate structure allow deprotonation and ring-closure to be deliberately steered, providing rare insight into selectivity control in FLP chemistry. Activating the C–Cl bond in benzyl chloride broadens the scope of activation to include polar σ-bonds. Taken together, these findings establish the GaOP system as a highly versatile and adaptable FLP, capable of reversible activation, bond cleavage and multistep substrate transformation.

Experimental section

General methods

All reactions and manipulations with air and moisture sensitive compounds were carried out under conventional Schlenk techniques or in a glove box using argon as an inert gas. Volatile compounds were handled in a vacuum line. The solvents n-hexane, toluene, toluene-d8 and benzene-d6 were dried over a Na/K alloy, distilled and degassed prior to use. Bis2Ga–O–PtBu2 (GaOP) was prepared according to a literature procedure.36 Benzaldehyde, cyclopentanone (CP), γ-butyrolactone (GBL), propylene oxide (PO), phenylacetylene, trimethylsilyl acetylene and benzyl chloride were degassed, dried over molecular sieves (4 Å) and distilled prior to use. Ethylene carbonate (EC) was used without further purification. NMR spectra were recorded using a Bruker Avance III 500, Avance III 500 HD, Ascend 500 neo2K or Ascend 500 neo3K spectrometer at ambient temperature unless otherwise stated. Chemical shifts were referenced to the residual proton or carbon signal of the solvent (benzene-d6: 1H: 7.16 ppm, 13C: 128.1 ppm; toluene-d8: 1H: 2.09 ppm) or externally (29Si: SiMe4, 31P: 85% H3PO4 in H2O). Elemental analyses were carried out using a HEKATECH EURO elemental analyzer.

Synthetic procedure

Bis2Ga–O–PtBu2 (GaOP) was dissolved in toluene (2 mL), the substrate was added and the reaction was stirred for 24 h at room temperature, unless otherwise stated. All volatiles were removed under reduced pressure and the residue was dried in vacuo.
Bis2Ga–O–PtBu2·PhCHO (1). Using benzaldehyde (5 μL, 4.8 mg, 45 μmol, 1.2 equiv.) and after an additional washing step with n-hexane, 1 was obtained as a colourless solid (24 mg, 37 μmol, 97%). 1H NMR (500 MHz, C6D6): δ [ppm] = −0.48 (s, 1H, GaCH), −0.26 (s, 1H, GaCH), 0.36–0.58 (m, 36H, Si(CH3)3), 0.84 (d, 3JP,H = 13.4 Hz, 9H, C(CH3)3), 1.15 (d, 3JP,H = 13.3 Hz, 9H, C(CH3)3), 5.85 (s, CHO), 7.02 (t, 3JH,H = 7.2 Hz, 1H, para-H), 7.14 (t, 3JH,H = 7.3 Hz, 2H, meta-H), 7.72 (d, 3JH,H = 7.6 Hz, 2H, ortho-H). 13C{1H} NMR (126 MHz, C6D6): δ [ppm] = 3.9 (s, Si(CH3)3), 4.6(s, Si(CH3)3), 4.6 (s, Si(CH3)3), 4.8 (s, Si(CH3)3), 6.6 (s, GaCH), 27.1 (s, C(CH3)3), 28.2 (s, C(CH3)3), 35.3 (s, C(CH3)3), 37.0 (s, C(CH3)3), 75.5 (d, 1JP,C = 40.4 Hz, CHO), 126.3 (s, ortho-C), 127.2 (s, para-C), 128.3 (s, meta-C), 141.6 (s, ipso-C). 29Si{1H} NMR (99 MHz, C6D6): δ [ppm] = −1.1 (s). 31P{1H} NMR (202 MHz, C6D6): δ [ppm] = 70.9 (s). Elemental analysis calcd (%) for C29H62GaO2PSi4 (Mr = 655.85): C 53.11, H 9.53; found C 53.57, H 9.54.
Bis2Ga–O–PtBu2·CP (2). Using cyclopentanone (10 μL, 9.5 mg, 113 μmol, 1.2 equiv.) and after an additional washing step with n-hexane, 2 was obtained as a colourless solid (58 mg, 91 μmol, 93%). 1H NMR (500 MHz, C6D6): δ [ppm] = −0.62 (s, 1H, GaCH), −0.39 (s, 1H, GaCH), 0.42–0.50 (s, 36H, Si(CH3)3), 0.99 (d, 3JP,H = 13.0 Hz, 9H, C(CH3)3), 1.11 (d, 3JP,H = 13.3 Hz, 9H, C(CH3)3), 1.48(m, 2H, CH2), 1.66 (m, 2H, CH2), 2.03 (m, 4H, CH2). 13C{1H} NMR (126 MHz, C6D6): δ [ppm] = 5.3 (s, Si(CH3)3), 6.8 (s, GaCH), 9.9 (s, GaCH), 23.5 (d, 2JP,C = 11.9 Hz, CH2), 23.8 (d, 2JP,C = 12.4 Hz, CH2), 28.2 (s, C(CH3)3), 28.8 (s, C(CH3)3), 35.8 (d, 1JP,C = 41.4 Hz, C(CH3)3), 37.5 (d, 1JP,C = 41.8 Hz, C(CH3)3), 41.1 (s, CH2), 42.5 (s, CH2), 87.0 (d, 1JP,C = 40.0 Hz, PCO). 29Si{1H} NMR (99 MHz, C6D6): δ [ppm] = −1.2 (s). 31P{1H} NMR (202 MHz, C6D6): δ [ppm] = 77.2 (s). Elemental analysis calcd (%) for C27H64GaO2PSi4 (Mr = 633.84): C 51.16, H 10.18; found C 50.78, H 10.14.
Bis2Ga–O–PtBu2·GBL (3). Using γ-butyrolactone (10 μL, 11.3 mg, 131 μmol, 1.0 equiv.) and after an additional washing step with n-hexane, 3 was obtained as a colourless solid (70 mg, 110 μmol, 87%). 1H NMR (500 MHz, C6D6): δ [ppm] = −0.38 (br. s, 1H, GaCH), −0.20 (br. s, 1H, GaCH), 0.45 (s, 36H, Si(CH3)3), 1.10 (m, 18H, C(CH3)3), 1.44 (m, 1H, CH2), 1.60 (br. m, 1H, CH2), 1.93 (br. s, 1H, CH2), 2.11 (br. s, 1H, CH2), 3.32 (br. s, 1H, CH2), 4.00 (br. s, 1H, CH2). 13C{1H} NMR (126 MHz, C6D6): δ [ppm] = 5.1 (s, Si(CH3)3), 9.2 (s, GaCH), 24.7 (s, CH2), 27.8 (s, C(CH3)3), 35.6 (d, 1JP,C = 47.5 Hz, C(CH3)3), 36.4 (d, 1JP,C = 48.5 Hz, C(CH3)3), 39.6 (s, CH2), 67.1 (s, OCH2), 109.9 (d, 1JP,C = 73.2 Hz, PCO2). 29Si{1H} NMR (99 MHz, C6D6): δ [ppm] = −1.5 (s). 31P{1H} NMR (202 MHz, C6D6): δ [ppm] = 74.1 (s). Elemental analysis calcd (%) for C26H62GaO3PSi4 (Mr = 635.82): C 49.12, H 9.83; found C 48.61, H 10.23.
Bis2Ga–O–PtBu2·EC (4). Using ethylene carbonate (9.1 mg, 103 μmol, 1.1 equiv.) and after an additional washing step with n-hexane, 4 was obtained in a colourless, resin-like form (52 mg, 82 μmol, 90%). 1H NMR (500 MHz, C6D6): δ [ppm] = −0.30 (br. s, 2H, GaCH), 0.45 (s, 36H, Si(CH3)3), 1.19 (m, 18H, C(CH3)3), 3.28 (br. s, 2H, CH2), 3.80 (br. s, 2H, CH2). 13C{1H} NMR (126 MHz, C6D6): δ [ppm] = 4.9 (s, Si(CH3)3), 9.0 (s, GaCH), 27.5 (s, C(CH3)3), 35.8 (br. m, C(CH3)3), 63.4 (s, OCH2), 122.7 (d, 1JP,C = 120.4 Hz, PCO3). 29Si{1H} NMR (99 MHz, C6D6): δ [ppm] = −1.3 (s). 31P{1H} NMR (202 MHz, C6D6): δ [ppm] = 70.6 (s). Elemental analysis calcd (%) for C25H60GaO4PSi4 (Mr = 637.79): C 47.08, H 9.48; found C 46.99, H 9.56.
Claisen addition ring-opening FLP adduct 5. Heating a solution of GaOP (55.7 mg, 101 μmol) and GBL (7.9 μL, 8.9 mg, 104 μmol, 1.0 equiv.) in toluene (3 mL) to 70 °C for 6 h, full conversion of the FLP to 5 was observed. After removing all volatiles in vacuo, and an additional washing step with n-hexane, 5 was obtained in a resin-like form (59.4 mg, 47 μmol, 92%). 1H NMR (500 MHz, C6D6): δ [ppm] = −0.48 (s, 2H, GaCH), −0.30 (s, 2H, GaCH), 0.34 (s, 36H, Si(CH3)3), 0.34 (s, 18H, Si(CH3)3), 0.36 (s, 18H, Si(CH3)3), 0.99 (d, 3JP,H = 15.0 Hz, 18H, C(CH3)3), 2.10 (tt, J = 11.9, 5.8 Hz, 2H, CH), 2.49 (m, 2H, CH), 2.49 (t, J = 8.3 Hz, 2H, CH), 3.68 (t, J = 8.2 Hz, 2H, CH), 4.07 (t, J = 6.0 Hz, 2H, CH2). 13C{1H} NMR (126 MHz, C6D6): δ [ppm] = 3.9/4.1/4.5 (s, Si(CH3)3), 7.5/10.3 (s, GaCH), 25.8 (s, C(CH3)3), 26.0 (s, CH2), 30.9 (s, CH2), 33.6 (d, 1JP,C = 59.0 Hz, C(CH3)3), 36.1 (s, CH2), 65.7 (s, OCH2), 67.8 (s, OCH2), 90.7 (s, CCO2Ga), 178.7 (s, CO2Ga), 188.0 (s, C[double bond, length as m-dash]O). 29Si{1H} NMR (99 MHz, C6D6): δ [ppm] = −1.4 (s), −0.5 (s), −0.3 (s). 31P{1H} NMR (202 MHz, C6D6): δ [ppm] = 63.2 (s).
Bis2Ga–O–PtBu2·EO (7). After heating a solution of GaOP (60.5 mg, 110 μmol) and EC (9.6 mg, 110 μmol, 1.0 equiv.) in toluene (3 mL) to 70 °C for 6 h, full conversion of the FLP to a mixture of Bis2Ga–O–PtBu2·CO2 (6) and 7 was observed. The chemical shifts of 6 are in accordance with the literature values.34 Crystals of 7, suitable for X-ray diffraction experiments, were obtained by the slow evaporation of a solution in C6D6. 1H NMR (500 MHz, C6D6): δ [ppm] = −0.40 (br. s, 2H, GaCH), 0.49 (s, 36H, Si(CH3)3), 0.87 (d, 3JP,H = 14.0 Hz, 18H, C(CH3)3), 4.10 (dt, 3JP,H = 18.5 Hz, 3JH,H = 5.6 Hz, 2H, OCH2); due to overlapping resonances, as well in 2D spectra, signals for PCH2 could not be localized exactly. 13C{1H} NMR (126 MHz, C6D6): δ [ppm] = 5.1 (s, Si(CH3)3), 10.5 (s, GaCH), 23.7 (d, 1JP,C = 54.5 Hz, PCH2), 25.9 (s, C(CH3)3), 26.2 (s, C(CH3)3), 33.6 (d, 1JP,C = 58.9 Hz, C(CH3)3), 35.4 (d, 1JP,C = 58.3 Hz, C(CH3)3), 60.5 (d, 2JP,C = 7.6 Hz, OCH2). 29Si{1H} NMR (99 MHz, C6D6): δ [ppm] = −0.5 (s). 31P{1H} NMR (202 MHz, C6D6): δ [ppm] = 75.1 (s).
Bis2Ga–O–PtBu2·PO (8). Using propylene oxide (10 μL, 8.3 mg, 143 μmol, 1.5 equiv.), the ring-opening adduct 8 crystallized from n-hexane at −18 °C, giving colourless needles (32 mg, 53 μmol, 55%). 1H NMR (500 MHz, C6D6): δ [ppm] = −0.49 (s, 1H, GaCH), −0.39 (s, 1H, GaCH), 0.42–0.56 (m, 36H, Si(CH3)3), 0.85 (d, 3JP,H = 13.9 Hz, 9H, C(CH3)3), 0.90 (d, 3JP,H = 14.0 Hz, 9H, C(CH3)3), 1.32 (ddd, J = 14.8, 3.7, 1.8 Hz, PCH2), 1.36 (dd, J = 5.7, 2.7 Hz, CH3), 1.63 (ddd, J = 14.8, 10.8, 8.8 Hz, PCH2), 4.35 (m, 1H, OCH). 13C{1H} NMR (126 MHz, C6D6): δ [ppm] = 4.8 (s, overlapped, from 1H–13C HSQC experiment, GaCH), 4.9 (s, Si(CH3)3), 5.2 (s, Si(CH3)3), 5.6 (s, Si(CH3)3), 5.6 (s, overlapped, from 1H–13C HSQC experiment, GaCH), 26.2 (s, C(CH3)3), 26.5 (s, C(CH3)3), 29.4 (d, 3JP,C = 13.6 Hz, CH3), 30.0 (d, 1JP,C = 54.5 Hz, PCH2), 35.1 (d, 1JP,C = 57.7 Hz, C(CH3)3), 35.7 (d, 1JP,C = 59.0 Hz, C(CH3)3), 65.6 (d, 2JP,C = 5.9 Hz, OCH). 29Si{1H} NMR (99 MHz, C6D6): δ [ppm] = −1.7 (s), −1.1 (s), −0.7 (s), −0.2 (s). 31P{1H} NMR (202 MHz, C6D6): δ [ppm] = 73.0 (s). Elemental analysis calcd (%) for C25H62GaO2PSi4 (Mr = 607.81): C 49.40, H 10.28; found C 49.14, H 10.53.
Bis2Ga–O–PtBu2·PhC[triple bond, length as m-dash]CH (9). Using phenylacetylene (10 μL, 9.3 mg, 91 μmol, 1.0 equiv.) and stirring at 70 °C, 9a was recrystallized from n-hexane at −18 °C, giving colourless needles of the deprotonation product (38 mg, 59 μmol, 64%). 1H NMR (500 MHz, C6D6): δ [ppm] = −0.31 (s, 2H, GaCH), 0.46 (s, 36H, Si(CH3)3), 0.95 (d, 3JP,H = 15.7 Hz, 18H, C(CH3)3), 5.98 (d, 1JP,H = 440.1 Hz, 1H, PH), 6.98 (t, 3JH,H = 7.4 Hz, 1H, para-H), 7.05 (t, 3JH,H = 7.3 Hz, 2H, meta-H), 7.62 (d, 3JH,H = 7.8 Hz, 2H, ortho-H). 13C{1H} NMR (126 MHz, C6D6): δ [ppm] = 4.6 (s, Si(CH3)3), 10.0 (s, GaCH), 26.0 (s, C(CH3)3), 33.8 (d, 1JP,C = 58.4 Hz, C(CH3)3), 109.7 (s, PhC[triple bond, length as m-dash]CGa), 126.1 (s, PhC[triple bond, length as m-dash]CGa), 127.6 (s, para-C), 128.4 (s, ipso-C), 128.6 (s, meta-C), 131.6 (s, ortho-C). 29Si{1H} NMR (99 MHz, C6D6): δ [ppm] = −0.9 (s). 31P{1H} NMR (202 MHz, C6D6): δ [ppm] = 67.3 (s). Elemental analysis calcd (%) for C30H62GaOPSi4 (Mr = 651.86): C 55.28, H 9.59; found C 54.61, H 9.88.
Bis2Ga–O–PtBu2·Me3SiC[triple bond, length as m-dash]CH (10). Using trimethylsilylacetylene (10 μL, 7.1 mg, 72 μmol, 1.0 equiv.), stirring at 70 °C and after an additional washing step with n-hexane, 10 was obtained in a colourless resin-like form (41 mg, 63 μmol, 92%). Deprotonation product (10a): 1H NMR (500 MHz, C6D6): δ [ppm] = −0.50 (s, 2H, GaCH), 0.27 (s, 9H, Si(CH3)3), 0.46 (s, 36H, CH(Si(CH3)3)2), 0.96 (d, 3JP,H = 15.6 Hz, 18H, C(CH3)3), 5.97 (d, 1JP,H = 442.2 Hz, 1H, PH). 13C{1H} NMR (126 MHz, C6D6): δ [ppm] = 0.3 (s, Si(CH3)3), 4.8 (s, Si(CH3)3), 8.2 (s, GaCH), 26.2 (s, C(CH3)3), 33.8 (d, 1JP,C = 58.3 Hz, C(CH3)3), 115.7 (s, Me3SiC[triple bond, length as m-dash]CGa), 134.8 (s, Me3SiC[triple bond, length as m-dash]CGa). 29Si{1H} NMR (99 MHz, C6D6): δ [ppm] = −22.9 (s), −0.7 (s). 31P{1H} NMR (202 MHz, C6D6): δ [ppm] = 66.9 (s). Elemental analysis calcd (%) for C27H66GaOPSi5 (Mr = 647.95): C 50.05, H 10.27; found C 49.81, H 10.37. Ring-closure product (10b): 1H NMR (500 MHz, C6D6): δ [ppm] = −0.50 (br. s, 2H, GaCH), 0.18 (s, 9H, Si(CH3)3), 0.38–0.43 (s, 36H, CH(Si(CH3)3)2), 1.02 (d, 3JP,H = 14.0 Hz, 18H, C(CH3)3), 9.25 (d, 3JP,H = 56.0 Hz, 1H, PCH). 13C{1H} NMR (126 MHz, C6D6): δ [ppm] = 2.0 (s, Si(CH3)3), 5.1/5.4 (s, Si(CH3)3), 27.5 (s, C(CH3)3), 35.1 (d, 1JP,C = 56.3 Hz, C(CH3)3), 139.9 (d, 1JP,C = 41.4 Hz, GaC[double bond, length as m-dash]CP), 206.7 (d, 2JP,C = 5.9 Hz, GaC[double bond, length as m-dash]CP); a signal for GaCH could not be observed. 29Si{1H} NMR (99 MHz, C6D6): δ [ppm] = −8.2 (s), −1.6 (s). 31P{1H} NMR (202 MHz, C6D6): δ [ppm] = 84.8 (s).
Bis2(Cl)Ga–O–P(Bn)tBu2 (11). Using benzyl chloride (10 μL, 9.1 mg, 72 μmol, 1.0 equiv.), stirring at 70 °C and after an additional washing step with n-hexane, 11 was obtained as a light yellow solid (44 mg, 65 μmol, 94%). 1H NMR (500 MHz, C6D6): δ [ppm] = −0.06 (s, 2H, GaCH), 0.36 (s, 36H, Si(CH3)3), 1.06 (d, 3JP,H = 13.6 Hz, 18H, C(CH3)3), 3.40 (d, 3JP,H = 14.0 Hz, 2H, CH2), 7.00 (t, 3JH,H = 7.3 Hz, 1H, para-H), 7.05 (t, 3JH,H = 7.6 Hz, 2H, meta-H), 7.39 (d, 3JH,H = 7.9 Hz, 2H, ortho-H). 13C{1H} NMR (126 MHz, C6D6): δ [ppm] = 4.3 (s, Si(CH3)3), 15.8 (s, GaCH), 27.5 (s, C(CH3)3), 31.2 (d, 1JP,C = 48.4 Hz, CH2), 36.7 (d, 1JP,C = 56.3 Hz, C(CH3)3), 127.0 (d, 4JP,C = 2.5 Hz, para-C), 128.7 (s, meta-C), 130.8 (d, 3JP,C = 5.0 Hz, ortho-C), 134.3 (d, 2JP,C = 6.0 Hz, ipso-C). 29Si{1H} NMR (99 MHz, C6D6): δ [ppm] = −1.3 (s). 31P{1H} NMR (202 MHz, C6D6): δ [ppm] = 63.6 (s). Elemental analysis calcd (%) for C29H63GaOPSi4 (Mr = 676.31): C 51.50, H 9.39; found C 51.67, H 9.59.

Author contributions

J. Buth: investigation, methodology, validation, visualization, and writing (original draft); B. Neumann, J.-H. Lamm and H.-G. Stammler: investigation (SCXRD); Y. V. Vishnevskiy (quantum chemical calculations) and N. W. Mitzel: funding acquisition, project administration, supervision, writing, reviewing and editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information (SI): experimental details, NMR spectra, crystallographic details, quantum chemical calculations (coordinates provided in a separate file), and references. See DOI: https://doi.org/10.1039/d6qi00380j.

CCDC 2528994–2529001 (for compounds 1, 2, 7, 8, 9a, 10a and 11) contain the supplementary crystallographic data for this paper.41a–h

Acknowledgements

The authors thank Marco Wißbrock and Dr Andreas Mix for recording NMR spectra, Barbara Teichner for performing elemental analyses and Hannah Koch for providing support with syntheses.

This work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation, grant MI 477/44-1, project no. 461833739 and grant VI 713/3-1, project no. 243500032). We acknowledge support by the Paderborn Center for Parallel Computing (PC2, HPC system Noctua 2) and by the Regional Computing Centre of the University of Cologne (RRZK, HPC system RAMSES) for providing computing time.

References

  1. D. W. Stephan, G. C. Welch, R. R. S. Juan and J. D. Masuda, Reversible, Metal-Free Hydrogen Activation, Science, 2006, 314, 1124–1126 CrossRef PubMed.
  2. D. W. Stephan, Frustrated Lewis Pairs, J. Am. Chem. Soc., 2015, 137, 10018–10032 CrossRef CAS PubMed.
  3. D. W. Stephan, Frustrated Lewis Pairs: From Concept to Catalysis, Acc. Chem. Res., 2015, 48, 306–316 CrossRef CAS PubMed.
  4. A. R. Jupp and D. W. Stephan, New Directions for Frustrated Lewis Pair Chemistry, Trends Chem., 2019, 1, 35–48 CrossRef CAS.
  5. N. Li and W. Zhang, Frustrated Lewis Pairs: Discovery and Overviews in Catalysis, Chin. J. Chem., 2020, 38, 1360–1370 CrossRef CAS.
  6. R. F. Heck, Acylation, methylation, and carboxyalkylation of olefins by Group VIII metal derivatives, J. Am. Chem. Soc., 1968, 90, 5518–5526 CrossRef CAS.
  7. W. S. Knowles and M. J. Sabacky, Catalytic asymmetric hydrogenation employing a soluble, optically active, rhodium complex, Chem. Commun., 1968, 1445 RSC.
  8. R. H. Grubbs and T. K. Brunck, Possible intermediate in the tungsten-catalyzed olefin metathesis reaction, J. Am. Chem. Soc., 1972, 94, 2538–2540 CrossRef CAS.
  9. S. Baba and E. Negishi, A novel stereospecific alkenyl-alkenyl cross-coupling by a palladium- or nickel-catalyzed reaction of alkenylalanes with alkenyl halides, J. Am. Chem. Soc., 1976, 98, 6729–6731 CrossRef CAS.
  10. T. Katsuki and K. B. Sharpless, The first practical method for asymmetric epoxidation, J. Am. Chem. Soc., 1980, 102, 5974–5976 CrossRef CAS.
  11. N. Miyaura and A. Suzuki, Palladium-Catalyzed Cross-Coupling Reactions of Organoboron Compounds, Chem. Rev., 1995, 95, 2457–2483 CrossRef CAS.
  12. C. P. Johnston, R. T. Smith, S. Allmendinger and D. W. C. MacMillan, Metallaphotoredox-catalysed sp3−sp3 cross-coupling of carboxylic acids with alkyl halides, Nature, 2016, 536, 322–325 CrossRef CAS PubMed.
  13. A. Y. Chan, I. B. Perry, N. B. Bissonnette, B. F. Buksh, G. A. Edwards, L. I. Frye, O. L. Garry, M. N. Lavagnino, B. X. Li, Y. Liang, E. Mao, A. Millet, J. V. Oakley, N. L. Reed, H. A. Sakai, C. P. Seath and D. W. C. MacMillan, Metallaphotoredox: The Merger of Photoredox and Transition Metal Catalysis, Chem. Rev., 2022, 122, 1485–1542 CrossRef CAS PubMed.
  14. J. Paradies, Mechanisms in Frustrated Lewis Pair-Catalyzed Reactions, Eur. J. Org. Chem., 2019, 283–294 CrossRef CAS.
  15. R. Zhou, Z. P. Tavandashti and J. Paradies, Frustrated Lewis Pair Catalyzed Reactions, SynOpen, 2023, 7, 46–57 CrossRef CAS.
  16. G. Ménard and D. W. Stephan, Room Temperature Reduction of CO2 to Methanol by Al-Based Frustrated Lewis Pairs and Ammonia Borane, J. Am. Chem. Soc., 2010, 132, 1796–1797 CrossRef PubMed.
  17. J. Chen, L. Falivene, L. Caporaso, L. Cavallo and E. Y.-X. Chen, Selective Reduction of CO2 to CH4 by Tandem Hydrosilylation with Mixed Al/B Catalysts, J. Am. Chem. Soc., 2016, 138, 5321–5333 CrossRef CAS PubMed.
  18. M. N. Khan, Y. Van Ingen, T. Boruah, A. McLauchlan, T. Wirth and R. L. Melen, Advances in CO2 activation by frustrated Lewis pairs: from stoichiometric to catalytic reactions, Chem. Sci., 2023, 14, 13661–13695 RSC.
  19. G.-Q. Chen, F. Türkyilmaz, C. G. Daniliuc, C. Bannwarth, S. Grimme, G. Kehr and G. Erker, Enamine/butadienylborane cycloaddition in the frustrated Lewis pair regime, Org. Biomol. Chem., 2015, 13, 10477–10486 RSC.
  20. G. C. Welch and D. W. Stephan, Facile Heterolytic Cleavage of Dihydrogen by Phosphines and Boranes, J. Am. Chem. Soc., 2007, 129, 1880–1881 CrossRef CAS PubMed.
  21. C. M. Mömming, E. Otten, G. Kehr, R. Fröhlich, S. Grimme, D. W. Stephan and G. Erker, Reversible Metal-Free Carbon Dioxide Binding by Frustrated Lewis Pairs, Angew. Chem., Int. Ed., 2009, 48, 6643–6646 CrossRef PubMed.
  22. E. Otten, R. C. Neu and D. W. Stephan, Complexation of Nitrous Oxide by Frustrated Lewis Pairs, J. Am. Chem. Soc., 2009, 131, 9918–9919 CrossRef CAS PubMed.
  23. D. W. Stephan and G. Erker, Frustrated Lewis pair chemistry of carbon, nitrogen and sulfur oxides, Chem. Sci., 2014, 5, 2625–2641 RSC.
  24. Z. Mo, E. L. Kolychev, A. Rit, J. Campos, H. Niu and S. Aldridge, Facile Reversibility by Design: Tuning Small Molecule Capture and Activation by Single Component Frustrated Lewis Pairs, J. Am. Chem. Soc., 2015, 137, 12227–12230 CrossRef CAS PubMed.
  25. D. Zhu, Z.-W. Qu and D. W. Stephan, Addition reactions and diazomethane capture by the intramolecular P–O–B FLP: tBu2POBcat, Dalton Trans., 2020, 49, 901–910 RSC.
  26. F. Breher, F. Krämer, J. Paradies and I. Fernández, A crystalline aluminium–carbon-based ambiphile capable of activation and catalytic transfer of ammonia in non-aqueous media, Nat. Chem., 2024, 16, 63–69 CrossRef PubMed.
  27. M. Shang, X. Wang, S. M. Koo, J. Youn, J. Z. Chan, W. Yao, B. T. Hastings and M. Wasa, Frustrated Lewis Acid/Brønsted Base Catalysts for Direct Enantioselective α-Amination of Carbonyl Compounds, J. Am. Chem. Soc., 2017, 139, 95–98 CrossRef CAS PubMed.
  28. D. W. Stephan and G. Erker, Frustrated Lewis Pair Chemistry: Development and Perspectives, Angew. Chem., Int. Ed., 2015, 54, 6400–6441 CrossRef CAS PubMed.
  29. C. Appelt, H. Westenberg, F. Bertini, A. W. Ehlers, J. C. Slootweg, K. Lammertsma and W. Uhl, Geminal Phosphorus/Aluminum-Based Frustrated Lewis Pairs: C[triple bond, length as m-dash]H versus C[triple bond, length as m-dash]C Activation and CO2 Fixation, Angew. Chem., Int. Ed., 2011, 50, 3925–3928 CrossRef CAS PubMed.
  30. R. G. Pearson, Hard and Soft Acids and Bases, J. Am. Chem. Soc., 1963, 85, 3533–3539 CrossRef CAS.
  31. H. Böhrer, N. Trapp, D. Himmel, M. Schleep and I. Krossing, From unsuccessful H2-activation with FLPs containing B(Ohfip)3 to a systematic evaluation of the Lewis acidity of 33 Lewis acids based on fluoride, chloride, hydride and methyl ion affinities, Dalton Trans., 2015, 44, 7489–7499 RSC.
  32. J. Possart and W. Uhl, Influence of the Lewis Acidity of Gallium Atoms on the Reactivity of a Frustrated Lewis Pair: Experimental and Theoretical Studies, Organometallics, 2018, 37, 1314–1323 CrossRef CAS.
  33. Y. Wang, Z. H. Li and H. Wang, Synthesis of an oxygen-linked germinal frustrated Lewis pair and its application in small molecule activation, RSC Adv., 2018, 8, 26271–26276 RSC.
  34. J. Buth, Y. V. Vishnevskiy, J.-H. Lamm, B. Neumann, H.-G. Stammler and N. W. Mitzel, Tuning reactivity through implementation of the HSAB concept in oxygen- and sulphur-bridged Al/P and Ga/P FLPs, Dalton Trans., 2026, 55, 2691–2701 RSC.
  35. L. Wickemeyer, N. Aders, A. Mix, B. Neumann, H.-G. Stammler, J. J. Cabrera-Trujillo, I. Fernández and N. W. Mitzel, Carbon dioxide reduction by an Al–O–P frustrated Lewis pair, Chem. Sci., 2022, 13, 8088–8094 RSC.
  36. J. Buth, M. J. Klingsiek, Y. V. Vishnevskiy, A. Mix, J.-H. Lamm, B. Neumann, H.-G. Stammler and N. W. Mitzel, Aluminum versus Gallium: Differences in Bis(bis(trimethylsilyl)methyl) Element Halides, Hydrides, and Oxygen-Bridged E/P Frustrated Lewis Pairs, ChemistryEurope, 2026, 4, e202500307 CrossRef CAS.
  37. L. Xing, W. Li, C. Wang, F. Gu, M. Xu, C. Tan and J. Yi, Theoretical Investigations on Oxidative Stability of Solvents and Oxidative Decomposition Mechanism of Ethylene Carbonate for Lithium Ion Battery Use, J. Phys. Chem. B, 2009, 113, 16596–16602 CrossRef CAS PubMed.
  38. Z. Lu, Z. Li, J. Duan, X. Li, C. Liu and W. Xiao, Revisit coupling of CO2 to ethylene carbonate with an integrated imidazolium and zinc halides catalyst by a study on its decomposition: Active center and mechanism, Chem. Eng. J., 2024, 495, 153687 CrossRef CAS.
  39. L. Wickemeyer, L. Hartmann, B. Neumann, H. Stammler and N. W. Mitzel, Differences in the Reactivity of Geminal Si−O−P and Al−O−P Frustrated Lewis Pairs, Chem. – Eur. J., 2023, 29, e202202842 CrossRef CAS PubMed.
  40. W. Uhl, H. R. Bock, M. Claesener, M. Layh, I. Tiesmeyer and E.-U. Würthwein, cis/trans Isomerism of Hydroalumination and Hydrogallation Products—Reflections on Stability and Rearrangement Mechanism, Chem. – Eur. J., 2008, 14, 11557–11564 CrossRef CAS PubMed.
  41. (a) CCDC 2528994: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qwmgq; (b) CCDC 2528995: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qwmhr; (c) CCDC 2528996: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qwmjs; (d) CCDC 2528997: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qwmkt; (e) CCDC 2528998: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qwmlv; (f) CCDC 2528999: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qwmmw; (g) CCDC 2529000: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qwmnx; (h) CCDC 2529001: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2qwmpy.

This journal is © the Partner Organisations 2026
Click here to see how this site uses Cookies. View our privacy policy here.