Extension of the Pnictinous Acid Series: Rational Synthesis of a Stibinous Acid and Exploration of Its Reactivity

Abstract

The oxyacids of the heavy group 15 elements remain underexplored despite the important role that main-group oxyacids play in academic and industrial chemistry. Higher-valent oxyacids of heavier main-group elements readily polymerize because of their ability to expand their coordination spheres and the polarity of their terminal E–O (E = P, As, Sb, Bi) bonds. As a result, no free R₂SbO₂H stibinic acids have yet been prepared. Lower-valent R₂SbOH stibinous acids should be less susceptible to such polymerization, but there are no reported rational syntheses of molecules belonging to this class. We report here the first controlled synthesis and isolation of a molecular stibinous acid, Dipp₂SbOH, where Dipp = 2,6-diisopropylphenyl. We also report an initial exploration of the chemistry of this class of compounds, including their Brønsted acidity and Lewis basicity. The hydroxyl group of Dipp₂SbOH is weakly acidic as probed via reaction with the very strong base DBU. We also prepared the analogous arsinous acid Dipp₂AsOH and demonstrated that it can support oxidation to the well-defined monomeric arsinic acid Dipp₂AsO₂H. We therefore investigated the oxidation of Dipp₂SbOH. Reaction with PhIO and H₂O₂ produced oxo-bridged clusters, suggesting the formation of a transient stibinic acid of the form Dipp₂SbO₂H.

Introduction

The bond between an O atom and an H atom can be very strong, with the bond dissociation enthalpy of the O–H bond in methanol (DH₂₉₈ = 104.2(9) kcal mol^–1) being greater than that of its C–H bonds (DH₂₉₈ = 96.06(15) kcal mol^–1). Although strong, the large electronegativity discrepancy between oxygen and hydrogen renders the O–H unit highly polar, contributing to the protic nature of hydroxy-substituted molecules. Examples include alcohols (R₃C–OH), phosphinous acids (R₂P–OH), sulfenic acids (RS–OH), and hypochlorous acid (Cl–OH). If the negative charge of the conjugate base formed upon deprotonation of these R _n E–OH species can be effectively distributed, then the substance can exhibit appreciable acidity. Resonance delocalization provides a highly effective means of distributing the negative charge and the presence of terminal O atoms attached to the E atom substantially increases acidity. From the list above, such compounds would include carboxylic acids (RCO₂H), phosphinic acids (R₂PO₂H), sulfinic acids (RSO₂H), and chlorous acid (ClO₂H). Each additional terminal O atom across which the negative charge can delocalize lowers the pK _a by approximately 5 units. These compounds are referred to as oxyacids or oxo acids.

Within group 15, the oxyacids have been comprehensively evaluated for N through As, but for Sb and Bi many gaps in fundamental understanding remain. Prior work has demonstrated that compounds containing terminal oxo groups bound to an Sb­(V) center readily polymerize but that sterically bulky groups can prevent this self-association process. − This behavior has been observed for Sb­(V) oxyacids as well, with stibonic acids existing as large cluster compounds. − Even large terphenyl groups have so far only been able to limit this association to dimerization. − The mechanism of this polymerization likely relies on the Lewis basic O atom of one molecule binding to the Lewis acidic Sb­(V) center of a separate molecule, which then primes the O atom of that molecule to attack a third, and so on. The situation for Sb­(III) oxyacids remains less clear. The Sb­(III) center in a putative stibinous acid would be significantly less Lewis acidic than an Sb­(V) center. Although this Lewis acidity could mediate secondary interactions such as pnictogen bonding, it is not likely to be strong enough to drive polymerization of the type seen with Sb­(V) compounds. Two molecules of R₂SbOH can, however, condense to eliminate water and form (R₂Sb)₂O with a bridging oxide, which is formally an anhydride of stibinous acid. Such species were observed to form spontaneously from the aerial oxidation of Sb­(II)-containing tetraaryldistibanes (R₂Sb)₂ when the aryl groups were phenyl, para-tolyl, or ortho-tolyl. It is noteworthy that the only known example of a well-characterized free stibinous acid has one very bulky substituent and all other examples of stibinous acids feature them associated with Lewis acids (Scheme ). − This collection of findings suggests that a kinetic stabilization approach may be helpful in accessing stibinous acids in a rational manner.

1. Prior Examples of Stibinous Acids.

We report here the rational synthesis, isolation, and characterization of Dipp₂SbOH, where Dipp = 2,6-diisopropylphenyl. The Dipp group was selected to provide a balance of kinetic stabilization while still allowing access to reactivity. We conducted an initial exploration of its Lewis basicity and Brønsted acidity. In solution, an equilibrium could be established between Dipp₂SbOH and the very strong Brønsted base 1,8-diazabicyclo[5.4.0]­undec-7-ene (DBU). In the solid state, a strong H-bond was established. A computational investigation of the conjugate base Dipp₂SbO^– demonstrated a significant Pn–O bond contraction and an electronic structure analysis provided an explanation for this phenomenon that was consistent with the computationally optimized structures of Dipp₂PnOH and Dipp₂PnO^– for Pn = P, As, and Sb, as well as experimental values from the limited literature on phosphinous and arsinous acids. For comparison, we also prepared the arsinous acid Dipp₂AsOH. This arsinous acid could be oxidized to the monomeric arsinic acid Dipp₂AsO₂H, which motivated an exploration of the oxidation of Dipp₂SbOH. The oxo-bridged species produced during the oxidation of Dipp₂SbOH are consistent with the formation of a transient stibinic acid of the form Dipp₂SbO₂H.

Results and Discussion

Computational Prediction of Pnictinous Acid Stability

Phosphinous acids of the type R₂POH, are well-known to favor the secondary phosphine oxide tautomer. , Indeed, R₂POH compounds can only be obtained in the phosphinous acid form if strongly electron-withdrawing R groups are used. − Possibly in analogy to this phosphorus-based work, the only two prior examples of R₂AsOH compounds that have been rigorously confirmed to exist in the arsinous acid form also bear strongly electron-withdrawing groups. , These arsinous acids were di([5]­trovacenyl)­arsinous acid and (F₅C₂)₂AsOH (vide infra), which raised the question of whether Dipp groups would favor the acid forms of the heavy pnictinous acids or whether electron-withdrawing groups would be needed for the whole family. Thermochemical analysis of computationally optimized structures of the different tautomers confirms that for Dipp₂POH, the pnictine oxide form is favored, whereas for Dipp₂AsOH and Dipp₂SbOH the pnictinous acid is favored (Table ). The pnictinous acid form is increasingly stabilized as the atomic number of the pnictine increases. This variation in equilibrium is consistent with a lesser extent of delocalization of O-based lone pairs into Pn–C antibonding orbitals as the Pn atomic number increases. The formation of weaker Pn⁺–O^– and Pn–H bonds with the heavier Pn elements simply does not compensate for the cleavage of the O–H bond for R₂AsOH and R₂SbOH. With the prediction that these species would be stable in the pnictinous acid form, we targeted the synthesis of Dipp₂AsOH.

1. Gibbs Free Energies of Secondary Pnictine Oxide/Pnictinous Acid Tautomerism of Dipp ₂ PnOH (Pn = P, As, Sb) .

Pn	ΔGrxn (kcal mol–1)
P	10.1
As	–21.7
Sb	–41.9

^aCalculated at the PBE0/def2-TZVPP level of theory.

Synthesis of Dipp₂AsOH and Dipp₂AsO₂H

Combination of DippBr and n-BuLi generates the corresponding aryllithium species in situ. Addition of 0.5 equiv of AsCl₃ to this aryllithium solution produces the diarylchloroarsine Dipp₂AsCl (Scheme ). Subsequent reaction of Dipp₂AsCl with NaOH afforded the arsinous acid Dipp₂AsOH. Minor shifts occur in the Dipp NMR resonances, but the product features a signal for the O-bound proton at 5.63 ppm in DMSO-d ₆. Treatment with H₂O₂ converted Dipp₂AsOH to the arsinic acid Dipp₂AsO₂H, which has a new diagnostic ν_AsO stretching band at 880 cm^–1 in the IR spectrum.

2. Synthesis of Dipp₂AsOH and Dipp₂AsO₂H, and Neutralization of Dipp₂AsO₂H with DBU.

Dipp₂AsOH and Dipp₂AsO₂H were both characterized by single-crystal X-ray diffraction (Figure ). One obvious, but nonetheless noteworthy feature of the structure of Dipp₂AsOH is that it exists as the arsinous acid and not as the secondary arsine oxide tautomer, consistent with the computational prediction described above. The crystal structure of Dipp₂AsOH features two molecules in the asymmetric unit that are related by a pseudoinversion center. The lower-symmetry space group P2₁ was ultimately selected over P2₁/c to afford a chemically reasonable hydrogen bonding description and because the Flack parameter refined to be statistically significantly different from 0.5. The pseudosymmetry produces large correlation matrix elements between the two molecules, however, and so similarity distance restraints were employed to avoid refinement artifacts. The molecules feature trigonal pyramidal As centers with As–OH bond lengths of 1.799(7) and 1.811(7) Å. These values agree well with those for the only two previously characterized free arsinous acids. The first to be structurally characterized was di([5]­trovacenyl)­arsinous acid, in which the organic substituent is an organometallic vanadium sandwich complex. In the crystal structure of di([5]­trovacenyl)­arsinous acid, there are four copies of the molecule in the asymmetric unit and the As–O bond lengths range from 1.793(3) to 1.813(3) Å. It was only very recently that the second structure of a free arsinous acid, (F₃C₂)₂AsOH, was reported. In this species, the As–O bond length is 1.766(1) Å.

1.

Molecular structures of (a) Dipp₂AsCl, (b) Dipp₂AsOH, (c) Dipp₂AsO₂H, and (d) (H-DBU)­[Dipp₂AsO₂]. Disorder and C-bound H atoms omitted for clarity. In instances where Z′ > 1, only one molecule is shown. Color code: As purple, Cl green, O red, N blue, C black, H white.

The oxidized tetrahedrally substituted Dipp₂AsO₂H crystallizes with a reciprocal H-bonded dimer in the asymmetric unit, a common solid-state motif for carboxylic acids and phosphinic acids. The product of the reaction between DBU and Dipp₂AsO₂H features the expected tetrahedral geometry about the As center, nearly equivalent As–O bond lengths, and a strong N–H···O hydrogen bond with an N···O distance of 2.637(3) Å.

Synthesis of Dipp₂SbOH

Encouraged by the well-behaved nature of Dipp₂AsOH and its capacity to readily support oxidation to the corresponding arsinic acid, we turned our attention to the synthesis of the stibinous acid Dipp₂SbOH. To date, there have been no rational syntheses of stibinous acids. As noted above, the only well-characterized examples have arisen as decomposition side-products formed during the investigation of other compounds. ,, In analogy to our synthesis of Dipp₂AsOH, we allowed two equivalents of DippLi to react with SbCl₃ and yield the chlorostibine Dipp₂SbCl as a bright yellow solid (Scheme and Figure ). This organolithium approach provides a nice complement to the organomagnesium approach that was used to prepare the related chlorostibine bearing two 2,4,6-triisopropylphenyl groups. Dipp₂SbCl did not undergo hydrolysis during the course of the aqueous synthetic workup and also survived open-air silica-gel chromatography.

3. Synthesis of Stibinous Acid Dipp₂SbOH.

2.

Molecular structures of (a) Dipp₂SbCl and (b) Dipp₂SbOH. Disorder and C-bound H atoms omitted for clarity. Color code: Sb teal, Cl green, O red, C black, H white.

Reaction with NaOH converted the yellow Dipp₂SbCl into a colorless solid, the IR spectrum of which showed a new band at 554 cm^–1. This band is consistent with ν_SbO for an Sb–O single bond. The IR data also showed a sharp band at 3550 cm^–1 that we assign to ν_OH. Single-crystal X-ray diffraction confirmed that this colorless compound was the stibinous acid Dipp₂SbOH (Figure ). As with Dipp₂AsOH, the space group of P2₁, the refinement of the Flack parameter to 0.474(6), and the presence of two copies of the molecule in the asymmetric unit all raise the possibility of missed higher symmetry in the crystal structure. Again, however, the symmetry relating the two molecules in the asymmetric unit was only local. The pseudoinversion center is present at (0.002, 0.499, 0.261) and none of the correlation matrix elements are large enough to indicate complications from this pseudosymmetry. The two Sb–O bond lengths are 1.978(4) and 1.986(4) Å.

Only one other free stibinous acid, 1 (Scheme ), has been structurally characterized before. Unfortunately, the report describes complications during the refinement. The systematic absences were consistent with space group P2₁/c, but the structure was ultimately refined in Pc with Z′ = 2. The authors refined the structure as a racemic twin and modeled the Sb–OH units of both molecules in the asymmetric unit as disordered. Nevertheless, inspection of pseudosymmetrically related anisotropic displacement parameters and bond metrics suggest that strong anticorrelations have skewed the model. The Sb–O bond lengths for the two molecules in the asymmetric unit are 1.814(13) and 2.065(5) Å. This large difference is chemically unrealistic, but we note that the average of these two values, 1.94 Å, agrees nicely with the values we obtained for Dipp₂SbOH.

The main-group adduct Ph₂SbOH·B­(C₆F₅)₃ (2, Scheme ) has an Sb–O bond length of 2.1086(10) Å. The Sb–O elongation, as compared to Dipp₂SbOH, is consistent with the weakening of this bond by interaction with the strong Lewis acid B­(C₆F₅)₃. For completeness, we also note that the only other stibinous acid structures, 3–5 (Scheme ), feature the Sb atom engaged in strong secondary interactions, including transition metal coordination, hydrogen bonding, and pnictogen bonding. − The Sb–O bond lengths of 1.943(5), 1.976(7), and 2.079(1) Å for 3 through 5, respectively, generally agree well with that of Dipp₂SbOH, but the large, varied, and simultaneous occurrence of different secondary interactions precludes a meaningful comparison at this time. It is also notable that stibinous acids with smaller substituents have, to date, only been obtained if there are strong interactions between the R₂SbOH species and a main-group or transition-metal Lewis acid; small substituents have afforded the (R₂Sb)₂O condensation products.

As with the arsinous acid described above, the computational data indicated that the presence of the compound in the stibinous acid form is not an artifact of solid-state stabilization but rather arises because this form is thermodynamically favored over the secondary stibine oxide. The thermodynamic preference is even greater than for the lighter congener (Table ). This structure maintains in solution and a ¹H NMR signal for the OH proton was observed at 4.22 ppm in DMSO-d ₆. Although the aromatic and alkyl resonances from the Dipp groups were relatively insensitive to solvent, the OH signal shifted to 2.23 ppm in CD₃CN, 1.01 ppm in CD₂Cl₂, and 0.93 ppm in CDCl₃. In addition to this strong solvent dependence, the signal was confirmed to come from the exchangeable OH motif through its disappearance upon addition of D₂O (see Supporting Information).

Brønsted Acidic Reactivity of Dipp₂SbOH

With the free stibinous acid in hand, we conducted a preliminary investigation of its chemistry. Given the prior lack of a reliable means of accessing stibinous acids, this is the first such exploration. As the stibinous acid moniker implies, we anticipated that the molecule would act as a Brønsted acid. Combination with the weak base Ph₃P in CD₂Cl₂ resulted in a negligible change in the hydroxyl ¹H NMR signal (Δδ = 0.02 ppm), indicating at most a very weak H-bonding interaction (Scheme ). The stronger base Et₃N resulted in a larger shift (Δδ = 0.3 ppm), consistent with a stronger, albeit still weak, reversible H-bonding interaction (Scheme ). The much stronger non-nucleophilic base DBU, resulted in a much more substantial effect (Figures S40 and S41). Half an equivalent of DBU resulted in a dramatic broadening of the signal with the fwhm increasing from 3 to 120 Hz (Figure S41). Further amounts of DBU result in the signal broadening out into the baseline, with it effectively disappearing upon addition of 4 equiv of DBU. This behavior is consistent with either strong reversible H-bonding in solution or dynamic protonation/deprotonation. The asymmetric unit of crystals grown from a 1:1 mixture of DBU and Dipp₂SbOH features two copies of strongly associated pairings of DBU and an Sb-based molecular fragment (Figure ). The short O···N distances of 2.772(2) and 2.754(3) Å indicate the strength of the association between the two and initially led us to think that a salt of the form (H-DBU)­[Dipp₂SbO] had formed. Close inspection of the difference Fourier synthesis of a model including all atoms except the acidic H atoms revealed the electron density maximum to lie closer to the O atom for both copies of the formula unit. The Sb–O bond lengths of 1.962(2) and 1.961(2) Å are statistically significantly shorter than the Sb–O bond lengths in the structure of Dipp₂SbOH (1.978(4) and 1.986(4) Å). We also note that similar results were obtained with Dipp₂AsOH (Figures S42 and S43).

4. Reactivity of Dipp₂SbOH with Brønsted Bases of Increasing Strength: (a) PPh₃, (b) NEt₃, and (c) DBU.

3.

Molecular structures of (a) Dipp₂SbOH·DBU and (b) trans-PdCl₂(Dipp₂SbOH)₂. Disorder and C-bound H atoms omitted for clarity. In instances where Z′ > 1, only one molecule is shown. Color code: Sb teal, Cl green, O red, N blue, Pt navy, C black, H white.

Pnictinite Electronic Structure

To understand the Sb–O bond length contraction observed in the crystal structure obtained from the reaction of Dipp₂SbOH and DBU, we computationally investigated both the stibinous acid and the stibinite conjugate base (Table ). The calculations reveal that there would indeed be an Sb–O bond contraction upon deprotonation of Dipp₂SbOH, but that it would be a much larger effect than that which we observed. This result further supported our decision to ultimately refine the structure as a cocrystal of Dipp₂SbOH and DBU. We were intrigued by the magnitude of the Sb–O length contraction that was predicted and so conducted a brief analysis of the electronic structures of these molecules. The Sb–O Wiberg bond index of Dipp₂SbOH is 0.61 and that of Dipp₂SbO^– is 0.96. This dramatic bond strengthening in the anion can be understood as arising from variation in the delocalization of O-based lone pairs into Sb–C antibonding orbitals. Indeed, the Sb–C antibonding NBOs of Dipp₂SbOH have occupancies of 0.07 and 0.06, whereas those of Dipp₂SbO^– both have occupancies of 0.11. If the Sb–O bond strengthening in Dipp₂SbO^– does indeed arise for such delocalizations, we would expect to see a lengthening of the Sb–C bonds in the crystallographically characterized H-bonded adduct. Although the effect is subtle, there is a bond lengthening with the Sb–C bonds of Dipp₂SbOH·DBU ranging from 2.190(2) to 2.198(2) Å and those of Dipp₂SbOH ranging from 2.181(5) to 2.193(4) Å.

2. Computationally Optimized Bond Metrics and NBO Parameters of Pnictinous Acids and Pnictinite Anions .

	Pn–O calculated bond length (Å)	Pn–O WBI	population of Pn–C antibonding orbitals	Pn–C calculated bond lengths (Å)
Dipp2SbOH	1.993	0.61	0.06, 0.07	2.193, 2.175
Dipp2SbO–	1.863	0.96	0.11, 0.11	2.249, 2.260
Dipp2AsOH	1.814	0.69	0.06, 0.07	1.977, 1.996
Dipp2AsO–	1.685	1.01	0.11, 0.12	2.049, 2.060
Dipp2POH	1.662	0.75	0.06, 0.07	1.849, 1.860
Dipp2PO–	1.533	1.08	0.11, 0.12	1.906, 1.909

^aCalculated at the PBE0/def2-TZVPP level of theory.

We can further anticipate that, if this origin of the Sb–O bond contraction in Dipp₂SbO^– is correct, the effect would be greater in arsinite/arsinous acid pairs and even greater in phosphinite/phosphinous acid pairs. The trend is indeed born out in the structural and NBO parameters obtained from computationally optimized geometries (Table ). As noted above, there are limited data available on arsinous acids and there is only one, very recently reported, example of an arsinous acid/arsinite pair that has been structurally characterized. (F₅C₂)₂AsOH has an As–O bond length of 1.766(1) Å, and the diethylammonium salt of (F₅C₂)₂AsO^– has a contracted As–O bond length of 1.701(4) Å.

Extension of the comparison to phosphinous acids is complicated because the tautomerism that interconverts phosphinous acids and secondary phosphine oxides generally greatly favors the latter. The use of electron-withdrawing perfluoroalkyl groups has permitted isolation of the only two phosphinous acids to date: (F₃C)₂POH and (F₅C₂)₂POH. − These volatile liquids have been structurally characterized by gas-phase electron diffraction but, for the sake of comparison, we will restrict ourselves to comparison of X-ray diffraction data. , The P–O bond length is 1.629(1) Å for (F₅C₂)₂POH and a range of 1.627(1) to 1.631(1) Å was obtained for (F₃C)₂POH because there are multiple copies of the acid in the asymmetric unit. Unfortunately, the phosphinite conjugate bases of these acids have not yet been explored. There are, however, many alkali metal and alkaline earth metal phosphinite salts that have been structurally characterized. Across a wide range of organic substituents and counterions, a fairly narrow range of P–O bond lengths emerge: 1.533(2) to 1.588(2) Å. − It is interesting to note that, because of the tautomerization shown in Table , these phosphinites are typically generated by the deprotonation of secondary phosphine oxides, which is implicated as a key step in the Pudovik reaction.

These data collectively indicate that there is contraction of the Pn–O bond length upon deprotonation of a pnictinous acid and that the effect is greater for the lighter congeners. The computational data support the proposal that the contraction arises from delocalization of the negative charge on the O atom of the R₂PnO^– pnictinite into Pn–C antibonding orbitals. The poorer energy match of the O-based lone pairs and the antibonding orbitals results in decreased delocalization of negative charge in the heavier congeners. The heavier pnictinites will therefore have a greater charge accumulation on the terminal O atom. This situation strongly parallels that observed for the Pn­(V) pnictine oxide species. ,

Lewis Basic Reactivity of Dipp₂SbOH

We next probed the Lewis basicity of Dipp₂SbOH with a pair of solution-phase acid–base association reactions (Scheme ). The presence of lone pairs on both the Sb and O atoms leaves open the possibility that either could act as a locus of Lewis basicity. Addition of excess B­(C₆F₅)₃ to a CDCl₃ solution of Dipp₂SbOH generates a new species with a single new set of shifted ¹H NMR resonances. Most notably, the OH signal shifts from 1.53 to 5.55 ppm (Figures S44 and S45). The ¹⁹F and ¹¹B NMR spectra show residual excess B­(C₆F₅)₃ and a new four-coordinate B-containing species with Δ _m–p (¹⁹F) = 6.9 ppm and δ­(¹¹B) = 1.16 ppm (Figures S46–S49). These features are consistent with the prior report of the formation of an adduct between B­(C₆F₅)₃ and Ph₂SbOH, which exhibited an OH ¹H NMR signal at 4.97 ppm, δ­(¹¹B) = −0.53 ppm Δ_m–p(¹⁹F) = 6.6 ppm in C₆D₆. Combination of Dipp₂SbOH with the less electrophilic Lewis acid B­(C₆H₅)₃ results in no reaction although slight shifts and mild broadening of the ¹¹B and ¹H NMR resonances could indicate a weak reversible interaction. This difference in behavior is consistent with B­(C₆F₅)₃ overcoming the steric bulk of the Dipp groups to form a strong Lewis adduct whereas the weaker Lewis acid interacts minimally.

5. Lewis Basic Reactivity of Dipp₂SbOH with (a) B­(C₆F₅)₃ and (b) PdCl₂ .

To assess the ability of the lone pair on the Sb atom to engage in Lewis basic reactivity, we combined it with PdCl₂ and saw complete spectroscopic conversion to a new symmetric product. Single-crystal X-ray diffraction analysis revealed coordination to the metal center through the Sb atom (Figure ). Only the trans isomer was observed, which could indicate that the stibinous acid exhibits a stronger trans effect than Cl^– but could also arise from the size of the former. The difference in reactivity with the B- and Pd-based Lewis acids is consistent with the expectation informed by Hard–Soft Acid–Base theory: the hard O atom serves as the Lewis basic partner for the hard B atom in B­(C₆F₅)₃ whereas the soft Pd­(II) metal center interacts more readily with the lone pair on the softer Sb atom.

Oxidation of Dipp₂SbOH

Following our exploration of the acid/base chemistry of the monomeric stibinous acid Dipp₂SbOH, we next sought to determine whether it would serve as a viable precursor for a monomeric stibinic acid. Stirring a suspension of PhIO in a DCM solution of Dipp₂SbOH resulted in gradual consumption of the insoluble hypervalent iodine polymer and ¹H NMR spectroscopic analysis of the reaction mixture confirmed complete consumption of Dipp₂SbOH. New resonances at 7.26 and 3.68 ppm, which are shifted downfield from those of the Sb­(III) starting material, are consistent with oxidation of the Sb center. Attempts to isolate this material consistently afforded solids that, upon redissolution, returned substantially more complex NMR spectra, indicating that product was decomposing. One of these decomposition products that could be consistently obtained in small amounts was crystallographically characterized and shown to be an oxo-bridged species featuring a triangular arrangement of Sb atoms (Scheme and Figure ). Each Sb center has a highly distorted octahedral geometry, binding to two Dipp groups and four O atoms. Two of these O atoms each bridge to a single adjacent Sb center, and the other two O atoms bridge to both adjacent Sb centers. The three μ₂ O atoms bridge along the edges of the Sb₃ triangle and the two μ₃ O atoms cap the two triangular faces. The structure is consistent with the oxidation of the Sb atoms to the +5 oxidation state. The lack of any counterions in the structure indicates that the molecule is neutral which would require that one of the bridging O atoms be a hydroxide and that the others are all oxides. We refined the structure with an H atom disordered across both μ₃ O atoms. We emphasize this point because the resulting formula Sb₃Dipp₆O₅H would be equal to 3­(Dipp₂SbO₂H)–H₂O, where Dipp₂SbO₂H is the targeted monomeric stibinic acid. This structure is therefore consistent with the proposal that a monomeric stibinous acid is a possible precursor to a monomeric stibinic acid, but that the Dipp groups simply are not big enough to prevent condensation (Scheme ).

6. Attempts to Obtain Stibinic Acid Dipp₂SbO₂H via PhIO Oxidation of Dipp₂SbOH or Dipp₂SbCl.

4.

Molecular structures of products obtained from the oxidation of Dipp₂SbOH (a and c) and Dipp₂SbCl (b). Disorder and C-bound H atoms omitted for clarity. For further clarity, in some instances, select Dipp groups are shown as lines. In instances where Z′ > 1, only one molecule is shown. Color code: Sb teal, Cl green, O red, C black, H white.

We suspect that the species initially formed in the reaction and observed by NMR spectroscopy is a dimeric species analogous to the dimeric stibonic acid reported by Beckmann and co-workers. This proposal was supported by the oxidation of Dipp₂SbCl with PhIO, which yielded the dimeric species (Dipp₂SbClO)₂ with trigonal bipyramidal Sb centers, terminal chloride groups, bridging oxide groups, and a dioxadistibetane core (Scheme and Figure ). Treatment of this species with 2 equiv of NaOH in THF afforded a solution with NMR spectra identical to those obtained upon oxidation of Dipp₂SbOH with PhIO. The product would also similarly decompose, and crystals of the trimeric species could be obtained from reaction mixtures that were allowed to stand at room temperature.

If the oxidation of Dipp₂SbOH was performed with aqueous H₂O₂, no appreciable quantities of a clean product could be isolated. Small amounts of a crystalline product were, however, repeatedly obtained from these reactions. Single-crystal X-ray diffraction revealed them to contain a complex decomposition product featuring four Sb atoms (Scheme and Figure ). Two of the Sb centers have a distorted octahedral geometry and are bound to the 1 position of only one Dipp group. This Dipp group has undergone CH activation at one of the isopropyl groups and is additionally bound to the Sb via an isopropoxide unit. The two Sb centers bearing single activated Dipp groups are linked by a μ₂-O^2– ligand and a μ₂:η¹,η¹-O₂ ^2–. Although it could not be obtained in a preparative fashion, this structure is noteworthy because the coordination spheres of the two Sb centers described above are completed by two bridging Dipp₂SbO₂ ^– groups. These bridging groups feature a tetrahedral Sb center bound to two Dipp groups and two O atoms. This is the stibinate conjugate base of the stibinic acid that has been targeted (Scheme ). These results are collectively consistent with the formation of monomeric intermediates, but we stress that such species have not yet been observed. We believe that Dipp groups may be at the cusp of being sufficiently bulky to prevent oligomerization.

7. Attempt to Obtain Stibinic Acid Dipp₂SbO₂H via H₂O₂ Oxidation of Dipp₂SbOH.

Conclusion

A robust route to access a stibinous acid, Dipp₂SbOH, has been developed, which permitted the properties and reactivity of this class of molecules to be preliminarily probed for the first time. The bulky Dipp groups appear to play a role in limiting condensation to form the anhydride (R₂Sb)₂O. Dipp₂SbOH exists in the pnictinous acid form as opposed to the secondary pnictine oxide, which is commonly observed for phosphinous acids. The hydroxyl group is only weakly acidic, not undergoing full deprotonation, even when treated with the strong base DBU. It can act as a Lewis base through either the O atom or the Sb atom depending on whether it interacts with a hard or soft Lewis acid. Oxidation does afford Sb­(V) compounds, which we speculate proceeds through monomeric stibinic acids, but the Dipp groups do not appear to be bulky enough to prevent oligomerizing self-association. Analysis of the products formed suggests, however, that this approach could be a viable one for accessing well-defined stibinic acids.

Experimental Section

Caution! n-Butyllithium is a pyrophoric substance and must be handled using proper needle and syringe techniques. All manipulations were performed on the smallest practical scale.

General Considerations

Reagents and solvents were purchased from commercial vendors and used as received unless otherwise specified. B­(C₆F₅)₃ was obtained from Combi-Blocks and purified by recrystallization from pentane. After recrystallization, an impurity remained and is identified in the ¹⁹F and ¹¹B NMR spectra shown in Figures S46–S49. This impurity does not impact the chemistry described. All reactions were performed under a N₂ atmosphere in an OMNI-lab glovebox or on a dual-manifold Schlenk line unless otherwise specified. All solvents were dried over 3 Å molecular sieves. NMR spectra were collected at room temperature using a Bruker Avance III HD 500 spectrometer equipped with a multinuclear Smart Probe. ¹H and ¹³C NMR signals are reported in ppm as chemical shifts from tetramethylsilane. These signals are referenced to the solvent (¹H: benzene-d ₅ δ = 7.16, DCM-d ₁ δ = 5.32, MeCN-d ₂ δ = 1.94, DMSO-d ₅ δ = 2.50, CHCl₃ δ = 7.26; ¹³C: benzene-d ₆ δ = 128.06, DCM-d ₂ δ = 53.84, MeCN-d ₃ δ = 1.32, 118.26, DMSO-d ₆ δ = 39.52, CDCl₃ δ = 77.16). ¹¹B signals are reported in ppm as chemical shifts from BF₃·OEt₂ (referenced to an external sample of BF₃·OEt₂). ¹⁹F NMR signals are reported in ppm as chemical shifts from CFCl₃ (referenced to an external sample of BF₃·OEt₂ at −152.8 ppm). ³¹P NMR signals are reported in ppm as chemical shifts from 85% H₃PO₄ (referenced to an external solution of triphenylphosphine in CDCl₃ at −5.35 ppm). NMR data were processed using Mestrenova (version 12.0.2). IR spectra were collected on KBr pellets by using a PerkinElmer Spectrum One FT-IR spectrometer. Elemental analyses were performed by NuMega Resonance Laboratories (San Diego, CA). The results of these combustion analyses are reported for transparency, but we note that reliability of this method has recently been called into question; , we present our spectroscopic and powder diffraction data as evidence of purity.

Synthesis of Dipp₂AsCl

2-Bromo-1,3-diisopropylbenzene (3.000 g, 12.44 mmol, 2.560 mL) was dissolved in Et₂O (approximately 20 mL). n-BuLi (12.44 mmol, 4.980 mL, 2.5 M in hexanes) was added via syringe at −78 °C and the mixture was left at −78 °C for 1 h before returning to room temperature. The reaction proceeded at room temperature for 5 h. AsCl₃ (1.130 g, 6.220 mmol, 0.5 mL) was added dropwise to the reaction mixture at −78 °C. The reaction mixture was then warmed to room temperature after 30 min and stirred overnight. The reaction mixture was opened to air, filtered to remove LiCl, and the filtrate was concentrated in vacuo to afford a yellow oily residue. The crude product was dissolved in a minimal amount of MeCN and stored at −30 °C for 2 d, yielding Dipp₂AsCl as light-yellow crystals (1.100 g, 41% yield). Anal. Calcd for C₂₄H₃₄AsCl: C, 66.59; H, 7.92. Found: C, 62.32; H, 7.76. IR (KBr, cm ^–1 ) ν_CH (iPr) 2975–2864. ¹ H NMR (500 MHz, C ₆ D ₆ ) δ 7.12 (q, J = 7.4 Hz, 2H, ArH), 7.04–6.97 (m, 4H, ArH), 3.88–3.72 (m, 4H, CH(CH₃)₂), 1.06 (m, 24H, CH­(CH ₃)₂) ppm. ¹ H NMR (500 MHz, CD ₂ Cl ₂ ) δ 7.32–7.26 (m, 2H, ArH), 7.17–7.12 (m, 4H, ArH), 3.60 (m, 4H, CH(CH₃)₂), 1.05 (m, 24H, CH­(CH ₃)₂) ppm. ¹ H NMR (500 MHz, CDCl ₃ ) δ 7.30–7.26 (m, 2H), 7.15–7.10 (m, 4H), 3.61 (m, 4H), 1.05 (m, 24H) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, C ₆ D ₆ ) δ 152.26 (ArC), 142.14 (ArC), 140.84 (ArC), 130.35 (ArC), 125.03 (ArC), 33.10 (CH­(CH₃)₂), 24.92 (CH­(CH₃)₂), 24.17 (CH­(CH₃)₂) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, CD ₂ Cl ₂ ) δ 152.40 (ArC), 142.25 (ArC), 141.00 (ArC), 130.47 (ArC), 125.22 (ArC), 33.32 (CH­(CH₃)₂), 33.02 (CH­(CH₃)₂), 25.05 (CH­(CH₃)₂), 24.89 (CH­(CH₃)₂), 24.28 (CH­(CH₃)₂) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, CDCl ₃ ) δ 151.93 (ArC), 141.78 (ArC), 129.96 (ArC), 124.73 (ArC), 32.88 (CH­(CH₃)₂), 32.59 (CH­(CH₃)₂), 24.86 (CH­(CH₃)₂), 24.12 (CH­(CH₃)₂) ppm. Bulk purity was confirmed by PXRD (see Supporting Information).

Synthesis of Dipp₂AsOH

In open air, Dipp₂AsCl (100 mg, 0.231 mmol) was dissolved in THF (3 mL). Then, 3 mL of a 1 M NaOH_(aq) solution (2.31 mmol, excess) was added. The reaction was stirred for 25 min, then extracted with DCM (2 × 4 mL), dried over Na₂SO₄, filtered, and the solvent was evaporated under reduced pressure. The residue was recrystallized from a mixture of DCM and pentane to obtain Dipp₂AsOH as colorless crystals (65 mg, 68% yield). Crystals suitable for single-crystal X-ray diffraction were obtained at −24 °C from a mixture of dichloromethane and pentane. Anal. Calcd for C₂₄H₃₅AsO: C, 69.55; H, 8.51. Found: C, 57.86; H, 7.51. IR (KBr, cm ^‑1 ) ν_OH 3430, ν_CH (iPr) 2960–2870, ν_As–OH 758. ¹ H NMR (500 MHz, CDCl ₃ ) δ 7.30–7.27 (m, 2H, ArH), 7.13 (d, J = 7.7 Hz, 4H, ArH), 3.63 (hept, J = 6.7 Hz, 4H, CH(CH₃)₂), 1.68 (s, 1H, OH), 1.06 (m, 24H, CH­(CH ₃)₂) ppm. ¹ H NMR (500 MHz, DMSO–d ₆ ) δ 7.22 (t, J = 7.6 Hz, 2H, ArH), 7.08 (d, J = 7.6 Hz, 4H, ArH), 5.63 (s, 1H, AsOH), 3.68 (hept, J = 6.6 Hz, 4H, CH(CH₃)₂), 1.01–0.91 (m, 24H, CH­(CH ₃)₂) ppm. ¹ H NMR (500 MHz, C ₆ D ₆ ) δ 7.20–7.16 (m, 2H, ArH), 7.06 (d, J = 7.7 Hz, 4H, ArH), 3.77 (hept, J = 6.6 Hz, 4H, CH(CH₃)₂), 1.09 (m, 24H, CH­(CH ₃)₂) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, CDCl ₃ ) δ 151.93 (ArC), 143.40 (ArC), 129.43 (ArC), 124.30 (ArC), 31.71 (CH­(CH₃)₂), 24.85 ((CH­(CH₃)₂), 24.20 (CH­(CH₃)₂) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, DMSO–d ₆ ) δ 151.32­(ArC), 143.20 (ArC), 128.79 (ArC), 123.66 (ArC), 30.57 (CH­(CH₃)₂), 24.45 (CH­(CH₃)₂), 23.87 (CH­(CH₃)₂) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, C ₆ D ₆ ) δ 152.22 (ArC), 143.85 (ArC), 129.72 (ArC), 124.51 (ArC), 31.90 (CH­(CH₃)₂), 25.01 (CH­(CH₃)₂), 24.33 (CH­(CH₃)₂) ppm. Bulk purity was confirmed by PXRD (see Supporting Information).

Synthesis of Dipp₂AsO₂H

In open air, Dipp₂AsOH (88 mg, 0.212 mmol) was dissolved in THF (3 mL). An excess of aqueous hydrogen peroxide solution (0.30 mL, 30% in water) was added dropwise at room temperature, and the reaction mixture was stirred overnight at room temperature. After completion of the reaction, the solvent was removed under reduced pressure. The residue was redissolved in water (5 mL) and extracted with DCM (3 × 4 mL). The combined organic layers were dried over anhydrous Na₂SO₄, filtered, and concentrated under reduced pressure to remove all volatile solvents, affording a colorless oil. The crude residue was recrystallized from pentane at −24 °C to yield Dipp₂AsO₂H as a white solid (59 mg, 65% yield). IR (KBr, cm ^‑1 ) ν_CH (iPr) 2958–2864, ν_AsO 880, ν_As–OH 753. ¹ H NMR (500 MHz, CDCl ₃ ) δ 7.36 (t, J = 7.8 Hz, 2H, ArH), 7.17 (d, J = 7.8 Hz, 4H, ArH), 4.05 (m, 4H, CH(CH₃)₂), 1.00 (d, J = 6.7 Hz, 24H, CH­(CH ₃)₂) ppm. ¹ H NMR (500 MHz, DMSO- d ₆ ) δ 7.44 (d, J = 7.7 Hz, 2H, ArH), 7.29 (d, J = 7.8 Hz, 4H, ArH), 4.08 (hept, J = 6.5 Hz, 4H, CH(CH₃)₂), 0.98 (d, J = 6.7 Hz, 24H, CH­(CH ₃)₂) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, CDCl ₃ ) δ 152.08 (ArC), 136.23 (ArC), 131.65 (ArC), 125.13 (ArC), 30.91 (CH­(CH₃)₂), 24.46 (CH­(CH₃)₂) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, DMSO- d ₆ ) δ 150.85 (ArC), 136.47 (ArC), 131.56 (ArC), 125.04 (ArC), 29.81 (CH­(CH₃)₂), 23.94 (CH­(CH₃)₂) ppm. Bulk purity was confirmed by PXRD (see Supporting Information).

Synthesis of Dipp₂SbCl

2-Bromo-1,3-diisopropylbenzene (3.170 g, 13.14 mmol, 2.710 mL) was dissolved in Et₂O (approximately 20 mL). n-BuLi (13.14 mmol, 5.260 mL, 2.5 M in hexanes) was added via syringe at −78 °C and the mixture was left at −78 °C for 1 h before returning to room temperature. The reaction proceeded at room temperature for 5 h. SbCl₃ (1.500 g, 6.570 mmol) was dissolved in Et₂O (approximately 20 mL) and was added dropwise to the reaction mixture at −78 °C. The reaction mixture was then warmed to room temperature after 30 min and stirred overnight. The reaction mixture was opened to air, treated with water (20 mL), and extracted with Et₂O (3 × 25 mL). The organic layer was dried over anhydrous Na₂SO₄ and concentrated under reduced pressure to remove all volatile solvents. The residue was recrystallized from MeCN to give Dipp₂SbCl as a yellow solid (1.60 g, 51% yield). Crystals suitable for single-crystal X-ray diffraction were obtained at −24 °C from MeCN. Anal. Calcd for C₂₄H₃₄SbCl: C, 60.09; H, 7.14. Found: C, 55.60; H, 7.15. IR (KBr, cm ^‑1 ) ν_CH (iPr) 2965–2864. ¹ H NMR (500 MHz, CD ₂ Cl ₂ ) δ 7.33–7.28 (m, 2H, ArH), 7.19–7.13 (m, 4H, ArH), 3.43–3.27 (m, 4H, CH(CH₃)₂), 1.09–1.03 (m, 24H, CH­(CH ₃)₂) ppm. ¹ H NMR (500 MHz, CDCl ₃ ) δ 7.28 (t, J = 7.5 Hz, 2H, ArH), 7.13 (d, J = 7.6 Hz, 4H, ArH), 3.37 (m, 4H, CH(CH₃)₂), 1.06 (m, 24H, CH­(CH ₃)₂) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, CD ₂ Cl ₂ ) δ 154.43 (ArC), 154.36 (ArC), 145.64 (ArC), 143.64 (ArC), 130.41 (ArC), 130.38 (ArC), 125.21 (ArC), 125.14 (ArC), 36.64 (CH­(CH₃)₂), 36.09 (CH­(CH₃)₂), 25.15 (CH­(CH₃)₂), 25.06 (CH­(CH₃)₂), 24.27 (CH­(CH₃)₂) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, CDCl ₃ ) δ 154.05 (ArC), 145.27 (ArC), 143.26 (ArC), 130.11­(ArC), 124.89 (ArC), 36.38 (CH­(CH₃)₂), 35.84 (CH­(CH₃)₂), 25.12 (CH­(CH₃)₂), 24.29 (CH­(CH₃)₂) ppm. Bulk purity was confirmed by PXRD (see Supporting Information).

Synthesis of Dipp₂SbOH

In open air, Dipp₂SbCl (275 mg, 0.573 mmol) was dissolved in THF (3 mL). Then, 3 mL of a 1 M NaOH_(aq) solution (3.000 mmol, excess) was added. Over 5 min, the light yellow color was discharged from the solution. The reaction was stirred for an additional 25 min, then extracted with DCM (2 × 4 mL), dried over Na₂SO₄, filtered, and the solvent was evaporated under reduced pressure. The residue was recrystallized from a mixture of DCM and pentane to obtain Dipp₂SbOH as colorless crystals (210 mg, 79% yield). Crystals suitable for single-crystal X-ray diffraction were obtained at −24 °C from a mixture of DCM and pentane. Anal. Calcd for C₂₄H₃₅SbO: C, 62.49; H, 7.65. Found: C, 61.94; H, 8.28. IR (KBr, cm ^‑1 ) ν_OH 3550, ν_CH (iPr) 2964–2859, ν_Sb–OH 554. ¹ H NMR (500 MHz, CD ₂ Cl ₂ ) δ 7.27 (t, J = 7.7 Hz, 2H, ArH), 7.13 (d, J = 7.6 Hz, 4H, ArH), 3.40 (hept, J = 6.6 Hz, 4H, CH(CH₃)₂), 1.06 (m, 24H, CH­(CH ₃)₂), 1.05 (s, 1H, OH) ppm. ¹ H NMR (500 MHz, DMSO–d ₆ ) δ 7.24 (t, J = 7.6 Hz, 2H, ArH), 7.10 (d, J = 7.6 Hz, 4H, ArH), 4.22 (s, 1H, OH), 3.44 (hept, J = 6.5 Hz, 4H, CH(CH₃)₂), 0.98 (m, 24H, CH­(CH ₃)₂) ppm. ¹ H NMR (500 MHz, CD ₃ CN) δ 7.26 (t, J = 7.6 Hz, 2H, ArH), 7.14 (d, J = 7.6 Hz, 4H, ArH), 3.46 (hept, J = 6.7 Hz, 4H, CH(CH₃)₂), 2.23 (s, 1H, OH), 1.06–1.01 (m, 24H, CH­(CH ₃)₂) ppm. ¹ H NMR (500 MHz, CDCl ₃ ) δ 7.31–7.24 (m, 2H, ArH), 7.13 (d, J = 7.6 Hz, 4H, ArH), 3.38 (hept, J = 6.7 Hz, 4H, CH(CH₃)₂), 1.07 (m, 24H, CH­(CH ₃)₂), 0.93 (s, 1H, OH) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, CD ₂ Cl ₂ ) δ 154.62 (ArC), 146.77 (ArC), 129.87 (ArC), 124.65 (ArC), 34.82 (CH­(CH₃)₂), 25.06 (CH­(CH₃)₂), 24.43 (CH­(CH₃)₂) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, DMSO–d ₆ ) δ 153.67 (ArC), 145.60 (ArC), 129.10 (ArC), 123.76 (ArC), 33.52 (CH­(CH₃)₂), 24.65 (CH­(CH₃)₂), 24.06 (CH­(CH₃)₂) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, CD ₃ CN) δ 155.30 (ArC), 147.13 (ArC), 130.37 (ArC), 125.08 (ArC), 35.01 (CH­(CH₃)₂), 25.15 (CH­(CH₃)₂), 24.52 (CH­(CH₃)₂) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, CDCl ₃ ) δ 154.23 (ArC), 146.49 (ArC), 129.69 (ArC), 124.45 (ArC), 34.66 (CH­(CH₃)₂), 25.04 (CH­(CH₃)₂), 24.43 (CH­(CH₃)₂) ppm. Bulk purity was confirmed by PXRD (see Supporting Information).

In Situ Deuteration of Dipp₂AsOH and Dipp₂SbOH

The pnictinous acid (20 mg) was transferred to an NMR tube under ambient atmosphere and dissolved in the corresponding deuterated solvent (CD₂Cl₂, DMSO-d ₆, CD₃CN, or CDCl₃). An initial ¹H NMR spectrum was recorded to establish the reference spectrum of the protonated species. A drop of D₂O was added directly to the NMR tube, and the mixture was gently shaken. ¹H NMR spectra were then acquired at defined time intervals to monitor H/D exchange.

NMR-Scale Reaction of Dipp₂SbOH and PPh₃

Dipp₂SbOH (17 mg, 0.037 mmol) and Ph₃P (10 mg, 0.037 mmol) were dissolved in CDCl₃. The resulting solution was allowed to stand at room temperature for 30 min, after which ¹H and ³¹P NMR spectra were recorded. ¹ H NMR (500 MHz, CD ₂ Cl ₂ ) δ 7.36–7.31 (m, 15H, PPh ₃), 7.31–7.25 (m, 2H, ArH), 7.14 (d, J = 7.7 Hz, 4H, ArH), 3.41 (hept, J = 6.7 Hz, 4H, CH(CH₃)₂), 1.07 (m, 24H, CH­(CH ₃)₂), 1.03 (s, 1H, OH) ppm. ³¹ P­{ ¹ H} NMR (202 MHz, CD ₂ Cl ₂ ) δ −5.51 ppm.

NMR-Scale Reaction of Dipp₂SbOH and NEt₃

Dipp₂SbOH (20 mg, 0.043 mmol) and Et₃N (4.4 mg, 5 μL, 0.043 mmol) were dissolved in CDCl₃. The resulting solution was allowed to stand at room temperature for 30 min, after which a ¹H NMR spectrum was acquired. ¹ H NMR (500 MHz, CD ₂ Cl ₂ ) δ 7.27 (t, J = 7.7 Hz, 2H, ArH), 7.13 (d, J = 7.7 Hz, 4H, ArH), 3.46–3.36 (m, 4H, CH(CH₃)₂), 2.48 (q, J = 7.1 Hz, 6H, NCH ₂), 1.33 (s, 1H, OH), 1.06 (m, 24H, CH­(CH ₃)₂), 1.00 (t, J = 7.2 Hz, 9H, NCH₂CH ₃) ppm.

NMR-Scale Reaction of Dipp₂SbOH and DBU

Dipp₂SbOH (20 mg, 0.043 mmol) and DBU (6.6 mg, 6.5 μL, 0.043 mmol) were dissolved in CDCl₃. The resulting solution was allowed to stand at room temperature for 30 min, after which a ¹H NMR spectrum was recorded. The reaction mixture was allowed to stand for 24 h, after which the solvent was slowly removed under reduced pressure. The residue was redissolved in a minimal amount of CH₂Cl₂ and allowed to stand undisturbed. Single crystals suitable for X-ray diffraction were obtained directly from this mixture. ¹ H NMR (500 MHz, CD ₂ Cl ₂ ) δ 7.25 (t, J = 7.7 Hz, 2H, ArH), 7.11 (d, J = 7.8 Hz, 4H, ArH), 5.33 (s, 1H, NH), 3.52–3.40 (m, 4H, CH(CH₃)₂), 3.18–3.16 (m, 6H, HNCH ₂ and NCH ₂), 2.34–2.32 (m, 2H, N=CCH ₂), 1.74 (m, 2H, HNCH₂CH ₂), 1.63–1.61 (m, 4H, N=CCH₂CH ₂CH ₂), 1.55 (m, 2H, NCH₂CH ₂), 1.05 (m, 24H, CH­(CH ₃)₂) ppm.

NMR-Scale Reaction between Dipp₂SbOH and B­(C₆F₅)₃

Dipp₂SbOH (20 mg, 0.043 mmol) and B­(C₆F₅)₃ (29 mg, 0.056 mmol) were dissolved in CDCl₃. The reaction mixture was allowed to stand at room temperature for 30 min, after which NMR spectra were recorded. The ¹¹B NMR spectrum displayed a resonance at δ = 1.16 ppm, consistent with the formation of a tetracoordinate boron center, supporting Lewis adduct Dipp₂SbOH·B­(C₆F₅)₃ in solution. Trace amounts of water result in additional minor signals corresponding to decomposition products. ¹ H NMR (500 MHz, CDCl ₃ ) δ 7.41 (t, J = 7.6 Hz, 2H, ArH), 7.23 (d, J = 7.6 Hz, 4H, ArH), 5.55 (s, 1H, OH), 2.93–2.76 (m, 4H, CH(CH₃)₂), 0.98 (m, 24H, CH­(CH ₃)₂) ppm. ¹¹ B­{ ¹ H} NMR (160 MHz, CDCl ₃ ) δ 1.16 ppm. ¹⁹ F­{ ¹ H} NMR (470 MHz, CDCl ₃ ) δ −132.27 (br s, 6F, o-F), −156.06 (br s, 3F, p-F), −162.93 (br s, 6F, m-F) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, CDCl ₃ ) δ 154.31 (ArC), 153.57 (ArC), 131.75 (ArC), 126.02 (ArC), 36.13 (CH­(CH₃)₂), 24.58 (CH­(CH ₃)₂), 24.14 (CH­(CH ₃)₂) ppm.

NMR-Scale Reaction between Dipp₂SbOH and BPh₃

Dipp₂SbOH (19 mg, 0.041 mmol) and BPh₃ (10 mg, 0.041 mmol) were dissolved in CDCl₃. The reaction mixture was allowed to stand at room temperature for 30 min, after which NMR spectra were recorded. The signals in the NMR spectra are not shifted from those of the individual species.

NMR-Scale Reaction between Dipp₂SbOH and PdCl₂

Dipp₂SbOH (15 mg, 0.032 mmol) and PdCl₂ (58 mg, 0.032 mmol) were dissolved in CD₃CN. The resulting solution was allowed to stand at room temperature for 2 d, after which NMR spectra were recorded. Single crystals suitable for X-ray diffraction were grown from this solution. ¹ H NMR (500 MHz, CD ₃ CN) δ 7.47 (t, J = 7.7 Hz, 2H, ArH), 7.28 (d, J = 7.7 Hz, 4H, ArH), 3.83 (s, 1H, OH), 3.53 (hept, J = 7.3 Hz, 4H, CH(CH₃)₂), 1.19 (d, J = 6.6 Hz, 12H, CH­(CH ₃)₂), 1.07 (d, J = 6.6 Hz, 12H, CH­(CH ₃)₂) ppm. ¹³ C­{ ¹ H} NMR (126 MHz, CD ₃ CN) δ 154.73 (ArC), 133.06 (ArC), 126.51 (ArC), 124.44 (ArC), 37.53 (CH­(CH₃)₂), 25.30 (CH­(CH₃)₂), 24.69 (CH­(CH₃)₂) ppm.

Oxidation of Dipp₂SbOH with PhIO

Dipp₂SbOH (80 mg, 0.173 mmol) was dissolved in DCM (3 mL) and PhIO (38 mg, 0.173 mmol) was added. The reaction mixture was stirred at room temperature for 5 h and then passed through a Celite pad to remove excess PhIO. The filtrate was stripped of solvent and the residue was taken up in DMSO-d ₆ for NMR analysis. Alternatively, the residue was taken up in DCM from which crystals suitable for X-ray diffraction were obtained. ¹ H NMR (500 MHz, DMSO- d ₆ ) δ 7.26 (t, J = 7.6 Hz, 2H, ArH), 7.13 (d, J = 7.8 Hz, 4H, ArH), 3.68 (m, 4H, CH(CH₃)₂), 3.49 (s, 1H, OH), 0.86 (s, 24H, CH­(CH ₃)₂) ppm.

Oxidation of Dipp₂SbCl with PhIO

Dipp₂SbCl (100 mg, 0.20 mmol) dissolved in DCM (7 mL), after which iodosobenzene (46 mg, 0.21 mmol) was added as a solid at room temperature. The solid was rapidly consumed to produce a faintly yellow hazy solution within 5 min. The reaction mixture was stirred overnight, then passed through a Celite pad to remove excess iodosobenzene. The solvent was removed from the filtrate and the residue was taken up in CDCl₃ for NMR analysis. Single crystals suitable for X-ray diffraction were grown when the solution was allowed to stand at room temperature.

Oxidation of Dipp₂SbOH with H₂O₂

Dipp₂SbOH (450 mg, 0.976 mmol) was dissolved in THF (3 mL). An excess of aqueous hydrogen peroxide solution (1.100 mL, 30% in water) was added dropwise at room temperature, and the reaction mixture was stirred overnight at room temperature. After completion of the reaction, the solvent was removed under reduced pressure. The residue was redissolved in water (5 mL) and extracted with DCM (3 × 4 mL). The combined organic layers were dried over anhydrous Na₂SO₄, filtered, and concentrated under reduced pressure to remove all volatile solvents, affording a colorless oil. The crude residue was taken up in DCM and cooled to −24 °C to precipitate a white solid. NMR spectra of this material indicate that it is a complex mixture of products, but small numbers of crystals of the compound shown in Figure and Scheme could be reproducibly grown from solutions of the product.

X-ray Crystallography

Crystals were grown as described above, selected under a microscope, loaded onto a MiTeGen polyimide sample loop using Type NVH Cargille immersion oil, and mounted onto a Rigaku XtaLAB Synergy-S single-crystal X-ray diffractometer. Some crystals were cooled under a stream of nitrogen and some data were collected at room temperature. Diffraction of Cu Kα radiation from a PhotonJet-S microfocus source was detected by using a HyPix6000HE hybrid photon counting detector. Screening, indexing, data collection, and data processing were performed with CrysAlisPro. The structures were solved using SHELXT and refined using SHELXL following established strategies. − All non-H atoms were refined anisotropically. C-bound H atoms were placed at calculated positions and refined with a riding model and coupled isotropic displacement parameters (1.2 × U _eq of attached C atom for nonmethyl C–H atoms and 1.5 × U _eq of attached C atom for methyl groups). Polar (OH and NH) H atoms were refined semifreely using distance restraints and coupled isotropic displacement parameters (1.5 × U _eq of attached atom). Structures collected at room temperature are displayed with thermal ellipsoids at the 30% probability level and structures collected at low temperature are displayed with ellipsoids at the 50% probability level.

Powder X-ray Diffraction

Bulk samples of Dipp₂AsCl, Dipp₂AsOH, Dipp₂AsO₂H, Dipp₂SbCl, and Dipp₂SbOH were ground using an agate mortar and pestle. The fine powders were each loaded onto a MiTeGen polyimide sample loop using Type NVH Cargille immersion oil and mounted onto a Rigaku XtaLAB Synergy-S single-crystal X-ray diffractometer. The diffraction of Cu Kα radiation was collected while the sample underwent a Gandolfi scan. Data collection and processing were performed using CrysAlis^Pro. Simulated powder diffractograms were generated from experimental crystal structures using Mercury. − These simulated diffractograms were compared with the experimentally measured powder diffractograms.

Computational Chemistry

All geometry optimizations and frequency calculations were performed in the gas phase using ORCA 6.1.0. − Geometry optimizations and frequency calculations were performed using crystallographic input coordinates at the PBE0/def2-TZVPP level of theory. −

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.inorgchem.6c02995.

• NMR spectra, IR spectra, and X-ray refinement details (PDF)

• Computationally optimized Cartesian coordinates (XYZ)

The authors declare no competing financial interest.