Synthesis of Unique Phosphazane Macrocycles via Steric Activation of

Article pubs.acs.org/IC

Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX

Synthesis of Unique Phosphazane Macrocycles via Steric Activation of C−N Bonds Yan X. Shi,† Katherine A. Martin,† Rong Z. Liang,† Daniel G. Star,† Yongxin Li,† Rakesh Ganguly,† Ying Sim,† Davin Tan,† Jesús Díaz,*,‡ and Felipe García*,† †

Downloaded via KAOHSIUNG MEDICAL UNIV on August 22, 2018 at 07:30:36 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

School of Physical and Mathematical Sciences, Division of Chemistry and Biological Chemistry, Nanyang Technological University, 21 Nanyang Link, 637371 Singapore ‡ Departamento de Química Orgánica e Inorgánica, Facultad de Veterinaria, Universidad de Extremadura, Av. de la Universidad s/n, 10003 Caceres, Spain S Supporting Information *

ABSTRACT: Herein we describe that oxidation reactions of the dimeric cyclophosphazanes, [{P(μ-NR)}2(μ-NR)]2, R = tBu (1), to produce a series of diagonally dioxidized products P4(μ-NtBu)6E2 [E = O (2), S (3), and Se (4)] and tetraoxidized frameworks. The latter display an unexpected C−N bond activation and cleavage to produce a series of novel phosphazane macrocyclic arrangements containing newly formed N−H bonds. Macromolecules P4(μ-NtBu)4(μ-NH)2O4 (5) and P4(μ-NtBu)3(μ-NH)3E4, E = S (6) and Se (7), dicleaved and tricleaved products, respectively, are rare examples of dimeric macrocycles containing NH bridging groups. Our theoretical and experimental studies illustrate that the extent to which these C−N bonds are cleaved can be controlled by modification of steric parameters in their synthesis, by adjusting either the steric bulk of the substituents in the parent framework or the size of the chalcogen element introduced during the oxidation process. Our findings represent new synthetic pathways for the synthesis of otherwise-elusive macrocycle arrangements within the phosphazane family.

■

INTRODUCTION The significant kinetic lability of bonds between main group elements other than carbon remains a major challenge in the design of robust inorganic molecular frameworks in modern main group chemistry. Except for specific families of compounds based on strong covalent bonds (silicate minerals, phosphates, phosphazene polymers, etc.), studies on the technological and industrial application of inorganic frameworks still lag far behind those of their organic counterparts. In addition, the tedious synthetic routes involved further handicap their production and, hence, their applications. During the last few decades, appreciable efforts have been dedicated to the synthesis of inorganic macrocyclic systems as prospective neutral or anionic ligands for coordination and host−guest chemistries (Figure 1).1−10 Among these, P−N phosphazane frameworks are of particular interest due to the robustness stemming from the thermodynamic stability of the P−N single bonds (70 kcal·mol−1; cf. C−C single bond energy 80 kcal·mol−1).11,12 Numerous synthetic methodologies have been developed for the rational design of macrocyclic phosph(III/III)azane and the fine-tuning of their steric and electronic properties.13−17 These systems are commonly generated by reacting the dimeric phosphazane [ClP(μ-NR)]2 with a broad range of inorganic or organic linkers, producing inorganic or hybrid © XXXX American Chemical Society

Figure 1. Examples of inorganic macrocycles.5,6,9

macrocyclic frameworks of formula [{P(μ-NR)}2(μ-X)]n or [{P(μ-NR)}2(μ-LL′)]n, (X = O, Se, and NR′ where R′ = H, i Pr, and tBu; LL′ = bifunctional organic linker; R = iPr, tBu, and Ar).14,18−23 In the case of inorganic macrocycles,6,19,21,24 it is wellestablished that their size is directly influenced by the nature and steric bulk of the bridging atoms or groups present within the macrocyclic backbone. For instance, tetrameric or pentameric P(III) macrocycles are formed when small NH Received: June 12, 2018

A

DOI: 10.1021/acs.inorgchem.8b01596 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 2. Tetrameric, pentameric, and hexameric cyclophosphazane-based macrocycles.25,29,30

the release of any increased steric strain imposed on the system. Following the successful purification of 1, we set out to explore the impact of increasing steric crowding by oxidation of its phosphorus centers with chalcogen atoms (Scheme 1),

groups or oxygen atoms act as bridging atoms, respectively;19,25,26 whereas dimeric macrocycles are favored when the steric bulk of the bridging groups is increased from NH to NiPr or N tBu groups.9,19,25−27 Selective formation of pentameric over tetrameric macrocyclic species can be achieved by halide templating during the condensation reactions leading to the formation of the desired macrocyclic arrangements (Figure 2a,b).24,25,28,29 Large mixed-valence macrocyclic arrangements (Figure 2c) can also be formed if the dimeric phosphazane [ClP(μ-NR)]2 reacts with its chalcogen-oxidized anionic counterparts [EP(S)(μ-NR)]22− (E = S or Se).30 Despite the increasing attention, their intrinsic bond lability has impeded phosphazane frameworks from becoming commonplace in technological applications. Previously, we have demonstrated the enhancement of air and hydrolytic stability of cyclophosph(III/III)azane frameworks upon oxidation of the phosphorus centers.31,32 This finding was also corroborated by the Stahl and Wright groups who reported similar conclusions in phosphazane metal complexes and macrocyclic arrangements, respectively.30,33 Subsequent to our initial findings on acyclic phosph(III/ III)azanes, we investigated the stability of macrocyclic phosphazane frameworks of the type [{P(μ-NtBu)}2(μNtBu)]2 (1) upon oxidation with selenium, which resulted in an unexpected C−N bond activation and cleavage process.34 Herein, we demonstrate that this process can be exploited rationally, by judiciously fine-tuning steric factors to produce a series of macrocyclic compounds with varying degrees of bond cleavage. The species obtained herein have proven elusive by standard synthetic routes. In addition, our experimental and computational mechanistic studies clearly demonstrate that steric factors play a crucial role in the degree of C−N bond activation achieved in these frameworks and exemplify how this can be used as a synthetic strategy to achieve unique phosphazane macrocycles.35

Scheme 1. Synthesis of Macrocycles P4(μ-NtBu)6E2 (E = O (2), S (3), and Se (4)

beginning with the smallest in the group, oxygen (1.54 Å van der Waals radius).36 Treatment of 1 with an excess of tBuOOH at room temperature for 8 h enabled the formation of the diagonally dioxidized product P4(μ-NtBu)6O2 (2), as indicated by various spectroscopic analyses. The proton decoupled 31 1 P{ H} NMR spectra of 2 revealed two sets of resonances at 79.9 (2JP−P = 32.4 Hz) and 12.6 (2JP−P = 32.4 Hz) ppm, corresponding to P(III) and P(V) centers, respectively, and its IR spectra featured a characteristic PO bond stretching frequency at 1267 cm−1. Two resonances, a doublet and a singlet at 1.77 (4JH−P = 4.4 Hz) and 1.55 ppm were observed by room-temperature 1H NMR that can be attributed to the bridging and P2N2 tert-butyl groups, respectively. Following isolation of the diagonally dioxidized macrocycle 2, we aimed to oxidize 1 with heavier, larger chalcogen atoms, specifically S, Se, and Te (1.85, 1.91, and 2.09 Å van der Waals radii, respectively).36 Reaction of 1 with excess (1:2.2) elemental S8 or Se8 powders under similar experimental conditions afforded a mixture of di- and tetraoxidized (as minor product) species. Purification by recrystallization provided the sulfur and selenium dioxidized compounds 3 and 4. Unfortunately, the formation of the Te-oxidized product was not observed when using elemental Te as the oxidant, even after prolonged reaction duration (over a month under refluxing conditions). The 31P{1H} NMR spectra for 3 and 4 exhibited similar sets of resonances to that of 2, corresponding to P(III) and P(V) centers at 96.2 (2JP−P = 37.4 Hz) and 57.4 (2JP−P = 34.0 Hz) ppm, and 101.8 (2JP−P = 34.0 Hz) and 45.0 (2JP−P = 35.6 Hz) ppm, respectively, consistent with the formation of diagonally dioxidized structures. In addition, the room-temperature 1H NMR spectroscopy obtained for 3 and 4

■

RESULTS AND DISCUSSION Synthesis of the Dioxidized Macrocycles 2−4. Our studies began with the synthesis of the phosphazane macrocycle [{P(μ-NtBu)}2(μ-NtBu)]2 (1) via the reaction between [ClP(μ-NtBu)]2 and [LiNtBuP(μ-NtBu)]2, as previously reported by Chivers et al.27 The product was initially obtained as cocrystals of 1·[LiNtBuP(μ-NtBu)]2 (1:1 ratio) from which we were then able to isolate 1 by recrystallization from THF. It is important to highlight that 1 possesses a highly rigid backbone, which, under standard reaction conditions, does not allow for substantial topological changes (i.e., only rotation around C−N single bonds is permitted) that might allow for B


Article

Inorganic Chemistry further corroborates the data observed for 31P{1H} NMR. Resonances attributed to the tert-butyl group protons at the bridging and P2N2 ring nitrogen substituents, respectively, are observed as a doublet and a singlet at 1.91 ppm (4JH−P = 5.2 Hz) and 1.66 ppm for 3 and 1.96 ppm (4JH−P = 5.6 Hz) and 1.71 ppm for 4. Anomalous 77Se Chemical Shift Observed for 4. Interestingly, the room-temperature 77Se NMR spectrum of 4 (ca. ∼98 ppm) bears greater resemblance to that of the singly bonded Me2P−Se−Me, rather than to any of the previously reported compounds containing PSe bonds, expected to be found in the negative region of spectrum (cf. −177.9 ppm in cis-[Me2N(Se)P(μ-NtBu)]2).37,38 This unexpected and pronounced downfield chemical shift can be rationalized by the close proximity between the Se atom and the N-bridging tert-butyl group, whereby the repulsive van der Waals effect of the latter over the relatively polarizable selenium atom gave rise to a resonance signal at a much higher chemical shift than expected (see SI, Figure S1.15).39 This effect is consistent and readily observed using 1H NMR spectroscopy, where a more downfield shift was observed for the tert-butyl protons on bridging nitrogen atoms, as compared to those within the P2N2 rings (see Figure 3). Such observations are indicative of even greater steric crowding present in the dioxidized species than in 1 (vide supra).40

produced a fully oxidized macrocycle in quantitative yield. This was further supported by in situ 31P{1H} NMR spectrum, which exhibited only one single resonance at −7.17 ppm. However, in contrast to the predicted 1H NMR spectrum of the expected product, [{P(O)(μ-NtBu)}2(μ-NtBu)]2, for which we would expect two single resonances corresponding to two distinct tert-butyl groups (cf. 1.54 and 1.50 ppm for 1), the presence of a sole singlet at 1.64 ppm was observed, indicating unexpected symmetrically and thus magnetically equivalent tert-butyl groups. Moreover, a broad resonance signal at 4.94 ppm with an intensity ratio of 36:2 (36 protons from the four tert-butyl groups), combined with the absorption band at 3348 cm−1 in the FTIR spectrum, suggests the presence of two newly formed N−H bonds. These spectroscopic data suggest the formation of a fully oxidized product, P4(μ-NtBu)4(μ-NH)2O4 (5), where, unexpectedly, two tert-butyl groups belonging to the ring-bridging nitrogen atoms of 1 had been cleaved (see Scheme 2). Scheme 2. Synthesis of P4(μ-NtBu)4(μ-NH)2O4 (5), P4(μNtBu)3(μ-NH)3S4 (6), and P4(μ-NtBu)3(μ-NH)3Se4 (7)

Subsequently, we attempted to obtain fully oxidized counterparts for the heavier chalcogen elements to assess the steric effect of a larger chalcogen on the observed cleavage process.34 A mixture of 1 with elemental S8 or Se8 (in a molar ratio of 1:4.4) was refluxed in toluene, and the reaction progress was monitored by in situ 31P{1H} NMR spectroscopy. In this case, the predicted fully oxidized and symmetrical products, P4(μ-NtBu)4(μ-NH)2E4 (E = S and Se), would be expected to display a single resonance within the range of ∼0− 50 ppm (cf. 146.0 ppm in isolated 1), similar to previously reported phosph(V/V)azane derivatives.31−33 To our surprise, the 31P{1H} NMR spectra of 6 and 7, S and Se, respectively, comprised of two sets of signals centered at 34.7 (dd, 2JP−P = 38.9 Hz, 2JH−P = 6.5 Hz) and 30.8 (dd, 2JP−P = 38.9 Hz, 2JH−P = 6.5 Hz) ppm and 24.7 (dd, 2JP−P = 59.9 Hz, 2 JH−P = 9.7 Hz) and 22.1 (dd, 2JP−P = 59.9 Hz, 2JH−P = 9.7 Hz) ppm, respectively, which is indicative of the formation of fully oxidized lower symmetry products (as compared to 1 or 4). The 1H NMR spectra of 6 and 7 revealed singlet resonances at 1.59, 1.56, and 1.23 ppm (intensity ratio 1:1:1) and 1.62 (broad) and 1.31 ppm (intensity ratio 2:1), respectively, which can be attributed to tert-butyl groups belonging to the nitrogen atoms of the P2N2 rings. The presence of two broad singlets with intensity ratio 2:1 at 5.16 and 4.13 ppm (6) and at 5.64 and 4.28 ppm (7), along with observed N−H IR absorptions bands at 3345 and 3398 cm−1 for 6 and 7, respectively, suggest the presence of newly formed N−H bonds. Compounds 6 and 7 were isolated as tetrasulfide- and tetraselenide-oxidized products, in which three tert-butyl groups, two on the ringbridging nitrogen atoms and one on the P2N2 ring, have been cleaved, as shown in Scheme 2. All attempts made to obtain the fully oxidized tellurium counterpart were unsuccessful,

Figure 3. 1H NMR spectra of 1−4 at room temperature.

Synthesis of the Tetraoxidized Macrocycles 5−7. Following the observation that a mixture of partially oxidized products was obtained on reacting 1 with a slight excess of the heavier chalcogens (i.e., elemental S8 and Se8), attempts to afford the fully oxidized species (i.e., tetraoxidation) were made so as to further increase the steric crowding within the phosphazane ring. Since we already demonstrated that the reaction of 1 with an excess of tBuOOH only affords the diagonally oxidized 2, we attempted full oxidation using 2 as starting material. Unfortunately, reacting 2 with 4.4 equiv of tBuOOH gave no indication of the reaction proceeding by in situ 31P{1H} NMR monitoring, even after prolonged reflux in toluene. Higher boiling point solvents, such as dimethylformamide (DMF) were also tested but led to no reaction or caused the decomposition of 1. Hence, we decided to explore the use of a stronger oxidizing agents, namely, 3-chloroperbenzoic acid (m-CPBA). Reaction of 1 with m-CPBA at room temperature for 3 h successfully C


Article

Inorganic Chemistry

work has shown that [Cl(O)P(μ-NtBu)]2 and [Cl(S)P(μNtBu)]2 exhibit low reactivity toward substitution reactions.31 In our work, no reaction was observed between [Cl(Se)P(μNtBu)]2 and [NH2(Se)P(μ-NtBu)]2 as monitored by in situ 31 1 P{ H} NMR spectroscopy, hence demonstrating that such a synthetic strategy cannot be used to create the desired macrocyclic compound 7 (Scheme 3, method B). Solid State Structures of Compounds 2−7. Diffraction quality crystals of 2−7 were obtained by cooling saturated solutions of the respective compounds in toluene at 7 °C overnight (Figures 4 and 5). X-ray diffraction analyses of the solid-state structures are consistent with spectroscopic data, with the molecular structures of compounds 2, 3, and 4 displaying a diagonally dioxidized macrocyclic framework comprised of alternating P(III) and P(V) atoms. The PE bond distances for the dioxidized products, 1.456 Å (average), 1.936 Å, and 2.098 Å for 2, 3, and 4, respectively, are comparable to previously reported analogues (cf. PO = 1.459 Å in cis-[(4-CN-PhO)(O)P(μ-NtBu)]2, PS = 1.927 Å in bis(sulfide) [Cy(H)N(S)P(μ-NtBu)]2, and PSe = 2.056 Å in cis-[(4-CN-PhO)(Se)P(μ-NtBu)]2),31 and the average ring is shorter than those observed in 1.27 The P2N2 rings in 2 are almost planar with puckering angle of 3.7° (cf. 1, 14.5°)34 and are nearly perpendicular to the plane of the macrocycle (dihedral angle of 89.1°); similar trends were also observed in 3 and 4. Closer inspection of the molecular structure as determined by X-ray diffraction crystallography revealed the existence of a close intramolecular hydrogen bonding interaction, C−H···E (E = O, S, Se), between chalcogen elements and the hydrogen atoms on the exogenous tert-butyl groups (P4(μ-NtBu)6O2 (2), P4(μ-NtBu)6S2 (3), and P4(μ-NtBu)6Se2 (4)). The measured C−H···E distances for 2, 3, and 4 are 3.0 Å, 3.4 Å, and 3.3 Å, respectively,47−50 indicating close contacts between the C−H and E within the molecule (sum of the vdW radius are 2.72, 3.0, and 3.10 Å, respectively). The observed short interatomic distances within the molecule further suggest that the repulsive van der Waals effect might be responsible for producing the anomalous chemical shift (ca. ∼98 ppm) observed in the 77Se NMR spectrum of 4 (vide supra). In addition, a C−N bond elongation for the bridging tert-butyl groups is also observed as the steric bulk of the chalcogen element increases, 1.508, 1.532, 1.549, and 1.551 Å from 1 to 4, respectively. For compounds 5, 6, and 7, X-ray diffraction data analysis confirms the sterically induced C−N bond cleavage (Figure 5). Compound 5, P4(μ-NtBu)4(μ-NH)2O4, displays a macrocyclic framework in which all phosphorus atoms have been oxidized, and both tert-butyl groups on the ring-bridging nitrogen atoms have been cleaved and replaced by N−H moiety (Figure 5a). In contrast, compounds 6 and 7 display three cleaved N−tBu bonds, two ring-bridging and one within the P2N2 unit (Figure 5b, c, respectively), with the greater extent of tBu cleavage in the latter likely attributed to the higher steric demand of sulfur and selenium. The PE (E = O, S, and Se) bond distances of 1.469, 1.919, and 2.093 Å for 5, 6, and 7, respectively, are comparable to previously reported analogues (cf. 1.47 Å in [tBu(H)N(O)P(μ-NtBu)]2, 1.927 Å in [Cy(H)N(S)P(μN t Bu)] 2 , and 2.091 Å in cis-[(μ-N t Bu) 2 (P(Se)NC4H8NMe)2]).51−53 It was also discovered that the average ring and bridging P−N bond distances (1.682 and 1.664 Å in 5, 1.686 and 1.687 Å in 6, and 1.700 and 1.651 Å in 7) are shorter than those observed in 1 (cf. 1, 1.781 and 1.742 Å, respectively),34 which, together with the increased steric

presumably due to a prohibitive level steric crowding between larger chalcogen elements and the three tert-butyl groups, two within the P2N2 rings and one on the bridging nitrogen atoms, surrounding the phosphorus centers in 1.41−43 In contrast to the anomalous chemical shift observed for 4, vide supra, where bulky bridging tert-butyl groups are present, the roomtemperature 77Se NMR spectrum of 7 displays a multiplet centered at −104.6 ppm, which is consistent and comparable to other reported analogues (cf. −180.4 ppm in cis-[C5H10N(Se)P(μ-NtBu)]2). The 77Se NMR signals, after the bridging tert-butyl groups have been cleaved, are within previously reported ranges for phosphazane selenide species, further supporting our initial hypothesis that the repulsive van der Waals effect might have significant influence on the anomalous chemical shift observed for compound 4. Can Compounds 5−7 Be Obtained via Standard Synthetic Routes? To assess the uniqueness of our synthetic strategy to afford chalcogen oxidized cyclophosphazane macrocyclic structures by sterically induced C−N cleavage, we decided to compare and evaluate the accessibility of compounds 5−7 using established synthetic methodologies. The standard synthetic method to produce such species typically involves the cyclization reaction of suitable building blocks and subsequent oxidation. Simple retrosynthetic analysis suggests that coupling reactions between [ClP(μNtBu)]2 with [H2NP(μ-NtBu)]2 and [(μ-NtBu)(μ-NH)(PNH2)2] with [ClP(μ-NtBu)]2 or [(μ-NtBu)(μ-NH)(PCl)2] with [H2NP(μ-NtBu)]2, followed by oxidation using appropriate oxidants would afford compounds 5, 6, and 7, respectively. However, it has been demonstrated that only reactions involving sterically demanding groups, such as tBu groups, produce dimeric macrocyclic arrangements. In cases where NH act as bridging groups between P2N2 rings, only tetrameric macrocycle arrangements [{P(μ-NtBu)}2(μ-NH)]4 have been previously obtained.26 Hence, in all the above cases, larger macrocycles are anticipated as the sole product obtained (Scheme 3, method A). While species containing NH bridges Scheme 3. Attempted Synthesis of Compound 7

within the P2N2 ring have been previously described, they are rare and require the presence of tetracoordinate phosphorus atoms.44−46 Unfortunately, attempts to synthesize species of the type [(μ-NtBu)(μ-NH)(PNH2)2] or [(μ-NtBu)(μ-NH)(PCl)2] were unsuccessful. An alternative synthetic route involves oxidation of these building blocks prior to the cyclization. However, previous D


Article

Inorganic Chemistry

Figure 4. Solid-state structures of 2 (a), 3 (b), and 4 (c). The tert-butyl units are drawn as wire frames in all the graphical representations. H atoms are omitted for clarity. Thermal ellipsoids are set at the 50% probability level. For selected bond lengths [Å] and angles [deg], see SI.

Figure 5. Solid-state structures of 5 (a), 6 (b), and 7 (c). The tert-butyl units are drawn as wire frames in all the graphical representations. H atoms are omitted for clarity. Thermal ellipsoids are set at the 50% probability level. For selected bond lengths [Å] and angles [deg], see SI.

our studies, presumably due to a faster rate of oxidation. In both cases, their corresponding 1H NMR spectra show signals consistent with the formation of isobutylene as the only detectable byproduct prior to the formation of any cleaved product (i.e., types V and VI, dicleavage intermediate, and 6 and 7, respectively). This is illustrated by the appearance of two new signals with intensity ratio 1:3 at 4.76 and 1.64 ppm (see Figure S1.44).54 Unfortunately, despite several attempts, the monocleaved species (type IV) was not observed throughout our spectroscopic studies. In contrast to previously reported C−N bond cleavage reactions, the observed process does not involve the use of either transition metal catalysts or strong acids.55 However, the overall process resembles a McLafferty rearrangement, typically observed in mass spectroscopy.56 Expansion to Less Sterically Hindered Species: Can the Degree of C−N Bond Activation Be Rationally Controlled? Our studies on the tert-butyl substituted system point toward steric strain release of the macrocyclic system as the driving force for the observed C−N bond cleavage.34 One can propose that other unusual macrocycles with a lower degree of bond cleavage (i.e., type IV, monocleaved) may be accessible when a less sterically encumbered macrocycle is present. In order to validate our hypothesis, the less sterically hindered dimeric phosphazane [{P(μ-NiPr)}2(μ-NiPr)]2 (8) was synthesized.28 It is important to note that frameworks of the type [{P(μ-NR)}2(μ-NR)]2 are thermodynamically less stable than their respective adamantoid isodesmic forms [P4(NR)6].27,40 Compound 8 is more susceptible to isomerization than 1; therefore milder reaction conditions were explored. For this reason, Se was chosen as the oxidizing agent for use in these experiments. Compound 8 was reacted with

demand of the chalcogen atom, provides a plausible explanation for the increase in number of C−N bonds cleaved when descending down the periodic group. Mechanistic Insights: 1H and 31P NMR Studies. To enable the exploitation of the observed C−N bond activation process for the synthesis of a wider range of unique phosphazane macrocycles, we set out to elucidate the reaction mechanism and the underlying factors that influence it. For this purpose, we monitored the reactions of 1 (in toluene-d8) with an excess of elemental sulfur or selenium in a sealed Young’s tap NMR tube43 using 1H and 31P{1H} NMR spectroscopies. Our 31P{1H} NMR studies suggest that the first two steps during the oxidation reaction of 1 with sulfur and selenium are identical (Figure S1.44 and S1.45). We observed the initial formation of mono-oxidized species (type I, in Scheme 4) as indicated by the appearance of signals at ∼39, 108, 147, and 154 ppm for sulfur and ∼54, 105, 143, and 152 ppm for selenium, corresponding to one P(V) and three P(III) centers, respectively. As the reaction proceeds, the signals belonging to dioxidized (type II) products 3 and 4 are observed at ∼57 and 96 ppm, and ∼45 and 101 ppm, respectively. The reaction pathways then diverge slightly between sulfur and selenium derivatives for subsequent oxidation steps (Figures S1.44 and S1.45). During the reaction of 1 with sulfur, the tetraoxidized structure, type V, with only two tertbutyl groups cleaved off (analogous to 5), can be observed prior to the final formation of compound 6 (i.e., type VI). This is evidenced by the appearance and subsequent disappearance of the single resonances at 37 ppm and 1.7 ppm in the 31P{1H} and 1H NMR spectra, respectively. Conversely, for selenium, the dicleaved (type V) intermediate was absent throughout E


Article

Inorganic Chemistry

Scheme 4. Proposed Mechanism for the Observed Steric Activation of C−N Bonds Based on 1H and 31P NMR Spectroscopic Studies

and two P2N2 isopropyl groups, suggesting symmetrically nonequivalent isopropyl groups on the two P2N2 rings. When compound 8 is reacted with 4.4 equiv of selenium under the same reaction conditions, the in situ 31P{1H} NMR spectrum revealed a signal pattern that is consistent with a fully oxidized product with only one iPr group having being cleaved, P4(μ-NiPr)5(μ-NH)Se4 (10). The 31P{1H} and 1H NMR spectra show complex signals at 55.3 and 30.7 ppm, and at 6.06 and 5.34 ppm (intensity ratio 1:2, respectively), supporting the successful formation of a monocleaved macrocycle. In contrast to our observations for 6 and 7, where three C− N bonds were cleaved and a 2:1 ratio (bridging to P2N2 ring ratio) for the newly formed NH groups was observed in the 1H NMR spectrum. Compound 10 displays a 1:2 ratio, which is equal to the relative ratio of bridging to P2N2 tert-butyl groups in the starting material 8, suggesting the coexistence of both bridging and P2N2 monocleaved structures following the original tert-butyl ratio displayed by 8. This was further corroborated by our single crystal X-ray diffraction studies showing the cleavage of only one iPr group, which is modeled to be disordered over the six nitrogen atoms present within the P4N6 backbone of 10 (Figure 6b, c). The mean P−N and PSe bond distances in 10 are 1.700 and 2.075 Å, respectively, with the P2N2 rings being almost planar (puckered by 4.7°) and perpendicular to the macrocyclic plane (ca. 89°). Unfortunately, compound 10 readily decomposes upon isolation impeding further characterization. Despite several efforts to synthesize analogues to 1 and 8 containing less bulky substituents (e.g., nPr, Et, or Me), the

2.2 equiv of elemental selenium at room temperature in THF for 24 h, affording the expected diagonally dioxidized P4(μNiPr)6Se2 (9) (Scheme 5). Scheme 5. Synthesis of P4(μ-NiPr)6Se2 (9) and P4(μNiPr)5(μ-NH)Se4 (10)

Compound 9 displays two resonances in the 31P{1H} NMR spectrum at 101.2 and 64.6 ppm, belonging to P(III) and P(V) centers, respectively. There are two multiplets at 5.16 and 3.82 ppm in 1H NMR spectrum of compound 9 with two and four protons corresponding to the bridging and P2N2 ring isopropyl groups, respectively. The three peaks at 1.56, 1.38, and 1.35 ppm with 12 protons for each can be attributed to the bridging F


Article

Inorganic Chemistry

Figure 6. (a) Solid-state structures of 9. The isopropyl units are drawn as wire frames. H atoms are omitted for clarity. Thermal ellipsoids are set at the 50% probability level. For selected bond lengths [Å] and angles [deg], see SI. (b, c) Structural representations of the two species present in 10. See Supporting Information for crystallographic data.

Table S2.15), which is also consistent with the experimentally observed trend. C−N Bond Activation: Dioxidized versus Tetraoxidized. To gain insights on the factors under which the C−N bond activation process is only observed for tetraoxidized species, we have computationally modeled the C−N bond activation and cleavage processes for di- and tetraoxidized systems, P4(μ-NtBu)6O2 and P4(μ-NtBu)6O4, types II and III, respectively, at the ωB97X-D/6-31G(d) level of theory. The relative Gibbs free energy comparison for the reaction pathway has been calculated for the two first C−N bond cleavage processes and is shown in Figure 8 for type II (i.e., compound 2) and Figure 9 for type III (hypothetical tetraoxidized uncleaved species). In both Figures 8 and 9, the red pathway denotes the first C−N bond cleavage at the P2N2 ring, while the blue pathway represents the cleavage at the bridging position. DFT calculations revealed some noteworthy results. In general, the calculated relative free energy for the first C−N bond cleavage for the dioxidized type II is always observed to be higher than that for their tetraoxidized type III counterparts. This trend is consistent for both of the cleavages at the P2N2 ring (TSIa) (66.7 vs 65.6 kcal·mol−1) and bridging positions (TSIb) (38.5 vs 25.6 kcal·mol−1), respectively (see Figure 8 and 9, respectively). The computational modeling is also consistent with our experimental observations, whereby the C−N bond cleavage was not observed for species originating from type II macrocycles (i.e., types VI′, V′, and VI′). Focusing on the tetraoxidized (P4(μ-NtBu)6O4) macrocycle, type III, the energy barrier for the first C−N bond scission to occur at the bridging position (TSIb, 25.6 kcal·mol−1) is energetically more favorable than for at the P2N2 ring (TSIa, 65.6 kcal·mol−1), as shown in Figure 9. Moreover, intrinsic reaction coordinate (IRC) calculations for the first cleavage process were carried out, which connect the transition states (TSIa and TSIb) with the starting material (III) and the proposed hydroxyphosphane intermediates (IntIa 31.6 kcal· mol−1 and IntIb 11.4 kcal·mol−1) formed following the C−N cleavage. These proposed hydroxyphosphane transient species would quickly undergo intramolecular proton transfer to form their corresponding phospazane products IVa-(4.6 kcal·mol−1) or IVb (−5.8 kcal·mol−1), respectively. Following the initial cleavage process, we only explored the subsequent second cleavage starting from the more thermodynamically favorable product IVb (In this case, bond scission at the bridging or P2N2 positions would led to two different dicleaved products (i.e., Vb1 and Vb2, respectively). Our calculations indicate that the second cleavage process occurring at the bridging position is

isolation of analogous species from the oxidation reactions were unsuccessful.

■

THEORETICAL STUDIES Throughout our investigations, it was revealed that when 1 is oxidized with different chalcogen elements, both di- and tetraoxidized frameworks can be readily isolated, with the latter displaying differing degrees of C−N cleavage that give rise to series of unique macrocyclic arrangements. To rationalize the energetics of the reaction and further understand the mechanism of the observed C−N bond cleavage, density functional theory (DFT) calculations were performed at the B3LYP/cc-pVDZ level of theory unless otherwise stated. To maintain consistency with the notation used for the proposed mechanism and the theoretically calculated species, for discussion purposes, type II and type III species correspond to uncleaved di- and hypothetical tetraoxidized species; types IV, V, and VI correspond to mono-, di-, and tricleavage tetraoxidized products, respectively. Their hypothetical dioxidized counterparts have been labeled as IV′, V′, and VI′, respectively. Dioxidation: cis versus trans. The relative free energies of the unoxidized reactants and dioxidized products were compared to assess their relative stabilities. The different isomers of the type II dioxidized species, IIARE−IICRE (R = iPr and tBu; E = O, S, and Se), whereby the chalcogen atoms are placed in different relative positions within the cage, were calculated (see Figure 7). The values obtained indicate that in

Figure 7. Different isomers of the dioxidized species.

all cases, the diagonally dioxidized isomer IIBRE, is the most favorable isomer (see Table S2.14), especially in the presence of tert-butyl substituent. Additionally, computational modeling indicates that the greater size of selenium and sulfur atoms, which correspond to greater steric bulk, increases the preference for the oxidation to occur at the most distant relative positions. This is consistent with the formation of compounds 2−4, whereby diagonally positioned phosphorus atoms were oxidized. Moreover, the C−N bond distance slightly elongates upon dioxidation (see G


Article

Inorganic Chemistry

Figure 8. Stepwise mechanistic pathway of the proposed sterically induced C−N bond cleavage reaction, with the calculated relative free energies of the intermediates and transition states shown in kcal·mol−1, with respect to the initial noncleavage product type II dioxidized P4(μ-NtBu)6O2 species.

Figure 9. Stepwise mechanistic pathway of the proposed sterically induced C−N bond cleavage reaction, with the calculated relative free energies of the intermediates and transition states shown in kcal·mol−1 with respect to the initial noncleavage product for type III tetraoxidized P4(μNtBu)6O4 species.

H


Article

Inorganic Chemistry more thermodynamically favored (TSIIb2, 16.9 kcal·mol−1) than the one in P2N2 ring (TSIIb1, 61.9 kcal·mol−1). Similar to first C−N cleavage, a hydroxyphosphane intermediate (IntIIb2, 3.0 kcal·mol−1) and an isobutylene molecule are formed prior to the formation of the final product Vb2 (−12.2 kcal·mol−1), which is fully consistent with the formation of the experimentally observed compound 5. Furthermore, noncovalent interaction (NCI)57,58 calculations (ωB97X-D/6-31G(d) level) conducted for the type II and type III species demonstrate that the steric hindrance associated with the latter is more pronounced (Figure 10). The

Å, respectively, while the P2N2 ring C−N bond distances are similar (ca. 1.49 Å). The Pairing Game in the Steric Activation of C−N Bonds: Substituents versus Chalcogens (Variations and Combinations). To explain the sterically driven nature of the observed C−N bond cleavage in the case of the fully oxidized species containing the studied triad of chalcogen atoms (i.e., O, S, and Se), the enthalpies of a series of hypothetical tetraoxidized products with increased steric crowding, IIIRE (R = nPr, iPr, and tBu; E = O, S, Se, and Te) were calculated. Within the context of our studies, reaction enthalpies provided similar trends as the previously obtained Gibbs free energies, but at a lower computational cost. This is because the inclusion of an entropic term, which inherently increases with increasing number of isobutylene molecules as byproducts of the C−N cleavage, would not account for the intrinsic stability of the calculated cleaved products. It would be observed that in the case of the less sterically crowded compounds IIInPrO and IIIiPrO, the cleavage of any C−N bond produces thermodynamically unfavorable species (see Table 1). However, this trend is reversed when steric crowding increases from IIInPrO to IIIiPrO and IIItBuO. For IIItBuO, the calculated enthalpies corresponding to the loss of an initial tert-butyl group, type IV species, are −7.64 and +3.25 kcal·mol−1 for bridging and P2N2 positions, respectively, which suggests that the first C−N cleavage would preferentially occur at the bridging nitrogen position. Subsequently, there are two possible scenarios in which the reaction could proceed to form the dicleaved type V species, either by a second cleavage taking place at a bridging position or within the P2N2 ring. Our calculations, −16.71 and −1.76 kcal·mol−1, respectively, suggest that the most favorable product formation results from the second cleavage occurring at the bridging C−N bond, which is consistent with the experimentally observed formation of compound 5. Any further C−N bond cleavage within VtBuO species produces thermodynamically less stable products, as shown in Table 1. Upon descending the group, the bond cleavage process becomes increasingly predominant as the chalcogen size increases. The enthalpy of di- and tricleaved species, types V and VI, is most favorable for sulfur and selenium (-69.85 and −69.64 kcal·mol−1 and −81.36 and −82.45 kcal·mol−1, respectively) and that of tri- and tetracleaved species is most favorable for tellurium (type V and a nonobserved type VII, −100.93 and −100.31 kcal·mol−1, respectively), which is

Figure 10. Noncovalent interaction computed at ωB97X-D/631G(D) level for type II and type III species. N-tert-butyl group on the P2N2 in the expansion has been omitted for clarity.

electronic density map of the NCI calculations enables visualization of the types of interactions formed between the functional groups in the molecules, with attractive interactions and repulsive interactions shown in blue and red, respectively. It is clear that strong repulsive interactions exist between the tert-butyl moiety and the two neighboring oxygen atoms in type III species, corroborating the observation that the longest C−N bond distance was found in the type III case. Thus, in the P4N6 macrocyclic ring, the C−N bond distance for the tetraoxidized type III and dioxidized type II are 1.55 and 1.53

Table 1. Relative Enthalpies (ΔH in kcal·mol−1) of Tetraoxidized Species with Increasing Degree of Cleavage Calculated at B3LYP/cc-pvdz Theory Levela P4(μ-NR)6O4 (IIIRO) no. cleavageb 0 1 1′ 2 1 + 1′ 3′ 4′ 5′ 6′

n

Pr

0.0 12.22 14.49 24.33 27.11 39.25 53.87 70.36 73.38

i

Pr

c 0.84 4.76 15.31 19.34 29.28 43.32 58.27 75.66

t

Bu

0.0 −7.64 3.25 −16.71c −1.76 −10.81 2.86 15.10 30.43

P4(μ-NR)6S4 (IIIRS) n

Pr

0.0 7.16 12.19 13.66 19.98 27.77 40.39 57.51 73.38

i

Pr

0.0 −0.61 7.07 3.56 9.11 11.67 21.65 38.29 53.70

t

P4(μ-NR)6Se4 (IIIRSe) Bu

0.0 −33.90 −19.84 −69.85c −43.25 −69.64c −64.45 −50.31 −38.15

n

Pr

0.0 7.74 13.91 12.94 20.95 26.89 39.11 54.96 70.44

i

Pr

0.0 −0.79c 7.35 1.00 7.55 8.50 17.60 34.69 49.83

t

Bu

0.0 −40.22 −27.31 −81.36 −53.78 −82.45c −78.81 −64.34 −52.66

P4(μ-NR)6Te4 (IIIRTe) n

Pr

0.0 5.17 12.84 8.36 18.43 21.12 33.63 50.18 65.79

i

Pr

0.0 −6.76 2.58 −10.76 −0.13 −3.28 7.75 23.86 39.99

t

Bu

0.0 −49.14 −40.88 −96.16 −70.34 −100.93 −100.31 −85.58 −75.16

Hypothetical species are italic. bCleavage at P2N2 positions are labeled with ′. cExperimentally observed.

a

I


Article

Inorganic Chemistry

Figure 11. (a) 31P{1H} NMR spectra of 5 (top), 6 (middle), and 7 (bottom) after exposure to air for 12 months; (b) 31P{1H} NMR spectra of 5 (top), 6 (middle), and 7 (bottom) in THF-d8/D2O = 4:1 for 2 weeks.

results present an exciting opportunity for future application of steric activation of bonds approach, as a strategy, to access these molecules.

corroborated by experimental observations in the case of selenium (i.e., compounds 6 and 7). Overall, our theoretical calculations are in good agreement with the observed experimental results and indicate a strong correlation between the combined steric bulk of the large chalcogen atoms and substituents on the N atoms with the overall extent of C−N bond scission. Air and Moisture Stability Studies. Last, but not least, the air and hydrolytic stabilities of compounds 5−7 were investigated using 1H and 31P NMR spectroscopies. Compounds 5−7 appear to be highly robust, without notable decomposition after exposure to air for 12 months or when dissolved in THF/H2O (4:1 ratio) mixtures for 2 weeks (Figure 11). This is aligned with our initial goal of stabilizing macrocyclic phosphazane frameworks macrocycle by oxidation of the phosphorus centers using chalcogen elements.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.8b01596. Experimental procedures, spectroscopic analyses, X-ray data, and selected bond lengths and angles (PDF) Theoretical methods, relative electronic energy, enthalpy, and entropy contributions of the complexes, enthalpy of all species relative to fully substituted ones, free energies of different isomers, bond distances of C− N bonds in 2, summaries of full coordinate reation paths for cleavage reactions, NCI computed for type 2 and 3 species, and Cartesian coordinates for complexes (PDF) Structure representations (XYZ format) of all complexes (ZIP)

■

CONCLUSIONS In summary, we have described the formation of di- and tetraoxidized dimeric cyclophosphazane frameworks [{P(μNR)}2(μ-NR)]2, R = tBu or iPr, that displayed an unexpected C−N bond activation and cleavage. Our experimental and theoretical studies suggest that the C−N bond activation is driven by steric crowding and strain release. More importantly, the observed bond cleavage process was exploited as a synthetic strategy to produce a series of unique phosphazane macrocyclic frameworks containing free unsubstituted mono-, di-, and tri-N−H moieties (P4(μ-NiPr)5(μ-NH)Se4 (10), P4(μNtBu)4(μ-NH)2O4 (5), and P4(μ-NtBu)3(μ-NH)3E4, E = S (6) and Se (7), respectively). These species open up new possibilities of using these macrocyclic phosphazane compounds as coordinative ligands to metal cations or as hydrogen bond donors for supramolecular chemistry inter alia.59 Overall, the results reported herein provide a novel proof-ofconcept for the use of steric activation of bonds for the rational synthesis of main group compounds that were previously unattainable by standard synthetic methods. Within the main group arena, there exists a large pool of “rigid” molecular arrangements containing diverse redox-active elements. Our

Accession Codes

CCDC 1432502 and 1527289−1527291 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/ cif, or by emailing [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

Felipe García: 0000-0002-9605-3611 Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. J


Article

Inorganic Chemistry Notes

novel tetranuclear gold(I) macrocycle, [{Au{(o-MeOC6H4O)P(μNtBu)}2}4](ClO4)4. Dalton Trans. 2009, 5478. (17) Roth, T.; Wadepohl, H.; Wright, D. S.; Gade, L. H. Chiral ditopic cyclophosphazane (CycloP) ligands: synthesis, coordination chemistry, and application in asymmetric catalysis. Chem. - Eur. J. 2013, 19, 13823. (18) García, F.; Goodman, J. M.; Kowenicki, R. A.; McPartlin, M.; Riera, L.; Silva, M. A.; Wirsing, A.; Wright, D. S. Selection of the cis and trans phosph(III)azane macrocycles[{P(μ-NtBu)}2(1-Y-2-NHC6H4)]2 (Y = O, S). Dalton Trans. 2005, 1764. (19) González-Calera, S.; Eisler, D. J.; Morey, J. V.; McPartlin, M.; Singh, S.; Wright, D. S. The selenium-based hexameric macrocycle [(Se=)P(μ-NtBu)2P(μ-Se)]6. Angew. Chem., Int. Ed. 2008, 47, 1111. (20) Kommana, P.; Kumara Swamy, K. C. Synthesis and structural characterization of a new class of macrocycles based on a cyclodiphosphazane skeleton. Inorg. Chem. 2000, 39, 4384. (21) Bond, A. D.; Doyle, E. L.; García, F.; Kowenicki, R. A.; McPartlin, M.; Riera, L.; Wright, D. S. Exo-metal coordination by a tricyclic [{P(μ-N-2-NC 5 H 4 )} 2 (μ-O)] 2 dimer in [{P(μ-N-2NC 5 H 4 )} 2 (μ-O)] 2 {CuCl·(C 5 H 5 N) 2 } 4 {2-NC 5 H 4 =2-pyridyl, C5H5N=pyridine). Chem. Commun. 2003, 2990. (22) García, F.; Kowenicki, R. A.; Kuzu, I.; Riera, L.; McPartlin, M.; Wright, D. S. The formation of dimeric phosph(III) azane macrocycles [{P(μ-NtBu}2·LL]2 [LL = organic spacer]. Dalton Trans. 2004, 2904. (23) Sim, Y.; Shi, Y.-X.; Ganguly, R.; Li, Y.; García, F. Mechanochemical Synthesis of Phosphazane-Based Frameworks. Chem. - Eur. J. 2017, 23, 11279. (24) García, F.; Kowenicki, R. A.; Kuzu, I.; McPartlin, M.; Riera, L.; Wright, D. S. The first complex of the pentameric phosphazane macrocycle [{P(μ-NtBu}2(μ-NH)]5 with a neutral molecular guest: Synthesis and structure of [{P(μ-NtBu}2(μ-NH)]5(CH2Cl2)2. Inorg. Chem. Commun. 2005, 8, 1060. (25) Bashall, A.; Doyle, E. L.; Tubb, C.; Kidd, S. J.; McPartlin, M.; Woods, A. D.; Wright, D. S. The tetrameric macrocycle [{P(μNtBu)}2NH]4. Chem. Commun. 2001, 2542. (26) Gonzalez-Calera, S.; Eisler, D. J.; Goodman, J. M.; McPartlin, M.; Singh, S.; Wright, D. S. Joining the crown family; the tetrameric, O-bridged macrocycle [{P(μ-NtBu)}2(μ-O)]4. Dalton Trans. 2009, 1293−1296. (27) Brask, J. K.; Chivers, T.; Krahn, M. L.; Parvez, M. {Li[P(NtBu)2]}4·0.25P4(NtBu)6: Solvent and substituent effects on the structures and reactivity of 1,3-Diaza-2-phosphaallyllithium Complexes. Inorg. Chem. 1999, 38, 290. (28) Bashall, A.; Bond, A. D.; Doyle, E. L.; García, F.; Kidd, S.; Lawson, G. T.; Parry, M. C.; McPartlin, M.; Woods, A. D.; Wright, D. S. Templating and selection in the formation of macrocycles containing [{P(μ-NtBu)2}(μ-NH)]n frameworks: observation of halide ion coordination. Chem. - Eur. J. 2002, 8, 3377. (29) Garcia, F.; Goodman, J. M.; Kowenicki, R. A.; Kuzu, I.; McPartlin, M.; Silva, M. A.; Riera, L.; Woods, A. D.; Wright, D. S. Selection of a pentameric host in the host-guest complexes {[{[P(μNtBu)]2(μ-NH)}5]·I}−[Li(thf)4]+ and [{[P(μ-NtBu)2(μ-NH)}5]· HBr-THF. Chem. - Eur. J. 2004, 10, 6066. (30) Plajer, A. J.; García-Rodríguez, R.; Benson, C. G. M.; Matthews, P. D.; Bond, A. D.; Singh, S.; Gade, L. H.; Wright, D. S. A modular approach to inorganic phosphazane macrocycles. Angew. Chem., Int. Ed. 2017, 56, 9242. (31) Shi, Y.-X.; Liang, R. Z.; Martin, K. A.; Weston, N.; GonzalezCalera, S.; Ganguly, R.; Li, Y.; Lu, Y.; Ribeiro, A. J. M.; Ramos, M. J.; Fernandes, P. A.; García, F. Synthesis and hydrolytic studies on the air-stable [(4-CN-PhO)(E)P(μ-NtBu)]2 (E = O, S, and Se) cyclodiphosphazanes. Inorg. Chem. 2015, 54, 6423. (32) Sim, Y.; Tan, D.; Ganguly, R.; Li, Y.; García, F. Orthogonality in main group compounds: a direct one-step synthesis of air- and moisture-stable cyclophosphazanes by mechanochemistry. Chem. Commun. 2018, 54, 6800. (33) Otang, M. E.; Lief, G. R.; Stahl, L. Alkoxido-, amido-, and chlorido derivatives of zirconium- and hafnium bis(amido)-

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS

■

REFERENCES

F. Garcı á would like to thank NTU start-up grant (M4080552), and MOE Tier 1 grant (M4011441) for financial support. D. Tan would like to thank A*STAR AME IRG (A1783c0003) for PDRF. J. D. thanks COMPUTAEX for granting access to LUSITANIA supercomputing facilities.

(1) Wright, D. S. In Comprehensive Inorganic Chemistry, Main Group Elements, Including Novel Gases; Chivers, T., Ed.; Elsevier, 2013; Vol. 1, p 953. (2) García, F.; Hehn, J. P.; Kowenicki, R. A.; McPartlin, M.; Pask, C. M.; Rothenberger, A.; Stead, M. L.; Wright, D. S. Reactions of Sn(NMe2)2 with alkali-metal tert-butylphosphides tBuPHM (M= Li, Na, K): Evidence for metal-induced modification of the tin(II) phosphinidene anions. Organometallics 2006, 25, 3275. (3) García, F.; Linton, D. J.; McPartlin, M.; Rothenberger, A.; Wheatley, A. E. H.; Wright, D. S. Synthesis and structure of [{Sb(μNCy)}2(μ-N)]3(Li·THF)3-(LiN= NH), containing a macrocyclic [{Sb(μ-NCy)}2N]33‑. trianion. Dalton Trans. 2002, 481. (4) Baxter, S. G.; Cowley, A. H.; Davis, R. E.; Riley, P. E. Crystal structure and low-temperature proton NMR spectrum of tetramesityldiphosphine. Evidence for the anticonformational preference in tetraaryldiphosphines. J. Am. Chem. Soc. 1981, 103, 1699. (5) Allan, R. E.; Beswick, M. A.; Raithby, P. R.; Steiner, A.; Wright, D. S. Build-up of an Al4P6Li4 cage from an Al4N4 cubane: synthesis and structure of [Li(OC 4 H 8 )]4 [{(AlMe)[μ-P(C 6H 11 )]} 2 {μ-P(C6H11)}]2·C6H5Me. J. Chem. Soc., Dalton Trans. 1996, 4153. (6) Nordheider, A.; Chivers, T.; Thirumoorthi, R.; Vargas-Baca, I.; Woollins, J. D. Planar P6E6 (E = Se, S) macrocycles incorporating P2N2 scaffolds. Chem. Commun. 2012, 48, 6346. (7) Allan, R. E.; Beswick, M. A.; Cromhout, N. L.; Paver, M. A.; Raithby, P. R.; Steiner, A.; Trevithick, M.; Wright, D. S. Fragmentation of an imido tin(II) cubane; syntheses and structures of heterobimetallic complexes containing tin(II) imido and phosphinidine anions. Chem. Commun. 1996, 1501. (8) Bryant, R.; James, S. C.; Jeffery, J. C.; Norman, N. C.; Orpen, A. G.; Weckenmann, U. Antimony imido and imido−amido compounds: a new route to an imidoantimony macrocycle. Dalton Trans. 2000, 4007. (9) Scherer, O. J.; Andres, K.; Krüger, C.; Tsay, Y.-H.; Wolmerhäser, G. P4(N-i-C3H7)6, a P4X6 molecule with and without adamantane structure. Angew. Chem., Int. Ed. Engl. 1980, 19, 571. (10) Beşli, S.; Yuksel, F.; Davies, D. B.; Kılıç, A. Conversion of a Cyclotriphosphazene to a Cyclohexaphosphazene by Ring Expansion. Inorg. Chem. 2012, 51, 6434−6436. (11) Goldwhite, H. Introduction to Phosphorus Chemistry; Press Syndicate of the University of Cambridge: Cambridge, U.K., New York, USA, Melbourne, Australia, 1981. (12) Luo, Y.-R. Comprehensive Handbook of Chemical Bond Energies; CRC: Boca Raton, FL, 2007. (13) Marsh, W. C.; Paddock, N. L.; Stewart, C. J.; Trotter, J. Dodeca(dimethylamido)cyclohexaphosphonitrile as a macrocyclic ligand. J. Chem. Soc. D 1970, 1190. (14) Dou, D.; Duesler, E. N.; Paine, R. T. Synthesis and structural characterization of new polyphosphorus ring and cage compounds. Inorg. Chem. 1999, 38, 788. (15) Chandrasekaran, P.; Mague, J. T.; Balakrishna, M. S. Tetranuclear rhodium(I) macrocycle containing cyclodiphosphazane [Rh2(μ-Cl)2(CO)2{(tBuNP(OC6H4OMe-o))2-κP]2 and its reversible conversion into trans-[Rh(CO)Cl{(tBuNP(OC6H4OMe-o))2-κP}2]. Organometallics 2005, 24, 3780. (16) Chandrasekaran, P.; Mague, J. T.; Balakrishna, M. S. Gold(I) complexes of cyclodiphosphazanes cis-{RP(μ-NtBu)}2: structure of a K


Article

Inorganic Chemistry

P(μ-NCy)]2, Showing Intermolecular Se···H−O Hydrogen Bonding. Eur. J. Inorg. Chem. 2011, 2011, 2264−2272. (53) Balakrishna, M. S.; Suresh, D.; Mague, J. T. Mono-, Bi-, Triand Tetranuclear Palladium(II), Copper(I), and Gold(I) Complexes of Morpholine- and N-Methylpiperazine-Functionalized Cyclodiphosph(III)azans, cis-[(tBuN-μ)2(PNC4H8X)2] (X = O, NMe). Eur. J. Inorg. Chem. 2010, 2010, 4201−4210. (54) The reaction with m-CPBA was difficult to control. The reaction was completed after only 20 min even at low temperature. There were no intermediate species observed during our studies using m-CPBA. (55) Ouyang, K.; Hao, W.; Zhang, W.-X.; Xi, Z. Transition-MetalCatalyzed Cleavage of C−N Single Bonds. Chem. Rev. 2015, 115, 12045−12090. (56) Nibbering, N. M. M. The McLafferty rearrangement: A personal recollection. J. Am. Soc. Mass Spectrom. 2004, 15, 956−958. (57) Contreras-García, J.; Johnson, E. R.; Keinan, S.; Chaudret, R.; Piquemal, J.-P.; Beratan, D. N.; Yang, W. NCIPLOT: A Program for Plotting Noncovalent Interaction Regions. J. Chem. Theory Comput. 2011, 7, 625−632. (58) Johnson, E. R.; Keinan, S.; Mori-Sánchez, P.; Contreras-García, J.; Cohen, A. J.; Yang, W. Revealing Noncovalent Interactions. J. Am. Chem. Soc. 2010, 132, 6498−6506. (59) Tan, D.; Ng, Z.-X.; Sim, Y.; Ganguly, R.; García, F. Ciscyclodiphosph(V/V)azanes as highly stable and robust main group supramolecular building blocks. CrystEngComm 2018, DOI: 10.1039/ C8CE00395E.

cyclodiphosph(V)azanes: Ligand ambidenticity and catalytic productivity. J. Organomet. Chem. 2016, 820, 98−110. (34) We have already reported the oxidation of 1 and 8 with elemental selenium, which are herein included for the completeness of our studies. Shi, Y.-X.; Liang, R. Z.; Martin, K. A.; Star, D. G.; Diaz, J.; Li, X. Y.; Ganguly, R.; García, F. Steric C-N bond activation on the dimeric macrocycle [{P(μ-NR)}2(μ-NR)]2. Chem. Commun. 2015, 51, 16468. (35) Niu, H.-C.; Plajer, A. J.; Garcia-Rodriguez, R.; Singh, S.; Wright, D. S. Designing the macrocyclic dimension in main group chemistry. Chem. - Eur. J. 2018, 24, 3073. (36) Mantina, M.; Chamberlin, A. C.; Valero, R.; Cramer, C. J.; Truhlar, D. G. Consistent van der Waals radii for the whole main group. J. Phys. Chem. A 2009, 113, 5806. (37) Keat, R.; Rycroft, D. S.; Thompson, D. G. Se-77 ChemicalShifts of Cyclodiphosphazane Selenides. Org. Magn. Reson. 1979, 12 (6), 391−392. Keat, R.; Rycroft, D. S.; Thompson, D. G. Selenium-77 Chemical-Shifts of Cyclodiphosphazane Selenides. Org. Magn. Reson. 1979, 12, 391−392. (38) McFarlane, W.; Wood, R. J. Nuclear magnetic doubleresonance studies of organo-selenium compounds. J. Chem. Soc., Dalton Trans. 1972, 1397−1402. (39) Li, S.; Allinger, N. L. Intramolecular van der Waals’ interactions and 1H chemical shifts: steric effects in some cyclic systems. Tetrahedron 1988, 44, 1339. (40) Shi, Y.-X.; Xu, K.; Clegg, J.; Ganguly, R.; Hirao, H.; Friscic, T.; García, F. The first synthesis of the sterically encumbered adamantoid phosphazane P4(NtBu)6: enabled by mechanochemistry. Angew. Chem., Int. Ed. 2016, 55, 12736. (41) Nordheider, A.; Woollins, J. D.; Chivers, T. Organophosphorus−Tellurium Chemistry: From Fundamentals to Applications. Chem. Rev. 2015, 115, 10378−10406. (42) Keat, R.; Thompson, D. G. Exchange effects on the 1H and 31P nmr spectra of cyclodiphosph(III)azane selenides and tellurides. J. Organomet. Chem. 1977, 141, C13−C16. (43) Scherer, O. J.; Schnabl, G. Elementorganische Amin/IminVerbindungen, XVI. Neue Diazadiphosphetidine. Chem. Ber. 1976, 109, 2996−3004. (44) Scherer, O. J.; Quintus, P.; Sheldrick, W. S. Elementorganische amin/iminverbindungen, XXVIII1) polycyclische phosphor-stic-stoffrhenium(I)-komplexe. Chem. Ber. 1987, 120, 1183. (45) Tirree, J.; Gudat, D.; Nieger, M.; Niecke, E. Reversible tautomeric transformation between a Bis(amino)cyclodiphosph(V)azene and a bis(imino)cyclodiphosph(V)azane. Angew. Chem., Int. Ed. 2001, 40, 3025. (46) Burck, S.; Gudat, D.; Nieger, M.; Schalley, C. A.; Weilandt, T. Phosphazene vs. diazaphospholene PN-bond cleavage in spirocyclic cyclodiphosphazenes. Dalton Trans. 2008, 3478. (47) Steiner, T. The hydrogen bond in the solid state. Angew. Chem., Int. Ed. 2002, 41, 48. (48) Potrzebowski, M. J.; Michalska, M.; Kozioł, A. E.; Każmierski, S.; Lis, T.; Pluskowski, J.; Ciesielski, W. Structural implications of C− H···S contacts in organophosphorus compounds. Studies of 1,6Anhydro-2-O-tosyl-4-S-(5,5-dimethyl-2-thioxa-1,3,2-dioxaphosphorinan-2-yl)-β-d-glucopyranose by X-ray and solid-state NMR methods. J. Org. Chem. 1998, 63, 4209. (49) Steiner, T. Unrolling the hydrogen bond properties of C-H···O interactions. Chem. Commun. 1997, 727. (50) Desiraju, G.; Steiner, T. The Weak Hydrogen Bond in Structure Chemistry and Biology; Oxford University Press, 2001. (51) Haagenson, D. C.; Lief, G. R.; Stahl, L.; Staples, R. J. N-versus O-silylation in cis-[(tBuHN)OP(μ-NtBu)2PO(NHtBu)] and [Me2Si(μ-NtBu)2PO(NHPh)]. Solid-state structures of their silylation products, of co-crystalline cis-[(tBuHN)OP(μ-NtBu)2PO(NHtBu)], and of {[Me2Si(μ-NtBu)2PO(N(SiMe3)Ph)] VCl3}. J. Organomet. Chem. 2008, 693, 2748−2754. (52) Chandrasekaran, P.; Mague, J. T.; Balakrishna, M. S. Synthesis and Derivatization of the Bis(amido)-λ3-cyclodi phosphazanescis[R(H)NP(μ-NR)]2, Including a Rare Example, trans-[tBu(H)N(Se)L


Synthesis of Unique Phosphazane Macrocycles via Steric Activation of

Recommend Documents