Dalton Transactions View Article Online
Published on 07 July 2015. Downloaded by California State University at Fresno on 21/07/2015 04:34:14.
PAPER
Cite this: DOI: 10.1039/c5dt02069g
View Journal
Synthesis and structures of [Sv(H)P(µ-NR)]2, potential building blocks for inorganic phosphorus–sulfur macrocycles† Callum G. M. Benson,a Vladislav Vasilenko,a,b Raúl García-Rodríguez,a Andrew D. Bond,a Silvia González Calera,a Lutz H. Gade*b and Dominic S. Wright*a The reactions of the chloro-phosph(III)azane dimers [ClP(μ-NR)]2 with LiSH give the dimers [Sv(H)-
Received 1st June 2015, Accepted 3rd July 2015 DOI: 10.1039/c5dt02069g www.rsc.org/dalton
P(μ-NR)]2 (III), which are potential new building blocks for inorganic macrocycles of the type [{P(μ-NR)}2(μ-S)]n. NMR spectroscopic studies and DFT calculations show that the preference for the cis or trans isomers of III is largely influenced by the steric demands of the R-group, with cis isomers being preferred for bulky substituents. This is an important factor in regard to applications in macrocycle synthesis since the cis arrangement is pre-organized for cyclisation.
Introduction Organic macrocycles (based on carbon backbones) such as crown ethers, calixarenes and porphyrins have a central role in modern coordination chemistry.1 In contrast, inorganic macrocycles (composed of non-carbon backbones) have been far less investigated, despite the potential to be more structurally and functionally diverse than their organic counterparts, given the broader range of bonding arrangements and the degree of electron donating or acceptor ability of the elements involved.2 A major factor hindering advances in this area is the lack of general synthetic methods for inorganic macromolecules, of the type that parallel those used in the organic arena. A rare example in which significant methodological inroads have been made is the P–N based phosph(III)azane macrocycles of the type [P(μ-NR)}2(μ-X)]n (X = NH, O).3 Underpinning their thermodynamic stability is the P–N single bond energy, which at 290 kJ mol−1 is not dissimilar to the C–C single bond energy (346 kJ mol−1). This development has required the synthesis of key dimeric precursors of the type [(HX)P(μ-NR)]2 (Fig. 1).4,5 For example, the amino-dimer I is accessible by direct reaction of NH3 with [ClP(μ-NtBu)]2 and adopts an all-cis structure which pre-organises it for cyclisation.4a Subsequent condensation with [ClP(μ-NtBu)]2 in the
a
Department of Chemistry, University of Cambridge, Lensfield Rd, Cambridge CB2 1EW, UK. E-mail: [email protected] b Anorganisch Chemisches Institut, Universität Heidelberg, Im Neuenheimer Feld 270, 69120 Heidelberg, Germany. E-mail: [email protected] † Electronic supplementary information (ESI) available. CCDC 1062561–1062563. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c5dt02069g
This journal is © The Royal Society of Chemistry 2015
Fig. 1 Structures of the key precursors I and II used in the synthesis of macrocycles [P(μ-NR)}2(μ-X)]n (X = NH, O), and the sulfur precursor (III) investigated in the current study.
Fig. 2 Structures of (a) the tetramer [{P(μ-NtBu)}2(μ-NH)]4 and (b) the pentamer [{P(μ-NtBu)}2(μ-NH)]5.
presence of Et3N as a Brønsted base gives the N–H bridged macrocyclic tetramer4a and pentamer4b,c shown in Fig. 2. Although controlled reaction of H2O with [ClP(μ-NtBu)]2 led to the O-analogue [Ov(H)P(μ-NtBu)]2 (II) (as the cis-PvO tautomer, Fig. 1b), this species is too unstable to be isolated and used as a precursor, dimerising rapidly to [{OvP(H)P(μ-NtBu)P}2(μ-O)] at ambient temperature.5a Instead, II can be ‘trapped’
Dalton Trans.
View Article Online
Published on 07 July 2015. Downloaded by California State University at Fresno on 21/07/2015 04:34:14.
Paper
Dalton Transactions
as the dianion [OvP(μ-NtBu)]22− by in situ reaction of H2O with nBuLi prior to in situ reaction with [ClP(μ-NtBu)]2. The product is the tetramer [{P(μ-NtBu)}2(μ-O)]4, which is the O-bridged analogue of the N–H bridged tetramer shown in Fig. 2a.5b The only related phosphazane macrocycles of the heavier Group 16 congeners (S, Se) reported are the hexamer [{P(μ-NtBu)PvSe}2(μ-Se)]6 6 and the trimer [(tBuNv)P(μ-NtBu)}2(μ-E2)]3 (E = S, Se),7 containing P(V)/P(III) or exclusively P(V) frameworks. In this paper we report the simple synthesis of sulfur phosph(V)azane dimers of the type [Sv(H)P(μ-NR)]2 (III, Fig. 1) and investigate the influence of the steric bulk of the R-group on the preference for the formation of cis and trans isomers and their stability towards decomposition and oligomerisation. This work represents an important step towards the ultimate synthesis of S-bridged phosph(III)azane and metalbridged macrocycles.
Results and discussion Pilot studies focused on the dimer [ClP(μ-NtBu)]2 as the precursor in order to establish the best experimental protocol for the synthesis of the sulfur dimer [Sv(H)P(μ-NtBu)]2 (IIIa). Following an analogous procedure to that employed earlier in the synthesis of the diamino-dimer I,4a a THF solution of [ClP(μ-NtBu)]2 was added dropwise to a THF solution of H2S at −78 °C. Stirring the reaction at ambient temperature, removal of the solvent under vacuum and extraction of the solid residue with toluene produced variable yields of IIIa (10–38%) as a powder after evaporation of the solvent under vacuum. However, the reaction of [ClP(μ-NtBu)]2 with an in situ prepared solution of LiSH in THF at −78 °C followed by the same workup produced significantly higher yields of pure IIIa (83%) (see Experimental section). The identity of IIIa was confirmed by multinuclear NMR (31P, 1H) and IR spectroscopy (see ESI†). The predominant cis isomer of IIIa was structurally characterised by X-ray diffraction of a crystal grown from a toluene solution at −5 °C (Fig. 3) (see Experimental section). Molecules
Fig. 3 X-ray crystal structure of the cis-isomer of IIIa. H-atoms (except those attached to P) have been omitted for clarity. Selected bond lengths (Å) and angles (°); P(1)–N(1) 1.6785(11), P(1)–N(2) 1.6817(11), P(1) ⋯P(2) 2.4991(4), P(2)–N(2) 1.6804(11), P(2)–N(1) 1.6799(11), P(1)–S(1) 1.9167(5), P(2)–S(2) 1.9143(5), P(1)–H(1) 1.322(17), P(2)–H(2) 1.314(18), N–P–N mean 83.88, P–N–P mean 96.10, N–PvS mean 122.15.
Dalton Trans.
Fig. 4 The fully H-coupled (c) IIIc and (d) IIId.
31
P NMR (298 K) spectra of (a) IIIa, (b) IIIb,
of cis-IIIa adopt the expected PV(H)vS (rather than PIII–SH) tautomer and have a planar P2N2 core arrangement. This isomer appears as a second-order AA′XX′ multiplet centred at δ 36.3 in the fully-coupled, room-temperature 31P NMR spectrum of IIIa, which contains only about 1% of the trans isomer (at δ 40.7, see Fig. 4a). The assignments of the 31P resonances for the cis and trans isomers are supported by DFT chemical shift calculations which show that the trans isomer should have a chemical shift ca. 2.7 ppm higher than the cis (exp. 4.4 ppm). The lower chemical shift found for the cis compared to the trans isomer is similar to the trend observed in bis(amino) cyclophosph(III)azanes [R2NP(μ-NR)]2.8 The presence of two isomers is also apparent in the room-temperature 1H NMR spectrum, which shows two tBu resonances [δ 1.38 (cis), 1.36 (trans)] and two second-order P-H multiplets [δ 8.40 (cis), 8.75 (trans)]. An in situ 31P NMR spectroscopic study of a sample of IIIa in toluene shows that it is gradually converted into transIIIa at higher temperature, but starts to decompose at ca. 333 K into a number of unidentified products. The final ratio of cis : trans isomers is ca. 7 : 1 at 333 K. However, the thermal decomposition of IIIa made van’t Hoff analysis of the
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 07 July 2015. Downloaded by California State University at Fresno on 21/07/2015 04:34:14.
Dalton Transactions
data unreliable. One noticeable feature of this process is that the conversion of trans-IIIa back to cis-IIIa is irreversible once the temperature is returned to room temperature. This indicates that the cis/trans inter-conversion has very high forward and reverse activation energies. Attempts to extend the method employed in the synthesis of IIIa to other dimers of this type [R = Dipp (2,6-iPr2C6H3), Mes (2,4,6-Me3C6H2), Ph2CH–, tBuCH2–] were hampered either by the instability of the product in solution at ambient temperature or the rapid formation of a variety of by-products. However, if worked up immediately after warming the reaction mixture to ambient temperature, then decomposition is minimised (see Experimental section). The new dimers [Sv(H)P(μ-NtBu)]2 [R = Mes (2,4,6-Me3C6H2) (IIIc), Ph2CH– (IIId)] can be obtained in moderate to good yields as powders using this method after removal of the solvent under vacuum (55–73%) (see ESI†). However, for R = Dipp (2,6-iPr2C6H3) the low purity of the powder obtained meant that a further crystallisation step was required, resulting in a lower final yield of 12% of the corresponding dimer IIIb. The 31P and 1H NMR spectra of IIIb, IIIc and IIId indicate that samples are of >95% purity. This synthetic approach is not reliable for less sterically bulky R groups (e.g., R = tBuCH2, 2-biphenyl), for which only minor yields of the dimers are obtained with a large number of impurities being present. Variable-temperature 31P NMR spectroscopic studies illustrate that the thermal stabilities of the new sulfur dimers is in the order IIIb, IIIc > IIIa > IIId. For example, whereas IIIb and IIIc are stable up to ca. 353 K, IIId decomposes significantly into other products at ca. 308 K. As illustrated in Fig. 4, which shows the H-coupled 31P NMR spectra of these species at ambient temperature, a further effect of the general reduction in the steric demands of the R-group is a switch from the cis isomer in the case of IIIa (1% trans, 99% cis), to a mixture of cis and trans for IIIb and IIIc (ca. 60% cis, 40% trans in IIIb, ca. 50% trans, 50% cis in IIIc) to predominantly the trans isomer in the case of IIIc (ca. 90% trans and 10% cis). Unlike IIIa, the 31 P NMR spectra of IIIb and IIIc show no changes in the ratios of the cis : trans isomers with increased temperature. As in the case of IIIa, the P–H region of the 1H NMR spectra of IIIb–IIId show similar second-order patterns for the P–H protons of their cis and trans isomers (see ESI†). The assignment of the 1H NMR spectrum of IIIb is particularly informative and was supported by a 2-D 31P/1H HSQC NMR experiment. For the cis diastereomer the two methyl groups of each iPr group are enantiotopic, whereas diastrereotopic methyl groups are found for the trans diastereomer (Fig. 5, top). Consequently, the room-temperature 1H NMR spectrum displays one septet resonance for the C–H protons of the iPr groups and two 1 : 1 Me resonances for the Me-groups of the i Pr substituents of the trans isomer. This is consistent with the C2 symmetric structure of the trans-IIIb revealed by singlecrystal X-ray analysis (Fig. 6), and results from restricted rotation of the Caryl–N and Caryl–C(iPr) bonds. In contrast, cisIIIb shows two C–H resonances for the iPr groups and two 1 : 1 resonances for the Me groups of the iPr substituents (Fig. 5,
This journal is © The Royal Society of Chemistry 2015
Paper
Fig. 5 Idealized C2h structure for trans-IIIb and C2v structure for cis-IIIb (top) and room-temperature 1H NOESY experiment showing throughspace correlation of Hb and Ha with the P–H protons (bottom).
Fig. 6 X-ray crystal structures of the trans-isomers of IIIb (top) and IIIc (bottom). H-atoms (except those attached to P) have been omitted for clarity. The molecules lies on a crystallographic inversion centre. A minor disorder component for the central core of IIIc is not shown; full details are given in the ESI.† Selected bond lengths (Å) and angles (°); IIIb: P(1)–N(1) 1.679(2), P(1)–N(1A) 1.676(2), P(1)⋯P(1A) 2.5055(16), P(1)– S(1) 1.8875(12), P(1)–H(1) 1.35(3), N–P–N 83.38(12), P–N–P 96.62(12), N–PvS mean 122.17; IIIc: P(1)–N(1) 1.680(3), P(1)–N(1A) 1.691(3), P(1)⋯P(1A) 2.5322(17), P(1)–S(1) 1.9084(13), N–P–N 82.63(13), P–N–P mean 97.38(13), N–PvS mean 121.2.
Dalton Trans.
View Article Online
Published on 07 July 2015. Downloaded by California State University at Fresno on 21/07/2015 04:34:14.
Paper
Dalton Transactions
top right). The 1H–1H NOESY NMR spectrum (Fig. 5, bottom) provides further support for the rigid structures of both isomers in solution by confirming that only one of the C–H (iPr) environments in the cis isomer is in close proximity to the P–H protons, whereas – for symmetry reasons – all of the C–H (iPr) protons of the trans isomer are close to the P–H protons. Similar restricted rotation was seen for IIIc where two separate broad signals are seen for the o-Me in the cis isomer whereas only one sharp signal is observed for the trans. Evidence for the less restricted nature of the Caryl–N bonds within the dimer IIIc was found by NOESY and EXSY 1H NMR as well as a variable-temperature NMR experiment. The singlecrystal X-ray structure of trans-IIIc is very similar to that of trans-IIIb (Fig. 6, bottom). Computational modelling We have performed DFT modelling of the cis and trans isomers of IIIa–d using the B3LYP/cc-pVTZ level of theory in THF solvent (PCM model).9–11 Frequency calculations of the optimized geometries were carried out, confirming the absence of imaginary frequencies. Thermodynamic parameters of all isomers are provided for 298 K and standard pressure (Table 1). The calculated thermodynamic data support the experimental observation from 31P NMR spectroscopy that the cis isomer is the most stable for bulkier R-groups (tBu, Dipp) but the trans isomer becomes most stable for less sterically encumbered R-groups (Mes, Ph2CH). Notably, depending on the steric demands of the substituents, the calculated cis/trans ratios range from strongly favoured cis (ΔG = 9.0 kJ/mol) to strongly favoured trans isomer (ΔG = −11.6 kJ mol−1). In contrast, the tautomers containing an SH group are all much higher in energy, regardless of the substituents attached to the P2N2-ring. It is clear, however, that this steric effect is a subtle one since the trend in the relative stabilities of the cis vs. trans isomers does not follow that expected on the basis of Tolman cone angle completely (for which the preference for the cis isomer should be IIIb > IIIc > IIIa > IIId). Although we cannot discount electronic factors as an influence on the preference for either isomer, we believe that the lower than expected steric influence of the Dipp and Mes groups of IIIb and IIIc is due to their perpendicular orientation with respect to the P2N2 rings. We next looked at possible mechanisms for the conversion of the cis isomer of IIIa to the trans isomer, in order to understand the apparent irreversibility of this isomerism. We found that typical intramolecular isomerization mechanisms such as
Table 1 Reaction energies (E, H, G, in kJ mol−1) for the conversion cis → trans for compounds IIIa–d. Thermodynamic parameters H and G were calculated for 298 K and standard pressure
Compound
ΔE
ΔH
ΔG
IIIa IIIb IIIc IIId
8.52 −0.05 −0.58 −10.31
8.42 −0.88 −1.03 −4.62
9.04 1.00 −4.22 −11.60
Dalton Trans.
(i) a concerted pathway in which cis-IIIa is converted into transIIIa directly via rotation of one P(H)vS fragment and (ii) a stepwise pathway involving P(H)vS/P–SH tautomerism and phosphorus lone pair inversion involve very high activation barriers (>150 kJ mol−1) and are therefore unlikely. In view of our findings in variable-temperature NMR experiments, it is probable that the relative cis/trans ratios of the different isomers are in part influenced by kinetic factors. Further mechanistic studies are required to resolve the nature of the transition states involved (intra- vs. intermolecular).
Conclusion The current work represents probably the most in-depth study of the cis and trans isomerism in any family of dimeric phosphazanes(III or V). We have shown that a series of dimers of the type [Sv(H)P(μ-NR)]2 can be readily obtained by the simple reactions of the chloro-dimers [ClP(μ-NR)]2 with LiSH, providing potential building blocks for the synthesis of S-bridged phosphazane macrocycles of the type [{P(μ-NR)}2(μ-S)]n. The relative thermodynamic stabilities of the cis and trans isomers of [Sv(H)P(μ-NR)]2 depend on the steric bulk of the R-group, so that more sterically encumbered groups result in a preference for the cis isomer while less sterically bulky groups favour the trans. This factor, and the apparently high activation energy between the isomers, is important in the condensation of [Sv(H)P(μ-NR)]2 into S-bridged macrocyclic phosphazanes since the cis-isomer is pre-organized for cyclisation while the trans is not.4 We are currently pursuing the metal-templated and non-templated condensation reactions of [Sv(H)P(μ-NR)]2 with [ClP(μ-NR)]2 to obtain macrocycles of the type [{P(μ-NR)}2(μ-S)]n, as well as coordination and deprotonation studies of the dimers themselves with a range of metals and metal bases ( potentially leading to metal-bridged macrocycles).
Experimental Synthesis All manipulations were carried out under dry, O2-free nitrogen on a vacuum-line, using standard inert-atmosphere techniques for isolation and characterisation. Chlorodimers were synthesized according to previously reported procedures.12 Synthesis of IIIa. nBuLi (6.2 mL, 10 mmol, in hexanes) was added to THF (40 ml) and the solution cooled to −78 °C. H2S gas was bubbled slowly though this solution for ca. 20 min. The reaction was allowed to warm to ambient temperature and the THF removed completely under vacuum, to give white powder of LiSH (this step ensures excess H2S is removed from the reaction). The solid was redissolved in THF (40 mL) and cooled to −78 °C. A solution of [ClP(μ-NtBu)]2 (1.375 g, 5 mmol) in THF (20 mL) was added slowly dropwise and the resulting mixture allowed to warm to ambient temperature and stirred for 72 h. The solvent was removed and the resulting solid was suspended in toluene (80 mL) and filtered. The
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 07 July 2015. Downloaded by California State University at Fresno on 21/07/2015 04:34:14.
Dalton Transactions
solvent was completely removed from the filtrate under vacuum and the semi-solid formed was washed with n-pentane (2 × 20 mL) and dried under vacuum. Yield of powder 1.15 g (83%). 1H NMR (500.2 MHz, CDCl3, +25 °C), cis isomer, δ /ppm = 8.40 (2H, P–H, AA′XX′ mult. apparent Jmax = 598 Hz), 1.38 (s, 18H, tBu); trans isomer, δ /ppm = 8.75 (P–H, AA′XX′ mult., cis isomer), 1.36 (s, tBu), (ratio of trans : cis ≈ 1 : 50 from P–H integration). 31P NMR (161.7 MHz, CDCl3, rel. 85% H3PO4/D2O. +25 °C), δ /ppm = 36.3 (AA′XX′ mult., cis isomer), 40.7 (AA′XX′ mult., trans isomer). IR (NaCl windows, Nujol mull), ν/cm−1 = 2360 (m) (P–H str.), 762 (s) (PvS str.). Synthesis of IIIb, IIIc, and IIId. The synthetic route employed was exactly the same as for IIIa but immediately after the addition of the dimer the reactions were allowed to warm to ambient temperature. At this point, the solvent was removed under vacuum and the solid extracted in the same manner as for IIIa. Yields of powders produced (200 mg scale for all): IIIb (200 mg of [ClP(μ-NDipp)]2, 30 mg of IIIb produced after further crystallisation from n-pentane, 15%), IIIc (200 mg of [ClP(μ-NMes)]2, 0.144 g of IIIc produced, 73%), IIId (200 mg of [ClP(μ-NCHPh2)]2, 0.109 g of IIId produced, 55%). Crystals of trans-IIIb were grown from pentane at −16 °C and crystals of trans-IIIc were grown from hexane/THF at room temperature. Unit cell analyses of crystalline samples of both indicate that only the trans isomers crystallise under these conditions. NMR: IIIb. 1H NMR (500.2 MHz, Tol d8, +25 °C), 1H NMR (500.2 MHz, CDCl3, +25 °C), cis isomer 9.04 (mult., Jmax = 593 Hz), 4.30 (sept., CHb Dipp), 3.00 (sept., CHa Dipp), 1.40 (d, Me Dipp), 0.86 (d, Me Dipp); trans isomer, 9.32 (mult., Jmax = 591 Hz), 3.78 (sept., CH Dipp), 1.10 (d, Me Dipp), 0.98 (d, Me Dipp). 31P NMR (161.7 MHz, Tol d8, rel. 85% H3PO4/D2O. +25 °C), δ /ppm = 53.1 (second-order mult., trans isomer, apparent 1JPH = 594 Hz), 45.05 (second-order mult., cis isomer, apparent 1JPH = 591 Hz) (ratio of trans : cis ≈ 1 : 1). IIIc: 1H NMR (500.2 MHz, Tol d8, +25 °C), cis isomer, 9.07 (2H, mult., P–H cis, Jmax = 592 Hz), 6.72 (2H, s., C–Haryl Mes cis, 6.61 (2H, s, C–Haryl Mes cis) 3.03 (s., 6H, o-Me Mes cis), 2.08 (s., 12H, p-Me Mes cis and trans), 1.95 (s., 6H, o-Me Mes cis); trans isomer, 9.29 (2H, mult., P–H trans, Jmax = 593 Hz), 6.65 (4H, s, C–Haryl Mes), 2.47 (12H, s, o-Me Mes trans). 31P NMR (161.7 MHz, Tol d8, rel. 85% H3PO4/D2O. +25 °C), δ /ppm = 44.6 (second-order mult., trans isomer, apparent 1JPH = 594 Hz; 31P{1H} d., 2JP–P 2.0 Hz), 36.7 (second-order mult., cis isomer, apparent 1JPH = 593 Hz) (ratio of trans : cis ≈ 1 : 1). IIId: 1H NMR (500.2 MHz, CDCl3, +25 °C), δ /ppm = 7.44 (2H, P–H, AA′XX′ mult. apparent Jmax = 550 Hz), 6.9–7.7 (overlapping mult. Ph of Ph2CH), 5.62 (dd., Ph2CH). 31P NMR (161.7 MHz, CDCl3, rel. 85% H3PO4/ D2O. +25 °C), δ = 40.0 (second-order mult., trans isomer, apparent 1JPH = 556 Hz), 38.2 (br. d., cis isomer, apparent 1 JPH = 579 Hz) (ratio of trans : cis ≈ 9 : 1). Satisfactory C, H and N analyses of IIIa–IIId could not be obtained. In particular, the N contents of each are found to be consistently and repeatedly low. This is a common problem in the analysis of phospha(III)zanes and is most likely to be due to incomplete combustion.4
This journal is © The Royal Society of Chemistry 2015
Paper
X-ray data ˉ, Z = 2, cis-IIIa, C8H20N2S2P2, M = 270.32, triclinic, space group P1 a = 5.9112(2), b = 9.9768(3), c = 12.1132(3) Å, α = 85.594(2), β = 88.593(2), γ = 79.0554(10)°, V = 699.27(4) Å3, µ(Mo-Kα) = 0.580 mm−1, ρcalc = 1.284 Mg m−3, T = 180(2) K. Total reflections 6232, unique 2952 (Rint = 0.026). R1 = 0.029 [I > 2σ(I)] and wR2 = 0.078 (all data). Trans-IIIb, C24H36N2P2S2, M = 478.61, monoclinic, space group P21/n, Z = 2, a = 8.3695(19), b = 10.1458(17), c = 15.541(3) Å, β = 101.646(7)°, V = 1292.5(4) Å3, µ(Mo-Kα) = 0.344 mm−1, ρcalc = 1.230 Mg m−3, T = 180(2) K. Total reflections 11 418, unique 1743 (Rint = 0.090). R1 = 0.039 [I > 2σ(I)] and wR2 = 0.087 (all data). Trans-IIIc, C18H24N2P2S2, M = 394.45, monoclinic, space group P21/c, Z = 2, a = 10.4215(3), b = 11.0475(3), c = 9.3492(3) Å, β = 109.462(2)°, V = 1014.89(5) Å3, µ(Cu-Kα) = 3.878 mm−1, ρcalc = 1.291 Mg m−3, T = 180(2) K. Total reflections 13 746, unique 1755 (Rint = 0.049). R1 = 0.046 [I > 2σ(I)] and wR2 = 0.110 (all data). Data for cis-IIIa were collected on a Nonius KappaCCD instrument, for trans-IIIb on a Bruker SMART X2S instrument and for trans-IIIc on a Bruker D8 QUEST diffractometer. All data were refined by full-matrix least squares on F2 using SHELXL-2014.13 The structure of trans-IIIc contains a minor disorder component for the central core, as described in the ESI.† CCDC 1062561–1062563 contain the supplementary crystallographic data. Computational modelling All calculations were performed with Gaussian 09 software suite of programs. Optimized Geometries of all the isomers were calculated in THF (PCM Model, cc-pVTZ basis set, B3LYP functional) and relative ΔG and enthalpies ΔH were determined at 298 K and standard pressure. In addition, relative electronic energies E were determined at 0 K and vacuum. Frequency calculations were performed for all stationary points, with no imaginary frequencies for minima and a single imaginary frequency for first-order saddle points (transition states). NMR calculations were carried out for the tBu dimer using the standard GIAO method as implemented in Gaussian 09. Here, optimized gas-phase geometries (B3LYP/cc-pVTZ) of all compounds were used. In the NMR calculation, the B3LYP functional was employed. The P atoms were described by the cc-pV5Z basis set, for all other atoms the cc-pVTZ basis set was used. The calculated chemical shifts were referenced vs. the experimental data of the cis (H)PvS isomer. For compounds where the two P atoms should be practically equivalent, an average chemical shift was calculated. Details of all of the models calculated are provided in the ESI.†
Acknowledgements We thank the ERC (Advanced grant for D.S.W., studentship for C.G.M.B.), Deutschlandstipendium (National Scholarship, V.V.) and the EU (Marie Curie Intra European Fellowship for R.G.-R.) for financial support. We also thank Dr J. E. Davies for collecting X-ray data for IIIa.
Dalton Trans.
View Article Online
Paper
Dalton Transactions
Published on 07 July 2015. Downloaded by California State University at Fresno on 21/07/2015 04:34:14.
Notes and references 1 (a) E. L. Doyle, L. Riera and D. S. Wright, Eur. J. Inorg. Chem., 2003, 3279; (b) M. S. Balakrishna, D. J. Eisler and T. Chivers, Chem. Soc. Rev., 2007, 36, 650; (c) J. S. Bradshaw, R. M. Izatt, A. V. Bardunov, C. Y. Zhu and J. K. Hathaway, in Comprehensive Supramolecular Chemistry, ed. J. L. Atwood, J. E. D. Davies, D. D. McNicol and F. Vögtle, Pergamon, Oxford, 1996, vol. 1, ch. 2; A. Pochini and R. Ungaro, in Comprehensive Supramolecular Chemistry, ed. F. Vögtle, Pergamon, Oxford, 1996, vol. 2, ch. 4; K. Odashima and K. Koga, in Comprehensive Supramolecular Chemistry, ed. F. Vögtle, Pergamon, Oxford, 1996, vol. 2, ch. 5. 2 D. S. Wright, Host–Guest Chemistry; p-Block Systems, in Comprehensive Inorganic Chemistry, Main Group Elements, Including Nobel Gases, ed. T. Chivers, 2013, vol. 1, ch. 1.30, p. 953. 3 S. Gonzalez Calera and D. S. Wright, Dalton Trans., 2010, 39, 5055. 4 (a) A. Bashall, E. L. Doyle, C. Tubb, S. J. Kidd, M. McPartlin, A. D. Woods and D. S. Wright, Chem. Commun., 2001, 2542; (b) A. Bashall, A. D. Bond, E. L. Doyle, F. Garcıa, S. J. Kidd, G. T. Lawson, M. C. Parry, M. McPartlin, A. D. Wood and D. S. Wright, Chem. – Eur. J., 2002, 8, 3377; (c) F. García, J. M. Goodman, R. A. Kowenicki, I. Kuzu, M. McPartlin, M. A. Silva, L. Riera, A. D. Woods and D. S. Wright, Chem. – Eur. J., 2004, 10, 6066. 5 (a) W. T. K. Chan, F. García, S. Gonzalez-Calera, M. McPartlin, J. V. Morey, R. E. Mulvey, S. Singh and
Dalton Trans.
6
7
8
9 10
11
12
13
D. S. Wright, Chem. Commun., 2008, 2251; (b) S. Gonzalez Calera, D. J. Eisler, J. M. Goodman, M. McPartlin, S. Singh and D. S. Wright, Dalton Trans., 2009, 1293. S. Gonzalez Calera, D. J. Eisler, J. V. Morey, M. McPartlin, S. Singh and D. S. Wright, Angew. Chem., Int. Ed., 2008, 47, 1111. A. Nordheider, T. Chivers, R. Thirumoorthi, I. Vargas-Baca and J. D. Woollins, Chem. Commun., 2012, 48, 6346. A. Tarassoli, M. L. Thomson, R. C. Haltiwanger, T. G. Hill and A. D. Norman, Inorg. Chem., 1988, 27, 3382 and references therein. M. J. Frisch, et al., Gaussian 09, Revision B.01, Gaussian, Inc., Wallingford CT, 2010. (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652; (b) C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter, 1988, 37, 785; (c) S. H. Vosko, L. Wilk and M. Nusair, Can. J. Phys., 1980, 58, 1200–1211; (d) P. J. Stephens, F. J. Devlin, C. F. Chabalowski and M. J. Frisch, J. Phys. Chem., 1994, 98, 11623. (a) T. H. Dunning Jr., J. Chem. Phys., 1989, 90, 1007; (b) D. E. Woon and T. H. Dunning Jr., J. Chem. Phys., 1993, 98, 1358; E. R. Davidson, Chem. Phys. Lett., 1996, 260, 514– 518. (a) T. Roth, H. Wadepohl, D. S. Wright and L. H. Gade, Chem. – Eur. J., 2013, 19, 13823; (b) T. Roth, V. Vasilenko, C. G. M. Benson, H. Wadepohl, D. S. Wright and L. H. Gade, Chem. Sci., 2015, 6, 2506. G. M. Sheldrick, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 2015, 71, 3.
This journal is © The Royal Society of Chemistry 2015
Published on 07 July 2015. Downloaded by California State University at Fresno on 21/07/2015 04:34:14.
PAPER
Cite this: DOI: 10.1039/c5dt02069g
View Journal
Synthesis and structures of [Sv(H)P(µ-NR)]2, potential building blocks for inorganic phosphorus–sulfur macrocycles† Callum G. M. Benson,a Vladislav Vasilenko,a,b Raúl García-Rodríguez,a Andrew D. Bond,a Silvia González Calera,a Lutz H. Gade*b and Dominic S. Wright*a The reactions of the chloro-phosph(III)azane dimers [ClP(μ-NR)]2 with LiSH give the dimers [Sv(H)-
Received 1st June 2015, Accepted 3rd July 2015 DOI: 10.1039/c5dt02069g www.rsc.org/dalton
P(μ-NR)]2 (III), which are potential new building blocks for inorganic macrocycles of the type [{P(μ-NR)}2(μ-S)]n. NMR spectroscopic studies and DFT calculations show that the preference for the cis or trans isomers of III is largely influenced by the steric demands of the R-group, with cis isomers being preferred for bulky substituents. This is an important factor in regard to applications in macrocycle synthesis since the cis arrangement is pre-organized for cyclisation.
Introduction Organic macrocycles (based on carbon backbones) such as crown ethers, calixarenes and porphyrins have a central role in modern coordination chemistry.1 In contrast, inorganic macrocycles (composed of non-carbon backbones) have been far less investigated, despite the potential to be more structurally and functionally diverse than their organic counterparts, given the broader range of bonding arrangements and the degree of electron donating or acceptor ability of the elements involved.2 A major factor hindering advances in this area is the lack of general synthetic methods for inorganic macromolecules, of the type that parallel those used in the organic arena. A rare example in which significant methodological inroads have been made is the P–N based phosph(III)azane macrocycles of the type [P(μ-NR)}2(μ-X)]n (X = NH, O).3 Underpinning their thermodynamic stability is the P–N single bond energy, which at 290 kJ mol−1 is not dissimilar to the C–C single bond energy (346 kJ mol−1). This development has required the synthesis of key dimeric precursors of the type [(HX)P(μ-NR)]2 (Fig. 1).4,5 For example, the amino-dimer I is accessible by direct reaction of NH3 with [ClP(μ-NtBu)]2 and adopts an all-cis structure which pre-organises it for cyclisation.4a Subsequent condensation with [ClP(μ-NtBu)]2 in the
a
Department of Chemistry, University of Cambridge, Lensfield Rd, Cambridge CB2 1EW, UK. E-mail: [email protected] b Anorganisch Chemisches Institut, Universität Heidelberg, Im Neuenheimer Feld 270, 69120 Heidelberg, Germany. E-mail: [email protected] † Electronic supplementary information (ESI) available. CCDC 1062561–1062563. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c5dt02069g
This journal is © The Royal Society of Chemistry 2015
Fig. 1 Structures of the key precursors I and II used in the synthesis of macrocycles [P(μ-NR)}2(μ-X)]n (X = NH, O), and the sulfur precursor (III) investigated in the current study.
Fig. 2 Structures of (a) the tetramer [{P(μ-NtBu)}2(μ-NH)]4 and (b) the pentamer [{P(μ-NtBu)}2(μ-NH)]5.
presence of Et3N as a Brønsted base gives the N–H bridged macrocyclic tetramer4a and pentamer4b,c shown in Fig. 2. Although controlled reaction of H2O with [ClP(μ-NtBu)]2 led to the O-analogue [Ov(H)P(μ-NtBu)]2 (II) (as the cis-PvO tautomer, Fig. 1b), this species is too unstable to be isolated and used as a precursor, dimerising rapidly to [{OvP(H)P(μ-NtBu)P}2(μ-O)] at ambient temperature.5a Instead, II can be ‘trapped’
Dalton Trans.
View Article Online
Published on 07 July 2015. Downloaded by California State University at Fresno on 21/07/2015 04:34:14.
Paper
Dalton Transactions
as the dianion [OvP(μ-NtBu)]22− by in situ reaction of H2O with nBuLi prior to in situ reaction with [ClP(μ-NtBu)]2. The product is the tetramer [{P(μ-NtBu)}2(μ-O)]4, which is the O-bridged analogue of the N–H bridged tetramer shown in Fig. 2a.5b The only related phosphazane macrocycles of the heavier Group 16 congeners (S, Se) reported are the hexamer [{P(μ-NtBu)PvSe}2(μ-Se)]6 6 and the trimer [(tBuNv)P(μ-NtBu)}2(μ-E2)]3 (E = S, Se),7 containing P(V)/P(III) or exclusively P(V) frameworks. In this paper we report the simple synthesis of sulfur phosph(V)azane dimers of the type [Sv(H)P(μ-NR)]2 (III, Fig. 1) and investigate the influence of the steric bulk of the R-group on the preference for the formation of cis and trans isomers and their stability towards decomposition and oligomerisation. This work represents an important step towards the ultimate synthesis of S-bridged phosph(III)azane and metalbridged macrocycles.
Results and discussion Pilot studies focused on the dimer [ClP(μ-NtBu)]2 as the precursor in order to establish the best experimental protocol for the synthesis of the sulfur dimer [Sv(H)P(μ-NtBu)]2 (IIIa). Following an analogous procedure to that employed earlier in the synthesis of the diamino-dimer I,4a a THF solution of [ClP(μ-NtBu)]2 was added dropwise to a THF solution of H2S at −78 °C. Stirring the reaction at ambient temperature, removal of the solvent under vacuum and extraction of the solid residue with toluene produced variable yields of IIIa (10–38%) as a powder after evaporation of the solvent under vacuum. However, the reaction of [ClP(μ-NtBu)]2 with an in situ prepared solution of LiSH in THF at −78 °C followed by the same workup produced significantly higher yields of pure IIIa (83%) (see Experimental section). The identity of IIIa was confirmed by multinuclear NMR (31P, 1H) and IR spectroscopy (see ESI†). The predominant cis isomer of IIIa was structurally characterised by X-ray diffraction of a crystal grown from a toluene solution at −5 °C (Fig. 3) (see Experimental section). Molecules
Fig. 3 X-ray crystal structure of the cis-isomer of IIIa. H-atoms (except those attached to P) have been omitted for clarity. Selected bond lengths (Å) and angles (°); P(1)–N(1) 1.6785(11), P(1)–N(2) 1.6817(11), P(1) ⋯P(2) 2.4991(4), P(2)–N(2) 1.6804(11), P(2)–N(1) 1.6799(11), P(1)–S(1) 1.9167(5), P(2)–S(2) 1.9143(5), P(1)–H(1) 1.322(17), P(2)–H(2) 1.314(18), N–P–N mean 83.88, P–N–P mean 96.10, N–PvS mean 122.15.
Dalton Trans.
Fig. 4 The fully H-coupled (c) IIIc and (d) IIId.
31
P NMR (298 K) spectra of (a) IIIa, (b) IIIb,
of cis-IIIa adopt the expected PV(H)vS (rather than PIII–SH) tautomer and have a planar P2N2 core arrangement. This isomer appears as a second-order AA′XX′ multiplet centred at δ 36.3 in the fully-coupled, room-temperature 31P NMR spectrum of IIIa, which contains only about 1% of the trans isomer (at δ 40.7, see Fig. 4a). The assignments of the 31P resonances for the cis and trans isomers are supported by DFT chemical shift calculations which show that the trans isomer should have a chemical shift ca. 2.7 ppm higher than the cis (exp. 4.4 ppm). The lower chemical shift found for the cis compared to the trans isomer is similar to the trend observed in bis(amino) cyclophosph(III)azanes [R2NP(μ-NR)]2.8 The presence of two isomers is also apparent in the room-temperature 1H NMR spectrum, which shows two tBu resonances [δ 1.38 (cis), 1.36 (trans)] and two second-order P-H multiplets [δ 8.40 (cis), 8.75 (trans)]. An in situ 31P NMR spectroscopic study of a sample of IIIa in toluene shows that it is gradually converted into transIIIa at higher temperature, but starts to decompose at ca. 333 K into a number of unidentified products. The final ratio of cis : trans isomers is ca. 7 : 1 at 333 K. However, the thermal decomposition of IIIa made van’t Hoff analysis of the
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 07 July 2015. Downloaded by California State University at Fresno on 21/07/2015 04:34:14.
Dalton Transactions
data unreliable. One noticeable feature of this process is that the conversion of trans-IIIa back to cis-IIIa is irreversible once the temperature is returned to room temperature. This indicates that the cis/trans inter-conversion has very high forward and reverse activation energies. Attempts to extend the method employed in the synthesis of IIIa to other dimers of this type [R = Dipp (2,6-iPr2C6H3), Mes (2,4,6-Me3C6H2), Ph2CH–, tBuCH2–] were hampered either by the instability of the product in solution at ambient temperature or the rapid formation of a variety of by-products. However, if worked up immediately after warming the reaction mixture to ambient temperature, then decomposition is minimised (see Experimental section). The new dimers [Sv(H)P(μ-NtBu)]2 [R = Mes (2,4,6-Me3C6H2) (IIIc), Ph2CH– (IIId)] can be obtained in moderate to good yields as powders using this method after removal of the solvent under vacuum (55–73%) (see ESI†). However, for R = Dipp (2,6-iPr2C6H3) the low purity of the powder obtained meant that a further crystallisation step was required, resulting in a lower final yield of 12% of the corresponding dimer IIIb. The 31P and 1H NMR spectra of IIIb, IIIc and IIId indicate that samples are of >95% purity. This synthetic approach is not reliable for less sterically bulky R groups (e.g., R = tBuCH2, 2-biphenyl), for which only minor yields of the dimers are obtained with a large number of impurities being present. Variable-temperature 31P NMR spectroscopic studies illustrate that the thermal stabilities of the new sulfur dimers is in the order IIIb, IIIc > IIIa > IIId. For example, whereas IIIb and IIIc are stable up to ca. 353 K, IIId decomposes significantly into other products at ca. 308 K. As illustrated in Fig. 4, which shows the H-coupled 31P NMR spectra of these species at ambient temperature, a further effect of the general reduction in the steric demands of the R-group is a switch from the cis isomer in the case of IIIa (1% trans, 99% cis), to a mixture of cis and trans for IIIb and IIIc (ca. 60% cis, 40% trans in IIIb, ca. 50% trans, 50% cis in IIIc) to predominantly the trans isomer in the case of IIIc (ca. 90% trans and 10% cis). Unlike IIIa, the 31 P NMR spectra of IIIb and IIIc show no changes in the ratios of the cis : trans isomers with increased temperature. As in the case of IIIa, the P–H region of the 1H NMR spectra of IIIb–IIId show similar second-order patterns for the P–H protons of their cis and trans isomers (see ESI†). The assignment of the 1H NMR spectrum of IIIb is particularly informative and was supported by a 2-D 31P/1H HSQC NMR experiment. For the cis diastereomer the two methyl groups of each iPr group are enantiotopic, whereas diastrereotopic methyl groups are found for the trans diastereomer (Fig. 5, top). Consequently, the room-temperature 1H NMR spectrum displays one septet resonance for the C–H protons of the iPr groups and two 1 : 1 Me resonances for the Me-groups of the i Pr substituents of the trans isomer. This is consistent with the C2 symmetric structure of the trans-IIIb revealed by singlecrystal X-ray analysis (Fig. 6), and results from restricted rotation of the Caryl–N and Caryl–C(iPr) bonds. In contrast, cisIIIb shows two C–H resonances for the iPr groups and two 1 : 1 resonances for the Me groups of the iPr substituents (Fig. 5,
This journal is © The Royal Society of Chemistry 2015
Paper
Fig. 5 Idealized C2h structure for trans-IIIb and C2v structure for cis-IIIb (top) and room-temperature 1H NOESY experiment showing throughspace correlation of Hb and Ha with the P–H protons (bottom).
Fig. 6 X-ray crystal structures of the trans-isomers of IIIb (top) and IIIc (bottom). H-atoms (except those attached to P) have been omitted for clarity. The molecules lies on a crystallographic inversion centre. A minor disorder component for the central core of IIIc is not shown; full details are given in the ESI.† Selected bond lengths (Å) and angles (°); IIIb: P(1)–N(1) 1.679(2), P(1)–N(1A) 1.676(2), P(1)⋯P(1A) 2.5055(16), P(1)– S(1) 1.8875(12), P(1)–H(1) 1.35(3), N–P–N 83.38(12), P–N–P 96.62(12), N–PvS mean 122.17; IIIc: P(1)–N(1) 1.680(3), P(1)–N(1A) 1.691(3), P(1)⋯P(1A) 2.5322(17), P(1)–S(1) 1.9084(13), N–P–N 82.63(13), P–N–P mean 97.38(13), N–PvS mean 121.2.
Dalton Trans.
View Article Online
Published on 07 July 2015. Downloaded by California State University at Fresno on 21/07/2015 04:34:14.
Paper
Dalton Transactions
top right). The 1H–1H NOESY NMR spectrum (Fig. 5, bottom) provides further support for the rigid structures of both isomers in solution by confirming that only one of the C–H (iPr) environments in the cis isomer is in close proximity to the P–H protons, whereas – for symmetry reasons – all of the C–H (iPr) protons of the trans isomer are close to the P–H protons. Similar restricted rotation was seen for IIIc where two separate broad signals are seen for the o-Me in the cis isomer whereas only one sharp signal is observed for the trans. Evidence for the less restricted nature of the Caryl–N bonds within the dimer IIIc was found by NOESY and EXSY 1H NMR as well as a variable-temperature NMR experiment. The singlecrystal X-ray structure of trans-IIIc is very similar to that of trans-IIIb (Fig. 6, bottom). Computational modelling We have performed DFT modelling of the cis and trans isomers of IIIa–d using the B3LYP/cc-pVTZ level of theory in THF solvent (PCM model).9–11 Frequency calculations of the optimized geometries were carried out, confirming the absence of imaginary frequencies. Thermodynamic parameters of all isomers are provided for 298 K and standard pressure (Table 1). The calculated thermodynamic data support the experimental observation from 31P NMR spectroscopy that the cis isomer is the most stable for bulkier R-groups (tBu, Dipp) but the trans isomer becomes most stable for less sterically encumbered R-groups (Mes, Ph2CH). Notably, depending on the steric demands of the substituents, the calculated cis/trans ratios range from strongly favoured cis (ΔG = 9.0 kJ/mol) to strongly favoured trans isomer (ΔG = −11.6 kJ mol−1). In contrast, the tautomers containing an SH group are all much higher in energy, regardless of the substituents attached to the P2N2-ring. It is clear, however, that this steric effect is a subtle one since the trend in the relative stabilities of the cis vs. trans isomers does not follow that expected on the basis of Tolman cone angle completely (for which the preference for the cis isomer should be IIIb > IIIc > IIIa > IIId). Although we cannot discount electronic factors as an influence on the preference for either isomer, we believe that the lower than expected steric influence of the Dipp and Mes groups of IIIb and IIIc is due to their perpendicular orientation with respect to the P2N2 rings. We next looked at possible mechanisms for the conversion of the cis isomer of IIIa to the trans isomer, in order to understand the apparent irreversibility of this isomerism. We found that typical intramolecular isomerization mechanisms such as
Table 1 Reaction energies (E, H, G, in kJ mol−1) for the conversion cis → trans for compounds IIIa–d. Thermodynamic parameters H and G were calculated for 298 K and standard pressure
Compound
ΔE
ΔH
ΔG
IIIa IIIb IIIc IIId
8.52 −0.05 −0.58 −10.31
8.42 −0.88 −1.03 −4.62
9.04 1.00 −4.22 −11.60
Dalton Trans.
(i) a concerted pathway in which cis-IIIa is converted into transIIIa directly via rotation of one P(H)vS fragment and (ii) a stepwise pathway involving P(H)vS/P–SH tautomerism and phosphorus lone pair inversion involve very high activation barriers (>150 kJ mol−1) and are therefore unlikely. In view of our findings in variable-temperature NMR experiments, it is probable that the relative cis/trans ratios of the different isomers are in part influenced by kinetic factors. Further mechanistic studies are required to resolve the nature of the transition states involved (intra- vs. intermolecular).
Conclusion The current work represents probably the most in-depth study of the cis and trans isomerism in any family of dimeric phosphazanes(III or V). We have shown that a series of dimers of the type [Sv(H)P(μ-NR)]2 can be readily obtained by the simple reactions of the chloro-dimers [ClP(μ-NR)]2 with LiSH, providing potential building blocks for the synthesis of S-bridged phosphazane macrocycles of the type [{P(μ-NR)}2(μ-S)]n. The relative thermodynamic stabilities of the cis and trans isomers of [Sv(H)P(μ-NR)]2 depend on the steric bulk of the R-group, so that more sterically encumbered groups result in a preference for the cis isomer while less sterically bulky groups favour the trans. This factor, and the apparently high activation energy between the isomers, is important in the condensation of [Sv(H)P(μ-NR)]2 into S-bridged macrocyclic phosphazanes since the cis-isomer is pre-organized for cyclisation while the trans is not.4 We are currently pursuing the metal-templated and non-templated condensation reactions of [Sv(H)P(μ-NR)]2 with [ClP(μ-NR)]2 to obtain macrocycles of the type [{P(μ-NR)}2(μ-S)]n, as well as coordination and deprotonation studies of the dimers themselves with a range of metals and metal bases ( potentially leading to metal-bridged macrocycles).
Experimental Synthesis All manipulations were carried out under dry, O2-free nitrogen on a vacuum-line, using standard inert-atmosphere techniques for isolation and characterisation. Chlorodimers were synthesized according to previously reported procedures.12 Synthesis of IIIa. nBuLi (6.2 mL, 10 mmol, in hexanes) was added to THF (40 ml) and the solution cooled to −78 °C. H2S gas was bubbled slowly though this solution for ca. 20 min. The reaction was allowed to warm to ambient temperature and the THF removed completely under vacuum, to give white powder of LiSH (this step ensures excess H2S is removed from the reaction). The solid was redissolved in THF (40 mL) and cooled to −78 °C. A solution of [ClP(μ-NtBu)]2 (1.375 g, 5 mmol) in THF (20 mL) was added slowly dropwise and the resulting mixture allowed to warm to ambient temperature and stirred for 72 h. The solvent was removed and the resulting solid was suspended in toluene (80 mL) and filtered. The
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 07 July 2015. Downloaded by California State University at Fresno on 21/07/2015 04:34:14.
Dalton Transactions
solvent was completely removed from the filtrate under vacuum and the semi-solid formed was washed with n-pentane (2 × 20 mL) and dried under vacuum. Yield of powder 1.15 g (83%). 1H NMR (500.2 MHz, CDCl3, +25 °C), cis isomer, δ /ppm = 8.40 (2H, P–H, AA′XX′ mult. apparent Jmax = 598 Hz), 1.38 (s, 18H, tBu); trans isomer, δ /ppm = 8.75 (P–H, AA′XX′ mult., cis isomer), 1.36 (s, tBu), (ratio of trans : cis ≈ 1 : 50 from P–H integration). 31P NMR (161.7 MHz, CDCl3, rel. 85% H3PO4/D2O. +25 °C), δ /ppm = 36.3 (AA′XX′ mult., cis isomer), 40.7 (AA′XX′ mult., trans isomer). IR (NaCl windows, Nujol mull), ν/cm−1 = 2360 (m) (P–H str.), 762 (s) (PvS str.). Synthesis of IIIb, IIIc, and IIId. The synthetic route employed was exactly the same as for IIIa but immediately after the addition of the dimer the reactions were allowed to warm to ambient temperature. At this point, the solvent was removed under vacuum and the solid extracted in the same manner as for IIIa. Yields of powders produced (200 mg scale for all): IIIb (200 mg of [ClP(μ-NDipp)]2, 30 mg of IIIb produced after further crystallisation from n-pentane, 15%), IIIc (200 mg of [ClP(μ-NMes)]2, 0.144 g of IIIc produced, 73%), IIId (200 mg of [ClP(μ-NCHPh2)]2, 0.109 g of IIId produced, 55%). Crystals of trans-IIIb were grown from pentane at −16 °C and crystals of trans-IIIc were grown from hexane/THF at room temperature. Unit cell analyses of crystalline samples of both indicate that only the trans isomers crystallise under these conditions. NMR: IIIb. 1H NMR (500.2 MHz, Tol d8, +25 °C), 1H NMR (500.2 MHz, CDCl3, +25 °C), cis isomer 9.04 (mult., Jmax = 593 Hz), 4.30 (sept., CHb Dipp), 3.00 (sept., CHa Dipp), 1.40 (d, Me Dipp), 0.86 (d, Me Dipp); trans isomer, 9.32 (mult., Jmax = 591 Hz), 3.78 (sept., CH Dipp), 1.10 (d, Me Dipp), 0.98 (d, Me Dipp). 31P NMR (161.7 MHz, Tol d8, rel. 85% H3PO4/D2O. +25 °C), δ /ppm = 53.1 (second-order mult., trans isomer, apparent 1JPH = 594 Hz), 45.05 (second-order mult., cis isomer, apparent 1JPH = 591 Hz) (ratio of trans : cis ≈ 1 : 1). IIIc: 1H NMR (500.2 MHz, Tol d8, +25 °C), cis isomer, 9.07 (2H, mult., P–H cis, Jmax = 592 Hz), 6.72 (2H, s., C–Haryl Mes cis, 6.61 (2H, s, C–Haryl Mes cis) 3.03 (s., 6H, o-Me Mes cis), 2.08 (s., 12H, p-Me Mes cis and trans), 1.95 (s., 6H, o-Me Mes cis); trans isomer, 9.29 (2H, mult., P–H trans, Jmax = 593 Hz), 6.65 (4H, s, C–Haryl Mes), 2.47 (12H, s, o-Me Mes trans). 31P NMR (161.7 MHz, Tol d8, rel. 85% H3PO4/D2O. +25 °C), δ /ppm = 44.6 (second-order mult., trans isomer, apparent 1JPH = 594 Hz; 31P{1H} d., 2JP–P 2.0 Hz), 36.7 (second-order mult., cis isomer, apparent 1JPH = 593 Hz) (ratio of trans : cis ≈ 1 : 1). IIId: 1H NMR (500.2 MHz, CDCl3, +25 °C), δ /ppm = 7.44 (2H, P–H, AA′XX′ mult. apparent Jmax = 550 Hz), 6.9–7.7 (overlapping mult. Ph of Ph2CH), 5.62 (dd., Ph2CH). 31P NMR (161.7 MHz, CDCl3, rel. 85% H3PO4/ D2O. +25 °C), δ = 40.0 (second-order mult., trans isomer, apparent 1JPH = 556 Hz), 38.2 (br. d., cis isomer, apparent 1 JPH = 579 Hz) (ratio of trans : cis ≈ 9 : 1). Satisfactory C, H and N analyses of IIIa–IIId could not be obtained. In particular, the N contents of each are found to be consistently and repeatedly low. This is a common problem in the analysis of phospha(III)zanes and is most likely to be due to incomplete combustion.4
This journal is © The Royal Society of Chemistry 2015
Paper
X-ray data ˉ, Z = 2, cis-IIIa, C8H20N2S2P2, M = 270.32, triclinic, space group P1 a = 5.9112(2), b = 9.9768(3), c = 12.1132(3) Å, α = 85.594(2), β = 88.593(2), γ = 79.0554(10)°, V = 699.27(4) Å3, µ(Mo-Kα) = 0.580 mm−1, ρcalc = 1.284 Mg m−3, T = 180(2) K. Total reflections 6232, unique 2952 (Rint = 0.026). R1 = 0.029 [I > 2σ(I)] and wR2 = 0.078 (all data). Trans-IIIb, C24H36N2P2S2, M = 478.61, monoclinic, space group P21/n, Z = 2, a = 8.3695(19), b = 10.1458(17), c = 15.541(3) Å, β = 101.646(7)°, V = 1292.5(4) Å3, µ(Mo-Kα) = 0.344 mm−1, ρcalc = 1.230 Mg m−3, T = 180(2) K. Total reflections 11 418, unique 1743 (Rint = 0.090). R1 = 0.039 [I > 2σ(I)] and wR2 = 0.087 (all data). Trans-IIIc, C18H24N2P2S2, M = 394.45, monoclinic, space group P21/c, Z = 2, a = 10.4215(3), b = 11.0475(3), c = 9.3492(3) Å, β = 109.462(2)°, V = 1014.89(5) Å3, µ(Cu-Kα) = 3.878 mm−1, ρcalc = 1.291 Mg m−3, T = 180(2) K. Total reflections 13 746, unique 1755 (Rint = 0.049). R1 = 0.046 [I > 2σ(I)] and wR2 = 0.110 (all data). Data for cis-IIIa were collected on a Nonius KappaCCD instrument, for trans-IIIb on a Bruker SMART X2S instrument and for trans-IIIc on a Bruker D8 QUEST diffractometer. All data were refined by full-matrix least squares on F2 using SHELXL-2014.13 The structure of trans-IIIc contains a minor disorder component for the central core, as described in the ESI.† CCDC 1062561–1062563 contain the supplementary crystallographic data. Computational modelling All calculations were performed with Gaussian 09 software suite of programs. Optimized Geometries of all the isomers were calculated in THF (PCM Model, cc-pVTZ basis set, B3LYP functional) and relative ΔG and enthalpies ΔH were determined at 298 K and standard pressure. In addition, relative electronic energies E were determined at 0 K and vacuum. Frequency calculations were performed for all stationary points, with no imaginary frequencies for minima and a single imaginary frequency for first-order saddle points (transition states). NMR calculations were carried out for the tBu dimer using the standard GIAO method as implemented in Gaussian 09. Here, optimized gas-phase geometries (B3LYP/cc-pVTZ) of all compounds were used. In the NMR calculation, the B3LYP functional was employed. The P atoms were described by the cc-pV5Z basis set, for all other atoms the cc-pVTZ basis set was used. The calculated chemical shifts were referenced vs. the experimental data of the cis (H)PvS isomer. For compounds where the two P atoms should be practically equivalent, an average chemical shift was calculated. Details of all of the models calculated are provided in the ESI.†
Acknowledgements We thank the ERC (Advanced grant for D.S.W., studentship for C.G.M.B.), Deutschlandstipendium (National Scholarship, V.V.) and the EU (Marie Curie Intra European Fellowship for R.G.-R.) for financial support. We also thank Dr J. E. Davies for collecting X-ray data for IIIa.
Dalton Trans.
View Article Online
Paper
Dalton Transactions
Published on 07 July 2015. Downloaded by California State University at Fresno on 21/07/2015 04:34:14.
Notes and references 1 (a) E. L. Doyle, L. Riera and D. S. Wright, Eur. J. Inorg. Chem., 2003, 3279; (b) M. S. Balakrishna, D. J. Eisler and T. Chivers, Chem. Soc. Rev., 2007, 36, 650; (c) J. S. Bradshaw, R. M. Izatt, A. V. Bardunov, C. Y. Zhu and J. K. Hathaway, in Comprehensive Supramolecular Chemistry, ed. J. L. Atwood, J. E. D. Davies, D. D. McNicol and F. Vögtle, Pergamon, Oxford, 1996, vol. 1, ch. 2; A. Pochini and R. Ungaro, in Comprehensive Supramolecular Chemistry, ed. F. Vögtle, Pergamon, Oxford, 1996, vol. 2, ch. 4; K. Odashima and K. Koga, in Comprehensive Supramolecular Chemistry, ed. F. Vögtle, Pergamon, Oxford, 1996, vol. 2, ch. 5. 2 D. S. Wright, Host–Guest Chemistry; p-Block Systems, in Comprehensive Inorganic Chemistry, Main Group Elements, Including Nobel Gases, ed. T. Chivers, 2013, vol. 1, ch. 1.30, p. 953. 3 S. Gonzalez Calera and D. S. Wright, Dalton Trans., 2010, 39, 5055. 4 (a) A. Bashall, E. L. Doyle, C. Tubb, S. J. Kidd, M. McPartlin, A. D. Woods and D. S. Wright, Chem. Commun., 2001, 2542; (b) A. Bashall, A. D. Bond, E. L. Doyle, F. Garcıa, S. J. Kidd, G. T. Lawson, M. C. Parry, M. McPartlin, A. D. Wood and D. S. Wright, Chem. – Eur. J., 2002, 8, 3377; (c) F. García, J. M. Goodman, R. A. Kowenicki, I. Kuzu, M. McPartlin, M. A. Silva, L. Riera, A. D. Woods and D. S. Wright, Chem. – Eur. J., 2004, 10, 6066. 5 (a) W. T. K. Chan, F. García, S. Gonzalez-Calera, M. McPartlin, J. V. Morey, R. E. Mulvey, S. Singh and
Dalton Trans.
6
7
8
9 10
11
12
13
D. S. Wright, Chem. Commun., 2008, 2251; (b) S. Gonzalez Calera, D. J. Eisler, J. M. Goodman, M. McPartlin, S. Singh and D. S. Wright, Dalton Trans., 2009, 1293. S. Gonzalez Calera, D. J. Eisler, J. V. Morey, M. McPartlin, S. Singh and D. S. Wright, Angew. Chem., Int. Ed., 2008, 47, 1111. A. Nordheider, T. Chivers, R. Thirumoorthi, I. Vargas-Baca and J. D. Woollins, Chem. Commun., 2012, 48, 6346. A. Tarassoli, M. L. Thomson, R. C. Haltiwanger, T. G. Hill and A. D. Norman, Inorg. Chem., 1988, 27, 3382 and references therein. M. J. Frisch, et al., Gaussian 09, Revision B.01, Gaussian, Inc., Wallingford CT, 2010. (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652; (b) C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter, 1988, 37, 785; (c) S. H. Vosko, L. Wilk and M. Nusair, Can. J. Phys., 1980, 58, 1200–1211; (d) P. J. Stephens, F. J. Devlin, C. F. Chabalowski and M. J. Frisch, J. Phys. Chem., 1994, 98, 11623. (a) T. H. Dunning Jr., J. Chem. Phys., 1989, 90, 1007; (b) D. E. Woon and T. H. Dunning Jr., J. Chem. Phys., 1993, 98, 1358; E. R. Davidson, Chem. Phys. Lett., 1996, 260, 514– 518. (a) T. Roth, H. Wadepohl, D. S. Wright and L. H. Gade, Chem. – Eur. J., 2013, 19, 13823; (b) T. Roth, V. Vasilenko, C. G. M. Benson, H. Wadepohl, D. S. Wright and L. H. Gade, Chem. Sci., 2015, 6, 2506. G. M. Sheldrick, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 2015, 71, 3.
This journal is © The Royal Society of Chemistry 2015
