Preparation and DFT Studies of  2 C,N Hypercoordinated Oxazoline Organotins: Monomer Constructs for Stable Polystannanes

: Tetraorganotin tin(IV) compounds containing a flexible or rigid ( 4 : Ph 3 Sn-CH 2 -C 6 H 4 -R; 7 : Ph 3 SnC 6 H 4 -R, R = 2-oxazolinyl) chelating oxazoline functionality were prepared in good yields by the reaction of lithiated oxazolines and Ph 3 SnCl. Reaction of 7 with excess HCl resulted in the isolation of the tin monochlorido compound, 9 (ClSn[Ph 2 ]C 6 H 4 -R). Conversion of the triphenylstannanes 7 and 4 into their corresponding dibromido species was successfully achieved from the reaction with Br 2 to yield 10 (Br 2 Sn[Ph]C 6 H 4 -R) and 11 (Br 2 Sn[Ph]-CH 2 -C 6 H 4 -R), respectively. X-ray crystallography of 4 , 7 , 9 , 10 , and 11 reveal that all structures adopt a distorted trigonal bipyramidal geometry around Sn in the solid state. Compound 4 , with an additional methylene spacer group, displays a comparatively long Sn–N bond distance compared to the dibromido tin species, 11 . Several DFT methods were compared for accuracy in predicting the solid-state geometries of compounds 4 , 7 , 9 – 11 . Compounds 10 and 11 were further converted into the corresponding dihydrides ( 12 : H 2 Sn[Ph]C 6 H 4 -R, 13 : H 2 Sn[Ph]-CH 2 -C 6 H 4 -R), via Br–H exchange, in high yield by reaction with NaBH 4 . Polymerization of 12 or 13 with a late transition metal catalyst produced a low molecular weight polystannane ( 14 : –[Sn[Ph]C 6 H 4 -R] n –, M w = 10,100 Da) and oligostannane ( 15 : –[Sn[Ph]-CH 2 -C 6 H 4 -R] n –, M w = 3200 Da), respectively. white Recrystallization 11 achieved of and layering followed yielding clear colourless 2.05 five tin


Introduction
Polystannanes are main group polymers consisting of chains of covalently bonded tin atoms. These materials possess unique chemical, optical, and electronic properties attributed to the delocalization of the formal Sn 5s electrons into σ-σ bonding orbitals along the backbone [1]. The degree of such delocalization in polystannanes is greater than that noted for comparable Group 14 polymers, (e.g., polysilanes and polygermanes). The Sn analogues display smaller band gaps, greater metallic character, and a visible red shift observed in the UV-Vis spectra [2][3][4]. A conductivity study by Caseri et al. [5] on undoped poly(di(3-propylphenyl)stannane) found an intrinsic conductivity of 3 × 10 −8 Scm −1 at 300 K. When Tilley et al. [6] cast films of poly(di-n-butylstannane) and poly(di-noctylstannane) doped with SbF5, conductivities of six to seven orders of magnitude greater (1 × 10 −2 and 3 × 10 −1 Scm −1 ) were reported. Unfortunately, both doped and undoped polymers were found to degrade rapidly when exposed to light and/or moisture.
A recent approach to address the sensitivity of polystannanes is to increase the coordination number at the tin center via a pendant functional ligand [7,8]. The benefit of the resulting hypercoordination is two-fold: first, the addition of a Lewis basic ligand provides additional electron density to the Lewis acidic Sn center and second, the occupation of an additional coordination site hinders nucleophilic attack by creating more congested tin units [9]. The nature of the dative interaction between the additional ligand and the Sn center has been previously rationalized in terms of a three center-four electron (3c-4e − ) bond, a concept used to describe hypervalent compounds that exceed the standard octet of valence electrons in Group 14 systems [10][11][12]. In hypercoordinated Sn (IV) species, 5-coordinate trigonal bipyramidal geometries are preferred and the 3c-4e − interaction is observed in the assigned axial position (Figure 1). A lone pair of electrons from Y is donated into the non-bonding molecular orbitals of tin and is prominent when X is an electronegative atom such as a halide. An early crystallographically characterized example was reported by van Koten et al. [13]. This complex featured an aryl-Sn bond in which the aromatic group contained an orthodimethylaminomethyl functionality. A  2 -C,N-bonding motif involving a 5-coordinate Sn atom was observed ( Figure 1: R = Ph, Y = NMe2, X = Br). Several examples of other Group 14 compounds displaying similar hypercoordination have been disclosed since that time [14][15][16].
Over the last several years, we have reported on the syntheses of asymmetric polystannanes incorporating flexible groups capable of hypercoordination, specifically propyl alkoxy chains containing a hydroxy, phenoxy, biphenyl oxy or oxy azo-benzene moieties [7,8]. These hypercoordinated polystannanes feature a notable increase in stability (>2 months) to both light and moisture compared to their 4-coordinate Sn analogues and display, in most instances, two unique 119 Sn NMR resonances. This is presumably due to the presence of both 4-and 5-coordinate tin centers attributed to the flexible side chain containing a donor atom that can move freely to, and away from, tin centers. Polystannanes bearing more rigid groups do not share the same level of flexibility, resulting from a presumed stronger dative interaction and greater steric bulk around the tin units. Recently, we disclosed details on the preparation of hypercoordinate polystannanes with either a benzyl  2 -C,N (dimethylamino) or a  2 -C,O (2-methoxymethyl) chelating group [17]. These materials are derived by dehydropolymerization of the respective -C,N and -C,O dihydride monomers using a late transition metal catalyst. This leads to the formation of modest molecular weight, amorphous polymers which are stable to both ambient light and moisture.
Another potential class of donor groups that may also be suitable for hypercoordination to Sn are oxazolines (Figure 2: 1a-b). This ring system, also known as 4,5-dihydro-2-oxazoles, are a subclass of azole heterocycles containing an imine nitrogen and an oxygen atom bound by an sp 2 hybridized carbon in a five-membered ring [18]. There have been a limited number of small molecule tin species bearing oxazoline groups reported in the literature [19][20][21][22][23][24][25]. Specifically, 2-phenyl-2-oxazolines are particularly interesting as they allow for a potential five-membered chelate ring to be formed when the tin center is bound at the ortho position of the aromatic group. The first complex of this nature was also reported by van Koten et al. for [2-(4,4-dimethyl-2-oxazoline)-5-methylphenyl]methylphenyltin bromide (Figure 2: 1c) [26]. X-ray crystallography reveals the nitrogen preferentially datively bonding to the tin center over that of the oxygen atom forming a distorted trigonal bipyramidal geometry around Sn with a relatively short Sn-N bond length (2.414 Å) [23]. The Staliński group later published a series of papers detailing organotin complexes bearing similar chiral oxazoline groups (Figure 2: 2a-g, 3a-e) [27][28][29][30]. Evidence of intramolecular Sn-N interactions were probed effectively by 1 H, 13 C, 15 N and 117 Sn NMR spectroscopy. Synthesis of these hypercoordinated compounds was achieved via bromine-lithium exchange from a 2-(2′bromophenyl)-2-oxazoline ring (n-BuLi) followed by reaction with a tin halide species. This work further evaluated the dative interaction structurally by comparing the series of ortho-(2a-e) substituted tin complexes to para-(3a-e) substituted tin analogues where dative interactions are presumably absent. Comparisons of the 117 Sn NMR chemical shifts of the series were thereafter given. This investigation revealed that the 117 Sn NMR chemical resonances of complexes 2a-e are shifted significantly upfield relative to the corresponding structural isomers 3a-e. Further evidence of the dative interaction comes from solution 15 N NMR spectroscopy of 2a-c where a single nitrogen resonance is detected accompanied by 117/119 Sn satellites. As expected, such tin satellite resonances are absent in 3a-3c. Single crystal X-ray analysis of 2a reveals a Sn-N bond distance of 2.89(9) Å, while for 2c, which has a bromido ligand trans to the Sn-N bond, possesses a significantly shorter Sn-N bond distance (2.39(2) Å) [27]. These bonds lengths fall well within the sum of the van der Waals radii (~3.72 Å). The shorter bond distance in 2c is likely a result of the electron withdrawing effects of the bromine on the tin center. One example of a methylene bridged tin oxazoline complex, 4, (Figure 2) was prepared by Parish and Bonnardel [31]. While no Sn NMR data was reported, the 1 H NMR spectrum displays a 2 J 119/117 Sn-H coupling constant of 75 Hz for the benzylic protons.
The current work focuses on the suitability of both more rigid and flexible oxazoline tin species as suitable monomers for preparation of light and moisture stable oxazoline containing polystannanes. This rigidity is modified by the incorporation of a methylene spacer group as shown in Figure 3 (cf. 4).

Scheme 1. Synthetic approaches to 7.
A potentially more flexible oxazoline tin species with a methylene spacer between tin and the aryl group was also pursued. Gschwend et al. [32] had previously demonstrated that the tolyl ring in 4,4-dimethyl-2-(o-tolyl)-4,5-dihydrooxazole (8) can be lithiated with n-BuLi (Scheme 2) selectively at the methyl group. This lithiated species was shown to be reactive towards various electrophiles; however, no examples with tin were reported. When the ruby red coloured solution of lithiated oxazoline 8 was reacted with Ph3SnCl, the mixture turned an opaque yellow colour. Following removal of the ethereal solvent, the crude mixture was redissolved in toluene and gravity filtered to remove insoluble materials. After drying in vacuo and trituration of the crude powder with CH3OH, white coloured crystals of 4 were recovered. The 1 H NMR (CDCl3) spectrum of 4 reveals a signal corresponding to the benzylic protons located at δ = 3.57 ppm along with 117/119 Sn satellites ( 2 J1H-117/119Sn = 75.5 Hz) in good agreement with the values reported by Parish [28].
X-ray diffraction studies of both 7 and 4 were undertaken and selected bond lengths and bond angles shown in Table 1. The geometry of the Sn center of 7 is distorted trigonal bipyramidal (τ5 = 0.92) [33] in nature with the equatorial bond angles around the tin center ranging from 102° to 117° and an axial bond angle of 172.20(6)°. The Sn1-C12 bond of 7 is slightly longer than the other Sn-C aryl bonds in this compound and is likely a result of it being located trans to the strongly donating nitrogen atom (Figure 4: top). Hypercoordinated tetraorganotins, such as 7, possessing a Sn→N interaction are relatively rare. Jurkschat and co-workers reported a Sn-N bond distance of 2.624(8) Å for the sterically confined azastannatrane, MeSn(CH2CH2CH2)3N [34]. The Sn-N bond distance of 7 is 2.762(1) Å and this is substantially longer than a typical covalent Sn-N bond (~2.15 Å) [35], but well within the sum of their van der Waals radii (3.72 Å).   (15) Two unique molecules of compound 4 (designated as 4′, 4″: Figure 4 bottom) were found within the unit cell, both with tin centers adopting a distorted trigonal bipyramidal geometry (τ5 = 0.80; 0.76, respectively). The equatorial angles around the tin center for both structures are between 103.6° and 116.1° and axial angles of 163.2(1) and 162.0(1)°. The Sn-N (4′: 3.176(4) Å, 4″: 3.234(4)) distances are significantly longer than in 7, a likely consequence of the additional -CH2-bridge. A comparison of the solid-state structures of 7, 4′ and 4″ reveal similar Sn-C bond lengths, with the exception of the Sn-Ph bond trans to the oxazoline of 7, which is elongated compared to the Sn-Ph bond trans to the oxazoline ligand of 4′ and 4″.

Halogenated Oxazoline Stannanes
Halogenation reactions of both 7 and 4, leading to dihalido reactive species were investigated. In a method first described by Pannell et al. [36], 7 was treated with one molar equivalent of a 1.0 M solution of HCl/Et2O for 1 h at RT (Scheme 3). After recrystallization from hexanes, the monochloridostannane 9 was recovered as white coloured powder in an 84% yield. 119 Sn NMR (CDCl3) analysis of 9 revealed a single tin resonance (δ = −226.0 ppm), as expected. A successive chlorination attempt to replace a second aryl substituent under the same conditions was unsuccessful (NMR: 1 H and 119 Sn). Addition of 3-or 4-fold excess of HCl also failed to convert 9 to the desired species. The use of a stronger chlorinating source, SOCl2, was also unsuccessful in conversion to the desired dichlorido species. This methodology was therefore abandoned.
Brominations of 7 and 4 (Scheme 4) were investigated as a possible alternative route to dihalide species. When 7 was reacted with two equivalents of Br2 (Scheme 4), 10 was obtained in good yield (79%). After solvent removal, a yellow coloured powder remained and 119 Sn NMR (CDCl3) analysis revealed a single resonance ( = −290.6 ppm), which corresponds closely to the shift reported for 2g ( Figure 2, δ117Sn = −288.5 ppm) [30]. In tin species with potentially hypercoordinating 2-methoxybenzyl or dimethylaminobenzyl groups, replacement of an alkyl or aryl group for an electronegative atom imparts a downfield shift of the 119 Sn NMR resonance [9]. However, in the case of 7, 9, and 10, the opposite effect is observed ( Figure 5) and upfield 119 Sn chemical shifts relative to their four coordinate tin analogs is observed. In the 1 H NMR spectra of compounds 7, 9, and 10, the methylene protons of the oxazoline display downfield shifts with each additional halide bound to Sn, possibly a result of inductive effects. Addition of two equivalents of Br2(l) to 4 in C6H6 was carried out under the same reaction conditions as detailed for 7. Analysis by NMR revealed the appearance of three tin resonances assigned to the starting material, 4 (the major signal), with two other unassigned resonances at −215 and −248 ppm. When the bromination of 4 was performed in the alternative solvent dry DCM (Scheme 4 bottom), a single 119 Sn tin resonance (CDCl3) was detected ( = −248.2 ppm). After removal of solvent, the crude powder was washed with MeOH, yielding 11 cleanly as a white coloured powder.
Crystals of 9, 10, and 11 suitable for single crystal X-ray diffraction analysis were obtained and this investigation revealed ( Figure 6) the bond lengths and bond angles shown in Table 2. Two unique molecules of 9 were observed (9′ and 9") within the unit cell of the crystal that differ slightly in terms of bond lengths and angles (vide infra).   (9) The geometry around the Sn center in both molecules of 9 is distorted trigonal bipyramidal in nature (9′: τ5 = 0.81, 9″: τ5 = 0.82) [35]. For both isomers, the angles of N1-Sn-Cl are almost 180°, while the angles between the equatorial atoms range from 117 to 121°. The angles between the axial and the equatorial phenyl groups range between 90 and 95°. The hypothesised 3c-4e − sharing between the nitrogen, tin and chlorine atoms is supported by an elongated Sn-Cl bond (2.49 Å) relative to a typical Sn-Cl bond length (2.41 Å) [33]. The shorter Sn-N bond length found in 9 (2.458 Å) relative to the triphenyltin analogue 7 (2.762 Å) may also be due to an enhanced 3c-4e − interaction. Staliński et al. reported on the crystal structure of a similar monobromido dimethyltin oxazoline ( Figure 2: 2i) complex that possesses a Sn-N bond distance of 2.39(2) Å [27], significantly shorter than that found in 9.
The geometry around the Sn centers of 10 (τ5 = 0.74) and 11 (τ5 = 0.89) are also distorted trigonal bipyramidal as described previously. Possible evidence for a 3c-4e − sharing is given by the relatively short Sn1-N1 bond lengths and the elongated apical Sn1-Br2 (10) interaction and the corresponding Sn1-Br1 (11) bond distances ( Table 2), compared to average covalent Sn-Br bond lengths (2.56 Å) [34]. The electron withdrawing effect of the apical bromine atom in 10 and 11 appears to draw the nitrogen closer to the tin center, resulting in a short Sn-N bond length of 2.383(3) Å for 10, and a slightly longer bond length in 11. For comparison, Novák et al. [37] and Švec and co-workers [38] independently reported related hypercoordinate phenyldichlorido and phenyldiiodidotin structures possessing a chelating benzyl amine substituent ( Figure 1). These compounds display Sn-N bond distances of 2.444(5) Å and 2.476(3) Å, which are 0.06 and 0.04 Å longer than that of 10 and 11, respectively.

DFT Studies
The seven crystallographically determined structures in this study (compounds 4 (4′,4″), 7, 9 (9′,9″), 10, 11) provide a small structurally-related dataset to compare with DFT models. We have previously summarized the problematic aspects of comparing calculated and experimental structural data [17]; nevertheless, it remains a useful approach for refining a DFT description of these hypercoordinate systems. Previous work has demonstrated that the LANL08d ECP for Sn and Br, and 6-31+G(d,p) for all other atoms provide accurate structures at moderate computational cost [39,40]. In a recent evaluation of selected density functionals, we found that B3PW91 [41] and PBE0 "PBE1PBE" [42] when supplemented with Grimme's D3 empirical dispersion function and Becke-Johnson damping (GD3BJ) [43] outperformed eight other methods in predicting hypercoordinate Sn geometries [17]. The use of diffuse functions on non-H atoms significantly improved accuracy as did augmenting methods with empirical dispersion. This latter finding was also reported in a recent comprehensive study [44], which included among its recommendations that M052X-GD3 [45,46] was effective for modelling general main-group thermochemistry, kinetics, and non-covalent interactions.
Experimental and calculated structures were compared by aligning them to maximize overlap. The central Sn atoms for the two structures were superimposed and then one structure was rotated relative to the other to minimize the mean sum of squared distances (MSSDs) between equivalent heavy (non-H) atoms on the two structures. Since the position of the Sn atoms is fixed, Sn is excluded from the average. The MSSD is a demanding metric for assessing accuracy of a calculated structure. Alignment of the experimental and calculated structures causes differences in bond lengths and angles at the Sn center to be propagated throughout the structure, increasing the distances between equivalent atoms at multiple locations. Small deviations are thus amplified, leading to larger mean values and greater discrimination between models for calculated structures. At the same time, the MSSD is acutely sensitive to small displacements in dihedral angles for phenyl ligands or any solidstate structural distortions due to packing forces or intermolecular associations between Sn atoms. To minimize the impact of phenyl ligand dihedral differences, only the Ph-C atoms directly bonded to Sn were included in the MSSD calculations. Table 3 shows the MSSD values determined for structures calculated by the three DFT methods for the seven compounds in this study. The largest MSSDs are found for 4 and 11, both compounds that contain an additional -CH2-bridge resulting in a twisted six-member hypercoordination ring, rather than the nearly planar five-member ring of the other compounds. The additional flexibility in these structures may allow for greater distortions in the solid state. Compound 11 is unique in the series in having a planar oxazoline ring in the solid state. All other structurally characterized compounds, and all calculated structures display a slight pucker in the oxazoline groups. Energy differences between planar and puckered oxazoline ring conformations are expected to be small enough to allow for stabilization of the higher energy conformer by packing forces. In the case of 10, observed crystallographic disorder is consistent with the presence of two oxazoline ring twist isomers that are resolvable due to the Sn hypercoordination. The MSSDs observed for 7, 9, and 10 are comparable with those measured for Sn complexes with phenyl ligands in the previous study [17]. In all cases, there is evidence of some distortion in the solid-state structures, typically seen in the bending of a phenyl ligand at the coordinating C atom, such that the Sn-C(Ph) bond no longer lies in the plane of the phenyl ring. The three DFT methods perform very similarly by the MSSD measure for this series of compounds. Since B3PW91-GD3BJ and PBE0-GD3BJ were among the best functionals considered in the earlier study, [17] this suggests that M05-2X-GD3 is also a good choice for modelling hypercoordinate Sn compounds. The substantial similarity in the results for the three calculation methods suggests that the treatment of dispersion is the dominant factor in overall predictive accuracy, minimizing differences in the quality of performance of the hybrid functionals. Overall, for this limited series of compounds, PBE0-GD3BJ slightly outperformed the other two methods. It is interesting to compare the performance in predicting individual bonds and angles. In most cases (4′, 4″, 7, 9′, 9″), M05-2X-GD3 resulted in the smallest mean unsigned error (MUE) in the Sn-N separation distances; B3PW91-GD3BJ produced smaller MUEs for 10 and 11. The Sn-C(Ar) bond MUE was most often smallest for PBE0-GD3BJ (4′, 4″, 7, 10, 11). Overall, M05-2X-GD3 provided better estimates of bond angles at the Sn center. The differences in performance for different bond and angle types is a typical complication of structural benchmarking studies, as is the requirement to average measures across a series of compounds. One of the advantages of the MSSD is that it combines the effects of all bond and angle estimates into one comprehensive measure that can be applied to an individual compound rather than an extended series. This makes it easier to pool observations from earlier studies as well as to evaluate new compounds not previously considered.

Oxazoline Stannane Dihydrides
The synthesis of dihydride 12 was successfully achieved using 1.5 mol equivalent of NaBH4 in EtOH, at −84 °C for 1 h from 10; albeit in relatively low yield (43%).
Compound 12 was initially isolated as a colourless viscous oil and requires cooling (−84 °C) to retard decomposition. Staliński et al. had also reported an analogous chiral stannane, 2h (Figure 2), shows sensitivity above −20 °C [30]. The thermal sensitivity of 12 was evident when a NMR sample of the compound was quickly (~1 min.) prepared in C6D6; the solution rapidly changes from colourless to a bright yellow coloured solution, with a precipitate being noted within a few hours. The 1 H NMR spectrum of clean 12 ( Figure S24)   Compound 13 (Scheme 5) was prepared via a traditional transformation of 11 using a large excess of NaBH4 in EtOH for 1 h at 0 °C. After work-up, a colourless viscous oil was recovered (86%). Compound 13, like 12, turned yellow in colour when dissolved in C6D6; however, clean NMR analysis was still possible. In the 1 H NMR spectrum ( Figure S49), the Sn-H signal appears at δ1H = 5.94 ppm with appropriate 117/119 Sn satellites. A single resonance (δ = −221.0 ppm) was observed in the 119 Sn NMR spectrum for 13, this is 30 ppm downfield relative to that of 12.

Synthesis of Polymers 14 and 15
It has been previously shown that Wilkinson's catalyst is ineffective for the polymerization of diaryl tin hydrides [47]. However, more recent reports have demonstrated that this catalyst is effective with mixed aryl/alkyl tin dihydrides that contain other donor atoms with potential to bind to Sn [7,8,48]. As a consequence of the thermal/solution sensitivity of monomers 12 and 13, dehydropolymerization attempts were carried out immediately after their preparation. Such polymerizations are often performed at RT in toluene; however, a temperature of 0 °C was selected (Scheme 6) to offset the rapid decomposition/oligomerization of these monomers prior to interaction with the catalyst. The recovered crude yellow coloured polymer mixture was cleaned by repeated re-precipitation of THF solution mixture (2 mL) being deposited into cold stirring hexanes (3 × 75 mL) and heptane (1 × 75 mL). A pale-yellow powder of 14 was recovered, which reveals a single, broad 119 Sn resonance ( = −268.1 ppm). This value is 20 ppm upfield shift relative to that of 12. The 1 H NMR spectrum showed the disappearance of the Sn-H signal (δ = 6.79 ppm) and a broadening of the other resonances, all characteristic of polymer formation. Unfortunately, repeated precipitation did not completely remove all of the impurities as identified by NMR spectroscopy. Gel permeation chromatography (GPC) was performed to estimate the molecular weight of 14, which shows a bimodal distribution (Figure 8 red). A molecular weight determination revealed that the polymer was of low Mw ( 10,100 Da) with a Ð of 1.73. This corresponds to short polymer chains with 27 repeat units.

General Considerations
All reagents and solvents were obtained from Sigma-Aldrich (Oakville, Canada) and used as received, unless otherwise indicated. Solvents were dried either through an MBraun solvent drying system, or through vacuum distillation and stored under an inert nitrogen atmosphere. All reactions were carried out under nitrogen atmosphere using Schlenk techniques, unless otherwise noted. 1 H NMR (400 MHz), 13 C NMR (100.6 MHz), and 119 Sn NMR (149.2 MHz) spectra were recorded on a Bruker Avance 400 NMR spectrometer (Bruker Corporation, Billerica, MA, USA) with a BBFO 5-mm direct probe. A 1 H pulse width of 30° was used, acquiring a spectral window of 8223 Hz (20 ppm) using a relaxation delay of 1s, acquisition time 3.98 s, 32 k points (16 scans). The 1 H 90° pulse width was 10.4 µs. A 13 C pulse width of 30° was used, acquiring a spectral window of 24,038 Hz (239 ppm) using a relaxation delay of 2 s, acquisition time 1.36 s, 32 k points (4096 scans). The 13 C 90° pulse width was 8.7 µs. A 119 Sn pulse width of 30° was used, 8.75 µs, acquiring a spectral window of 100,000 Hz (670 ppm) using a relaxation delay of 1s, acquisition time 0.33 s, 32 k points (15,360 scans) with inverse gated proton decoupling. All results were analysed on MestReNova LITE 5.2.5 software (Mestrelab Research, S. L., Santiago Compostelea, Spain). Chemical shifts were calculated using the chemical deuterated standards as a reference for 1 H and 13 C. The 119 Sn chemical shifts were referenced to SnMe4 as an external standard. All J coupling values are reported as absolute values. All NMR assignments of small molecules were made using 2D NMR spectroscopy (COSY, HMBC, HSQC) experiments. Time-of-flight mass spectrometry analyses were performed at the AIMS Mass Spectrometry Laboratory, University of Toronto using a JMS-T1000LC mass spectrometer (JEOL, Inc., Peabody, MA, USA) equipped with a Direct Analysis in Real Time (DART) ionization source (DART-SVP, Ionsense, Inc., Saugus, MA, USA). The DART source was operated with He gas and the temperature was adjusted in the range 100-400 C. Isotopic distributions for the observed ionic species were calculated using the Mass Center utility (JEOL) and were in good agreement with the measured mass spectra. Molecular weights of the polymers were determined by GPC using a Viscotek Triple Model 302 Detector system. GPC columns were calibrated versus polystyrene standards (American Polymer Standards). A flow rate of 1.0 mL/min. was used with ACS grade THF as the eluent. GPC samples were prepared using 3-10 mg of each polymer per mL THF, and filtered using a 0.45 µm filter. All reactions were carried out under a nitrogen atmosphere using Schlenk techniques, unless otherwise described. The X-ray diffraction data for single crystals of compounds 4, 7, 9, 10, and 11 were collected on a Bruker Kappa APEX-DUO diffractometer using monochromated Mo-Kα radiation (Bruker Triumph) and were measured using a combination of ϕ scans and ω scans ( Table 4). The data were processed using APEX2 and SAINT programs. Absorption corrections were carried out using SADAB. The structures were solved using SHELXT [49] and refined using SHELXL-2013 [49] for fullmatrix least-squares refinement that was based on F 2 . For all structures, H atoms were included in calculated positions and allowed to refine in a riding-motion approximation with Uiso tied to the carrier atom.

Computational Details
The Gaussian 16 suite of programs was used for all geometric optimizations and frequency calculations [50]. DFT was implemented with three density functionals (B3PW91, PBE1PBE, and M05-2X) supplemented with Grimme's D3 empirical dispersion function and for B3PW91 and PBE1PBE, Becke-Johnson damping [43]. The LANL08d basis set (obtained from the Basis Set Exchange) [51] was used for Sn and Br, and the 6-31+G(d,p) basis set for all other atoms, as a reasonable compromise between accuracy and computational cost [39,40]. Tight convergence criteria and the superfine integration grid was used for all geometry optimizations. Frequency calculations were used in every case to verify that a potential energy minimum had been located. Cartesian coordinates for experimental structures were extracted from crystallographic information files (.cif) with OpenBabelGUI [52].

Synthesis of 2-(2-Bromophenyl)-4,4-dimethyl-4,5-dihydrooxazole (5)
Method 1: Thionyl chloride (3.60 mL, 49.8 mmol) and 3 drops of DMF were added to a 250 mL round bottom flask containing 2-bromobenzoic acid (5.00 g, 24.9 mmol) in 100 mL of dry DCM. The reaction mixture was heated to reflux temperature for 2 h. The solvent and excess thionyl chloride were then removed under reduced pressure to afford a yellow coloured liquid. The product was dissolved in 30 mL of dry DCM and then added dropwise to a 250 mL round bottom flask containing a stirring solution of distilled 2-amino-2-methyl-1-propanol (2.66 g, 29.9 mmol) and NEt3 (5.

Synthesis of 4,4-Dimethyl-2-(2-(triphenylstannyl)phenyl)-4,5-dihydrooxazole (7):
Method 1: Compound 5 (0.565 g, 2.22 mmol) and 15 mL of dry Et2O were added to a dry 100 mL Schlenk flask. The solution was cooled to −84 °C and 1.6 M n-BuLi/hexane (1.52 mL, 2.45 mmol) was added dropwise. The reaction mixture was stirred for 2 h at −84 °C, and then was allowed to warm to 0 °C before the dropwise addition of a solution containing Ph3SnCl (0.857 g, 2.22 mmol) in 20 mL of Et2O. The reaction mixture was stirred for 1 h at 0 °C. The solvent was removed under reduced pressure and 20 mL of toluene was added. The mixture was filtered via gravity filtration and the solvent was removed under reduced pressure to yield a pale yellow coloured powder. The crude product was triturated with MeOH and the yellow supernatant decanted from the solid. The residual solvent was removed under reduced pressure to yield 7 as a white coloured powder. Yield: 0.86 g, 73.9%.
Method 2: Compound 6 (5.00 g, 28.5 mmol) and 50 mL of dry Et2O were added to a dry 250 mL Schlenk flask. The solution was cooled to −84 °C and 1.3 M sec-BuLi/cyclohexane (24.14 mL, 31.39 mmol) was added dropwise. The reaction mixture stirred for 1.5 h at −84 °C and then allowed to warm to 0 °C before the dropwise addition of a solution of Ph3SnCl (11.0 g, 28.5 mmol) in 20 mL of Et2O. The reaction mixture was stirred for 1 h at 0 °C. The solvent was then removed under reduced pressure and 40 mL of toluene was added. The mixture was filtered via gravity filtration and the solvent removed under reduced pressure to yield a pale yellow coloured powder. The crude product was triturated with MeOH and the yellow supernatant decanted from the solid. The residual solvent was removed under reduced pressure to yield 7 as a white coloured powder. Recrystallization was achieved by dissolving the powder (0.5 g) in DCM (1 mL) and layering with hexanes (1.5 mL). The

Synthesis of 4,4-Dimethyl-2-(2-(phenylstannyl)phenyl)-4,5-dihydrooxazole (12)
A suspension of 11 (0.50 g, 0.94 mmol) in 15 mL of dry EtOH was added dropwise to a solution of NaBH4 (0.053 g, 1.42 mmol) in 5 mL of dry EtOH at −84 °C. The solution was stirred at −84 °C for 1 h. Cold hexane (15 mL) was added to the solution and then quenched with 5 mL of chilled degassed distilled water. The mixture was extracted and the organic layer dried over anhydrous MgSO4. The solution was filtered and the solvent was removed under reduced pressure to yield the product 12 as a colourless viscous oil. Yield = 0.151 g, 43%. 1

Synthesis of Polymer (14)
A solution of 12 (0.327 g, 0.879 mmol) in 3 mL of dry toluene was added dropwise to a dry Schlenk wrapped in Al foil containing Wilkinson's catalyst (0.033 g, 0.035 mmol) in 10 mL of dry toluene at 0 °C. The solution was stirred at 0 °C for 4 h, after which the solvent was removed under reduced pressure. The crude product was dissolved in minimal dry THF (2 mL) and added dropwise to cold stirring hexanes (3 × 75 mL) and heptane (1 × 75 mL), and the residual solvent removed under reduced pressure to yield a pale-yellow powder of 14. Yield = 0.052 g, 15.9%. 1

Synthesis of 4,4-Dimethyl-2-(o-tolyl)-4,5-dihydrooxazole (8)
Thionyl chloride (5.35 mL, 73.4 mmol) and 3 drops of DMF were added to a 250 mL round bottom flask containing o-toluic acid (5.00 g, 36.7 mmol) in 100 mL of dry DCM. The reaction mixture was heated to reflux temperature for 2 h. The solvent and excess SOCl2 were then removed under reduced pressure to afford a yellow coloured liquid. The product was dissolved in 30 mL of dry DCM and then added dropwise to a 250 mL round bottom flask containing a stirring solution of distilled 2-amino-2-methyl-1-propanol (4.26 g, 47.7 mmol) and NEt3 (7.68 mL, 55.1 mmol) in 100 mL of dry DCM at 0 °C. The reaction mixture was then stirred for 4 h at RT. The crude sample was washed once with 30 mL of 1 M solution of aqueous HCl and 2 × 30 mL of a saturated aqueous brine solution. The organic layer was then dried over anhydrous MgSO4, filtered, and the solvent removed under reduced pressure to afford the N-(1-hydroxy-2-methylpropan-2-yl)-2-methylbenzamide intermediate as a white coloured powder in a 69.9% yield. The benzamide (5.32 g, 25.7 mmol) was dissolved in SOCl2 (15.0 mL, 206 mmol) at 0 °C. The solution stirred at room temperature for 24 h, followed by addition of 100 mL of Et2O. The precipitate was collected via vacuum filtration and dissolved in 50 mL of DCM. The solution was washed with a saturated aqueous solution of NaHCO3 until the organic layer was neutral and then dried over anhydrous MgSO4. After filtration, the solvent was removed under reduced pressure to yield 8 as a yellow coloured liquid. Yield: 4.07 g, 58.6%. NMR data ( 1 H, 13 C) matches reported literature data within experimental error [32]. 1

Synthesis of 4,4-Dimethyl-2-(2-((triphenylstannyl)methyl)phenyl)-4,5-dihydrooxazole (4)
Compound 8 (0.50 g, 26.4 mmol) and 10 mL of dry Et2O were added to a dry 50 mL Schlenk flask. The solution was cooled down to 0 °C and 1.6 M n-BuLi/hexane (1.82 mL, 2.91 mmol) was added dropwise to produce a dark red opaque solution and allowed to stir for 20 min. A solution of Ph3SnCl (1.02 g, 26.4 mmol) in 15 mL of Et2O was added dropwise to the lithiated mixture to produce a yellow coloured opaque solution and stirred for 1 h at 0 °C. The solvent was removed under reduced pressure and 10 mL of toluene was added. The mixture was filtered via gravity filtration and the solvent was removed under reduced pressure to yield a pale yellow coloured powder. The crude product was triturated with MeOH and the yellow supernatant removed. The residual solvent from the solid was removed under reduced pressure to yield white coloured crystals of 4. 1 H and 13 C NMR experimental data are in good agreement with the reported literature [31]. Yield: 1.10 g, 77.4%. 1

Synthesis of 2-(2-((Dibromo(phenyl)stannyl)methyl)phenyl)-4,4-dimethyl-4,5-dihydrooxazole (11)
A solution of 4 (2.46 g, 4.57 mmol) and 20 mL of DCM was added to a 100 mL Schlenk flask and cooled to 0 °C. A solution of Br2 (0.466 mL, 9.14 mmol) in 10 mL of DCM was added dropwise and the reaction mixture stirred for 1 h at 0 °C. The solvent was removed under reduced pressure to yield a brown-yellow coloured powder. The crude product was triturated with MeOH and the yellow solution was decanted off. The residual solvent was removed under reduced pressure to afford the product as a white coloured powder. Recrystallization of 11 (0.500 g) was achieved by dissolving the powder in a minimal amount of DCM (0.7 mL) and layering with hexanes (1.0 mL), followed by cooling to −20 °C overnight, yielding clear colourless crystals. Yield 2.05 g, 82.3%. 1

Synthesis of 4,4-Dimethyl-2-(2-((phenylstannyl)methyl)phenyl)-4,5-dihydrooxazole (13)
A suspension of 11 (0.50 g, 0.919 mmol) in 15 mL of dry EtOH was added dropwise to a solution of NaBH4 (0.174 g, 4.70 mmol) in 5 mL of dry EtOH at 0 °C. The solution continued to stir at 0 °C for 1 h. Cold hexane (15 mL) was added to the solution and then quenched with 5 mL of chilled degassed distilled water. The mixture was extracted, and the organic layer was dried over anhydrous MgSO4. The solution was filtered, and the solvent was removed under reduced pressure to yield the product as a colourless viscous oil. Yield = 0.305 g, 85.9%. 1

Synthesis of Polymer (15)
A solution of 13 (0.432 g, 1.119 mmol) in 3 mL of dry toluene was added dropwise to a dry Schlenk wrapped in Al foil containing Wilkinson's catalyst (0.0414 g, 0.045 mmol) in 10 mL of dry toluene at 0 °C. The solution continued to be stirred at 0 °C for 4 h and then the solvent was removed under reduced pressure. The crude product was dissolved in minimal dry THF (2 mL) and added dropwise to cold stirring hexanes (3 × 75 mL) and heptane (1 × 75 mL). The residual solvent was removed under reduced pressure to yield a pale-yellow powder of 15. Yield = 0.096 g, 22.2%. 1

Conclusions
Sets of both oxazolines containing triphenyl stannanes, dibromidooxazoline stannanes, and stannane dihydrides were prepared and characterized. A monochloridooxazoline stannane was also prepared. Crystallographic studies of five different  2 -C,N tin-oxazoline compounds adopt a preferentially distorted trigonal bipyramidal geometry around Sn, even in the absence of a strongwithdrawing ligand trans to the Sn-N bond. Short Sn-N bonds were observed for the halo containing tin oxazolines. The oxazoline organotin monomers form an interesting class of compounds for investigating Sn hypercoordination by DFT modelling, particularly as they have two potentially coordinating atoms (N,O). We are currently extending these studies to consider the relative energy differences between oxazoline coordination isomers, the impact of solvation, and the feasibility of predicting 119 Sn chemical shifts in these systems by NMR calculations. Evidence for Sn oxazoline polymers prepared by transition metal catalyzed dehydrocoupling of their respective dihydrides was demonstrated; however, purification and isolation proved difficult. The difficulty in preparing high molecular weight examples stems in part from the extreme sensitivity of the dihydride monomers, which begin to decompose prior to polymerization. Efforts to identify more flexible, yet stable oxazoline tin derivatives is now underway.