Molybdenum complexes derived from the oxydianiline [(2-NH2C6H4)2O]: Synthesis, characterization and ε- caprolactone ROP capability

The reaction of Na2MoO4 with 2,2/-oxydianiline (2-aminophenylether), (2-NH2C6H4)2O, LH4, in DME (DME=1,2dimethoxyethane) in the presence of Et3N and Me3SiCl afforded either the bis(imido) molybdenum(VI) complex {Mo(L)Cl2(DME)} (1), where L = (2-NC6H4)2O, or the molybdenum(V) salt [Mo(L/)Cl4][Et3NH] (2), where L/ = [(2-NH2C6H4)(2-NC6H4)O], depending on the work-up method employed. The same diamine reacted with in-situ [Mo(NtBu)2Cl2(DME)] afforded a tetra-nuclear complex 10 [Mo4Cl3(NtBu)3(OSiMe3)(μ4-O)(L)2(L/)2]·2MeCN (3·2MeCN). The crystal structures of 1, 2 and 3·2MeCN have been determined. The structure of the bis(imido) complex 1 contains two unique molecules paired up via weak π-stacking, whereas the structure of 2 contains a chelating amine/imido ligand, and is made up of discrete units of two cations and two anions which are interacting via Hbonding. The tetra-nuclear structure 3 contains four different types of distorted octahedral molybdenum centre, and a bent Me3SiO group thought to originate from the precursor synthesis. Complexes 1 3 have been screened for their ability to ring open polymerize (ROP) ε15 caprolactone. For 1 and 3 (not 2), conversion rates were good (> 90 %) at high temperatures (100 oC) over 6 24 h, and the polymerization proceeded in a living manner.


Introduction
Entry into molybdenum(VI) organoimido [Mo=NR] chemistry, which has proved pivotal in designing new ring opening metathesis polymerization (ROMP) catalysts over the years, [1] is convenient via the one-pot 'sodium molybdate' route, [2] or when basicity permits, the 'tert-butylimido exchange' route.[3] Such methods have allowed for the introduction of a range of organoimido groups with varying electronic and steric properties, which ultimately can be used to affect/control the polymerization processes.In the 'molybdate prep', a limited number of potentially chelating anilines/amines have been employed, namely types I or II (see scheme 1), which have resulted in bis(imido) type ligation forming 7, 8 or 9 membered chelate rings.[4] By contrast, the use of highly functionalized anilines, such as 2-aminoterephthalic acid III, instead have been shown to form triethylammonium salts, the anions of which bear an imido group together with a chelate ligand derived from the functionalized aniline.[5] An intriguing example is the use of 2diphenylphosphinoaniline [1,2-(NH2)(PPh2)C6H4] IV which, when employed in the 'sodium molybdate' route, results in a novel MoNC2PN six-membered ring derived from P-N coupling, whereas bridging imido/terminal phosphine ligation results from the use of IV in the 'imido exchange' route.[6] With this in mind, we have turned our attention to the use of the potentially chelating 2,2 / -oxydianiline (2-NH2C6H4)2O, and find that its use in the 'sodium molybdate' route affords either a mononuclear Mo(VI) bis(imido) complex 1 or a Mo(V) salt complex 2 bearing a chelating amine/imide ligand in the anion, 45 depending on the work-up employed.By contrast, the 'imido exchange' route affords a more complicated tetra-nuclear product 3 that contains four different molybdenum coordination environments (see scheme 2).The molecular structures of these complexes are reported herein.A search of the Cambridge 50 Crystallographic database (CSD) revealed 62 hits for motifs derived from (2-NH2C6H4)2O binding to any metal; the majority of these examples contained larger (mostly Schiff-like) ligands incorporating the backbone of the dianiline.Indeed, there was only one recent structural report of complexes (of palladium) 55 incorporating (2-NH2C6H4)2O type ligands or the deprotonated version thereof.[7] Despite the great success of molybdenum species in ROMP, the use of molybdenum in the ROP of cyclic esters, typically lactides or lactones, has received far less attention.[8] Given this, we 60 have screened 1 -3 for their ability to act as catalysts in the ROP of ε-caprolactone (in the presence of benzyl alcohol).Interest in poly(ε-caprolactone) stems from it use as a biodegradable polymer.[9] [journal], [year], [vol], 00-00 |1 5

Synthesis
The reaction of Na2MoO4 with (2-NH2C6H4)2O (LH4) in refluxing DME (DME=1,2-dimethoxyethane) in the presence of Et3N and Me3SiCl afforded, following filtration, cooling, and prolonged standing (1 -2 days) at ambient temperature, red crystals of the bis(imido) complex {Mo(L)Cl2(DME)} (1), where L = (2-NC6H4)2O, in good yield (ca. 60 %).A single crystal was subjected to X-ray diffraction and the molecular structure is shown in Figure 1, with selected bond lengths and angles given in the caption; crystallographic data are given in Table 2.There are two very similar molecules of {Mo(L)Cl2(DME)} in the asymmetric unit.The molybdenum centre has a distorted octahedral geometry with trans chlorides and cis-chelating bis(imide) ligation.The Mo -N -C angles are about 153 o , which is at the lower end of the range associated with linear imido groups.[10] The N(1) -Mo(1) -N(2) bite angle is 101.21 (11) o , which is similar to the bite angle observed for the 8membered ring formed in the chelating bis(imide) complex {Mo[(2-NC6H4)2CH2]Cl2(DME)} [100.8( 5) o ] derived from I (n = 1).[4a] The two unique molecules pair up into weakly πstacked units with a centroid-to-centroid distance of 3.767 Å (see Figure 2).There are a number of intermolecular C-H•••Cl interactions at ca. 2.8 Å.   2) was isolated in ca.30 % yield.The IR spectrum contained broad weak stretches in the v(NH/NH2)region (ca.3068/3320 cm -1 ), whilst signals in the 1 H NMR spectrum were broad and uninformative, consistent with the presence of a Mo(V) centre.Reduction to Mo(V) during such syntheses has been noted previously, for example when using diphenylglycine.[5] The X-band EPR spectrum (see Fig. S1) of 2 was reminiscent of that recorded for [MoOCl5] -. [11] Crystals of 2 suitable for an X-ray diffraction study were grown from acetonitrile on prolonged standing at ambient temperature.The molecular structure is shown in Figure 3 with selected bond lengths and angles given in the caption.Crystallographic data are presented in Table 2.The complex is a salt comprising an Et3NH cation and an anion in which a chelating amine/imide ligand is bound to the molybdenum(V) centre (Figure 3).The NH2 groups are involved in H-bonding to two of the chlorides of a neighbouring anion giving a total of four H-bonds between anions.A third chloride and one of the others on each molybdenum is involved in asymmetric, bifurcated, H-bonding with the cation.Given this, N(1), Cl(1), Cl(2) and Cl(4) are not subject to any disorder, whereas Cl(3) and the imido group at N(2), which are not involved in any H-bonding, are able to swap sites (major:minor disorder components = 0.805:0.195(4)).Overall, the structure is made up of discrete units of two cations and two anions, all H-bonded with the Cl(3) and with imido groups able to occupy either site (see Figure 4).Use of the same dianiline, but using the 'imido-exchange' route via in-situ generated [Mo(NtBu)2Cl2(DME)] led, following work-up, to the isolation of a purple crystall suitable for X-ray crystallography.We tried to grow single crystals suitable for Xray diffraction from toluene, but were unsuccessful.However, when the toluene was removed and these samples were taken up in acetonitrile, single crystal of X-ray diffraction quality were eventually obtained.The molecular structure revealed the structure [Mo4Cl3(NtBu)3(OSiMe3)(μ4-O)(L)2(L / )2]•2MeCN (3•2MeCN) and this is shown in Figure 5, with selected bond lengths and angles given in the caption.All four molybdenum centres adopt different distorted octahedral geometries, but all are bound to the central µ4-oxo group.Given the complex nature of the structure, the core of 3 is also presented in Figure 6.The complexity of the structure is also reflected in the 1 H NMR spectrum (for low field region, see Fig. S2), in which a number of the peaks are subject to ring currents which results in unusual high field chemical shifts for aromatic protons (as well as the tert-butyl/siloxide groups).[12] In the solid state structure, for Mo(1) and Mo(4), there is a chelating L / ligand present, but thereafter the coordination environments differ in that Mo(1) has a chloride and a near linear imido group [Mo(1) -N(1) -C(1) = 168.5(5) o ], whereas Mo(4) has a bent siloxide group [Mo(4) -O(2) -Si(1) = 142.15(18) o ] and two nitrogens of a chelating (2-NC6H4)2O ligand which also acts as a bridge between Mo(2) and Mo(3).In the case of Mo(2) and Mo(3), the difference is essentially due to the nature of the bonding of the nitrogens derived from the precursor oxydianiline.For Mo(2), the 'chelates' are only bound in η 1 -fashion, with that also bound to Mo(1) doing so via an NH2 moiety, whereas for Mo(3) there is a bidentate chelating (2-NC6H4)2O ligand present as well as an oxydianiline derived ligand binding in η 1 -fashion.To complete the distorted octahedral environment, both Mo(2) and Mo(3) also bear a chloride as well as a linear tert-butylimido group [Mo(2/3) -N(2/3) -C(5/9) = 168.5(5)and 168.5 (5) o ].In the structure, there are long bonds from the metals to the nitrogen atoms N(6) and N(9) at ca. 2.14 Å, however their near trigonal planar nature (as opposed to pyramidal) is consistent with a lack of protonation.There is some intramolecular H-bonding involving Cl(1) and Cl(3) with N(1)H2, Cl(2) with N(4)H2 and O(2) with N(4)H2.The methyl groups of the OSiMe3 group and the tBu group at C(1) are disordered over two sets of positions: major:minor components 64(2):36(2) and 62.8:37.2(8)%,respectively.Whilst the origin of the central µ4-oxo is thought to be fortuitous oxygen, the siloxide group is thought to arise via excess Me3SiCl present in the system.We note that silyl ester group formation via Me3SiCl attack has previously been reported for the 'sodium molybdate' route when using anthranilic acid as the amine source.[5]   A search of the CSD revealed 1024 hits for µ4-O Mo species, the majority of which are hexamolybdate type species, whilst narrowing the search to those complexes also bearing an imido group gives 115 hits (again, many of these are imido functionalized hexamolybdates).[13] Ring opening polymerization of ε-caprolactone Complexes 1 -3 have been screened for their ability to act as catalysts in the ring open polymerization of ε-caprolactone and the results are depicted in Table 1.For 1 -3, runs were conducted in the presence of benzyl alcohol (BnOH), however, given the presence of the silyloxide group in the tetra-nuclear complex 3, this complex was also screened in the absence of BnOH.Results using 2 were not encouraging and so this complex is not discussed further.For catalytic systems utilizing 1 and 3, at temperatures below 80 o C, little or no activity was observed, however at higher temperatures, good conversions over 1 h were possible.All the polycaprolactone polymers (PCLs) obtained possessed a narrow distribution with unimodal characteristics [Mw/Mn = 1.10 -1.82].There is a near linear relationship between monomer conversion and the Mn values for the polymers, which is indicative of a living polymerization process (see Figs. S3 and S4).
We have also investigated the effect of the ε-CL/Mo molar ratio on the catalytic behaviour (entries 7 -9 for 1 and 13 and 18 -20 for 3).When the molar ratio ε-CL:Mo was increased from 125 or 250 to 1000, the molecular weight for 1 initially decreased (from ca 21 x 10 3 g mol -1 ) and then remained relatively constant (ca 13 x 10 3 g mol -1 ), whereas for 3, the molecular weight gradually increased from ca 5 to 53 x 10 3 g mol -1 with a gradual increase in the conversion rate, but with little change of molecular weight distribution (1.23 -1.65).
In general, the resulting polymer molecular weight was lower than expected, which indicates that in most cases, there were significant trans-esterification reactions occurring as evidenced in the MALDI-TOF mass spectrum for run 10 (see Figure S5).However, in cases where there is better agreement between observed and calculated molecular weight, the MALDI-TOF mass spectrum reveals only a single population of peaks (see Figure S6 for run 18).In the 1 H NMR spectra of the PCL (Figures S7 and S8), signals at around 7.34 and 5.15 ppm (C6H5CH2-), and 3.62 ppm (CH2CH2OH), with an integral ratio 5:2:2, indicated that the polymer chains are capped by a benzyl group and a hydroxy end group.This suggests that the polymerization occurs through insertion of a benzyl alkoxy group into CL. 13C NMR data (Figures S9 and S10) also revealed peaks at 128.2 (C6H5CH2-), 66.2 ppm (C6H5CH2-), 62.6 ppm (CH2CH2OH) assigned to the respective end groups.The aforementioned MALDI-TOF spectra also revealed the presence of the benzyloxy initiating group and a series of peaks separated by 114.14 mass units.
The limited number of Mo systems reported thus far tend to only be active at elevated temperatures, for example ammonium decamolybdate can function as part of a melt at 150 o C afforded In terms of the end group for the PCL obtained when using 3/no BnOH, we can only tentatively propose a siloxide as there is a peak at about 0 ppm that we can assign to the siloxide (run under grease-free conditions), however the polymer seems to form a coating which hampers attempts to obtain the MALDI-TOF mass spectrum and so we are only able to see low molecular weight (see ESI).Table 1.ROP studies using complexes 1 -3.The structure of 3 contains four different octahedral molybdenum environments.Complexes 1 and 3 were shown to be capable of the ROP of ε-caprolactone at temperatures ≥ 80 o C with good conversions in the presence of benzyl alcohol, affording moderate molecular weight (Mn = ca 5 -30 x 10 3 g mol -1 ) polymers.Silyloxide 3 is also capable of the ROP of ε-caprolactone with good conversion in the absence of benzyl alcohol.

Experimental
General: All manipulations were carried out under an atmosphere of dry nitrogen using conventional Schlenk and cannula techniques or in a conventional nitrogen-filled glove box.DME was refluxed over sodium and benzophenone.Toluene was refluxed over sodium.Acetonitrile was refluxed over calcium hydride.All solvents were distilled and degassed prior to use.IR spectra (nujol mulls, KBr windows) were recorded on a Nicolet Avatar 360 FT IR spectrometer; 1 H NMR spectra were recorded at room temperature on a Varian VXR 400 S spectrometer at 400 MHz or a Gemini 300 NMR spectrometer or a Bruker Advance DPX-300 spectrometer at 300 MHz.The 1 H NMR spectra were calibrated against the residual protio impurity of the deuterated solvent.The EPR of 2 was recorded on a Bruker EMX -10/12 spectrometer.Elemental analyses were performed by the elemental analysis service at the London Metropolitan University or at Sichuan University.Matrix Assisted Laser Desorption/Ionization Time of Flight (MALDI-TOF) mass spectrometry was performed in a Bruker autoflex III smart beam in linear mode, and the spectra were acquired by avaeraging at least 100 laser shots.2,5-Dihydroxybenzoic acid was used as the matrix and THF as solvent, except in the case of the sample form 3 with no benzyl alcohol for which 9-nitroanthracene was employed as matrix.Sodium chloride was dissolved in methanol and used as the ionizing agent.Samples were prepared by mixing 20 µl of matrix solution in THF (2 mg/ml) with 20 µl of matrix solution (10 mg/ml) and 1 µl of a solution of ionizing agent (1 mg/ml).Then 1 ml of these mixtures was deposited on a target plate and allowed to dry in air at ambient temperature.The precursor [Mo(NtBu)2Cl2(DME)] was prepared by the literature method, [14] whilst the oxydianiline was prepared by the method of Randell et al. [15] Synthesis To Na2MoO4 (3.00 g, 14.6 mmol) and (2-NH2C6H4)2O (2.92 g, 14.6 mmol) in DME (200 ml) were added Et3N (7.73 ml, 58.3 mmol) and Me3SiCl (14.95 ml, 116.5 mmol), and the system was heated to 90 o C for 12 h.On cooling, filtration followed by concentration to about half volume afforded small red prisms of 1 on prolonged standing (1 -2 days) at ambient temperature.Yield 3.96 g, 60 %.C16H18N2O3Cl2Mo requires C 42.40, H 4.00, N 6.18 %.Found: C 41.98, H 4.10, N 6.18 %.IR: 3090w, 3058w,

Ring opening polymerization
Typical polymerization procedures in the presence of one equivalent of benzyl alcohol (Table 1, run 1) are as follows.A toluene solution of 1 (0.030 mmol, 1.0 mL toluene) and BnOH (0.030 mmol) were added into a Schlenk tube in the glove-box at room temperature.The solution was stirred for 2 min, and then εcaprolactone (2.5 mmol) along with 1.5 mL toluene was added to the solution.The reaction mixture was then placed into an oil bath pre-heated at 80 °C, and the solution was stirred for the prescribed time (12 h).The polymerization mixture was then quenched by addition of an excess of glacial acetic acid (0.2 mL) into the solution, and the resultant solution was then poured into methanol (200 mL).The resultant polymer was then collected on filter paper and was dried in vacuo.

Crystallography
Diffraction data for 1 were collected on an Agilent Xcalibur Eos CCD diffractometer at 148(2)K [16].Diffraction data for 2 and 3•2MeCN were measured on a Bruker SMART 1000 CCD diffractometer at 150(2)K.[17] Corrections were made for absorption and for Lorentz and Lp effects.[16,17] The structures were solved by superflip (1)

Table 2 :
(14)irect methods (2 and 3.2MeCN) and refined on F 2 by full-matrix-least squares.[17-20]Hatoms on nitrogen were freely refined with similarity restraints on N-H bond lengths, those on C atoms were modelled using a riding model.In 2 atoms Cl(3) and N(2)/O(1)/C(7)>C(12)were modelled as two-fold positionally disordered with major component 80.5(4)%.In 3•2MeCN the whole tBu group at C(1) was modelled with two fold disorder with major component 62.8(8)% and the methyl groups at Si(1) were similarly disordered with major component 64(2)%.One MeCN of crystallisation was modelled at full occupancy, while those at N(13) and N(14)were both modelled as two-fold disordered with the two components adding up to half an MeCN molecule.CCDC 1013319, 1013320, and 1061146 contain the supplementary crystallographic data for this paper.These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif. Crystallographic data for complexes 1, 2 and 3•2MeCN [journal], [year], [vol], 00-00 |8