Distorted zinc coordination polyhedra in bis(1-ethoxy-2-{[(2-methoxyethyl)imino]methyl}propan-1-olato)zinc, a possible CVD precursor for zinc oxide thin films

The coordination polyhedra of the zinc atoms in the title complex display long Zn—O bonds as parts of distorted trigonal bipyramids and octahedra.


Chemical context
Zinc oxide is of considerable current interest in materials science because it is a semiconductor with a band gap of 3.37 eV and it possesses high electron mobility, a high exciton binding energy of 60 meV, strong room-temperature luminescence, photoelectric response, high transparency, and high photocatalytic activity (Ganesh et al., 2017;Das & Sarkar, 2017). As a result of these favorable properties, ZnO has potential applications in solar cells, sensors, ultra-violet laser diodes, actuators, field-emission devices, photocatalysts and piezoelectric devices (Galstyann et al., 2015;Hong et al., 2017). The identification of a viable technique that is capable of depositing zinc oxide thin films of high purity and high quality is a significant challenge. Metal-organic chemical vapor deposition (MOCVD) has proven to be a promising method for depositing high-quality ZnO thin films at a high growth rate over a large area (Malandrino et al., 2005). The success of the MOCVD process depends heavily on the precursor. An 'ideal' MOCVD precursor should be volatile, exhibit a sufficiently large temperature window between evaporation and film deposition, and decompose without the incorporation of residual impurities. Diethyl zinc, Zn(C 2 H 5 ) 2 , in combination with an oxygen source, H 2 O, or ROH is the traditional precursor for depositing ZnO thin films (Smith, 1983). As a result of the pyrophoric nature of the alkyl zinc reagents and the gas-phase pre-reaction that results in precursor decomposition and film contamination, alternative precursors such as alkoxide, dialkyl zinc precursors of acetate and acetylacetonate have been employed (Sato et al., 1994). The drawback with these precursors is that impurities are often incorporated in the deposited ZnO films. These disadvantages have resulted in a search for single-source precursors. A single-source precursor is one that has the oxygen already present in the precursor, thereby eliminating the need for an external oxygen source.
The synthesis of two thermally stable ketoiminato zinc complexes [Zn{[(CH 2 ) x OCH 3 ]NC(CH 3 ) C(H)C(CH 3 ) O} 2 ] (1: x = 2; 2: x = 3) were reported with melting points as low as 330 K (Barreca et al., 2010;Bekermann et al., 2010a,b). In another case, ketoiminato zinc complexes that incorporate ether O-donor atoms have shown promise (Cosham et al., 2015). With these favorable results in mind, we decided to further explore the -enaminoalkoxyester ligand platform. Our research group has demonstrated that high-quality ZnO thin films with fewer impurities can be accomplished by utilizing Zn-bis--iminoesterate complexes (Matthews et al., 2006;Onakoya et al., 2011;Gbemigun et al., 2019). Studies have shown that the organic ligand attached to the N moiety of the zinc complex has a significant effect on the level of carbon incorporated into the deposited ZnO thin film (Manzi et al., 2015), thus the investigation of such compounds with different substituents at the N atom is of significant interest in improving precursors for these ZnO films. Herein, the synthesis, characterization and crystal structure of the title compound 1 are reported.

Structural commentary
The synthesis of [Zn(C 9 H 16 NO 3 ) 2 ] (1), was carried out by the direct reaction of 1a with diethyl zinc in a 2:1 molar ratio under an inert atmosphere of nitrogen utilizing Schlenk techniques to afford white single crystals of complex 1. The 1 H-NMR and 13 C-NMR spectra of 1 contain the characteristic resonances in the expected regions. The 1 H-NMR spectrum in particular shows the absence of the N-H resonance ( = 8.63) that was present in the free ligand (1a), indicating the absence of any starting material. Generally, the introduction of a Lewis acidic metal center into the ligand sphere results in the proton and carbon resonances being shifted downfield (Matthews et al., 2006). This was not observed in this study: in going from the free ligand (1a) to complex 1 most of the proton and carbon resonances were slightly shifted upfield. This incon-sistency suggests that the electron density in the chelate ring of 1 is not completely delocalized around the ring. If complete delocalization was observed, the carbon atoms and protons in the complex would have been deshielded and the resonances would have been shifted downfield.
The title complex, C 18 H 32 N 2 O 6 Zn, 1, crystallizes in the monoclinic space group P2 1 /c with eight molecules in the unit cell, thus two in the asymmetric unit (Z 0 = 2 and named as A and B for the purposes of discussion), which have adopted different metal-ion coordinations and conformations (Table 1). In molecule A (Fig. 1   Both ketoimine chelate rings are almost planar (r.m.s. deviations of 0.018 and 0.026 Å for molecule A and 0.002 and 0.014 Å for molecule B) with the zinc atoms deviating from the respective planes by 0.089 (6)/0.220 (6) Å and 0.248 (2)/ 0.030 (7) Å for A and B, respectively. The dihedral angles between the chelate planes in 1 are 71.4 (1) and 77.3 (1) for the A and B molecules, respectively.

Figure 3
Packing diagram for 1 showing both the intra-and intermolecular C-HÁ Á ÁO interactions.

Experimental
All chemicals were purchased from Aldrich and used without further purification. The 1 H and 13 C-NMR spectra were recorded with a Bruker AVANCE 400MHz Ultra Shield TM NMR spectrometer. Chemical shifts for 1 H (400MHz) and 13 C (100MHz) were referenced to CDCl 3 and reported in ppm. Thermogravimetric analyses were performed under a nitrogen atmosphere at 1atm using a Perkin-Elmer thermogravimetric analyzer series 7 at a heating rate of 10 C min À1 . All manipulations were carried out using oven dried, standard reflux glassware consisting of a condenser connected to a roundbottom flask. Distillation was performed using oven-dried micro-still apparatus.

Synthesis and crystallization
Synthesis of ethyl-3-N-(2-methoxyethylamino)but-2-enoate (1a) Ethyl acetoacetate (5.00 g, 38.42 mmol) and 2-methoxyethylamine (5.77 g, 76.84 mmol) were added to a 100 ml round-bottom flask via syringe. The solution was refluxed for 1h with constant stirring. The resulting mixture was allowed to cool to room temperature and approximately 30 ml of hexane was added to dissolve the product. The solution was then dried over anhydrous sodium sulfate. The resulting mixture was then filtered, and the solvent was evaporated in vacuo to afford a viscous yellow oil. This crude product was then purified via vacuum distillation to afford a viscous light-yellow oil (1a)  Synthesis and crystallization of [Zn (C 9 H 16 NO 3 ) 2 ] (1) 50ml of dried hexanes, ethyl-3-N-(2-methoxyethylamino) butanoate (1a) (6.87 g, 36.5 mmol) and a stir bar were added to a 250 ml Schlenk flask under an inert atmosphere of nitrogen. The mixture was degassed with N 2 gas for approximately fifteen minutes then diethyl zinc (2.25 g, 18.25 mmol) was added. The resulting mixture was refluxed for 4 h with constant stirring. The solvent was removed in vacuo at room temperature to afford a viscous yellow oil. The yellow oil was recrystallized from a solution in dry hexanes for 48 h at 243 K to afford white needle-like crystals. The hexanes were removed using a cannula and the white needle-like crystals were purified by washing with cold 10 ml portions of dried hexanes (yield 71.7%, 5.73 g), m.p. 311.0-311.

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 3. This was a highly airsensitive compound and the best available crystal was chosen. However it was non-merohedrally twinned with multiple components. Integration and refinement using the hklf5 (twinned) file was not successful so the hklf4 file was used. Consequently there are two significant difference peaks in chemically unreasonable positions. A face-indexed absorption correction was applied but there are still some residual peaks near the metal atoms. For one of the asymmetric molecules there is disorder in some of the ethyl substituents. These were constrained to have similar metrical parameters and refined with occupancy factors of 0.717 (13)/0.283 (13) and 0.68 (3)/ 0.32 (3). A riding model was used for the H atoms with atomic displacement parameters = 1.2U eq (C) [1.5U eq (CH 3 )], with C-H bond lengths ranging from 0.95 to 0.99 Å .

Funding information
Funding for this research was provided by: National Science Foundation, Directorate for Mathematical and Physical Sciences (grant No. CHE-0619278).

Bis(1-ethoxy-2-{[(2-methoxyethyl)imino]methyl}propan-1-olato)zinc
Crystal data Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes.