A Novel Surfactant Synthesis of BaTi0.85Zr0.15O3 (BTZ): Highly Efficient Catalyst for Synthesis of 1,5-benzodiazepine Derivatives under Solvent-free Conditions

INTRODUCTION
Due to pharmacological and industrial properties, benzodiazepines and their polycyclic derivatives are very significant compounds [1]. Particularly, 1,5-benzodiazepines are useful precursors for the synthesis of some fused ring benzodiazepine derivatives, such as triaxolo-, triazolo-, oxadiazolo-, oxazino- or furano-benzodiazepines [2]. Many reagents have been reported in the literature for the synthesis of benzodiazepine including AlTP and AlMP [3], H3PMo12O40 [4], boric acid [5], Clay (KSF and K10)-supported heteropoly acids [6], glycerol [7] and BF3–H2O [8]. However, many of these methodologies are associated with several shortcomings, such as long reaction times, low yields, drastic reaction conditions, co-occurrence of several side products and very expensive reagents. Moreover, the main disadvantage of the reported methods is that the catalysts are destroyed in the work-up procedure and cannot be recovered or reused.
Metal oxide nanomaterials represent a growing asset in many industries, especially with their heightened chemical, physical, and electronic properties compared with their bulk counterparts. Metal oxide nanomaterials are versatile materials that can be used in applications such as environmental remediation, medical technology, energy, water treatment, and personal care products [9-13]. Barium titanate (BaTiO3) is a perovskite type (ABO3) ferroelectric material that has been known since the 1940s [14] However, it still attracts much attention as a promising and environmentally friendly material for a variety of electronic devices such as capacitors, memory storage systems, piezoelectric, pyroelectric and microwave components [15,16]. Among the possible modifications, the substitution of Ti+4 ion by the larger ionic radius Zr+4 in the B-site leads to the solid solution compound BaTi1-xZrxO3 (BTZx) [17].
Herein, we wish to report a suitable method for the use of BaTi0.85Zr0.15O3 (BTZ) as heterogeneous catalysts for the synthesis of 1,5-benzodiazepine derivatives (Fig. 1).

MATERIALS AND METHODS
Chemicals and Instruments
Hexadecylamine 98% (HAD), Titanium(IV) isopropoxide 97% and zirconium butoxide was purchased from Sigma-Aldrich; chemicals. All other reagents were of analytical grade and used without treatment.
A Perkin Elmer Spectrum 65 spectrophotometer was applied to record infrared spectra (400–4000 cm-1) from KBr pellets. Cu Kα (1.54056 Å) radiation with automatic control was employed to gain powder XRD diffraction patterns on a XRD, Bruker D8 ADVANCE and PW1830 instrument. Adsorption/desorption of nitrogen at liquid nitrogen temperature was operated to determine BET specific surface areas and pore volumes of the catalysts with a Micromeritics BELSORT mini ΙΙ instrument. The samples were outgassed at 623 K for 12h under a vacuum of 10-4 Pa prior to the adsorption measurements. Catalyst pore sizes were achieved from the peak positions of the distribution curves detected by the adsorption branches of the isotherms. In temperature programmed desorption of ammonia (NH3-TPD; Nano sort-NS91), 0.1 g of the catalyst was taken in a U-shaped, flow-through, quartz sample tube. Thermogravimetric analysis (TG) and differential thermal analysis (DTA) were carried out on a BAHR-STA-504 apparatus. Thermal analyses were utilized in the range of 25oC-800 oC, with a heating rate of 10 K.min-1.

Preparation of BaTi0.85Zr0.15O3 (BTZ)
BaTiO3 (0.8 g) was homogeneously dispersed in ethanol (97.4 mL) by ultrasonication, followed by the addition of 1.0 g of hexadecylamine (HAD), as surfactant, and 2 mL of ammonia, and stirring at room temperature for 30 min to form a uniform dispersion. Then, 1.1 mL of titanium isopropoxide and 1.7 mL of zirconium butoxide was added to the dispersion under stirring; the milky white mixture was kept static after 2 h 5 and then aged at room temperature overnight. The white powder sol was collected by centrifugation then washed with water and ethanol several times. To prepare hollow mesoporous BTZ nano spheres with Cubic framework, a solvothermal treatment of the precursor beads was performed. A 1.6 g of the BTZ precursor spheres was dispersed in a 20 mL ethanol and 10 mL water mixture with an ammonia concentration of 0.5 M. Then the resulting mixtures were sealed within a Teflon-lined autoclave (50 mL) and heated at 160 °C for 16 h. After centrifugation and ethanol washing, the air-dried powders were calcined at 600 °C for 4 h in air to remove HDA templates for characterization. The experimental procedure is presented in the flowcharts of Fig. 2. 

General procedure for synthesis of 1,5-benzodiazepine derivatives in solution
The reactions were carried out by taking a 1:2.5 mole ratio mixture of 1,2-phenylenediamine and the ketone in the presence of catalytic amount of BTZ nano spheres (0.1 g) in a mortar with a pestle at room temperature for the appropriate time. The progress of the reaction was monitored by GC. After completion of the reaction, 10 ml of CH2Cl2 was added to the reaction mixture and the catalyst was recovered by filtration. The organic layer was concentrated and the products was purified by silica gel column (100–200 mesh) and eluted with EtOAc:n-hexane (2:8) to afford pure compound in 80–98% yield. The wet catalyst was recycled and no appreciable change in activity was noticed after a few cycles. The spectral data of the some of the compounds are given below.
 
RESULTS AND DISSCUSSION
Characterization 
Fig. 3 presents the FT-IR spectrum of BTZ nano spheres with an absorption region of 400–4000 cm-1. Several bands were observed for BTZ nano spheres in the FT-IR spectrum. It is well known that two kinds of OH- groups, such as a surface adsorbed OH- and a lattice OH- group can be observed in bariumtitanate [18]. The broad low-intensity peak with the maximum around 3407 cm-1 has been assigned to O-H stretching modes of surface adsorbed water corresponding to the stretching vibrations of weakly bound water interacting with its environment via hydrogen bonding and to stretching vibrations of hydrogen-bonded OH groups. The 1618 cm-1 peak was due to the deformation mode of absorbed H2O molecules, assigned to the bending vibration. FTIR for pure BTZ sample shows that a strong peak around 596 cm-1 appeared, which was assigned to TiO6 stretching vibration that connected to the barium ion and Ti-O stretching vibration along the polar axis of spontaneous polarization in BTZ with tetragonal phase like BaTiO3 [19,20].
Fig. 4 displayed showed the XRD pattern of BaTi0.85Zr0.15O3, and the diffraction peaks of BaTi0.85Zr0.15O3 corresponded well in position with the standard pattern, indicating the phase purity of the sample. The strong peaks (2q) included 22.63, 31.97, 39.33, 45.68, 51.33, 56.63, 66.13, 70.78, and 75.48, which corresponded to (010), (011), (111), (020), (102), (112), (022), (212), and (130) crystal planes, based on JCPDS card No. 74-1963. The sample showed good crystallinity. The crystallite size of the calcined powder was determined as 12 nm from the full-width half maximum (FWHM) of the (111) crystallographic plane using Scherrer’s relation [21], with a rhombohedral structure [22].
The interplanar distance (d) and the crystallite size (D) of the as-prepared sample are estimated using the Bragg (1) and the Scherrer formula (2), respectively [23]:

Where n is a positive integer, λ is the wavelength of the Cu Kα radiation (λ = 0.15406 nm), k = 0.89 (the Scherrer constant), β is the width of the peak (full width at half maximum (FWHM)), and θ is the diffraction angle. The average interplanar distance and crystallite size of the sample estimated using the Bragg and the Scherrer formula from its XRD patterns are 0.2044 nm and 18.61 nm, respectively. 
The dislocation density (δ) and strain (ε) are calculated using equations (3,4) [24]. The calculated crystallite size and other structural parameters are shown in Table 1.

According to Williamson-Hall (W-H) equation (5) [25], a plot is drawn with 4sinq along the x-axis and bcosq along the y-axis for as-prepared BTZ as shown in Fig. 3.

From the linear fit to the data, the crystalline size was estimated from the y-intercept, and the strain e, from the slope of the fit. Equation represents the UDM (Uniform Deformation Model), where the strain was assumed to be uniform in all crystallography directions, thus considering the isotropic nature of the crystal, where the material properties are independent of the direction along which they are measured. The uniform deformation model for BTZ nanoparticles is shown in Fig. 5. 
The N2 adsorption–desorption isotherms of BTZ sample is shown in Fig. 6. In accordance with the IUPAC nomenclature, both samples show typical IV type isotherms and H1 type hysteresis looped at high relative pressures [26]. It can be seen from Fig. 1, that all adsorption and desorption isotherms display a sharp rise at medium relative pressure P/P0 of 0.2–0.8.
The SEM image of BTZ  (Fig. 7) shows smooth shericals and well-defined edges. The average particle size of these BTZ powders is about 50 nm and exhibits homogeneous distribution of the size and shape. The size distribution is shown in Fig. 6e. The maximum frequency of the particles, size is under 50 nm.
Fig. 8 shows thermal stabilities of BTZ determined by thermogravimetric analysis in the range of 25–1000 oC. The 3.72 % mass loss of BTZ in the temperature range of 25–150 oC is attributed to was attributed to dehydration reactions. A smooth weight loss step of over 10.21 % was also observed as the temperature increased from 250 to 700 oC, which occurs due to the combustion/oxidation of BTZ reactions in that temperature range [27]. Approximately 86.07 % of the starting weight remained after the decomposition, which indicates the formation of the metal oxides. 
To further investigate the acidity sites of BTZ, we carried out the temperature-programmed desorption of NH3-TPD (Fig. 9) which explains three peaks. Two peaks at around 231 and 870 oC implies the weak range acidic sites and another peak at around 528 oC displays the strong acidic sites. These strong acidic sites are disclosed due to Lewis acidity of ZrO2 nanoparticles [28]. 
Synthesis of 1,5-benzodiazepine derivatives
The best experimental procedure was obtained via optimization of the catalytic amount of BTZ, various solvents and reaction temperature (Table 2). The optimal conditions have been established with 0.05 g of BTZ in CH3CN under reflux condition. 
The synthesis of 2,3-dihudro-1H-1,5-benzodiazepines by the condensation of 1,2-phenylenediamine with ketones catalyzed by BTZ in acetonitrile and under reflux conditions within 10 min in 85–98% yield.  It is noteworthy to mention that by starting from an unsymmetrical ketone, such as 2-butanone (Table 3, entry c), the ring closure occurs selectively only from one side of carbonyl group yielding a single product. Cyclic ketones, such as cyclopentanone and cyclohexanone (Table 3, entries e,f) also reacted effectively to produce the corresponding fused ring benzodiazepines in CH3CN or in the absence of solvent. 
A reasonable pathway for the reaction of diamine with ketone in the presence of supported BTZ is also presented by Fig. 10.
The recovery and reusability of the supported catalysts have been investigated. We have noticed that after the addition of ETOH to the reaction mixture, BTZ can be easily recovered quantitatively by simple filtration. The wet catalyst was recycled (the nature of the recovered catalysts has been followed by NAA, XRD and FTIR spectra), but no appreciable change in the catalytic activity was observed up to four cycles if calcination of reused catalyst was occurred (Fig. 11).

 CONCLUSION
In summary, novel and efficient BTZ was prepared under hydrothermal conditions and characterized by FTIR, XRD, SEM, BET, and TGA, and NH3-TPD analysis. The catalytic activity of BTZ was investigated for synthesis of 1,5-benzodiazepine derivatives by
reaction of o-phenylenediamine with different ketones. The significant advantages of this procedure are short reaction times, high yields, mild reaction conditions and reusability of catalyst for several times.

CONFLICT OF INTEREST
The authors declare that there is no conflict of interests regarding the publication of this manuscript.

ACKNOWLEDGEMENTS
We gratefully thank Shahreza Branch, Islamic Azad University for financial support.