Optimization of yield in the synthesis of 2-Chloroquinoline- 3-carbaldehyde by box - hunter designed experiments via the vilsmeier - haack reaction

60 HNUE JOURNAL OF SCIENCE DOI: 10.18173/2354-1059.2017-0055 Chemical and Biological Science 2017, Vol. 62, Issue 10, pp. 60-65 This paper is available online at OPTIMIZATION OF YIELD IN THE SYNTHESIS OF 2-CHLOROQUINOLINE- 3-CARBALDEHYDE BY BOX - HUNTER DESIGNED EXPERIMENTS VIA THE VILSMEIER - HAACK REACTION Duong Ba Vu Faculty of Chemistry, Ho Chi Minh City University of Education Abstract. 2-chloroquinoline-3-carbaldehyde (QCl) is one of key organic compounds for synthesizing numerous products which possess biological potential. According to Vilsmeier - Haack, QCl was prepared by refluxing a mixture of acetanilide (A) and phosphoryl chloride (P) in N,N-dimethylformamide (D). The yield of this reaction was optimized by Box - Hunter model as a Central Composite (Response Surface) Design for three factors: the molar ratio of A : P : D was 1 : 14 : 3; the temperature of waterpaths for heating refluxed system reached at around 100 o C and the reaction time lasted about 14 hours. In this condition, the optimized yield of QCl synthesis was predicted to be approximately 80.0% in comparison with the actual one of 77.0%. Keywords: Quinoline, least squares, Box - Hunter design, reponse surface design, central composite design. 1. Introduction It is widely considerable that chloro- and carbaldehyde functions are two active sides of a 2-chloroquinoline-3-carbaldehyde (QCl) molecule as a highly activecompound [1]. The former group is vulnerable to leave by substitution reaction and the latter is also easy to be oxidated to carboxylic acids or be condensed with hydrazine, hydrazide, (thio)semicarbazide analogues. Those resulted in products possessing numerous interesting biological properties such as antifungal, antibacterial, and antiprotozoic, antineoplastics as well as antitumour [1-5]. Therefore, QCl is a center compound for many organic syntheses of heterocylic derivatives. Although there are currently many ways to obtain QCl, the Vilsmeier - Haack reaction is the popular process in almost researches. According to this method, a mixture of acetanilide (A) and phosphoryl chloride (P) in N,N-dimethylformamide (D) is refluxed inhot waterpaths for specific time [1, 5]. In fact, many factors, for instance, the kind of reactants, the timeand temperature of reaction, even ratios of the agents, are recognized to Received October 22, 2017. Revised November 16, 2017. Accepted November 23, 2017. Contact Duong Ba Vu, e-mail address: vudb@hcmup.edu.vn Optimization of yield in the synthesis of 2-chloroquinoline-3-carbaldehyde 61 affect the yield of QCl synthesis. In our range of knowledge, there are no any reports associated with optimization in yield of this reaction. In 2005, Ambika Srivastava and his partner used the one-factor-at-a-time (OFAT) experiments to investigate whether the number of mole of P influenced the objective funtion in the fact that other conditions were kept fixed [5]. However, this process exited lots of disadvantages preventing for researchers to have a comprehensive vision of this reaction. Nowadays mathematic techniques have supported many new methods - designed experiments (DE) to tackle this problem. DE is a effective way to assess impacts of two or more factors on a response. In comparison with OFAT, DE requires less the number of experiments to get observations to estimate effect results in reduced variability. DE also provides the information about the interaction between factors. Therefore, the prediction of response is improved in entire factor space. This surface response methodology was applied in Vietnam for many fields such as inorganic materials, biochemical production, extraction of organic compounds from natural products and barely in organic synthesis [6-10]. For all upward sides, in this paper, the yield of QCl synthesis is a response function optimized by Box - Hunter design in variety of three factors: the molar number (n, mmoles) of P; the temperature (T, o C) of waterpaths for heating refluxed system and the reaction time (t, hours). 2. Content 2.1. Experiments * Chemicals Acetanilide C6H5NHCOCH3 (98.0% purity) and phosphoryl chloride POCl3, d=0,945g/mL (99.0%) were purchased from Merck, Germany.N,N-dimethylformamide HCON(CH3)2, DMF, d = 1.68g/mL (99.5%), glacial acetic acid CH3COOH (98.0%), sodium azide NaN3 (99.0%), ethyl acetate CH3COOC2H5 and diethyl ether (C2H5)2O were produced from Xilong Company, China. * Statistical experimental design method Table 1. The limited levels of three-factors response surface design Variables Levels Difference +1.682 +1 0 -1 -1.682 x1 100 95 90 85 80 5 x2 18.73 16 12 8 5.27 4 x3 187.3 160 120 80 572 40 The Box-Hunter surface response design was applied to figure out the optimizing point of target function y with three variables presented in Table 1 and Table 2. The data all were analysed to conform the surface response graph by Mode 5.0 software. The factors (T, t, n and the yield) were coded by x1, x2, x3and y respectively. Table 1 represented their limited levels. All center points in this investigation for three factors were chosen based on the one-factor-at-a-time experiments of [6]. By using waterpath for QCl synthesis, the target compound started to be formed at 85 o C for over 12 hours [6]. Duong Ba Vu 62 It was not cyclic addition when the agent mixture was refluxed at above 100 o C for a long time. In case of high temperature (>100 o C), flamable reduction-oxidation reaction may take place dangerously. * Investigation in the impact of R, T, t by Box - Hunter design QCl was also prepared by the process described in [5] where 10 mmoles of A and 30 mmoles of D were fixed during 20 runs. This is designed to estimate the second order regression modelling. The details of each run were tabulated in Table 1. The effect of T was studied in the range of 85 - 100 o C. The effect of t was investigated from 8 hours to 16 hours. The effect of the last variable, n, was observed from 5.72 mmoles to 187.3 mmoles. The raw production was recrystallized from ethylacetate. Then, the needle yellow substance was dried at the room temperature for 4 hours before its weigh was scaled three times and was determined its range of boiling temperature. That of QCl is about 143 - 145 o C. Next came to qualitative experiments for each production to check the chemical properties of NoQCl. The white precipitation appeared when QCl reacted to NaN3 in DMF/CH3COOH. Table 2. Variables were varied in QCl synthesis No. T t n (volume of P) The weigh of pure QCl (grams) The yield (%) 1 85 8 80 (7.3 mL) 0.20 10.47 2 95 8 80 (7.3 mL) 0.77 40.48 3 85 16 80 (7.3 mL) 0.33 17.02 4 95 16 80 (7.3 mL) 1.08 56.34 5 85 8 160 (14.6 mL) 0.48 25.13 6 95 8 160 (14.6 mL) 1.12 58.59 7 85 16 160 (14.6 mL) 0.83 43.46 8 95 16 160 (14.6 mL) 1.54 80.42 9 80 12 120 (11.0 mL) 0.74 38.69 10 100 12 120 (11.0 mL) 1.38 71.99 11 90 5.27 120 (11.0 mL) 0.51 26.60 12 90 18.73 120 (11.0 mL) 1.09 57.17 13 90 12 5.72 (5.2 mL) 0.28 14.87 14 90 12 187.3 (17.2 mL) 1.21 63.56 15 90 12 120 (11.0 mL) 1.32 69.11 16 90 12 120 (11.0 mL) 1.29 67.70 17 90 12 120 (11.0 mL) 1.16 60.68 18 90 12 120 (11.0 mL) 1.33 69.79 19 90 12 120 (11.0 mL) 1.19 62.56 20 90 12 120 (11.0 mL) 1.20 62.67 Optimization of yield in the synthesis of 2-chloroquinoline-3-carbaldehyde 63 N H O POCl3/ t0 N CHO Cl QCl DMF Scheme 1. Synthetic equation of QCl 2.2. Results and discussion Table 3. The matrices of 20 runs No. xo x1 x2 x3 x1x2 x1x3 x2x3 x1 2 x2 2 x3 2 y (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) 1 + - - - + + + + + + 10.47 2 + + - - - - + + + + 40.48 3 + - + - - + - + + + 17.02 4 + + + - + - - + + + 56.34 5 + - - + + - - + + + 25.13 6 + + - + - + - + + + 58.59 7 + - + + - - + + + + 43.46 8 + + + + + + + + + + 80.42 9 + -1.682 0 0 0 0 0 2.828 0 0 38.69 10 + +1.682 0 0 0 0 0 2.828 0 0 71.99 11 + 0 -1.682 0 0 0 0 0 2.828 0 26.60 12 + 0 +1.682 0 0 0 0 0 2.828 0 57.17 13 + 0 0 -1.682 0 0 0 0 0 2.828 14.87 14 + 0 0 +1.682 0 0 0 0 0 2.828 63.56 15 + 0 0 0 0 0 0 0 0 0 69.11 16 + 0 0 0 0 0 0 0 0 0 67.70 17 + 0 0 0 0 0 0 0 0 0 60.68 18 + 0 0 0 0 0 0 0 0 0 69.79 19 + 0 0 0 0 0 0 0 0 0 62.56 20 + 0 0 0 0 0 0 0 0 0 62.67 By least square method, the regession equation, y, was constructed as a function of x1, x2 and x3: 2 2 2 1 22 3 1 365.52 14.33 8.35 12.09 4.15 8.91 9.85y x x x x x x       (1) Duong Ba Vu 64 For confidence level of 95%, (1) was fitting to the experimental data because F = 2.78 < F0.95(8;5) = 4.9. Thepattern in figure 1 showed the temperature was the greatest contribution for yield of QCl preparation in comparison with others (Figure 1). This can be interpreted due to increase of successful collision among reactant molecules in an environment which had a high vicosity with a three-fold excessive amount of P compared to the amount of D. In cases of other factors, those coefficents for two kinds of variables were nearly the same. The excess of P and the prolonged time aimed at completing reaction to get efficient yield. Figure 1. The effect of T at three points In next step, equation (1) was differentiated by three variables to get a system of three linear equations. Solving this kind of system involved the optimal point. 1 1 1 2 2 2 2 2 3 3 3 9014.33 8.30x 0 1.73x 98.65 1005 13.88 1412 8.35 17.82x 0 0.47 x 4 140 120 83.56%0.61 12.09 19.70x 0 40 x o y T x T C t hoursy t x n m moles n H yy x                                                          Therefore, the yield expected to reach the highest values at approximately 80% in conditions that the molar number of P was 140 mmoles, the reaction mixture was refluxed in boiling waterpaths for 14 hours. The amount of P was more than that in [5] because the electron density in A molecule was less than that in m-methoxyacetanilde due to lack of induced effect of methoxy group. Besides, [5] did not studied the effect of T and t to theyield of reaction. In fact, when QCl was produced in this condition, the yield was in range of 70 - 75% that is approximately near to the optimization point because of some errors from recrystallization of raw product and analytical equipment. Optimization of yield in the synthesis of 2-chloroquinoline-3-carbaldehyde 65 3. Conclusion The Box - Hunter model provided that the yield of QCl synthesis can get about 80% in case of that the molar ratio of A : P : D was 1 : 14 : 3; the temperature of waterpaths for heating refluxed system reached at around 100 o C and the reaction time lasted about 14 hours. This is also an important result for next steps in synthesis of antitụmor ligands and desirable complexes. REFERENCES [1] B. F. Abdel-wahab, R. E. Khidre, and A. A. Farahat, 2012. 2-Chloroquinoline-3- carbaldehydes: synthesis , reactions and applications. Vol., No. I, pp. 211-276. [2] A. Shubin and I. D. Dergachev, 2013. Complex formation of Sn(II) with L-cysteine: an IR, DTA/TGA and DFT Investigation. Spectrochim. ACTA PART A Mol. Biomol. Spectrosc., No. Ii, [3] H. Badwaik, S. Venkataraman, D. Thakur, M. Vinod, and J. Raamamurthy, 2011. Antibacterial Evaluation of Novel 2-Aryl-Quinoline -3- Carbaldehyde Derivatives. Vol. 4, No. 6, pp. 893-895. [4] R. Musiol et al., 2008. Investigating biological activity spectrum for novel quinoline analogues 2: Hydroxyquinolinecarboxamides with photosynthesis-inhibiting activity. Bioorganic Med. Chem., Vol. 16, No. 8, pp. 4490-4499. [5] O. Afzal et al., 2015. A review on anticancer potential of bioactive heterocycle quinoline. Eur. J. Med. Chem., Vol. 97, pp. 871-910. [6] A. Srivastava and R. M. Singh, 2005. Vilsmeier-Haack reagent : A facile synthesis of 2-chloro-3-formylquinolines from N-arylacetamides and transformation into different functionalities, Vol. 44, No. September, pp. 1868-1875. [7] V. C. Zitrom, 1999. One-Factor-at-a-Time Versus Designed Experiments. [8] Le Thi Thanh Huong et al., 2009. Optimization of transesterification of tra fat with KOH/γ-Al2O3 catalyst using response surface methodology. Journal of Science and Technology Development, 12(13), pp.69-76. [9] Nguyen Thi Yen et al, 2013. Optimization of medium components for hyrogen productionof anaerobic bacterial strain Thermoanaerobacterium aciditolerans Trau Dat isolated from vietnam by using response surfacemethodology. Journal of Biology, 35(4), pp.469-476. [10] Ho Ba Vuong et al, 2015. Optimization of Polyphenol Extraction from Guava Leaves by Response Surface Methodology. Journal of Science and Technology Development, 13(7), pp.1144-1152.