DOI : 10.17577/IJERTV2IS3253
- Open Access
- Total Downloads : 329
- Authors : Nashid Sharif, Md. Zahangir Alam
- Paper ID : IJERTV2IS3253
- Volume & Issue : Volume 02, Issue 03 (March 2013)
- Published (First Online): 20-03-2013
- ISSN (Online) : 2278-0181
- Publisher Name : IJERT
- License:
This work is licensed under a Creative Commons Attribution 4.0 International License
Screening Of Reaction Variables Through Plackett-Burman Experimental Design In Alkylation Of P-Cresol With Cyclohexylchloride
Nashid Sharif, Md. Zahangir Alam*
Department of Applied Chemistry and Chemical Engineering University of Dhaka, Dhaka 1000, Bangladesh
Abstract
A Plackett-Burman experimental design was used to identify the reaction variables that influence the alkylation of p-cresol with cyclohexylchloride in presence of anhydrous aluminium chloride. Reaction variables such as temperature, molar ratio of p-cresol to cyclohexylchloride, amount of catalyst, addition time, stirring time, speed of stirring were changed at two levels and cyclohexyl p-cresol was evaluated as the response. Statistical analysis in this study reveals that temperature, molar ratio of p-cresol to cyclohexylchloride and amount of anhydrous aluminium chloride were identified as the major variables to influence the alkylation of p-cresol with cyclohexylchloride.
Keywords: Plackett-Burmann Experimental Design, alklylation, p-Cresol, Cyclohexylchloride.
-
Introduction
Alkylation of phenol and its derivatives with different alkenes and alcohols has been of interest because of industrial importance of the alkylated phenols [1-5]. Since a large number of factors are involved in any chemical process, a preliminary screening should be performed. Plackett-Burman design helps to identify the most significant variables for a certain system with only few experiments, but it can not give the optimum value for each variable [6-9]. This design has the advantage of limiting the number of experiments that are to be performed. Screening techniques such as factorial designs allow the analyst to select which factors are significant and at what levels. Such techniques are vital in determining initial factor significance for subsequent optimization. The P-B design has been used by several researchers for screening the important variables in pharmaceuticals [10-14], biotechnology research and development [15- 20]. Pluckett-Burman experimental design used in this study to screen out the important variables of alkylation of p-cresol is a 2-run design (n = 12) with 11 variables
(k). The highest (+) and lowest () levels of the variables were chosen based the previous experiments. In this study, P-B experimental design is used for the screening of significant variables of alkylation of p- cresol with cyclhexylchloride in presence of aluminium chloride.
-
Experimental
All Alkylation of p-cresol with cyclohexylchloride in presence of aluminium chloride was done according to the method described elsewhere [21]. P-B experimental design was used to screen out the significant variables that affect the yield cyclohexyl p- cresol.
-
Results and Discussion
Alkylation of p-cresol with cyclohexylchloride in presence of anhydrous aluminium chloride was carried out in random manner. Reaction variables such as temperature, molar ratio of p-cresol to cyclohexylchloride, amount of catalyst, addition time, stirring time, speed of stirring were changed at two levels and cyclohexyl p-cresol was evaluated as the response. Six potential variables were considered to have an influence on the yield and selected for screening experiments. These factors and the selected experimental levels are listed in Table 1. Since there were six factors, a 12-trial Plackett-Burman design was suitable. This design had a nominal capacity of 11 variables or factors. The five unassigned factors (X7 through X11) were used in the computation to get some measure of the experimental error.
Data analysis was carried out as described by Akhnazarova and Katarov [22]. The first row in the design matrix was obtained from Akhnazarova and Katarov [22] for n = 12 and K = 11 and given as + +
+ + + +. The remaining design matrix was generated row (or column)-wise from the first one by moving the elements of the row (or column) to the right
(or down) one position and placing the last element of the first row (or column) in the first position. A third row (or column) was produced from the second similarly and the process continued until row (or column) K was generated. A row of minus signs was added to complete the design (18, 19).
Table 1. Experimental range and levels of independent variables
Variable
(+) Level
(-) Level
Temperature, °C, X1
150
130
PC:CHC, X2
8:1
6:1
Amount of sulphuric acid, % by wt. of PC, X3
15
5
Time of addition, ta, X4
2
1
Time of stirring, ts, X5
2
1
Speed of stirring, X6 rpm
600
200
X7 – X11, Unassigned factors used to calculate standard deviation. Response, Y: % yield of CMP
Trial
Mean
X1
X2
X3
X4
X5
X6
Unassigned Factors
Response yield, %
UFE X7
UFE X8
UFE X9
UFE X10
UFE X11
1
+
+
+
+
+
+
+
61.5
2
+
+
+
+
+
+
+
79.8
3
+
+
+
+
+
+
+
81.7
4
+
+
+
+
+
+
+
92.5
5
+
+
+
+
+
+
+
56.2
6
+
+
+
+
+
+
+
47.3
7
+
+
+
+
+
+
+
39.6
8
+
+
+
+
+
+
+
69.7
9
+
+
+
+
td>
+
+
+
46.5
10
+
+
+
+
+
+
+
79.8
11
+
+
+
+
+
+
+
81.7
12
+
35.2
Sum + S
771.5
417.1
420.1
485.2
388.9
386.5
388.5
387.4
381.2
384.3
392.3
386.3
Sum S
0
354.4
351.4
286.3
382.6
385.0
383.0
384.1
390.3
387.2
379.2
385.2
Sum + S & S
771.5
771.5
771.5
771.5
771.5
771.5
771.5
771.5
771.5
771.5
771.5
771.5
Difference
771.5
62.7
68.7
198.9
6.3
1.5
5.5
3.3
91
2.9
13.1
1.1
Effect
64.29
10.455*
11.45*
33.15*
1.05
0.25
0.92
0.55
1.52
0.48
2.18
0.18
(UFE)2
0.3025
2.3104
0.2304
4.7524
0.0324
Trial
Mean
X1
X2
X3
X4
X5
X6
Unassigned Factors
Response yield, %
UFE X7
UFE X8
UFE X9
UFE X10
UFE X11
1
+
+
+
+
+
+
+
61.5
2
+
+
+
+
+
+
+
79.8
3
+
+
+
+
+
+
+
81.7
4
+
+
+
+
+
+
+
92.5
5
+
+
+
+
+
+
+
56.2
6
+
+
+
+
+
+
+
47.3
7
+
+
+
+
+
+
+
39.6
8
+
+
+
+
+
+
+
69.7
9
+
+
+
+
+
+
+
46.5
10
+
+
+
+
+
+
+
79.8
11
+
+
+
+
+
+
+
81.7
12
+
35.2
Sum + S
771.5
417.1
420.1
485.2
388.9
386.5
388.5
387.4
381.2
384.3
392.3
386.3
Sum S
0
354.4
351.4
286.3
382.6
385.0
383.0
384.1
390.3
387.2
379.2
385.2
Sum + S & S
771.5
771.5
771.5
771.5
771.5
771.5
771.5
771.5
771.5
771.5
771.5
771.5
Difference
771.5
62.7
68.7
198.9
6.3
1.5
5.5
3.3
91
2.9
13.1
1.1
Effect
64.29
10.455*
11.45*
33.15*
1.05
0.25
0.92
0.55
1.52
0.48
2.18
0.18
(UFE)2
0.3025
2.3104
0.2304
4.7524
0.0324
Table 2. Placket-Burman experimental design for screening significant variables.
,
The experimental design and the calculations are illustrated in Table 2. Each of the 12 trials of the design was listed in horizontal lines. The vertical columns labeled X1 through X11 indicated the label of the factor in each trial. In regard to the design, in the 12 trials each factor was at a high + level for 6 trials and at a low () level for 6 trials. The yield for each trial was indicated in the Y column on the right.
The Sum+'s line was then comuted by adding the yield values for all lines where the factor was at a + level. (Example: X1 factor 61.5 + 79.8 + 92.5 +
56.2 + 47.3+79.8 = 417.1). This operation was
continued across the table for all factors, including the five unassigned factors. In a similar way, the Sum's line was computed. The next line simply totaled the Sum+'s and Sums to check to the arithmetic. The next line was the difference between the Sum+s and the Sums for each factor. This represented the total difference in yield for the six trials where the factor was at the plus level, from the six trials where the factor was at a minus level. The last line represented the average effects of the factor at the plus level and was computed by dividing the difference by 6, the number of plus signs in the column. The absolute values of the calculated factor effects related to their relative importance. X3- amount of catalyst was the most important variable.
In order to determine whether a factor effect was significant, experimental error was considered. The minimum value for factor effect to be significant was computed using the five unassigned factor effects X7 through X11. Each unassigned factor effect was squared, totaled, divided by 5, the number of unassigned factors. The square root of this number multiplied by a magic number gave the minimum significant factor effect MIN. The magic number used in this computation (2.57) came from a table of probability points of the t-distribution corresponding to five degrees of freedom (five unassigned factors) and the 95% confidence level.
The effect of each variable was determined according to the following equation.
Exi = (Mi+ Mi)/N where Exi = variable main effect
Mi+ and Mi = response of alkylation when the variable was at high level and low level respectively.
N = is the number of trials divided by two.
In every trial except the 12th, 6 variables are at a high level and 5 variables are at low level.
From the analysis it is observed that temperature, molar ratio of p-cresol to
cyclohexylchloride and amount of catalyst were the most significant variables in the alkylation of p- cresol with cyclohexylchloride. Addition time of cyclohexylchloride to the p-cresol-AlCl3 mixture and stirring time after the addition of cyclohexylchloride and stirring speed either had no effect or an effect so small that it was obscured by the experimental error and interaction effects. Thus P-B design provides an efficient and reliable method for screening several reaction variables of alkylation of p-cresol with cyclohexylchloride with the minimum possible number of experiments.
-
Conclusion
In this study alkylation of p-cresol with cyclohexylchloride was carried out in presence of aluminium chloride and Plackett-Burman experimental design was used to screen out the significant variables for these alkylations. Three variables: temperature, molar ratio p-cresol to cyclohexylchloride and amount of aluminium chloride were identified as significant variables. The highest experimentally yield was found to be 96.9% under the following reaction conditions: temperature 1500C, molar ratio of p-cresol to cyclohexylchloride 8:1, amount of aluminium chloride 15 % by wt. of p-cresol, addition time 2h and stirring time 1h.
10. References
-
Lebedev NN, Chemistry and Technology of Basic Organic and Petrochemical Synthesis, Mir Publishers, Moscow, 1984.
-
A.M. Ravikovich, Antioxidants for minerals and synthetic lubricating oils, Chemistry and Technology of Fuels and Oils, 1984. 11:pp. 68-71.
-
T. Sato, G. Sekiguchi, T. Adschiri and K. Arai, Non- catalytic and selective alkylation of phenol with propan- 2-ol in supercritical water, Chem. Commun., 2001. pp. 1566-1567.
-
A.V. Krishnan, K. Ojha and N.C. Pradhan. Alkylation of phenol with tertiary butyl alcohol over zeolites, Org. Proc. Res. Develop, 2002. pp. 132-137.
-
S. Subramanian, A. Mitra, C.V.V. Satyanarayana and
D.K. Chakrabarty, Para-selective butylation of phenol over silicoaluminophosphate molecular sieve SAPO-11 catalyst, Appl. Catalysis A: General, 1997, pp. 229-240.
-
R.L. Plackett and J.P. Burman, The design of optimum multifactorial experiments, Biometrika, 1946, pp. 305-325.
-
M. DeMeo, M. Laget, R. Phan-Tan-Luu, D. Mathieu and G. Dumenil, Application of experimental design for optimization of medium and culture conditions for fermentation, Bioscience, 1985, pp. 99-102.
-
X. Yu, S.G. Hallett, J. Sheppard and A.K. Watson, Application of Plackett-Burman experimental design to evaluate nutritional requirements for the production of Colletorichum coccodes spores, Appl. Microbiol. Biotechnol., 1997, pp. 301-305.
-
J.L. Goupy, Methods for Experimental Design, Principles and Applications for Physicists and Chemists, Elsevier, Amsterdam, 1993.
-
G. Srinubabu, B.V.V. Ratnam, A.A. Rao and R.N. Rao, Development and validation of LC-MS/MS method for the quantification of oxcarbazepine in human plasma using an experimental design, Chemical & Pharmaceutical Bulletin, 2008. pp. 28-33.
-
J. Gabrielsson, M. Sjoestroem, N.-O. Lindberg, A.-
C. Pihl and T. Lundstedt, Robustness Testing of a Tablet Formulation Using Multivariate Design, Drug Development and Industrial Pharmacy, 2006, pp. 297- 307.
-
A. Vatanara, A.R. Najafabadi, K. Gilani, R. Asgharian, M. Darabi and M. Rafiee-Tehrani, A Plackett – Burman design for screening of the operation variables in the formation of salbutamol sulphate particles by supercritical antisolvent, J. Supercritical Fluids, 2007, pp. 111-116.
-
B. Rambali, L. Baert, D. Thone and D.L. Massart, Using experimental design to optimize the process parameters in fluidized bed granulation, Drug Development and Industrial Pharmacy, 2001, pp. 47-55.
-
D. Rege Bhagwant, J. Gawel and J.H. Kou, Identification of critical process variables for coating actives onto tablets via statistically designed experiments, International Journal of Pharmaceutics, 2002, pp. 87-94.
-
S.B. Imandi, B.V.V. Ratnam, S.S. Rao, B.S. Ramalakshmi and G.H. Rao, Application of statistical experimental designs for the optimization of medium constituents for the production of citric acid from pineapple waste, Bioresource Technology, 2008, pp. 4445-4450.
-
Z.J. Xiao, P.H. Liu, J.Y. Qin and P. Xu, Statistical optimization of medium components for enhanced acetoin production from molasses and soybean meal hydrolysate, Applied Microbiology and Biotechnology, 2007, pp. 61-68.
-
Y.R. Abdel-Fattah, H.A. El-Enshasy, N.A. Soliman and H. El-Gendi, Bioprocess development for production of alkaline protease by Bacillus pseudofirmus Mn6 through statistical experimental designs, Journal of Microbiology and Biotechnology, 2009, pp. 378-386.
-
M. Venkateshwar, K. Chaitanya, M. Altaf, B. Hameeda and M.G. Redd, Evaluation of nitrogenous media components by Plackett – Burman statistical design for -D-fructofuranosidase production by Saccharomyces sp. strain GVT263, Canadian Journal of Microbiology, 2009, pp. 405-409.
-
J. Ren, W.T. Lin, Y.-J. Shen, J.-F. Wang, X.-C. Luo and M.-Q. Xie, Optimization of fermentation media for nitrite oxidizing bacteria using sequential statistical design, Bioresource Technology, 2008, pp. 7923-7927.
-
R. Aravindan and T. Viruthagiri, Sequential optimization of culture medium composition for extracellular lipase production by Bacillus sphaericus using statistical methods, Journal of Chemical Technolonlogy and Biotechnology, 2007. pp. 460-470.
-
M.Z. Alam, A. Alam, M. Kamruzzaman, S. Kurihara and M. Saha, Development of a mathematical model by means of experimental design for alkylation of m-cresol with cyclopentene, Chemical Engineering Journal, 2008. pp. 598-602.
-
Akhnazarova S. and V. Katarov, Experiment Optimization in Chmistry and Chemical Engineering, MIR Publishers, Moscow, 1982.
Dedication
The authors would like to dedicate this research article to the memory of late Professor Manoranjan Saha of Applied Chemistry and Chemical Engineering, University of Dhaka, Bangladesh who was the designer of this research.