Application of Taguchi and response surface methodology for biodiesel production from alkali catalysed transesterification of waste cooking oil.

Reddy, B. Sidda ; Kumar, J. Suresh ; Reddy, K. Vijaya Kumar 等

Introduction

India is looking at renewable alternative fuel sources to reduce its dependence on foreign fuels. India currently imports about 72% of its petroleum requirements and has been hit hard by rapidly increasing cost and uncertainty. Biodiesel used as alternative diesel engine fuel has become increasingly important due to the diminishing petroleum reserves and the environmental consequences of exhaust gases from petroleum-fueled engines. Biodiesel possess several distinct advantages over petro-diesel: high lubricity, non toxic, non hazardous, non-flammable, it comes from renewable sources and as such, it does not contribute to new carbon dioxide emission, it is biodegradable, its combustion products have reduced levels of particulates, carbon monoxide, sulfur oxides, hydrocarbons and soot [1-2].

Numerous studies have been conducted on biodiesel production and emission testing in the past two decades. Most of the current challenges are targeted to reduce its production cost, as the cost of biodiesel is still higher than its petrodiesel counterpart. This opens a golden opportunity for the use of waste or recycled oils as its production feedstock. Everywhere in the world, there is an enormous amount of waste lipids generated from restaurants, food processing industries and fast food shops everyday. Reusing of these waste greases can not only reduce the burden of the government in disposing the waste, maintaining public sewers and treating the oily wastewater, but also lower the production cost of biodiesel significantly [3]. Transesterification (or alcoholysis) is a process of displacement of an alcohol group from an ester by another alcohol. In vegetable oil almost 90-95% is glycerides, which are basically esters of glycerol and fatty acids [4]. Previous publications reported the use of methyl, ethyl, and butyl alcohols for the transesterification of rape oil, sunflower oil, cotton seed oil, peanut oil, soybean oil and palm oil to produce methyl, ethyl and butyl esters. The transesterification were enhanced by the use of potassium hydroxide, sodium hydroxide, sodium methoxide or sodium ethoxide as catalysts [5-7]. Chitra, venkatachalam and Sampathrajan [8] conducted the experiments to maximize the biodiesel production from alkali-catalised transesterification of Jatropha curcus oil. They found the average biodiesel yield of 96% in large-scale study in the Tamil Nadu Agricultural University biodiesel pilot plant. Refaat [9] was studied the feasibility of the production of biodiesel considering the variables affecting the yield and characterizes of the biodiesel from waste/recycled oil. The best yield percentage was obtained using a methanol/oil molar ratio of 6:1, potassium hydroxide as catalyst (1%) and 65[degrees]C temperature for one hour. Zorica Knezevic [10] was studied the effects of various factors on the methanolysis of sunflower oil by a commercial lipozyme lipase from Rhizomucor miehei in a solvent free system using Response surface methodology based on central composite

rotatable design. It would seem that the reaction temperature and the amount of water predominantly determined the conversion process while the methanol/oil molar ratio had no significant influence on the reaction rate. The temperature and amount of water showed negative interactive effects on the observed reaction rate per amount of enzyme. The highest yield of "10.15 mol. [kg.sup.-1]" enzyme was observed at 45[degrees]C with a 6:1 methanol to oil molar ratio and with no added water in the system. Seung Wook Kim [11] was investigated the reusability of immobilized lipases and optimized the molar ratio (methanol/oil) and methanol feeding method. Gui, Lee and Bhatia [12] was studied the preparation of biodiesel fuel from refined palm oil using non catalytic transesterification reaction in supercritical ethanol. The optimum conditions for maximum biodiesel production were studied using design of experiments or specially response surface methodology coupled with central composite design. Saifuddin and Chua [13] determined the optimum conditions for the transesterification of waste cooking oil to produce ethyl ester. Lalita Attanatho [14] studied the factors affecting the synthesis of biodiesel from crude kernel oil using [2.sup.4] (four variables, each with two levels) factorial experimental design. The study showed that the catalyst concentration was the most important factor affecting the methyl ester yield, room temperature was the optimum temperature for the synthesis of biodiesel from crude palm kernel oil with 1% NaOH catalyst, 1:3 mass ratio of methanol to oil, 120 minute reaction time which gave 72.77% production yield and 99.27% methyl ester concentration.

In the present study, the effect of process parameters in the production of methyl ester (biodiesel) from waste cooking oil is evaluated using signal to noise ratio. The optimum process parameters (conditions) for the transesterification of waste cooking oil to produce methyl ester were determined. They were: reaction temperature, reaction time, amount of catalyst and amount of methanol. The second order response surface model is developed for predicting the methyl ester yield. The predicted and measured values are fairly close to each other. Their proximity to each other indicates the developed model can be effectively used to predict the methyl ester yield in transesterification of biodiesel.

Materials and Methods

Preparation of Waste Cooking Oil Methyl Ester (WCME)

The raw material (i.e., waste cooling oil) collected from the several restaurants (in Nandyal) in India. The used cooking oil was filtered to remove food residues and solid precipitate in the oil. During the transesterification process, three operations were studied namely (a) transesterification, (b) phase separation and (c) washing. Fig 1 shows the transesterification process; Common alcohols used in this process are short chain alcohols, most notably methanol and ethanol [15]. The most commonly used catalysts are sodium hydroxide and potassium hydroxide. In this study methanol was used and methyl esters were produced. After the reaction was complete, the reaction products separated into two layers; the ester product formed the upper layer and the by-product glycerine formed the lower layer.

The residual catalyst and unreacted excess alcohol were distributed between the two phases. After separation of the phases, the catalyst and alcohol were washed from the ester with water.

[FIGURE 1 OMITTED]

Taguchi Method

Taguchi techniques were developed by Taguchi and Konishi [16]; these techniques have been utilized widely in engineering analysis to optimize the performance characteristics within the combination of design parameters. Taguchi technique is also power tool for the design of high quality systems. It introduces an integrated approach that is simple and efficient to find the best range of designs for quality, performance and computational cost [17].

In Taguchi technique, three-stages such as system design, parameter design and tolerance design are employed. System design consists of the usage of scientific and engineering information required for producing a part. Tolerance design is employed to determine and to analyze tolerances about the optimum combinations suggested by parameter design. Parameter design is used to obtain the optimum levels of process parameters for developing the quality characteristics and to determine the product parameter values depending optimum process parameter values [18]. In this study, parameter design is implemented to achieve the optimum levels of process parameters leading to maximum biodiesel yield during the transesterification process.

Steps in Taguchi Parameter Design

Taguchi method is a scientifically disciplined mechanism for evaluating and implementing improvements in products, processes, materials, equipment and facilities. These improvements are aimed at improving the desired characteristics and simultaneously reducing the number of defects by studying the key variables controlling the process and optimizing the procedures or design to yield the best results. Taguchi proposed a standard procedure for applying his method for optimizing any process.

Step 1: Determine the quality characteristic to be optimized

The first step in the Taguchi method is to determine the quality characteristic to be optimized. The quality characteristic is a parameter whose variation has a critical effect on process or product quality. It is output or the response variable to be observed. Examples are biodiesel yield, weight, cost, corrosion, target thickness, surface roughness, strength of a structure and electro magnetic radiation etc.

Step 2: Identify the noise factors and test conditions

The next step is to identify the noise factors that can have a negative impact on system performance and quality. Noise factors are those parameters which are either uncontrollable or are too expensive to control. Noise factors include variations in environmental operating conditions, deterioration of components with usage and variation in response between products of same design with the same input.

Step 3: Identify the control parameters and their alternative levels

The third step is to identify the control parameters thought to have significant effects on the quality characteristic. Control parameters are those design factors that can be set and maintained. The levels for each test parameter must be chosen at this point. The number of levels, with associated test levels, for each test parameter defines the experimental region.

Step 4: Design the matrix experiment and define the data analysis procedure

The next step is to design the matrix experiment and define the data analysis procedure. First, the appropriate orthogonal arrays for the noise and control parameters to fit a specific study are selected. Taguchi provides many standard orthogonal arrays and corresponding linear graphs for this purpose.

After selecting the appropriate orthogonal arrays, a procedure to simulate the variation in the quality characteristic due to the noise factors needs to be defined. Taguchi proposes orthogonal array based simulation to evaluate the mean and the variance of a product response resulting from variations in noise factors.

Step 5: Conduct the matrix experiment

The next step is to conduct the matrix experiment and record the results.

Step 6: Analyze the data and determine the optimum levels

After the experiments have been conducted, the optimal test parameter configuration within the experiment design must be determined. To analyze the results, the Taguchi method uses a statistical measure of performance called signal to noise (S/N) ratio borrowed from electrical control theory. The S/N ratio developed by Dr. Taguchi is a performance measure to choose control levels that best cope with noise. The S/N ratio takes both mean and the variability into account. In its simplest form S/N ratio is the ratio of the mean (signal) to the standard deviation (noise). The S/N equation depends on the criterion for the quality characteristic to be optimized. While there are many different possible S/N ratios, three of them are considered standard and is given in below.

Smaller-the-better

Nominal-the-best

Larger-the-best

Step 7: Predicting the performance at these levels

Using the Taguchi method for parameter design, the predicted optimum setting need not correspond to one of the rows of the matrix experiment. This is often the case when highly fractioned designs are used therefore, as the final step; an experimental confirmation is run using the predicted optimum levels for the control parameters being studied.

Response Surface Methodology

Response Surface Methodology (RSM) is one of the Total Quality Management Tools or Techniques that can be used in problem based learning. RSM is a collection of mathematical and statistical techniques that are useful for modeling, analysis and optimizing the process in which response of interest is influenced by several variables and the objective. RSM uses quantitative data from appropriate experiments to determine and simultaneously solve multivarient equations. The response surface methodology comprises regression surface fitting to obtain approximate responses, design of experiments to obtain minimum variances of the responses and optimization using the approximated responses [19].

In statistical modeling to develop an appropriate approximating model between the response 'Y' and independent variables {[x.sub.1], [x.sub.2], ------- [x.sub.n]} in general, the relation ship is written in the form of

Y = f([x.sub.1], [x.sub.2], -------[x.sub.n]) + [epsilon]; (1)

Where the form of the true response function Y is unknown and perhaps very complicated and [epsilon] is a term that represents other sources of variability not accomplished for in Y. Usually [epsilon] includes effects such as measurement error on response, back ground noise, the effect of the other variables and so on. Usually a is treated as statistical error, often assuming it to have a normal distribution with mean zero and variance [[sigma].sup.2].

E(y) = [??] = E[f([x.sub.1], [x.sub.2], -------[x.sub.n])] + E[epsilon] = f([x.sub.1], [x.sub.2], -------[x.sub.n]) (2)

The variables [x.sub.1], [x.sub.2], -------[x.sub.n] in equation (2) are usually called the natural variables, because they are expressed in the natural units of measurements such as degrees, Celsius, pounds/square inch etc. In much RSM work it is convenient to transform the natural variables to coded variables [x.sub.1], [x.sub.2], ------- [x.sub.n], which are usually defined to be dimensionless with mean zero and the same standard deviation. In terms of the coded variables the response function (2) will be written as

[??] = f([x.sub.1], [x.sub.2], -------[x.sub.n]); (3)

is called response surface. In most of the RSM problems, the form of relationship between the response and the independent variable is unknown. Thus the first step in RSM is to find a suitable approximation for the true functional relationship between Y and set of independent variables employed. Usually a second order model is utilized in response surface methodology.

Y = [[beta].sub.0] + [K.summation over (j=1)] [[beta].sub.j][X.sub.j] + [k.summation over (j=1)] [[beta].sub.jj][X.sub.j.sup.2] + [summation over (I<)] [k.summation over (j=2)] [[beta].sub.jj][X.sub.i][X.sub.j] (4)

The [beta] coefficients, used in the above model can be calculated by means of using least squares technique. The second order model is normally used when the response function is not known or non linear.

Experimental Details

The transesterification experiments were performed using 250 g (for each experiment) of waste cooking oil. In this study, methanol was the alcohol of choice and sodium hydroxide (NaOH) as catalyst.

The DOE has been implemented to select the process parameters that could result in maximum biodiesel yield. In this study, the biodiesel yield was investigated by considering the process parameters, reaction temperature, reaction time, amount of Methanol and amount of catalyst. Therefore, a DOE setup was considered, each with three levels and then [3.sup.4] = 81 runs were required in the experiments for four independent variables. But using Taguchi's [L.sub.9] orthogonal array the number of experiments reduced to 9 experiments from 81 experiments. The process parameters used and their levels chosen are given in Table 1.

The [L.sub.9] orthogonal array is shown in Table 2, the Experimental conditions are generated by three levels in the design of experiments [20] and results are presented in Table 3.

Results and Discussion

Effect of Control Parameters on Biodiesel Yield (%) (Analysis of S/N Ratio)

In Taguchi method, the term "signal" represents the desirable value and "noise" represents the undesirable value. The objective of using Signal-To-Noise ratio is a measure of performance to develop products and processes insensitive to noise factors. The S/N ratio indicates the degree of the predictable performance of a product or process in the presence of noise factors. Process parameter settings with the highest S/N ratio always yield the optimum quality with minimum variance. The S/N ratio for each parameter level is calculated by averaging the S/N ratio's obtained when the parameter is maintained at that level. The signal to noise ratio is calculated using the equation (5) for Larger the better. Table 4 shows the S/N ratio's obtained for different parameter levels

S/N = - 10[log.sub.10] (1/n [summation] 1/[Y.sub.i.sup.2]) (5)

Where n is the number of Output data sets which is equal to 9, and [Y.sub.i] is the Output value for the ith data set.

The calculated S/N ratio for four factors on the biodiesel yield (%) of transesterification process for each level is shown in Fig 2.

[FIGURE 2 OMITTED]

From Table 4 and Fig 2, the optimum process parameters found corresponding to he highest signal to noise ratio. The optimum levels found as reaction temperature (level 2), reaction time (level1), amount of methanol (level 1) and amount of NaOH (level 1). The optimum process parameters for the maximum biodiesel yield (%) can be established at:

Reaction temperature : 50[degrees]C

Reaction time : 60[degrees]C

Amount of methanol : 25 grams

Amount of NaOH : 1.25 grams

It is emphasized that these conditions only provide maximum biodiesel yield (%) among the process conditions tested. As shown in Table 4 and Fig 2 reaction temperature is the most dominant parameter on the biodiesel yield (%) followed by amount of NaOH, amount of methanol and reaction time.

Predicting the optimum performance eat these levels

Using the aforementioned data, one could predict the optimum (maximum) biodiesel yield using the process conditions as

* Predicted mean and predicted signal to noise ratio

Predicted mean and S/N ratio has been calculated from the following expression

[[eta].sub.predicted] = [[eta].sub.m] + [k.summation over (j=1)] ([[eta].sub.j] - [[eta].sub.m]) (6)

Where [[eta].sub.m] = grand mean of S/N ratio,

[[eta].sub.j] = mean S/N ratio at optimum level,

k = number of main design parameters that affect the quality characteristics.

Predicted S/N ratio has been calculated from the response Table 4. Table 5 shows the predicted mean and signal to noise ratio for the optimal process parameters.

Response Surface Analysis

The second order response surface representing biodiesel yield (%) can be expressed as a function of process parameters such as reaction temperature (A), reaction time (B), amount of methanol (C) and amount of NaOH (D). The relationship between the biodiesel yield (%) and the process parameters has been expressed as follows:

Biodiesel Yield (%) = [[beta].sub.0] + [[beta].sub.1] (A) + [[beta].sub.2] (B) + [[beta].sub.3] (C) + [[beta].sub.4] (D) + [[beta].sub.5] (A)(B) + [[beta].sub.6] (A)(C) + [[beta].sub.7] (A)(D) + [[beta].sub.8] (B)(C) + [[beta].sub.9] (B)(D) + [[beta].sub.10] (C)(D) + [[beta].sub.11] [(A).sup.2] + [[beta].sub.12] [(B).sup.2] + [[beta].sub.13] [(C).sup.2] + [[beta].sub.14] [(D).sup.2] (7)

From the experimental data for biodiesel yield (%), the response function has been determined after excluding the insignificant terms in un-coded units as:

Biodiesel Yield (%) = -234.132+22.3757(A)-0.210722(B)-13.6732(C)-10.2147(D) -0.228100[(A).sup.2]+0.149728[(C)sup.2]+0.0432000(A)(C) (8)

The adequacy of the developed model can be verified by using [R.sup.2] value after estimating the sum of squares (SS) and mean square (MS). The quantity [R.sup.2] called as coefficient of determination is used to determine the adequacy of the developed model. 0 [less than or equal to] [R.sup.2] = 1. The [R.sup.2] value is the variability in the data accounted for by the model in percentage.

[R.sup.2] = 1 - SS Error/SS total (9)

The coefficient of determination is calculated using the above expression and is 100% for the present investigation, which shows the high correlation that, existing between the experimental and predicted values. The results of ANOVA for the response function, biodiesel Yield is presented in Table 6. This analysis is carried out for a level of significance of 5%, i.e., for a level of 95% confidence. From the analysis of Table 6, it is apparent that, the F-calculated value is greater than the F- table value [21] ([F.sub.0.05, 7, 1] = 237) and hence the second order model response function developed is quite ample.

Equation (8) is plotted in Fig 3(a)-(i) as contours for each of the response surfaces at different hold values of methanol and NaOH (methanol as: 25g, 37.5g and 50g; NaOH as:1.25g, 2.5g and 3.75g). These response contours can help in the prediction of biodiesel at any zone of the experimental domain [19]. It can be seen from Fig 3(a)-(i) that biodiesel yield increases with an increase in reaction temperature up to 56[degrees]C and decreases with an increase of reaction time. It was observed that an increase of amount of methanol, biodiesel yield increases and decreases with the amount of catalyst (NaOH) increases.

[FIGURE 3 OMITTED]

Effect of Various Process Parameters On Biodiesel Yield

Effect of Reaction Temperature

The temperature variations adopted in this study were 40[degrees]C, 50[degrees]C and 60[degrees]C. The results clearly indicate that the maximum ester yield was obtained at 50[degrees]C temperature. The variation in reaction temperature versus ester yield (%) is shown in Fig 4 (a). It clearly shows that the ester yield increases with increase in reaction temperature and further increase in reaction temperature the ester yield decreases. The reaction temperature should always be below the boiling point (65[degrees]C) of methanol.

Effect of Reaction Time

In order to optimize the reaction time, the different reaction times selected for this study were 60, 90 and 120 min. The results clearly indicate that the biodiesel yield increases with the reaction time up to 90 min and further increase in reaction time yester yield decreases. The biodiesel yield was found to be maximum at 90min as shown in Fig 4 (b). The variation in reaction time versus ester yield percentage is shown in Fig 4 (b).

Effect of Methanol Quantity

To optimize the amount of Methanol required for the reaction, experiments were conducted with 25 gms, 37.5 gms and 50 gms Methanol. The results clearly indicate that the optimum concentration of Methanol required for effective transesterification of waste cooking oil was 50 gms. Moreover, it was found that biodiesel yield increases as the concentration of Methanol increases. The variation in the methanol concentration versus ester yield percentage is shown in Fig 4 (c).

Effect of NaOH concentration

The catalyst NaOH concentration variations adopted in this study were 1.25 gms, 2.5 gms, and 3.75 gms. The results clearly indicate that the optimum concentration of NaOH required for effective transesterification was 1.25 gms. From Fig 4(d), it was observed that, if NaOH concentration increases, the yester yield decreases. The variation in NaOH concentration versus ester yield percentage is shown Fig 4 (d).

Validation of experimental results

To predict and verify the improvement of biodiesel yield in the transesterification process, verification tests are used. Fig 5 shows the comparisons of values of biodiesel yield from prediction and from actual experiment. From Fig 5 it is clear that, waste cooking oil methyl ester yield between the experimental and predicted values shows a linear relation ship.

[FIGURE 4(a) OMITTED]

[FIGURE 4(b) OMITTED]

[FIGURE 4(c) OMITTED]

[FIGURE 4(d) OMITTED]

Conclusions

The waste cooking oil methyl ester was produced through transesterification process by varying the different process parameters using Taguchi's orthogonal array. Based on the experimental and analytical results, the following conclusions are drawn

(1) Taguchi parameter design provides a systematic procedure that can effectively identify the optimum biodiesel yield in the transesterification process.

(2) it reduces the process variability using relatively small number of experimental runs and costs to achieve the maximum biodiesel yield.

(3) The effect of process parameters on the biodiesel yield has been evaluated using Taguchi method and optimal process parameters for optimum biodiesel yield have been determined with the help of signal to noise ratio.

(4) The reaction temperature is the most dominant parameter on the biodiesel yield (%) followed by amount of NaOH, amount of methanol and reaction time.

(5) A second order response surface model for biodiesel yield has been developed from the observed data. The predicted and measured values are fairly close, which indicates that the developed model can be effectively used to predict the biodiesel yield in the transesterification process with 95% confidence intervals.

References

[1] Sharma, R. P., 2004, "Prospects and current status of biodiesel as an alternative fuel in India", Proceedings of Indo-Swedish Workshop on Biofuel conversion II, pp. 19-34.

[2] Encinar, J. M., Gonzales, J. F., Rodriguez, J. J., Tejedor, A., 2002, Energy fuels, Vol. 16, pp. 443.

[3] Alcantara, R., Amores, J., Canoira, L., Fidalgo, E., Franco, M. J., Navarro, A., 2000, "Catalytic production of biodiesel from soybean oil, used frying oil and tallow", Biomass and bioenergy, Vol 18, issue No. 6, pp. 515-527.

[4] Quick, G. R., Woodmore, P. J., Wilson, B. T., 1983, "Engine evaluation of linseed oil and derivatives", In: Bagby, M. O, Pryde, E. H. (Eds.), Vegetable Oils Diesel Fuel: Seminar III, ARM-NC-28, US Department of Agriculture: Peoria, IL, Vol. 138.

[5] Fangrui, Ma., and Milford, A., 1999, Hanna Biodiesel production: a review. Bioresource Technology, Vol. 70, pp. 1-15.

[6] Filiz, K., 1999, "Vegetable Oil Fuels: A Review. Energy Sources", Vol. 21, pp. 221-231.

[7] Srivastava, A., Prasad, R., 2000, "Triglyceridesbased diesel fuels. Renewable and Sustainable Reviews", Vol. 4, pp. 111-133.

[8] Chitra, P., Venkatachalam P., and Sampathrajan, A., 2005, "Optimisation of experimental conditions for biodiesel production from alkali-catalysed transesterificatin of Jatropha curcus oil", Energy for sustainable development, Vol 9, No.3, pp. 13-18.

[9] Refaat, A. A., Attia, N. K., Sibak, H. A., El Sheltawy, S. T., ElDiwani, G. I., 2008, "Production optimization and quality assessment of biodiesel from waste vegetable oil" Int. J. Environ. Sci. Tech., Vol.5, No.1, pp. 75-82.

[10] Zorica Knezevic, Nevena Ognjanovic and Dejan Bezbradica, 2008, "Optimization of the production of biodiesel by a commercial immobilized lipase in a solvent-free system using a response surface methodology", J. Serb. Chem. Soc. Vol. 73 No.2 pp. 147-156.

[11] Lee, Jong Ho, Dong Hwan Lee, Jung Soo Lim, Byung-Hwan Um, Chulhwan Park and Seung Wook Kim, 2008, " Optimization of the process for biodiesel production using a mixture of immobilized Rhizopus oryzae and Candida rugosa Lipases", J. Microbiol. Biotechnol, Vol 18, No.12, pp. 1927-1931.

[12] Gui, M. M., Lee, K. T., and Bhatia, S., 2008, "Production of palm oil biodiesel using supercritical ethanol transesterification", International symposium on environmental management: hazardous-environmental management toward sustainability, Thailand, pp. 80-81.

[13] Saifuddin, N. and K. H. Chua, 2004, "Production of Ethyl Ester (Biodiesel) from used Frying Oil: Optimization of Transesterification Process using Microwave Irradiation", Malaysian Journal of Chemistry, Vol. 6, No. 1, 077-082.

[14] Lalita Attanatho, Sukunya Magmee and Peesamai Jenvanitpanjakul, 2004, "Factors Affecting the Synthesis of Biodiesel from Crude Palm Kernel Oil", Proceedings of the Joint International Conference on "Sustainable Energy and Environment (SEE)", Hua Hin, Thailand, pp. 359-361.

[15] Zongba, Li. B. Z, 2007, "Biodiesel production by direct methanolysis of oleaginous microbial biomass", Journal of Chemical Technology and Biotechnology, Vol. 82:775-780.

[16] Taguchi, G., Konishi, S., 1987, "Taguchi methods, orthogonal arrays and linear graphs, tools for quality" American supplier institute. American Supplier Institute, pp. 8-35.

[17] Taguchi, G., 1990, Introduction to quality engineering. New York: Mc Graw- Hill.

[18] Yong, WH., Tarng, YS., 1998, "Design optimization of cutting parameters for turning operations based on Taguchi method". J Mater Process Technol, Vol. 84:122-9.

[19] Montgomery DC, Design and Analysis of experiments, John Willey and sons, New York, 1991.

[20] MINITAB user guide, Minitab inc., 2003-2004.

[21] Miller, I., Freund, JE., 1985, Probability and statistics for engineers. New Delhi: Prentice Hall of India.

(1) B. Sidda Reddy, (2) J. Suresh Kumar, (3) K. Vijaya Kumar Reddy and (4) A. Aruna Kumari

(1) Assistant professor, (2,4) Assoc. Professor, and (3) Professor

(1) Department of Mechanical Engineering, R. G. M. Engineering College, Nandyal-518 501. A.P, India. E-mail: bsrrgmcet@gmail.com.

(2,3,4) Department of Mechanical Engineering, J.N.T.U. H College of Engineering, Kukat pally, Hyderabad, A.P, India

Table 1: Process parameters and their levels.

Controllable                                Levels
parameters                 Units        1      2      3

Reaction temperature (A)   [degrees]C   40     50     60
Reaction time (B)          min          60     90     120
Amount of Methanol (C)     grams        25     37.5   50
Amount of NaOH (D)         grams        1.25   2.5    3.75

Table 2: [L.sub.9] orthogonal array design.

 A     B     C     D

 1     1     1     1
 1     2     2     2
 1     3     3     3
 2     1     2     3
 2     3     1     2
 2     2     3     1
 3     1     3     2
 3     2     1     3
 3     3     2     1

Table 3: Experimental conditions and results.

Filtered     A     B     C      D     Biodiesel   Yield
Oil (gms)                              (grams)     (%)

   250      40    60    25     1.25    164.32    65.73
   250      40    90    37.5   2.5     33.83     13.53
   250      40    120   50     3.75    24.16     9.66
   250      50    60    37.5   3.75    106.32    42.52
   250      50    90    50     1.25    202.98    81.2
   250      50    120   25     2.5     173.98    69.2
   250      60    60    50     2.5     173.98    69.2
   250      60    90    25     3.75    115.99    46.4
   250      60    120   37.5   1.25    111.15    44.46

Table 4: Response table for Signal--To--Noise ratio:
Larger the better.

Level     A        B        C        D

  1     26.23    35.24   35.7455   35.50
  2     35.86    31.38    29.39    32.08
  3     34.361   29.82    31.56    28.53
Delta    9.63    5.42     6.11      7.3
Rank      1        4        3        2

Table 5: Results of the confirmation experiment for biodiesel yield.

Optimal Process                   Experimental          Predicted
parameters                  Mean (%)   S/N ratio   Mean (%)  S/N ratio
                                       (dB)                  (dB)

50[degrees]C-60[degrees]C   94.74%     39.53       100.397   45.9834
-25 g-1.25g