Sea Almond as a Promising Feedstock for Green Diesel: Statistical Optimization and Power Rate Law Based Chemical Kinetics of Its Consecutive Irreversible Methanolysis

*Chizoo Esonye, Okechukwu Donminic Onukwuli, Akuzuo Uwaoma Ofoefule, Cyril Sunday Ume and Nkiruka Jacintha Ogbodo*

### **Abstract**

For successful industrial scale-up and effective cost analysis of transesterification process, presentation of complimentary research data from process optimization using statistical design techniques, chemical kinetics and thermodynamics are essential. Full factorial central composite design (FFCCD) was applied for the statistical optimization of base methanolysis of sea almond (*Terminalia catappa)* seed oil using response surface methodology (RSM) coupled with desirability function analysis on quadratic model. Reaction time had the most significant impact on the biodiesel yield. Optimum conditions for biodiesel yield of 93.09 wt% validated at 92.58 wt% were 50.03°C, 2.04 wt% catalyst concentration, 58.5 min and 4.66 methanol/oil molar ratio with overall desirability of 1.00. Ascertained fuel properties of the FAME were in compliance with international limits. GC–MS, FTIR and NMR characterizations confirmed unsaturation and good cold-flow qualities of the biodiesel. Based on power rate law, second-order kinetic model out-performed firstorder kinetic model. Rate constants of the triglyceride (TG), diglycerides (DG) and monoglycerides (MG) hydrolysis were in the range of 0.00838–0.0409 wt%/min while activation energies were 12.76, 15.83 and 22.43 kcal/mol respectively. TG hydrolysis to DG was the rate determining step. The optimal conditions have minimal error and would serve as a springboard for industrial scale-up of biodiesel production from *T. catappa* seed oil.

**Keywords:** kinetics, methanolysis, optimization, response surface methodology, sea almond

#### **1. Introduction**

The application of biodiesel as an alternative energy source to petrodiesel due to its various established renewable advantages has been reported by many researches [1].

**102**

Prunus

[1] Bal JS. Fruit Growing. Ludhiana, Punjab: Kalyani Publishers; 2006

[2] Chadha KL, Pareek OP. Advances in Horticulture. Vol. 4. New Delhi: Malhotra Publishing House; 1993

[3] Chadha TR. Text Book of Temperate

Fruits. New Delhi: ICAR; 2001

Most importantly the biodiesel production from low cost feedstocks (mostly from agro-waste) that are readily and widely available, with high oil yield, non-food competing and underutilized are key parameters that make them satisfactory to EU sustainable biofuel directives. Various methods have been established as ways of converting vegetable oils into petrodiesel-replaceable-form for application in diesel engines (DE). It is very important to highlight that pyrolysis as thermal degradation of vegetable oil produces more bio-gasoline than biodiesel fuel, micro-emulsion results have been only on short term while dilution of vegetable oil with petrodiesel requires very low concentration of vegetable oil. Transesterification is therefore the major chemical process that involves the conversion of fatty acids or triglycerides in vegetable oils to biodiesel (alkyl esters). Structures of chemical building blocks (CBB) involved in transesterification process are presented in **Figure 1**. Although transesterification of vegetable oils can be conducted with both homogeneous (acid or base) and heterogeneous catalysts, base methanolysis always provide much faster rates [2] and cheaper process [1] and more predominantly applied for industrial purposes and large scale biodiesel production. However, currently the major challenge of biodiesel application as a replacement to petrodiesel is its industrial production sustainability. This can be achieved through detailed established transesterification viability and most importantly feasibility data.

Consequently, the successful scale-up of laboratory results in transesterification requires so much information obtained through optimization and kinetics studies. Hence, for effective cost analysis of transesterification process, a holistic presentation of research data from process optimization using statistical design techniques and knowledge of chemical kinetics are essential in establishing the optimum conditions, feed compositions, degree of conversion and recycling as well as reaction mechanisms. It has been established that one factor at a time (O-F-AT) has obvious challenges of non-reliability of obtained results, non-depiction of the interactive effects of the independent variables and ineffectiveness due to the existence of multiple experimental run [3]. Therefore, many researches on optimization and modeling of transesterification process have been established through the application of such soft computing techniques like response surface methodology (RSM), artificial neural network (ANN) and integrated models (IM) [4]. It is very interesting to write that RSM has undisputable edge over others due to its ability to navigate the design space, flexibility, robustness in establishing the optimum condition with the help of desirability function and capability to minimize the number of experimental runs needed to give adequate evidence for statistically acceptable result [4]. Other obvious advantages are its availability in most statistical software, as an asset in statistical quality control, expression and inferential statistics, reliability, gage repeatability and reproducibility studies and process ability as well as improved grappling output. The integration of RSM with desirability function has been reported to have a high a potential over conventional RSM [5]. In the RSM design of experiment, different types have been applied such as full factorial, fractional factorial, Box-Behken, Placket-Burman, central composite rotatable design (CCRD) etc. However, full factorial central composite design (FFCCD) has the advantage of providing double factor axial points at a fixed distance from the centre and significant replicate points at the centre. This has a resultant effect of providing a better reduced-cost approach in obtaining optimal response with least number of experimental runs.

Also, it has been reported that lack of the vital kinetics data of many nonconventional biodiesel feedstocks possess great challenges on their industrial scale process application, reactor design, simulation and control [1]. Although, many researchers have previously reported the kinetics of base-catalyzed transesterification of conventional feedstocks, those works have dwelled more on the reversible

consecutive mechanisms using complex kinetic models. Such works on sunflower oil ethanolysis [6], jatropha oil methanolysis (Kuma et al., 2011), African pear seed oil [1], mixed crude oil palm oil methanolysis [7] buttress the above point. It is noteworthy that the complexity in kinetic models proposed in the above reports challenges their industrial translation while simplified kinetic models suffice for practical

*Sea Almond as a Promising Feedstock for Green Diesel: Statistical Optimization and Power Rate…*

*DOI: http://dx.doi.org/10.5772/intechopen.93880*

*Structures of chemical building blocks involved in transesterification process. (a) Idealized fatty acid; (b) idealized soap; (c) glycerol; (d) alcohols used in biodiesel production; (e) methyl ester; (f) ethyl ester;*

*(g) generalized ester structure; (h) Trigyceride. (i) Cetane versus ethyl ester.*

**Figure 1.**

**105**

*Sea Almond as a Promising Feedstock for Green Diesel: Statistical Optimization and Power Rate… DOI: http://dx.doi.org/10.5772/intechopen.93880*

**Figure 1.**

Most importantly the biodiesel production from low cost feedstocks (mostly from agro-waste) that are readily and widely available, with high oil yield, non-food competing and underutilized are key parameters that make them satisfactory to EU sustainable biofuel directives. Various methods have been established as ways of converting vegetable oils into petrodiesel-replaceable-form for application in diesel engines (DE). It is very important to highlight that pyrolysis as thermal degradation of vegetable oil produces more bio-gasoline than biodiesel fuel, micro-emulsion results have been only on short term while dilution of vegetable oil with petrodiesel requires very low concentration of vegetable oil. Transesterification is therefore the major chemical process that involves the conversion of fatty acids or triglycerides in vegetable oils to biodiesel (alkyl esters). Structures of chemical building blocks (CBB)

involved in transesterification process are presented in **Figure 1**. Although

transesterification viability and most importantly feasibility data.

experimental runs.

**104**

Prunus

transesterification of vegetable oils can be conducted with both homogeneous (acid or base) and heterogeneous catalysts, base methanolysis always provide much faster rates [2] and cheaper process [1] and more predominantly applied for industrial purposes and large scale biodiesel production. However, currently the major challenge of biodiesel application as a replacement to petrodiesel is its industrial production sustainability. This can be achieved through detailed established

Consequently, the successful scale-up of laboratory results in transesterification requires so much information obtained through optimization and kinetics studies. Hence, for effective cost analysis of transesterification process, a holistic presentation of research data from process optimization using statistical design techniques and knowledge of chemical kinetics are essential in establishing the optimum conditions, feed compositions, degree of conversion and recycling as well as reaction mechanisms. It has been established that one factor at a time (O-F-AT) has obvious challenges of non-reliability of obtained results, non-depiction of the interactive effects of the independent variables and ineffectiveness due to the existence of multiple experimental run [3]. Therefore, many researches on optimization and modeling of transesterification process have been established through the application of such soft computing techniques like response surface methodology (RSM), artificial neural network (ANN) and integrated models (IM) [4]. It is very interesting to write that RSM has undisputable edge over others due to its ability to navigate the design space, flexibility, robustness in establishing the optimum condition with the help of desirability function and capability to minimize the number of experimental runs needed to give adequate evidence for statistically acceptable result [4]. Other obvious advantages are its availability in most statistical software, as an asset in statistical quality control, expression and inferential statistics, reliability, gage repeatability and reproducibility studies and process ability as well as improved grappling output. The integration of RSM with desirability function has been reported to have a high a potential over conventional RSM [5]. In the RSM design of experiment, different types have been applied such as full factorial, fractional factorial, Box-Behken, Placket-Burman, central composite rotatable design (CCRD) etc. However, full factorial central composite design (FFCCD) has the advantage of providing double factor axial points at a fixed distance from the centre and significant replicate points at the centre. This has a resultant effect of providing a better reduced-cost approach in obtaining optimal response with least number of

Also, it has been reported that lack of the vital kinetics data of many nonconventional biodiesel feedstocks possess great challenges on their industrial scale process application, reactor design, simulation and control [1]. Although, many researchers have previously reported the kinetics of base-catalyzed transesterification

of conventional feedstocks, those works have dwelled more on the reversible

*Structures of chemical building blocks involved in transesterification process. (a) Idealized fatty acid; (b) idealized soap; (c) glycerol; (d) alcohols used in biodiesel production; (e) methyl ester; (f) ethyl ester; (g) generalized ester structure; (h) Trigyceride. (i) Cetane versus ethyl ester.*

consecutive mechanisms using complex kinetic models. Such works on sunflower oil ethanolysis [6], jatropha oil methanolysis (Kuma et al., 2011), African pear seed oil [1], mixed crude oil palm oil methanolysis [7] buttress the above point. It is noteworthy that the complexity in kinetic models proposed in the above reports challenges their industrial translation while simplified kinetic models suffice for practical purposes. Consequently, methanolysis reaction has been proposed to constitute three consecutive irreversible stages, more especially by the usual condition of using high methanol to oil ratio (>3:1) which shifts the reaction methyl to the right [8, 9].

*Terminalia catappa*; belongs to *combietaccea* family with meridional Asian origin. It occurs in nature and widespread in the sub-tropical zones of India. It is called sea almond or tropical almond or Indian almond. In Nigeria, it is grown basically for

ornamental purposes [10]. It has been reported that the major works on *Terminalia catappa*, has focused mainly on the investigations of phyto-chemical, biological and medicinal application of its leaves, bark and fruit extracts with little or no attention to its seed oil industrial application [11]. Similar to other almonds like Iranian bitter almond and sweet almond, sea almond contains high amount of oil (>60%) [4, 12]. This is similar in quantity to what is observed in other established viable biodiesel feedstocks such as sunflower, peanut and rape seed [11]. Although empirical nonlinear kinetics model of oil extraction as well as synthesis of transformer oil from seeds of *T. catappa* has been reported [11], process optimization and the kinetics of its seed oil methanolysis based on irreversible model under consecutive mechanism has not been reported. A pictorial representation of the *T. catappa* is shown in **Figure 2**. It is therefore the aim of this study to investigate and establish the optimal conditions, chemical kinetics and thermodynamic data for the production of biodiesel from *T. catappa* (sea almond) for its biofuel application relevance. This research is believed to compliment *T. catappa* seed oil's bio-lubricant potential as previously reported [11]. Additionally, the relevant characterizations through the application of nuclear magnetic resonance, gas chromatographic - mass spectrometry, Fourier transform infrared spectrometry analysis of the biodiesel

*Sea Almond as a Promising Feedstock for Green Diesel: Statistical Optimization and Power Rate…*

All the reagents used were all of analytical grade and purchased from the popular

The ripped fruits were collected from Abakaliki city of Nigeria. They were subsequently washed to remove dirt before the pulp was peeled out to release the kernel. The kernels were placed on solar drier for one (1) week. The seeds were extracted by cracking the kernels. Electric milling machine was used to grind the seeds into microsized meals before being sieved using an electric powered mechanical sieve to obtain a fine size of the meal. The remaining moisture in the sieved ground meal was removed

The oil extraction followed the same method previously applied by the authors [1] but with slight modification. The extracted oil was further degummed by mixing the raw oil with 3 wt% by weight of warm water and the mixture was mechanically agitator coupled with using magnetic stirrer for 30 minutes at a temperature of 60°

The quality of the seed oil was determined in accordance with Association of Official Analytical Chemist [14] method. Other properties such as moisture, viscosity and density content were ascertained by using oven method, Oswald viscometer

C to ensure that the emulsifiers were easily separated from the oil [13].

BriDGe-Head Chemical market in Onitsha, Anambra State Nigeria.

were conducted and reported.

*DOI: http://dx.doi.org/10.5772/intechopen.93880*

**2. Materials and methods**

**2.2 Biomass collection and preparation**

*2.2.1 Sourcing of seeds/seed meal preparation*

**2.3 Oil extraction and degumming**

by further sun drying the meal for a period of 5 days.

**2.4 Physico-chemical characterization of the oil**

**2.1 Reagents**

**107**

**Figure 2.**

*Sea almond fruit biomass, a. the fruit, b. fruit cut section, c. dried fruit pulp, d. inner seed with coat. e. the seed, f. the fruit husk, g. the ground pulp (raffinnate and 600 μm particle size).*

*Sea Almond as a Promising Feedstock for Green Diesel: Statistical Optimization and Power Rate… DOI: http://dx.doi.org/10.5772/intechopen.93880*

ornamental purposes [10]. It has been reported that the major works on *Terminalia catappa*, has focused mainly on the investigations of phyto-chemical, biological and medicinal application of its leaves, bark and fruit extracts with little or no attention to its seed oil industrial application [11]. Similar to other almonds like Iranian bitter almond and sweet almond, sea almond contains high amount of oil (>60%) [4, 12]. This is similar in quantity to what is observed in other established viable biodiesel feedstocks such as sunflower, peanut and rape seed [11]. Although empirical nonlinear kinetics model of oil extraction as well as synthesis of transformer oil from seeds of *T. catappa* has been reported [11], process optimization and the kinetics of its seed oil methanolysis based on irreversible model under consecutive mechanism has not been reported. A pictorial representation of the *T. catappa* is shown in **Figure 2**. It is therefore the aim of this study to investigate and establish the optimal conditions, chemical kinetics and thermodynamic data for the production of biodiesel from *T. catappa* (sea almond) for its biofuel application relevance. This research is believed to compliment *T. catappa* seed oil's bio-lubricant potential as previously reported [11]. Additionally, the relevant characterizations through the application of nuclear magnetic resonance, gas chromatographic - mass spectrometry, Fourier transform infrared spectrometry analysis of the biodiesel were conducted and reported.
