Contents lists available at ScienceDirect Fuel journal homepage: www.elsevier.com/locate/fuel Full Length Article Process parameter optimization of biodiesel production from algal oil by response surface methodology and artificial neural networks Akshay Garg, Siddharth Jain ⁎ Department of Mechanical Engineering, College of Engineering Roorkee, Roorkee, India ARTICLE INFO Keywords: ANN RSM Biodiesel Transesterification Algae oil ABSTRACT Biodiesel production from algae oil at low temperature as well as a comparative study of response surface methodology (RSM) and artificial neural networks (ANN) for the modeling of yield and process parameters was carried out in this research work. Box–Behnken experimental design was adopted and the three process para- meters considered were methanol to oil percentage (v/v) (20–60%), catalyst concentration (0–2 wt%) and re- action time (60–180 min) at constant temperature of 50 °C. The results of the present work indicate that ANN has good predictability as compared to RSM. A significant quadratic regression model with value of R 2 of 0.99 and 0.96 was obtained in case of ANN and RSM respectively. 1. Introduction The price of petroleum products and the environmental concerns about pollution coming from the transport exhausts are increasing ra- pidly. The most feasible way to solve these problems is by using al- ternative fuels. Among alternative fuels, biodiesel is the most promising fuel for the transport sector that is predominantly produced from bio- mass[1]. Biodiesel is the mono-alkyl ester of long-chain fatty acids derived from renewable lipid feedstocks such as edible oils and non- edible oils[2]. Besides minimizing the amount of carbon dioxide, un- burnt hydrocarbons and particulate matter released in the environment, biodiesel is biodegradable and contributes a minimal amount of net greenhouse gases or sulfur to the atmosphere [3]. Biodegradation is degradation caused by biological activity, particularly by enzyme ac- tion leading to significant changes in the material's chemical structure. The biodegradability of biodiesel is dependent on the chemical struc- ture of the fatty acids. Biodegradable fuels such as biodiesels have an expanding range of potential applications and they are environmentally friendly. Therefore, there is growing interest in degradable diesel fuels that degrade more rapidly than conventional disposable [4]. The bio- diesel can be prepared by transesterification of vegetable oils with al- cohol in the presence of the catalyst to form fatty acid alkyl esters and glycerol[5]. The possibilities of production of biodiesel from edible oil resources in India is almost economically not feasible, as primary need is to first meet the demand of edible oil that is already imported. India production of oil seed is around 9.3% of world’s seed production and contributes as the fourth largest edible oil producing country [1–2]. Even then, about 50% of edible oil is imported for accommodating the domestic needs and therefore the non-edible resources like Jatropha, Pongamia, Neem, etc. are currently used for biodiesel production [2]. However, their poor oil yield is further making them an unattractive source for biodiesel production. Consequently, the microalgae have attracted the attention of many researchers as a source of biodiesel due to 30 times more oil content in a short period of time from low-grade land areas as compared to other oilseed crops[6]. Many species of mi- croalgae have high lipid content ranging from 20% to 80% of their dry weight[7]. Many researchers have done work on biodiesel production from algae. Cakirca et al. [8] has done work transesterification of algal oil using CaO based catalyst and a yield of 90% was obtained when the reaction was performed with methanol/microalgae oil molar ratio of 6:1, catalyst amount of 3%, and reaction temperature at 65 °C for 3 h. In this study, a high temperature of 850 °C was used for the preparation of the catalyst. Nematian et al. [9] has reported the transesterification of chlorella vulgaris oil using nano biocatalyst and a conversion of 69.8% (w/w) was reported. Teo el al. [10] has reported the result of transes- terification of algae using nano catalyst. The highest biodiesel yield of 99.0% was obtained for 3% of Ca(OCH 3 ) 2 as catalyst at methanol to oil molar ratio of 30:1 and reaction time of 3 h at 80 °C. On the other hand it was observed that empirical relationships based on modeling of the process are of great importance while optimizing the process para- meters. Response surface methodology (RSM) is an empirical modeling method used to develop the relationship between a set of controllable experimental factors and the observed results[11]. An artificial neural https://doi.org/10.1016/j.fuel.2020.118254 Received 25 February 2020; Received in revised form 17 April 2020; Accepted 27 May 2020 ⁎ Corresponding author. Tel.: +917579469566. E-mail addresses: arthjain2001@gmail.com, siddharthajain@coer.ac.in (S. Jain). Fuel 277 (2020) 118254 0016-2361/ © 2020 Elsevier Ltd. All rights reserved. T