A single step non-catalytic esterification of palm fatty acid distillate (PFAD) for biodiesel production

Abstract

In this work, the single step method for non-catalytic esterification of palm fatty acid distillate (PFAD), which is readily applicable to actual production of biodiesel, was investigated. In this method, the esterification reaction is accomplished in a single step by ensuring water-free reaction conditions and the acid value is reduced to below 0.5 (mg KOH/g) which has not been possible in previous methods. The reaction was completed (<0.5 mg KOH/g) within 180 min at relatively high temperature (>250 °C) enough to be above boiling point of water and at moderate pressure (0.85–1.20 MPa) without any catalyst. The effects of temperature, methanol feed rate and pressure on a semi-batch reaction were investigated and the optimal values of these variables were found (temperature: 290 °C, pressure: 0.85 MPa, feed rate: 2.4 g/min). The acid value was reduced from 191.4 to 0.36 (mg KOH/g) just in 180 min at these conditions. From the kinetic study on non-catalytic esterification of PFAD, it was found that the activation energy is 17.74 kJ/mol and the frequency factor is 2.12 min⁻¹.

Introduction

As an alternative diesel fuel, biodiesel is made out of renewable biomass such as vegetable oils, animal fats, and waste grease. It is well known that the utilization of biodiesel is carbon–neutral and contributable to the reduction of CO₂ emission [1], [2]. Due to its biodegradability, nontoxic property and low emission profiles, biodiesel has attracted lots of attention and has been considered environmentally beneficial [3], [4]. Because biodiesel contains no sulfur but about 11% of oxygen (w/w), we can expect SOx free emission and a significant reduction of pollutant emissions such as unburned hydrocarbons, carbon monoxide and particulate matters compared with petroleum-derived diesel fuel [5], [6]. In addition to the environmental benefits, primary advantage of biodiesel is that none- or few modifications of current diesel engine system are needed to adopt biodiesel blends or even pure biodiesel [7].

In general, biodiesel, fatty acid methyl ester (FAME), is efficiently produced from the transesterification of vegetable oil or animal fats or from esterification of fatty acids with short chain alcohols in the presence of homogeneous or heterogeneous alkali- and/or acid–based catalysts [8]. Among catalytic processes, alkali-catalyzed transesterification reaction is much faster than acid-catalyzed transesterification and is popular in commercial production [9], [10]. But alkali-catalyzed transesterification is suitable only for biodiesel production from feedstock containing low level of free fatty acid (FFA) such as refined vegetable oils. For efficient alkaline transesterification reactions, the feedstock should contain no more than 1 wt% FFA [11]. If the FFA level is too high, the soap generated during the transesterification inhibits separation of glycerol from the methyl esters after the reaction resulting in emulsion formation during the water wash and in induction of significant loss and poor quality of FAME product in the products [12]. Therefore, some low quality feedstock which is unrefined and much cheaper than the refined oil such as used cooking oils (2–7% FFA), animal fats (5–30%), palm fatty acid distillate (PFAD, 85–95%) and trap grease (−100%) is definitely unavailable for alkaline transesterification processes.

Several two-steps processes for the low-cost feedstock containing high content of free fatty acid (FFA) were then proposed [13], [14], [15], [16], [17], [18], [19]. In these processes, the first step is reduction of FFA content in the feedstock by esterification with methanol and acid catalyst such as sulfuric acid as a pre-treatment. The second step consists of transesterification process in which triglyceride portion of the feedstock reacts with methanol and base catalysts, usually sodium or potassium hydroxide, to form ester and glycerol. A major disadvantage of homogeneous catalyzed esterification reaction involving strong acid such as sulfuric acid is the difficulty in catalyst recovery and treatment, which generates a large amount of waste water, increasing the overall operation cost of the process. Furthermore, because the esterification process requires the equipment to be made of high-priced anti-corrosive material and multiple reaction stages to convert FFA to FAME sufficiently so that the product is suitable for the transesterification (FFA < 1%), the capital cost for the process is estimated not as competitive as traditional transesterification processes using refined oils.

Recently, heterogeneous acid catalysts have been more widely favored over homogeneous ones since they are more separable and thus easier to recover. Especially there have been various studies on independent esterification processes without requiring transesterification based on the heterogeneous catalysts [20], [21], [22], [23], [24], [25], [26], [27]. However, processes employing heterogeneous catalysts have not been widely commercialized. Competitiveness of those processes may be improved by increase of stability of catalysts during long-term operation and by application of efficient treatment methods for wastes generated during the catalyst-regenerating process.

There have been only a few studies on non-catalytic esterification and/or transesterification reactions which lead to much simpler purification and environmentally friendly processes [28], [29], [30], [31], [32], [33]. Diasakou et al. [28] investigated the non-catalytic thermal transesterification of soybean oil with methanol and performed study on reaction kinetics. Kusdiana and Saka [29] reported the effects of molar ratio of reactants and reaction temperature on the catalyst-free transesterification of rapeseed oil and proposed a simple method for the analysis of reaction kinetics. Yujaroen et al. [30] investigated the effects of temperature, molar ratio of methanol to fatty acid components and water content in the feed on the esterification of palm fatty acid distillate (PFAD). Most of these studies were conducted under pressurized conditions, i.e., supercritical or subcritical conditions of methanol. However the processes revealed in these works are not easily applicable to actual production of biodiesel due to significantly high production and capital costs required. In addition, these processes require severe operating conditions such as high pressure, high temperature and high molar ratio of methanol including uncertain safety aspects. On the other hand, Yamazaki et al. [31] studied the non-catalytic alcoholysis of sunflower oil for biodiesel production in a semi-batch bubble column reactor under atmospheric pressure. They analyzed the effects of reaction temperature, methanol feed flow rate, pressure, stirring rate and initial feed volume on the transesterification reaction. But, results of any in-depth studies which quantify the effect of operating conditions on the non-catalytic atmospheric alcoholysis were not shown in their work. Joelianingsih et al. [32] performed a kinetic study on the non-catalytic transesterification of palm oil in a semi-batch bubble column reactor at atmospheric pressure. They also investigated the effects of reaction temperatures on the rate constant, conversion, yield of methyl esters (ME) and composition of the reaction product. In another work by Joelianingsih et al. [12], the non-catalytic methyl esterification of oleic acid in a bubble column reactor was investigated and a kinetic study on the esterification was also revealed. However, the atmospheric transesterification and esterification investigated in these studies have a serious problem of significant decrease in yield of methyl ester only a small portion of which is recovered into reaction product. The yields of methyl ester revealed in the transesterification [32] and the esterification [12] are no more than 27.43% and 48.50%, respectively. Also, the conversions in these works only amount to 55.07% and 58.50%. It seems that the reason for such a low reactivity is a low solubility of methanol in the liquid phase at relatively high temperature (200–290 °C) and atmospheric pressure. The low yield and the conversion of the atmospheric reactions revealed in the works make it impossible for the processes to be applied to actual biodiesel production.

In the present work, a novel method for the non-catalytic esterification of fatty acids, especially of palm fatty acid distillate (PFAD), is developed. In this method, the acid value of the product is reduced to below 0.5 (mg KOH/g) which has not been reached by previous works within 180 min just in a single step reaction. For this reason the proposed method is readily applicable to real biodiesel production. By keeping the reaction condition at an optimal water-free range, i.e. at high enough to be above boiling point of water and also at below bubble points of methyl esters, not only reaction is completed beyond equilibrium but also no loss of methyl ester yield is achieved. The effects of temperature, pressure and methanol feed rate (i.e., molar ratio of methanol to fatty acids) on reactivity in a semi-batch CSTR (Continuously Stirred Tank Reactor) were also investigated. From the kinetic study on the non-catalytic esterification of PFAD, values of activation energy and frequency factor for the reaction were determined as well. Meanwhile, PFAD, used as raw material of the esterification reaction in this work, is a byproduct being inevitably generated in purification process of palm oil refinery, and so the price of PFAD is much cheaper than other refined oils which are major feed stocks for most of current biodiesel plants. Also, because PFAD consists of 85–95% fatty acids and 5–15% triglycerides both of which are available for biodiesel production, recently there have been a few trials that directly utilize PFAD as a feedstock for biodiesel production [27], [30], [33], but none of which have possibility of leading to actual production in commercial scale. The process for the non-catalytic single-step esterification of PFAD proposed in this work is readily applicable to actual biodiesel production and can be one of the most competitive processes due to its simplicity, excellent reaction yield and use of low-priced feedstock.