Elsevier

Energy

Volume 199, 15 May 2020, 117438
Energy

Optimization of food waste hydrothermal liquefaction by a two-step process in association with a double analysis

https://doi.org/10.1016/j.energy.2020.117438Get rights and content

Highlights

  • Bio-oil production from alkaline pretreatment and hydrothermal liquefaction is studied.

  • Influences of operating factors are analyzed by Taguchi method and analysis of variance (ANOVA).

  • The highest energy yield is 56.55% and the highest HHV of bio-oil is 25.12 MJ/kg.

  • A highly linear correlation between the effect of Taguchi analysis and F value of ANOVA is observed.

  • TG-FTIR indicates that the bio-oil’s composition under optimal operation is more uniform.

Abstract

Bio-oil production from food waste, consisting of pineapple peel, banana peel, and watermelon peel, is investigated by a two-step process, namely, an alkaline pretreatment process with K2CO3 (10 wt% of the dry feedstock) followed by a hydrothermal liquefaction (HTL) process. Meanwhile, the Taguchi method is introduced to maximize the energy yield of the two-step process. Four parameters in the Taguchi approach are taken into account; they are the pretreatment temperature and time as well as the liquefaction temperature and holding time. The optimal combination of the four parameters gives the highest energy yield of 56.55%. The higher heating value of the bio-oil is 25.12 MJ/kg, yielding a 45.88% improvement when compared to the HHV of the dry-basis feedstock. A double analysis, namely, the Taguchi approach and analysis of variance (ANOVA), suggests that the liquefaction temperature plays the most influential role in the energy yield, and a strong linear relationship (R2 ≈ 0.99) is exhibited between the effect in the Taguchi approach and the F value in ANOVA. The experiments of thermogravimetric analysis coupled with Fourier-transform infrared spectroscopy indicate that the composition of the bio-oil from the optimal operation is more uniform.

Introduction

On account of the rapid growth in the global energy demand, which is accompanied by the environmental problems caused by fossil energy utilization, the hydrothermal liquefaction (HTL) technology for bio-oil production has attracted much interest in recent years. HTL is a thermochemical process that converts wet feedstocks into liquid fuels with subcritical or critical water. In general, the HTL process is carried out at temperatures between 250 and 374 °C accompanied by higher operating pressures from 4 to 22 MPa to maintain the solvents [1], normally with the holding time of 1 min–2 h, in the liquid phase for a highly reactive environment [2,3].

HTL is composed of several complex reactions that occur between the feedstock and water. The HTL reaction mechanism can be divided into three main steps [4,5]: (1) the feedstock is first hydrolyzed (depolymerization) into small fragments (polar oligomers and monomers) via the cleavage of chemical bonds with the elevated temperatures; (2) further decomposition reactions, including dehydration, decarboxylation, and deamination, among these fragments are triggered when the surrounding temperature and pressure reach the desired reaction conditions, thereby generating higher reactive compounds; and (3) these reactive compounds will undergo repolymerization or condensation to form bio-oils or higher molecular weight compounds like char. The crucial operating parameters such as temperature, holding time, and catalyst would have different effects on the liquefaction mechanism. In addition, the process depends on the nature of biomass, which is a complex mixture of carbohydrates, lignin, proteins, and lipids. Consequently, the reaction chemistry and mechanism of biomass liquefaction are also complicated [6,7]. It is difficult and time-consuming to optimize the liquefaction process by understanding the mechanism alone.

The normal design of the experiment methods, including one-factor-at-a-time and full-factorial experiments, can be used to determine the optimal process condition. However, the Taguchi method is developed for experiment design to analyze the impact of various parameters of a process on the mean and variance of performance quality, which determines the proper function of a process [8]. The advantage of this method is that it is not necessary to examine all the combinations of parameters but just a couple of sets of the combination. This can save time and cost to a great extent. In the Taguchi method, an orthogonal array is used for optimizing the different parameters affecting the process and analyzing the sensitivity of each parameter. The Taguchi method has been employed in several different research fields such as gasification [9], wind power [10], thermoelectric generation [11], and pyrolysis [12], and showed good performances. For instance, Sathish et al. [13] studied the optimization of the transesterification process for the production of manilkara zapota methyl ester (MZME). Four parameters of reaction temperature, reaction time, methanol-to-oil molar ratio, and catalyst concentration were considered. The experimental study revealed that the reaction temperature of 50 °C, holding time of 90 min, 6:1 M ratio, and 1 wt% of catalyst concentration gave the optimal process performance, while the methanol-to-oil molar ratio was the most influential parameter governing the biodiesel production. Wang et al. [14] carried out a supercritical conversion process (similar to the HTL process) using sewage sludge in a high-pressure reaction system. They considered four operating conditions of the reaction temperature, residence time, sodium carbonate catalyst content, and feedstock filling ratio. Their Taguchi approach revealed that the reaction temperature and holding time were the dominant factors on the bio-oil yield. The maximum bio-oil yield of 39.73% was attained under the reaction temperature of 375 °C and the reaction holding time of 0 min.

It is known that food wastes are a potential feedstock for bio-oil production from HTL. Tomoaki et al. [15] used cabbages, boiled rice, boiled and dried sardine, butter, and the shell of a short-necked clam to represent the vegetables, rice and bread, animals, shells of eggs, and shellfishes in the food waste, respectively. Their experiments were conducted with three temperatures (250, 300, and 340 °C), three holding times (0.1, 0.5, and 2 h), and without or with a catalyst of sodium carbonate (4 wt%). The highest oil yield of 27.6% was obtained at the reaction temperature of 340 °C, holding time of 0.5 h, and catalyst addition of 4 wt%. Maag et al. [16] studied HTL of food waste with a reaction temperature of 300 °C and a residence time of 60 min. The effects of homogeneous catalyst Na2CO3 and heterogeneous catalyst CeZrOx on bio-oil production and product distribution were compared with each other. The results showed that the food waste catalyzed by CeZrOx could produce bio-oil with higher HHV (31.20 MJ/kg) and energy recovery (38.8%). Aierzhati et al. [17] operated the HTL processes by use of food waste from a campus dining hall, which consisted of high lipid group (salad dressing and cream cheese), high protein group (beef and chicken), and high carbohydrate group (hamburger bun, vegetable, and fruit peels). The effects of reaction temperature (280–360 °C) and residence time (10–60 min) on bio-oil production and product distribution were investigated. The results suggested that the biochemical organization of feedstock significantly affected both the bio-oil yield and different product distributions and properties. The salad dressing exhibited the best performance for bio-oil production with a maximum yield of 78.1% and an HHV of 39.68 MJ/kg. The studies using food wastes and model food wastes as feedstocks are summarized in Table 1.

Though some HTL studies have been conducted, these studies lacked systemic designs for experiments and the appropriate combination of the operating parameters [18,19], especially in using food wastes as feedstocks. It is also known that catalytic pretreatment has a significant effect on bio-oil production [15,20]. Homogeneous alkali salts such as sodium carbonate (Na2CO3) have been widely investigated and reported to improve HTL carbon yield, and this improvement is attributed to suppressing coke formation [16,[21], [22], [23]]. The main purpose of adding Na2CO3 is to reduce acid-catalyzed decomposition of carbohydrates, phenol aldehyde cross-linking reactions, and to enhance aldol condensation reactions. During HTL, if the pH value of the sample slurry drops below 4, dehydration, polymer formation, and char formation are easily triggered. Therefore, a buffering agent is employed to inhibit the formation of high molecular weight compounds and solid products, and to facilitate bio-oil production. Under the same concentrations of K2CO3 and Na2CO3, their pH values are close to each other, implying that they may have similar catalytic effects. However, the effects of pretreatment temperature and duration with K2CO3 on food waste liquefaction have not been investigated yet. So this study uses this catalyst to pretreat the feedstocks. Therefore, the primary objective of this study is to optimize the energy yield of bio-oil from the food waste which is composed of three kinds of fruit peels (pineapple peel, banana peel, and watermelon peel). To achieve this target, a two-step process through the combination of alkaline pretreatment and an HTL process is practiced. The Taguchi approach is introduced to identify the key process parameters where the pretreatment temperature and residence time, as well as the liquefaction temperature and holding time, are taken into account. Moreover, a double analysis, namely, the Taguchi method and ANOVA, is performed to explore the relationship between the effect in the Taguchi approach and the F value in the analysis of variance. Based on the results, the optimal strategy for HTL operation will be obtained which is beneficial to the development of the waste-to-fuel technology.

Section snippets

Material

The three common fruit peels of pineapple peel (Ananas comosus cv. Tainon 17, PP), banana peel (Musa basjoo, BP), and watermelon peel (Citrullus lanatus, WP) were used in this study. The three fruits can be harvested all year round in Taiwan. According to the statistics year book of agricultural production in Taiwan [24], the production amounts of pineapple, banana, and watermelon in 2017 were 553,531, 356,017, and 210,661 tons, respectively. The three peels were collected from the local

Bio-oil yield, higher heating value, and energy yield

The distribution of the products from each case was tabulated in Table 3. Since the objective of this study is to optimize the energy yield of bio-oil, the analysis of products was emphasized on bio-oils.

The profile of the bio-oil yield from the feedstock undergoing the two-step process of alkaline pretreatment followed by HTL is showed in Fig. 3a. Overall, the bio-oil yield is in the range of 28.4–39.6 wt% with the maximum yield occurring at Case 8 (39.6 wt%). It is noteworthy that bio-oil

Conclusions

The optimization of alkaline pretreatment coupling an HTL process, namely, a two-step process, of a food waste composed of three fruit peels (pineapple peel, banana peel, and watermelon peel) has been investigated. The influences of four operating factors of pretreatment temperature (30–100 °C), pretreatment time (30–120 min), liquefaction temperature (300–350 °C), and liquefaction holding time (0–60 min) on the energy yield of the process are analyzed by the Taguchi method and analysis of

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors acknowledge the financial support from the Metal Industries Research & Development Centre, and Ministry of Science and Technology (MOST 106-2923-E-006-002-MY3 and 109-3116-F-006-016-CC1), Taiwan, R.O.C., for this research.

References (57)

  • Yan Wang et al.

    Experimental study of the bio-oil production from sewage sludge by supercritical conversion process

    International Journal of Integrated Waste Management, Science and Technology

    (2013)
  • Tomoaki Minowa et al.

    Oil production from garbage by thermochemical liquefaction

    Biomass Bioenergy

    (1995)
  • A. Aierzhati et al.

    Experimental and model enhancement of food waste hydrothermal liquefaction with combined effects of biochemical composition and reaction conditions

    Bioresour Technol

    (2019)
  • W.-H. Yan et al.

    Composition of the bio-oil from the hydrothermal liquefaction of duckweed and the influence of the extraction solvents

    Fuel

    (2016)
  • T.H. Pedersen et al.

    Continuous hydrothermal co-liquefaction of aspen wood and glycerol with water phase recirculation

    Appl Energy

    (2016)
  • Y. Zhu et al.

    Techno-economic analysis of liquid fuel production from woody biomass via hydrothermal liquefaction (HTL) and upgrading

    Appl Energy

    (2014)
  • Y. Zhu et al.

    Development of hydrothermal liquefaction and upgrading technologies for lipid-extracted algae conversion to liquid fuels

    Algal Research

    (2013)
  • A.B. Ross et al.

    Hydrothermal processing of microalgae using alkali and organic acids

    Fuel

    (2010)
  • Y. Hu et al.

    Production of low-nitrogen bio-crude oils from microalgae pre-treated with pre-cooled NaOH/urea solution

    Fuel

    (2017)
  • Y. Qu et al.

    Experimental study on the direct liquefaction of Cunninghamia lanceolata in water

    Energy

    (2003)
  • LAd Macedo et al.

    Influence of potassium carbonate addition on the condensable species released during wood torrefaction

    Fuel Process Technol

    (2018)
  • W.-H. Chen et al.

    A comprehensive analysis of food waste derived liquefaction bio-oil properties for industrial application

    Appl Energy

    (2019)
  • T.H. Pedersen et al.

    Production of fuel range oxygenates by supercritical hydrothermal liquefaction of lignocellulosic model systems

    Biomass Bioenergy

    (2015)
  • Z. Li et al.

    Alkaline pretreatment and the synergic effect of water and tetralin enhances the liquefaction efficiency of bagasse

    Bioresour Technol

    (2015)
  • H.-M. Liu et al.

    Alkaline pretreatment and hydrothermal liquefaction of cypress for high yield bio-oil production

    J Anal Appl Pyrol

    (2014)
  • H. Mazaheri et al.

    Influence of temperature on liquid products yield of oil palm shell via subcritical water liquefaction in the presence of alkali catalyst

    Fuel Process Technol

    (2013)
  • L. Qian et al.

    Hydrothermal liquefaction of sewage sludge under isothermal and fast conditions

    Bioresour Technol

    (2017)
  • J.-J. Lu et al.

    Investigation on the ignition and burnout temperatures of bamboo and sugarcane bagasse by thermogravimetric analysis

    Appl Energy

    (2015)
  • Cited by (45)

    View all citing articles on Scopus
    View full text