Elsevier

Renewable Energy

Volume 133, April 2019, Pages 11-22
Renewable Energy

Upgrading bio-oil by catalytic fast pyrolysis of acid-washed Saccharina japonica alga in a fluidized-bed reactor

https://doi.org/10.1016/j.renene.2018.09.103Get rights and content

Highlights

  • Catalytic pyrolysis of acid-washed S. japonica was carried out in a bubbling fluidized-bed reactor.

  • Liquid yield showed an increase in the order raw S. japonica < 1% AW < 5% AW.

  • Acid-washed sample using HZSM-5 catalyst resulted in 35.37–39.05 wt% liquid yield.

  • The HHVs of bio-oil from catalytic pyrolysis of S. japonica sample was in the range 24.98–27.19 MJ/kg.

Abstract

Macroalgae contain significant amounts of inorganic compounds (K, Na, Mg, and Ca) that cause problems during thermal processing (ash fouling or agglomeration of char and bed material). The macroalga Saccharina japonica (S.J) was demineralized with diluted acid solution to remove inorganic species before using this matter for fast pyrolysis in a bubbling fluidized-bed reactor. The effect of acid washing on the product yield and the quality of bio-oil was systematically investigated. When the pyrolysis temperature increased from 400 to 500 °C, the bio-oil yield was 39.70–45.36 wt%. The major compounds in the bio-oil were levoglucosan and di-anhydromannitol. The fractional catalytic pyrolysis of pre-treated S. japonica using HZSM-5 catalyst (calcined at 550 °C) was investigated in fluidized-bed reactor under the same conditions. Pyrolysis of the pretreated sample with catalyst resulted in 35.37–39.05 wt% liquid yield, which was lower than that from using silica sand. Using HZSM-5 catalytic pyrolysis bio-oil showed a decrease in dianhydromannitol and 2-furyl methyl ketone, and an increase in aromatic compounds such as derivatives of phenol, indole, and naphthalene. The effect of pretreatment by acid washing, and the effect of this catalyst on the pyrolysis product distribution and composition were clarified in relation to previous work.

Introduction

Macroalgae offer great promise as feedstock for biofuel production. The production of biofuel from macroalgae based on thermal conversion via pyrolysis has received a lot of attention as a promising biofuel technique. The pyrolysis bio-oils have many advantages, and they are considered an attractive replacement for fossil fuel [1,2]. However, the bio-oils produced from both macroalgae and terrestrial biomass feedstock have high moisture content and a much larger amount of oxygen than fossil fuels do [3]. The presence of these compounds can cause variation in the bio-oil properties (i.e., high acidity, corrosion of reactors, instability, high viscosity, low heating value, and immiscibility with conventional hydrocarbon fuels) and thus the reduce quality of the bio-oil [3]. Consequently, upgrading of bio-oil is necessary for its direct use as transportation fuel. The main purposes when upgrading bio-oil are reduction or removal of oxygen, and the cracking of high molecular weight components into smaller ones. Two main approaches have been presented in previous studies for the partial or total removal of oxygen atoms: catalytic cracking and catalytic hydrotreatment [[4], [5], [6], [7]]. Upgrading of bio-oil can be performed simultaneously with fast pyrolysis in the presence of catalyst, and this combined process is known as catalytic fast pyrolysis.

Catalytic fast pyrolysis is an updated pyrolysis method by which biomass can be converted into higher quality bio-oil by upgrading the pyrolysis vapors before condensation via a series of reactions such as dehydration and decarboxylation at atmospheric pressure [8,9]. Under catalytic upgrading, the properties of the bio-oil are elevated by removing oxygen in the forms of CO2, CO, and H2O [9]. This technique has received the attention of many researchers in recent years. Huang et al. reported the conversion of pine sawdust to advanced biofuel over the catalyst HZSM-5 using a two-stage catalytic pyrolysis reactor, showing an increase in the HHV of the bio-oil from 21.29 MJ/kg (for non-catalyst) to 24.80 MJ/kg (for HZSM-5 catalyst). The author concluded that the catalytic pyrolysis process has a good potential for production of advanced biofuel [10]. The catalytic pyrolysis of algal biomass over acid zeolite catalyst in a fixed-bed reactor for hydrocarbon production was investigated by Suchithra et al. [11]. The author reported that the catalytic fast pyrolysis over HZSM-5 was reduced nitrogen and oxygen content via deoxygenation and denitrogenation reactions in bio-oil. Lorenzetti et al. [12] also studied the effect of HZSM-5 on the pyrolysis of lignocellulosic and proteinaceous biomass. They found that HZSM-5 was effective in producing the aromatic hydrocarbons and in decreasing the amount of O and N-containing compounds.

In addition to organic compounds, macroalgae also contains inorganic compounds (mainly alkali and alkaline earth metals such as K, Na, Mg and Ca), which have been found to have catalytic effects during pyrolysis [[13], [14], [15]]. Although these components are known natural catalysts within the biomass structure, these are also the main cause of problems during thermal processing such as ash fouling or agglomeration of char and bed material (in fluidized-bed reactor) [13,15]. This problem, therefore, might limit the production of bio-fuels from macroalgae using continuous processes like pyrolysis or gasification [15]. Hence, it is desirable to develop a pre-treatment method for removing inorganic compounds from macroalgae before subjecting this bio material to thermal processing.

Different from other previous research, for this work, we used dilute acid washing as the method for pre-treating S. japonica macroalga sample and then the fast pyrolysis and catalytic fast pyrolysis were studied using acid-washed S. japonica as raw material. Firstly, the fast pyrolysis experiments of acid-washed S. japonica were carried out in a bubbling fluidized-bed reactor using silica sand as fluidized-bed material, with nitrogen as a carrier gas, to investigate the effect of acid washing on the product distribution and bio-oil composition. Secondly, experiments with HZSM-5 as a bed material were performed to study and clarify the catalytic influence on the pyrolysis of acid-washed S. japonica. Then these results were compared to those obtained from pyrolysis of the original S. japonica in our previous work [16].

Section snippets

Samples and catalyst preparation

S. japonica biomass was ground with a knife mill to 300–500 μm in length. The moisture and ash contents of S. japonica samples as determined in a previous study were 6.90 and 20.21 wt%, respectively [16,17]. The biomass samples were dried at 105 °C for 12 h before being used in the experiments.

Nitric acid (HNO3) was used as diluted acid solution to demineralize inorganic species (mainly Na, K, Ca, and Mg) from S. japonica samples. Specifically, 1 g of biomass sample was immersed in 10 mL of a

Material characterization

The characteristics of the acid-washed S. japonica samples were analyzed and compared to those of the raw material in previous work [17], as presented in Table 1. After acid washing of the S. japonica sample, the ash content decreased significantly from 20.21 to 7.23 wt% (for S.J-1% AW) and to 3.14 wt% (for S.J-5% AW). This represented 64.23 and 84.46% reductions in inorganics. The removal rates of Na, K, Ca, Mg and P were (78.9, 70.4, 5.0, 47.5, and 34.2 wt%), respectively, for S.J-1% AW, and

Conclusions

The effect of acid washing on the product yield and the quality of bio-oil were systematically investigated. The liquid yield showed an increase in the order original S. japonica < S.J-1% AW < S.J-5% AW, and the char yield exposed a decrease in the order original S. japonica > S.J-1% AW > S.J-5% AW. When the pyrolysis temperature was increased from 400 to 500 °C, the bio-oil yield was between 39.70 and 45.36 wt%. The major components of the bio-oil were levoglucosan and di-anhydromannitol. In

Acknowledgements

This work was financially supported by the Ministry of Oceans and Fisheries of the Republic of Korea (Project No. 20140559).

References (40)

Cited by (0)

View full text