Behavior of selected hydrolyzed and dehydrated products during hydrothermal carbonization of biomass

Highlights

• An innovative slurry sampling system was designed for hydrothermal reaction system.

• Elemental carbon increases rapidly after 90 min for HTC cellulose at 230 °C.

• The liquid state reactions continued for hours for HTC of poplar and straw.

• Selective hydrolyzed compounds’ degradation follow first-order reaction kinetics.

Abstract

In this study, effects of reaction temperature and reaction time on both solid hydrochar and HTC process liquid products were studied for hydrothermal carbonization (HTC) of cellulose, wheat straw, and poplar. A novel slurry sampling system was designed and used with an 18.6 L Parr reactor for 0–480 min in 200, 230, and 260 °C. Sugars (sucrose, glucose, and fructose), HMF, and furfural were found maximum in lower HTC temperature and time. However, they degrade following first order degradation kinetics. Activation energies of total sugars (glucose, fructose, sucrose, and xylose), furfural, and HMF for straw and poplar were 95–127, 130–135, and 74–90 kJ mol⁻¹, respectively and individuals were lower for HTC of cellulose than others. Organic acids (acetic acid, formic acid, and lactic acid) and phenolic compounds (phenol, catechol, and guaiacol) were increasing with higher HTC severity.

Introduction

In current practice, hydrothermal carbonization (HTC) is considered as one of the most effective thermochemical pretreatment processes, where biomass is treated with hot compressed water (180–280 °C) for 5 min to 8 h (Reza et al., 2013a, Reza et al., 2014a, Reza et al., 2014b, Reza et al., 2014c, Reza et al., 2014d). During HTC, volatile oxygen-rich compounds of biomass are removed from the structure, leaving behind stable carbon-rich micro–nano particles (hydrochar) (Jin, 2014). Subcritical water has maximum ionic product in temperature range of 200–280 °C (Bandura and Lvov, 2006). Hydrolysis of the extractives, hemicellulose, and cellulose probably occur first, followed by dehydration, decarboxylation, condensation, polymerization, and aromatization in the liquid phase (Funke and Ziegler, 2010, Reza et al., 2014a). Although overall reaction chemistry follows mainly the stated reaction paths, each of the individual HTC reaction has their own reaction kinetics and is probably catalyzed by one another (Reza et al., 2013b). Moreover, liquid phase HTC reactions are responsible for the production of so-called liquid biocrude (Reza et al., 2014a, Reza et al., 2014d), which is considered as the primary precursor of solid hydrochar. Thus, quality of hydrochar heavily depends on the composition of HTC process liquor (Libra et al., 2011).

Hydrolyzed and dehydrated products during HTC like HMF, furfural, phenol, and their derivatives are potentially toxic and hazardous products, which make HTC process liquor as waste water and thus, requires proper treatment prior to discharge into the environment, even after successive recycling (Uddin et al., 2013). Such toxic volatile organic compounds (VOCs) are reported in the solid hydrochar especially for lower HTC temperatures (e.g., 200–230 °C) (Becker et al., 2013). Presence of selective VOCs in hydrochar prohibits its use for soil amendment (Libra et al., 2011). It is also reported in previous studies that with the increase of HTC temperature (e.g., 250–270 °C) and a prolonged reaction time (6–8 h) many of the VOCs were not observed in hydrochar (Becker et al., 2013). This might be one possible reason for practicing HTC at higher temperature for a prolonged time, which unfortunately also makes HTC economically less viable. Recently, Wirth and Mumme (2013) reported on anaerobic digestion of high strength HTC process liquor for biogas production, which can potentially contribute to the overall economics besides treating waste water. However, high capital cost and proof-of-concept (for HTC process liquid) are hindering anaerobic digestion of such high strength process liquid.

In contrary, many researchers reported HMF, furfural, lactic acid, levulinic acid, levoglucosan, and other hydrolyzed products as platform chemicals or building blocks (Bozell et al., 2000; Jin, 2014; Reza et al., 2014d, Yan et al., 2013). HMF can be dehydrogenated into DHMF and then liquid alkanes in presence of catalysts (Olcay et al., 2013). Furfurals can be used for perfumes and preservatives, while levulinic acid and levoglucosan can be converted into liquid fuels as well (Win, 2005). Lactic acid can be converted into poly-lactic acid (PLA), which is well known as biodegradable plastic (Oda et al., 1997). Thus, hydrolyzed and dehydrated products have market value at their semi-pure to pure form (>70%). Hoekman et al. (2013) reported that up to 4–10% of the dry feedstock can be found in HTC process liquid as valuable hydrolyzed intermediates after 30 min of reaction time for 215–255 °C for various feedstocks. Reza et al. (2014c) also reported that up to 18% of dry maize silage can be found as furfural, HMF, lactic acid in HTC process liquid treated at 200 °C for 20 min. Both literatures also reported that many of the valuable hydrolyzed products degrade with higher HTC temperature and longer reaction time. So, knowledge about degradation chemistry and kinetics of hydrolyzed intermediates within the liquid phase of specific feedstock are necessary for a comprehensive HTC process and product optimization.

The objective of this work was to study HTC process liquid obtained by varying HTC reaction conditions. One of the main problems of such study is the lack of sampling scope, which allows taking samples during the HTC reaction. Most of the researchers used 100–2000 ml batch pressurized reactors (Becker et al., 2013, Becker et al., 2014, Hoekman et al., 2011, Hoekman et al., 2013, Reza et al., 2014a, Reza et al., 2014b, Reza et al., 2014c), where it is difficult to install such high pressure high temperature sampling system and justify with the overall HTC reaction. Thus, the first objective of this work was to design a high pressure high temperature slurry sampling system on an 18.6 L (5 gal) Parr reactor. Therefore, several different reaction conditions were implied in this study by changing biomass feedstock, reaction time, and reaction temperature, and both solid and liquid phase were studied. Further objective of this work was to prepare a degradation model for selective hydrolyzed products and validate for various lignocellulosic feedstocks.