Hygroscopic transformation of woody biomass torrefaction for carbon storage

Applied Energy 231 (2018) 768–776

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

Hygroscopic transformation of woody biomass torrefaction for carbon storage ⁎

T

⁎

Wei-Hsin Chena,b, , Bo-Jhih Lina,c, Baptiste Colinc, Jo-Shu Changb,d, , Anélie Pétrissansc, Xiaotao Bie, Mathieu Pétrissansc a

Department of Aeronautics and Astronautics, National Cheng Kung University, Tainan 701, Taiwan Research Center for Energy Technology and Strategy, National Cheng Kung University, Tainan 701, Taiwan c Université de Lorraine, Inra, LERMaB, F88000 Epinal, France d Department of Chemical Engineering, National Cheng Kung University, Tainan 701, Taiwan e Department of Chemical and Biological Engineering, University of British Columbia, Vancouver V6T 1Z3, Canada b

H I GH L IG H T S

G R A P H I C A L A B S T R A C T

change and hygroscopic trans• Color formation of poplar and fir from torrefaction are analyzed.

color difference linearly in• Total creases with increasing mass loss or torrefaction severity.

transformation of bio• Hygroscopic mass is evaluated by equilibrium

• •

moisture content (EMC) and contact angle. Hygroscopicity reduction extent (HRE) can reach up to 57.39% at 230 °C. Carbon, hydrogen, and oxygen removals from torrefaction can be predicted by color change and HRE.

A R T I C LE I N FO

A B S T R A C T

Keywords: Torrefaction Hygroscopicity Equilibrium moisture content (EMC) Contact angle Color change Devolatilization

Biochar is a potential medium for carbon storage, so its production and storage have been considered as is a crucial route to effectively achieve negative CO2 emissions. Meanwhile, torrefaction is a thermochemical conversion process for producing biochar. Biochar is featured by its hydrophobicity, which makes it different from its parent biomass with hygroscopicity and is conducive to material storage. To evaluate the hygroscopic transformation of biomass from torrefaction, two woody biomass materials of poplar (hardwood) and fir (softwood) are torrefied at temperatures of 200–230 °C, and the variations of color, equilibrium moisture content, and contact angle of raw and torrefied samples are examined. The results indicate that the total color difference of torrefied woods increases linearly with increasing mass loss. The hygroscopicity reduction extent in torrefied fir is higher than in torrefied poplar, and can be increased by up to 57.39% at 230 °C. The tests of the contact angle suggest that the hygroscopicity of the raw woods is evidently exhibited, whereas the angles of the torrefied woods are in the range of 94–113°, showing their hydrophobic surfaces (> 90°). The decarbonization, dehydrogenation, and deoxygenation phenomena of the biomass during torrefaction are also analyzed. It is found that the three indexes can be correlated well by the total color difference and hygroscopicity reduction

⁎ Corresponding authors at: Department of Aeronautics and Astronautics; Department of Chemical Engineering, National Cheng Kung University, Tainan 701, Taiwan. E-mail addresses: [email protected], [email protected] (W.-H. Chen), [email protected] (J.-S. Chang).

https://doi.org/10.1016/j.apenergy.2018.09.135 Received 29 May 2018; Received in revised form 3 September 2018; Accepted 12 September 2018 0306-2619/ © 2018 Elsevier Ltd. All rights reserved.

Applied Energy 231 (2018) 768–776

W.-H. Chen et al.

extent. A comprehensive study on the improvement of hydrophobicity of produced biochar has been performed, which clearly shows the potential of carbon storage and negative CO2 emissions by biochar.

1. Introduction

varies from 1 to 6%, depending on torrefaction severity. Torrefied biomass has showed its feature of high hydrophobicity [19]. Strandberg et al. [20] examined the EMC of untreated and torrefied spruce woods at 20 °C along with 65% relative humidity, and pointed out that the EMC of the torrefied wood was decreased by 50% or more when compared to the EMC of the raw wood. Chen et al. [21] conducted the torrefaction of cotton stalk at different temperatures, and indicated that the EMC decreased obviously with increasing torrefaction temperature. The EMC of the raw material was 10.8% at 25 °C along with 50% relative humidity, whereas the EMC values of the cotton stalk torrefied at 220, 250, and 280 °C were 7.1, 5.6, and 4.3%, respectively. Kambo and Dutta [22] produced biochars via torrefaction and hydrothermal carbonization. Their results indicated that the EMC values of biochars were in the range of 3.52–7.54%, and were all lower than their parent biomass. Mei et al. [23] used a pilot scale rotary kiln to torrefy cedarwood in flue gas and N2. The EMC value of torrefied cedarwood was reduced by approximately 12–43% when compared with untreated cedarwood. Moreover, compared with N2 torrefaction, the biomass torrefied in the flue gas had a lower EMC value, presumably owing to the oxidation reactions which obviously destroyed biomass structure. The literature reviewed above has provided some of impressive results concerning the EMC of torrefied biomass. Nevertheless, some crucial information in the variation of hygroscopicity from torrefaction remains insufficient. For example, the contact angle is also a crucial measure to response the change of biomass hygroscopicity from torrefaction. But very few studies examined the contact angle of torrefied biomass. In addition, another important physical quantity accompanied by torrefaction is the color change of biomass. With increasing the torrefaction severity, biomass has a trend to become darker. It has been underlined that there was a strong correlation between the EMC and color change when biomass was thermally treated [24]. The improvement in the hydrophobicity of biomass is a pivotal consequence from torrefaction, which will play an important role for biochar production and carbon sequestration, thereby achieving the development of negative CO2 emission technologies (NETs). For this reason, a comprehensive study on the hygroscopic behavior of biomass undergoing torrefaction is carried out in this study where the color change, EMC, and contact angle of torrefied woods are simultaneously considered. The observed phenomena of color change and hygroscopic transformation will be discussed. Furthermore, the correlations between element removals and color change and hygroscopic transformation will be established. In industry, the developed correlations can give a simple tool to produce biochar in accordance with the requirement of hydrophobicity, element removals, or color change, thereby providing a useful insight into the development of NETs.

The sustainability of fuel resources and environment is an issue that is of considerable concern in the world currently. Considering global warming, it is desired to reach the target to “bend down” the greenhouse gas (GHG) emission curve by 2020 [1]. Another goal after Paris Agreement is to limit the global average temperature increment less than 2 °C [2]. The development of biofuels is regarded as an effective countermeasure to reduce fossil fuel consumption and CO2 emissions [3]. Compared with fossil fuels, biofuels have the following advantages: (1) they can be easily obtained and converted from biomass which is characterized by the short life cycle; (2) their combustion is based on the carbon-dioxide cycle (carbon neutral); and (3) they are sustainable and more environment friendly [4]. Biomass is an abundant and available bioresource, and can be supplied from agricultures, forests, industries, and lignocellulosic residues [5]. By means of the thermochemical methods, biomass can be converted into different types of biofuels such as biochar, bio-oil, bioethanol, biodiesel, and syngas [6]. Biochar is a promising alternative fuel to replace coal which generates highest CO2 emissions during combustion when compared with oil and natural gas [7]. It has been pointed out that biochar could reduce net GHG emissions when it was co-fired with coal in power plants [8]. Meanwhile, biochar is a potential material to stably store and fix carbon in soil for negative carbon emissions [9]. While biochar plays a role as a long-term sink for atmospheric CO2 in carbon sequestration process, CO2 emissions could be reduced by up to 84% [10]. To date, biochar has been produced and utilized for several thousand years and is well known as charcoal (when produced from woody biomass). The applications of biochar are very multiple, ranging from energy production [10], building and furniture materials [11], bio-adsorbent for wastewater treatment to soil amendments [12]. Torrefaction is a mild pyrolysis process where biomass is thermally degraded in an inert atmosphere at temperatures of 200–300 °C to produce biochar (torrefied biomass), and can be categorized into light, mild, and severe torrefaction with corresponding temperature ranges of approximately 200–235, 235–275, and 275–300 °C, respectively [13]. Torrefaction is regarded as a pretreatment method to improve the physical, chemical, and biochemical properties of raw biomass [7]. In addition, biochar produced from torrefaction has been proposed as a suitable feedstock for co-combustion, gasification, and thermochemical fuel production [14]. Raw biomass is a hygroscopic material in nature, attributing to the cell wall polymers containing hydroxyl (eOH) groups. The groups absorb moisture into the walls and hold water molecules through hydrogen bonding [6]. This high hygroscopic nature results in biomass being characterized by low calorific value, low dimensional stability, and poor durability [6,15]. These drawbacks also give rise to poor conversion efficiency of biomass, thereby limiting its utilization as fuels and causing high costs for biomass collection, storage, and transportation [16]. After biomass undergoes torrefaction, the produced biochar possesses lower moisture content and higher calorific value. These changes are mainly due to the removal of eOH functional groups during torrefaction. As a result, the hygroscopicity of pretreated biomass is obviously lowered [17], and biochar’s resistance to microbial degradation is intensified [18], rendering easier and cheaper handling and storage of torrefied biomass. The biochar with longer durability implies, in turn, that the ability of carbon sequestration is enhanced. The equilibrium moisture content (EMC) is a common indicator to evaluate the hygroscopicity of biomass. The EMC of torrefied biomass

2. Experimental methodology 2.1. Material preparation and torrefaction Two common European wood species with the dimensions of 60 cm × 17 cm × 2.2 cm were adopted in this study; they were poplar (Populus nigra) and fir (Abies pectinata). Before torrefaction, the wood boards were dried at 103 °C in an oven until mass stabilization. In order to obtain torrefied wood boards with uniform surface and interior, light torrefaction rather than mild and severe torrefaction which would cause intense destroy on the biomass surface [25], was carried out. Another advantage of light torrefaction was that the obtained results were able to provide useful insights into the application of wood treatment for producing sustainable biochar materials [26]. The light 769

Applied Energy 231 (2018) 768–776

W.-H. Chen et al.

angle apparatus, which consisted of a video measuring system and a high-resolution CCD camera (1/3″ CCD B/W Camera). A commercial software (FTA32) was connected with the system to measure and record the drop shape.

torrefaction was performed in a pilot-scale reactor at the temperature range of 200–230 °C along with a low heating rate (0.2 °C min−1) and a low pressure (200 hPa) for 1000 min to ensure the high homogeneity of torrefied biomass. Some studies have reported that the biochars obtained from low pressure pyrolysis had a higher fuel reactivity than atmospheric pyrolysis biochars [27], and could also provide treated products with greater homogeneity [28]. After obtaining torrefied woods, the raw and torrefied samples were crushed by a four-blade cracker and sieved to particle sizes ranging from 0.36 to 1.7 mm for analysis. For the measurements of EMC and contact angle, the samples were cut to the dimensions of 20 mm × 20 mm × 20 mm. The sieved and cut wood samples were dried in an oven at 105 °C for 24 h to provide a dry basis for examinations. The experiments were repeated at least twice to ensure measurement quality. The relative error of mass loss between the experiments was controlled below 5%, and the average values were displayed.

3. Results and discussion 3.1. Solid yield and heating value The indexes of the solid yield, mass loss, the enhancement factor of higher heating value (HHV), and energy yield have been widely utilized to evaluate the torrefaction performance [13]. The formulas of these indexes are expressed as the following:

Solid yield (%) =

Weight torrefied Weight raw

× 100

(2)

Mass loss (%) = 100−Solid yield 2.2. Analysis

Enhancement factor = A number of analyses such as elemental (ultimate), calorific, proximate, and thermogravimetric analyses were performed to determine the basic properties of raw and torrefied woods. The elemental analysis was carried out using an elemental analyzer (PerkinElmer 2400 Series II CHNS/O Elemental Analyzer) to measure the weight percentages of C, H, and N in the biomass, while the weight percentage of O was obtained by difference (i.e., O = 100-C-H-N). The calorific analysis was fulfilled through a bomb calorimeter (IKA C5000). The proximate analysis was performed according to the ASTM standard procedures (i.e., E871 for moisture content; E872 for volatile matter; E1534 for fixed carbon; and D1102 for ash) [29]. The non-isothermal thermogravimetric analysis (TGA) of raw and torrefied wood were examined by a thermogravimetric analyzer (SDT Q600 TGA, TA Instruments). In each run, around 6 mg of the sample was loaded in an Al2O3 crucible, and N2 at a flow rate of 100 mL min−1 was used as the sweep gas. In TGA, the heating process proceeded from room temperature to 105 °C (heating rate = 20 °C min−1), followed by holding the temperature for 10 min to remove surface water in the sample for providing a basis of the analysis. Afterwards, the sample was heated from 105 to 800 °C at the same heating rate. To ensure the experimental quality, prior to performing experiments, the measuring instruments were periodically calibrated. Prior to testing EMC, the dried block samples were placed in a humidity chamber at 25 °C with 55% relative humidity for 24 h. The EMC is defined as follows:

EMC (%) =

m wet −mdried × 100 mdried

(3)

HHVtorrefied HHVraw

(4)

Energy yield (%) = Solid yield × Enhancement factor

(5)

The subscripts “raw” and “torrefied” represent raw and torrefied woods, respectively. The values of the indexes of raw and torrefied woods are listed in Table 1. The solid yields of torrefied woods are in the range of 83.08–96.18%. The lower solid yield implies the higher torrefaction severity. During light torrefaction (200–230 °C), the mass loss mechanisms are dominated by the thermal degradation of hemicelluloses [26]. It is noteworthy that the solid yield of torrefied poplar is always lower than that of torrefied fir at the same torrefaction conditions, ascribing to the different proportions of hemicelluloses in the two wood species [31]. Poplar pertains to hardwood species which come from angiosperm plants. The hemicelluloses in hardwood mainly consist of glucuronoxylan, xyloglucan, and glucomannan. On the other hand, fir belongs to softwood species originating from gymnosperm plants, and their hemicelluloses mainly comprise xyloglucan, arabinoglucuronoxylan, and galactoglucomannan [32]. The glucuronoxylan in hardwood is a strongly acetylated component, and the acetyl groups are attached to the glucomannan backbone. The stronger deacetylation in hardwood during torrefaction can accelerate the biomass degradation, which is responsible for the lower solid yield of the poplar [31]. The HHVs of the raw and torrefied woods are between 16.29 MJ kg−1 and 18.31 MJ kg−1. The HHVs of the fir are slightly higher than those of the poplar, resulting from the higher contents of resin or extractives (with higher calorific values) in softwood [33]. The enhancement factor, standing for energy densification from torrefaction, is in the range of 1.021–1.124, while the energy yield is in the range of 93.38–98.20%. These data are in a good agreement with the results of light torrefaction in the literature [23,29]. For example, Mei et al. [23] examined the torrefaction of cedarwood in a rotary kiln

(1)

where m wet is the weight of humidified wood and mdried is the weight of dried wood The measurement of the contact angle was based on the sessile drop method [30] to observe the profile of a deposited drop on a solid surface. The observation was performed using an optical contact Table 1 Solid yield, HHV, and proximate analysis. Material

Torrefaction temperature (°C)

Solid yield (%)

HHV (MJ kg−1)

Enhancement factor

Energy yield (%)

Proximate analysis (wt%, dry basis) Fixed carbon

Volatile matter

Ash

Poplar (Populus nigra)

Raw 200 210 220 230

100 91.73 88.25 85.61 83.08

16.29 17.23 17.56 17.83 18.31

1.00 1.06 1.08 1.09 1.12

100 97.05 95.13 93.66 93.38

14.70 18.04 19.68 20.93 22.70

84.74 81.40 79.59 78.27 76.46

0.56 0.57 0.73 0.80 0.84

Fir (Abies pectinata)

Raw 200 210 220 230

100 96.18 93.92 90.36 87.71

16.55 16.90 17.08 17.61 18.26

1.00 1.02 1.03 1.06 1.10

100 98.20 96.93 96.14 96.74

14.28 16.49 17.18 18.87 20.03

85.53 83.31 82.59 80.88 79.72

0.19 0.20 0.23 0.24 0.25

770

Applied Energy 231 (2018) 768–776

W.-H. Chen et al.

(a) Poplar Raw

(b) Fir Raw Pore

200 °C

200 °C

Crack

Pore

210 °C

210 °C

220 °C

210 °C

230 °C

230 °C Pore

Fig. 1. TGA and DTG distributions of (a) poplar and (b) fir before and after torrefaction. Fig. 2. SEM images (1000X magnification) of (a) poplar and (b) fir before and after torrefaction.

reactor at temperatures between 200 and 230 °C, and the enhancement factor was in the range of 1.05–1.16. Zhang et al. [29] performed the torrefaction of biomass wastes (spent coffee grounds, Chinese medicine residue, and microalga residue) in a fixed bed reactor, and showed that the energy yield of light torrefaction was between of 90.58 and 99.73%. The results of proximate analysis are also shown in Table 1. The volatile matter (VM) contents in the raw poplar and fir are 87.74 and 85.53%, respectively, while their fixed carbon (FC) contents are 14.70 and 14.28%, respectively. After torrefaction, the VM of the torrefied woods decrease, as a consequence of devolatilization mechanism involved [6]. On the other hand, the increase in the FC of the torrefied woods is owing to the formation of carbonaceous materials from the carbonization mechanism and thermal cross-linking reactions during torrefaction [6,26].

peaks of the untreated poplar and fir, stemming from the decomposition of cellulose, develop at 361 and 372 °C, respectively. Meanwhile, the small shoulders by the peaks, due to the thermal degradation of hemicelluloses, are exhibited. The temperature of a shoulder can be obtained from the second derivative of a DTG (2nd DTG) where its value is zero [34]. Mathematically, this value (=0) stands for the inflection point or the junction of the concave up and concave down curves. According to this definition, the degradation temperatures of hemicelluloses in the raw poplar and fir are located at 315 and 329 °C, respectively. When the temperature is higher than 400 °C, a continuous and slow decomposition zone, resulting from the thermal degradation of lignin, is observed. After undergoing torrefaction, the zero value of the 2nd DTG cannot be found, regardless of which torrefaction temperature is. This reveals the pronounced destruction of hemicelluloses from the torrefaction. Compared to the torrefied fir, the torrefied poplar samples have sharper peaks. It has been illustrated earlier that the hemicellulose degradation of the poplar during torrefaction is more drastic than that of the fir. This implies, in turn, that relatively more cellulose is retained in the torrefied poplar. This is the reason why the torrefied poplar has higher peak. The obtained results are in line with past studies [26,35,36] which investigated the thermal treatment of wood materials at temperatures of 200–240 °C, and concluded that: (1) the hemicelluloses were significantly degraded during treatment; (2) the amorphous cellulose was decomposed at this temperature range, leading to the increase in the

3.2. TGA and SEM The thermogravimetric analysis (TGA) and derivative thermogravimetric (DTG) analysis are able to provide quantitative observations on the thermal degradation of hemicelluloses, cellulose, and lignin in biomass [25]. The thermal decomposition temperatures of hemicelluloses, celluloses, and lignin are in the ranges of 200–315, 315–400, and 160–900 °C, respectively [6]. Fig. 1 displays the TGA and DTG curves of raw and torrefied woods. The thermal decompositions of the untreated poplar and fir occur at temperatures between 200 and 600 °C, and the thermal degradations of hemicelluloses, cellulose, and lignin in the woods can be clearly identified. The DTG curves suggest that the 771

Applied Energy 231 (2018) 768–776

W.-H. Chen et al.

Fig. 3. Profiles of mass loss versus (a) lightness L*, (b) chrome value a*, (c) chrome value b* and (c) total color difference (ΔE ∗).

the color of torrefied biomass changed from brown to black at the temperature range of 150–300 °C [39]. In this study, the colors of the raw and torrefied woods were measured by a colorimeter (Chroma meter CR-410, Konica Minolta) based on the three dimensional CIELAB color space, which has been adopted to quantify the color change of treated biomass [40,41]. In the CIELAB color space, the vertical coordinate for lightness L* represents the position on the black-white axis (L* = 0 for total black and L* = 100 for pure white). The chromatic coordinates a* and b* are characterized by the position of horizontal plane. The chrome value of a* stands for the position on the green-red axis (+a* for red and −a* for green), while the chrome value of b* responds the position on the blue-yellow axis (+b* for yellow and −b* for blue) [40,42]. The total color difference ΔE ∗ is expressed by [40]:

crystallinity of cellulose; and (3) the degree of polymerization in cellulose and lignin were lessened. Overall, the peak intensity of torrefied poplar is higher than that of torrefied fir. In order to provide a deeper insight into the impact of torrefaction upon biomass structure, the scanning electron microscope (SEM, Hitachi S-3000 N) is applied to observe the surface morphology of the raw and torrefied woods. The SEM images of the samples at a 1000 magnification are shown in Fig. 2. In the poplar (Fig. 2a), the welldefined porous ovals are observed in raw material [37]. When the torrefaction temperature is higher, the ovals start to crack. The damaged fibre structure is also observed. For the torrefaction temperature of 230 °C, the ovals are significantly destroyed, attributing to the severe degradation. For the raw fir (Fig. 2b), a homogeneous fibrillary organization in the external surface with tiny pores can be observed. After torrefaction, the cell-wall structures are characterized by bigger and destroyed pores, elucidating the profound collapse of the cell walls. This is assigned to the release of VM, namely, the devolatilization mechanism, in the course of torrefaction [38]. The change in the surface morphology of the wood materials from torrefaction such as destroyed cell wells and increased porosity can improve their grindability and the reactivity for solid-gas reactions, such as gasification and combustion [25,38].

ΔE ∗ =

(L2∗−L1∗)2 + (a2∗−a1∗)2 + (b2∗−b1∗)2

(6)

where subscripts 1 and 2 represent raw and torrefied samples, respectively. It has been underlined that mass loss (ML) is an effective measure to indicate torrefaction severity [13]. The higher the ML, the higher the torrefaction severity. For this reason, the profiles of L*, a*, b*and ΔE ∗ versus ML are plotted in Fig. 3 to examine the correlation between color change and torrefaction severity. For the raw poplar, its lightness (L*) is 86.10, and after torrefaction the value decreases significantly, ranging from 31.09 to 35.68 (Fig. 3a). For the fir, its L* decreases from 80.75 (raw) to 30.58 (at 230 °C of torrefaction). For the two woods, there is a clear linear relationship between L* and torrefaction temperature, and decreasing L* implies that the color of torrefied biomass becomes darker. Esteves [43] performed the thermal treatment of pine and eucalypt, and showed that the lightness decreased significantly with

3.3. Color change Biomass torrefaction accompanied by color change, stemming from devolatilization and carbonization mechanisms, is a remarkable feature where the color of biomass is modified from light brown to dark brown or black, depending on torrefaction severity. It has been reported that 772

Applied Energy 231 (2018) 768–776

W.-H. Chen et al.

(a) EMC 8 7 6

Poplar

EMC (%)

5 4 3 2

Fir

1 0

0

2.5

5

7.5

10

12.5

15

17.5

20

Mass loss (%)

(b) Hygroscopicity reduction extent , HRE 100

HRE (%)

80

60

40

20

0

0

2.5

5

7.5

10

12.5

15

17.5

Fig. 5. Profiles of contact angle on the surface of raw and torrefied (a) poplar and (b) fir.

20

Mass loss (%)

byproducts are formed from the decomposition of hemicelluloses [43,44]; (2) the cross-linking reactions, condensation reactions, and oxidation reactions from cleavage of lignin β-O-4 ether bonds and aromatic methyl groups in lignin lead to the formation of oxidative products like quinones, which facilitate color change during treatment [44,45]; (3) the enzyme-mediated (Maillard) reactions between polysaccharides such as sugars, phenolic compounds, and amino acids are triggered during the thermal degradation [41]; and (4) the oxidative reactions between extractives (in woody biomass) and the atmosphere in the course of treatment are driven [44].

Fig. 4. Profiles of mass loss versus (a) EMC, and (b) hygroscopicity reduction extent (HRE).

increasing treatment time and temperature. The values of a* in the poplar and fir increase at the lower temperature of 200 °C, followed by decline at higher temperatures (Fig. 3b). The maximum of a* is 5.91 for the poplar, and 7.88 for the fir. For the vlaues of b* in the poplar and fir, they both decrease with rising torrefaction severity (Fig. 3c), resembling the results of other studies [41,43]. González-Peña et al. [41] reported that the changes of a* and b* were more complex and depended on wood species, hence it was more difficult to identify the chemical reactions during treatment. The total color difference (ΔE ∗) of the two woods is in the range of 41.49–51.66 and increase after torrefaction (Fig. 3d). Overall, L* is the dominant factor in determining ΔE ∗ inasmuch as its variation before and after torrefaction is by far greater than the alternations of the others (i.e., a* and b*). The color variation in Fig. 3d is observably correlated to ML. This arise from the fact that the color change of torrefied woods are mainly caused by the thermal degradation of biopolymer [24]. The color change of lignocellulosic biomass from torrefaction is attributed to a number of reactions during torrefaction: (1) the color

3.4. EMC and contact angle Equilibrium moisture content (EMC) and contact angle are two important measures to show the hydrophobicity of biomass. In this study, the hygroscopicity reduction extent (HRE) is introduced to quantify the reduction of biomass hygroscopicity from torrefaction, and is defined as:

HRE (%) = (1−

773

EMCtorrefied ) × 100 EMCraw

(7)

Applied Energy 231 (2018) 768–776

W.-H. Chen et al.

Table 2 Elemental analysis, DC, DH and DO. Raw and torrefied wood

Elemental analysis (wt%, dafa)

DCc (%)

DHd (%)

DOe (%)

C

H

N

Ob

Poplar Raw 200 °C 210 °C 220 °C 230 °C

46.61 49.44 50.54 51.02 52.25

6.32 6.27 6.12 5.95 5.88

0.28 0.28 0.36 0.38 0.45

46.79 44.01 42.98 42.65 41.42

0 2.71 4.48 6.52 7.13

0 8.93 14.63 19.57 22.87

0 13.73 19.08 22.15 26.67

Fir Raw 200 °C 210 °C 220 °C 230 °C

47.16 48.36 49.35 50.73 51.83

6.41 6.17 6.05 5.98 5.94

0.23 0.24 0.27 0.34 0.38

46.20 45.23 44.33 42.95 41.86

0 1.39 1.75 2.84 3.66

0 7.35 11.37 15.72 18.74

0 5.86 9.93 16.05 20.59

a b c d e

Dry-ash-free. By difference. Decarbonization. Dehydrogenation. Deoxygenation.

results in removing eOH and eCOOH groups and further decreasing hydrogen bonding with water after torrefaction; (2) tar condenses inside the pores, thereby obstructing the passage of moist air through the solid and then avoiding water vapor condensation; (3) the apolar character of condensed tar on the solid also prevents the condensation of water vapor inside the pores [6,36,46]. Most importantly, the observed hygroscopicity transformation of woods after torrefaction can markedly improve their dimensional stability and avoid the biodegradation from microorganism [36]. Paul et al. [48] examined the correlation between the EMC change and fungal resistance of thermally treated wood materials, and discovered that the fungal decay of treated wood tended to disappear where the EMC of treated wood was decreased by around 37–40% compared to untreated wood. These improvements can efficiently enhance the storage and transport of produced biochar. Alternatively, the prolonged storage life of torrefied wood implies the enhancement of carbon sequestration ability.

The larger the HRE value, the more hydrophobic the torrefied biomass. The profiles of EMC and HRE versus ML are shown in Fig. 4. A pronounced drop in the EMC occurs after torrefaction (Fig. 4a), and the EMC has a trend to decrease linearly with increasing torrefaction severity. The hydrophobicity of the fir is more sensitive to ML and its HRE can reach up to 57.39% at 230 °C (Fig. 4b). The obtained results are consonant with other studies [22,23]. Strandberg et al. [20] discovered that the EMC of torrefied biomass (spruce wood) decreased by at least 50% when biomass was torrefied at temperature higher than 260 °C. The transient profiles of the contact angles of raw and torrefied woods are shown in Fig. 5. It has been pointed out that the contact angle close to 0° corresponded to a hydrophilic surface (with clearly hygroscopic property). If the angle was less than 90°, the sample was more hydrophilic than hydrophobic. Once the angle was above 90°, the surface was hydrophobic [20]. Fig. 5 shows the apparent absorption of water droplet into the raw poplar and fir, rendering their hygroscopic nature. For the raw poplar, the initial contact angle is 63.7° and the angle becomes 0° at around 3 s. As regard to the raw fir, the initial contact angle is 86.5°, and the angle is close to 0° after 19 s. Overall, it appears that poplar has higher hygroscopicity when compared to fir. Water absorbed by the woody materials is mainly due to the presence of hydroxyl groups (eOH) which attract and hold water molecules through hydrogen bonding. In wood materials, hemicelluloses are more hydrophilic than cellulose and lignin. Hemicelluloses and the noncrystalline region of cellulose chains can attract water easier, owing to the availability of hydroxyl groups [43]. The carboxylic acid groups (eCOOH) in hemicelluloses are also active to absorb water [46]. The higher water absorption speed in the raw poplar is attributed to the specially large water-conducting pores (called vessels) and higher content of carboxylic acid groups in hardwood species [47]. Unlike the raw woods, the contact angles of torrefied samples can last for a longer time and are always greater than 90°. The higher the torrefaction severity, the larger the contact angle. The contact angles of the torrefied poplar and fir are in the ranges of 94.9–107.0° and 103.4–113.0°, respectively. The contact angles of the torrefied fir (Fig. 5b) are always greater than those of torrefied poplar (Fig. 5a), despite the lower ML of the former. This observation can be owing to the significant tar components present during fir torrefaction where the tar condensates on the torrefied wood surface, thereby increasing hydrophobic properties [20]. This is also the reason why the HRE of fir is higher than poplar (Fig. 4b). In summary, the reduction in the hygroscopic behavior of torrefied woods can be assigned to the following reasons: (1) the degradation of hemicelluloses and amorphous cellulose

3.5. Removal of C, H, and O and correlations In the earlier discussion, the changes of color and hygroscopicity of woody biomass are mainly attributed to the devolatilization during torrefaction. In order to evaluate the characteristics of devolatilization during torrefaction; decarbonization (DC), dehydrogenation (DH), deoxygenation (DO) are analyzed. According to the elemental analysis of the wood samples (Table 2), the original carbon amount (OC) in a raw material is expressed as [49]

OC (g) = W0 × 10−2 × YC,0

(8)

where W0, and YC,0 denote the weight (g) of sample (in dry-ash-free basis) and mass fraction of carbon, respectively, and the subscript 0 stands for raw material. The residual carbon amount (RC) in torrefied wood is determined in accordance with the following formula:

R C (g) = W0 × SY × 10−2 × YC,t

(9)

where SY designates solid yield, and the subscript t denotes treated wood. Consequently, decarbonization (DC), which represents the carbon loss percentage in the wood from torrefaction, is calculated by

R DC(%) = ⎛1− c ⎞ × 100 ⎝ OC ⎠ ⎜

⎟

(10)

DH and DO are also determined using the same way. The values of DC, DH, and DO are listed in Table 2, which shows 774

Applied Energy 231 (2018) 768–776

W.-H. Chen et al.

(a)

(b) 10

10 8 y = 0.8121x - 38.614 2 R = 0.9827

6 4 2

Decarbonization (%)

Decarbonization (%)

Poplar

Fir y = 0.2004x - 7.1834 2 R = 0.8742

4 2

30

30

25

25

20

Dehydrogenation (%)

Dehydrogenation (%)

Total color difference (E*)

y = 1.0136x - 35.234 2 R = 0.9717

15 10 5

y = 2.4618x - 116.360 2 R = 0.9933

0 40 42.5 45 47.5 50 52.5 55 57.5 60

Deoxygenation (%)

30

15 10 5

y = 2.185x - 97.511 2 R = 0.9777

Total color difference (E*)

50

55

60

y = 0.4745x - 7.4846 2 R = 0.8836

35

40

25

45

HRE (%)

50

55

60

y = 1.4628x - 39.034 2 R = 0.9412

20 15 10

y = 0.6312x - 14.536 2 R = 0.9147

5

0 40 42.5 45 47.5 50 52.5 55 57.5 60

45

HRE (%)

y = 1.6735x - 51.502 2 R = 0.9859

5

30

20

40

10

35

y = 1.3043x - 49.342 2 R = 0.9412

35

15

35

25

y = 0.1015x - 2.0333 2 R = 0.9304

20

0 30

Total color difference (E*)

y = 0.5586x - 17.488 2 R = 0.9987

6

0 30

0 40 42.5 45 47.5 50 52.5 55 57.5 60

Deoxygenation (%)

8

0 30

35

40

45

HRE (%)

50

55

60

Fig. 6. Profiles and linear regressions of (a) total color difference (ΔE ∗) and (b) hygroscopicity reduction extent (HRE) versus decarbonization, dehydrogenation, and deoxygenation.

predict mass loss of treated biomass [24,26,41]. Nguyen et al. [24] performed the thermal pretreatment of bamboo in the temperature range of 130–220 °C, and found that the mass loss of bamboo was wellcorrelated by color change and EMC in that the R2 values were in the range of 0.87–0.93 and 0.74–0.86, respectively. As a consequence, this study suggests that, aside from mass loss [29], the total color difference and HRE can also be used to predict carbon, hydrogen, and oxygen removals in biomass from torrefaction.

that the general trend is ranked as DO > DH > DC, stemming from the removal of moisture and light volatiles during torrefaction [50]. This reflects the more significant impact of torrefaction upon oxygen and hydrogen than on carbon. However, the DH is larger than DO for the fir torrefied at 200 and 210 °C. This can be explained by stronger dehydration and more release of extractives which mainly consist of aliphatic, alicyclic, and phenolic compounds [51]. It has been reported that the extractives in softwood (5–11%) were higher than hardwood (2–4.5%) [34], and the maximum thermal degradation rate of extractives occurred at approximately 205 °C [32]. In addition, the existence of extractives indeed influences the biomass pyrolysis behavior and its products, especially when pyrolyzing extractives-rich biomass [32]. The profiles of DC, DH, and DO versus ΔE ∗ and HRE are plotted in Fig. 6. The profiles are characterized by strongly linear relationship (R2 > 0.87), especially for the poplar (R2 is in the range of 0.9412–0.9933). The slops of regression lines of the poplar are greater than those of the fir, elucidating more severe degradation of the former during torrefaction. Some studies have used EMC and color change to

4. Conclusions The color change and hygroscopic transformation of two woody biomass materials (poplar and fir) during torrefaction have been investigated in this study to provide useful insights into biochar storage and carbon sequestration. The measured color change based on CIELAB color space indicates that lightness (L*) is the dominant factor in determining the total color difference (ΔE ∗) . The value of ΔE ∗ of the two woods is in range of 41.49–51.66 and increases with mass loss or torrefaction severity. A measure of the hygroscopicity reduction extent is 775

Applied Energy 231 (2018) 768–776

W.-H. Chen et al.

introduced to quantify the hygroscopic transformation of biomass. Overall, the hygroscopicity reduction extent of torrefied poplar and fir is in the range of 34.50–57.39%, suggesting the strong transformation of the biomass from hygroscopic nature to hydrophobic behavior. This significantly diminishes the possibility of microbial degradation of biochar, and the decreased hygroscopicity can effectively prolong storage period of torrefied biomass, thereby resulting in the enhancement of carbon sequestration ability. The contact angles of the raw poplar and fir decays rapidly when water is dropped on the boards. The contact angels of all the torrefied samples are higher than 90° (in the range of 94–113°), showing the hydrophobic surfaces. In addition, the contact angles of the torrefied fir are greater than those of torrefied poplar, ascribing to more tar present and condensed during the torrefaction of the fir. The extents of decarbonization, dehydrogenation, and deoxygenation from torrefaction are also evaluated, and a strong linear relationship (R2 > 0.87) between the three indexes versus ΔE ∗or hygroscopicity reduction extent is exhibited. These results imply, in turn, that the variations of color change and hygroscopicity reduction extent can be used to predict carbon, hydrogen, and oxygen removals for biomass in torrefaction. Overall, the obtained results suggest that torrefaction is a suitable route to produce hydrophobic biochar which can against fungi attach and is conducive carbon storage, thereby achieving negative emissions technologies.

upgrade solid characteristics. Bioresour Technol 2017;244:741–9. [19] Iroba KL, Baik OD, Tabil LG. Torrefaction of biomass from municipal solid waste fractions II: Grindability characteristics, higher heating value, pelletability and moisture adsorption. Biomass Bioenergy 2017;106:8–20. [20] Strandberg M, Olofsson I, Pommer L, Wiklund-Lindströma S, Åberg K, Nordin A. Effects of temperature and residence time on continuous torrefaction of spruce wood. Fuel Process Technol 2015;134:387–98. [21] Chen D, Zheng Z, Fu K, Zeng Z, Wang J, Lu M. Torrefaction of biomass stalk and its effect on the yield and quality of pyrolysis products. Fuel 2015;159:27–32. [22] Kambo HS, Dutta A. Strength, storage, and combustion characteristics of densified lignocellulosic biomass produced via torrefaction and hydrothermal carbonization. Appl Energy 2014;135:182–91. [23] Mei Y, Liu R, Yang Q, Yang H, Shao J, Draper C, et al. Torrefaction of cedarwood in a pilot scale rotary kiln and the influence of industrial flue gas. Bioresour Technol 2015;177:355–60. [24] Nguyen CT, Wagenführ A, Phuong LX, Dai VH, Bremer M, Fischer S. Effects of thermal modification on the properties of two vietnamese bamboo species. Part I: Effects on physical properties. BioResources 2012;7(4):5355–66. [25] Chen WH, Cheng WY, Lu KM, Huang YP. An evaluation on improvement of pulverized biomass property for solid fuel through torrefaction. Appl Energy 2011;88:3636–44. [26] Chaouch M, Dumarcay S, Pétrissans A, Pétrissans M, Gérardin P. Effect of heat treatment intensity on some conferred properties of different European softwood and hardwood species. Wood Sci Technol 2013;47:663–73. [27] Yang X, Yuan C, Xu J, Zhang W. Co-pyrolysis of Chinese lignite and biomass in a vacuum reactor. Bioresour Technol 2014;173:1–5. [28] Ferrari S, Allegretti O, Cuccui I, Moretti N, Marra M, Todaro L. A revaluation of Turkey oak wood (Quercus cerris L.) through combined steaming and Thermo-vacuum treatments. BioResources 2013;8(4):5051–66. [29] Zhang C, Ho SH, Chen WH, Xie Y, Liu Z, Chang JS. Torrefaction performance and energy usage of biomass wastes and their correlations with torrefaction severity index. Appl Energy 2018;220:598–604. [30] Pétrissans M, Gérardin P, El bakali I, Serraj M. IWettability of heat-treated wood. Holzforschung 2005;57(3):301–7. [31] Candelier K, Chaouch M, Dumarcay S, Pétrissans A, Pétrissans M, Kamdem P, et al. Utilization of thermodesorption coupled to GC–MS to study stability of different wood species to thermodegradation. J Anal Appl Pyrol 2011;92:376–83. [32] Wang S, Dai G, Yang H, Luo Z. Lignocellulosic biomass pyrolysis mechanism: a state-of-the-art review. Prog Energy Combust Sci 2017;62:33–86. [33] Telmo C, Lousada J. Heating values of wood pellets from different species. Biomass Bioenergy 2011;35:2634–9. [34] Grnli MG, Vrhegyi G, Di Blasi C. Thermogravimetric analysis and devolatilization kinetics of wood. Ind Eng Chem Res 2002;41(17):4201–8. [35] Esteves BM, Pereira HM. Wood modification by heat treatment: a review. Bioresources 2009;4(1):370–404. [36] Pelaez-Samaniego MP, Yadamam V, Lowell E, Espinoza-Herrera R. A review of wood thermal pretreatments to improve wood composite properties. Wood Sci Technol 2013;47:1285–11239. [37] Granados DA, Ruiz RA, Vega LY, Chejne F. Study of reactivity reduction in sugarcane bagasse as consequence of a torrefaction process. Energy 2017;139:818–27. [38] Ramos-Carmona S, Martínez JD, Pérez JF. Torrefaction of patula pine under air conditions: a chemical and structural characterization. Ind Crops Prod 2018;118:302–10. [39] Tumuluru JS, Sokhansanj S, Hess JR, Wright CT, Boardman RD. A review on biomass torrefaction process and product properties for energy applications. Ind Biotechnol 2011;7(5):384–401. [40] Tooyserkani Z, Sokhansanj S, Bi X, Lim J, Lau A, Saddler J, et al. Steam treatment of four softwood species and bark to produce torrefied wood. Appl Energy 2013;103:514–21. [41] González-Peña MM, Hale MDC. Colour in thermally modified wood of beech, Norway spruce and Scots pine. Part 1: Colour evolution and colour changes. Holzforschung 2009;63:385–93. [42] Agudo JE, Pardo PJ, Sánchez H, Pérez ÁL, Suero MI. A low-cost real color picker based on arduino. Sensors 2014;14:11943–56. [43] Esteves B, Velez Marques A, Domingos I, Pereira H. Heat-induced colour changes of pine (Pinus pinaster) and eucalypt (Eucalyptus globulus) wood. Wood Sci Technol 2008;5:369–84. [44] Dubey MK, Pang S, Walker J. Effect of oil heating age on colour and dimensional stability of heat treated Pinus radiate. Eur J Wood Wood Prod 2011;69:255–62. [45] Wikberg H, Maunu SL. Characterisation of thermally modified hard- and softwoods by 13C CPMAS NMR. Carbohydr Polym 2004;58:461–6. [46] Järvinen T, Agar D. Experimentally determined storage and handling properties of fuel pellets made from torrefied whole-tree pine chips, logging residues and beech stem wood. Fuel 2014;129:330–9. [47] Stokke DD. Fundamental aspects of wood as a component of thermoplastic composites. J Vinyl Add Tech 2003;9(2):96–104. [48] Paul W, Ohlmeyer M, Leithoff H. Thermal modification of OSB-strands by a one-step heat pre-treatment-influence of temperature on weight loss, hygroscopicity and improved fungal resistance. Holz Roh Werkst 2007;65:57–63. [49] Chen YC, Chen WH, Lin BJ, Chang JS, Ong HC. Impact of torrefaction on the composition, structure and reactivity of a microalga residue. Appl Energy 2016;181:110–9. [50] Wang L, Barta-Rajnai E, Skreiberg Ø, Khalil R, Czégény Z, Jakab E, et al. Effect of torrefaction on physiochemical characteristics and grindability of stem wood, stump and bark. Appl Energy 2018;227:137–48. [51] Valette N, Perrot T, Sormani R, Gelhaye E, Morel-Rouhier M. Antifungal activities of wood extractives. Fungal Biology Rev 2017;31:113–23.

Acknowledgments The authors gratefully acknowledge the financial supports (MOST 106-2923-E-006-002-MY3) of the Ministry of Science and Technology in Taiwan, R.O.C., as well as the financial support from Laboratory of Excellence ARBRE (ANR-11-LABX-0002-01) and Lorraine Region Council in France. References [1] Baykara SZ. Hydrogen: a brief overview on its sources, production and environmental impact. Int J Hydrogen Energy 2018;23:10605–14. [2] Murphy F, McDonnell K. Investigation of the potential impact of the Paris Agreement on national mitigation policies and the risk of carbon leakage; an analysis of the Irish bioenergy industry. Energy Policy 2017;104:80–8. [3] Nicodème T, Berchem T, Jacquet N, Richel A. Thermochemical conversion of sugar industry by-products to biofuels. Renew Sustain Energy Rev 2018;88:151–9. [4] Gaurav N, Sivasankari S, Kiran GS, Ninaw A, Selvin J. Utilization of bioresources for sustainable biofuels: a review. Renew Sustain Energy Rev 2017;73:205–14. [5] Cai J, He Y, Yu X, Banks SW, Yang Y, Zhang X, et al. Review of physicochemical properties and analytical characterization of lignocellulosic biomass. Renew Sustain Energy Rev 2017;76:309–22. [6] Chen WH, Peng J, Bi XT. A state-of-the-art review of biomass torrefaction, densification and applications. Renew Sustain Energy Rev 2015;2015(44):847–66. [7] Huang YF, Cheng PH, Chiueh PT, Lo SL. Leucaena biochar produced by microwave torrefaction: fuel properties and energy efficiency. Appl Energy 2017;204:1018–25. [8] Ohlemüller P, Ströhle J, Epple B. Chemical looping combustion of hard coal and torrefied biomass in a 1 MWth pilot plant. Int J Greenhouse Gas Control 2017;65:149–59. [9] Smith P. Soil carbon sequestration and biochar as negative emission technologies. Glob Change Biol 2016;22(3):1325–11314. [10] Yu KL, Lau BF, Show PL, Ong HY, Ling TC, Chen WH, et al. Recent developments on algal biochar production and characterization. Bioresour Technol 2017;246:2–11. [11] Gupta S, Kua HW, Low CY. Use of biochar as carbon sequestering additive in cement mortar. Cem Concr Compos 2018;87:110–29. [12] Gan YY, Ong HC, Show PL, Ling TC, Chen WH, Yu KL, et al. Torrefaction of microalgal biochar as potential coal fuel and application as bio-adsorbent. Energy Convers Manage 2018;165:152–62. [13] Chen WH, Hsu HJ, Kumar G, Budzianowski WM, Ong HC. Predictions of biochar production and torrefaction performance from sugarcane bagasse using interpolation and regression analysis. Bioresour Technol 2017;246:12–9. [14] Ciolkosz D, Wallace R. A review of torrefaction for bioenergy feedstock production. Biofuels, Bioprod Biorefin 2011;5(3):317–29. [15] Pétrissans A, Hamada J, Chaouch M, Gérardin P, Pétrissans M. Modeling and numerical simulation of wood torrefaction. Innovation Woodworking Indust Eng Des 2014;5:26–32. [16] Zhang Y, Geng P, Liu R. Synergistic combination of biomass torrefaction and cogasification: reactivity studies. Bioresour Technol 2017;245:225–33. [17] Kumar L, Koukoulas AA, Mani S, Satyavolu J. Integrating torrefaction in the wood pellet industry: a critical review. Energy Fuels 2017;31:37–54. [18] Iáñez-Rodríguez I, Martín-Lara MÁ, Blázquez G, Pérez A, Calero M. Effect of torrefaction conditions on greenhouse crop residue: optimization of conditions to

776