Effect of alkaline pre-extraction of hemicelluloses and silica on kraft pulping of bamboo (Neosinocalamus affinis Keng)

Industrial Crops and Products 91 (2016) 66–75

Contents lists available at ScienceDirect

Industrial Crops and Products journal homepage: www.elsevier.com/locate/indcrop

Effect of alkaline pre-extraction of hemicelluloses and silica on kraft pulping of bamboo (Neosinocalamus affinis Keng) Zhaoyang Yuan a,∗ , Nuwan S. Kapu a , Rodger Beatson a,b , Xue Feng Chang b , D. Mark Martinez a a b

Department of Chemical & Biological Engineering, University of British Columbia, 2360 East Mall, Vancouver, BC V6T 1Z4, Canada Chemical & Environmental Technology, British Columbia Institute of Technology, 3700 Willingdon Ave., Vancouver, BC V5G 3H2, Canada

a r t i c l e

i n f o

Article history: Received 19 February 2016 Received in revised form 13 June 2016 Accepted 19 June 2016 Keywords: Alkaline pre-extraction Bamboo Hemicellulose Kraft pulping Pulp strength Silica

a b s t r a c t Commercial bamboo chips were pre-treated with sodium hydroxide (NaOH) solutions to completely extract silica and partially extract hemicelluloses prior to kraft pulping. Reaction temperatures of 80–100 ◦ C, times of 1–5 h, and NaOH charges of 6–18% were explored. With increasing pre-extraction severity, all silica and up to 50% of hemicelluloses in raw chips could be extracted without degrading cellulose and lignin. The chips from select extractions were cooked using the kraft process with varying effective alkali (EA) charges. Pre-extraction resulted in significant improvement in the delignification of chips during subsequent kraft pulping, offering an option to reduce the EA charge or the H-factor. The pulp yield was similar to the control while the drainage resistance of pulp from pre-extracted chips was slightly improved. Physical strength properties of pulps made from pre-extracted chips showed lower tensile index and higher tear index as compared with the control runs. Moreover, silica was no more a problem for chemical recovery and production of high-grade pulp. Extracted silica and hemicelluloses in the alkaline extraction liquor (AEL) can be used as a potential raw material for value-added products. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Non-wood materials are attracting growing attention for use in conventional pulp and paper making and as a feedstock for the bio-based economy due to diminishing forest resources and increasing consumption of lignocellulosic products (Farrell et al., 2000; Machmud et al., 2013; Salmela et al., 2008). Of the nonwoods, bamboo, abundantly distributed in Asia and South America, is considered to be among the most attractive feedstocks due to its rapid growth and comparable chemical composition with woods (Luo et al., 2013; Scurlock et al., 2000; Sixta, 2006). Moreover, unlike cereal straw, the kraft based technologies, similar to those used for woody raw materials, are generally used for the production of pulp and paper products from bamboo (Vu Mân et al., 2004). However, compared to wood, bamboo contains a much higher level of silica. The high silica level creates problem in the recovery cycle of the kraft pulping process and high amounts of residual silica in dissolving pulps, cause poor filterability and interfere with the downstream conversion of dissolving pulps into other products

∗ Corresponding author. E-mail address: [email protected] (Z. Yuan). http://dx.doi.org/10.1016/j.indcrop.2016.06.019 0926-6690/© 2016 Elsevier B.V. All rights reserved.

(Liese, 1987; Salmela et al., 2008; Tsuji et al., 1965). Considerable efforts have been expended in trying to solve the silica issue. These efforts have been mainly focused on preserving silica in the final pulp and desilication of black liquor rather than removing silica from the raw material entering the pulping process (Jahan et al., 2006; Kopfmann and Hudeczek, 1988; Pan et al., 1999; Tsuji et al., 1965). Unfortunately, until now, none of this work has led to a commercial process. Recently, the proposed integrated forest biorefinery (IFBR) concept has been advanced as a means of addressing concerns over energy security and climate change. According to the IFBR concept, hemicelluloses are partially or completely pre-extracted under acidic or alkaline conditions prior to the pulping stages for the generation of value-added products such as bioethanol, furfural, acetone, or to be used as papermaking additives (Bai et al., 2012; Hamzeh et al., 2013; Liu et al., 2013; Mao et al., 2008). Moreover, pre-extraction of lignocellulosic biomass has been shown to be an efficient way of removing hemicelluloses while also improving the digestibility of the residual biomass and preserving pulp quality (Van Heiningen, 2006). Thus, a novel way to solve the silica problems encountered when pulping bamboo could involve preextraction of silica along with hemicelluloses prior to the pulping processes.

Z. Yuan et al. / Industrial Crops and Products 91 (2016) 66–75

Previous studies on the pre-treatment of bamboo chips prior to pulping were mainly focused on the extraction of hemicelluloses/lignin for the production of kraft pulp, high-grade dissolving pulp or fermentable sugar (Leenakul and Tippayawong, 2010; Luo et al., 2013; Sathitsuksanoh et al., 2010; Xiao et al., 2014). Limited investigation has been conducted in extracting and recovering silica from bamboo. Additionally, bamboo silica, like silica in cereal straw or rice husks, could be an excellent resource for silica-based products such as silica gel and silica nanoparticles, which could be used in various high-value products such as thixotropic agents, pharmaceuticals, film substrates, electric and thermal insulators composite filler, etc. (Kalapathy et al., 2002; Zhang et al., 2013). Within this context, the pre-extraction of silica prior to subsequent commercial pulping may not only be able to solve the silica problems of using bamboo in kraft pulping but also would add value and increase revenue to the mill. Therefore, the kraft-based bamboo pulping process could be improved by adding a pre-extraction stage. In addition, even though bamboo belongs to the grass family, having a woody stem it is structurally quite different from other lignocellulosic materials such as softwoods, hardwoods and cereal straw. Bamboo stem is composed of three parts: epidermal (outmost cell layer of the stem), mid-cortex (the region between epidermal layer and inner cortex), and inner cortex (the portion encircling the hollow center of the culm). Due to the unique location in the bamboo stem and their functions, these three regions differ substantially in chemical composition including extractives, ash and silica content (Chand et al., 2006). These differences will affect the bamboo feedstock processing methods in kraft-based pulping and biorefinery applications. It is thought that it may be beneficial to remove one or two regions of the bamboo stem (bamboo epidermal or inner cortex) before pulping. However, the removal of bamboo parts as waste will reduce the useful biomass fraction and cause environmental problems (Li et al., 2014). Accordingly, before optimizing the pulping processes, the chemical composition and silica mass distribution in the bamboo stem needs to be quantified. In the work reported in this paper, the chemical composition and silica mass distribution of the bamboo stem (Neosinocalamus affinis Keng) were quantified to provide insights into promising bamboo stem processing methods for the preparation of bamboo chips. Subsequently, alkaline pre-treatment was carried out to completely extract silica and partially separate hemicelluloses from bamboo chips prior to kraft pulping. The objectives of the work described here were to study the effects of pre-treatment conditions (NaOH charge, temperature and time) on silica and hemicelluloses preextraction and to investigate the effect of alkaline pre-extraction on the properties of kraft pulps such as silica content, yield, freeness, and tensile and tear strength indices by comparing with the properties of the non-extracted counterpart. A modified pulping process was proposed and a mass balance of the main components from pre-extraction to kraft pulping was established. Our results suggest that the modified pulping scheme fits well with the IFBR concept.

2. Materials and methods 2.1. Raw materials and chemicals Fresh bamboo trees (5 years old) of Neosinocalamus affinis Keng were collected in September, 2012 from a natural forest in Sichuan Province, China. Each bamboo stem was evenly cut into three parts along the length of the bamboo stem: top-middle-bottom. All samples were washed thoroughly with distilled water to remove the dust and other impurities from the surface. This washing operation was performed 10 times, and then samples were dried for 48 h at

67

room temperature. Subsequently, the samples were cut into small strips with a razor blade and ground using a Wiley Mill. The sample powder that passed through a 40-mesh sieve yet retained on a 60-mesh sieve was collected in glass jars for further use. For the analysis of chemical composition in different layers along the radial direction, the stems were separated manually with a plane block (Lee Valley Tools Ltd. Canada) to three layers: epidermal layer (≈0.1 mm), mid-cortex (≈3 mm) and inner cortex (≈0.2 mm). The three fractions of bamboo samples were ground, screened and stored as previously described. Bamboo chips prepared from 3 to 7 year old trees were provided by the Lee & Man Paper Manufacturing Ltd. China. The obtained chips were washed with distilled water at a liquid-to-wood ratio of 20 L/kg using a laboratory mixer to remove impurities, such as soil and sand. The washed chips were air dried for approximately 24 h and stored at 4 ◦ C until used for subsequent alkaline pretreatment and kraft cooking experiments. The chips were analyzed with respect to lignin content, carbohydrate composition, extractives, ash and silica content. All chemicals used in this study were reagent grade and were purchased from Fisher Scientific, Canada.

2.2. Alkaline pre-treatments Alkaline pre-extraction experiments were carried out in 4 silicate glass bottles of 2 L capacity immersed in a laboratory-scale heated oil bath. A series of alkaline pre-extraction experiments were conducted over two temperatures (80 and 100 ◦ C), sodium hydroxide (NaOH) charges (6–18%) and reaction times (1–5 h). In all alkaline pre-extraction experiments, the liquid-to-wood ratio was kept constant at 8 L/kg. For an alkaline pre-treatment run, bamboo chips of 100 g oven dried (o.d.) and the calculated volume of distilled water and NaOH solution (stock concentration of 100 g/L) were mixed and placed in a reactor. Subsequently, the reactor was placed in the oil bath pre-heated to the target temperature. After the pre-treatment, the vessels were rapidly cooled down in an ice/water bath. The pre-treated bamboo chips were separated from the liquor through vacuum filtration. The liquor was collected and stored at 4 ◦ C for the composition analysis. The treated chips were washed thoroughly with distilled water and stored at 4 ◦ C for component analysis and kraft cooking. All experiments were performed in triplicate.

2.3. Kraft pulping Pulping of pre-extracted and untreated (without pre-extraction) bamboo chips by the kraft process was conducted in four 300 mL stainless steel reactors in an oil bath under conditions covering the range of practical interest. For all cooks, the temperature was raised to 165 ◦ C in 85 min and held at 165 ◦ C for 75 min. The liquid-towood ratio and sulfidity were fixed at 4 L/kg and 25% (percentage of Na2 S, expressed as Na2 O), respectively. The effective alkali (EA) (expressed as Na2 O) was varied in the range of 13–19% (based on oven dried wood mass). For each kraft cook, 45 g o.d. extracted or non-extracted bamboo chips and the calculated volume of cooking chemicals and distilled water were placed in the reactor and mixed for 10 min. Afterwards, the cooking process was carried out according to the conditions being investigated. Upon completion of a cook, the reactor was rapidly cooled and kraft pulp was recovered using vacuum filtration. The kraft pulp was thoroughly washed with distilled water until the pH of the filtrate reached neutral. Then, the pulp was disintegrated, screened, and filtered to measure total yield, screened yield, and rejects of the kraft cooking. All experiments were performed in triplicate.

68

Z. Yuan et al. / Industrial Crops and Products 91 (2016) 66–75

2.4. Analytical methods The moisture content of solid samples was measured by drying at 105 ± 2 ◦ C to constant weight. The contents of water and solvent extractives of bamboo culm samples and bamboo chips were determined using a Soxhlet extractor according to TAPPI T 204 cm-97. Carbohydrates and lignin content of the solids was determined after air drying. Lignin content was calculated as the sum of Klason lignin and acid soluble lignin. Klason lignin content of bamboo chips and pulp was determined using National Renewable Energy Laboratory (NREL) standard protocols (Sluiter et al., 2012). Briefly, 40–60 mesh samples were subjected to a two-step sulfuric acid (H2 SO4 ) hydrolysis protocol to digest the polysaccharides into monomeric sugars. After hydrolysis, Klason lignin was separated through filtration and weighed after drying at 105 ± 2 ◦ C. Acid soluble lignin in the hydrolysate (after removing Klason lignin) was measured at wavelength 205 nm using a UV–vis spectrophotometer (Dence, 1992). Monomeric sugars were determined using a Dionex ICS 5000+ HPLC (high performance liquid chromatography) system equipped with an AS-AP autosampler and an electrochemical detector (Thermo Fisher Scientific, MA, USA) following the NREL methods (Sluiter et al., 2012). The monomeric sugars were separated on a Dionex Carbopac SA10 analytical column (Thermo Fisher Scientific, MA, USA) at 45 ◦ C using 1 mM NaOH as the mobile phase. Fucose was used an internal standard. The ash content of bamboo chips and pulp was determined according to TAPPI T211 om-02. Silica content of bamboo chips and pulp was measured gravimetrically, using a method modified from Ding et al. (2008). Briefly, about 5 g of dried and powdered bamboo sample was completely ashed at 550 ◦ C. After cooling, 10 mL of dilute HCl (6 mol/L) was added to the ash to precipitate silica and dissolve acid-soluble ash. The resultant solution was gently boiled to near dryness in a boiling water bath. HCl treatment was repeated three times in about 30 min, after which another 15 mL of dilute HCl (6 mol/L) was added to the solution. After 2 more min, the solution was filtered off through No. 42 ashless filter paper (Fisher Scientific, Canada). The precipitate was rinsed 5–6 times with 1 mol/L HCl solution and 5–6 times with hot distilled water (≈50 ◦ C). Both the filter paper with the precipitate was ashed at 700 ◦ C and calcined at 1000 ◦ C in a muffle furnace to reach a constant weight. The resultant silica residue was weighed to determine silica and ash content. All measurements were run in triplicate. Before measuring the properties of the extracts and black liquors, the liquors were filtered to remove potential fines material. Total solid content of the extracts and black liquors was determined by vacuum drying at 45 ◦ C for 48 h. Silica content of liquors was measured by using the silicon molybdenum blue photometric method (Tong et al., 2005). For the determination of lignin and carbohydrates content of the extracts, the filtered AELs were neutralized using dilute sulfuric acid. Then the samples were autoclaved with 4% (w/w) H2 SO4 for 60 min. The analysis was continued as described for the analysis of the solid samples. Lignin content of the black liquors was determined gravimetrically by acid precipitation and centrifugation (Rocha et al., 2012). The chemical contents in extracts were determined in mg/ml and then converted into percentage of original oven dry wood weight by multiplying with the value of the liquid-to-wood ratio divided by 10. 2.5. Evaluation of pulps The kappa number of screened pulps was determined according to TAPPI T236 om-99. The fines content of the pulps was measured with a Fibre Quality Analyzer (Op Test Equipment Inc., ON, Canada) based on TAPPI T271 om-07. The pre-extracted and untreated bamboo pulps were beaten in a laboratory disc refiner (PFI mill) at

different revolutions according to TAPPI T248 sp-00. The freeness (drainability) of the pulps was determined according to TAPPI T227 om-99 (Canadian Standard Method). Standard handsheets of about 60 g/m2 were made by TAPPI T 205 sp-02. The handsheets were tested for tensile and tear strength properties using TAPPI T 220 sp-01.

3. Results and discussion 3.1. Characterization of the raw material Table 1 shows the results of the analysis of chemical composition of different fractions along the length (top, middle, bottom) of the bamboo stem and layers (epidermal, mid-cortex, inner cortex) across the cross section of the original bamboo stem. The composition of each component analyzed (glucan, xylan, galactan, arabinan, lignin, extractives, ash, silica), is expressed as the average mass percentage of this component in the oven dry solids, determined from at least three tests. With regards to the composition along the length of the bamboo stem, the cellulose and lignin content of the three fractions along the length of the stem (top, middle, bottom) were similar at around 46.5% and 23.5%, respectively, while the hemicellulose (mainly xylan) content of the bamboo at the top of the stem (24.8%) was the highest, followed by that in the middle (23.4%). Hemicellulose content was the lowest at the bottom of the stem (21.4%). Arabinan and galactan content in all three parts were less than 1% while mannan was undetectable by the HPLC methodology used. The main difference in chemical composition among the three parts along the stem was found in the silica content. At the top of the stem the silica content was 0.91% which decreased by almost 50% to a value of 0.47% at the bottom of the stem. This is in agreement with previous studies that reported decreasing silica levels from the apical to basal portions of the stem (Collin et al., 2012). The most important observation was that the ratio of silica mass in the three parts was about 1:1:1 due to the difference in the weight fraction of bamboo biomass in each region. The chemical composition and silica mass distribution of different layers (epidermal, mid-cortex, inner cortex) along the radial direction of the stem are shown in Table 1. It was found that the glucan content of the bamboo mid-cortex (49.4%) was much higher than that of the epidermal layer (44.3%) and the inner cortex (43.3%). Lignin content was found to be highest in the bamboo epidermal portion (28.5%), followed by the inner cortex (24.1%) and mid-cortex (23.2%). The hemicellulose (xylan, arabinan and galactan) content in the epidermal region, mid-cortex, and inner cortex were comparable at 22.3%, 23.5% and 24.4%, respectively. The highest extractives content was in the bamboo inner cortex (8.4%), followed by the mid-cortex (5.2%) and the epidermal layer (4.8%). The ash and silica contents of the bamboo epidermal part were higher than those of the bamboo mid-cortex and inner cortex. The difference in silica content (1.3%) was especially high with that in the epidermal region being about seven-times higher than in the mid-cortex or the bamboo inner cortex. However, the majority of silica mass in the bamboo was located in bamboo mid-cortex (63%) because this region accounts for about 88% of the biomass while the epidermal portion and the inner cortex account for only 6% each. According to the composition analysis results, the removal of bamboo epidermal layer or inner cortex does not result in a significant decrease of silica amount input into the pulping processes; in contrast, it would increase the capital cost and cause environmental problems by the increasing industrial wastes. Moreover, compared with bamboo mid-cortex, bamboo epidermal part and inner cortex have comparable concentrations of hemicelluloses, which can be pre-extracted for bioconversion. Consequently, rather than removing biomass of epidermal region and inner cortex the

Z. Yuan et al. / Industrial Crops and Products 91 (2016) 66–75

69

Table 1 Chemical composition of original bamboo stem (Neosinocalamus Affinis Keng).

Length direction Cross section

Location

Glucan (%)

Top Middle Bottom Epidermal Mid-cortex Inner cortex

46.57 46.71 46.89 44.32 49.44 43.35

± ± ± ± ± ±

1.52 1.85 1.94 1.56 2.12 1.85

Xylan (%)

Galactan (%)

24.82 ± 1.02 23.43 ± 0.87 21.41 ± 0.98 21.53 ± 1.12 22.61 ± 1.26 23.37 ± 1.55

0.58 0.44 0.35 0.18 0.20 0.20

± ± ± ± ± ±

0.12 0.25 0.14 0.04 0.04 0.02

Arabinan (%) 0.63 0.52 0.41 0.62 0.77 0.91

± ± ± ± ± ±

Table 2 Chemical characteristics of commercial bamboo chips (Neosinocalamus Affinis Keng). Property

Bamboo

Moso Bamboo (Li et al., 2012)

Aspen (Jun et al., 2012)

Loblolly Pine (Huang and Ragauskas, 2013)

Glucan, % Xylan, % Galactan, % Arabinan, % Mannan, % Klason lignin, % Acid soluble lignin, % Extractives, % Ash, % Silica, %

47.30 20.30 0.52 0.84 – 24.40 0.91

44.4 22.9 – 0.2 – 22.0 –

44.10 19.60 1.80 0.80 2.00 21.2 3.02

45.50 7.30 2.10 1.40 9.80 27.70 –

4.60 2.11 1.12

7.10 2.40 –

– 0.60 –

3.20 – –

whole bamboo stem should be used as the raw material in pulping or biorefinery processes. According to above results, bamboo chips were prepared from the bamboo stem. Table 2 compares the chemical composition of the commercial bamboo chips used in this work to that reported in the literature for other cellulosic biomasses such as different bamboo species, softwood and hardwood. As shown in Table 2, glucan content of the commercial chips was 47.3%, which was a little higher than the reported data for Moso bamboo and the woods (Huang and Ragauskas, 2013; Jun et al., 2012; Li et al., 2012). The higher glucan content should translate into a higher chemical pulping yield. In terms of the hemicelluloses in bamboo isxylan rich with a similar xylan content (20%) to that of the hardwood, aspen. The Klason lignin content in the studied bamboo was 24.4%, which was slightly higher than other reported results for Moso bamboo and aspen, but lower than that of softwood (Loblolly Pine) (Table 2). Lower lignin content means easier delignification in kraft pulping process. Water and solvent extractives were 4.6%, lower than the 7.1% found for Moso bamboo (Li et al., 2012). The ash and silica content of bamboo chips (2.11% and 1.12%, respectively) were much higher than those of woods in which the silica content is less than ˇ 0.01% while the ash content is 0.3–1.0% (Torelli and Cufar, 1995). These data illustrate both the advantage and challenge of using bamboo as an alternative feedstock for pulping applications. In terms of the challenge, for example, on a daily basis, approximately 34 tons/day of soluble silicates will be generated from a 1500 ton chips/day bamboo kraft pulp mill and 70% of these silicates will be introduced into the chemical recovery circuit of a typical kraft mill using bamboo (Salmela et al., 2008). In addition to the challenges posed by silica, high overall ash content will interfere with other process steps; for example high ash levels can lead to acid neutralization reactions during pre-hydrolysis with dilute acid or water/steam used to remove hemicellulose in dissolving pulp and biofuels production (Kapu and Trajano, 2014). 3.2. Alkaline extraction of bamboo chips In this study, the goal of alkaline pre-treatment prior to pulping is to extract maximum silica and hemicelluloses from commercial

0.05 0.11 0.13 0.13 0.14 0.14

Lignin (%) 23.14 23.75 23.56 28.52 23.21 24.08

± ± ± ± ± ±

0.21 0.32 0.20 0.22 0.31 0.26

Extractives (%) 5.64 4.12 3.83 4.78 5.24 8.41

± ± ± ± ± ±

0.72 0.64 0.81 0.44 0.36 0.29

Ash (%) 1.35 1.31 1.28 2.12 0.93 1.26

± ± ± ± ± ±

Silica (%) 0.10 0.12 0.13 0.08 0.04 0.04

0.91 0.60 0.47 1.32 0.17 0.20

± ± ± ± ± ±

0.05 0.02 0.04 0.03 0.02 0.05

Silica mass fraction 0.36 0.34 0.30 0.32 0.63 0.05

bamboo chips while minimizing the loss of cellulose and lignin. The presence of lignin in the spent pre-treament liquor hampers its utilization in bioethanol or xylitol production because lignin degradation products inhibit the growth and metabolic activity of micro-organisms used in bioconversion processes. Moreover, silica recovery through lowering the pH of spent liquor is negatively affected due to co-precipitation of lignin (Minu et al., 2012; Shi et al., 2011). Since high temperature low alkali charge preextraction processes have several drawbacks such as high capital investment cost and the low molecular-mass of the hemicelluloses extracted-which renders the hemicelluloses difficult to concentrate and separate (Jun et al., 2012; Yoon et al., 2011), high alkali charge (6–18%) and relatively lower reaction temperatures were investigated in this study. Table 3 shows the results of extractions in terms of chemical composition of pre-treated chips prior to pulping. Results indicated that the yield depended on all three factors, extraction temperature, NaOH charge and time. For example, the use of 6% NaOH resulted in chip yields between 92% and 97% at 80 ◦ C while the yield decreased to 88–94% at 100 ◦ C. With regards to the composition changes, as expected, alkaline extraction at low temperatures (<100 ◦ C) resulted in little loss of cellulose and lignin while significantly reducing the hemicelluloses (xylan) and silica contents. On the basis of the chip yield and composition analysis, the actual loss of the different bamboo components during the alkaline extraction was calculated (calculation not shown). By increasing the alkali charge up to 18% (based on the original chips), the calculated cellulose mass fraction loss was 0.1–1.2% (based on the o.d. original chip mass), which would not significantly affect the pulping yield. Its high crystallinity and limited accessibility towards chemicals make cellulose very recalcitrant towards degradation under mild conditions such as those utilized in this study (Engström et al., 2006). The mass fraction loss of lignin, calculated from the data presented in Table 3, was 0.2–1.3% (based on the original o.d. chip mass). The results show that the treatment time and temperature used in this study did not significantly degrade the lignin in bamboo chips, which means the lignin co-precipitation can be avoided during the silica recovery from the AEL, resulting in high purity silica particles. The loss of galactan and arabinan, did not contribute much to yield loss as the content in the starting material was low (total mass fraction less than 1.5% in raw chips); the majority of the extracted hemicelluloses was xylan. According to the data shown in Tables 2 and 3, it can be concluded that up to 50% of the original xylan in raw chips was extracted during alkaline pre-treatment and that the amount of extracted xylan increased with increasing NaOH charge, reaction temperature and reaction time. For example, about 30% of original xylan mass was extracted by treating bamboo chips at 80 ◦ C with 18% NaOH charge for 5 h. In contrast, at 100 ◦ C with the same NaOH charge (18%), about 50% of xylan was extracted in 5 h (Table 3). With alkaline extraction, the amount of silica removal from bamboo chips increased with increasing NaOH charge and temperature. For example, treating bamboo chips for 5 h with 6% NaOH

70

Z. Yuan et al. / Industrial Crops and Products 91 (2016) 66–75

Table 3 Chemical composition of bamboo chips after pre-treatments with NaOH under varying conditions. Tmax (◦ C)

NaOH charge (%)

Time at Tmax (h)

Yield (%)

Glucan (%)

Xylan (%)

Galactan (%)

Arabinan (%)

Lignin (%)

Silica (%)

80

6

1 2 3 4 5

96.5 95.2 94.3 93.1 92.0

48.85 ± 0.65 49.26 ± 0.44 49.82 ± 0.59 50.38 ± 0.28 50.94 ± 0.45

20.65 ± 0.41 20.17 ± 0.33 19.62 ± 0.42 19.33 ± 0.51 19.13 ± 0.44

0.52 ± 0.05 0.54 ± 0.06 0.51 ± 0.06 0.59 ± 0.05 0.50 ± 0.08

0.81 ± 0.04 0.76 ± 0.08 0.71 ± 0.07 0.72 ± 0.1 0.71 ± 0.06

26.18 ± 0.24 26.11 ± 0.36 26.20 ± 0.25 26.38 ± 0.33 26.53 ± 0.42

0.58 ± 0.03 0.36 ± 0.02 0.23 ± 0.01 0.21 ± 0.02 0.18 ± 0.01

12

1 2 3 4 5

93.5 92.2 90.7 89.1 88.4

50.25 ± 0.84 51.78 ± 0.56 51.65 ± 0.47 52.61 ± 0.65 52.93 ± 0.42

20.86 ± 0.43 19.85 ± 0.55 18.74 ± 0.67 17.73 ± 0.49 17.26 ± 0.51

0.49 ± 0.07 0.56 ± 0.04 0.58 ± 0.05 0.60 ± 0.08 0.41 ± 0.04

0.73 ± 0.06 0.70 ± 0.05 0.65 ± 0.07 0.61 ± 0.08 0.64 ± 0.04

26.75 ± 0.26 26.85 ± 0.37 27.13 ± 0.22 27.45 ± 0.56 27.50 ± 0.89

0.42 ± 0.02 0.23 ± 0.01 0.14 ± 0.02 0.08 ± 0.01 0.01 ± 0.01

18

1 2 3 4 5

92.4 90.3 88.5 85.7 84.3

50.57 ± 0.53 51.79 ± 0.78 52.87 ± 0.82 54.66 ± 0.43 55.54 ± 0.65

20.67 ± 0.42 19.58 ± 0.68 18.34 ± 0.79 17.59 ± 0.46 16.48 ± 0.54

0.42 ± 0.05 0.47 ± 0.05 0.36 ± 0.05 0.38 ± 0.04 0.43 ± 0.06

0.68 ± 0.07 0.67 ± 0.04 0.72 ± 0.05 0.76 ± 0.06 0.73 ± 0.05

26.96 ± 0.48 27.36 ± 0.75 27.75 ± 0.64 28.48 ± 0.87 28.78 ± 0.92

0.25 ± 0.03 0.10 ± 0.02 0.05 ± 0.02 0.02 ± 0.01 ND

6

1 2 3 4 5

93.7 92.1 90.4 88.7 87.7

49.95 ± 0.55 50.84 ± 0.68 51.79 ± 0.58 52.76 ± 0.66 53.37 ± 0.44

20.76 ± 0.53 20.04 ± 0.51 19.26 ± 0.42 18.49 ± 0.65 17.45 ± 0.37

0.37 ± 0.08 0.39 ± 0.09 0.33 ± 0.07 0.52 ± 0.05 0.43 ± 0.06

0.79 ± 0.03 0.74 ± 0.07 0.71 ± 0.1 0.65 ± 0.06 0.70 ± 0.08

26.80 ± 0.55 27.10 ± 0.46 27.33 ± 0.72 27.63 ± 0.65 27.72 ± 0.26

0.34 ± 0.02 0.27 ± 0.01 0.19 ± 0.01 0.14 ± 0.01 0.11 ± 0.03

12

1 2 3 4 5

91.6 90.3 88.5 86.1 84.2

51.13 ± 0.67 51.79 ± 0.62 52.98 ± 0.73 54.22 ± 0.53 55.47 ± 0.42

20.56 ± 0.59 19.08 ± 0.48 17.74 ± 0.55 16.58 ± 0.44 15.31 ± 0.46

0.48 ± 0.04 0.43 ± 0.07 0.58 ± 0.1 0.62 ± 0.04 0.51 ± 0.06

0.77 ± 0.06 0.71 ± 0.04 0.64 ± 0.04 0.60 ± 0.03 0.55 ± 0.05

27.30 ± 0.25 27.48 ± 0.42 27.81 ± 0.81 28.35 ± 0.53 28.75 ± 0.46

0.23 ± 0.02 0.16 ± 0.02 0.08 ± 0.01 0.02 ± 0.005 ND

18

1 2 3 4 5

89.4 87.2 85.8 84.5 82.3

52.32 ± 0.34 53.58 ± 0.21 54.39 ± 0.41 54.16 ± 0.54 55.07 ± 0.64

20.51 ± 0.53 18.68 ± 0.52 16.66 ± 0.61 14.62 ± 0.45 12.06 ± 0.38

0.39 ± 0.05 0.37 ± 0.07 0.44 ± 0.08 0.43 ± 0.06 0.55 ± 0.07

0.70 ± 0.10 0.64 ± 0.09 0.60 ± 0.07 0.57 ± 0.03 0.59 ± 0.06

27.98 ± 0.25 28.34 ± 0.38 28.57 ± 0.59 28.77 ± 0.57 29.10 ± 0.64

0.05 ± 0.02 0.01 ± 0.004 ND ND ND

100

Note: ND-not detected. Yield-weight percentage of recovered mass after pre-treatment to non pre-extracted biomass, both in dry matter.

charge at 80 ◦ C and 100 ◦ C, about 85% and 90% of initial silica in bamboo chips could be removed, respectively. In contrast, at 80 ◦ C using 12% and 18% NaOH charge, up to 95% of silica could be removed within 3 h. When treating bamboo chips at 100 ◦ C with 18% NaOH charge, nearly 96% of silica was removed in 1 h. As shown in Table 3, after the removal of more than 96% of silica mass, the silica content of treated chips was about 0.04% or even less (based on treated o.d. chip mass), which means that even the silica impact on high purity dissolving-grade pulp can be eliminated (Sixta, 2006). With such low amount of silica (≤0.04%) in treated bamboo chips, the adverse effect of silica on the chemical recovery process of kraft pulping can be significantly resolved. Therefore, treated bamboo chips with more than 96% of silica removal could be readily used for subsequent production of kraft pulp or dissolving-grade pulp. 3.3. Analysis of the AEL To further assess the effects of alkaline pre-treatment on the dissolution of bamboo components, the sugars, soluble lignin and silica in the extract liquors were analyzed (Table 4). An interesting observation is that the sum of chip yield (Table 3) and solid contents of the AEL (Table 4) was larger than 100% in all experimental runs. The excess values of the mass balance might be partly due to the sodium ions bound to dissolved components such as acetic or uronic acid. On the other hand, the dissolved sodium compounds determined as residual alkali (data not shown) in the AEL also contributed to the total solid content. As shown in Table 4, varying the severity of the pre-treatment had little effect on the extraction of glucan and lignin from bamboo chips into the AEL. The low glucan (0.1–0.9%) and lignin (0.1–1.2%) content (based on original o.d. chip mass) in the AEL also confirmed that the cellulose and lignin are resistant to the studied alkaline pre-extraction con-

ditions. The low lignin content in the AEL not only preserves the heating value of black liquor obtained from kraft pulping but also avoids the lignin co-precipitation during silica recovery process. Thus, the precipitated silica could be easily separated and purified for further applications such as production of silica nanoparticles or silica gel. Clearly, this alkaline pre-treatment favored the silica removal and recovery during kraft pulping of bamboo chips. The other goal of the alkaline pre-treatment was to extract hemicelluloses prior to pulping. Even at extraction conditions that resulted in larger than 96% silica removal, the contents of galactan and arabinan in the AEL were less than 0.6% of the original o.d. chip mass; in contrast, depending on the alkaline pre-treatment conditions employed, the xylan content in the AEL increased significantly from 4.7% to 9.2% (based on the original o.d. chip mass), which corresponded to 23–45% of initial xylan in bamboo chips. Moreover, the data in Tables 3 and 4 show that the hemicellulose content in both biomass residuals and AELs was 94.2–98.6% of initial hemicelluloses in raw chips for all experimental runs, showing a reasonable agreement of mass balance. The 2–6% xylan not accounted for might be lost during chip washing after the pre-treatment or through xylan degradation into products undetected by the methodology used. In the alkaline pre-extraction process, the ester linkages between polysaccharides and lignin were hydrolyzed (Lehto and Alén, 2013), resulting in the release of hemicelluloses. Moreover, it has been reported that the hemicelluloses extracted with alkaline pre-treatment were mostly in the oligomeric or polymeric form due to the stopping of alkaline peeling reactions at the branches in the molecular chains (Taherzadeh and Karimi, 2008; Yoon and Van Heiningen, 2010). Compared to hemicelluloses in the monomeric form, hemicelluloses in the oligomeric or polymeric form have potential higher economic applications such as barrier films, hydro-

Z. Yuan et al. / Industrial Crops and Products 91 (2016) 66–75

71

Table 4 Sugars, soluble lignin and silica in alkaline pre-extraction liquors (% of original oven dried chip mass). Tmax (◦ C)

NaOH charge (%)

Time at Tmax (h)

Glucan (%)

Xylan (%)

Galactan (%)

Arabinan (%)

Lignin (%)

Silica (%)

TS (%)

80

6

1 2 3 4 5

0.18 ± 0.11 0.40 ± 0.09 0.36 ± 0.18 0.41 ± 0.08 0.45 ± 0.12

0.34 ± 0.12 0.79 ± 0.25 1.48 ± 0.23 2.05 ± 0.20 2.40 ± 0.18

0.01 ± 0.01 ND 0.02 ± 0.01 0.02 ± 0.01 0.05 ± 0.01

0.05 ± 0.01 0.11 ± 0.02 0.15 ± 0.02 0.15 ± 0.02 0.15 ± 0.03

ND 0.40 ± 0.05 0.50 ± 0.10 0.65 ± 0.10 0.85 ± 0.05

0.52 ± 0.04 0.75 ± 0.02 0.88 ± 0.02 0.90 ± 0.02 0.92 ± 0.03

8.44 9.76 10.66 11.82 12.68

12

1 2 3 4 5

0.35 ± 0.21 0.51 ± 0.17 0.52 ± 0.12 0.53 ± 0.07 0.60 ± 0.18

0.50 ± 0.10 1.80 ± 0.20 3.10 ± 0.21 4.24 ± 0.10 4.66 ± 0.05

ND ND ND ND 0.12 ± 0.03

0.12 ± 0.03 0.17 ± 0.02 0.24 ± 0.01 0.25 ± 0.04 0.25 ± 0.02

0.25 ± 0.05 0.45 ± 0.11 0.65 ± 0.05 0.78 ± 0.08 0.92 ± 0.12

0.70 ± 0.02 0.88 ± 0.02 0.96 ± 0.03 1.02 ± 0.03 1.08 ± 0.02

17.24 18.75 20.21 21.82 22.13

18

1 2 3 4 5

0.55 ± 0.09 0.60 ± 0.13 0.60 ± 0.05 0.58 ± 0.04 0.61 ± 0.14

1.04 ± 0.08 2.46 ± 0.16 4.01 ± 0.07 5.06 ± 0.14 6.13 ± 0.15

0.11 ± 0.02 0.07 ± 0.03 0.15 ± 0.04 0.17 ± 0.02 0.14 ± 0.02

0.18 ± 0.03 0.20 ± 0.03 0.18 ± 0.02 0.17 ± 0.02 0.20 ± 0.02

0.36 ± 0.03 0.58 ± 0.02 0.70 ± 0.05 0.80 ± 0.13 0.94 ± 0.10

0.86 ± 0.03 1.01 ± 0.02 1.02 ± 0.04 1.07 ± 0.03 1.10 ± 0.02

24.49 25.68 27.39 30.28 31.59

6

1 2 3 4 5

0.44 ± 0.06 0.44 ± 0.10 0.41 ± 0.05 0.40 ± 0.10 0.50 ± 0.15

0.71 ± 0.10 1.60 ± 0.05 2.74 ± 0.11 3.73 ± 0.12 4.85 ± 0.10

0.13 ± 0.04 0.14 ± 0.02 0.18 ± 0.04 ND 0.10 ± 0.02

0.08 ± 0.02 0.13 ± 0.03 0.18 ± 0.02 0.23 ± 0.03 0.20 ± 0.02

0.15 ± 0.04 0.30 ± 0.05 0.60 ± 0.02 0.75 ± 0.05 0.90 ± 0.10

0.75 ± 0.05 0.85 ± 0.02 0.90 ± 0.05 0.96 ± 0.03 1.01 ± 0.03

12.26 12.86 14.56 16.27 17.19

12

1 2 3 4 5

0.50 ± 0.13 0.56 ± 0.13 0.50 ± 0.18 0.65 ± 0.10 0.75 ± 0.10

1.20 ± 0.06 3.01 ± 0.14 4.44 ± 0.10 5.76 ± 0.14 7.21 ± 0.08

0.06 ± 0.03 0.10 ± 0.03 ND ND 0.07 ± 0.02

0.10 ± 0.03 0.18 ± 0.02 0.24 ± 0.04 0.28 ± 0.04 0.34 ± 0.03

0.20 ± 0.12 0.41 ± 0.08 0.64 ± 0.07 0.80 ± 0.10 1.02 ± 0.12

0.87 ± 0.04 0.94 ± 0.03 1.02 ± 0.03 1.06 ± 0.0.05 1.09 ± 0.03

19.32 20.65 22.43 23.87 25.72

18

1 2 3 4 5

0.62 ± 0.05 0.65 ± 0.13 0.69 ± 0.17 0.79 ± 0.14 0.91 ± 0.21

1.69 ± 0.05 3.84 ± 0.05 5.70 ± 0.16 7.80 ± 0.20 9.20 ± 0.10

0.14 ± 0.03 0.16 ± 0.03 0.12 ± 0.02 0.14 ± 0.02 0.05 ± 0.02

0.19 ± 0.02 0.25 ± 0.03 0.27 ± 0.05 0.30 ± 0.05 0.31 ± 0.04

0.26± 0.03 0.55 ± 0.06 0.74 ± 0.05 0.92 ± 0.09 1.20 ± 0.05

1.05 ± 0.02 1.08 ± 0.03 1.10 ± 0.03 1.10 ± 0.03 1.10 ± 0.03

27.48 29.69 31.08 32.39 32.59

100

Note: ND-not detected. TS-total solids.

gel, paper additives, and biofuels (Persson et al., 2009; Chirat et al., 2012; Hamzeh et al., 2013; Bai et al., 2012). Detailed experiments on the recovery of silica and hemicelluloses from the AEL are being carried out and will be described in a subsequent paper. The AEL after silica and hemicelluloses removal would be sent to the chemical recovery circuit of a typical kraft process to recover inorganic chemicals such as sodium hydroxide. The impact of the pre-treatments on subsequent kraft pulping is discussed below. 3.4. Kraft pulping Chips from three representative alkaline pre-treatment runs with residual silica content less than 0.05% were selected for subsequent kraft pulping. The three pre-treatment runs were those using 18% NaOH at 100 ◦ C for 1 and 5 h and 18% NaOH at 80 ◦ C for 3 h. To improve the efficiency when applied in a mill process, higher NaOH charge and shorter reaction times were used for the pre-treatment These conditions have the potential to extract higher levels of hemicelluloses and silica with less cellulose degradation (see previous discussion in Sections 3.2 and 3.3). Since increasing effective alkali (EA) charge has been considered to be more important than increasing H-factor in removing lignin during the kraft pulping (Sixta, 2006), EA charge was chosen as the variable while reaction temperature and time were kept constant. In the kraft cooking process, four levels of effective alkali (EA) charge were studied to investigate the effectiveness of alkaline pre-extraction on the kraft pulping of bamboo. To compare the kraft pulping process of extracted bamboo chips and original chips on a uniform basis, the alkali charged to the treated chips was adjusted according to the analysis of the residual alkali in the extract liquor. Table 5 shows the effect of alkaline pre-extraction on the kraft pulping of bamboo chips. It should be noted that the pulping yield was expressed as the overall pulp yield (measured based on the

initial oven dry mass of bamboo chips). As shown in Table 5, at all EA charges, kraft pulps from alkaline pre-treated chips had lower kappa numbers (lower residual lignin content in the kraft pulp) than the non pre-extracted bamboo chips, showing that the alkaline pre-treatment had a positive effect on delignification during kraft cooking. This might be due to the fact that the removal of hemicelluloses/lignin during alkaline pre-treatment process resulted in chips having a more open structure thus improving the accessibility of the cooking chemicals to lignin in chips and improving the rate of diffusion of degraded lignin into the black liquor. For example, a kappa number of 16.0 could be achieved by pulping extracted chips at100 ◦ C for 1 h with 17% EA while 19% EA was needed when pulping with control chips (without pre-treatment) under the same conditions. Increasing the pre-treatment severity (longer time and higher temperature) also decreased the kappa number of pulp obtained under the same kraft cooking conditions (results of chips treated at 100 ◦ C for 1 and 5 h, respectively in Table 5). Moreover, lower kappa numbers (lower lignin contents) translate into lower demand for bleaching chemicals and, hence lower bleaching costs. The total pulp yield of extracted chips was generally slightly lower than that of the control under the same kraft cooking conditions (Table 5). It should be noted that pulps from extracted chips had lower kappa numbers (lower lignin content), which accounted for 0.3–0.6% of the pulp yield. In addition, a lower rejects content in the pulp could be obtained with alkaline pre-extracted (chipsbelow 0.5%) compared to the 1.3–1.5% of the controls; this could be related to the fact that extracted chips have a more open structure enabling better penetration of cooking chemicals resulting in a more even cook. The screened pulp yield obtained from the extracted chips was similar to that of the control or even higher. Total pulp yield decreased with increasing EA charge for both the pre-treated chips and control. Similar results have been obtained

72

Z. Yuan et al. / Industrial Crops and Products 91 (2016) 66–75

Table 5 Effect of alkaline pre-extraction and effective alkali charge on kraft pulping of bamboo. Pre-extracted bamboo chips

EA (%)

Rejectsa (%)

Total yielda (%)

Kappa number

Residual silica (%) in pulpa

Silica in BLa (%)

CSF (ml)

Fines (%)

Control

13 15 17 19

2.3 ± 0.2 1.8 ± 0.4 1.5 ± 0.5 1.3 ± 0.3

55.1 ± 1.4 53.9 ± 1.2 53.6 ± 1.7 53.1 ± 1.2

30.1 ± 0.5 22.6 ± 0.2 18.8 ± 0.6 15.9 ± 0.5

0.19 ± 0.03 0.16 ± 0.01 0.14 ± 0.02 0.13 ± 0.03

0.91 ± 0.03 0.95 ± 0.04 0.95 ± 0.03 0.97 ± 0.04

549 ± 10 538 ± 8 544 ± 12 531 ± 7

17.4 18.6 19.1 19.8

80 ◦ C 3h

13 15 17 19

0.5 ± 0.2 0.4 ± 0.3 0.2 ± 0.1 ≈0.1

54.3 ± 1.8 53.1 ± 1.2 52.2 ± 1.5 51.0 ± 0.8

27.2 ± 0.6 20.7 ± 0.7 16.5 ± 0.4 13.9 ± 0.2

≈0.01 ≈0.01 ND ND

≈0.01 ≈0.01 ≈0.04 ≈0.02

629 ± 12 624 ± 9 608 ± 13 618 ± 11

14.2 14.4 14.4 14.7

100 ◦ C 1h

13 15 17 19

0.3 ± 0.1 ≈0.1 ≈0.1 ND

52.3 ± 2.3 52.7 ± 1.3 52.1 ± 1.5 51.4 ± 1.2

27.6 ± 1.2 19.8 ± 0.9 16.4 ± 0.3 12.9 ± 0.8

≈0.01 ND ND ND

≈0.01 0.04 ± 0.02 ≈0.02 0.04 ± 0.01

635 ± 8 624 ± 8 619 ± 12 620 ± 11

13.8 14.0 14.4 14.6

100 ◦ C 5h

13 15 17 19

≈0.1 ND ND ND

52.2 ± 1.6 51.5 ± 1.9 50.9 ± 1.0 50.3 ± 1.0

22.4 ± 1.0 16.1 ± 0.7 13.8 ± 0.6 11.1 ± 0.4

ND ND ND ND

ND ND ND ND

649 ± 5 640 ± 16 620 ± 9 612 ± 12

13.1 13.6 14.0 14.2

a

Calculations were based on original oven dried chip mass. ND-not detected. BL-black liquor.

by kraft pulping of alkaline pre-extracted aspen chips (Jun et al., 2012). The initial kraft pulp (brownstock), drainage resistance is an important parameter as it strongly affects the downstream operations such as pulp washing. In this study, the drainage resistance of the brownstock was determined as Canadian Standard Freeness (CSF). As shown in Table 5, the CSF of pre-extracted pulps were in the range of 600–700 mL whereas the freeness of the non-extracted pulps ranged from 520–560 mL. Additionally, the measured amount of fines, determined as fibrous materials with sizes between 0.07 and 0.2 mm, was lower in pre-extracted pulps than in control pulps (Table 5); these results were in agreement with studies on kraft pulping of hemicellulose extracted sugar maple (Duarte et al., 2011). Therefore, the higher CSF values of extracted pulps were in accordance with the decrease in the fines content. Higher CSF means faster rates of water drainage during brownstock washing, which improves the mill efficiency. The most significant observation was the very low residual silica content in kraft pulp from pre-extracted chips. As can be seen in Table 5, the residual silica content of the pulp from pre-extracted chips was below 0.02% while it was 0.15–0.19% (based on original o.d. chip mass) in the control. One reason for the high residual silica content of the pulp from non-extracted chips might be that the silicates dissolved during pulping adhere onto the fiber surface and are not removed during subsequent pulp washing. High silica contents in kraft pulp make it unsuitable for use in high grade products such as ashless filter paper or facial tissue. In addition to the challenges of silica in the kraft pulp, high silica content in black liquor also causes problems in the chemical recovery process such as scaling of evaporators, decrease in causticizing efficiency, and the generation of large amount of solid waste (calcium silicate mixed with calcium carbonate). Moreover, silica in the black liquor is difficult to remove because of high lignin content. Therefore, alkaline pre-treatment is a promising approach to solve the silica problems when pulping bamboo chips.

3.5. Pulp physical properties It is important to assess the impact of the extraction process on the physical properties of resulting pulps. As indicated in Fig. 1 the initial freeness of the control pulp was much lower than the initial freeness of the pulps from chip pre-extracted with 19% EA. The rate of freeness drop with PFI refining was similar for all the pulps. These results indicate that pulps from alkaline pre-extracted chips

Fig. 1. Plot of pulp freeness (CSF) versus PFI mill revolution for pulps obtained from kraft pulping of extracted and non-extracted chips with 19% EA.

need more refining energy to attain the same level of freeness as the control. This agrees well with the generalized experience that pulps with low content of hemicelluloses and fines are difficult to beat to a target freeness due to the small degree of internal fibrillation with increased refining (Walton et al., 2010). Similar refining results were also obtained in kraft pulping with extracted and nonextracted chips with 15%, 17% and 19% EA (data not shown). The strength properties of all pulps were determined at the CSF of 425 mL. Plots of tensile and tear indices against EA charge of handsheets of pulps from pre-extracted chips and the control pulp are shown Figs. 2 and 3, respectively. Tensile strength index of pulps from alkaline pretreated bamboo chips initially increased with increasing EA charge, thereafter it decreased. For the control samples, the tensile index increased with increasing EA charge. These results confirmed the result that the delignification rate of pre-extracted chips was faster than the control. With the higher removal of lignin from treated chips at lower EA charge, more bonding could be formed among cellulosic fibres than in the case of the control, resulting in the improvement of physical properties of handsheets. However, with continually increasing the EA charge, compared to the control pulps, the handsheet strength of pulps from extracted bamboo chips decreased significantly. One

Z. Yuan et al. / Industrial Crops and Products 91 (2016) 66–75

73

removal of lignin and hemicelluloses making the resultant kraft pulps to contain more cellulose per gram handsheet. Results of the kraft pulping with extracted and non-extracted chips show that the alkaline pre-extraction is suitable for production of both papermaking fiber and dissolving pulps (high purity cellulose). Moreover, the pre-extraction process removes hemicelluloses and silica from the black liquor cycle which will improve the efficiency of chemical recovery. The hemicellulose and silica removed during the process can potentially be converted into other value-added products. 3.6. Proposed modification to kraft pulping process and mass balance

Fig. 2. Effect of pre-extraction and effective alkali charge on the tensile index of pulp.

Fig. 3. Effect of pre-extraction and effective alkali charge on the tear index of pulp.

likely reason is the lower hemicellulose content in extracted pulps, resulting in fewer bonding. It can be seen in Fig. 2 that the tensile strength index of pulps from chips extracted using milder pre-treatment conditions (80 ◦ C for 3 h and 100 ◦ C for 1 h) decreased at EA charges higher than 17%. This decrease began at 15% EA in the pulps from chips extracted at high severity (100 ◦ C for 5 h). This is probably due to the loss of different amounts of hemicelluloses during the alkaline pre-extraction processes. With the increase in the intensity of alkaline pre-extraction, more hemicelluloses were removed, resulting in higher cellulose/hemicelluloses ratio, which will form larger macrofibrils during handsheet making (Molin and Teder, 2002; Walton et al., 2010), resulting in lower bonding. In addition, at the highest tensile index of handsheets, hemicelluloses contents of the pulps were 7.01%, 5.45%, 5.78% and 5.12% (based on original o.d. chip mass) for non-treated chips and chips pre-treated under conditions of 80 ◦ C for 3 h, 100 ◦ C for 1 h, and 100 ◦ C for 5 h, respectively. Thus, it confirmed the assumption that hemicellulose content plays an important role in the strength properties of paper. Fig. 3 shows that the tear index of extracted pulp was better than that of non-extracted pulp. The overall improvement of tear strength with alkaline pre-extraction can be explained by the

Based on our findings, we propose a process for the integration of low temperature alkaline pre-extraction of silica and hemicelluloses from bamboo chips into a commercial kraft pulping process as illustrated in Fig. 4. In a typical kraft pulp mill, alkali is readily abundant as kraft white liquor, so the integration of this pre-extraction stage could be achieved without major capital investment or extensive process changes. In the proposed process scheme, washed bamboo chips are treated with alkali solution under atmospheric conditions from which a liquor rich in silica and hemicelluloses is produced. Silica and hemicellulose fractions are recovered from this AEL as potential starting materials for value added products. The treated AEL is mixed with the black liquor for the recovery of inorganic chemicals such as NaOH. Pre-extracted chips are then processed through kraft pulping or pre-hydrolysis kraft pulping to produce high-grade kraft pulp or dissolving pulp. Taking the 4 h pre-extraction at 100 ◦ C with 18% NaOH as an example, the mass balance of the main components of bamboo around the whole system from pre-treatment to kraft cooking can be determined (Fig. 4). Mass balance of each fraction is expressed in terms of dried material mass. With regards to the pre-treatment stage, the sum of recovered organics (cellulose, hemicelluloses, lignin) and inorganic (silica) in both biomass residual and AEL corresponded to 98.4% of those in the raw bamboo chips. The 1.6% material loss is due to the degradation of lignin and/or hemicelluloses into unidentified products. The analysis of the AEL showed that about 37% of hemicelluloses (mainly xylan) and 99% of silica contained in raw bamboo chips were extracted, showing a good revenue source for the mill and a novel way for solving the silica problems. Then pretreated bamboo chips (without washing) were subjected to kraft cooking, performed at 15% EA charge and 25% sulfidity. Based on 100 g o.d. mass of chips entering the preextraction-kraft pulping process, the mass of obtained pulp was determined to be 51.6 g, in which masses of cellulose, hemicelluloses, and lignin were 43.8 g, 4.8 g, and 2.7 g, respectively. The recovered masses of cellulose and hemicelluloes in the brownstock corresponded to 95.6% and 36.4% of the two components in the pretreated chips, respectively, showing reasonable agreement with previous studies (Pinto et al., 2005; Vu Mân et al., 2004). The losses of cellulose and hemicelluloses during delignification might be due to peeling reactions and alkaline degradation, respectively (Rocha et al., 2012). For the material balance of lignin during kraft pulping, as shown, the total amount of precipitated lignin in the black liquor and lignin in the final pulp was 19.9 g, which was equivalent to 81.9% of the lignin in pretreated chips. The 18.1% loss of lignin in raw chips during kraft cooking might be due to water washing of the kraft pulp and incomplete lignin precipitation during black liquor acidification. Finally, with regards to the main components (cellulose, hemicelluloses and lignin), the overall recovery of the proposed system from pre-extraction to kraft pulping showed a good mass balance of 82.7%. The advantage of our proposed approach is that the method decouples silica from organics-rich black liquor enabling silica

74

Z. Yuan et al. / Industrial Crops and Products 91 (2016) 66–75

Fig. 4. Proposed process of kraft pulping with silica and hemicelluloses extraction unit with NaOH.

recovery without substantial loss of lignin and hemicelluloses. A further study on the recovery of silica and hemicelluloses in the liquid fraction will be discussed in a forthcoming paper. 4. Conclusions Analysis of the distribution of chemical components in the bamboo stem showed that the whole stem should be used for commercial chips preparation. Alkaline pre-extraction of bamboo chips prior to pulping processes was demonstrated to be an effective way to extract silica and hemicelluloes selectively without degrading cellulose and lignin. Under the studied extraction conditions, almost 100% of silica and up to 50% of hemicelluloses in bamboo chips were extracted. Taking account into the consideration for efficiently dissolution of silica and hemicelluloses removal was more efficient at a pre-treatment temperature at 100 ◦ C compared to 80 ◦ C. Pre-extraction of the chips improved delignification during kraft pulping even at lower effective alkali charges. Pulp from pre-extracted bamboo chips showed similar screened yield to that of non-extracted chips while initial drainage resistance (CSF) improved slightly. The tensile strength index of the pulp actually benefited from the alkaline pre-treatment at low EA charges. The tear strength index of the kraft pulp was improved by the alkaline pre-extraction. Silica was not detected in the kraft pulp and black liquor obtained with pre-extracted bamboo chips indicating that the process provides a solution to silica problems encountered in bamboo pulping. To make the proposed process fit well into the integrated forest biorefinery concept, investigations into the isolation, purification and transformation of silica and hemicelluloses from the AEL into valuable products are being carried out in our laboratory. Acknowledgements This work was funded by Lee & Man Paper Manufacturing Limited. The authors are grateful to Elaine Woo, British Columbia Institute of Technology, and Professors Jack Saddler, Valdeir Arantes and Phillip Evans, University of British Columbia, Vancouver, for providing devices and technical support in this study. References Bai, L., Hu, H., Xu, J., 2012. Influences of configuration and molecular weight of hemicelluloses on their paper-strengthening effects. Carbohydr. Polym. 88, 1258–1263.

Chand, N., Jain, D., Nigrawal, A., 2006. Investigation on gradient dielectric characteristics of bamboo (Dentroclamus strictus). J. Appl. Polym. Sci. 102, 380–386. Chirat, C., Lachenal, D., Sanglard, M., 2012. Extraction of xylans from hardwood chips prior to kraft cooking. Process Biochem. 47, 381–385. Collin, B., Doelsch, E., Keller, C., Panfili, F., Meunier, J.D., 2012. Distribution and variability of silicon, copper and zinc in different bamboo species. Plant Soil 351, 377–387. Dence, C.W., 1992. The determination of lignin. In: Dence, C.W. (Ed.), Methods in Lignin Chemistry. Springer-Verlag, Berlin, pp. 33–61. Ding, T., Zhou, J., Wan, D., Chen, Z., Wang, C., Zhang, F., 2008. Silicon isotope fractionation in bamboo and its significance to the biogeochemical cycle of silicon. Geochim. Cosmochim. Acta 72, 1381–1395. Duarte, G.V., Ramarao, B.V., Amidon, T.E., Ferreira, P.T., 2011. Effect of hot water extraction on hardwood kraft pulp fibers (Acer saccharum, sugar maple). Ind. Eng. Chem. Res. 50, 9949–9959. Engström, A.C., Ek, M., Henriksson, G., 2006. Improved accessibility and reactivity of dissolving pulp for the viscose process: pretreatment with monocomponent endoglucanase. Biomacromolecules 7, 2027–2031. Farrell, E.P., Führer, E., Ryan, D., Andersson, F., Hüttl, R., Piussi, P., 2000. European forest ecosystems: building the future on the legacy of the past. For. Ecol. Manage. 132, 5–20. Hamzeh, Y., Ashori, A., Khorasani, Z., Abdulkhani, A., Abyaz, A., 2013. Pre-extraction of hemicelluloses from bagasse fibers: effects of dry-strength additives on paper properties. Ind. Crops Prod. 43, 365–371. Huang, F., Ragauskas, A., 2013. Extraction of hemicellulose from loblolly pine woodchips and subsequent kraft pulping. Ind. Eng. Chem. Res. 52, 1743–1749. Jahan, M.S., Lee, Z.Z., Jin, Y., 2006. Organic acid pulping of rice straw, part-I: cooking. Turk. J. Agric. For. 30, 231–239. Jun, A., Tschirner, U.W., Tauer, Z., 2012. Hemicellulose extraction from aspen chips prior to kraft pulping utilizing kraft white liquor. Biomass Bioenergy 37, 229–236. Kalapathy, U., Proctor, a., Shultz, J., 2002. An improved method for production of silica from rice hull ash. Bioresour. Technol. 85, 285–289. Kapu, N.S., Trajano, H.L., 2014. Review of hemicellulose hydrolysis in softwoods and bamboo. Biofuels. Bioprod. Biorefin. 8, 857–870. Kopfmann, K., Hudeczek, W., 1988. Desilication of black liquor: pilot plant tests. Tappi J. 71, 139–147. Leenakul, W., Tippayawong, N., 2010. Dilute acid pretreatment of bamboo for fermentable sugar production. J. Sustainable Energy Environ. 1, 117–120. Lehto, J., Alén, R., 2013. Alkaline pre-treatment of hardwood chips prior to delignification. J. Wood Chem. Technol. 33, 77–91. Li, Z., Jiang, Z., Fei, B., Pan, X., Cai, Z., Yu, Y., 2012. Ethanol organosolv pretreatment of bamboo for efficient enzymatic saccharification. BioResources 7, 3452–3462. Li, Z., Jiang, Z., Fei, B., Cai, Z., Pan, X., 2014. Comparison of bamboo green, timber and yellow in sulfite, sulfuric acid and sodium hydroxide pretreatments for enzymatic saccharification. Bioresour. Technol. 151, 91–99. Liese, W., 1987. Research on bamboo. Wood Sci. Technol. 21, 189–209. Liu, H., Hu, H., Jahan, M.S., Ni, Y., 2013. Furfural formation from the pre-hydrolysis liquor of a hardwood kraft-based dissolving pulp production process. Ind. Eng. Chem. Res. 52, 3974–3979. Luo, X., Ma, X., Hu, H., Li, C., Cao, S., Huang, L., 2013. Kinetic study of pentosan solubility during heating and reacting processes of steam treatment of green bamboo. Bioresour. Technol. 130, 769–776. Machmud, M.N., Fadi, F., Fuadi, Z., Kokarkin, C., 2013. Alternative fiber sources from Gracilaria Sp and Eucheuma Cottonii for papermaking. Int. J. Sci. Eng. 6, 1–10.

Z. Yuan et al. / Industrial Crops and Products 91 (2016) 66–75 Mao, H., Genco, J., Van Heiningen, A., Pendse, H., 2008. Technical economic evaluation of a hardwood biorefinery using the “near-neutral” hemicellulose pre-extraction process. J. Biobased Mater. Bioenergy 2, 177–185. Minu, K., Jiby, K.K., Kishore, V.V.N., 2012. Isolation and purification of lignin and silica from the black liquor generated during the production of bioethanol from rice straw. Biomass Bioenergy 39, 210–217. Molin, M., Teder, A., 2002. Importance of cellulose/hemicelluloses ratio for pulp strength. Nordic Pulp Paper Res. J. 17, 14–19. Pan, X.J., Sano, Y., Ito, T., 1999. Atmospheric acetic acid pulping of rice straw. II: behavior of ash and silica in rice straw during atmospheric acetic acid pulping and bleaching. Holzforschung 53, 49–55. Persson, T., Ren, J.L., Joelsson, E., Jönsson, A.S., 2009. Fractionation of wheat and barley straw to access high-molecular-mass hemicelluloses prior to ethanol production. Bioresour. Technol. 100, 3906–3913. Pinto, P.C., Evtuguin, D.V., Neto, C.P., 2005. Structure of hardwood glucuronoxylans: modifications and impact on pulp retention during wood kraft pulping. Carbohydr. Polym. 60, 489–497. Rocha, G.J.M., Martín, C., da Silva, V.F., Gómez, E.O., Gonc¸alves, A.R., 2012. Mass balance of pilot-scale pretreatment of sugarcane bagasse by steam explosion followed by alkaline delignification. Bioresour. Technol. 111, 447–452. Salmela, M., Alén, R., Vu, M.T.H., 2008. Description of kraft cooking and oxygen-alkali delignification of bamboo by pulp and dissolving material analysis. Ind. Crops Prod. 28, 47–55. Sathitsuksanoh, N., Zhu, Z.G., Hoc, T.J., Bai, M.D., Zhang, Y.H.P., 2010. Bamboo saccharification through cellulose solvent-based biomass pretreatment followed by enzymatic hydrolysis at ultra-low cellulase loadings. Bioresour. Technol. 101, 4926–4929. Scurlock, J.M.O., Dayton, D.C., Hames, B., 2000. Bamboo: an overlooked biomass resource? Biomass Bioenergy 19, 229–244. Shi, H., Fatehi, P., Xiao, H., Ni, Y., 2011. A combined acidification/PEO flocculation process to improve the lignin removal from the pre-hydrolysis liquor of kraft-based dissolving pulp production process. Bioresour. Technol. 102, 5177–5182.

75

Sixta, H., 2006. Handbook of Pulp. Wiley-VCH, Weinheim, pp. 325–365 and 1009–1067. Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., Crocker, D., 2012. Determination of Structural Carbohydrates and Lignin in Biomass. Version 2012. National Renewable Energy Laboratory, USA. Taherzadeh, M.J., Karimi, K., 2008. Pretreatment of lignocellulosic wastes to improve ethanol and biogas production: a review. Int. J. Mol. Sci. 9, 1621–1651. Tong, G.L., Lu, Q., Wang, J., 2005. Spectrophotometric determination of silicon in rice straw and black liquor after rice straw alkaline pulping by silicomolybdate blue. China Pulp Paper Ind. 26, 64–66. ˇ Torelli, N., Cufar, K., 1995. Mexican tropical hardwoods. Comparative study of ash and silica content. Holz als Roh-und werkstoff 53, 61–62. Tsuji, H., Isono, Z., Ono, K., 1965. Studies on dissolving bamboo pulp. Bull. Univ. Osaka Prefect. Ser. B: Agric. Biol. 16, 89–104. Van Heiningen, A.R.P., 2006. Converting a kraft pulp mill into an integrated forest biorefinery. Pulp Pap. Can. 107, 38–43. Vu Mân, T.H., Pakkanen, H., Alen, R., 2004. Delignification of bamboo (Bambusa procera acher): part 1. Kraft pulping and the subsequent oxygen delignification to pulp with a low kappa number. Ind. Crops Prod. 19, 49–57. Walton, S.L., Hutto, D., Genco, J.M., Van Walsum, G.P., Van Heiningen, A.R.P., 2010. Pre-extraction of hemicelluloses from hardwood chips using an alkaline wood pulping solution followed by kraft pulping of the extracted wood chips. Ind. Eng. Chem. Res. 49, 12638–12645. Xiao, X., Bian, J., Li, M.F., Xu, H., Xiao, B., Sun, R.C., 2014. Enhanced enzymatic hydrolysis of bamboo (Dendrocalamus giganteus Munro) culm by hydrothermal pretreatment. Bioresour. Technol. 159, 41–47. Yoon, S.H., Van Heiningen, A., 2010. Green liquor extraction of hemicelluloses from southern pine in an integrated forest biorefinery. J. Ind. Eng. Chem. 16, 74–80. Yoon, S.H., Tunc, M.S., Van Heiningen, A., 2011. Near-neutral pre-extraction of hemicelluloses and subsequent kraft pulping of southern mixed hardwoods. Tappi J. 10, 7–15. Zhang, X., Zhao, Z., Ran, G., Liu, Y., Liu, S., Zhou, B., Wang, Z., 2013. Synthesis of lignin-modified silica nanoparticles from black liquor of rice straw pulping. Powder Technol. 246, 664–668.