Optimization of cellulose nanocrystals from bamboo shoots using Response Surface Methodology.

Cellulose-based advanced materials, such as cellulose nanocrystals (CNC), have high potential application for drug delivery system. In this study, the CNC were produced from bamboo shoots using acid hydrolysis process. The delignification of bamboo shoots was conducted using alkali and hydrogen peroxide pretreatment processes. The operating condition of the production of CNC from bamboo shoots was optimized using Response Surface Methodology (RSM) based on the yield and crystals recovery as the responses. The optimum CNC yield of 50.67 ± 0.74% with a crystals recovery of 77.99 ± 1.14% was obtained at the sulfuric acid concentration of 54.73 wt% and a temperature of 39 °C from the optimization based on the yield. This optimization has been validated to confirm the accuracy.

Introduction

Bamboo is a rapidly growing plant that is widely cultivated in Asia, including Indonesia. The shoots of this plant are edible and used as one of the important ingredients of domestic foods. Bamboo shoots (BS) have a unique characteristic; their odor is unpleasant and pungent. BS are lignocellulosic material, which consists of lignin, hemicelluloses, and cellulose. The compositions based on the dry matter of BS were 5–11% of lignin, 24–27% of hemicelluloses, and 23–35% of cellulose (Li et al., 2015).

In recent years, the research interests of BS are mainly on the polysaccharides constituents (Chen et al., 2018a, Chen et al., 2018b; Chen et al., 2018a, Chen et al., 2018b; He et al., 2016; Zhang et al., 2018), the bioactive agents contents like antioxidant, phenolic compound, and angiotensin converting enzyme inhibition activity (Gong et al., 2016; Park and Jhon, 2010), and the pyrolysis products of BS (Ye et al., 2015). As compared to the other lignocellulosic materials, the cellulose content of BS is relatively high, while the low lignin content facilitates the easy treatment to extract the cellulose content. So far, the utilization of BS as the raw material for cellulose nanocrystals (CNC) production has never been studied.

Cellulose nanocrystals (CNC) are one of the advanced cellulose-based materials. CNC are an advanced nano-scale material with a needle or rod-like crystals structure. CNC have a diameter size of 5 nm up to 30 nm and length size of 100 nm up to several μm. CNC possess good physical characteristics as well as their chemical characteristics, such as reactive hydroxyl groups, high crystallinity, large surface area, good thermal stability, good mechanical properties, biocompatible, biodegradable, and non-toxic (Collazo-Bigliardi et al., 2018; Qing et al., 2016; Zainuddin et al., 2017). Due to their excellence properties, CNC can be used for advanced utilization such as drug delivery, tissue engineering, composite material, protein or enzyme immobilization, emulsifier, and others nanomaterial synthesis (Wijaya et al., 2017). Previous studies have reported the CNC production from various natural resources, such as rice straw (Lu and Hsieh, 2012), corncob (Silvério, et al., 2013), sugarcane bagasse (Teixeira et al., 2011), sugar palm fiber (Ilyas et al., 2018), sago seed shell (Naduparambath et al., 2018; Naduparambath and Purushothaman, 2016), passion fruit peels (Wijaya et al., 2017), and groundnut shell (Bano and Negi, 2017). The cellulose content of BS also can be converted to micro- and nano-scale material, such as CNC, and currently there is no study utilizes the BS for the CNC production and its subsequent use as a drug carrier in drug delivery system.

CNC can be produced using various methods, such as steam explosion treatment (Abraham et al., 2011; Cherian et al., 2010; Deepa et al., 2011), ionic liquids (Man et al., 2011; Tan et al., 2015), ultrasound technique (Li et al., 2012b), high pressure homogenization (Kawee et al., 2018; Lee et al., 2018; Li et al., 2012a; Wang et al., 2015), acid hydrolysis (Coelho et al., 2018; Collazo-Bigliardi et al., 2018; Gu et al., 2017; Robles et al., 2018), enzymatic hydrolysis (Chen et al., 2018c; Henriksson et al., 2007), and combined treatments. Acid hydrolysis is the most popular and common method for CNC production. The advantages of this method are its simplicity and the convenient removal process of the amorphous cellulose region using acid solution (Phanthong et al., 2018). Sulfuric acid is the most common acid solution used to produce CNC with a negative charge, which stable in suspension (Gu et al., 2017).

In this study, the preparation of CNC from BS was conducted using acid hydrolysis method, and sulfuric acid was used as the hydrolyzing agent. The optimization of the acid hydrolysis process was also studied using Response Surface Methodology (RSM) to determine the optimum conditions of the recovery of CNC from BS.

Experimental

Materials

Bamboo shoots (Dendrocalamus asper) were obtained from the local market in Surabaya, East Java, Indonesia. BS were harvested after having a height of 30–50 cm from the ground. BS were repeatedly washed and thinly sliced before they were dried under sunlight for three days. Dried BS were ground and sieved through a 100-mesh sieve. All chemicals, i.e., sodium hydroxide, hydrogen peroxide (50%), and sulfuric acid (98%), were purchased from Sigma-Aldrich (Singapore) as an analytical grade.

Pretreatment of BS

Cellulose from BS was obtained by the following steps: delignification and bleaching steps. For the delignification process, BS powder (10 g) were soaked using 200 mL of 20% (w/v) sodium hydroxide solution. The mixture was heated at 90 °C for 4 h under continuous stirring. The solid residue was washed with distilled water until the pH was neutral. Then, it was dried using oven at 50 °C for overnight. For the bleaching process, the dried residue was soaked using 4% (w/v) sodium hydroxide solution at 50 °C under continuous stirring. Then, 50% hydrogen peroxide with the same volume of sodium hydroxide was added dropwise. The bleaching process was done for 60 min under continuous stirring. The pretreated bamboo shoots (PBS) were washed with distilled water until neutral pH and then dried at 50 °C for overnight.

Isolation of CNC

CNC were isolated from the cellulose of BS by acid hydrolysis process. The acid hydrolysis process was carried out using different concentrations of sulfuric acid (45, 55, 65, and 75 wt%) at various temperatures (30, 45, and 60 °C) for 60 min. The PBS mass to sulfuric acid volume ratio was 1:20. After the hydrolysis process completed, excess cold distilled water was added to the mixture to stop the hydrolysis process. The suspension was centrifuged at 6000 rpm for 10 min to remove the acid solution. The precipitate was washed using distilled water until neutral pH was reached. The neutral suspension was sonicated in an ice bath for 20 min and dried using freeze dryer at a pressure of 0.08 mbar and temperature of –42 °C.

Characterizations

Determination of chemical composition

The chemical compositions of BS and PBS, such as the hot water soluble, lignin, hemicelluloses, and cellulose content, were determined at different stages of treatment (Datta, 1981). 1 g of sample was refluxed with 150 mL of distilled water at 100 °C for 2 h to remove hot water soluble content. The dried solid residue was then refluxed with 150 mL of 0.5 M sulfuric acid solution at 100 °C for 2 h to remove hemicelluloses content. After that, the dried solid residue was treated with 10 mL of 72% (v/v) sulfuric acid solution at room temperature for 4 h. Then, it was diluted to 0.5 M sulfuric acid solution and refluxed at 100 °C for 2 h to remove cellulose content. The ash content was determined using furnace at a temperature of 600 °C for 6 h, while the weight loss during furnacing was the lignin content. The moisture content was determined using moisture balance.

Thermal gravimetric analysis (TGA)

The material stability under thermal condition was analyzed using a Perkin Elmer Diamond TG/DTA thermal analyzer. The TGA analysis was carried out in a nitrogen gas flow of 150 mL/min with a constant heating and cooling rate of 10 °C/min at a temperature range of 25–600°.

X-ray diffraction (XRD) analysis

The crystallinity of CNC was analyzed using PANalytical X'Pert PRO X-ray diffractometer with a monochromated high-intensity Cu Kα radiation (1.5406 Å) operating at 40 kV and 30 mA. The XRD pattern was recorded using a step size of 0.02°/C at the angle range of 5–60°. The crystallinity index percentage was expressed by this equation below (Ditzel et al., 2017):where is the minimum value of the 110 and 200 lattices diffractions which exhibits the amorphous region, while is the maximum value of the 200 lattice which exhibits the total of amorphous and crystalline regions (Wijaya et al., 2017).

Scanning electron microscopy (SEM) analysis

The SEM analysis was used to observe the microstructure of CNC. A fine auto coater (JFC-1200, JEOL, Ltd., Japan) was used to coat the material using a thin conductive platinum layer under argon atmosphere. The SEM analysis was carried out using a JEOL JSM-6390 field emission SEM that was operated at an accelerating voltage of 5 kV.

Fourier transform infrared spectroscopy (FTIR) analysis

The FTIR analysis was used to determine the functional groups in the BS, PBS, and CNC. The change of the functional groups based on the treatments was used to know the effects of the treatments on the material's structure. The FTIR analysis was carried out using FTIR Shimadzu 8400S that used KBr in pelleting the materials. The spectra were recorded at the wavenumber range of 4000–400 cm−1.

Optimization

RSM was used to analyze the optimum condition of the acid hydrolysis process in the CNC production, and Minitab 16 statistical software (Minitab Inc., ITS Surabaya, Indonesia) was employed for this purpose. RSM was also used to describe the significance of the independent variables toward the response, where the significance was determined when the p-value < 0.05. The central composite design with two independent variables (sulfuric acid concentration and temperature) was used for the optimization. Based on the full factorial design with two repetitions, 24 independent experiments were carried out. The influence of independent variables toward the response is given by the following equation (Muthukumaran et al., 2018):where is the constant, while , , and are the linear, square, and interaction effects coefficient of the two independent variables. Here, and indicate the independent variables used in this research.

As the responses in this statistical analysis are the yield of CNC and the crystals recovery of CNC from PBS. The yield and crystals recovery of CNC was calculated by these following equations:where and are the mass and the crystals mass of CNC, while and are the mass and the crystals mass of PBS, respectively. and are the crystallinity index of CNC and PBS, respectively.

Results and discussions

Cellulose from BS

The chemical compositions of BS and their pretreated form are presented in Table 1. The compositions of BS are consistent with the previous studies; they have high cellulose and low lignin content. With high cellulose content indicates that BS are very potential the raw material for CNC production. The change of chemical composition between BS and their pretreated form is obvious as seen in Table 1. After the pretreatment, the cellulose content significantly increases from 33.14 ± 0.34% to 74.34 ± 0.40%, while a significant decrease of the hemicelluloses and lignin contents is observed. The reduction of lignin content from 9.72 ± 0.45% to 0.63 ± 0.41% makes the subsequent process (acid hydrolysis) become easier.

Table 1

The chemical composition of BS and its pretreated form.

Component (in wt%)	This study		BS
	BS	PBS	Refa	Refb	Refc
Moisture	12.85 ± 0.10	6.95 ± 0.17	-	-	-
Hot water soluble	12.71 ± 0.56	3.29 ± 0.37	-	-	-
Hemicelluloses	22.79 ± 1.09	4.82 ± 0.18	21.3	24.8	13.0
Cellulose	33.14 ± 0.34	74.34 ± 0.40	22.8	27.5	31.1
Lignin	9.72 ± 0.45	0.63 ± 0.41	1.8	5.2	5.3
Ash	8.79 ± 0.42	9.98 ± 0.18	-	-	-

(Yang et al., 2014).

(Li et al., 2015).

(Chang et al., 2013).

Fig. 1 shows the thermal decomposition behavior of BS and PBS, a substantial weight loss of both materials during the heating process is observed. In the temperature range of 35–150 °C, the evaporation of moisture content occurred as detected by the weight loss, while the cellulose content was degraded at the temperature range of 220–365 °C. In the TGA curve of PBS, the substantial weight loss is observed in a temperature region that indicates the degradation of a significant amount of cellulose in PBS. The maximum degradation of cellulose was achieved at temperature of 325.3 °C. This phenomenon supports the result in Table 1. Due to the hemicelluloses and lignin reduction by the pretreatment process, the curve region of hemicelluloses and lignin are not seen in the PBS curve that means that the hemicelluloses and lignin contents are low enough. This phenomenon indicates that the pretreatment process was successful.

Fig. 1

TGA (A) and DTG (B) curves of bamboo shoots and its pretreated form.

Isolation of CNC from PBS

XRD analysis was conducted for every product of the acid hydrolysis process. The XRD patterns of CNC were displayed in Fig. 2, and matched up with the standard of the cellulose I crystals (JCPDS No. 00-050-22411). The XRD pattern of CNC has fours existed peaks at 2θ of 15.5°, 16.4°, 22.8°, and 35° assigned to the 10, 110, 200, and 004 planes, respectively. The crystallinity index was calculated using Eq. (1), where the result of PBS was 53.15%, and the highest result of CNC was 84.55%. The results indicate that the acid hydrolysis process successfully removes the amorphous region of PBS. The BS-CNC crystallinity index is high enough compared to the CNC isolated from other natural materials, such as from rice husk (CrI = 90%), coffee husk (CrI = 92%) (Collazo-Bigliardi et al., 2018), blue agave waste (CrI = 85–88%) (Tang et al., 2017), sugar palm fibres (CrI = 85.90%) (Ilyas et al., 2018), eucalyptus cellulose (CrI = 75%) (Gu et al., 2017), sago seed shells (CrI = 69%) (Naduparambath et al., 2018), and elephant grass (CrI = 72–77%) (Nascimento and Rezende, 2018). The crystallinity index of the CNC depends on the amount of cellulose crystals region in the raw material. Several natural resources, like kinds of husk, have higher amount of cellulose crystals region compared to any kind of shoots or grass. Therefore, the crystallinity index of the CNC from BS is lower than the CNC from rice husk and coffee husk.

Fig. 2

The XRD patterns of pretreated bamboo shoots (A) and CNCs which produced by acid hydrolysis process using [H2SO4] of 45 wt% at temperatures of 30 °C (B), 45 °C (C), and 60 °C (D); using [H2SO4] of 55 wt% at temperatures of 30 °C (E), 45 °C (F), and 60 °C (G); and using [H2SO4] of 65 wt% at temperatures of 30 °C (H), 45 °C (I), and 60 °C (J).

The crystallinity index was used for the calculation of the yield of CNC (Eq. 3), and the crystals recovery of CNC (Eq. 4). The yields and crystals recoveries of CNCs are summarized in Table 2. At the sulfuric acid concentration of 65 wt% and temperature of 45 °C, the highest yield and crystals recovery of CNC were 51.86 ± 2.04% and 82.50 ± 3.24%, respectively. The high crystals recovery of CNC indicates that the amorphous region was hydrolyzed into sugars while the crystalline region remains intact. At higher sulfuric acid concentration (75 wt%), both of amorphous and crystalline region of cellulose were hydrolyzed into sugars.

Table 2

The yields and crystals recoveries of CNCs in every combined variables.

Yields of CNC (%)
Temp. (°C)	[H2SO4] (wt%)
	45	55	65	75
30	42.97 ± 1.00	41.14 ± 0.58	48.45 ± 2.67	n.d.
45	41.78 ± 0.60	40.79 ± 0.19	51.86 ± 2.04	n.d.
60	40.30 ± 0.45	39.88 ± 0.56	40.22 ± 1.51	n.d.
Crystals recoveries of CNC (%)
Temp. (°C)	[H2SO4] (wt%)
	45	55	65	75
30	51.37 ± 1.19	57.89 ± 0.82	71.31 ± 3.93	n.d.
45	57.48 ± 0.82	61.69 ± 0.29	82.50 ± 3.24	n.d.
60	49.87 ± 0.56	57.54 ± 0.81	60.30 ± 2.27	n.d.

n.d.: not detected.

The morphology of CNC from BS is depicted in Fig. 3. The interconnected rod-like structure of the CNC has been proven by this SEM figure. The interconnected crystals structure makes CNC has high stability in suspension. The sonication treatment also assists the cavitation of CNC so that it is less precipitated in suspension by the formation of interconnected nano-scale cellulose structure.

Fig. 3

Scanning electron microscopy of CNC.

The functional groups of BS, PBS, and CNC have been analyzed using FTIR method, and the spectra are shown in Fig. 4. The appearance of O–H, C–H, and C–O stretching vibration peaks at 3337 cm−1, 2890 cm−1, and 1045 cm−1, respectively, represents the cellulose content in all of the samples, (Gu et al., 2017; Hernandez et al., 2018). The peak of O–H stretching vibration in the CNC has stronger transmittance compared to PBS and BS samples. The intensity of this hydrogen bonding increase is due to the removal of some parts of the amorphous region and the breakdown of some β-glycosidic bonds in the cellulose structure. The removal of some parts of the amorphous region is supported by the peak reduction of CH2 bonding at the sixth C in the cyclic chain of glucose (at a wavenumber of 1325 cm−1). The peak at 897 cm−1 indicates the β-glycosidic linkage between the monomers of cellulose (Naduparambath et al., 2018). The absorbed water of all samples is observed at the wavenumber of 1647 cm−1 (Xing et al., 2018).

Fig. 4

FTIR spectra of bamboo shoots, pretreated bamboo shoots, and CNC.

The success of the pretreatment process is indicated by the reduction of lignin and hemicelluloses content in the BS and PBS samples. The aromatic skeletal vibration of lignin is shown at 1516 cm−1, while the C=O and C–O stretching of hemicelluloses are at 1729 cm−1 and 1264 cm−1, respectively (Xing et al., 2018). In the FTIR spectrum of PBS, the peak that correspond to lignin could not be observed, a strong indication that the lignin content is low. While, the peaks transmittance correspond to hemicelluloses are significantly reduced. After the acid hydrolysis process, the hemicelluloses peaks in the FTIR spectra of CNC disappear completely due to the breakdown of hemicelluloses into pentose sugars. The C–H bonds of the crystalline cellulose is indicated by the vibrations at 1379-1311 cm−1 (Hernandez et al., 2018). In the FTIR spectra of CNC, the S=O and C–O–S vibrations at 1250 cm−1 and 833 cm−1, respectively, could not be detected which indicates the CNC contains low sulfur content (Gu et al., 2017; Xing et al., 2018).

Response surface methodology

Model fitting

The regression coefficients along with the Fisher value (F-value) and probability (p-value) of the model fitting are given in Table 3. The F-value and p-value are used to specify the fitness and significance of the model fitting. From the F-value, the statistical significance is considered if the F-value is higher than the critical F-value at a level of 5%. The significance of each fitting parameter toward the response value is also determined by the p-value < 0.05. The significance parameter is the sulfuric acid concentration as the independent variable for acid hydrolysis process. Therefore, the mathematical models for the yield and crystals recovery of CNC are given by the following equations:

Table 3

Regression coefficients, F-value, and p-values (significance levels) for the yield and crystals recovery of CNC as the responses.

Parameter	The yield of CNC (%)			Crystals recovery of CNC (%)
	Coefficient	F-value	p-value	Coefficient	F-value	p-value
α_0	-305.108	22.22	0.001	-597.725	20.65	0.000
α_1	12.526	36.17	0.000	21.879	48.44	0.000
α_2	0.674	0.18	0.678	2.779	1.32	0.266
α_11	-0.114	46.95	0.000	-0.194	59.11	0.000
α_22	-0.009	0.32	0.581	-0.031	1.66	0.214
α_12	0.000	0.00	0.978	-0.002	0.01	0.912

= constant; = [H2SO4] (linear parameter); = temperature (linear parameter); = [H2SO4]2 (quadratic parameter); = temperature2 (quadratic parameter); and = [H2SO4] × temperature (interaction parameter).

Effects of sulfuric acid concentration and temperature on the yield and crystals recovery of CNC

The contour plots of the yield and crystals recovery of CNC are shown in Fig. 5. Fig. 5A describes the effect of sulfuric acid concentration towards the yield of CNC. These contour plots are also used to attest the model fitting results about the significance of both variables. In the range of 45–75 wt%, the yield of CNC is found to be very varied, The yield of CNC >50% was obtained around 55 wt% of the sulfuric acid concentration. At a sulfuric acid concentration higher or lower than 55%, the decrease of the yield of CNC is observed. Temperature has no significant influence on the yield of CNC.

Fig. 5

The contour plots for the yield (A) and crystals recovery (B) of CNC.

For the crystals recovery of CNC, a similar phenomenon was observed. The sulfuric acid concentration has a more significant effect than the temperature as shown in Fig. 5B. The crystals recovery of CNC higher than 70% was obtained at a sulfuric acid concentration around 55–60 wt%. The temperature has no significant effect on the crystals recovery of CNC.

Optimization of the acid hydrolysis process

The “Response Optimizer” tool of the Minitab 16 statistical software was used to optimize the acid hydrolysis process for the production of CNC from BS. Both of the independent variables were optimized to satisfy each response which evaluated by the composite desirability value. The value equals to 1 indicates to the very satisfying solution, while 0 indicates that the response is out of the acceptable limits. For both optimizations, the deciding goal was the maximum solution with the lower response value of 0 and the target or upper response value of 100.

As shown in Fig. 6A, the optimum condition is a process using sulfuric acid concentration of 54.73 wt% at a temperature of 39 °C, where the maximum yield of CNC reaches 51.30%. The composite desirability of 0.51 indicates that only a part of the optimized solution satisfies the goal of this optimization. While for the crystals recovery of CNC as shown in Fig. 6B, the sulfuric acid concentration of 56.21 wt% and the temperature of 43 °C is the optimum condition, with the crystals recovery of CNC was 77.48%. This optimized solution has high satisfaction due to its composite desirability of 0.77.

Fig. 6

Optimization results for the yield (A) and crystals recovery (B) of CNC.

Acid hydrolysis process was conducted under the optimum condition to validate the optimization results as shown in Table 4. Based on their optimum condition, the yield of CNC obtained was 50.67 ± 0.74% with an error of 1.21%, while the crystals recovery of CNC was 76.25 ± 1.93% with an error of 1.60%. It indicates the high accuracy of the optimization results. From the optimization based on yield, CNC have CrI of 81.80%, while it is 83.65% from the optimization based on crystals recovery. This CrI indicates the quantitative indicator of crystallinity, which means the high CrI represents the high crystallinity of this CNC. The enhancement of CrI affects the hardness, density, transparency, and diffusion of the material.

Table 4

Validation of the optimization results.

Optimization	Validation results
	CrI (%)	Yield of CNC (%)	Crystals recovery of CNC (%)
Based on yielda	81.80	50.67 ± 0.74%	77.99 ± 1.14%
Based on crystals recoveryb	83.65	48.44 ± 1.23%	76.25 ± 1.93%

using sulfuric acid concentration of 54.73 wt% at a temperature of 39 °C.

using sulfuric acid concentration of 56.21 wt% at a temperature of 43 °C.

Conclusion

This study utilized bamboo shoots (BS), a potential lignocellulosic material with high enough cellulose content and low lignin content, as the raw material for the production of CNC. The CNC have the structure of the rod-like crystals. In this CNC production, the crystals recovery of 77.99 ± 1.14% and the yield of 50.67 ± 0.74% were obtained using a sulfuric acid concentration of 54.73% and a temperature of 39 °C. As the results, CNC from BS have a potential application for composite material, drug delivery, emulsifier, enzyme immobilization, nanomaterial synthesis, and tissue engineering.

Declarations

Author contribution statement

Christian J. Wijaya: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Wrote the paper.

Suryadi Ismadji, Setiyo Gunawan, Hakun W. Aparamarta: Conceived and designed the experiments; Contributed reagents, material, and analysis tools or data; Analyzed and interpreted the data.

Funding statement

This work was supported by The Directorate General of Resources for Science, Technology, and Higher Education, Ministry of Research, Technology, and Higher Education of Republic Indonesia (No. 851/PKS/ITS/2019) and Widya Mandala Surabaya Catholic University.

Competing interest statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.