Chemical Engineering Journal 429 (2022) 132452

Contents lists available at ScienceDirect

Chemical Engineering Journal

journal homepage: www.elsevier.com/locate/cej

Enzymatic hydrolysis enhancement of oil palm empty fruit bunch by

Peracetic-Sulfuric acid pretreatment
Libertus Darus a, *, Susana Susana a, Halasan Sihombing b, Amaliyah Rohsari Indah Utami c,
Maizirwan Mel d
a
Study Program of Crops Processing Technology, Pontianak State Polytechnic, Jl. A. Yani I, Pontianak, Kalimantan Barat 78124, Indonesia
b
Study Program of Electrical Engineering, Pontianak State Polytechnic, Jl. A. Yani I, Pontianak, Kalimantan Barat 78124, Indonesia
c
Engineering Physics, School of Electrical Engineering, Telkom University, Bandung, Jawa Barat 40257, Indonesia
d
Department of Biotechnology Engineering, Faculty of Engineering, International Islamic University Malaysia (IIUM), Gombak, Kuala Lumpur 50728, Malaysia

A R T I C L E I N F O A B S T R A C T

Keywords: The use of Oil palm empty fruit bunch (OPEFB) for bioethanol production requires lignin and hemicellulose
Oil palm empty fruit bunch removals to provide more enzyme accessibility during the subsequent enzymatic hydrolysis of cellulose to
Peracetic acid glucose. The well-known Dilute sulfuric acid (DS)-pretreatment degraded mainly the hemicellulose with minimal
Sulfuric acid
delignification. Oppositely, Peracetic acid (PA) degraded selectively the lignin, leaving most of the hemicellulose.
Closed reactor
Electromagnetic irradiation
In this study, the addition of PA 200 mM to H2SO4 100 mM (called PS) under electromagnetic irradiation (140 ◦ C
Microstructure for 5 min) led to the high effectiveness of not only hemicellulose removal (88.5%) but also delignification
(81.3%). This effective pretreatment enriched the OPEFB quality (cellulose composition) to 83.5% from 49.9% in
unpretreated, compared to 62.9% in DS-pretreated OPEFB. The PS-pretreatment enhances the enzymatic di­
gestibility to 77.0% (at 35 EGU cellulase and 30 U β-glucosidase/g cellulose), 1.8- and 11.9-times higher than
those of DS-pretreated and unpretreated OPEFB, respectively. This high effectiveness of PS-pretreatment pro­
vides a potential way towards economically feasible bioethanol production from OPEFB.

1. Introduction pretreatments for lignocellulosic biomass. During Organosolv-

pretreatment, both lignin and hemicellulose are successfully removed
The scarcity of fossil energy and global warming motivate the utili­ from OPEFB [8]. However, this method involved a high temperature
zation of renewable energy sources such as wind, water, solar, (210 ◦ C) and another process for solvent recovery. DS-pretreatment
geothermal, and biomass. Oil palm empty fruit bunch (OPEFB), ligno­ effectively removes hemicellulose with a low cellulose degradation but
cellulosic biomass produced by palm oil mill is abundantly available performs a low delignification [9–10]. The residual lignin in DS-
with low price but not optimally utilized. OPEFB containing cellulose pretreated solid undesirably affected the following step by adsorbing
can be used as a raw material for second-generation bioethanol through and blocking enzymes during cellulose hydrolysis [11–12]. An
pretreatment, enzymatic hydrolysis, and fermentation [1,2]. However, increasing delignification was achieved during DS-pretreatment of
the complex lignin and hemicellulose structures are recalcitrant to cel­ OPEFB using a microwave reactor (110 ◦ C, 80 min) [13]. In contrast,
lulose enzymatic hydrolysis to glucose [3–5]. The complexity becomes Peracetic acid (PA)-pretreatment selectively degraded lignin from pop­
the primary economic barrier for lignocellulosic bioethanol production lar biomass with low hemicellulose removal [14]. Double-step pre­
[6]. Pretreatment is required to remove the lignin and hemicellulose treatments of OPEFB with PA (35 ◦ C, 3–9 h) followed with alkaline
with minimal cellulose degradation to provide cellulose-rich biomass. peroxide (35–45 ◦ C, 10–12 h) removed 68.7–91.5% lignin significantly
This high-quality biomass can be easily accessed and digested by [15–16]. Nevertheless, the double-step process is less practical, and the
cellulase at a minimal enzyme loading [7]. Mild conditions, including residual hemicellulose after PA-pretreatment also prevented the en­
time, chemical dosage and temperature, and simple process are neces­ zymes from accessing cellulose in the subsequent process [3–5].
sary to create an economically feasible pretreatment. Recently, Kundu et al. [17] reported that a thermally-assisted single-step
Organosolv and Dilute sulfuric acid (DS) are known as effective pretreatment (90 ◦ C, 5 h) of Hard Wood Eucalyptus regnans with PA

* Corresponding Author.
E-mail address: [email protected] (L. Darus).

https://doi.org/10.1016/j.cej.2021.132452
Received 13 July 2021; Received in revised form 4 September 2021; Accepted 10 September 2021
Available online 16 September 2021
1385-8947/© 2021 Elsevier B.V. All rights reserved.
L. Darus et al. Chemical Engineering Journal 429 (2022) 132452

Fig. 1. Component loss of pretreated OPEFB from the initial content in unpretreated OPEFB at different PA concentrations (A), H2SO4 concentrations (B), and
pretreatment times (C). An error bar showed a standard deviation from a triplicate experiment.

addition to DS (called PS) led to 91.1% delignification, 74.9% hemi­ 2. Materials and methods
cellulose removal, and retained 73.8% cellulose in pretreated solid [17].
More interestingly, a microwave-assisted PS-pretreatment of Yellow 2.1. Materials
Poplar biomass (120 ◦ C) took only 5 min to perform both delignification
(90.4%) and hemicellulose removal (85.7%) with high-quality cellulose OPEFB obtained from PT. Pundi Lahan Khatulistiwa (Kubu Raya-
(76.0%) in the pretreated solid [18]. West Kalimantan-Indonesia) was cut (±5 cm), dried (cabinet dryer,
According to Rezania et al. [6], a suitable pretreatment selection 50 ◦ C, moisture content < 9%), hammer-milled and sieved (vibration
should be based on biomass structure. Moreover, Geng et al. [14] shaker: 40–60 mesh). The powder was re-dried to a constant weight and
observed that PA-delignifications of different biomasses needed specific stored in a desiccator before analysis and pretreatment. Chemicals
pretreatment conditions. To our knowledge, the PS-pretreatment eval­ including H2O2 (CAS Number 7722–84-1, 34.5–36.5%), acetic acid (CAS
uation for OPEFB has not yet been reported. In this paper, OPEFB pre­ Number 64–19-7, ≥99.8%), and H2SO4 (CAS Number 7664–93-9,
treatments were optimized for PA and H2SO4 concentrations, 95–97%) were purchased from Sigma-Aldrich. PA was prepared by
pretreatment times, and temperatures in a closed reactor system under mixing H2O2 and acetic acid, according to Lee et al. [18]. Mixing the
electromagnetic irradiation. The optimization was based on pretreat­ prepared PA and H2SO4 provided a working-acid solution. Cellulase
ment effectiveness (delignification, hemicellulose removal, cellulose (Novozymes, Viscozyme® Cassava CL) and β–glucosidase enzymes
enrichment) and enzymatic hydrolysis digestibility. Finally, the (Creative Enzymes, DIA-195) had activities of 700 EGU/g and 60 U/mg,
component composition, hydrolysis digestibility, and microstructural respectively.
characteristics were compared for PS and DS-pretreated OPEFB under
similar conditions.
2.2. Chemical pretreatment optimizations

The OPEFB powder (5 g) was pretreated with the 100-mL working-

acid solution in a 150-mL pressurized-wall glass reactor (WBFY-201,
ZZKD, China). The glass reactor was equipped with a Teflon cup and

2
L. Darus et al. Chemical Engineering Journal 429 (2022) 132452

reached the desired values. Further, to determine final optimized con­

ditions of chemical pretreatment, a higher temperature (140 ◦ C) was
applied at a fixed (previously optimized) pretreatment time for the
possibility of lowering the PA and H2SO4 concentrations. Finally, the
major component compositions (hemicellulose, lignin, and cellulose),
OPEFB qualities (percentage of cellulose), cellulose digestibilities, and
microstructural characteristics were compared between the DS and PS-
pretreated OPEFB under similar conditions.

2.3. Enzymatic hydrolysis

Enzymatic hydrolyses were conducted for OPEFB resulted from PS-

pretreatment (at final optimized conditions), DS-pretreatment (under
similar conditions to PS-pretreatment), and unpretreated one according
to the National Renewable Energy Laboratory (NREL) Standard [19].
The amount of OPEFB solids (equivalent to 0.4 g cellulose) was prepared
in a 25-mL Erlenmeyer to make a slurry with the addition of 0.4 mL of
sodium azide 20 mg/mL, cellulase loadings (5, 20, and 35 EGU/g cel­
lulose), β-glucosidase loading (30 U/g cellulose) and citrate buffer (50
Fig. 2. Component loss altered by a higher temperature and lower concentra­
mM, pH 4.8) to make a total liquid of 20 mL. A fixed β-glucosidase
tion of acids than the previously optimized conditions (PA 300 mM, H2SO4 150 loading was used since the dosage variation has little effect on hydrolysis
mM, 120 ◦ C, 5 min). Error bars showed standard deviation from triplicate digestibility [20–21]. The slurries were hydrolyzed using a Shaking-
experiments. water bath incubator (WNB-22, Memmert, Germany) at 150 rpm,
50 ◦ C. Before the enzyme addition, the slurries were first situated for one
h in an incubator (50 ◦ C). Two hundred μL of aliquots were collected
after 1, 2, 3, 4, and 5-day enzymatic hydrolyses, centrifuged at 10,000
rpm for 5 min to determine the glucose content in the supernatant and
further to calculate the cellulose digestibility (%). All experiments
(pretreatment and enzymatic hydrolysis) were conducted in triplicate.

2.4. Chemical analyses and calculations

The main component contents (lignin, hemicellulose, and cellulose)

in the pretreated and unpretreated OPEFB were determined using NREL
standard method [22]. After two-step hydrolysis of the NREL protocol,
the residue was analyzed gravimetrically for acid-insoluble lignin (after
correction for ash and protein). The filtrate was analyzed for sugar
(xylose and glucose) to determine further hemicellulose and cellulose
contents, respectively by HPLC refractive-index detector (UltiMate
3000, Thermo Scientific, USA) with column: Aminex HPX-87H (300 ×
7.8 mm, 5 μm; Bio-Rad) and solvent: sulfuric acid 5 mM (0.6 mL.min− 1,
60 ◦ C). Acid-soluble lignin in the filtrate was analyzed using a UV–Vis
Spectrophotometer (UV-1800, Shimadzu, Japan) at 240 nm wavelength
and 1 nm spectral resolution. The acid-soluble lignin together with the
Fig. 3. Component compositions of PS-pretreated (at different PA concentra­
tions) and DS-pretreated OPEFB at similar pretreatment conditions: H2SO4 100 acid-insoluble lignin was then counted as the total lignin.
mM, 5 min, 140 ◦ C. Error bars showed standard deviation from triplicate The major component losses were calculated as the percentage of
experiments. component reduction after pretreatment from the initial component
content in the unpretreated OPEFB. The major component compositions
temperature sensor, stirred with a magnetic bar (400 rpm). A modified in the pretreated and unpretreated OPEFB were calculated as the
home-type microwave (EMM 2007X-800 W, Electrolux, Indonesia) with component amount (g) per 100 g unpretreated OPEFB. The quality of
improved temperature control irradiated the reactor electromagneti­ OPEFB referred to the cellulose percentage from the total amount of
cally. A thermocouple was inserted into the solution (inside the reactor), hemicellulose, lignin, and cellulose. Cellulose digestibility was calcu­
connected to the temperature control, and maintained the desired lated as the percentage of the 0.9 × generated glucose content after
temperature. Following the irradiation, the reactor reached room tem­ enzymatic hydrolysis from the cellulose content before enzymatic hy­
perature through water immersion (approximately within 3 min). The drolysis. The 0.9 is an anhydro correction for cellulose calculation from
OPEFB slurry was filtered through a Whatman-1 paper with a vacuum glucose content [22]. The analyses were conducted in duplicate for each
pump. The solid residues were rinsed with distilled water until the sample.
filtrate reached pH-7, dried to a constant weight (vacuum oven, 40 ◦ C),
cooled, and stored in a desiccator for analysis. 2.5. Microstructural characterizations
Optimization of chemical pretreatment conditions was based on the
effectiveness of delignification and hemicellulose removal and the low The chemically pretreated OPEFB with PS and DS (at final optimized
degree of cellulose loss. The concentrations of PA (0, 100, 200, 300, and conditions) and unpretreated OPEFB were characterized for their
400 mM), H2SO4 (0, 50, 100, and 150 mM) and pretreatment times (5, microstructural characteristics: surface morphology, surface area and
30, 60, and 90 min) were optimized stepwise at fixed temperature pore size, crystallinity, and functional group. The surface morphologies
(120 ◦ C). The pretreatment times were counted after the temperatures were observed by a scanning electron microscope/SEM (FlexSem 1000,
Hitachi, Japan). The surface area and pore size were analyzed using

3
L. Darus et al. Chemical Engineering Journal 429 (2022) 132452

Fig. 4. The effect of pretreatments (PS and DS) on enzymatic hydrolysis (cellulose digestibility) at different incubation times and cellulase loadings (at fixed
β-glucosidase loading of 30 U/g cellulose). The OPEFB solids were previously pretreated at similar conditions: H2SO4 100 mM, 5 min, 140 ◦ C. Error bars showed
standard deviation from a triplicate experiment.

Brunauer Emmet Teller/BET, conducted by Gas Sorption Surface Area was removed at PA 0 mM when the only H2SO4 100 mM was available. It
and Pore Size Analyzer (QUADRASORB evoTM, Quantachrome, USA) N2 was found that PA introduction (100–400 mM) to H2SO4 100 mM did
at 77 K on Carbon. The crystallinity was determined using Powder X-ray not affect the hemicellulose removal degree. This occurrence was also
Diffractometer/XRD (XPert MPD, Philips, USA) within 5–60◦ , at reported by Lee et al. [18], most likely due to the same readily available
0.04◦ s− 1. The crystallinity index/CrI was calculated according to Segal H2SO4 concentration (100 mM) during pretreatment. Higher PA con­
et al. [23], indicating the relative composition of crystalline cellulose in centrations detrimentally led to higher cellulose losses (3.0 to 9.9, 11.3,
the OPEFB solid. Finally, the functional group was characterized using 14.7, and 27.9%, respectively) due to chemical hydrolysis by PA. For its
Fourier Transform Infrared Spectroscopy/FTIR (Nicolet 6700, Thermo relatively high delignification and hemicellulose removal and the low
Fisher Scientific, USA) at 4000 to 500 cm− 1 and resolution 8 cm− 1. cellulose loss, the PA 300 mM was chosen to optimize the other
variables.
3. Results and discussions On the other hand, higher H2SO4 concentrations (0 to 50, 100 and
150 mM) at fixed conditions (PA 300 mM, 5 mins, 120 ◦ C) led to higher
3.1. PS-pretreatment optimization hemicellulose removal (20.1 to 70.1, 78.6 and 86.2%, respectively)
(Fig. 1B). This revealed the H2SO4 catalytic activity to hydrolyze
The unpretreated OPEFB contained 15.2% hemicellulose, 25.8% hemicellulose chemically [9–10,13]. Attractively, an increasing
lignin, 40.9% cellulose, and 18.1% other components. The PS- delignification was noticed from 65.8% at H2SO4 0 mM to 80.5% after
pretreatment at higher PA concentrations (0 to 100, 200, 300, and the first H2SO4 introduction (50 mM) to PA (at a fixed concentration of
400 mM) and fixed conditions (H2SO4 100 mM, 5 mins, 120 ◦ C) led to 300 mM). This provides evidence that the hemicellulose removal
higher delignification (17.8 to 40.9, 65.9, 82.9 and 91.3%, respectively) affected the increasing delignification, similar to other reports [18,24].
(Fig. 1A), showing the PA reactivity to lignin [14–16,18]. Hemicellulose As expected, no further increasing delignification rate was performed by

4
L. Darus et al. Chemical Engineering Journal 429 (2022) 132452

Table 1 hemicellulose removal (88.5%) was higher, and PA concentration used

Optimum performances of pretreatment and enzymatic hydrolysis after PS- (200 mM) was milder than those in that study (85.0% and 300 mM,
pretreatment of different lignocellulosic biomass and different pretreatment of respectively). The raised energy cost due to the application of a higher
OPEFB. temperature (140 ◦ C) than those reported in Lee et al. [18] (120 ◦ C) and
Biomass Methods and Removal (%): Cellulose (%): Ref. the NREL standard method (121 ◦ C) [22] might be balanced by the
Conditions Lignin/ Quality/ integration of renewable energy sources [25]. The 5-min pretreatment at
Hemicellulose Digestibility
140 ◦ C, PA 300 mM, and H2SO4 100 mM might be considered for its
Eucalyptus Thermally-assisted 91.1 / 74.9 73. 8 / ——— [17] highest delignification but had disadvantages for the higher cellulose
regnans PS-pretreatment loss and PA consumption.
(90 ◦ C; 300 min)
Yellow Microwave- 90.0 / 85.0 75.6 / 81.7 [18]
Poplar assisted PS- 3.2. Comparison of PS and DS-pretreatments under similar conditions
pretreatment
(120 ◦ C; 5 min) PS-pretreatment with PA additions (300 and 200 mM) into H2SO4
OPEFB Microwave- 81.3 / 88.5 83.5 / 77.0 This
100 mM (called P300S100 and P200S100) and DS-pretreatment (only
assisted PS- study
pretreatment H2SO4 100 mM available) were conducted at similar conditions (5 min,
(140 ◦ C; 5 min) 140 ◦ C). The effects of PS and DS-pretreatments on component
OPEFB Organosolv: 89.6 / 84.6 80.0 / 94.1 [8] composition, enzymatic hydrolysis, and microstructural characteristic
Sulfuric
of OPEFB were compared.
acid–ethanol
(210 ◦ C; 90 min)
OPEFB Bisulfite (180 ◦ C; 40.2 / 50.8 63.6 / 95.0 [27] 3.2.1. Effect on component composition
30 min) The unpretreated OPEFB contained lignin, hemicellulose, and cel­
OPEFB Choline chloride- 88.0 / >95 71.4 / ——— [28] lulose of 25.8, 15.2, and 40.9 g per 100 g unpretreated OPEFB, respec­
lactic acid (120 ◦ C;
tively (Fig. 3). The PS-pretreatment chemically hydrolyzed lignin and
480 min)
OPEFB Two-stage: acetic 59.5 / 53.6 65.3 / 72.9 [29] hemicellulose, while the DS-pretreatment exclusively hydrolyzed
acid/aqueous hemicellulose. Based on 100 g unpretreated OPEFB, the P300S100 and
ammonia (170/ P200S100 resulted in considerably lower lignin (1.7 and 4.8 g, respec­
220 ◦ C; 15/45 min) tively) than that with DS (21.3 g), comparable hemicellulose (1.9 and
1.8 g, respectively) to that with DS (1.7 g), and lower cellulose (28.7 and
increasing H2SO4 concentration to 50, 100 and 150 mM. Similar to the 33.2 g, respectively) than that with DS (38.9 g). These led to the higher
PA concentration effect, higher H2SO4 concentrations (0 to 50, 100, and quality of P300S100-pretreated (89.1%) and P200S100-pretreated (83.5%)
150 mM) led to higher (but still relatively low) cellulose loss (7.3 to than those of DS-pretreated (62.9%) and unpretreated OPEFB (49.9%).
12.2, 14.7, and 17.7%, respectively) (Fig. 1B). For these reasons, H2SO4 The P300S100 and P200S100 enriched the unpretreated OPEFB quality (1.8
150 mM was preferred for further experiments. and 1.7 times, respectively) more effectively than the DS did (1.3 times).
Fig. 1C shows that delignification and hemicellulose removal for 5 Interestingly, the qualities of PS-pretreated OPEFBs reported in this
min were relatively similar to those for the longer pretreatment times study (140 ◦ C, 5 min) were comparable to that of Organosolv-pretreated
(30, 60, and 90 min) at fixed conditions (PA 300 mM, H2SO4 150 mM, OPEFB (84.8%) conducted at a higher temperature (210 ◦ C) and a longer
120 ◦ C). This might be because the PA and H2SO4 concentrations which time (90 min) [8].
became very low within the 5-min [18]. However, the longer pretreat­
ment times (30, 60, and 90 min) detrimentally increased the cellulose 3.2.2. Effect on cellulose digestibility
degradations (25.3, 29.0, and 35.4%, respectively) due to chemical The enzymatic hydrolyses were conducted for the OPEFB solids
hydrolyses by PA and H2SO4 acids. For its lowest cellulose loss (17.7%), (P300S100, P200S100, DS, and unpretreated) at different cellulase loadings:
the 5-min one was selected as the optimal condition. 5, 20, and 35 EGU/g cellulose (at fixed β-glucosidase loading 30 U/g
The previously optimized variables (PA 300 mM, H2SO4 150 mM, 5 cellulose) and incubation times: 1, 2, 3, 4, and 5 days. Their enzymatic
min) at 120 ◦ C were further used as the basis to observe the effect of a digestibilities were compared and shown in Fig. 4.
higher temperature on PA and H2SO4 dosages. As shown in Fig. 2, The P300S100 provided higher digestible cellulose than the P200S100
increasing temperature from 120 to 140 ◦ C positively led to higher did at all cellulase loadings. The enzyme mixture more successfully
hemicellulose removal (86.2 to 96.6%) and delignification (80.1 to digested the PS (P300S100 and P200S100)-pretreated than the DS-
92.2%), but undesirably led to high cellulose loss (17.7 to 33.6%). pretreated and unpretreated OPEFB. At 35 EGU cellulase/g cellulose,
A lower concentration of H2SO4 100 mM (than 150 mM) at fixed cellulose digestion (5-day) increased from 6.5% in unpretreated to 43.3,
conditions (PA 300 mM, 5 min, 140 ◦ C) caused a lower hemicellulose 77.0, and 93.0% in DS, P200S100, and P300S100-pretreated OPEFB,
removal (96.6 to 87.9%), without significant effect on delignification respectively. The higher cellulose digestibility by the enzyme mixture
(92.2 to 93.5%), and a lower but still high cellulose loss (33.6 to 30.0%). (cellulase and β-glucosidase) was in line with the higher OPEFB quality,
Further, a lower concentration of PA 200 mM (than 300 mM) at fixed as shown in Fig. 3. Therefore, it is predictable that the OPEFB quality
conditions (H2SO4 100 mM, 5 min, 140 ◦ C) decreased delignification enrichment during PS-pretreatment increased the enzyme access to
(93.5 to 81.3%), without significant effect on hemicellulose removal cellulose and further enhanced cellulose hydrolysis conversion to
(87.9 to 88.5%), and importantly led to a lower cellulose loss (30.0 to glucose [7]. The lower digestibility of DS-pretreated than the PS-
18.9%). Hence, the temperature of 140 ◦ C and other conditions (5-min, pretreated OPEFB might be affected by the rest or re-deposited lignin,
PA 200 mM, and H2SO4 100 mM) were chosen as the final optimal which absorbed and blocked the enzymes, thus preventing their activity
conditions for its high performances (delignification, hemicellulose on cellulose [11–12,26].
removal, and cellulose loss), comparable to the basic versions. These The cellulose digestibility in PS and DS-pretreated solids slightly
optimal H2SO4 concentration (100 mM or ~ 1 wt%) and pretreatment increased over a longer incubation time (1 to 5-day) at all enzyme
time (5 min) are milder than those with the two-step hydrolysis of NREL loadings. For example, at a cellulase loading of 35 EGU/g cellulose, the
standard method (4 wt% sulfuric acid; 1 h) [22]. Although delignifica­ cellulose digestion of P200S100-pretreated OPEFB increased from 58.4 to
tion effectiveness of this PS-pretreated OPEFB (81.3%) was lower than 70.4, 73.7, 73.0, and 77.0%, respectively. A higher cellulase loading
that achieved during PS-pretreatment of Yellow Poplar (90.0%) [18], served a slightly higher digestibility for all pretreated OPEFB. At
increasing loading from 5 to 20 and 35 EGU/g cellulose, the cellulose

5
L. Darus et al. Chemical Engineering Journal 429 (2022) 132452

Fig. 5. SEM images of unpretreated (A), DS-pretreated (B), and P200S100-pretreated OPEFB (C) at 75× (1) and 2,500 × magnification (2). The pretreatments were
conducted at fixed conditions: H2SO4 100 mM, 5 min, 140 ◦ C.

digestion of P200S100-pretreated OPEFB (5-day) increased from 67.9 to (at 5 EGU cellulase/g cellulose) of P200S100-pretreated OPEFB and the 5-
74.0 and 77.0%, respectively. day hydrolysis (at 35 EGU cellulase/g cellulose) of DS-pretreated OPEFB
The P200S100 decreased the enzyme consumption and incubation had comparable cellulose digestibility (47.3 and 43.3%, respectively).
time compared to those with the DS. For example, the 1-day hydrolysis Overall, the P200S100 (140 ◦ C for 5 min) was preferable as an optimal

6
L. Darus et al. Chemical Engineering Journal 429 (2022) 132452

Table 2 enzymatic hydrolysis (35 EGU cellulase/g cellulose, 30 U β-glucosi­

Surface area and porosity of the unpretreated, DS-pretreated, and P200S100- dase/g cellulose), P200S100-pretreated OPEFB had 1.8- and 11.9- times
pretreated OPEFB. The pretreatments were conducted at fixed conditions: 5 min, higher digestibility than those of DS-pretreated and unpretreated
140 ◦ C. The value is the average of two analyses. OPEFB, respectively.
OPEFB Solid Surface Area (m2/g) Pore Size (nm) PS-pretreatment has also been considered a promising method for its
Unpretreated 1.835 7.001 effectiveness in delignification and hemicellulose removal from Yellow
DS-pretreated 3.502 9.966 Poplar and Eucalyptus regnans, as shown in Table 1. Although the cel­
P200S100-pretreated 5.821 18.170 lulose digestibility of OPEFB in this study was slightly lower, this
P200S100-pretreatment provides advantages for a relatively lower tem­
perature (140 ◦ C), shorter time (5 min), and simpler process than those
in other methods, including Organosolv (210 ◦ C; 90 min), Bisulfite
(180 ◦ C; 30 min) and Two-stage with acetic acid followed by aqueous
ammonia (170/220 ◦ C; 15/45 min).

3.2.3. Effect on microstructural characteristics

The microstructural characteristics of P200S100-pretreated, DS-
pretreated, and unpretreated OPEFB are compared for their surface
morphology, surface area, porosity, crystallinity, and functional group.
Structural changes after pretreatment displayed consistency with the
effectiveness of delignification and hemicellulose removal. The previ­
ously hammer-milled and sieved (40–60 mesh) unpretreated OPEFB had
a naturally structured, compacted, and neat surface with some flakes
spread on the surface (Fig. 5.A1-5.A2). The DS-pretreatment partly
destroyed the surface, displaying irregular breaks but still preserved the
general structures of the unpretreated OPEFB (Fig. 5.B1-5.B2). A previ­
ous article on DS-pretreated bamboo residues also showed similar
changes in surface morphology [11].
On the other hand, P200S100-pretreatment resulted in an entirely
different form of solid surface with separated cellulose fibers (Fig. 5.C1-
5.C2).
As expected, the P200S100-pretreatment extensively destroyed surface
Fig. 6. FTIR spectra of functional groups related to cellulose (green dashed structure into smaller fragments and led to cellulose exposure, thus
line), hemicellulose (blue dashed line), and lignin (red dashed line) in P200S100- increasing enzyme access for cellulose hydrolysis to glucose. The
pretreated (A), DS-pretreated (B), and Unpretreated OPEFB (C). The pretreat­ destruction was also observed after PA, followed by alkaline peroxide-
ment was conducted at fixed conditions: 5 min, 140 ◦ C.
pretreatment of OPEFB [16], PS-pretreatment of Hard Wood [17], and
Yellow Poplar [18].
The changes in the surface area and pore size are consistent with the
effectiveness of lignin and hemicellulose removals in the P200S100-pre­
treated, DS-pretreated, and unpretreated OPEFB. BET surface area and
pore size of P200S100-pretreated OPEFB (5.821 m2/g and 18.170 nm)
were higher and longer than those of DS-pretreated OPEFB (3.502 m2/g
and 9.966 nm) and unpretreated OPEFB (1.835 m2/g and 7.001 nm),
respectively (Table 2). Lignin and hemicellulose removals’ effects on the
increased surface area and pore size were similarly reported in other
lignocellulosic biomass pretreatments [30–32]. The higher surface area
and pore size caused cellulose exposure, allowing higher cellulose con­
version to glucose by cellulase enzyme [33].
Fig. 6 shows the absorption bands related to the functional groups
associated with lignin, hemicellulose, and cellulose in the PS-pretreated,
DS-pretreated, and unpretreated OPEFB. The changes in absorption
bands were logical with the effectiveness of lignin and hemicellulose
removals and the OPEFB quality. After P200S100-pretreatment, the ab­
sorption bands of lignin and hemicellulose-related functional groups
decreased. They were shown by the disappearance of the C-O-C func­
tional group in hemicellulose, the weaker stretches of lignin-related
bands, and the weaker stretches of lignin-hemicellulose linkage (C =
Fig. 7. The crystallinity of P200S100-pretreated (A), DS-pretreated (B), and
O), respectively at 1230 cm− 1 [10,17–18], 1300–1600 cm− 1 [34–37],
Unpretreated OPEFB solid (C). CrI stands for Crystallinity Index, indicating a
and 1700 cm− 1 [17–18,32,35,37].
relative composition of crystalline cellulose in OPEFB solid. Imax and Iam stand
for the maximum intensity of crystalline and minimum intensity of amorphous In opposite to lignin and hemicellulose, the cellulose-related bands
components in OPEFB solid. The pretreatment was conducted at fixed condi­ increased in P200S100-pretreated OPEFB as observed by Kundu et al. [17]
tions: 5 min, 140 ◦ C. and Lee et al. [18] after PS-pretreatment of Yellow Poplar and Hard
Wood Eucalyptus regnans, respectively. The more vigorous stretches
chemical pretreatment considering the relatively high effectiveness of related to cellulose were observed for C-O–H (alcohols), C-O-C (skeletal
lignin and hemicellulose removal, low cellulose loss, lower chemical ring vibration), C-O-C (glycosidic bond), and C–H at 1030, 1100, 1160
consumption (PA, H2SO4), and high cellulose digestibility. After 5-day cm− 1 [17–18,38], and 2916 cm− 1 [17–18,36,38], respectively. These
suggest removing lignin and hemicellulose during PS-pretreatment,

7
L. Darus et al. Chemical Engineering Journal 429 (2022) 132452

leaving a relatively higher amount of cellulose. The absorption band at Development, the Ministry of Research, Technology, and Higher Edu­
3000–3600 cm− 1 was also associated with the O–H stretching region of cation of Indonesia with Grant Number:150/SP2H/LT/DRPM/2019.
cellulose [38]. A more substantial stretch at 3330 cm− 1 than that in the
unpretreated OPEFB implied that the O–H functional group related to REFERENCES
crystalline cellulose was interfered with the pretreatment. This sub­
stantial stretch near 3330 cm− 1 has already been shown in the previous [1] S.H. Mood, A.H. Golfeshan, M. Tabatabaei, G.S. Jouzani, G.H. Najafic, M. Gholami,
M. Ardjmand, Lignocellulosic biomass to bioethanol, a comprehensive review with
works after white-rot fungus followed by phosphoric acid-pretreatment
a focus on pretreatment, Renew. Sust. Energ. Rev. 27 (2013) 77–93, https://doi.
[34] and microwave-assisted oxalic acid-pretreatment [36] of OPEFB. org/10.1016/j.rser.2013.06.033.
DS-pretreatment caused weaker stretches of hemicellulose-related [2] A.F.P. Harahap, J.R.H. Panjaitan, C.A. Curie, M.Y.A. Ramadhan, P. Srinophakun,
M. Gozan, Techno-economic evaluation of hand sanitiser production using oil palm
and stronger stretches of lignin and cellulose-related functional groups
empty fruit bunch-based bioethanol by simultaneous saccharification and
than those in unpretreated OPEFB. These spectra indicated the removal fermentation (SSF) process, Appl. Sci. 10 (2020) 5987, https://doi.org/10.3390/
selectivity of DS-pretreatment to hemicellulose, leaving relatively app10175987.
higher lignin and cellulose composition in pretreated solid [10,17–18]. [3] X. Zhao, L. Zhang, D. Liu, Biomass recalcitrance. Part I: the chemical compositions
and physical structures affecting the enzymatic hydrolysis of lignocellulose,
The diffractogram intensities in the regions at approximately 18◦ and Biofuels, Bioprod. Bioref. 6 (4) (2012) 465–482, https://doi.org/10.1002/bbb.
22◦ represented the OPEFB amorphous components (lignin, hemicellu­ v6.410.1002/bbb.1331.
lose, and disordered cellulose) and crystalline cellulose, respectively [4] R.H. Narron, H. Kim, H. Chang, H. Jameel, S. Park, Biomass pretreatments capable
of enabling lignin valorization in a biorefinery process, Curr. Opin. Biotech. 38
[32,39]. The CrI in the P200S100-pretreated OPEFB (74.5%) was higher (2016) 39–46, https://doi.org/10.1016/j.copbio.2015.12.018.
than those in the DS-pretreated OPEFB (56.5%) and the unpretreated [5] Q. Yu, X. Zhuang, W. Wang, W. Qi, Q. Wang, X. Tan, X. Kong, Z. Yuan,
OPEFB (33.2%) (Fig. 7). As also occurred in other works [32,40], the CrI Hemicellulose and lignin removal to improve the enzymatic digestibility and
ethanol production, Biomass Bioenerg. 94 (2016) 105–109, https://doi.org/
differences were affected by delignification, hemicellulose removal, and 10.1016/j.biombioe.2016.08.005.
cellulose loss and in line with the quality of pretreated and unpretreated [6] S. Rezania, B. Oryani, J. Cho, A. Talaiekhozani, F. Sabbagh, B. Hashemi, P.
OPEFB. F. Rupani, A.A. Mohammadi, Different pretreatment technologies of lignocellulosic
biomass for bioethanol production: An overview, Energy. 199 (2020) 117457,
https://doi.org/10.1016/j.energy.2020.117457.
4. Conclusions [7] A. George, A. Brandt, K. Tran, S.M.S.N.S. Zahari, D. Klein-Marcuschamer, N. Sun,
N. Sathitsuksanoh, J. Shi, V. Stavila, R. Parthasarathi, S. Singh, B.M. Holmes,
T. Welton, B.A. Simmons, J.P. Hallett, Design of low-cost ionic liquids for
A 5-min pretreatment with PA 200 mM and H2SO4 100 mM at 140 ◦ C
lignocellulosic biomass pretreatment, Green Chem. 17 (3) (2015) 1728–1734,
performed highly delignification (81.3%) and hemicellulose removal https://doi.org/10.1039/C4GC01208A.
(88.5%) with minimal cellulose loss (18.9%). It was found that the [8] K. Mondylaksita, J.A. Ferreira, R. Millati, W. Budhijanto, C. Niklasson, M.
extended time did not affect pretreatment performances, except the J. Taherzadeh, Recovery of high purity lignin and digestible cellulose from oil palm
empty fruit bunch using low acid-catalyzed organosolv pretreatment, Agronomy.
cellulose loss. Five-day enzymatic hydrolysis using 35 EGU cellulase and 10 (2020) 674, https://doi.org/10.3390/agronomy10050674.
30 U β-glucosidase/g cellulose converted 77.0% of cellulose from the [9] Z.N. Akhlisah, R. Yunus, Z.Z. Abidin, B.Y. Lim, D. Kania, Pretreatment methods for
P200S100-pretreated OPEFB to glucose, 1.8- and 11.9-times higher than an effective conversion of oil palm biomass into sugars and high-value chemicals,
Biomass Bioenerg. 144 (2021) 105901, https://doi.org/10.1016/j.
cellulose digestibility of the DS-pretreated and unpretreated OPEFB, biombioe.2020.105901.
respectively. Moreover, the hydrolysis of P200S100-pretreated OPEFB at a [10] T.-C. Hsu, G.-L. Guo, W.-H. Chen, W.-S. Hwang, Effect of dilute acid pretreatment
lower enzyme loading and shorter incubation time achieved comparable of rice straw on structural properties and enzymatic hydrolysis, Bioresour. Technol.
101 (13) (2010) 4907–4913, https://doi.org/10.1016/j.biortech.2009.10.009.
cellulose digestibility with that in the DS-pretreated OPEFB. Micro­ [11] W. Lin, J. Yang, Y. Zheng, C. Huang, Q. Yong, Understanding the effects of different
structural changes in surface morphology, surface area, porosity, crys­ residual lignin fractions in acid-pretreated bamboo residues on its enzymatic
tallinity, and the functional group were consistent with the pretreatment digestibility, Biotechnol Biofuels. 14 (1) (2021), https://doi.org/10.1186/s13068-
021-01994-y.
effectiveness and OPEFB quality. The highly effective P200S100-pre­
[12] D. Gomes, J. Cunha, E. Zanuso, J. Teixeira, L. Domingues, Strategies towards
treatment with mild conditions (pretreatment time, chemical concen­ reduction of cellulases consumption: Debottlenecking the economics of
tration, and temperature) and relatively simple process provide cost and lignocellulosics valorization processes, Polysaccharides. 2 (2021) 287–310,
https://doi.org/10.3390/polysaccharides2020020.
economic benefits. The integration with renewable energy sources paves
[13] J. Akhtar, C.L. Teo, L. Lai, N. Hassan, A. Idris, R. Aziz, Factors affecting
a potential way towards economically feasible bioethanol production delignification of oil palm empty fruit bunch by microwave-assisted dilute acid/
from OPEFB. alkali pretreatment, Bioresources. 10 (2014) 588–596, https://doi.org/10.15376/
BIORES.10.1.588-596.
[14] W. Geng, R.A. Venditti, J.J. Pawlak, H. Chang, Effect of delignification on
CRediT authorship contribution statement hemicellulose extraction from switchgrass, poplar, and pine and its effect on
enzymatic convertibility of cellulose-rich residues, BioRes. 13 (2018) 4946–4963,
Libertus Darus: Conceptualization, Funding acquisition, Project https://doi.org/10.15376/BIORES.13.3.4946-4963.
[15] B. Yimlamai, W. Choorit, Y. Chisti, P. Prasertsan, Cellulose from oil palm empty
administration, Methodology, Investigation, Supervision, Data curation, fruit bunch fiber and its conversion to carboxymethylcellulose, J. Chem. Technol.
Writing – original draft. Susana Susana: Methodology, Validation, Su­ Biotechnol. 96 (6) (2021) 1656–1666, https://doi.org/10.1002/jctb.
pervision, Writing - review & editing. Halasan Sihombing: Resources, v96.610.1002/jctb.6689.
[16] H. Hermansyah, D.N. Putri, A. Prasetyanto, Z.B. Chairuddin, M.S. Perdani,
Methodology, Supervision, Supervision, Writing - review & editing. M. Sahlan, M. Yohda, Delignification of oil palm empty fruit bunch using peracetic
Amaliyah Rohsari Indah Utami: Methodology, Validation, Writing - acid and alkaline peroxide combined with the ultrasound, Int. J. Technol. 10
review & editing. Maizirwan Mel: Methodology, Validation, Writing - (2019) 1523–1532, https://doi.org/10.14716/ijtech.v10i8.3464.
[17] C. Kundu, S.P. Samudrala, M.A. Kibria, S. Bhattacharya, One-step peracetic acid
review & editing. pretreatment of hardwood and softwood biomass for platform chemicals
production, Sci Rep. 11 (1) (2021), https://doi.org/10.1038/s41598-021-90667-9.
Declaration of competing interest [18] H.R. Lee, R.J. Kazlauskas, T.H. Park, One-step pretreatment of yellow poplar
biomass using peracetic acid to enhance enzymatic digestibility, Sci. Rep. 7 (1)
(2017), https://doi.org/10.1038/s41598-017-12542-w.
The authors declare that they have no known competing financial [19] M. Selig, N. Weiss, Y. Ji, Enzymatic saccharification of lignocellulosic biomass
interests or personal relationships that could have appeared to influence [electronic resource]: laboratory analytical procedure (LAP): issue date, 3/21/
2008 / M, N. Weiss, and Y. Ji National Renewable Energy Laboratory Golden, Colo,
the work reported in this paper.
Selig, 2008 http://purl.access.gpo.gov/GPO/LPS94124.
[20] S.W. Pryor, N. Nahar, β-glucosidase supplementation during biomass hydrolysis:
Acknowledgments How low can we go? Biomass Bioenerg. 80 (2015) 298–302, https://doi.org/
10.1016/j.biombioe.2015.06.005.
[21] Z. Zhang, M. Wang, R. Gao, X. Yu, G. Chen, Synergistic effect of thermostable
This work was supported by the Directorate of Research and Com­ β-glucosidase TN0602 and cellulase on cellulose hydrolysis, 3 Biotech. 7 (2017) 54.
munity Service, the Directorate General of Strengthening Research and https://dx.doi.org/10.1007%2Fs13205-017-0672-2.

8
L. Darus et al. Chemical Engineering Journal 429 (2022) 132452

[22] A. Sluiter, B. Hames, R. Ruiz, C. Scarlata, J. Sluiter, D. Templeton, D. Crocker, [32] K.K. Valladares-Diestra, L.P.S. Vandenberghe, L.A.Z. Torres, V.S. Nishida, A.
Determination of structural carbohydrates and lignin in biomass [electronic Z. Filho, A.L. Woiciechowski, C.R. Soccol, Imadazole green solvent pre-treatment
resource]: laboratory analytical procedure (LAP): issue date, 4/25/2008 / A, as a strategy for second-generation bioethanol production from sugarcane bagasse,
Sluiter et al. National Renewable Energy Laboratory Golden, Colo, 2008 http:// Chem. Eng. J. 420 (2021), 127708, https://doi.org/10.1016/j.cej.2020.127708.
purl.access.gpo.gov/GPO/LPS94089. [33] T. Lan, Y. Jiang, W. Zheng, S. Wang, S. Sang, H. Li, Comprehensively
[23] L. Segal, J.J. Creely, A.E. Martin, C.M. Conrad, An empirical method for estimating understanding enzymatic hydrolysis of lignocellulose and cellulase–lignocellulose
the degree of crystallinity of native cellulose using the X-Ray Diffractometer, Text. adsorption by analyzing substrates’ physicochemical properties, Bioenerg. Res. 13
Res. J. 29 (10) (1959) 786–794, https://doi.org/10.1177/004051755902901003. (4) (2020) 1108–1120, https://doi.org/10.1007/s12155-020-10141-8.
[24] H.e. Tan, R. Yang, W. Sun, S. Wang, Peroxide− acetic acid pretreatment to remove [34] Isroi, M. Ishola, R. Millati, S. Syamsiah, M. Cahyanto, C. Niklasson, M. Taherzadeh,
bagasse lignin prior to enzymatic hydrolysis, Ind. Eng. Chem. Res. 49 (4) (2010) Taherzadeh, Structural changes of oil palm empty fruit bunch (OPEFB) after fungal
1473–1479, https://doi.org/10.1021/ie901529q. and phosphoric acid pretreatment, Molecules. 17 (12) (2012) 14995–15012,
[25] M. Martín, I.E. Grossmann, Optimal integration of renewable based processes for https://doi.org/10.3390/molecules171214995.
fuels and power production: Spain case study, Appl. Energy. 213 (2018) 595–610, [35] S.H. Anita, Fitria, N.N. Solihat, F.P. Sari, L. Risanto, W. Fatriasari, E. Hermiati,
https://doi.org/10.1016/j.apenergy.2017.10.121. Hermiati, Optimization of microwave-assisted oxalic acid pretreatment of oil palm
[26] L. Kumar, V. Arantes, R. Chandra, J. Saddler, The lignin present in steam empty fruit bunch for production of fermentable sugars, Waste Biomass Valor. 11
pretreated softwood binds enzymes and limits cellulose accessibility, Bioresour. (6) (2020) 2673–2687, https://doi.org/10.1007/s12649-018-00566-w.
Technol. 103 (1) (2012) 201–208, https://doi.org/10.1016/j. [36] N.N. Solihat, F.P. Sari, L. Risanto, S.H. Anita, F. Fitria, W. Fatriasari, E. Hermiati,
biortech.2011.09.091. Disruption of oil palm empty fruit bunches by microwave-assisted oxalic acid
[27] L. Tan, Y. Yu, X. Li, J. Zhao, Y. Qu, Y.M. Choo, S.K. Loh, Pretreatment of empty pretreatment, J. Math. Fund. Sci. 49 (3) (2017) 244, https://doi.org/10.5614//j.
fruit bunch from oil palm for fuel ethanol production and proposed biorefinery math.fund.sci.2017.49.310.5614/j.math.fund.sci.2017.49.3.3.
process, Bioresour. Technol. 135 (2013) 275–282, https://doi.org/10.1016/j. [37] X. Tao, J. Li, P. Zhang, M. Nabi, S. Jin, F. Li, S. Wang, J. Ye, Reinforced acid-
biortech.2012.10.134. pretreatment of triarrhena lutarioriparia to accelerate its enzymatic hydrolysis, Int.
[28] Y.T. Tan, G. Ngoh, A.S.M. Chua, Evaluation of fractionation and delignification J. Hydrog. Energy. 42 (29) (2017) 18301–18308, https://doi.org/10.1016/j.
efficiencies of deep eutectic solvents on oil palm empty fruit bunch, Ind. Crop. ijhydene.2017.04.149.
Prod. 123 (2018) 271–277, https://doi.org/10.1016/j.indcrop.2018.06.091. [38] S.M. Nomanbhay, R. Hussain, K. Palanisamy, Microwave-assisted alkaline
[29] D.Y. Kim, Y.S. Kim, T.H. Kim, K.K. Oh, Two-stage, acetic acid-aqueous ammonia, pretreatment and microwave assisted enzymatic saccharification of oil palm empty
fractionation of empty fruit bunches for increased lignocellulosic biomass fruit bunch fiber for enhanced fermentable sugar yield, J. Sustain. Bioenergy Syst.
utilization, Bioresour. Technol. 199 (2016) 121–127, https://doi.org/10.1016/j. 03 (01) (2013) 7–17, https://doi.org/10.4236/jsbs.2013.31002.
biortech.2015.09.049. [39] Hemansi, R. Gupta, V.K. Aswal, J.K. Saini, Saini, Sequential dilute acid and alkali
[30] Y. Su, X. Yu, Y. Sun, G. Wang, H. Chen, G. Chen, Evaluation of screened lignin- deconstruction of sugarcane bagasse for improved hydrolysis: Insight from small
degrading fungi for the biological pretreatment of corn stover, Sci. Rep. 8 (1) angle neutron scattering (SANS), Renew. Energy. 147 (2020) 2091–2101, https://
(2018), https://doi.org/10.1038/s41598-018-23626-6. doi.org/10.1016/j.renene.2019.10.003.
[31] K. Ninomiya, A.R.I. Utami, Y. Tsuge, K. Kuroda, C. Ogino, T. Taima, J. Saito, [40] L.M.F. Pardo, J.G.S. Mendoza, J.E.L. Galán, Influence of pretreatments on
M. Kimizu, K. Takahashi, Pretreatment of bagasse with a minimum amount of crystallinity and enzymatic hydrolysis in sugar cane residues, Braz. J. Chem. Eng.
cholinium ionic liquid for subsequent saccharification at high loading and co- 36 (2019) 131–141, https://doi.org/10.1590/0104-6632.20190361s20180093.
fermentation for ethanol production, Chem. Eng. J. 334 (2018) 657–663, https://
doi.org/10.1016/j.cej.2017.10.113.