Australian Journal of Basic and Applied Sciences, 11(3) Special 2017, Pages: 114-119

AUSTRALIAN JOURNAL OF BASIC AND

APPLIED SCIENCES
ISSN:1991-8178 EISSN: 2309-8414
Journal home page: www.ajbasweb.com

Simulation of Dilute Acid Hydrolysis of Wood Sawdust for Xylose

Production using Aspen Plus
N.H. Ibrahim, W.H.W. Ibrahim, A.M. Mimi Sakinah, N. Harun, Mohamad Rizza Othman

Faculty of Chemical Engineering & Natural Resources, Universiti Malaysia Pahang, Lebuh Raya Tun Razak, 26300 Gambang, Pahang,
Malaysia.

Address For Correspondence:

Wan Hanisah Wan Ibrahim, Faculty of Chemical Engineering & Natural Resources, University Malaysia Pahang, Lebuh Raya Tun Razak,
26300 Gambang, Pahang, Malaysia.
E-mail: [email protected], Tel : 09-5492905

ARTICLE INFO ABSTRACT

Article history: Dilute acid hydrolysis is one of the most effective and inexpensive pretreatment
Received 18 September 2016 methods for lignocellulosic biomass. Generally in hydrolysis process, monomeric
Accepted 21 January 2017 sugars and soluble sugar oligomers (xylose and glucose) from the cell wall matrix are
Available online 26 January 2017 released into the hydrolysate. The aim of this study is to develop and simulate process
model including heating, hydrolysis, neutralization, separation and cooling process for
xylose production using Aspen Plus simulator. Based on experimental study for xylose
production, a flowsheet has been proposed for a possible commercial production.
Keywords: Sawdust was used as a feedstock for the simulation as it has above 29% of xylan
Sawdust; Simulation; Aspen Plus; Acid content, which is a promising source for xylose recovery. The hydrolysis of
Hydrolysis; Kinetic reaction hemicellulose and cellulose to its monomeric constituents and possible degradation of
these monomers were modeled by a series of first-order reactions. A kinetic and
stoichiometric based reactor was used for dilute acid hydrolysis and neutralization
process. The main reactions involved were implemented for both reactor with specified
kinetic parameter and conversion factor. The results from the simulation were
compared with experimental study from literature. Results from the developed process
model show a good agreement with experimental study with 2.4% overall mean error
for xylose production and 4.5% overall mean error for glucose production.

INTRODUCTION

The production of fuel, gas and specialty chemical product from lignocellulosic feedstock has emerged as a
worldwide market that is becoming increasingly important, drawing scientists and public attention. The
biotechnology route for these productions has gaining interest among researchers due to several driving force;
the global warming, the increase of the crude oil price and legislative restrictions regarding the use of non-
renewable energy source (Procentese et al., 2014). Many researchers agree that using lignocellulosic material as
a renewable energy sources in these productions is a great potential solution for this global issue. Licht FO
(2006) was reported, less than 4% of the ethanol is produced synthetically from oil, while the rest is produced by
fermentation from bio resources. The low cost and renewable nature of lignocellulosic material such as corn
stover, sawdust and sugarcane bagasse represents the ideal feedstock for their conversion into value-added
products by biotechnological routes. The biotechnological route present several advantages on the chemical one,
such as moderate conditions of temperature, pH, pressure, high product selectivity and product biodegradation
(Martínez & Santos, 2012). Therefore alternatives for the use of these materials to avoid the environment
degradation were investigated. Meranti wood sawdust (MWS) is a lignocellulosic material that is abundantly
wasted throughout the year in Malaysia. The high amount of xylan (29.22%) content in MWS makes this
Open Access Journal
Published BY AENSI Publication
© 2017 AENSI Publisher All rights reserved
This work is licensed under the Creative Commons Attribution International License (CC BY).
http://creativecommons.org/licenses/by/4.0/

To Cite This Article: N.H. Ibrahim, W.H.W. Ibrahim, A.M. Mimi Sakinah, N. Harun, Mohamad Rizza Othman., Simulation of Dilute Acid
Hydrolysis of Wood Sawdust for Xylose Production using Aspen Plus. Aust. J. Basic & Appl. Sci., 11(3): 114-119, 2017
115 N.H. Ibrahim et al, 2017
Australian Journal of Basic and Applied Sciences, 11(3) Special 2017, Pages: 114-119

biomass sufficient for xylose production (Rafiqul & Sakinah, 2012). The hemicellulosic fraction of MWS is
easily hydrolyzed by dilute acid pretreatment process to produce xylose and glucose that can be used as a
potential substrate for microbial growth and for the production of specialty chemicals. Benefit of using xylose as
carbon and energy source for microorganisms are the reduction of commercial xylose utilization and the
preparation of industrially important enzyme xylose reductase that enhances the economy of the bioprocess. In
the world of software development, process simulation is becoming a requirement for all major process designs.
Therefore necessity of simulation software vendor’s work closely with the biobased producers to make sure that
the software can help the production plants perform at peak operation condition. Established bioprocess
simulators for these processes including Aspen Plus, HYSYS and SuperPro Designer that allow for the efficient
and optimal process design, regulatory compliance and operational analysis of the process can benefit the
production of biobased product. The proposed simulator; Aspen Plus will be selected as the best simulator tool
for production of xylose in commercial scale. Recently Aspen Plus is the most widely used commercial process
simulation software for steady state simulation as Aspen Plus offers many advantages as a process simulator
(Luyben, 1997). Previous study by (Procentese et al., 2014) using Aspen Plus discussed the economic potential
(EP) for developed process model including upstream, downstream and fermentation section for butanol
production by conversion of hardwood biomass. The EP of the process for the proposed flowsheet is about
1,254 k€/yr by determining the butanol production rate at 251 kg/h and assuming that cost of the raw materials
is negligible. Study by (García et al., 2013) using Aspen Plus to simulate different bio-transformation stages for
different bio products; ethanol, xylitol, lactic acid and citric acid from pretreated lignocellulosic feedstock
showed that the developed simulations allowed to evaluated mass and energy balances for each bioprocess
which resulting in 133, 310 162 and 157 kg/hr of bioproduct outputs. The ethanol production process resulted in
the lowest cooling and heating utilities required (2.6 and 3.0 kW) as compared to other three products. The
major advantage over many other software packages is it can easily handles solids. Besides the thermodynamic
models and the unit operation models are already built in, so there is no need to program them individually.
Even with all the built in abilities, Aspen plus is easily customized when required (Wooley et al., 1999).

MATERIALS AND METHODS

Dilute acid hydrolysis (DAH) simulation was modeled as a steady state process. The process was simulated
with Aspen Plus software version 8.0, employing the following stages: mixing, heating, dilute acid hydrolysis
reactor, cooling, neutralisation reactor and solid separation. This process model was applied to the hydrolysis of
wood sawdust. Two inlet streams were inserted; the solid biomass and liquid substrate (sulphuric acid and
water).
NRTL (Non –Random Two Liquid) model is the most appropriate thermodynamic model for the
components properties estimation owing to system characteristic which is operation pressures lower than 10 atm
and system compounds are no-electrolyte polar substance (Peralta-Ruíz et al., 2012). According to the
compositions of the wood sawdust it was necessary to create hypothetical components (those that are not present
in the database of the simulator). The biomass input stream was assumed to consist of pure elemental solids
which are xylan, cellulose, and lignin and modeled as a combined solid feed stream and the mass fraction of
each components are implemented according to Rafique and Sakinah (2012). Xylan, cellulose, lignin, calcium
oxide and calcium sulphate was classes as conventional inert solids (CI solids) where these components are inert
to phase equilibrium and salt precipitation/solubility and possible to react with conventional components (Aspen
Tech User Manuals, 2000). The properties provided to the simulator for the conventional inert solids (cellulose,
lignin, calcium oxide and calcium sulphate) were implemented according to Aden (2011) by input the molecular
weight. Xylan component was defined by specified the heat of formation (DHSFRM), molecular weight and
imported the molecular structure in component specification column. All these components were located in the
CISOLID substream. For this study, component of extractive and ash is assumed to be inert thus the values were
not added in simulation. The components specified in the biomass input stream was listed in table 1
respectively.

Table 1: Sawdust component

Mass Fraction
Component
(%w/w)
Hemicellulose (xylan) 0.31
Cellulose (gulcan) 0.41
Lignin 0.25
Extractives 0.03

For acid hydrolysis process, all the equations used in this study were inserted as kinetic reaction (based on
rate constant and activation energy) and conversion reactions (based on stoichiometric) were inserted for
neutralisation process and calculation was made in order to obtain the conversion for the optimum conditions of
116 N.H. Ibrahim et al, 2017
Australian Journal of Basic and Applied Sciences, 11(3) Special 2017, Pages: 114-119

previous work. For the separation process, separator was chosen as a solid separation unit owing the properties
for solid liquid separation.
In this work it was assumed that only xylose was produced from the xylan breakdown and only the fast
hydrolyzing fraction was included. The reaction of xylan hydrolysis was considered by using first order reaction
kinetics dependence on reactant concentration with an Arrhenius temperature associated for the first order rate
constants (ki) following equation 1. Decomposition rate of cellulose to glucose was also considered by using
first order reaction kinetics. Where ki0 is the pre-exponential factor and Ea is the activation energy.
−𝐸𝑎
ki = 𝑘ᵢ₀𝑒 𝑅𝑇 (1)

From the above equation, the reaction constant ki increases exponentially with temperature and pre-
exponential factor ki0 is the link between temperature and acid concentration variables in the reactions.
Therefore, the kinetic model can be correlated with temperature and acid concentration through the empirical
equation below;

𝑘ᵢ₀ = 𝑏ᵢ[𝐶𝑎]ᵅⁱ (2)

Where bi and ɑᵢ is the regression parameter determined from experiment study and Ca is sulphuric acid
concentration (% w/w). The ki0 values of sawdust material for hydrolysate products can be calculated using
generalized model developed by Rafique & Sakinah (2012) as below;

k10 = 0.01778 Ca1.2453 (3)

k20 = 0.00276 Ca 0.44 (4)
k30 = 0.00662 Ca 0.8135 (5)
k40 = 0.00288 Ca 0.59 (6)
k50 = 0.03003 Ca 0.5411 (7)

Two inlet stream; 3 g/min of sawdust and 24 g/min of dilute sulfuric acid (6% w/w) were mixed and the
outlet stream temperature were heat up to 130⁰C in a heat exchanger (heater). The outlet solution are then feed
to kinetic reactor for hydrolysis process. The optimal temperature and residence time in the hydrolysis reactor
were claimed to be 130⁰C and 60 min respectively. Kinetic parameters used for the reaction process are based
on experimental data and details are listed in the table 2. The hydrolysate solution of the reactor were cooled
down to 25⁰C using heat exchanger (cooler) prior to filtration in order to facilitate gypsum formation and
removal from hydrolysate solution. The hydrolysate solution are then feed to conversion reactor for
neutralization process where calcium oxide flakes are added at 25⁰C and atmospheric pressure and the acid-base
reactions took place to form salt and water. The input of conversion value were calculated based on initial
content of xylan (29g xylan/100g MWS on o.d.b.) and cellulose (41g cellulose/100g MWS on o.d.b) in MWS.
The remaining sulfate in the hydrolysate solution were then separated using solid separator (separator) prior to
fermentation process. Temperature of hydrolysis reactor were then varies to 105⁰C, 115⁰C and 125⁰C to
validate the developed simulation model result with experimental result.

Table 2: Kinetic parameter consider for the reaction in pretreatment reactor

130⁰C
Reaction
K1 (min-1) Ea (kJ/mol)
Xylan → Xylose
C5H8O4 + H2O → C5H10O5 0.16258 116.430

Cellulose → Glucose
C6H10O5 + H2O → C6H12O6 0.03006 30.862

Xylan → Furfural
C5H10O5 → C5H4O2 + 3H2O 0.00842 11.351

Xylan → Acetic Acid

C5H8O4 + H2O → 2.5C2H4O2 0.08012 13.049

RESULT AND DISCUSSION

The Process Flowsheet Window and the Model Library in Aspen Plus allow the user to construct the
flowsheet graphically (Aspen Tech User Manuals, 2003). The model library is equipped with an array of process
equipments, modifiers and connectors from which a process plant and its sections can be designed.
117 N.H. Ibrahim et al, 2017
Australian Journal of Basic and Applied Sciences, 11(3) Special 2017, Pages: 114-119

Figure 1 shows the detailed process flow diagram of xylose production from wood sawdust. The
hydrolysate solution produced from dilute acid hydrolysis process were contains fermentable sugars; xylose,
glucose, furfural and acetic acid respectively. Products from the partial conversion of the lignin; phenolic
compounds, ferulic acid and from xylan and cellulose; HMF, oligomer, arabinose were not included but they are
assumed to have the same reactions and conversions as lignin, furfural and acetic acid. In the hydrolysis stage,
xylan and cellulose are hydrolyzed to mono sugars (xylose and glucose). Severe hydrolysis condition (high
temperature, strong acid concentration and longer residence time) may further hydrolyzed part of the produced
xylose to furfural. Acetic acid is also released from the liberation of the acetyl groups associated with xylan.

Temperature (F)

77 COOLER Pressure (psia)

15

77
DACID 77 15
S7 29
77
320 S10
15
S6 15

77 266 S8
77 77
MIXER
S4 15
15 15 15
NETRLZTN
SEPARATO
S5 77
SAWDUST S11
S9 15
HEATER

HYDRLYS

Fig. 1: Flowsheet of model development process for the xylose production using Aspen Plus

Model validation for xylose and glucose production are plotted in figure 2 and figure 3 respectively. Results
show a good agreement with previous experimental study. In this process 6.04, 9.09, 15.08 and 14.95 kg/hr of
xylose were produced from the conversion of xylan. The highest xylose concentration produced is 15.08 g/L at
125°C. Xylose concentration is rapidly increased at temperature 115°C to 125°C and start to decrease at
temperature 130°C. The efficiency of the overall process model was calculated from the value of product
produced in filtrate stream. Simulation result was validate with experimental result to evaluate the error of the
process model. The highest error was at 105°C with 3.5% and the smallest error was at 130°C with 1.3%.
Overall the mean error is 2.4%. Glucose formation also show a good trend with experimental study by Rafique
and Sakinah (2012). In this process 1.11, 2.31, 2.54 and 4.06 kg/hr of glucose were produced from the
conversion of cellulose. The highest glucose concentration produced is 4.06 g/L at 130°C. Glucose
concentration is rapidly increased at temperature 125°C to 130°C and expected to continuously increased with
increased temperature. The highest error was at 105°C with 5.1% and the smallest error was at 130°C with
4.0%. Overall the mean error for glucose production is 4.5%. The energy consumed for heating was 141.5 kJ/hr
respectively. This results are on a same order of magnitude as those presented by Souza et al., (2013), where
154.4 kJ/hr were consumed to heat the banana waste (pulp) to 125⁰C. High concentration of xylose and glucose
is prior to fermentation process for the various bio products; ethanol and xylitol. From the result it was evident
that the developed process model were essential to produce high xylose concentration and low concentration of
inhibitors were produced. Generally, higher temperature (>130⁰C) is preferable for cellulose hydrolysis and
longer hydrolysis process will result in decreasing xylose concentration due to degradation to inhibitor
(Chiaramonti et al., 2012). Under this condition, xylan was decomposed very fast to xylose and degraded
completely to furfural. A considerable amount of xylose was released within 40-80 min and temperature
between 120-130⁰C was critical to xylose recovery. Higher temperature could enhance the productivity of the
xylose and glucose due to increase in collision of the sulphuric acid between feedstock. When temperature is
raised to a certain level, part of hydrogen bond fracture among feedstock lead to the unstable structure thus
increase the accessibility of acid molecules to the inner layers of the feedstock.
118 N.H. Ibrahim et al, 2017
Australian Journal of Basic and Applied Sciences, 11(3) Special 2017, Pages: 114-119

experimental simulation
16.00
Xylose Concentration (G/L)
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
105 110 115 120 125 130 135
Temperature °C

Fig. 2: Production of xylose from experimental study by Rafique and Sakinah (2012) and from simulation using
Aspen Plus

experimental simulation
4.50
4.00
Glucose Concentration (G/L)

3.50
3.00
2.50
2.00
1.50
1.00
0.50
0.00
105 110 115 120 125 130 135
Temperature °C

Fig. 3: Production of Glucose from experimental study by Rafique and Sakinah(2012) and from simulation
using Aspen Plus

In general, furfural and acetic acid formation increased with operating temperature. In the hydrolysis of
wood sawdust, furfural is generated as the main decomposition product with high pentose content. The
concentration of furfural formed at 105⁰C, 115⁰C, 125⁰C and 130⁰C was 0.0037, 0.0665, 0.2372 and 0.3865 g/L
respectively. Although the degradation of xylose to furfural was observed at all condition, the low amount of
furfural formed can be negligible compared to the maximum amount of xylose produced 15.08 g/L at 125⁰C.
Acetic acid is also produced in hydrolysis of wood sawdust. This product is generated for the hydrolysis of the
acetyl groups present in the hemicellulosic heteropolymer (Kim et al., 2012). The degradation reactions of
acetic acid was slowly increased with increasing temperature. 5.04 g/L of acetic acid was recovered at 105⁰C
and the concentration was slightly increased to 5.16, 5.24 and 5.44 g/L at 115⁰C, 125⁰C and 130⁰C. The low
concentration of acetic acid showed that at low temperatures below 130 ⁰C, the only reaction conditions found
to result in acetyl removal. The higher temperature and longer reactions times would be required to obtain
significant acetic acid production.

Conclusion:
MWS biomass contains above 29% xylan, which is a promising source for xylose production. Simulation of
MWS hydrolysis was carried out with sulfuric acid under optimum experimental parameter to obtain a high
concentration of xylose with low byproducts in the resulting hydrolysate. The maximum recovery of xylose is
15.08 g/L which was achieved when the reaction was conducted at 125⁰C for 60 min with 6% H2SO4 and LSR
of 8 g/g. With 2.4% mean error it can be conclude that the simulation for xylose production is valid. For glucose
production, the highest concentration produced is 4.06 g/L at 130°C. It was found that xylose production was
119 N.H. Ibrahim et al, 2017
Australian Journal of Basic and Applied Sciences, 11(3) Special 2017, Pages: 114-119

highly affected by temperature, acid concentration and residence time. The use of MWS to produce xylose will
not only solve the disposal problem but also give back a high value product to the wood industry. The adequate
result of the model validation by using Aspen Plus is a sequences method to examine the sensitivity of process
parameter to process alternatives, equipment and operating cost reductions, and advances in conversion
research.

ACKNOWLEDGEMENT

The authors wish to thank to the Graduate Research Scheme (Grant no. GRS 150385), UMP for the
financial assistance provided to complete this project.

REFERENCES

Aspen, Aspen Plus: Getting Started Building and Running a Process Model., 2003. Aspen Technology.
Aspen, Aspen Plus: Introduction to Flowsheet Simulation., 2000. Aspen Technology.
Chiaramonti, D., M. Prussi, S. Ferrero, L. Oriani, P. Ottonello, P. Torre, and F. Cherchi, 2012. Review of
pretreatment processes for lignocellulosic ethanol production, and development of an innovative method.
Biomass and Bioenergy, 46: 25-35.
García, A., I. Egüés, C. Sánchez, Z. Barta and J. Labidi, 2013. Study of Different Bio-Processing Pathways
in a Lignocellulosic Biorefinery by Process Simulation. Chemical Engineering Transactions, 35: 505-510.
Humbird, D., R. Davis, A. Aden, L. Tao, C. Kinchin, and D. Hsu, 2011. Process Design and Economics for
Biochemical Conversion of Lignocellulosic Biomass to Ethanol. NREL Report TP-5100-47764.
Kim, J.S., W.I. Choi, M. Kang, J.Y. Park and J. Lee, 2012. Kinetic Study of Empty Fruit Bunch Using Hot
Liquid Water and Dilute Acid. Appl. Biochem. Biotechnol., 167: 1527-1539.
Martínez, E.A. and J.A.F. Santos, 2012. Influence of the use of rice bran extract as a source of nutrients on
xylitol production. Ciência E Tecnologia de Alimentos, 32(2): 308-313.
Peralta-Ruíz, Y., Y. Pardo, Á. González-Delgado and V. Kafarov, 2012. Simulation of bioethanol production
process from residual microalgae biomass. Computer Aided Chemical Engineering, 30: 1048-1052.
Procentese, A., T. Guida, F. Raganati, G. Olivieri, P. Salatino and A. Marzocchella, 2014. Process
simulation of biobutanol production from lignocellulosic feedstocks. Chemical Engineering Transactions, 38:
343-348.
Rafiqul, I.S.M. and A.M.M. Sakinah, 2012. Design of process parameters for the production of xylose from
wood sawdust. Chemical Engineering Research and Design, 90(9): 1307-1312.
Souza, P.K., N. De Sellin, O. Souza and C. Marangoni, 2013. Simulation of Dilute Acid Hydrolysis of
Banana Waste for Ethanol Production : Comparison between the Use of Fruits, Peel and Pseudostem. Chemical
Engineering Transactions, 32: 1141-1146.
Wooley, R., M. Ruth, J. Sheehan, H. Majdeski and A. Galvez, 1999. Lignocellulosic Biomass to Ethanol
Process Design and Economics Utilizing Co-Current Dilute Acid Prehydrolysis and Enzymatic Hydrolysis
Current and Futuristic Scenarios. NREL Report TP-580-26157.