Bioresource Technology 249 (2018) 268–275

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

An integrated green process: Subcritical water, enzymatic hydrolysis, and MARK

fermentation, for biohydrogen production from coconut husk
⁎
Maktum Muharja, Fitri Junianti, Dian Ranggina, Tantular Nurtono, Arief Widjaja
Department of Chemical Engineering – Institut Teknologi Sepuluh Nopember, Surabaya 60111, Indonesia

G RA P H I C A L AB S T R A C T

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

Keywords: The objective of this work is to develop an integrated green process of subcritical water (SCW), enzymatic
Biohydrogen hydrolysis and fermentation of coconut husk (CCH) to biohydrogen. The maximum sugar yield was obtained at
Coconut husk mild severity factor. This was confirmed by the degradation of hemicellulose, cellulose and lignin. The tendency
Subcritical water of the changing of sugar yield as a result of increasing severity factor was opposite to the tendency of pH change.
Enzymatic hydrolysis
It was found that CO2 gave a different tendency of severity factor compared to N2 as the pressurizing gas. The
Fermentation
result of SEM analysis confirmed the structural changes during SCW pretreatment. This study integrated three
steps all of which are green processes which ensured an environmentally friendly process to produce a clean
biohydrogen.

1. Introduction Coconut husk (Cocos nucifera) (CCH) which is produced about

5.1 million tons per year is the one of abundant crop residue in
The world faces the progressive depletion of its energetic resources Indonesia (Mahmud and Ferry, 2005). CCH which has the main con-
that mainly based on non-renewable fuels. One renewable solution is tains 26.72% cellulose, 17.73% hemicellulose can be a potential feed-
the use of biohydrogen form of biomass (Sánchez and Cardona, 2008). stock candidate for biohydrogen production (Sangian et al., 2015a).
In agricultural dominating countries like Indonesia, the crop residue However, because of the highness of lignin content (about 41.19%), it is
and waste which is renewable, sustainable, biodegradable, and en- not easy work to directly convert CCH into biohydrogen.
vironmentally friendly have great potential for biohydrogen produc- Production of biohydrogen was well performed from lignocellulosic
tion. material using acid pretreatment (Gonzales et al., 2016b; Kumar et al.,

⁎
Corresponding author.
E-mail address: [email protected] (A. Widjaja).

http://dx.doi.org/10.1016/j.biortech.2017.10.024
Received 15 July 2017; Received in revised form 5 October 2017; Accepted 6 October 2017
Available online 09 October 2017
0960-8524/ © 2017 Elsevier Ltd. All rights reserved.
M. Muharja et al. Bioresource Technology 249 (2018) 268–275

2017a). An acid or alkaline solution is often used in the pretreatment

process due to its low price, simplicity of the process, resulting in a high
yield. Although acid and alkaline hydrolysis are relatively fast methods
that produce high sugars concentrations, the reaction medium must be
neutralized after the process, generating solid waste (Prado et al.,
2016). The subcritical water (SCW) technique is one of the most pro-
mising technologies, which is essentially break down the recalcitrant
structure of the lignocellulosic material in order to increase the acces-
sibility of cellulose and hemicellulose polymers to cellulolytic enzymes
(Koppram et al., 2014; Sangian et al., 2015b). One of the best strategies
to convert coconut husk into fermented sugars is by using enzymatic
hydrolysis – due to its low energy requirement and less pollution caused Fig. 1. The design of subcritical water apparatus.
– however, the major problem of enzymatic hydrolysis is the low ac-
cessibility of cellulose, due to the rigid association on cellulose with 2.2. Methods
lignin. It is also required to alter the biomass macroscopic and micro-
scopic size and structure as well as its sub-microscopic chemical com- 2.2.1. Subcritical water pretreatment
position and structure so that hydrolysis of the carbohydrate fraction to Fig. 1 shows apparatus design of SCW that is adapted from work by
monomeric sugars can be achieved more rapidly and with greater yields Ju et al. (2011). It consists of three main parts, a subcritical reactor,
(Mosier et al., 2005). In other words, SCW is crucially and costly unit heater, and control devices. They are a stainless-steel reactor with inner
process in converting coconut husk into hydrogen. volume about 100 ml in total, a tube of carbon dioxide (CO2), pressure
Due to the severe process conditions, industrial application of this regulator valve, output ball valve, jacket heater, a temperature con-
processes suffers from various challenges (Toor et al., 2011). Depend on troller (PID), and a pressure gauge. PID and pressure indicator were
the operational conditions, the degradation products are also formed, connected to the reactor. The process was run under batch mode.
both from sugars (furan and its derivates and weak acids) and, to a less Briefly, 6 g of coconut husk and 120 ml of deionized water were
extent, from lignin (phenolics). These compounds may also inhibit the added to the reactor. CO2 was supplied to the reactor until reached
later fermentation processes, leading to lower hydrogen yields and 80 bar and set the temperature at 150, 160, 180, 200 °C with a reaction
productivities and, therefore, might be required to carry out a prior time of 60 min when the temperature inside the reactor reached the
detoxification treatment (Gírio et al., 2010). Despite these limitations, desired temperature of the reaction. In the end of the process, the re-
the SCW technique has been proved to be an environmentally benign actor was immediately cooled down by cooling water until 30 °C and
medium for a number of chemical and related processes in the last few the pressure was released instantaneously using the ball valve. The
decades (Wen et al., 2009). The challenge of this pretreatment method extracted sample then filtered using filter paper, and the residue was
is the adequate fractionation of hemicelluloses, cellulose and lignin, washed with deionized water. It was then dried in the oven at 60 °C for
together with a minor degradation, in order to get maximal fermenta- 2 d until constant weight and stored for analysis.
tion yields and rates. In order to make comparison, nitrogen gas was also employed in the
Although subcritical water process is able to degrade hemicellulose present study.
into dissolved sugars in the liquid fraction of SCW, there is still a re-
sidue of cellulose in the solid fraction. Therefore combining subcritical
2.2.2. Enzymatic hydrolysis
water and enzymatic hydrolysis was investigated to degrade the re-
Enzymatic hydrolysis was conducted at the optimal condition based
maining cellulose in the SCW solid fraction into glucose which can be
on previous work (Sangian et al., 2015a). The results of the solid from
fermented into hydrogen (Ren et al., 2009). The combination of su-
SCW-treated were hydrolyzed using commercial cellulase from Asper-
percritical CO2 and enzymatic hydrolysis has been studied by Alinia
gillus niger (Sigma Aldrich, Japan) and Commercial endo-1,4-β-xylanase
et al. (2010) and Gao et al. (2010) by using a mixture of commercial
from Trichoderma longibrachiatum (Sigma Aldrich, Mexico). Enzymatic
cellulase and β-glucosidase enzymes to enhance sugar yield from rice and
hydrolysis was conducted at following reaction conditions: pH of 3, the
wheat straw. Kumar et al. (2017b) analyses biohydrogen production
temperature of 60 °C, with incubation in a shake flask at 125 rpm for
from lignocellulosic biomass using dark fermentation and studied the
48 h.
bottlenecks associated with the pretreatment. However, as long as the
literature studied, there has been no publication that reported studies of
applying integration of three subsequent green processes, subcritical 2.2.3. Fermentation
water-enzymatic hydrolysis-fermentation to produce hydrogen from Enterobacter aerogenes NBRC 13534 was used in this fermentation
coconut husk. studies. E. aerogenes was cultured on potato dextrose agar (glucose
The objective of this work is to develop methods for efficient SCW 20 g L−1, peptone 10 g L−1, yeast extract 5 g L−1, and agar 16 g L−1) at
and enzymatic hydrolysis of coconut husk into fermentable sugars and 30 °C. The hydrolysate was obtained from SCW-treated coconut husk
fermentation of hydrolysate to hydrogen. The effect of the severity that has the highest total reducing sugar (TRS) concentration.
factor on sugar yield was also discussed to determine the best condition Inoculums were prepared by acclimation process, inoculating 10 oses of
of SCW. E. aerogenes to 8 mL pre-seed medium which added 0.35 g L−1
FeSO4·7H2O and incubating for 14–16 h. After incubation, the cell
2. Materials and methods concentration was confirmed about 10 million cells/mL. The fermen-
tation process was conducted on the 50 ml of working volume of fer-
2.1. Materials mentor which connected to hydrogen bag (CEL Scientific Tedlar Gas
Sampling Bags, USA), incubated at 37 °C, adjusted to pH 7 using 4 M
Coconut husk was obtained in Manado, North Sulawesi, Indonesia, NaOH, agitated at 125 rpm for 48 h, and insured no leak for the anae-
it was firstly dried under the sunlight for 2 d, then in an oven at 105 °C. robic process.
The material was milled and screened by using screener (Retsch GmBh
Rheinische Strade 36 4278, Haan, Germany) to obtain a particle size of 2.2.4. Analytical methods
120 mesh. After that it was washed with distilled water, dried at 105 °C The composition of hemicellulose, cellulose and lignin at the un-
until constant weight and stored in a desiccator. treated and SCW-treated condition were analyzed gravimetrically

269
M. Muharja et al. Bioresource Technology 249 (2018) 268–275

which was previously reported by Datta (1981). Liquid fractions and the bonds from each of its monomers (i.e. xylose, arabinose, fructose)
solid residues were thoroughly analyzed. Liquid fraction from SCW and through reactions of dissolution and hydrolysis to water-soluble sugars
enzymatic hydrolysis were measured by using Vis-Spectrophotometer (oligomers & monomers). This is attributed to amorphous and short
(CECIL 1001, Cambridge, United Kingdom) based on DNS method to polymer chains structure of the hemicellulose which is very easily
determine the concentration of TRS (Miller, 1959). While the solid broken down by high temperatures (Gírio et al., 2010). The same trend,
fraction after SCW pretreatment was analyzed by XRD and SEM assay to Mohan et al. (2015) has obtained the highest sugar yield of 42.21% in
know the structure and morphology of crystal. Solid and liquid hy- the bamboo hydrolysis at a temperature of 180 °C and reaction time of
drolysate were taken at 6 h interval during 48 h for measurement of 25 min (SF of 3.75). It was reported that under these conditions, the
sugar and hydrogen concentration using DNS and Gas Chromatograph rate of xylose yield was highest because of the rapid decomposition
(GC-2010 Plus Shimadzu, Kyoto, Japan), respectively. The values of pH process even at low temperatures.
of solid hydrolysate were measured by pH meter (pH 5+ Eutech, Ayer Moreover, the TRS yield tends to decrease with increasing SF from
Rajah Crescent, Singapore). 3.25 to 7.67. These decreasing TRS yield from 14.71% to 2.94% may
due to further degradation of sugar into furfural and HMF and acetyl
2.2.5. Statistical analysis and calculations groups during subcritical water hydrolysis that was not measured in
One-way analysis of variance (ANOVA) was carried out to analyze this study. Zhang and Wu (2013) reported that inhibitor concentration
significance level of the various severity factor (SF) by using the increases along with the increasing temperature after subcritical CO2
Minitab 16 statistical software (Minitab Inc., ITS Surabaya, Indonesia). pretreatment of sugarcane bagasse. At the same operating condition,
Significance was assumed if the p < .05. All the running was con- HMF concentration increased from 0.01 g/L to 0.46 g/L at 140 °C to at
ducted and analyzed in triplicate. 180 °C for 100 min (SF of 3.18 and 4.36), respectively, as well as fur-
The severity factor is usually represented as a log function using fural from 0.01 to 6.19 g/L and acetic acid of 0.69 to 5.41 g/L. The same
following equation: result was obtained by Prado et al. (2014) that furfural mass increased
from 7.49 to 23.33 g/100 g raw material at 212 °C (SF of 4.77) to 259 °C
LogR0 = Log [t exp((T −Tr )/14.75)] (1) (SF of 6.16), respectively. Likewise, HMF, 4-hydroxybenzoic and va-
where t corresponds to SCW pretreatment time (min), T and Tr is the nillin increased with increasing temperature.
pretreatment and reference (100 °C) temperature (°C), respectively. The Fig. 2 also shows the changing of pH values as result of increasing
value of 14.75 is an empirical parameter of activation energy from severity factor. Increasing the sugar yield when severity factor was
pseudo-first-order kinetics (Monlau et al., 2014a). increased was accompanied by decreasing pH value. This could be ex-
TRS yield is defined by Eq. (2) plained that CO2 as the pressurizing gas dissolve in the medium re-
sulting formation weak acid HCO3- (Yang et al., 2017). This will help
mass of reducing sugar obtained (g ) the degradation of hemicellulose and cellulose to form sugar. Further
TRS yield = × 100%
mass of initial coconut husk (g ) (2) increase of severity factor resulting in decreasing sugar yield was fol-
lowed by a slight increase of pH values. This may be attributed to the
formation of a degradation product of lignin such as phenolic com-
3. Results and discussion pound (Ciftci and Saldana, 2015; Naseeruddin et al., 2013). In the
subcritical water process, the high SF decrease pH as increasing number
3.1. Pretreatment and hydrolysis in subcritical water of hydronium ion. Indeed, the maximum H+ and OH− concentrations
are reached at the highest SF of 7.67 (at 300 °C) which favoring acid/
3.1.1. The effect of severity factor of SCW hydrolysis on the total reducing base catalyzed reactions and allowing water to act as a catalyst (Prado
sugar yield et al., 2016). These results are consistent with the study by Khuwijitjaru
Fig. 2 presents the effect of severity factor of SCW hydrolysis on the et al. (2012). They reported a decreasing of pH values during subcritical
total reducing sugar (TRS) yield obtained after SCW hydrolysis applied. water hydrolysis of coconut meal from 6 to 3 with increasing tem-
As shown in Fig. 2, at low-intermediate severity factor (SF) of SCW, the perature from 100 (SF of 1.78) to 200 °C (SF of 4.72).
sugar yield recovered gradually increases with increasing SF of 2.37 to The effect of severity factor to degradation of lignocellulose com-
3.25 from 5.04% to 14.71%, respectively. Based on ANOVA results, the positions of CCH after SCW hydrolysis applied was shown in Fig. 3. The
SF significantly influenced TRS yield of CCH during SCW pretreatment degradation of hemicellulose increases gradually from 0.255 to 0.767 g
due to p < .05. In this condition, hemicellulose firstly starts breaking at low SF of 2.367 to mild SF condition of 3.54, respectively, (see
Fig. 3a). The figure shows that the rate of hemicellulose decomposition
is faster than cellulose because hemicellulose has an amorphous mo-
lecular structure whereas cellulose is crystalline. Further increase of the
SF value causes a slight decrease of the hemicellulose. The same trend
was also obtained by Gonzales et al. (2016a) on the reports of hemi-
cellulose (xylose) recovery after acid pretreatment applied on empty
palm fruit bunch, rice husk and pine tree wood. The hemicellulose re-
covery in these three types of biomass increases during the low to mild
combined severity factor and then decreases with increasing SF value.
The higher severity condition corresponds to further reaction between
xylose and a pseudo-lignin compound which lead to an increase the
acid-insoluble lignin fraction (Monlau et al., 2014a).
As shown in Fig. 3b, the degradation of cellulose increased by in-
creasing SF value. Nguyen Thi et al. (2017) also reported a decreasing
of cellulose content under higher SCW temperature. Although the de-
gradation cellulose increases with increasing temperature, it must be
highly controlled considering that at high temperatures, the rate of
glucose decomposition and other oligomers is the faster than cellulose
hydrolysis rate (Zhao et al., 2011). This will allow for further de-
Fig. 2. The TRS yield and pH behavior of liquid fraction after SCW hydrolysis applied.
gradation of sugars and oligomers into compounds that can inhibit both

270
M. Muharja et al. Bioresource Technology 249 (2018) 268–275

Fig. 3. Degradation of lignocellulose (hemicellulose (a), cellulose (b), and lignin (c)) content of solid fraction after SCW hydrolysis.

enzymatic hydrolysis and hydrogen fermentation processes.

The same trend as cellulose is shown by lignin degradation in the
SCW hydrolysis process under various SF conditions as shown in
Fig. 3c. The similar result was reported by Ciftci and Saldana (2015) in
the study of SCW effect on the hydrolysis of sweet blue lupin. The lignin
content in solid residue decreased as increasing of severity condition up
to 5 times from 22.9% to 4.6% from 160 °C to 220 °C, respectively.
Moreover, these conditions lead to the breakdown of a strong bond
between carbohydrate and lignin which is indicated by increasing
phenolic compound up to 10 times from 0.07 to 0.72 mg gallic acid/g
lupin hull. The result confirmed the previous result of the tendency of
pH change in Fig. 2. The degradation of hemicellulose, cellulose and
lignin described in Fig. 3 confirmed well with the result of optimal
sugar in Fig. 2 under mild SF condition.
Fig. 4 shows the comparison of TRS yield under different types of
pressurizing gas CO2 and N2 as a function of severity factor. Under CO2
gas, the TRS yield increased significantly from 16.25% to 20.64% at low
(SF of 2.66) to mild (SF of 3.25) condition, respectively. However, the
decrease of TRS yield to 14.99% obtained when SCW hydrolysis con-
dition was increased at the higher value of SF 4.13. The similar result
was reported by Weiqi et al. (2013) using eucalyptus. They reported Fig. 4. The comparison of TRS yield after SCW hydrolysis pressured by CO2 (a), and N2
that TRS yield increases significantly from 4.74 to 13.09 g xylose/100 g (b).
raw material under SF condition of 3.07 and 3.66, respectively, then

271
M. Muharja et al. Bioresource Technology 249 (2018) 268–275

Table 1
Literature review on various pretreatment in TRS yield.

Substrate Pre-treatment Severity Factor Reactor Type TRS yield (%) References

Eucalyptus Liquid hot water 3.66 Batch 13.79 Weiqi et al. (2013)
Coconut (Cocos nucifera L.) meal CO2-SbCWH 4.22 Batch 13.6 Khuwijitjaru et al. (2012)
Coconut Husk CO2-SbCWH 6.10 Semi-batch 11.7 Prado et al. (2014)
Corn Stover CO2-SbCWH 4.02 Batch 20.81 King et al. (2012)
Corn cob 4.02 23.38
Switch grass 3.72 8.23
Coconut Husk CO2-SbCWH 3.25 Batch 20.64 This Study
Coconut Husk N2-SbCWH 4.13 Batch 11.8 This Study

gradually decreases to 8.36 g xylose/100 g raw material at SF of 4.25. Although the clustering of low, medium and high SF in Fig. 5 cor-
This might be explained by the breaking of the polysaccharide chain relates to the clustering of operating pressure of SCW of 40, 80 and
(i.e. arabinan, xylan and glucan) into oligosaccharides in the mild se- 160 bar, respectively, it is difficult to say that pressure may give effect
verity of subcritical water. Increasing severity caused further degrada- to the sugar yield since, at this increasing pressure, the temperature was
tion of oligosaccharides to their monomers but simultaneously sugar increased accordingly. Effect of pressure has been studied and found
degradation compounds were formed (Prado et al., 2016). This reveals that it gave no significant effect to sugar yield during SCW (data not
the existence of an optimal severity factor in order to obtain the best shown). Prado et al. (2016) in the review of the pressure effect of SCW
yield of sugar during SCW. revealed similarly that in general, the pressure is not a key parameter of
In contrast to CO2, surprisingly, TRS yield on SCW hydrolysis the pretreatment condition.
pressured by N2 shows a constant increase even as it is elevated to the The decreasing of sugar after attaining its highest value occurred
most severe conditions (SF of 4.13). Table 1 shows a comparison of the because enzymatic hydrolysis has a characteristic of having high initial
sugar yield according to the literature review. The highest TRS yield rapid rate followed by slower and sometimes incomplete hydrolysis
using N2 in the present study was obtained at SF of 4.13. Öztürk et al. (Gabhane et al., 2014). A study by Ramos-gonza (2017) showed that
(2010) reported that no significant difference between the two gases for the enzyme activity decreased continually and only reached 20% of its
hydrolysis of kenaf stems by subcritical water. Although the yield by initial activity after having pre-incubation at 60 °C for 4 h.
using N2 was lower than CO2, N2 can be a promising gas for future SCW Table 2 summarizes the result of TRS yield at various SF. Mono-
hydrolysis. This is because N2 has lower price and inert in addition to meric sugar liberated from cellulose and hemicellulose by the catalytic
the constant increase of sugar yield even at increasing severity factor. action of enzyme contributed significantly to the total reducing sugar
obtained. It can be seen that the highest yield of total reducing sugar
3.1.2. Characterization of SCW solid residue was obtained at mild SF. The total TRS yield was significantly influ-
3.1.2.1. Scanning Electron Micrograph (SEM) analysis. SEM was used to enced by the SF based on ANOVA analysis (p < .05).
compare morphological changes in coconut husk before and after
pretreatment. As shown by SEM images of SCW-treated at 150 °C, 3.3. Hydrogen from fermentation process
80 bar (SF of 3.25), the structure of untreated lignocellulose is smooth,
tight and contiguous, whereas after pretreatment it showed extensive Fermentation was performed using substrate from the SCW hydro-
anomalous porosity and more rough, hollow and amorphous. These lysate and enzymatic hydrolysate. Both substrates were chosen from the
structural changes help to expose the fasciculi to enzyme access and best SCW condition which gave the highest total sugar concentration.
resulting in high enzymatic hydrolysis yield (Sangian and Widjaja, Time courses of sugar consumption and accumulative H2 yield for 48 h
2017). fermentation are shown in Fig. 6a. The concentration increased gra-
dually accompanied by the decreasing of sugar concentration. The H2
3.1.2.2. X-ray diffraction (XRD) analysis. According to diffraction yield of 0.175 mol/mol of sugar consumed was obtained. Fig. 6b shows
pattern, the value of crystalline index (CrI) decreased from 45.8% to the hydrogen yield obtained after fermentation of SCW solid fraction of
42.5% after the SCW pretreatment applied at 80 bar and 150 °C (SF of coconut husk for 48 h. It was obtained the total hydrogen yield of
3.25). This was derived from the difference between the crystalline 0.104 mol/mol of sugar consumed of an enzymatic hydrolysate. The
region at 2θ of 21.59° and 22.35° and the amorphous region at 2θ of maximum total hydrogen yield of variable 80 bar and 150 °C (SF of
18.76° and 18.14°. This phenomenon could be explained by the 3.25) H2 was 0.279 mol /mol of sugar consumed. Table 3 shows the
degradation of crystalline cellulose (Sangian and Widjaja, 2017). comparison of biohydrogen yield according to recently published re-
ports. The yield of the present study was lower than the others.
3.2. Sugar from enzymatic hydrolysis There are several factors which resulted in a low hydrogen yield.
Sivagurunathan et al. (2017) reported that the presence of a furanic and
Enzymatic hydrolysis performance is influenced by hemicellulose phenolic compound in the hydrolysate can cause a shifting of the
contents, porosity, the crystallinity of cellulose and lignin content acetate-butyrate pathway to the lactate-propionate pathway. Biohy-
(Zhang and Wu, 2013). Time course of sugar yield on solid residue was drogen production was depended on the acetate-butyrate pathway.
shown on Fig. 5 clustered according to the degree of SF, i.e. low, Exposed to high severity pretreatment, the abundant lignin monomers
medium and high SF. It could be seen from this figure that higher SF in the coconut husk (i.e., syringyl and guaiacyl units) are predictably
tended to result in lower sugar yield on the enzymatic hydrolysis of released (Monlau et al., 2014b). The syringyl and guaiacyl compound
solid residue. Since two kinds of enzymes cellulase and xylanase were subsequently undergoing further degradation into vanillin and syr-
employed in the present study, reducing sugars obtained from the en- ingaldehyde, respectively, which are the inhibitors of both enzymatic
zymatic process were derived from their polymer cellulose and hemi- and fermentation hydrolysis processes (Barakat et al., 2012). Moreover,
cellulose. From Table 2, the higher SF tended to give the lower total furfural, which is a derived compound of pentose, causes severe in-
amount of hemicellulose and cellulose. The higher the total amount of hibition by shifting pathway of acetate-butyrate to lactate-propionate
hemicellulose and cellulose in the solid residue the higher the reducing even at very small doses of 0.1 g/l (Liu et al., 2015).
sugar produced by the enzymatic hydrolysis. One of the attempt to increase the yield of hydrogen is to avoid the

272
M. Muharja et al. Bioresource Technology 249 (2018) 268–275

Fig. 5. Time course of sugar yield on solid residue at (a) low SF (0–3.25), (b) medium SF (3.25–4.13), (c) high SF (4.72–7.67).

Table 2 H2 production (Sivagurunathan et al., 2017). The detoxification

Comparison between the TRS yield of SCW and Enzymatic Hydrolysis. method, however, increases the operational cost of the process and it is
even possible to lose sugar in the hydrolysate (Monlau et al., 2014b).
Variable SF Lignocellulose mass (g) TRS Yield (%)
Therefore, a strategy to minimize the formation of inhibitor will be
P T Hemi- Cellulose Total SCW Enzymatic Total needed to develop an optimal process prior to fermentation.
cellulose E. aerogenes used in the present study is a strict anaerobe. The
presence of air leakage even in a very few amount may give serious
40 100 1.78 0.667 1.255 2.270 5.04a 6.56 11.60a
40 130 2.66 0.443 1.264 1.876 7.65b 8.60 16.25b
effect to the bacteria. Using this kind of microbes will need a very strict
40 140 2.96 0.344 1.329 2.150 10.67c 7.15 17.82c control to avoid the system from leakage whether during operation or
80 150 3.25 0.207 1.261 2.262 14.71d 5.94 20.64d during sampling. Pachapur et al. (2015) reviewed a co-culture strate-
80 160 3.54 0.156 1.186 2.165 9.52e 5.47 14.99e gies to increase biohydrogen production. Co-culture consisting of
80 180 4.13 0.239 1.135 2.162 6.38f 5.51 11.89f
Clostridium, Enterobacter, and photo-fermenting bacteria was reported
160 200 4.72 0.248 0.911 1.766 4.69g 4.41 9.10g
160 220 5.31 0.306 0.746 1.607 3.21h 4.01 7.21h to increase H2 yield. The use of mixed cultures that have more resistant
160 260 6.49 0.368 0.570 1.446 2.87i 3.70 6.57i characteristics of toxicity is promoted as an attempt to counter the
160 300 7.67 0.343 0.384 1.264 2.94i 3.51 6.45j negative effects of by-products (Sivagurunathan et al., 2017). The
a–j
anaerobic facultative bacteria, Clostridium sp., is a resistant bacteria
Means denoted with the same letter in the same column are statistically not sig-
genus to inhibitory influences (Kumar et al., 2016; Quéméneur et al.,
nificantly different at p = .05.
The bold letter denoted the highest value.
2012).
Optimizing the operation condition such as pH and temperature will
negative effects of inhibitors formed during SCW. Detoxification is a give an optimal condition for microbial growth. The presence of hy-
proposed process by addition of a neutralizing pH agent such us Ca drogen product in the reaction mixture was also reported to have in-
(OH)2 (Datar et al., 2007; Nissilä et al., 2012), activated carbon (Cheng hibition effect to the fermentation process (Bundhoo and Mohee, 2016).
et al., 2015), and by using anionic resin (Li et al., 2011). This detox- Optimization of reactor design will be required to reduce hydrogen
ification process is highly recommended for more stable and efficient accumulation in the system.

273
M. Muharja et al. Bioresource Technology 249 (2018) 268–275

Fig. 6. Time course of H2 yield and sugar concentration (a) liquid fraction and (b) solid of SCW hydrolysate.

Table 3
Literature review of biohydrogen yield.

Substrate Pre-treatment SF of Pre-treatment Reactor Type Hydrogen Yield References

Empty palm fruit bunch Dilute acid in autoclave 2.04 Batch 0.96 mol H2/mol TS Gonzales et al. (2016a)
Rice Husk Dilute acid in autoclave 1.86 Batch 1.25 mol H2/mol TS
Pine tree wood Dilute acid in autoclave 2.01 Batch 0.99 mol H2/mol TS

Agave tequila bagasse Dilute acid, Enzymatic hydrolysis 5.10 Batch 1.4 mol H2/mol hexose Arreola-Vargas et al. (2016)
De-oiled rice bran Dilute acid 4.9 Batch 0.94 mol H2/mol TS Azman et al. (2016)
Oat Straw Dilute acid, Enzymatic hydrolysis 4.28 Batch 2.39 mol H2/mol RS Arreola-Vargas et al. (2015)
Coconut Husk CO2-SbCWH, Enzymatic hydrolysis 3.25 Batch 0.279 mol H2/mol RS This Study

4. Conclusions and methane production from acid and enzymatic hydrolysates of Agave tequilana
bagasse. Int. J. Hydrogen Energy 41, 897–904. http://dx.doi.org/10.1016/j.ijhydene.
2015.11.016.
Hydrogen has been successfully prepared from coconut husk using Arreola-Vargas, J., Razo-Flores, E., Celis, L.B., Alatriste-Mondragón, F., 2015. Sequential
integration of three green processes: SCW, enzymatic hydrolysis and hydrolysis of oat straw and hydrogen production from hydrolysates: role of hydro-
lysates constituents. Int. J. Hydrogen Energy 40, 10756–10765. http://dx.doi.org/10.
fermentation. It has been revealed from the study of severity factor that 1016/j.ijhydene.2015.05.200.
the maximum sugar yield of 20.64% was obtained at mild severity Azman, N.F., Abdeshahian, P., Kadier, A., Nasser Al-Shorgani, N.K., Salih, N.K.M.,
factor under subcritical water pretreatment at 80 bar, 150 °C, 60 min. Lananan, I., Hamid, A.A., Kalil, M.S., 2016. Biohydrogen production from de-oiled
rice bran as sustainable feedstock in fermentative process. Int. J. Hydrogen Energy
The result was confirmed by the degradation of hemicellulose, cellulose 41, 145–156. http://dx.doi.org/10.1016/j.ijhydene.2015.10.018.
and lignin. The tendency of pH changing during SCW pretreatment was Barakat, A., Monlau, F., Steyer, J.P., Carrere, H., 2012. Effect of lignin-derived and furan
in opposite manner to the tendency of changing of sugar yield. compounds found in lignocellulosic hydrolysates on biomethane production.
Bioresour. Technol. 104, 90–99. http://dx.doi.org/10.1016/j.biortech.2011.10.060.
Enzymatic hydrolysis was found to contribute significantly to the total
Bundhoo, M.A.Z., Mohee, R., 2016. Inhibition of dark fermentative bio-hydrogen pro-
sugar obtained from CCH. Types of pressurizing gas determine the se- duction: a review. Int. J. Hydrogen Energy 41, 6713–6733. http://dx.doi.org/10.
verity factor tendency for sugar yield. 1016/j.ijhydene.2016.03.057.
Cheng, J., Lin, R., Song, W., Xia, A., Zhou, J., Cen, K., 2015. Enhancement of fermentative
hydrogen production from hydrolyzed water hyacinth with activated carbon detox-
Acknowledgements ification and bacteria domestication. Int. J. Hydrogen Energy 40, 2545–2551. http://
dx.doi.org/10.1016/j.ijhydene.2014.12.097.
The authors are grateful to the Higher Education Department of Ciftci, D., Saldana, M.D.A., 2015. Hydrolysis of sweet blue lupin hull using subcritical
water technology. Bioresour. Technol. 194, 75–82. http://dx.doi.org/10.1016/j.
Indonesia Government for financial support of this study and the biortech.2015.06.146.
Biochemical Engineering Lab of ITS Surabaya that gave authors wide Datar, R., Huang, J., Maness, P.C., Mohagheghi, A., Czernik, S., Chornet, E., 2007.
access in conducting many experiments. Hydrogen production from the fermentation of corn stover biomass pretreated with a
steam-explosion process. Int. J. Hydrogen Energy 32, 932–939. http://dx.doi.org/10.
1016/j.ijhydene.2006.09.027.
Appendix A. Supplementary data Datta, R., 1981. Acidogenic fermentation of lignocellulose-acid yield and conversion of
components. Biotechnol. Bioeng. 23, 2167–2170. http://dx.doi.org/10.1002/bit.
260230921.
Supplementary data associated with this article can be found, in the
Gabhane, J., Prince William, S.P.M., Gadhe, A., Rath, R., Vaidya, A.N., Wate, S., 2014.
online version, at http://dx.doi.org/10.1016/j.biortech.2017.10.024. Pretreatment of banana agricultural waste for bio-ethanol production: individual and
interactive effects of acid and alkali pretreatments with autoclaving, microwave
heating and ultrasonication. Waste Manage. 34, 498–503. http://dx.doi.org/10.
References
1016/j.wasman.2013.10.013.
Gao, M., Xu, F., Li, S., Ji, X., Chen, S., Zhang, D., 2010. Effect of SC-CO2 pretreatment in
Alinia, R., Zabihi, S., Esmaeilzadeh, F., Kalajahi, J.F., 2010. Pretreatment of wheat straw increasing rice straw biomass conversion. Biosyst. Eng. 106, 470–475. http://dx.doi.
by supercritical CO2 and its enzymatic hydrolysis for sugar production. Biosyst. Eng. org/10.1016/j.biosystemseng.2010.05.011.
107, 61–66. http://dx.doi.org/10.1016/j.biosystemseng.2010.07.002. Gírio, F.M., Fonseca, C., Carvalheiro, F., Duarte, L.C., Marques, S., Bogel-Łukasik, R.,
Arreola-Vargas, J., Flores-Larios, A., González-Álvarez, V., Corona-González, R.I., 2010. Hemicelluloses for fuel ethanol: a review. Bioresour. Technol. 101, 4775–4800.
Méndez-Acosta, H.O., 2016. Single and two-stage anaerobic digestion for hydrogen http://dx.doi.org/10.1016/j.biortech.2010.01.088.
Gonzales, R.R., Sivagurunathan, P., Kim, S.H., 2016a. Effect of severity on dilute acid

274
M. Muharja et al. Bioresource Technology 249 (2018) 268–275

pretreatment of lignocellulosic biomass and the following hydrogen fermentation. Biotechnol. 168, 2160–2169. http://dx.doi.org/10.1007/s12010-012-9925-z.
Int. J. Hydrogen Energy 41, 21678–21684. http://dx.doi.org/10.1016/j.ijhydene. Öztürk Ilker, I., Irmak, S., Hesenov, A., Erbatur, O., 2010. Hydrolysis of kenaf (Hibiscus
2016.06.198. cannabinus L.) stems by catalytical thermal treatment in subcritical water. Biomass
Gonzales, R.R., Sivagurunathan, P., Parthiban, A., Kim, S.H., 2016b. Optimization of Bioenergy 34, 1578–1585. http://dx.doi.org/10.1016/j.biombioe.2010.06.005.
substrate concentration of dilute acid hydrolyzate of lignocellulosic biomass in batch Pachapur, V.L., Sarma, S.J., Brar, S.K., Le Bihan, Y., Soccol, C.R., Buelna, G., Verma, M.,
hydrogen production. Int. Biodeterior. Biodegrad. 113, 22–27. http://dx.doi.org/10. 2015. Co-culture strategies for increased biohydrogen production. Int. J. Energy Res.
1016/j.ibiod.2016.04.016. 39, 1479–1504. http://dx.doi.org/10.1002/er.3364.
Ju, Y.H., Huynh, L.H., Kasim, N.S., Guo, T.J., Wang, J.H., Fazary, A.E., 2011. Analysis of Prado, J.M., Forster-Carneiro, T., Rostagno, M.A., Follegatti-Romero, L.A., Maugeri Filho,
soluble and insoluble fractions of alkali and subcritical water treated sugarcane ba- F., Meireles, M.A.A., 2014. Obtaining sugars from coconut husk, defatted grape seed,
gasse. Carbohydr. Polym. 83, 591–599. http://dx.doi.org/10.1016/j.carbpol.2010. and pressed palm fiber by hydrolysis with subcritical water. J. Supercrit. Fluids 89,
08.022. 89–98. http://dx.doi.org/10.1016/j.supflu.2014.02.017.
Khuwijitjaru, P., Watsanit, K., Adachi, S., 2012. Carbohydrate content and composition of Prado, J.M., Lachos-Perez, D., Forster-Carneiro, T., Rostagno, M.A., 2016. Sub- and su-
product from subcritical water treatment of coconut meal. J. Ind. Eng. Chem. 18, percritical water hydrolysis of agricultural and food industry residues for the pro-
225–229. http://dx.doi.org/10.1016/j.jiec.2011.11.010. duction of fermentable sugars: a review. Food Bioprod. Process. 98, 95–123. http://
King, J.W., Srinivas, K., Guevara, O., Lu, Y.W., Zhang, D., Wang, Y.J., 2012. Reactive high dx.doi.org/10.1016/j.fbp.2015.11.004.
pressure carbonated water pretreatment prior to enzymatic saccharification of bio- Quéméneur, M., Hamelin, J., Barakat, A., Steyer, J.P., Carrre, H., Trably, E., 2012.
mass substrates. J. Supercrit. Fluids 66, 221–231. http://dx.doi.org/10.1016/j. Inhibition of fermentative hydrogen production by lignocellulose-derived compounds
supflu.2012.02.010. in mixed cultures. Int. J. Hydrogen Energy 37, 3150–3159. http://dx.doi.org/10.
Koppram, R., Tomás-Pejó, E., Xiros, C., Olsson, L., 2014. Lignocellulosic ethanol pro- 1016/j.ijhydene.2011.11.033.
duction at high-gravity: challenges and perspectives. Trends Biotechnol. 32, 46–53. Ramos-gonza, M.A.M.R., 2017. Cellulases immobilization on chitosan-coated magnetic
http://dx.doi.org/10.1016/j.tibtech.2013.10.003. nanoparticles : application for Agave Atrovirens lignocellulosic biomass hydrolysis.
Kumar, G., Sivagurunathan, P., Chen, C.-C., Lin, C.-Y., 2016. Batch and continuous bio- Bioprocess Biosyst. Eng. 9–22. http://dx.doi.org/10.1007/s00449-016-1670-1.
genic hydrogen fermentation of acid pretreated de-oiled jatropha waste (DJW) hy- Ren, Y., Wang, J., Liu, Z., Ren, Y., Li, G., 2009. Hydrogen production from the monomeric
drolysate. RSC Adv. 6, 45482–45491. http://dx.doi.org/10.1039/C6RA05628H. sugars hydrolyzed from hemicellulose by Enterobacter aerogenes. Renewable Energy
Kumar, G., Sivagurunathan, P., Sen, B., Kim, S.H., Lin, C.Y., 2017a. Mesophilic con- 34, 2774–2779. http://dx.doi.org/10.1016/j.renene.2009.04.011.
tinuous fermentative hydrogen production from acid pretreated de-oiled jatropha Sánchez, Ó.J., Cardona, C.A., 2008. Trends in biotechnological production of fuel ethanol
waste hydrolysate using immobilized microorganisms. Bioresour. Technol. 240, from different feedstocks. Bioresour. Technol. 99, 5270–5295. http://dx.doi.org/10.
137–143. http://dx.doi.org/10.1016/j.biortech.2017.03.059. 1016/j.biortech.2007.11.013.
Kumar, G., Sivagurunathan, P., Sen, B., Mudhoo, A., Davila-Vazquez, G., Wang, G., Kim, Sangian, H., Widjaja, A., 2017. Effect of pretreatment method on structural changes of
S.-H., 2017b. Research and development perspectives of lignocellulose-based bio- coconut coir dust. Bioresources 12, 8030–8046. http://dx.doi.org/10.15376/biores.
hydrogen production. Int. Biodeterior. Biodegrad. 119, 225–238. http://dx.doi.org/ 12.4.8030-8046.
10.1016/j.ibiod.2016.10.030. Sangian, H.F., Kristian, J., Rahma, S., Dewi, H.K., Puspasari, D.A., Agnesty, S.Y.,
Li, Y.C., Wu, S.Y., Chu, C.Y., Huang, H.C., 2011. Hydrogen production from mushroom Gunawan, S., Widjaja, A., 2015a. Preparation of reducing sugar hydrolyzed from
farm waste with a two-step acid hydrolysis process. Int. J. Hydrogen Energy 36, high-lignin coconut coir dust pretreated by the recycled ionic liquid [mmim][dmp]
14245–14251. http://dx.doi.org/10.1016/j.ijhydene.2011.06.102. and combination with alkaline. Bull. Chem. React. Eng. Catal. 10, 8–22. http://dx.
Liu, Z., Zhang, C., Wang, L., He, J., Li, B., Zhang, Y., Xing, X.H., 2015. Effects of furan doi.org/10.9767/bcrec.10.1.7058.822.
derivatives on biohydrogen fermentation from wet steam-exploded cornstalk and its Sangian, H.F., Ranggina, D., Ginting, G.M., Purba, A.A., Gunawan, S., Widjaja, A., 2015b.
microbial community. Bioresour. Technol. 175, 152–159. http://dx.doi.org/10. Study of the preparation of sugar from high-lignin lignocellulose applying subcritical
1016/j.biortech.2014.10.067. water and enzymatic hydrolysis: synthesis and consumable cost
Mahmud, Z., Ferry, Y., 2005. Prospek pengolahan hasil samping buah kelapa. Perspektif evaluation|Subcritical water and enzymatic hydrolysis: synthesis and consumable
4, 55–63. cost evaluation. Sci. Study Res. Chem. Chem. Eng. Biotechnol. Food Ind. 16, 13–27.
Miller, G.L., 1959. Use of dinitrosalicylic acid reagent for determination of reducing Sivagurunathan, P., Kumar, G., Mudhoo, A., Rene, E.R., Saratale, G.D., Kobayashi, T., Xu,
sugar. Anal. Chem. 31, 426–428. http://dx.doi.org/10.1021/ac60147a030. K., Kim, S.H., Kim, D.H., 2017. Fermentative hydrogen production using lig-
Mohan, M., Banerjee, T., Goud, V.V., 2015. Hydrolysis of bamboo biomass by subcritical nocellulose biomass: an overview of pre-treatment methods, inhibitor effects and
water treatment. Bioresour. Technol. 191, 244–252. http://dx.doi.org/10.1016/j. detoxification experiences. Renewable Sustainable Energy Rev. 77, 28–42. http://dx.
biortech.2015.05.010. doi.org/10.1016/j.rser.2017.03.091.
Monlau, F., Sambusiti, C., Barakat, A., Quéméneur, M., Trably, E., Steyer, J., Carrère, H., Toor, S.S., Rosendahl, L., Rudolf, A., 2011. Hydrothermal liquefaction of biomass: a re-
2014a. Do furanic and phenolic compounds of lignocellulosic and algae biomass view of subcritical water technologies. Energy 36, 2328–2342. http://dx.doi.org/10.
hydrolyzate inhibit anaerobic mixed cultures? A comprehensive review. Biotechnol. 1016/j.energy.2011.03.013.
Adv. http://dx.doi.org/10.1016/j.biotechadv.2014.04.007. Weiqi, W., Shubin, W., Liguo, L., 2013. Combination of liquid hot water pretreatment and
Monlau, F., Sambusiti, C., Barakat, A., Quéméneur, M., Trably, E., Steyer, J., Carrère, H., wet disk milling to improve the efficiency of the enzymatic hydrolysis of eucalyptus.
2014b. Do furanic and phenolic compounds of lignocellulosic and algae biomass Bioresour. Technol. 128, 725–730. http://dx.doi.org/10.1016/j.biortech.2012.08.
hydrolyzate inhibit anaerobic mixed cultures? A comprehensive review. Biotechnol. 130.
Adv. 32, 934–951 10.1016/j.biotechadv.2014.04.007. Wen, D., Jiang, H., Zhang, K., 2009. Supercritical fluids technology for clean biofuel
Mosier, N., Wyman, C., Dale, B., Elander, R., Lee, Y.Y., Holtzapple, M., Ladisch, M., 2005. production. Prog. Nat. Sci. 19, 273–284. http://dx.doi.org/10.1016/j.pnsc.2008.09.
Features of promising technologies for pretreatment of lignocellulosic biomass. 001.
Bioresour. Technol. 96, 673–686. http://dx.doi.org/10.1016/j.biortech.2004.06.025. Yang, T., Wang, J., Li, B., Kai, X., Li, R., 2017. Effect of residence time on two-step li-
Naseeruddin, S., Srilekha Yadav, K., Sateesh, L., Manikyam, A., Desai, S., Venkateswar quefaction of rice straw in a CO2 atmosphere: differences between subcritical water
Rao, L., 2013. Selection of the best chemical pretreatment for lignocellulosic sub- and supercritical ethanol. Bioresour. Technol. 229, 143–151. http://dx.doi.org/10.
strate Prosopis juliflora. Bioresour. Technol. 136, 542–549. http://dx.doi.org/10. 1016/j.biortech.2016.12.110.
1016/j.biortech.2013.03.053. Zhang, H., Wu, S., 2013. Subcritical CO2 pretreatment of sugarcane bagasse and its en-
Nguyen Thi, B.T., Ong, L.K., Nguyen Thi, D.T., Ju, Y.H., 2017. Effect of subcritical water zymatic hydrolysis for sugar production. Bioresour. Technol. 149, 546–550. http://
pretreatment on cellulose recovery of water hyacinth (Eichhornia crassipe). J. Taiwan dx.doi.org/10.1016/j.biortech.2013.08.159.
Inst. Chem. Eng. 71, 55–61. http://dx.doi.org/10.1016/j.jtice.2016.12.028. Zhao, Y., Wang, H.T., Lu, W.J., Wang, H., 2011. Combined supercritical and subcritical
Nissilä, M.E., Li, Y.C., Wu, S.Y., Puhakka, J.A., 2012. Dark fermentative hydrogen pro- conversion of cellulose for fermentable hexose production in a flow reaction system.
duction from neutralized acid hydrolysates of conifer pulp. Appl. Biochem. Chem. Eng. J. 166, 868–872. http://dx.doi.org/10.1016/j.cej.2010.11.058.

275