Bioresource Technology 172 (2014) 241–248
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Ethanol and lignin production from Brazilian empty fruit bunch biomass Jegannathan Kenthorai Raman ⇑, Edgard Gnansounou Bioenergy and Energy Planning Research Group (BPE), Ecole Polytechnique Fédérale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland
h i g h l i g h t s Brazil Government is promoting oil palm plantation on degraded land for biofuels. Brazil EFB has 33.5% glucan, 26.8% xylan, 21.2% lignin and 2.8% ash. Optimized dilute acid treatment conditions (160 °C, 1.025% acid, 10.5 min). Predicted maximum ethanol production was 51.1 g/kg dry EFB at low enzyme loading. 224 g lignin and 3.7 l xylose rich liquid per kg EFB are available as co-products.
a r t i c l e
i n f o
Article history: Received 6 August 2014 Received in revised form 2 September 2014 Accepted 6 September 2014 Available online 16 September 2014 Keywords: Biorefinery Brazil EFB Ethanol Dilute acid pretreatment
a b s t r a c t Brazil Government is promoting palm plantations to use degraded land for biofuels. Palm production is expected to increase 35 per cent in future and there would be profuse biomass available that needs to be handled efficiently. Therefore, in this study the potential of EFB from Brazil as raw material for biorefinery was explored by compositional analysis and pretreatment conditions optimization to produce ethanol and co-products. EFB from Brazil contains significant cellulose, hemicellulose, lignin and low ash content. The optimized dilute sulfuric acid pretreatment conditions for efficient cellulose and hemicellulose separation were 160 °C temperature, 1.025% v/v acid concentration, 10.5 min and 20% solid loading. Under optimum pretreatment process conditions, low enzyme loading (10 FPU, 20 IU cellulase and glucosidase enzyme/g glucan) and 15% solid loading, 51.1 g ethanol, 344.1 g solid residue (65% lignin and 24.87 MJ/kg LHV) and 3.7 l xylose rich liquid could be produced per kg dry EFB. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Increasing energy demand, security and environmental concerns are the reasons to explore more renewable energy resources. Among the renewable energy resources, biofuels have proved to be one of the options to replace fossil fuels in transport sector and at the same time mitigate carbon emission in the environment. There was an immense development in biofuel research and its commercialization in the last decade which led to use of blended biodiesel from vegetable oil and bioethanol from food crops with conventional fossil based diesel and petroleum in developed and developing countries. However, these advancements are seen as the primary reason for food price hike and deforestation in industrialized and less industrialized countries (Braun, 2007; Thompson and Meyer, 2013). To tackle this situation, the focus of biofuel research ⇑ Corresponding author at: Bioenergy and Energy Planning Research Group, GC A3 444 (Bâtiment GC), ENAC IIC GR-GN, EPFL, Station 18, CH-1015 Lausanne, Switzerland. Tel.: +41 216936025; fax: +41 216932863. E-mail address: jegannathan.kenthorairaman@epfl.ch (J.K. Raman). http://dx.doi.org/10.1016/j.biortech.2014.09.043 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.
has shifted towards non food based resources and agricultural waste like biomass residues for ethanol production and, non edible oil crops and algae for biodiesel production. In case of bioethanol, though the technological advancement has been achieved to convert cellulose present in biomass residues to bioethanol via pretreatment, enzymatic saccharification and fermentation, the process does not seem to be economically viable for its commercialization (Carriquiry et al., 2011). However, a large amount of wastes containing xylose and lignin are generated in these processes that could be utilized to produce valuable products like furfural and lignin based products which could increase the overall economic status of the processes, and enable this technology (biorefinery) to be commercially available (European Commission, 2006). Oil palm is cultivated in tropical regions mainly for its oil used in food, household products and biodiesel. In 2012, oil palm was cultivated in 17.2 million ha across the world (FAOSTAT, 2014). In Brazil, oil palm cultivation for biodiesel production is envisaged as a promising way to use degraded land and the Government is promoting oil palm plantations under the sustainable production
242
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248
of palm oil program (UNEP, 2011; Villela et al., 2014). Brazil produced 0.62 per cent of palm oil globally in 2012 and it is expected to increase by 35 per cent in future (UNEP, 2011; FAOSTAT, 2014) and if so there would be profuse biomass available that needs to be handled efficiently to avoid air pollution due to biomass burning (Sudiyani et al., 2013; Abdullah and Sulaiman, 2013). Empty fruit bunch (EFB) is the biomass residue left over after extracting palm oil from fresh fruit bunches and it is rich in cellulose, hemicellulose and lignin. The composition of EFB varies based on the geographic conditions and the compositional analysis studies reported so far are from south-east Asia (Malaysia, Indonesia) (Sudiyani et al., 2013; Kim and Kim, 2013; Hamzah et al., 2011; Han et al., 2011; Tan et al., 2013; Jeon et al., 2014), Colombia (Piarpuzán et al., 2011) and Africa (Chiesa and Gnansounou, 2014), and there are no studies reported on EFB from Brazil. Hence, this study aims at exploring the compositional analysis and opportunity for bioethanol and co-products (lignin) production from Brazilian EFB. On the other hand, most of the studies in the past focused only on ethanol production from cellulose and hemicellulose. In this work a biorefinery approach capable of producing multiple products was examined to harness the benefits of other components which could be produced from EFB. In the case of ethanol production from cellulose and hemicelluloses, the pretreatment process conditions have to be optimized to retain the maximum amount of cellulose and hemicellulose with less inhibitors such as furfural and acetic acid. A severe pretreatment condition would generate high inhibitors, therefore, the pretreatment conditions have to be less severe to limit the inhibitors but components separation may not be efficient at that conditions. Whereas, if severe conditions are used to make the separation efficient, the inhibitors generated would be more and a specific type of microbial strain which has high tolerance to inhibitors has to be used to ferment glucose and xylose or the inhibitors have to be reduced by other processes. Whereas in biorefinery context, only cellulose fraction would be used for ethanol production and as the hemicelluloses fraction would be used for products like furfural, there is no concern of such inhibitors problems and pretreatment process conditions capable to separate the components efficiently could be achieved. Therefore in general the negative effect of pretreatment process over the subsequent processes would be less for that biorefinery process (multiproduct) compared to the ethanol production process (single product) and this could be an advantage of biorefinery approach. As this study is being conducted under the frame work of a larger project involving the synthesis of multiple products (Bioethanol, lignin and furfural) only results of bioethanol and lignin production streams using dilute acid pretreatment are presented here. Bioethanol and co-products from biochemical method involves pretreatment, enzymatic saccharification and fermentation. Pretreatment is one of the most important processes in the biorefinery approach where the strong linkages between the biomass components are disintegrated to be raw materials for individual product through the subsequent processes. Several pretreatment methods such as dilute acid treatment, alkali treatment, steam explosion, hot water treatment have been reported and each process has its own advantages and disadvantages (Alvira et al., 2010). The objective of this study is to separate the cellulose, hemicellulose and lignin fraction present in the EFB in order to convert them in specific products in the framework of a biorefinery scheme. Therefore dilute acid pretreatment was chosen owing to its ability to separate cellulose and hemicellulose fraction apart in milder reaction conditions (Chiesa and Gnansounou, 2014; Hu et al., 2010). However, the optimum conditions (Temperature, time and acid concentration) of dilute sulfuric acid pretreatment are not general for all the biomass and all goals, as it depends on the nature of the biomass and to the desirability function of the process. The conditions
have to be optimized for each particular plant residue and desirability function (Chiesa and Gnansounou, 2014). Consequently, in this research work, where the objective is to maximize both the ethanol and furfural production the pretreatment process parameters were optimized using a statistical design to determine the best operational conditions. The optimal conditions would not be the same compared to the case where only the total yield of monomeric sugars has to be maximized. Hence one of the novelties of the paper is to find out the optimal conditions of the whole process when both ethanol and furfural have to be produced and the feedstock is the oil palm EFB. Following the pretreatment, the biomass rich in cellulose and lignin were subject to enzymatic saccharification at low enzyme loading to hydrolyze the cellulose fraction to glucose, which was further fermented by Saccharomyces cerevisiae to bioethanol. Finally, the sensitivity of the optimal conditions to solid loading at pretreatment and the enzyme loading on saccharification were studied, and the composition and heating value of lignin rich solid residue from saccharification were analyzed, and the mass balance of whole process was presented. 2. Methods 2.1. Materials Dried oil palm EFB (Elaeis guineensis) was provided by Federal University of Pará, Brazil. Fresh fruit bunch (approximate tree age-20–25 years) was harvested from the palm plantation located between Thailand and Moju (about 2°450 S and 48°500 W) for palm oil production by the AGROPALMA company. EFB (moisture content 20–25%) collected after pressing from the palm oil production plant was washed, dried for 24 h at 60 °C, ground in a mill to a size range 3–4 cm and stored in a polythene bag until sending to the Ecole Polytechnique Fédérale de Lausanne (EPFL), Switzerland. At EPFL, the sample was further milled to pass through 80–20 mesh and stored in a zip lock bag until use. Sugars, ethanol, organic acid standards, Cellulase (CellulastÒ 1.5L) and b-glucosidase (powder from almonds) enzymes were procured from Sigma–Aldrich, Switzerland. Other chemicals used in this study were of analytical grade. 2.2. Compositional analysis Compositional analysis of raw, pretreated, saccharified EFB samples were carried out according to the standard procedures for biomass compositional analysis of National Renewable Energy Laboratory (NREL) (Sluiter et al., 2010; NREL, 2014). Sugars were analyzed by high performance liquid chromatography (HPLC) (Agilent Technologies, Germany) equipped with an Aminex HPX-87P column and refractive index detector (RID) with Millipore water as the mobile phase at flow rate 0.5 ml/min. The column and detector temperature were maintained at 80 and 55 °C respectively. Ethanol, furfural, hydroxymethylfurfural (HMF), acetic acid were analyzed by same HPLC equipped with an Aminex HPX-87H column and RID detector with 5 mM sulfuric acid as the mobile phase at flow rate 0.6 ml/min. The column and detector temperature were maintained at 65 °C and 55 °C respectively. Moisture, ash and insoluble lignin were analyzed by gravimetric method. Soluble lignin was analyzed using spectrophotometer method at 320 using 30 l/g cm as the absorptivity coefficient (Sluiter et al., 2011). 2.3. Dilute acid pretreatment Design expert v 9 was used to create a Box–Behnken design (Tables 2 and 3) for the dilute acid pretreatment experiments. Three parameters temperature (100–180 °C), residence time
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248
(1–20 min) and acid concentration (0.05–2% v/v) were included in the design. There were 15 sets of experiments, including 3 center points. Experiments were carried out in a 300 ml high pressure reactor (Parr, USA). Biomass and acid solution in the required ratio were taken in the reactor and the mixture was allowed to soak for 5 min followed by pretreatment process at varied temperature, residence time and acid. After allowing the mixture to react at particular temperature and residence time the reactor was cooled down to room temperature by circulating cold water into the cooling tube available in the reactor. Later the mixture was vacuum filtered to separate the liquid and solid fractions. After separating the liquid fraction, the solid fraction in the filter was washed with 800 ml water and the moist pretreated EFB, liquid fraction was stored in a refrigerator until further use. Sugars, organic acids, lignin and ash composition of the pretreated solid fraction and liquid fraction were analyzed using the protocols mentions above. 2.4. Saccharification and fermentation Cellulase and b-glucosidase enzyme activity were analyzed using filter paper assay and cellobiose assay respectively (Adney and Baker, 2008; Zhang and Hong, 2009). Required amount of moist pretreated biomass and citrate buffer (50 mmol, pH 4.8) were taken in a 50 ml conical flask with stopper cork, the mixture was sterilized in an autoclave at 121 °C for 20 min, cooled down to 50 °C in an incubator and the required volume of enzymes (cellulase – FPU/g glucan and b-glucosidase – IU/g glucan) respectively were added aseptically and the conical flasks containing total saccharification mixture volume of 20 ml were incubated in a shaker incubator at 50 °C at 150 rpm for 72 h (Selig and Weiss, 2008). After saccharification the mixture was vacuum filtered to separate the glucose rich liquid and lignin rich solid fraction. The glucose present in the liquid fraction was analyzed to report the glucose yield (% theoretical maximum) and the solid fraction was analyzed for its chemical composition. In a 50 ml conical flask with stopper, fermentation media (10 ml) containing liquid fraction (rich in glucose) from saccharification, yeast extract (1% w/v) and peptone (2% w/v) were sterilized at 121 °C for 20 min. Inoculum containing S. cerevisiae strain (5 g/l) was added to the fermentation media aseptically and the flasks were incubated in a shaker incubator at 30 °C and 130 rpm for 48 h and its ethanol content after fermentation was analyzed to report the ethanol yield (% theoretical maximum). 2.5. Solid lignin recovery and analysis The solid fraction remaining after saccharification (rich in lignin) was dried in the oven overnight at 45 °C. Its composition (cellulose, hemicellulose, lignin and ash) were measured according to protocol mentioned above and its elemental composition (C H, N, O, ash) were analyzed by combustion method to calculate its heating value according to the formula proposed by Sokhansanj (2011). 3. Results and discussion 3.1. EFB composition Results from the compositional analysis revealed that EFB is rich in glucan (33.5%) and xylan (26.8%) with low amount of arabinan (2%) and galactan (0.2%). The lignin content was found to be 20.1% (acid insoluble) and 1.1% (acid soluble) along with 12.9% extractives, 5.1% acetyl groups and 2.8% ash. The glucan and xylan, lignin constitutes the major part of Brazilian EFB similar to other regions EFB (Piarpuzán et al., 2011; Kim and Kim, 2013). Ash content is similar to EFB from Malaysian region (3%) (Han et al., 2011)
243
but low compared to Benin (5.4%) (Chiesa and Gnansounou, 2014) and Indonesia (6.69%) (Sudiyani et al., 2013). This difference in chemical composition could be due to the geographic conditions, cultivation practice, harvesting season and post harvest treatment. However, in general biomass with low ash content is preferable as ash increases the waste stream which needs to be treated or disposed. 3.2. Dilute acid pretreatment, saccharification and fermentation – impact of process conditions The objective of this pretreatment is to remove the xylan and acetyl fraction from raw biomass to the maximum, at the same time retaining and untie glucan and lignin linkage in solid fraction. Xylan or xylooligomers, if present in the pretreated solid could inhibit the cellulose activity during enzymatic hydrolysis and it is beneficial to remove them prior to saccharification (Qing et al., 2010; Liao et al., 2005). To explain the effect of the process conditions, Combined Severity factor (CSF) was used based on the below equation (Chum et al., 1990).
CSF ¼ logR0 pH ¼ log½t eðT r T 0 =xÞ pH
ð1Þ
where pH refers to the final pH of the acidic solution, Tr and t are the treatment temperature and time, T0 and x are constants (100 °C and 14.75, respectively) (Overend et al., 1987). The results (Table 1) show that the mass recovery (ratio of solid fraction mass to initial mass) after pretreatment decreases upon increase in CSF indicating that the conditions were less effective in separating the cellulose, hemicellulose and lignin components at less severity and more effective at high severity due to high solubilization of hemicellulose at elevated conditions. However, the compositional analysis of the pretreated biomass shows that at a higher severity than 1.36 most of the glucan and all the xylan were separated from the solid fraction, this shows that the biomass is susceptible to elevated conditions i.e. >1.36 CSF. Between CSF 1.26 and 1.36 the glucan recovery (81–113% initial) were high, and xylan and acetyl content were low in the pretreated biomass meaning that at this CSF range corresponding to 140–180 °C, acid concentration 1.025% v/v, 1–10.5 °C the pretreatment was effective in solubilisation or removal of xylan from EFB. Beyond this range the conditions were found to be harsh for the biomass resulting in low glucan recovery and below this range the conditions only slightly affected the components. This trend was confirmed by the composition analysis results of the liquid fraction, where an opposite trend was found (Table 2). The maximum xylan and organic acids were found in CSF range 1.26–1.36. In the case of lignin, most of the pretreatment conditions retained the initial lignin content. However, unlike other components the lignin recovery was high (155, 137% initial lignin content) at severe conditions CSF > 1.36. The high solid lignin content could be the false lignin or pseudo-lignin. At severe pretreatment conditions, part of the lignin in biomass is liquefied and reacts with carbohydrates to form acid insoluble solid products through acid catalyzed dehydration know as pseudo-lignin which contains aliphatic, unsaturated and carbonyl carbon functionalities (Li et al., 2007; Sannigrahi et al., 2011). The acetyl components in the pretreated solid biomass decreased upon increase in severity. At the CSF range 1.26–1.36 it was below 0.8% compared to 5.1% in raw biomass. Based on the compositional analysis results of the pretreated solid and liquid fraction, the region around CSF 1.26–1.36 could be of interest under the present experimental setup. However, the results of the saccharification and fermentation have to be observed to know the effect of inhibitors present in the pretreated solids. Following the pretreatment, the solid fraction containing 0.6 g glucan from all the experiments (except two) were subject to
244
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248
Table 1 Composition of pretreated biomass (solid fraction) relative to dilute acid treatments on various conditions based on the experimental design. Temp. °C -Raw100 140 140 100 180 100 100 140 140 140 140 180 140 180 180 160a a
Acid conc. (% v/v)
Time (min)
CSF
% Mass recovery
0.05 0.05 0.05 1.025 0.05 1.025 2 2 1.025 1.025 1.025 1.025 2 2 1.025 1.025
10.5 1 20 1 10.5 20 10.5 1 10.5 10.5 10.5 1 20 10.5 20 10.5
2.43 2.37 1.05 0.86 0.01 0.3 0.32 0.49 1.26 1.27 1.29 1.36 1.77 2.71 2.76 1.33
97.5 96.6 92.5 90.2 74.2 82.4 81.7 65 64 62.9 63.3 49.3 61.3 35.8 36.6 54.9
Glucan (% pretr. solid)
Xylan (% pretr. solid)
Lignin (% pretr. solid)
Glucan recovery (% initial)
Xylan recovery (% initial)
Lignin recovery (% initial)
Acetyl (% pretr. solid)
33.5 36.6 41.1 40.3 38.7 52.0 42.3 32.9 41.2 59.1 59.3 58.3 47.3 44.1 1.7 16.0 59.3
24.8 25.3 28.2 26.8 22.1 14.3 18.6 12.3 3.6 4.7 5.0 4.8 0 0 0 0 1
21.2 21.1 22.7 22.8 25.4 27.8 25.0 26.9 32.8 33.4 34.2 34.3 47.8 32.5 91.6 79.5 36.7
– 106 119 111 104 115 104 80 80 113 111 110 70 81 2 17 97
– 99.3 109.9 99.9 80.4 42.9 61.8 40.5 9.4 12.1 12.6 12.1 0 0 0 0 2.3
– 97 104 99 108 97 97 104 101 101 101 102 111 94 155 137 95
5.1 2.7 4.33 4.39 3.74 2.46 3.02 1.9 0.55 0.75 0.78 0.8 0.19 0.21 0.17 0.06 0.4
At optimum pretreatment conditions.
Table 2 Composition of pretreated biomass (liquid fraction) to dilute acid treatments on various conditions based on the experimental design.
a
Temp. °C
Acid conc. (% v/v)
Time (min)
CSF
Glucose mg/100 ml
Xylose mg/100 ml
AA mg/100 ml
HMF mg/100 ml
Furfural mg/100 ml
100 140 140 100 180 100 100 140 140 140 140 180 140 180 180 160a
0.05 0.05 0.05 1.025 0.05 1.025 2 2 1.025 1.025 1.025 1.025 2 2 1.025 1.025
10.5 1 20 1 10.5 20 10.5 1 10.5 10.5 10.5 1 20 10.5 20 10.5
2.43 2.37 1.05 0.86 0.01 0.30 0.32 0.49 1.26 1.27 1.29 1.36 1.77 2.71 2.76 1.33
7.0 12.4 10.3 9.0 21.7 9.6 17.0 66.5 95.7 82.5 96.7 596.1 106.8 403.3 873.6 554.60
9.5 28.9 62.4 228.9 1346.8 775.2 870.6 1420.2 1836.8 1592.5 1861.1 437.1 994.0 0.0 0.0 1118.4
17.2 17.6 25.7 127.2 123.5 181.2 244.9 458.2 408.0 396.3 466.6 190.6 413.0 429.7 369.6 475.7
0.5 2.8 0.9 0.4 1.1 0.6 0.9 1.7 1.5 1.5 1.9 18.7 1.9 23.3 35.8 5.4
0 1.3 2.6 21.3 46.3 27.7 37.6 160.7 156.3 145.5 153.0 268.0 250.3 275.0 359.2 480.1
At optimum pretreatment conditions.
saccharification at 20 FPU and 40 IU/g glucan loading of cellulose and b-glucosidase enzyme respectively. For the last two high CSF experiments, the pretreated solid biomass had very low glucan, high false lignin content and low mass recovery. Performing saccharification with pretreated solid equivalent to 0.6 g glucan from these two experiments would increase the saccharification solid loading more than 100%, which restricted the execution of these two experiments further. The results from other experiments (Table 3) show glucose yield (% theoretical maximum) (47–67.8%) in the CSF range (0.49–1.77) upon saccharification. This could be due to high glucan and low xylan in the pretreated biomass and, easy access of enzyme to the biomass with high surface area and weak linkage between the components owing to pretreatment at this severity conditions. The low glucose yield for CSF < 0.49 could be attributed to the less effectiveness of the pretreatment in recalcitrance of the biomass components, which restricted the enzyme access to biomass. In addition, high amount of xylan present in pretreated biomass could have inhibited cellulase enzyme activity (Qing et al., 2010). Results from fermentation (Table 3) of saccharified hydrolysate from the runs CSF range (1.26–1.36) show bioethanol yield (% theoretical maximum) and ethanol production per g/g raw sample of 48–71% and 0.49–0.63 respectively. The high ethanol yield for this CSF range could be attributed to the high glucose content and low acetyl groups in the pretreated biomass.
To find the optimal conditions, a statistical analysis of the mass recovery, glucose and ethanol yields results obtained according to experimental design were performed: a quadratic model has been fitted and an ANOVA analysis has been carried out to validate the models. The models were found to be significant with p value < 0.05. For mass recovery, glucose yield and ethanol yield models, the parameters A,B,C,AB,BC,B2,C2, A,B,C,AB,BC,B2,C2 and A,B,AB,B2,C2 were found to be significant (A = Temperature, B = Concentration, C = Time). The model equations in terms of actual factors are
Mass recovery% ¼ 109:59 þ 0:17334 Temperature 19:73 Concentration 0:85 Time 0:15 Temperature Concentration 2
þ 12:24 Concentration
ð2Þ
Glucose yield ð%theoretical maximumÞ ¼ 20:04 þ 0:091 Temperature 43:72 Concentration þ 2:01 Time þ 0:55 Temperature Concentration þ 0:093 Concentration Time 2
2
8:43 Concentration 0:094 Time
ð3Þ
245
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248 Table 3 Results of saccharification and fermentation process relative to dilute acid treatments on various conditions based on the experimental design. Temp. °C
Acid conc (% v/v)
Time (min)
CSF
Glucose yield (% theoretical max.) upon sacc.
Ethanol yield (% theoretical max.) upon ferment.
Glucose g/g dry EFB upon PT and sacc.
Ethanol g/g dry EFB upon PT, sacc. and ferment.
100 140 140 100 180 100 100 140 140 140 140 180 140 180 180 160a
0.05 0.05 0.05 1.025 0.05 1.025 2 2 1.025 1.025 1.025 1.025 2 2 1.025 1.025
10.5 1 20 1 10.5 20 10.5 1 10.5 10.5 10.5 1 20 10.5 20 10.5
2.43 2.37 1.05 0.86 0.01 0.30 0.32 0.49 1.26 1.27 1.29 1.36 1.77 2.71 2.76 1.33
5.1 3.9 6.5 6.9 23.0 10.2 13.5 53.1 47.3 47.1 47.2 67.8 59.2 NA NA 60.3
1.7 0.1 0.9 0.2 38.9 0.3 9.4 52 48.4 48.2 47.1 70.8 44.5 NA NA 65.6
0.02 0.02 0.03 0.03 0.10 0.04 0.04 0.16 0.20 0.19 0.19 0.17 0.18 NA NA 0.22
0 0 0 0 0.019 0.000 0.002 0.041 0.049 0.048 0.046 0.063 0.040 NA NA 0.072
NA – not available. Pretreatment (PT) at 10% solid loading. Saccharification (sacc.) at 3% glucan loading, 15% total solid loading, 20 FPU cellulase and 40 IU glucosidase/g glucan (72 h). Fermentation (ferment.) at 2% peptone and 1% yeast extract supplement, 5 g/l yeast (48 h). a At optimum pretreatment conditions.
Fig. 1. Response surface for mass recovery, glucan, glucose yield and ethanol yield for EFB acid pretreatments as a function of Temperature and acid concentration when the time is fixed to 10.5 min.
246
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248
Ethanol yield ð%theoretical maximumÞ
at 10 FPU (Cellulase) and 20 IU (b-glycosidase)/g glucan were 45.9% and 60.9% respectively.
¼ 29:79 0:071 Temperature 21:98 Concentration þ 3:71 Time þ 0:47 Temperature
3.4. Lignin recovery and mass balance
2
Concentration 8:69 Concentration 0:17 2
Time
ð4Þ
The response surface plots shown in Fig. 1(A–D) provides one solution (160 °C, 1.025% v/v and 10.5 min) at which the maximum ethanol that could be produced from raw EFB corresponding to the 53% mass recovery, 52.6% glucan content in pretreated solid, 63% glucose yield and 66.7% ethanol yield (Theoretical maximum). Experiment performed in these pretreatment conditions (160 °C, 1.025% v/v and 10.5 min) showed 54.9% mass recovery, 59.3% glucan content in pretreated solid, 60.3% glucose yield and 65.6% ethanol yield (Theoretical maximum) which were close to the values obtained from response surface plots. The compositional analysis of the pretreated biomass at the optimum condition showed highest glucan content (59.3%), low xylan content (1%) and 36.74% lignin (Table 1) indicating high removal of xylan and retention of glucan and lignin. 3.3. Solid and enzyme loading – sensitivity analysis Pretreatment and saccharification with high solid loading and low enzyme would be an advantage to reduce the process cost. Therefore to study the sensitivity of the pretreatment process to solid loading, experiments were conducted at optimized condition by varying the solid loading from 10% to 20% (The maximum solid that could be loaded in the reactor). The results show (Fig. 2) that there was no significant difference in the mass recovery and composition at 20% loading in comparison to 10% for the solid fraction after pretreatment. However, as expected the glucose, xylose and acetyl group composition in liquid fraction of 20% loading were almost doubled (Fig. 3) meaning that the effectiveness of the pretreatment was same at both 10 and 20% solid loading. Consequently from the economical point of view pretreatment at 20% solid loading would be preferable. To study the sensitivity of enzyme loading to saccharification, pretreated solid (at optimized conditions) was subject to saccharification at low enzyme loading. The results (Fig. 4) shows that when enzyme dosage was reduced by 50% from 20 to 10 FPU/g glucan (Cellulase) and 40 to 20 IU/g glucan (b-glycosidase) the glucose yield reduced by 15%. An economic trade-off will have to be made between the cost gained due to enzyme savings and ethanol production loss at low enzyme loading (Ramachandriya et al., 2013). Glucose and ethanol yield upon saccharification and fermentation
Fig. 2. Composition of pretreated solid fraction biomass – solid loading impact on optimized pretreatment conditions (160 °C, 1.025% v/v and 10.5 min).
From the solid fraction of the saccharification process performed at 15% solid loading and low enzyme loading (10 FPU (Cellulase) and 20 IU (b-glycosidase)/g glucan), the compositional analysis and elemental analysis were performed to know the amount of solid lignin content and its heating value. Solid biomass of 0.344 g/g raw EFB rich in lignin (65.2%) with a heating value of 24.87 MJ/kg was available as solid waste after saccharification. This lignin rich and low ash (7.3%) content solid fraction could be used in the boiler to generate the heat required for the biorefinery or it could be used as raw material for biobased resins. However, 27.2% of glucan content present in the solid fraction had not been hydrolyzed in the saccharification process due to low enzyme loading. Increasing the enzyme loading could be an option to reduce glucan loss and improve ethanol production. The mass balance of the whole process of converting raw EFB to ethanol and lignin is shown in flow diagram (Fig. 5) based on the experimental process carried out under the pretreatment, low enzyme loading (10P FPU and 20 IU) saccharification and fermentation conditions. 51.1 g of ethanol, 344.1 g of lignin rich solid biomass and 3.7 l xylose rich (2.8% w/v) liquid could be produced from 1 kg of dry EFB. Ethanol production per kg raw material of this study is similar compared to 53 g ethanol/kg EFB at 25 FPU cellulase/g substrate reported by Piarpuzán et al. (2011). However, it is low compared to the value reported in other literatures (Table 4).
Fig. 3. Composition of pretreated liquid fraction – solid loading impact on optimized pretreatment conditions (160 °C, 1.025% v/v and 10.5 min).
Fig. 4. Enzyme loading impact on saccharification (FPU – cellulase, IU – bglucosidase/g glucan).
247
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248
Fig. 5. Mass balance results of bioethanol and lignin production from EFB.
Table 4 Comparison of present study results with literature studies. Composition of EFB (glucan, xylan and lignin)%
Pretreatment method
Enzyme dose (FPU)
Sugar subject to fermentation
Ethanol yield g/kg EFB
References
37.3, 32.7, 36.8, 34.6, 39.3,
Alkali (NaOH) Alkali (NaOH) Sodium bisulphite Alkali (NaOH) Alkali (NaOH)
34 50 20 40 25
Glucose Glucose Glucose Glucose Glucose
120 146 192 144 53
Formalin Sodium bisulphite Dilute acid (H2SO4)
15 15 10
Glucose Glucose and xylose Glucose
166 171 51
Sudiyani et al. (2013) Han et al. (2011) Tan et al. (2013) Jeon et al. (2014) Piarpuzán et al. (2011) Cui et al. (2014) Cheng et al. (2014) Present study
14.6, 21.4, 19.3, 17.1, 36.5,
31.7 23.5 17.6 26.4 22.3
37, 15, 17.4 30.5, 16, 17.1 33.5, 24.8, 20.1
Whereas the pure lignin content (224 g/kg raw EFB) produced in this study is higher than the value (140 g/kg raw EFB) reported by Cui et al. (2014). The difference in the results could be due to difference in enzyme dosage, composition of EFB, pretreatment methods and the sugars subject to fermentation (only glucose or both glucose and xylose) (Table 4). 4. Conclusion Studies using statistical design show that cellulose and hemicellulose in EFB could be effectively separated using dilute sulfuric acid pretreatment at 160 °C temperature, 1.025% v/v acid concentration, and 10.5 min. 51.1 g ethanol, 344.1 g solid residue and 3.7 l xylose rich syrup could be produced per kg EFB. Thus EFB from Brazil could be a potential raw material for biorefinery to produce ethanol, lignin and xylose that could be used for transport fuel, solid fuel and furfural production or animal nutrition respectively. EFB generated in Brazil in future could be managed by biorefinery concept to produce valuable products.
Acknowledgements Authors acknowledge Prof. Luiz França (UFPA) for his support with EFB samples and VALE International Inc. and École Polytechnique Fédérale de Lausanne (EPFL) for financial support. References Abdullah, N., Sulaiman, F., 2013. The Oil Palm Wastes in Malaysia, Biomass Now – Sustainable Growth and Use. In: Matovic, M.D., (Ed.), ISBN: 978-953-51-1105-4, InTech. doi: 10.5772/55302. http://www.intechopen.com/books/biomass-nowsustainable-growth-and-use/the-oil-palm-wastes-in-malaysia. Adney, B., Baker, J., 2008. Measurement of Cellulase Activities – Laboratory Analytical Procedure (LAP), Technical Report. NREL/TP-510-42628. National Renewable Energy Laboratory ed. Golden, CO. Alvira, P., Tomás-Pejó, E., Ballesteros, M., Negro, M.J., 2010. Pretreatment technologies for an efficient bioethanol production process based on enzymatic hydrolysis: a review. Bioresour. Technol. 101, 4851–4861. Braun, J.V., 2007. The World Food Situation – New Driving Forces and Required Actions. International Food Policy Research Institute, Washington. Carriquiry, M.A., Du, X., Timilsina, G.R., 2011. Second generation biofuels: economics and policies. Energy Policy 39, 4222–4234.
248
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248
Cheng, J., Leu, S.Y., Zhu, J.Y., Jeffries, T.W., 2014. Ethanol production from nondetoxified whole slurry of sulfite-pretreated empty fruit bunches at a low cellulase loading. Bioresour. Technol. 164, 331–337. Chiesa, S., Gnansounou, E., 2014. Use of Empty Fruit Bunches from the Oil Palm for bioethanol production: a thorough comparison between dilute acid and dilute alkali pretreatment. Bioresour. Technol. 159, 355–364. Chum, H.L., Johnson, D.K., Black, S.K., Overend, R.P., 1990. Pretreatment – catalyst effects and the combined severity parameter. Appl. Biochem. Biotechnol. 24–25, 1–14. Cui, X., Zhao, X., Zeng, J., Loh, S.K., Choo, Y.M., Liu, D., 2014. Robust enzymatic hydrolysis of formiline-pretreated oil palm empty fruit bunches (EFB) for efficient conversion of cellulose to sugars and ethanol. Bioresour. Technol. 166, 584–591. European Commission, 2006. EUR 22066 – Biofuels in the European Union – A Vision for 2030 and Beyond. European Commission. Luxembourg. . FAOSTAT, 2014. Food and Agriculture Organization of the United Nations. . Hamzah, F., Idris, A., Shuan, T.K., 2011. Preliminary study on enzymatic hydrolysis of treated oil palm (Elaeis) empty fruit bunches fibre (EFB) by using combination of cellulase and b 1–4 glucosidase. Biomass Bioenergy 35, 1055–1059. Han, M., Kim, Y., Kim, S.W., Choi, G.W., 2011. High efficiency bioethanol production from OPEFB using pilot pretreatment reactor. J. Chem. Technol. Biotechnol. 86, 1527–1534. Hu, F., Jung, S., Ragauskas, A., 2010. Impact of pseudolignin versus dilute acidpretreated lignin on enzymatic hydrolysis of cellulose. ACS Sustain. Chem. Eng. 1, 62–65. Jeon, H., Kang, K.E., Jeong, J.S., Gong, G., Choi, J.W., Abimanyu, H., 2014. Production of anhydrous ethanol using oil palm empty fruit bunch in a pilot plant. Biomass Bioenergy 67, 99–107. Kim, S., Kim, C.H., 2013. Bioethanol production using the sequential acid/alkalipretreated empty palm fruit bunch. Renewable Energy 54, 150–155. Li, J., Henriksson, G., Gellerstedt, G., 2007. Lignin depolymerisation/ repolymerization and its critical role for delignification of aspen wood by steam explosion. Bioresour. Technol. 98, 3061–3068. Liao, W., Wen, Z., Hurley, S., Liu, Y., Liu, C., Chen, S., 2005. Effects of hemicellulose and lignin on enzymatic hydrolysis of cellulose from dairy manure. Appl. Biochem. Biotechnol. 121, 1017–1030. NREL, 2014. Standard Procedures for Biomass Compositional Analysis. . Overend, R.P., Chornet, E., Gascoigne, J.A., 1987. Fractionation of lignocellulosics by steam-aqueous pretreatments [and discussion]. Philos. Trans. R. Soc. London, Ser. A 321, 523–536.
Piarpuzán, D., Quintero, J.A., Cardona, C.A., 2011. Empty fruit bunches from oil palm as a potential raw material for fuel ethanol. Biomass Bioenergy 35, 1130–1137. Qing, Q., Yang, B., Wyman, C.E., 2010. Xylooligomers are strong inhibitors of cellulose hydrolysis by enzyme. Bioresour. Technol. 101, 9624–9630. Ramachandriya, K.D., Wilkins, M., Atiyeh, H.K., Dunford, N.T., Hiziroglu, S., 2013. Effect of high dry solids loading on enzymatic hydrolysis of acid bisulfite pretreated Eastern redcedar. Bioresour. Technol. 147, 168–176. Sannigrahi, P., Kim, D.H., Jung, S., Ragauskas, A., 2011. Pseudo-lignin and pretreatment chemistry. Energy Environ. Sci. 4, 1306–1310. Selig, M., Weiss, N., Ji, Y., 2008. Enzymatic Saccharification of Lignocellulosic Biomass – Laboratory Analytical Procedure (LAP), Technical Report National. Renewable Energy Laboratory NREL/TP-510-42629, ed. Golden, CO. Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., Crocker, D., 2011. Determination of Structural Carbohydrates and Lignin in Biomass. Laboratory Analytical Procedure (LAP). Technical Report NREL/TP-510-42618. National Renewable Energy Laboratory ed. Golden, CO. Sluiter, J.B., Ruiz, R.O., Scarlata, C.J., Sluiter, A.D., Templeton, D.W., 2010. Compositional analysis of lignocellulosic feedstocks. 1. Review and description of methods. J. Agric. Food Chem. 58, 9043–9053. Sokhansanj, S., 2011. The Effect of Moisture on Heating Values. Biomass Energy Data Book – 2011. Center for Transportation Analysis. Oak Ridge National Laboratory. . Sudiyani, Y., Styarini, D., Triwahyuni, E., Kiky Sudiyarmanto, C., Sembiring Aristiawan, Y., Abimanyu, H., Han, M.H., 2013. Utilization of biomass waste empty fruit bunch fiber of palm oil for bioethanol production using pilot-scale unit. Energy Procedia 32, 31–38. Tan, L., Yu, Y., Li, X., Zhao, J., Qu, Y., Choo, Y.M., Loh, S.K., 2013. Pretreatment of empty fruit bunch from oil palm for fuel ethanol production and proposed biorefinery process. Bioresour. Technol. 135, 275–282. Thompson, W., Meyer, S., 2013. Second generation biofuels and food crops: coproducts or competitors? Global Food Secur. 2, 89–96. UNEP, 2011. Oil palm plantations: threats and opportunities for tropical ecosystems. UNEP Global Environmental Alert Service (GEAS). . Villela, A.A., Jaccoud, D.B., Rosa, L.P., Freitas, M.V., 2014. Status and prospects of oil palm in the Brazilian Amazon. Biomass Bioenergy 67, 270–278. Zhang, Y.H., Hong, J., Ye, X., 2009. Cellulase assays. In: Mielenz, J.R. (Ed.), Biofuels: Methods and Protocols Methods in Molecular Biology, vol. 581. Humana Press, NJ, pp. 213–231.
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Ethanol and lignin production from Brazilian empty fruit bunch biomass Jegannathan Kenthorai Raman ⇑, Edgard Gnansounou Bioenergy and Energy Planning Research Group (BPE), Ecole Polytechnique Fédérale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland
h i g h l i g h t s Brazil Government is promoting oil palm plantation on degraded land for biofuels. Brazil EFB has 33.5% glucan, 26.8% xylan, 21.2% lignin and 2.8% ash. Optimized dilute acid treatment conditions (160 °C, 1.025% acid, 10.5 min). Predicted maximum ethanol production was 51.1 g/kg dry EFB at low enzyme loading. 224 g lignin and 3.7 l xylose rich liquid per kg EFB are available as co-products.
a r t i c l e
i n f o
Article history: Received 6 August 2014 Received in revised form 2 September 2014 Accepted 6 September 2014 Available online 16 September 2014 Keywords: Biorefinery Brazil EFB Ethanol Dilute acid pretreatment
a b s t r a c t Brazil Government is promoting palm plantations to use degraded land for biofuels. Palm production is expected to increase 35 per cent in future and there would be profuse biomass available that needs to be handled efficiently. Therefore, in this study the potential of EFB from Brazil as raw material for biorefinery was explored by compositional analysis and pretreatment conditions optimization to produce ethanol and co-products. EFB from Brazil contains significant cellulose, hemicellulose, lignin and low ash content. The optimized dilute sulfuric acid pretreatment conditions for efficient cellulose and hemicellulose separation were 160 °C temperature, 1.025% v/v acid concentration, 10.5 min and 20% solid loading. Under optimum pretreatment process conditions, low enzyme loading (10 FPU, 20 IU cellulase and glucosidase enzyme/g glucan) and 15% solid loading, 51.1 g ethanol, 344.1 g solid residue (65% lignin and 24.87 MJ/kg LHV) and 3.7 l xylose rich liquid could be produced per kg dry EFB. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Increasing energy demand, security and environmental concerns are the reasons to explore more renewable energy resources. Among the renewable energy resources, biofuels have proved to be one of the options to replace fossil fuels in transport sector and at the same time mitigate carbon emission in the environment. There was an immense development in biofuel research and its commercialization in the last decade which led to use of blended biodiesel from vegetable oil and bioethanol from food crops with conventional fossil based diesel and petroleum in developed and developing countries. However, these advancements are seen as the primary reason for food price hike and deforestation in industrialized and less industrialized countries (Braun, 2007; Thompson and Meyer, 2013). To tackle this situation, the focus of biofuel research ⇑ Corresponding author at: Bioenergy and Energy Planning Research Group, GC A3 444 (Bâtiment GC), ENAC IIC GR-GN, EPFL, Station 18, CH-1015 Lausanne, Switzerland. Tel.: +41 216936025; fax: +41 216932863. E-mail address: jegannathan.kenthorairaman@epfl.ch (J.K. Raman). http://dx.doi.org/10.1016/j.biortech.2014.09.043 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.
has shifted towards non food based resources and agricultural waste like biomass residues for ethanol production and, non edible oil crops and algae for biodiesel production. In case of bioethanol, though the technological advancement has been achieved to convert cellulose present in biomass residues to bioethanol via pretreatment, enzymatic saccharification and fermentation, the process does not seem to be economically viable for its commercialization (Carriquiry et al., 2011). However, a large amount of wastes containing xylose and lignin are generated in these processes that could be utilized to produce valuable products like furfural and lignin based products which could increase the overall economic status of the processes, and enable this technology (biorefinery) to be commercially available (European Commission, 2006). Oil palm is cultivated in tropical regions mainly for its oil used in food, household products and biodiesel. In 2012, oil palm was cultivated in 17.2 million ha across the world (FAOSTAT, 2014). In Brazil, oil palm cultivation for biodiesel production is envisaged as a promising way to use degraded land and the Government is promoting oil palm plantations under the sustainable production
242
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248
of palm oil program (UNEP, 2011; Villela et al., 2014). Brazil produced 0.62 per cent of palm oil globally in 2012 and it is expected to increase by 35 per cent in future (UNEP, 2011; FAOSTAT, 2014) and if so there would be profuse biomass available that needs to be handled efficiently to avoid air pollution due to biomass burning (Sudiyani et al., 2013; Abdullah and Sulaiman, 2013). Empty fruit bunch (EFB) is the biomass residue left over after extracting palm oil from fresh fruit bunches and it is rich in cellulose, hemicellulose and lignin. The composition of EFB varies based on the geographic conditions and the compositional analysis studies reported so far are from south-east Asia (Malaysia, Indonesia) (Sudiyani et al., 2013; Kim and Kim, 2013; Hamzah et al., 2011; Han et al., 2011; Tan et al., 2013; Jeon et al., 2014), Colombia (Piarpuzán et al., 2011) and Africa (Chiesa and Gnansounou, 2014), and there are no studies reported on EFB from Brazil. Hence, this study aims at exploring the compositional analysis and opportunity for bioethanol and co-products (lignin) production from Brazilian EFB. On the other hand, most of the studies in the past focused only on ethanol production from cellulose and hemicellulose. In this work a biorefinery approach capable of producing multiple products was examined to harness the benefits of other components which could be produced from EFB. In the case of ethanol production from cellulose and hemicelluloses, the pretreatment process conditions have to be optimized to retain the maximum amount of cellulose and hemicellulose with less inhibitors such as furfural and acetic acid. A severe pretreatment condition would generate high inhibitors, therefore, the pretreatment conditions have to be less severe to limit the inhibitors but components separation may not be efficient at that conditions. Whereas, if severe conditions are used to make the separation efficient, the inhibitors generated would be more and a specific type of microbial strain which has high tolerance to inhibitors has to be used to ferment glucose and xylose or the inhibitors have to be reduced by other processes. Whereas in biorefinery context, only cellulose fraction would be used for ethanol production and as the hemicelluloses fraction would be used for products like furfural, there is no concern of such inhibitors problems and pretreatment process conditions capable to separate the components efficiently could be achieved. Therefore in general the negative effect of pretreatment process over the subsequent processes would be less for that biorefinery process (multiproduct) compared to the ethanol production process (single product) and this could be an advantage of biorefinery approach. As this study is being conducted under the frame work of a larger project involving the synthesis of multiple products (Bioethanol, lignin and furfural) only results of bioethanol and lignin production streams using dilute acid pretreatment are presented here. Bioethanol and co-products from biochemical method involves pretreatment, enzymatic saccharification and fermentation. Pretreatment is one of the most important processes in the biorefinery approach where the strong linkages between the biomass components are disintegrated to be raw materials for individual product through the subsequent processes. Several pretreatment methods such as dilute acid treatment, alkali treatment, steam explosion, hot water treatment have been reported and each process has its own advantages and disadvantages (Alvira et al., 2010). The objective of this study is to separate the cellulose, hemicellulose and lignin fraction present in the EFB in order to convert them in specific products in the framework of a biorefinery scheme. Therefore dilute acid pretreatment was chosen owing to its ability to separate cellulose and hemicellulose fraction apart in milder reaction conditions (Chiesa and Gnansounou, 2014; Hu et al., 2010). However, the optimum conditions (Temperature, time and acid concentration) of dilute sulfuric acid pretreatment are not general for all the biomass and all goals, as it depends on the nature of the biomass and to the desirability function of the process. The conditions
have to be optimized for each particular plant residue and desirability function (Chiesa and Gnansounou, 2014). Consequently, in this research work, where the objective is to maximize both the ethanol and furfural production the pretreatment process parameters were optimized using a statistical design to determine the best operational conditions. The optimal conditions would not be the same compared to the case where only the total yield of monomeric sugars has to be maximized. Hence one of the novelties of the paper is to find out the optimal conditions of the whole process when both ethanol and furfural have to be produced and the feedstock is the oil palm EFB. Following the pretreatment, the biomass rich in cellulose and lignin were subject to enzymatic saccharification at low enzyme loading to hydrolyze the cellulose fraction to glucose, which was further fermented by Saccharomyces cerevisiae to bioethanol. Finally, the sensitivity of the optimal conditions to solid loading at pretreatment and the enzyme loading on saccharification were studied, and the composition and heating value of lignin rich solid residue from saccharification were analyzed, and the mass balance of whole process was presented. 2. Methods 2.1. Materials Dried oil palm EFB (Elaeis guineensis) was provided by Federal University of Pará, Brazil. Fresh fruit bunch (approximate tree age-20–25 years) was harvested from the palm plantation located between Thailand and Moju (about 2°450 S and 48°500 W) for palm oil production by the AGROPALMA company. EFB (moisture content 20–25%) collected after pressing from the palm oil production plant was washed, dried for 24 h at 60 °C, ground in a mill to a size range 3–4 cm and stored in a polythene bag until sending to the Ecole Polytechnique Fédérale de Lausanne (EPFL), Switzerland. At EPFL, the sample was further milled to pass through 80–20 mesh and stored in a zip lock bag until use. Sugars, ethanol, organic acid standards, Cellulase (CellulastÒ 1.5L) and b-glucosidase (powder from almonds) enzymes were procured from Sigma–Aldrich, Switzerland. Other chemicals used in this study were of analytical grade. 2.2. Compositional analysis Compositional analysis of raw, pretreated, saccharified EFB samples were carried out according to the standard procedures for biomass compositional analysis of National Renewable Energy Laboratory (NREL) (Sluiter et al., 2010; NREL, 2014). Sugars were analyzed by high performance liquid chromatography (HPLC) (Agilent Technologies, Germany) equipped with an Aminex HPX-87P column and refractive index detector (RID) with Millipore water as the mobile phase at flow rate 0.5 ml/min. The column and detector temperature were maintained at 80 and 55 °C respectively. Ethanol, furfural, hydroxymethylfurfural (HMF), acetic acid were analyzed by same HPLC equipped with an Aminex HPX-87H column and RID detector with 5 mM sulfuric acid as the mobile phase at flow rate 0.6 ml/min. The column and detector temperature were maintained at 65 °C and 55 °C respectively. Moisture, ash and insoluble lignin were analyzed by gravimetric method. Soluble lignin was analyzed using spectrophotometer method at 320 using 30 l/g cm as the absorptivity coefficient (Sluiter et al., 2011). 2.3. Dilute acid pretreatment Design expert v 9 was used to create a Box–Behnken design (Tables 2 and 3) for the dilute acid pretreatment experiments. Three parameters temperature (100–180 °C), residence time
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248
(1–20 min) and acid concentration (0.05–2% v/v) were included in the design. There were 15 sets of experiments, including 3 center points. Experiments were carried out in a 300 ml high pressure reactor (Parr, USA). Biomass and acid solution in the required ratio were taken in the reactor and the mixture was allowed to soak for 5 min followed by pretreatment process at varied temperature, residence time and acid. After allowing the mixture to react at particular temperature and residence time the reactor was cooled down to room temperature by circulating cold water into the cooling tube available in the reactor. Later the mixture was vacuum filtered to separate the liquid and solid fractions. After separating the liquid fraction, the solid fraction in the filter was washed with 800 ml water and the moist pretreated EFB, liquid fraction was stored in a refrigerator until further use. Sugars, organic acids, lignin and ash composition of the pretreated solid fraction and liquid fraction were analyzed using the protocols mentions above. 2.4. Saccharification and fermentation Cellulase and b-glucosidase enzyme activity were analyzed using filter paper assay and cellobiose assay respectively (Adney and Baker, 2008; Zhang and Hong, 2009). Required amount of moist pretreated biomass and citrate buffer (50 mmol, pH 4.8) were taken in a 50 ml conical flask with stopper cork, the mixture was sterilized in an autoclave at 121 °C for 20 min, cooled down to 50 °C in an incubator and the required volume of enzymes (cellulase – FPU/g glucan and b-glucosidase – IU/g glucan) respectively were added aseptically and the conical flasks containing total saccharification mixture volume of 20 ml were incubated in a shaker incubator at 50 °C at 150 rpm for 72 h (Selig and Weiss, 2008). After saccharification the mixture was vacuum filtered to separate the glucose rich liquid and lignin rich solid fraction. The glucose present in the liquid fraction was analyzed to report the glucose yield (% theoretical maximum) and the solid fraction was analyzed for its chemical composition. In a 50 ml conical flask with stopper, fermentation media (10 ml) containing liquid fraction (rich in glucose) from saccharification, yeast extract (1% w/v) and peptone (2% w/v) were sterilized at 121 °C for 20 min. Inoculum containing S. cerevisiae strain (5 g/l) was added to the fermentation media aseptically and the flasks were incubated in a shaker incubator at 30 °C and 130 rpm for 48 h and its ethanol content after fermentation was analyzed to report the ethanol yield (% theoretical maximum). 2.5. Solid lignin recovery and analysis The solid fraction remaining after saccharification (rich in lignin) was dried in the oven overnight at 45 °C. Its composition (cellulose, hemicellulose, lignin and ash) were measured according to protocol mentioned above and its elemental composition (C H, N, O, ash) were analyzed by combustion method to calculate its heating value according to the formula proposed by Sokhansanj (2011). 3. Results and discussion 3.1. EFB composition Results from the compositional analysis revealed that EFB is rich in glucan (33.5%) and xylan (26.8%) with low amount of arabinan (2%) and galactan (0.2%). The lignin content was found to be 20.1% (acid insoluble) and 1.1% (acid soluble) along with 12.9% extractives, 5.1% acetyl groups and 2.8% ash. The glucan and xylan, lignin constitutes the major part of Brazilian EFB similar to other regions EFB (Piarpuzán et al., 2011; Kim and Kim, 2013). Ash content is similar to EFB from Malaysian region (3%) (Han et al., 2011)
243
but low compared to Benin (5.4%) (Chiesa and Gnansounou, 2014) and Indonesia (6.69%) (Sudiyani et al., 2013). This difference in chemical composition could be due to the geographic conditions, cultivation practice, harvesting season and post harvest treatment. However, in general biomass with low ash content is preferable as ash increases the waste stream which needs to be treated or disposed. 3.2. Dilute acid pretreatment, saccharification and fermentation – impact of process conditions The objective of this pretreatment is to remove the xylan and acetyl fraction from raw biomass to the maximum, at the same time retaining and untie glucan and lignin linkage in solid fraction. Xylan or xylooligomers, if present in the pretreated solid could inhibit the cellulose activity during enzymatic hydrolysis and it is beneficial to remove them prior to saccharification (Qing et al., 2010; Liao et al., 2005). To explain the effect of the process conditions, Combined Severity factor (CSF) was used based on the below equation (Chum et al., 1990).
CSF ¼ logR0 pH ¼ log½t eðT r T 0 =xÞ pH
ð1Þ
where pH refers to the final pH of the acidic solution, Tr and t are the treatment temperature and time, T0 and x are constants (100 °C and 14.75, respectively) (Overend et al., 1987). The results (Table 1) show that the mass recovery (ratio of solid fraction mass to initial mass) after pretreatment decreases upon increase in CSF indicating that the conditions were less effective in separating the cellulose, hemicellulose and lignin components at less severity and more effective at high severity due to high solubilization of hemicellulose at elevated conditions. However, the compositional analysis of the pretreated biomass shows that at a higher severity than 1.36 most of the glucan and all the xylan were separated from the solid fraction, this shows that the biomass is susceptible to elevated conditions i.e. >1.36 CSF. Between CSF 1.26 and 1.36 the glucan recovery (81–113% initial) were high, and xylan and acetyl content were low in the pretreated biomass meaning that at this CSF range corresponding to 140–180 °C, acid concentration 1.025% v/v, 1–10.5 °C the pretreatment was effective in solubilisation or removal of xylan from EFB. Beyond this range the conditions were found to be harsh for the biomass resulting in low glucan recovery and below this range the conditions only slightly affected the components. This trend was confirmed by the composition analysis results of the liquid fraction, where an opposite trend was found (Table 2). The maximum xylan and organic acids were found in CSF range 1.26–1.36. In the case of lignin, most of the pretreatment conditions retained the initial lignin content. However, unlike other components the lignin recovery was high (155, 137% initial lignin content) at severe conditions CSF > 1.36. The high solid lignin content could be the false lignin or pseudo-lignin. At severe pretreatment conditions, part of the lignin in biomass is liquefied and reacts with carbohydrates to form acid insoluble solid products through acid catalyzed dehydration know as pseudo-lignin which contains aliphatic, unsaturated and carbonyl carbon functionalities (Li et al., 2007; Sannigrahi et al., 2011). The acetyl components in the pretreated solid biomass decreased upon increase in severity. At the CSF range 1.26–1.36 it was below 0.8% compared to 5.1% in raw biomass. Based on the compositional analysis results of the pretreated solid and liquid fraction, the region around CSF 1.26–1.36 could be of interest under the present experimental setup. However, the results of the saccharification and fermentation have to be observed to know the effect of inhibitors present in the pretreated solids. Following the pretreatment, the solid fraction containing 0.6 g glucan from all the experiments (except two) were subject to
244
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248
Table 1 Composition of pretreated biomass (solid fraction) relative to dilute acid treatments on various conditions based on the experimental design. Temp. °C -Raw100 140 140 100 180 100 100 140 140 140 140 180 140 180 180 160a a
Acid conc. (% v/v)
Time (min)
CSF
% Mass recovery
0.05 0.05 0.05 1.025 0.05 1.025 2 2 1.025 1.025 1.025 1.025 2 2 1.025 1.025
10.5 1 20 1 10.5 20 10.5 1 10.5 10.5 10.5 1 20 10.5 20 10.5
2.43 2.37 1.05 0.86 0.01 0.3 0.32 0.49 1.26 1.27 1.29 1.36 1.77 2.71 2.76 1.33
97.5 96.6 92.5 90.2 74.2 82.4 81.7 65 64 62.9 63.3 49.3 61.3 35.8 36.6 54.9
Glucan (% pretr. solid)
Xylan (% pretr. solid)
Lignin (% pretr. solid)
Glucan recovery (% initial)
Xylan recovery (% initial)
Lignin recovery (% initial)
Acetyl (% pretr. solid)
33.5 36.6 41.1 40.3 38.7 52.0 42.3 32.9 41.2 59.1 59.3 58.3 47.3 44.1 1.7 16.0 59.3
24.8 25.3 28.2 26.8 22.1 14.3 18.6 12.3 3.6 4.7 5.0 4.8 0 0 0 0 1
21.2 21.1 22.7 22.8 25.4 27.8 25.0 26.9 32.8 33.4 34.2 34.3 47.8 32.5 91.6 79.5 36.7
– 106 119 111 104 115 104 80 80 113 111 110 70 81 2 17 97
– 99.3 109.9 99.9 80.4 42.9 61.8 40.5 9.4 12.1 12.6 12.1 0 0 0 0 2.3
– 97 104 99 108 97 97 104 101 101 101 102 111 94 155 137 95
5.1 2.7 4.33 4.39 3.74 2.46 3.02 1.9 0.55 0.75 0.78 0.8 0.19 0.21 0.17 0.06 0.4
At optimum pretreatment conditions.
Table 2 Composition of pretreated biomass (liquid fraction) to dilute acid treatments on various conditions based on the experimental design.
a
Temp. °C
Acid conc. (% v/v)
Time (min)
CSF
Glucose mg/100 ml
Xylose mg/100 ml
AA mg/100 ml
HMF mg/100 ml
Furfural mg/100 ml
100 140 140 100 180 100 100 140 140 140 140 180 140 180 180 160a
0.05 0.05 0.05 1.025 0.05 1.025 2 2 1.025 1.025 1.025 1.025 2 2 1.025 1.025
10.5 1 20 1 10.5 20 10.5 1 10.5 10.5 10.5 1 20 10.5 20 10.5
2.43 2.37 1.05 0.86 0.01 0.30 0.32 0.49 1.26 1.27 1.29 1.36 1.77 2.71 2.76 1.33
7.0 12.4 10.3 9.0 21.7 9.6 17.0 66.5 95.7 82.5 96.7 596.1 106.8 403.3 873.6 554.60
9.5 28.9 62.4 228.9 1346.8 775.2 870.6 1420.2 1836.8 1592.5 1861.1 437.1 994.0 0.0 0.0 1118.4
17.2 17.6 25.7 127.2 123.5 181.2 244.9 458.2 408.0 396.3 466.6 190.6 413.0 429.7 369.6 475.7
0.5 2.8 0.9 0.4 1.1 0.6 0.9 1.7 1.5 1.5 1.9 18.7 1.9 23.3 35.8 5.4
0 1.3 2.6 21.3 46.3 27.7 37.6 160.7 156.3 145.5 153.0 268.0 250.3 275.0 359.2 480.1
At optimum pretreatment conditions.
saccharification at 20 FPU and 40 IU/g glucan loading of cellulose and b-glucosidase enzyme respectively. For the last two high CSF experiments, the pretreated solid biomass had very low glucan, high false lignin content and low mass recovery. Performing saccharification with pretreated solid equivalent to 0.6 g glucan from these two experiments would increase the saccharification solid loading more than 100%, which restricted the execution of these two experiments further. The results from other experiments (Table 3) show glucose yield (% theoretical maximum) (47–67.8%) in the CSF range (0.49–1.77) upon saccharification. This could be due to high glucan and low xylan in the pretreated biomass and, easy access of enzyme to the biomass with high surface area and weak linkage between the components owing to pretreatment at this severity conditions. The low glucose yield for CSF < 0.49 could be attributed to the less effectiveness of the pretreatment in recalcitrance of the biomass components, which restricted the enzyme access to biomass. In addition, high amount of xylan present in pretreated biomass could have inhibited cellulase enzyme activity (Qing et al., 2010). Results from fermentation (Table 3) of saccharified hydrolysate from the runs CSF range (1.26–1.36) show bioethanol yield (% theoretical maximum) and ethanol production per g/g raw sample of 48–71% and 0.49–0.63 respectively. The high ethanol yield for this CSF range could be attributed to the high glucose content and low acetyl groups in the pretreated biomass.
To find the optimal conditions, a statistical analysis of the mass recovery, glucose and ethanol yields results obtained according to experimental design were performed: a quadratic model has been fitted and an ANOVA analysis has been carried out to validate the models. The models were found to be significant with p value < 0.05. For mass recovery, glucose yield and ethanol yield models, the parameters A,B,C,AB,BC,B2,C2, A,B,C,AB,BC,B2,C2 and A,B,AB,B2,C2 were found to be significant (A = Temperature, B = Concentration, C = Time). The model equations in terms of actual factors are
Mass recovery% ¼ 109:59 þ 0:17334 Temperature 19:73 Concentration 0:85 Time 0:15 Temperature Concentration 2
þ 12:24 Concentration
ð2Þ
Glucose yield ð%theoretical maximumÞ ¼ 20:04 þ 0:091 Temperature 43:72 Concentration þ 2:01 Time þ 0:55 Temperature Concentration þ 0:093 Concentration Time 2
2
8:43 Concentration 0:094 Time
ð3Þ
245
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248 Table 3 Results of saccharification and fermentation process relative to dilute acid treatments on various conditions based on the experimental design. Temp. °C
Acid conc (% v/v)
Time (min)
CSF
Glucose yield (% theoretical max.) upon sacc.
Ethanol yield (% theoretical max.) upon ferment.
Glucose g/g dry EFB upon PT and sacc.
Ethanol g/g dry EFB upon PT, sacc. and ferment.
100 140 140 100 180 100 100 140 140 140 140 180 140 180 180 160a
0.05 0.05 0.05 1.025 0.05 1.025 2 2 1.025 1.025 1.025 1.025 2 2 1.025 1.025
10.5 1 20 1 10.5 20 10.5 1 10.5 10.5 10.5 1 20 10.5 20 10.5
2.43 2.37 1.05 0.86 0.01 0.30 0.32 0.49 1.26 1.27 1.29 1.36 1.77 2.71 2.76 1.33
5.1 3.9 6.5 6.9 23.0 10.2 13.5 53.1 47.3 47.1 47.2 67.8 59.2 NA NA 60.3
1.7 0.1 0.9 0.2 38.9 0.3 9.4 52 48.4 48.2 47.1 70.8 44.5 NA NA 65.6
0.02 0.02 0.03 0.03 0.10 0.04 0.04 0.16 0.20 0.19 0.19 0.17 0.18 NA NA 0.22
0 0 0 0 0.019 0.000 0.002 0.041 0.049 0.048 0.046 0.063 0.040 NA NA 0.072
NA – not available. Pretreatment (PT) at 10% solid loading. Saccharification (sacc.) at 3% glucan loading, 15% total solid loading, 20 FPU cellulase and 40 IU glucosidase/g glucan (72 h). Fermentation (ferment.) at 2% peptone and 1% yeast extract supplement, 5 g/l yeast (48 h). a At optimum pretreatment conditions.
Fig. 1. Response surface for mass recovery, glucan, glucose yield and ethanol yield for EFB acid pretreatments as a function of Temperature and acid concentration when the time is fixed to 10.5 min.
246
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248
Ethanol yield ð%theoretical maximumÞ
at 10 FPU (Cellulase) and 20 IU (b-glycosidase)/g glucan were 45.9% and 60.9% respectively.
¼ 29:79 0:071 Temperature 21:98 Concentration þ 3:71 Time þ 0:47 Temperature
3.4. Lignin recovery and mass balance
2
Concentration 8:69 Concentration 0:17 2
Time
ð4Þ
The response surface plots shown in Fig. 1(A–D) provides one solution (160 °C, 1.025% v/v and 10.5 min) at which the maximum ethanol that could be produced from raw EFB corresponding to the 53% mass recovery, 52.6% glucan content in pretreated solid, 63% glucose yield and 66.7% ethanol yield (Theoretical maximum). Experiment performed in these pretreatment conditions (160 °C, 1.025% v/v and 10.5 min) showed 54.9% mass recovery, 59.3% glucan content in pretreated solid, 60.3% glucose yield and 65.6% ethanol yield (Theoretical maximum) which were close to the values obtained from response surface plots. The compositional analysis of the pretreated biomass at the optimum condition showed highest glucan content (59.3%), low xylan content (1%) and 36.74% lignin (Table 1) indicating high removal of xylan and retention of glucan and lignin. 3.3. Solid and enzyme loading – sensitivity analysis Pretreatment and saccharification with high solid loading and low enzyme would be an advantage to reduce the process cost. Therefore to study the sensitivity of the pretreatment process to solid loading, experiments were conducted at optimized condition by varying the solid loading from 10% to 20% (The maximum solid that could be loaded in the reactor). The results show (Fig. 2) that there was no significant difference in the mass recovery and composition at 20% loading in comparison to 10% for the solid fraction after pretreatment. However, as expected the glucose, xylose and acetyl group composition in liquid fraction of 20% loading were almost doubled (Fig. 3) meaning that the effectiveness of the pretreatment was same at both 10 and 20% solid loading. Consequently from the economical point of view pretreatment at 20% solid loading would be preferable. To study the sensitivity of enzyme loading to saccharification, pretreated solid (at optimized conditions) was subject to saccharification at low enzyme loading. The results (Fig. 4) shows that when enzyme dosage was reduced by 50% from 20 to 10 FPU/g glucan (Cellulase) and 40 to 20 IU/g glucan (b-glycosidase) the glucose yield reduced by 15%. An economic trade-off will have to be made between the cost gained due to enzyme savings and ethanol production loss at low enzyme loading (Ramachandriya et al., 2013). Glucose and ethanol yield upon saccharification and fermentation
Fig. 2. Composition of pretreated solid fraction biomass – solid loading impact on optimized pretreatment conditions (160 °C, 1.025% v/v and 10.5 min).
From the solid fraction of the saccharification process performed at 15% solid loading and low enzyme loading (10 FPU (Cellulase) and 20 IU (b-glycosidase)/g glucan), the compositional analysis and elemental analysis were performed to know the amount of solid lignin content and its heating value. Solid biomass of 0.344 g/g raw EFB rich in lignin (65.2%) with a heating value of 24.87 MJ/kg was available as solid waste after saccharification. This lignin rich and low ash (7.3%) content solid fraction could be used in the boiler to generate the heat required for the biorefinery or it could be used as raw material for biobased resins. However, 27.2% of glucan content present in the solid fraction had not been hydrolyzed in the saccharification process due to low enzyme loading. Increasing the enzyme loading could be an option to reduce glucan loss and improve ethanol production. The mass balance of the whole process of converting raw EFB to ethanol and lignin is shown in flow diagram (Fig. 5) based on the experimental process carried out under the pretreatment, low enzyme loading (10P FPU and 20 IU) saccharification and fermentation conditions. 51.1 g of ethanol, 344.1 g of lignin rich solid biomass and 3.7 l xylose rich (2.8% w/v) liquid could be produced from 1 kg of dry EFB. Ethanol production per kg raw material of this study is similar compared to 53 g ethanol/kg EFB at 25 FPU cellulase/g substrate reported by Piarpuzán et al. (2011). However, it is low compared to the value reported in other literatures (Table 4).
Fig. 3. Composition of pretreated liquid fraction – solid loading impact on optimized pretreatment conditions (160 °C, 1.025% v/v and 10.5 min).
Fig. 4. Enzyme loading impact on saccharification (FPU – cellulase, IU – bglucosidase/g glucan).
247
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248
Fig. 5. Mass balance results of bioethanol and lignin production from EFB.
Table 4 Comparison of present study results with literature studies. Composition of EFB (glucan, xylan and lignin)%
Pretreatment method
Enzyme dose (FPU)
Sugar subject to fermentation
Ethanol yield g/kg EFB
References
37.3, 32.7, 36.8, 34.6, 39.3,
Alkali (NaOH) Alkali (NaOH) Sodium bisulphite Alkali (NaOH) Alkali (NaOH)
34 50 20 40 25
Glucose Glucose Glucose Glucose Glucose
120 146 192 144 53
Formalin Sodium bisulphite Dilute acid (H2SO4)
15 15 10
Glucose Glucose and xylose Glucose
166 171 51
Sudiyani et al. (2013) Han et al. (2011) Tan et al. (2013) Jeon et al. (2014) Piarpuzán et al. (2011) Cui et al. (2014) Cheng et al. (2014) Present study
14.6, 21.4, 19.3, 17.1, 36.5,
31.7 23.5 17.6 26.4 22.3
37, 15, 17.4 30.5, 16, 17.1 33.5, 24.8, 20.1
Whereas the pure lignin content (224 g/kg raw EFB) produced in this study is higher than the value (140 g/kg raw EFB) reported by Cui et al. (2014). The difference in the results could be due to difference in enzyme dosage, composition of EFB, pretreatment methods and the sugars subject to fermentation (only glucose or both glucose and xylose) (Table 4). 4. Conclusion Studies using statistical design show that cellulose and hemicellulose in EFB could be effectively separated using dilute sulfuric acid pretreatment at 160 °C temperature, 1.025% v/v acid concentration, and 10.5 min. 51.1 g ethanol, 344.1 g solid residue and 3.7 l xylose rich syrup could be produced per kg EFB. Thus EFB from Brazil could be a potential raw material for biorefinery to produce ethanol, lignin and xylose that could be used for transport fuel, solid fuel and furfural production or animal nutrition respectively. EFB generated in Brazil in future could be managed by biorefinery concept to produce valuable products.
Acknowledgements Authors acknowledge Prof. Luiz França (UFPA) for his support with EFB samples and VALE International Inc. and École Polytechnique Fédérale de Lausanne (EPFL) for financial support. References Abdullah, N., Sulaiman, F., 2013. The Oil Palm Wastes in Malaysia, Biomass Now – Sustainable Growth and Use. In: Matovic, M.D., (Ed.), ISBN: 978-953-51-1105-4, InTech. doi: 10.5772/55302. http://www.intechopen.com/books/biomass-nowsustainable-growth-and-use/the-oil-palm-wastes-in-malaysia. Adney, B., Baker, J., 2008. Measurement of Cellulase Activities – Laboratory Analytical Procedure (LAP), Technical Report. NREL/TP-510-42628. National Renewable Energy Laboratory ed. Golden, CO. Alvira, P., Tomás-Pejó, E., Ballesteros, M., Negro, M.J., 2010. Pretreatment technologies for an efficient bioethanol production process based on enzymatic hydrolysis: a review. Bioresour. Technol. 101, 4851–4861. Braun, J.V., 2007. The World Food Situation – New Driving Forces and Required Actions. International Food Policy Research Institute, Washington. Carriquiry, M.A., Du, X., Timilsina, G.R., 2011. Second generation biofuels: economics and policies. Energy Policy 39, 4222–4234.
248
J.K. Raman, E. Gnansounou / Bioresource Technology 172 (2014) 241–248
Cheng, J., Leu, S.Y., Zhu, J.Y., Jeffries, T.W., 2014. Ethanol production from nondetoxified whole slurry of sulfite-pretreated empty fruit bunches at a low cellulase loading. Bioresour. Technol. 164, 331–337. Chiesa, S., Gnansounou, E., 2014. Use of Empty Fruit Bunches from the Oil Palm for bioethanol production: a thorough comparison between dilute acid and dilute alkali pretreatment. Bioresour. Technol. 159, 355–364. Chum, H.L., Johnson, D.K., Black, S.K., Overend, R.P., 1990. Pretreatment – catalyst effects and the combined severity parameter. Appl. Biochem. Biotechnol. 24–25, 1–14. Cui, X., Zhao, X., Zeng, J., Loh, S.K., Choo, Y.M., Liu, D., 2014. Robust enzymatic hydrolysis of formiline-pretreated oil palm empty fruit bunches (EFB) for efficient conversion of cellulose to sugars and ethanol. Bioresour. Technol. 166, 584–591. European Commission, 2006. EUR 22066 – Biofuels in the European Union – A Vision for 2030 and Beyond. European Commission. Luxembourg. . FAOSTAT, 2014. Food and Agriculture Organization of the United Nations. . Hamzah, F., Idris, A., Shuan, T.K., 2011. Preliminary study on enzymatic hydrolysis of treated oil palm (Elaeis) empty fruit bunches fibre (EFB) by using combination of cellulase and b 1–4 glucosidase. Biomass Bioenergy 35, 1055–1059. Han, M., Kim, Y., Kim, S.W., Choi, G.W., 2011. High efficiency bioethanol production from OPEFB using pilot pretreatment reactor. J. Chem. Technol. Biotechnol. 86, 1527–1534. Hu, F., Jung, S., Ragauskas, A., 2010. Impact of pseudolignin versus dilute acidpretreated lignin on enzymatic hydrolysis of cellulose. ACS Sustain. Chem. Eng. 1, 62–65. Jeon, H., Kang, K.E., Jeong, J.S., Gong, G., Choi, J.W., Abimanyu, H., 2014. Production of anhydrous ethanol using oil palm empty fruit bunch in a pilot plant. Biomass Bioenergy 67, 99–107. Kim, S., Kim, C.H., 2013. Bioethanol production using the sequential acid/alkalipretreated empty palm fruit bunch. Renewable Energy 54, 150–155. Li, J., Henriksson, G., Gellerstedt, G., 2007. Lignin depolymerisation/ repolymerization and its critical role for delignification of aspen wood by steam explosion. Bioresour. Technol. 98, 3061–3068. Liao, W., Wen, Z., Hurley, S., Liu, Y., Liu, C., Chen, S., 2005. Effects of hemicellulose and lignin on enzymatic hydrolysis of cellulose from dairy manure. Appl. Biochem. Biotechnol. 121, 1017–1030. NREL, 2014. Standard Procedures for Biomass Compositional Analysis. . Overend, R.P., Chornet, E., Gascoigne, J.A., 1987. Fractionation of lignocellulosics by steam-aqueous pretreatments [and discussion]. Philos. Trans. R. Soc. London, Ser. A 321, 523–536.
Piarpuzán, D., Quintero, J.A., Cardona, C.A., 2011. Empty fruit bunches from oil palm as a potential raw material for fuel ethanol. Biomass Bioenergy 35, 1130–1137. Qing, Q., Yang, B., Wyman, C.E., 2010. Xylooligomers are strong inhibitors of cellulose hydrolysis by enzyme. Bioresour. Technol. 101, 9624–9630. Ramachandriya, K.D., Wilkins, M., Atiyeh, H.K., Dunford, N.T., Hiziroglu, S., 2013. Effect of high dry solids loading on enzymatic hydrolysis of acid bisulfite pretreated Eastern redcedar. Bioresour. Technol. 147, 168–176. Sannigrahi, P., Kim, D.H., Jung, S., Ragauskas, A., 2011. Pseudo-lignin and pretreatment chemistry. Energy Environ. Sci. 4, 1306–1310. Selig, M., Weiss, N., Ji, Y., 2008. Enzymatic Saccharification of Lignocellulosic Biomass – Laboratory Analytical Procedure (LAP), Technical Report National. Renewable Energy Laboratory NREL/TP-510-42629, ed. Golden, CO. Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., Crocker, D., 2011. Determination of Structural Carbohydrates and Lignin in Biomass. Laboratory Analytical Procedure (LAP). Technical Report NREL/TP-510-42618. National Renewable Energy Laboratory ed. Golden, CO. Sluiter, J.B., Ruiz, R.O., Scarlata, C.J., Sluiter, A.D., Templeton, D.W., 2010. Compositional analysis of lignocellulosic feedstocks. 1. Review and description of methods. J. Agric. Food Chem. 58, 9043–9053. Sokhansanj, S., 2011. The Effect of Moisture on Heating Values. Biomass Energy Data Book – 2011. Center for Transportation Analysis. Oak Ridge National Laboratory. . Sudiyani, Y., Styarini, D., Triwahyuni, E., Kiky Sudiyarmanto, C., Sembiring Aristiawan, Y., Abimanyu, H., Han, M.H., 2013. Utilization of biomass waste empty fruit bunch fiber of palm oil for bioethanol production using pilot-scale unit. Energy Procedia 32, 31–38. Tan, L., Yu, Y., Li, X., Zhao, J., Qu, Y., Choo, Y.M., Loh, S.K., 2013. Pretreatment of empty fruit bunch from oil palm for fuel ethanol production and proposed biorefinery process. Bioresour. Technol. 135, 275–282. Thompson, W., Meyer, S., 2013. Second generation biofuels and food crops: coproducts or competitors? Global Food Secur. 2, 89–96. UNEP, 2011. Oil palm plantations: threats and opportunities for tropical ecosystems. UNEP Global Environmental Alert Service (GEAS). . Villela, A.A., Jaccoud, D.B., Rosa, L.P., Freitas, M.V., 2014. Status and prospects of oil palm in the Brazilian Amazon. Biomass Bioenergy 67, 270–278. Zhang, Y.H., Hong, J., Ye, X., 2009. Cellulase assays. In: Mielenz, J.R. (Ed.), Biofuels: Methods and Protocols Methods in Molecular Biology, vol. 581. Humana Press, NJ, pp. 213–231.
